(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Proceedings_of_the_sixth_international_soil_correlation_meeting(Vi_iscom)"

Proceedings of the Sixth International 
Soil Correlation Meeting (VI ISCOM): 

Characterization, Classification, and 

Utilization of Cold Aridisols 

and Vertisols 

Montana, Idaho, and Wyoming, United States, 
and Saskatchewan, Canada 

August 6-18, 1989 
March, 1991 

Edited by J.M. Kimble 

Soil Conservation Service 
Soil Management Support Services 

USDA 





Soil Conservation Service Soil Management Support Services 

USDA USAID 



Correct Citation: 

Kimble, J.M. 1990. Proceedings of the Sixth International Soil Correlation Meeting (VI ISCOM) - 

Characterization, Classification, and Utilization of Cold Aridisols and Vertisols. USD A, Soil 

Conservation Service, National Soil Survey Center, Lincoln, NE. 



Copies of this publication may be obtained from: 

National Leader World Soil Resources National Leader for Soil Classification 

Soil Management Support Services Soil Survey Division 

Soil Survey Division USDA-SCS 

USDA-SCS P.O. Box 2890 

P.O. Box 2890 Washington D.C- 20013 USA 

Washington D.C. 20013 USA 



Proceedings of the 
Sixth International Soil Correlation Meeting (VI ISC 

Characterization, Classification, and Utilization of Cold Aridisols and ^ 

Montana, Idaho, and Wyoming, United States, and Saskatchewan, Cana 

August 6-18, 1989 
March, 1991 



Organized by: 

USDA, SOS, SMSS, Lincoln, NE and Washington, DC 

Agriculture Canada 

USDA, SCS, Montana, Idaho, and Wyoming 

USDI, BLM, Montana and Wyoming 

University of Saskatchewan, Saskatoon, Saskatchewan, Canada 

Sponsored by: 

USDA, SCS, Washinton, D.C. 
USAID, Washington, D.C. 



Editor 

J.M. Kimble 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Workshop Committees 



Steering Committee 

H. Eswaran, National Leader SMSS, SCS, Washington, DC 

J. Witty, National Leader, Soil Classification, SCS, Washington, DC 

Juan Comerma, Chairman of ICOMERT 

Ahmet Osman, Chairman of ICOMID 

Organizing Committee 

J. Witty, National Leader, Soil Classification, SCS, Washington, DC (Chairman) 
J. Kimble Research Soil Scientist, SMSS, SSIV, NSSC, SCS, Lincoln, NE 
T.D. Cook, Soil Mangement Specialist, SMSS, SCS, Washington, DC 
A. Mermut, University of Saskatchewan, Saskatoon, Saskatchewan, Canada 
D. Acton, Agriculture Canada, Saskatoon, Saskatchewan, Canada 

Organized by 

USDA, SCS, SMSS, Lincoln, NE and Washington, DC 

Agriculture Canada 

USDA, SCS, Montana, Idaho, and Wyoming 

USDI, BLM, Montana and Wyoming 

University of Saskatchewan, Saskatoon, Saskatchewan, Canada 



SIXTH iNTERNATIONAl S Oil C LASSIFICATION WORKSHOP - ill - 



Preface 



The Sixth International Soil Correlation Meeting (VI ISCOM) was organized by the Soil 
Mangement Support Services in conjunction with the Soil Survey Division, Soil Conserva- 
tion Service, the SCS Soil Staffs and BLM Soils Staffs in Montana, Idaho, and Wyoming, 
Agriculture Canada, Saskatoon, Saskatchewan, Canada, and the University of Sas- 
katchewan, Saskatoon, Saskatchewan, Canada. The meeting was held August 6-18, 1989. 

The meeting addressed cold Aridisols and Vertisols. Proposed Keys developed by the 
chairmen of the respective committees were tested in the field and discussed. Modifications 
of the proposals were presented and new proposals developed. Based on the discussion, a 
revision of the Vertisol order was developed and submitted to John Witty. Proposals to 
change the Aridisol order are still under consideration. 

One of the more interesting and far-reaching proposals was the one to drop the Aridisol 
order and include soils which now fall in it in the other orders, based on their gentic horizons. 
This proposal created a very active discussion in which very strong feelings were expressed. 
We thank all of the tour participants for such frank discussions. Without them, the meeting 
could not have been as successful as it was. Having a large number of participants with a 
common interest interacting as a group helps develop new ideas and a better understanding 
of old ones. 

This Proceedings was assembled to allow others to have the benefit of the material pre- 
sented on the tour by the various authors. It represents a current reference to the latest 
thinking on Aridisols and Vertisols. The idea to drop the Aridisol order was not accepted at 
this time, but it did generate a great deal of discussion and made many of the participants 
look at the idea. In fact, many thought it had merit. It even lead to a proposal to also drop 
the Mollisol order, which may have been suggested in jest but has merit if we consider how 
the other orders are defined. As editor, I feel both of these topics will arise again and cause 
many soil scientists to examine the concepts of both Aridisols and Mollisols, two orders which 
seem not to follow the ideas of Soil Taxonomy. 

J.M. Kimble, Editor 



- iv - SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



Acknowledgements 

The editor wishes to thank all of the authors who prepared manuscripts and presented 
them to the tour participants and then sent them in to be published in this Proceedings. 
Thanks also are due to all of the participants who took part in the peer review of the manu- 
scripts. Without the efforts of all concerned, the maunscripts would not have ended up in the 
form which they did. Special thanks is also given to Maria Lemon, Ph.D., of the Editor Inc., 
for completing an English edit of all of the manuscripts and for preparing the proceedings in 
its final format. 

Thanks also go to Kristen Stuart, Anthony Flores, and Nancy Martinez for handling all 
of the mailings for the manuscripts' technical edits and for correcting the final manuscripts. 
Without their time and efforts, this Proceedings would not have been completed. 

Special thanks also are given to G. Durnbush, Director, Midwest National Technical 
Center, for making so many of the National Soil Survey Centers employees available to take 
part and help in conducting the tour and in preparing the manuscripts for this Proceedings. 
The State conservationists and the soils staffs from the SCS offices in Montana, Idaho, and 
Wyoming are thanked for all of their efforts, as are the many other soil scientists who helped 
make the VI ISCOM a sucess. 

Similar thanks go to the State BLM diredctors in Montana and Wyoming and their soil 
staffs for active and enthusiastic support for the tour. We also thank Dr. Acton, Agriculture 
Canada, and Dr. Mermut, University of Scakatchewan, for their assistance in all phases of 
ISCOM VI. 

We also wish to recognize the steering committe members for their encouragement and 
guidance. Special thanks go to Dr. Juan Comerma for preparing the Vertisol proposals and 
to Dr. A. Osman for preparing the Aridisol proposals. We also would like to regonize both Dr. 
J. Witty and Dr. H. Eswaran for their contributions to both proposals. Without the efforts of 
Dr. Witty and his staff, the many proposals put forth would not be considered, and much 
careful consideration and testing is required before the necessary changes can be made to 
Soil Taxonomy to reflect the contributions of such a wide and diverse group of participants 
and committe members. 

We are genuinely indebted and grateful to the all these individuals and staffs; however, 
we do not want to overlook the many others, too numerous to name, who contributed to 
making the ISCOM such a sucess. We gratefully recognize their contributions. 

The Editor, 

for all of the organizing committee 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP -v- 

Contents 

Classification of Vertic Intergrades: 
Macromorphological and Micromorphological Aspects 

by W.A. Blokhuis, L.P. Wilding, and M.J. Kooistra 1 

Desertification: 

Concept, Evaluation, Status 

by T.G.Boyadgiev 8 

II. Soil Formation in the Arid Regions of Israel 

by J.Dan 12 

Physical Properties Affecting the Productivity and Management 
of Clay Soils in Saskatchewan 

by E. de Jong and J. A. Elliott 29 

Alternative Classification of Soils 
with Aridic Soil Moisture Regimes 

by R. J. Engel, J.E. Witty, and J.D.Nichols 39 

Soil Forming Processes in Soils with Cryic and Frigid 
Soil Temperature Regimes in Idaho 

byM.A. Fosberg, A.L. Falen,R.R. Blank, and KW. Hippie 43 

Altocryic Aridisols in China 

by Gong Zitong and Gu Guoan 54 

Aridisols of Spain 

by J. Herrero and J. Porta 61 

Aridisols of New Zealand 

by A.E. Hewitt and F.G. Beecroft 67 

Australian Vertisols 

byKF.Isbell 73 

Properties and Classification 
of Cold Aridisols in Montana 

by C. Wang and Tom Keck, Jerry Nielsen, Robert Richardson, and Gordon Decker 81 

Effects of Minerology and Climate on the Development 

of Vertic Properties in Clayey Soils of Central and Eastern Canada 

byC.R. deKimpe, C.J.Acton, C. Wang, and M. C. Nolin 92 

Procedures and Rationale for Development 

of Adequate and Comprehensive Field Soil Data Bases 

byR.Langohr 100 

The Soils of the Desertic and Arid Regions of Chile 

by Walter Luzio-Leighton 104 

A Review of Recent Research on Swelling Clay Soils in Canada 

by A.R. Mermut, D.F. Acton, and C. Tarnocai 112 

Thermal Regime and Morphology of Clay Soils in Manitoba, Canada 

byG.F. Mills, R.G.Eilers, and H. Veldhuis 122 

Periglacial Features as Sources of Variability 
in Wyoming Aridisols 

byL.C.Munn 133 

A Comparison of Land Use and Productivity 

of Clay and Loam Soils within the Interior Plains of Western Canada 

by C. Onofrei, J. Dumanski, R.G. Eilers, and R.E. Smith 138 

Reclamation Management and Techniques 
for Cold Entisols in Southwestern Wyoming 

byF.E. Par ady and Norman E. Hargis 146 



-vi- CONTENTS 



Vertisols of New Caledonia: Distribution; Morphological, 
Chemical, and Physical properties; and Classification 

by P. Podwojewski and A.G. Beaudou 151 

Salinity Development, Recognition, and Management 
in North Dakota Soils 

by J.L. Richardson, D.G. Hopkins, B.E. Seelig, and M.D. Sweeney 159 

Aridisols of Argentina 

by C.O. Scoppa and R.M. di Giacomo 166 

Classification of Aridic Soils, Past and Present: 
Proposal of a Diagnostic Desert Epipedon 

byA.Souirji 175 

Report on Site Specific Soil-Climate-Vegetation Relationships: 
Cold Vertisols-Aridisol ISCOM Tour (Montana - Idaho - Wyoming) 

by G.J. Staidl and S.G. Leonard 185 

Soil Management Research on the Clay Soils 
of Southwestern Ontario - A Review 

by J. A. Stone 197 

Soil Classification Related Properties of Salt-Affected Soils 

byl.Szabolcs 204 

Clayey Soils of Northern Canada and the Cordillera 

by C. Tarnocai, G.F. Mills, H. Veldhuis, H. Luttmerding, and A. Green 208 

Vertisols of France 

by Daniel Tessier, Ary Bruand, and Yves-Mary Cabidoche 227 

Close Interval Spatial Variability of Vertisols: 
A Case Study in Texas 

by L.P. Wilding, D. Williams, W. Miller, T. Cook, andH. Eswaran 232 

Geomorphology of Cold Deserts 

by Robert C. Palmquist 248 



Classification of Vertic Intergrades: Macromorphological and 

Micromorphological Aspects 

W.A. Blokhuis*, L.R Wilding, and M.J* Kooistra 1 

Abstract 

Differentiation between typic and vertic subgroups of Alfisols, Incepti- 
sols, and Mollisols has been based mainly on the shrink-swell potential of 
the soil material, expressed as a coefficient of linear extensibility (COLE) 
and/or potential linear extensibility (PLE). Clay plasticity index and liquid 
limit also have been used as differentiae. Other soil characteristics are of- 
ten different between vertic and typic subgroups, or between vertic sub- 
groups and Vertisols, but these are insufficiently significant to be used as 
differentiating criteria. 

Micromorphological data do not consistently support any separation 
between vertic intergrades and Vertisols or between vertic and typic sub- 
groups in other soil orders. Some forms of plasma separations and voids 
may seem to be characteristic of Vertisols, but they are not unique and can 
not be used as differentiae. 

This paper proposes to differentiate between two kinds of intergrades: 
firstly, soils that fulfill the present requirements on COLE/PLE, have vertic 
characteristics in the surface soil but lack such characteristics in the B ho- 
rizon because of insufficient wet-dry cycles to generate shear failure, and, 
secondly, soils with a vertic soil structure at some depth but lacking vertic 
characteristics in the surface soil. 



Introduction: Present and Former 
Soil Taxonomy Criteria 

Vertisols occur in the KEY to Soil Orders of 
Soil Taxonomy (Soil Survey Staff, 1987) subse- 
quent to Histosols, Spodosols, and Oxisols, as 
"Other soils that: 

1. Do not have a lithic or paralithic contact, 
petrocalcic horizon, or duripan within 50 
cm of the surface; 

and 

2. After the soil to a depth of 18 cm has been 
mixed, as by plowing, have 30 percent or 
more clay in all subhorizons to a depth of 50 
cm or more; 

and 

3. Have, at some time in most years unless ir- 
rigated or cultivated, open cracks at a 
depth of 50 cm that are at least 1 cm wide 
and extend upward to the surface or to the 
base of the plow layer or surface crust; 

and 

4. Have one or more of the following: 

a. Gilgai; 

b. At some depth between 25 cm and 1 m, 
slickensides close enough to intersect; 

c. At some depth between 25 cm and 1 m, 



Agricultural University, Wageningen, The Netherlands; 
Texas A & M University, USA; The Winand Staring Centre for 
Integrated Land, Soil and Water Research, Wageningen, The 
Netherlands. 

*Corresponding author 



wedge-shaped natural structural ag- 
gregates that have their long axes 
tilted 10 to 60 degrees from the hori- 
zontal." 

If the requirement on gilgai is waived (in ac- 
cordance with ICOMERT proposals), criterion 4 
is essentially one on Vertic structure'. We use 
the term Vertic structure 7 when wedge-shaped 
or parallelepiped structural aggregates and/or 
intersecting slickensides are present. If a pedon 
satisfies both the requirements on cracking (3) 
and on soil structure (4), those on soil depth (1) 
and on clay content (2) normally will be satisfied 
as well. 

Vertic subgroups have been defined in the 
Orders Alfisols, Aridisols, Entisols, Inceptisols, 
Mollisols and Ultisols (Soil Survey Staff, 1987). 
Vertic subgroups differ from typic subgroups in 
the same great group by having: 

1. Cracks at some period in most years that 
are 1 cm or more wide at a depth of 50 cm, 
that are at least 30 cm long in some part, 
and that extend upward to the soil surface 
or to the base of an Ap horizon (in some 
vertic Alfisols: to the base of an albic hori- 
zon); 

and 

2. More than 35 percent clay in horizons that 
have total thickness of 50 cm; 

and 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



3. A coefficient of linear extensibility (COLE) 
of more than 0.05, 0.07 or 0.09 (depending 
on the soil moisture regime of the great 
group) in a horizon or horizons at least 50 
cm thick, and a potential linear extensibil- 
ity (PLE) of 6 cm or more in the upper 150, 
125 or 100 cm, respectively, of the soil, or in 
the whole soil if a lithic or paralithic contact 
is deeper than 50 cm but shallower than 
150, 125 or 100 cm, respectively (in some 
great groups there is no requirement for 
COLE). 

The mimimum COLE requirements are 0.09 
for udic and aquic, 0.07 for ustic, and 0.05 for 
xeric and aridic soil moisture regimes. Different 
COLE requirements have been introduced in 
order to reflect different moisture gradients: if 
the soil moisture changes are large or frequent, 
a smaller COLE would produce the same move- 
ment in the soil as a higher COLE would do in 
an environment where these changes are small 
or infrequent (Soil Survey Staff, 1975). 

The criteria on cracking are practically the 
same for vertic subgroups and for Vertisols, but 
there are differences in the requirements on clay 
content in certain sections of the profile. Swell- 
shrink characteristics must be shown in Verti- 
sols in a specific subsoil structure, whereas in 
vertic subgroups requirements on COLE and 
PLE indicate a potential to swell-shrink. 

The Soil Taxonomy concept of vertic inter- 
grades is one of soils that lack sufficient evi- 
dence of soil movement to meet the definition of 
Vertisols but that have high shrink-swell poten- 
tial. In most vertic subgroups, wet-dry cycles 
are insufficiently expressed to generate shear 
failure. A vertic subgroup may develop into a 
Vertisol with time, and under changing environ- 
mental conditions. One could think of the fol- 
lowing examples: 

1. A floodplain, in which the surface soils have 
sufficiently developed to produce structural 
aggregates and cracks, but where the sub- 
surface soil is still in an 'Entisol-' or Incep- 
tisol-stage/ perhaps not yet fully ripened 
(assuming adequate COLE). 

2. A smectitic clay soil under aridic conditions, 
where the sparse and erratic rainfall has 
created sufficient dry/wet alternations in 
the surface soil to generate cracks but has 
seldom moistened the subsurface soil to 
such an extent that swelling pressures and 
shear failure are generated. 



If vertic soils are potential Vertisols, it is not 
surprising that there are many common proper- 
ties. Bartelli and McCormack (1976) emphasize 
the great similarity between Vertisols and soils 
in vertic subgroups in such characteristics as in- 
stability because of swelling, clay mineralogy, 
COLE value, clay content, plasticity index, liq- 
uid limit, 15-bar (ISOOkPa) water and cation ex- 
change capacity. Both Vertisols and vertic soils 
belong to group CH (clayey soils with high liquid 
limits) in the Unified Soil Classification System 
(DeMent and Bartelli, 1969). 

In Soil Taxonomy (Soil Survey Staff, 1975) 
Lithic Vertic and Paralithic Vertic subgroups 
were recognised in Mollisols and Inceptisols. 
These had a lithic or paralithic contact between 
25 and 50 cm depth and horizons totalling 25 cm 
or more in thickness that had either more than 
35% montmorillonitic clay or COLE >0.05 or 
0.07 (depending on soil moisture regime). In the 
3rd edition of the KEYS (Soil Survey Staff, 
1987) these subgroups have been waived. 

Vertic Integrades: The Need for a 
Wider Concept 

No provision is made in Soil Taxonony for the 
many Alfisols, Mollisols, and Inceptisols that 
have a vertic structure in some part of the solum 
- usually accompanied by relatively high COLE 
values and high clay contents - but that do not 
meet the requirements on cracking and there- 
fore can not be placed in a vertic subgroup. 

In clayey soils it is the combination of crack- 
ing and a vertic structure that characterizes 
Vertisols. There would be some logic in defining 
soils that have either the morphological surface 
soil characteristics (i.e., cracking) or the mor- 
phological subsurface soil characteristics (i.e., a 
vertic structure) as vertic intergrades. In Soil 
Taxonomy, however, only the first category is 
recognized - with the additional requirement 
(COLE, PLE) that the soil material has a poten- 
tial to develop a vertic structure. A wider con- 
cept of vertic intergrades that includes non- 
cracking clay soils with a vertic structure in the 
subsoil is worth considering. 

When scanning soil reports, excursion guides, 
and the like on vertic 'intergrades', one finds 
that the term 'vertic' often has been applied 
rather loosely to soils with cracks, soils with 
pressure faces, soils with slickensides. How- 
ever, there is always one vertic symptom: either 
the cracking or the vertic subsoil structure. And 



BLOKHUIS, WILDING, and KOOISTRA: CLASSIFICATION OP VERTIC INTERGRADES 



this seems, from the point of view of macromor- 
phology, a reasonable entrance. 

Of the nine pedons described as vertic sub- 
groups in India (Kooistra, 1982), none is con- 
vincingly vertic according to the present Soil 
Taxonomy criteria. The name vertic has been 
applied when there are cracks and/or when 
there are features of a vertic structure. COLE 
values given for some of these pedons meet the 
requirements for vertic subgroups. 

Vertic subgroups of Alfisols, Inceptisols, and 
Mollisols in Texas, USA, were described by Hall- 
mark et al. (1986). Profile and site descriptions 
and analytical data, including COLE, were 
given. Cracks have been recorded in only three 
of the eleven vertic intergrades, but one has to 
realise that cracks were described only when 
they were apparent at the time of field descrip- 
tion. COLE was always above the critical value. 
Slickensides, pressure faces, and parallelepi- 
peds - alone or in combination - occurred at some 
depth in ten of the eleven pedons. 

If rules are to follow practice, then these ob- 
servations underline the need for a wider con- 
cept of the vertic subgroup than the present one. 

Macro- and Micromorphological 
Considerations 

There are many suggestions in the literature 
that the obvious differences in macromorphol- 
ogy between Vertisols and vertic subgroups (ac- 
cording to the present definitions) are matched 
with differences in micromorphology. However, 
if one tries to generalize the observations that 
have been made, there is the problem that vertic 
subgroups may, or may not, have a vertic struc- 
ture. 

The nine benchmark soils of India, described 
as vertic subgroups (Kooistra, 1982), had 
weakly to moderately developed vosepic and/or 
skelsepic plasmic fabric. Only two pedons in 
this group had a distinct vertic macrostructure 
and masepic plasmic fabric in some part. 

Plasmic fabric of Vertisols described in the 
same study was usually moderately to strongly 
developed vo-skelsepic, whereas four Vertisols 
had, in addition, masepic plasmic fabric. Skelse- 
pic plasmic fabric only occurred in connection 
with coarse fragments over 4 mm diameter. 

Dasog et al. (1987) described four pedons with 
vertic properties in Saskatchewan, Canada: two 
Vertisols and two Argic Vertic Cryoborolls. The 
main differences between Vertisols and vertic 
intergrades were in macromorphology (slicken- 



sides, versus prismatic structure with illuvia- 
tion argillans) and in micromorphology (masepic 
and lattisepic plasmic fabric were common 
throughout most of the solum in Vertisols, 
whereas they were few in vertic subgroups, and 
confined to the BC horizon). There were minor 
differences in other characteristics (CEC, clay 
content, clay mineralogy, COLE, pH, liquid 
limit, plasticity index), and, if these were taken 
into consideration as well, next to the macro- 
and micromorphological features, Vertisols dif- 
fered clearly from Vertic Borolls. However, it 
was difficult to single out one or two of these 
properties as diagnostic criteria. 

Bui and Mermut (1988) measured the total 
length, the area percentage (total length per 
unit area, in cm/cm2), and the orientation (in 10 
degree-intervals) of planar voids, using an im- 
age analyser, in three Vertisols from India, four 
from Ghana, and one from Saskatchewan, as 
well as three vertic subgroups from Sas- 
katchewan. The vertic soils had columnar or 
prismatic macrostructure and most planar voids 
were oriented vertically, whereas in Vertisols 
the dominant orientation of planar voids was 
subhorizontal and oblique. 

Mermut et al. (1988) state that vertic sub- 
groups lack planar vosepic, skelsepic, and mase- 
pic plasmic fabric. 

Nettleton et al. (1983) tried to find micromor- 
phological parameters for turbation. They stud- 
ied eight Vertisols and six soils in vertic sub- 
groups (two Torrifluvents and four Hapludalfs) 
in thirteen counties across the United States. 
Physical and chemical characteristics (including 
linear extensibility) were similar in Vertisols, 
Torrifluvents, and the Bt horizons of the 
Hapludalfs (other horizons of the Hapludalfs 
were not investigated). There were differences 
in macromorphology: vertic subgroups had 
cracks, but lacked intersecting slickensides; and 
in micromorphology: the Torrifluvents had a 
silasepic plasmic fabric, the Hapludalfs had an 
omnisepic plasmic fabric, and the Vertisols had 
skelsepic plasmic fabric in the upper and mase- 
pic in the lower part of the solum. One of the 
Hapludalfs had a few slickensides and areas 
with masepic plasmic fabric. Note that vosepic 
plasmic fabric was not described in any of the 
soils discussed. 

Masepic plasmic fabric, with minor areas 
with skelsepic and vosepic plasmic fabric, was 
characteristic for Chromic Pelloxererts de- 
scribed by Nettleton and Sleeman (1985). 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Masepic plasmic fabric was common, next to 
vosepic and skelsepic plasmic fabric, in the 
Branyon series (microlow), a Udic Pellustert in 
Texas (Hallmark et al.,1986). 

In calcareous lower Bw horizons of Vertisols, 
the plasmic fabric was crystic, but decalcified 
thin sections revealed mostly masepic plasmic 
fabrics (Wilding and Drees, 1989). 

The above citations suggest that certain mi- 
cromorphological features, notably masepic 
plasmic fabric - but to a smaller extent also the 
vosepic, skelsepic, and lattisepic types - and pla- 
nar voids are characteristic of Vertisols and are 
rare, weakly developed or absent in vertic sub- 
groups. However, other observations show that 
these features also are found in vertic subgroups 
and in non-vertic soils, and in some cases are 
more strongly developed in non-vertic soils than 
in related Vertisols. 

Many of the soils reported by McCormack and 
Wilding (1974; 1975), Ritchie et al. (1974), 
Smith and Wilding (1972), Smeck et al. (1981), 
Stahnke et al. (1983), Rabenhorst and Wilding 
(1986), Sobecki and Wilding (1983), and Wilding 
and Tessier (1988) had skelsepic, masepic, and 
lattisepic plasmic fabrics. Some had, in addi- 
tion, vosepic plasmic fabric. The soils studied 
were Vertisols, vertic subgroups in other orders, 
and non-vertic clayey, fine-loamy and loamy 
soils. Most of the non-vertic soils belonged to 
aquic suborders or subgroups. They included 
Aerie and Mollic Ochraqualfs, Aerie Fragiaq- 
ualfs, Typic Argiaquolls, and Aquic and Typic 
Paleustalfs. 

Yerima et al. (1987) found skelsepic and 
masepic, next to insepic, mosepic, and argillase- 
pic plasmic fabric, in a Vertic Argiustoll in El 
Salvador. 

Nettleton and Sleeman (1985) described some 
non-vertic clayey soils that had plasma modifi- 
cations like Vertisols, viz. masepic, lattisepic, 
and vosepic plasmic fabric. 

Blokhuis et al. (1989) gave examples of ripen- 
ing clay sediments that had a stronger develop- 
ment of vosepic and masepic plasmic fabric than 
was usually found in Vertisols and in vertic sub- 
groups. Unistrial and omnisepic plasmic fabrics 
were also common in such sediments. 

Linear Extensibility 

Studies by Nettleton et al. (1969) showed that 
soils with argillic horizons that had omnisepic, 
lattisepic, and masepic plasmic fabrics had 
COLE values in the range 0.05-0.06, whereas in 



similar soils that had mosepic plasmic fabric 
COLE values were mostly around 0.03. Nettle- 
ton et aL (1983) found that Vertisols with mase- 
pic plasmic fabric usually had COLE 0.06-0.16, 
but some Vertisols without masepic plasmic fab- 
ric had COLE 0.04-0.06, i.e., below the require- 
ments on COLE for vertic subgroups. 

McCormack and Wilding (1975) and Yule and 
Ritchie (1980) showed that expansive soils con- 
tinue to swell at moisture tensions below 1/3-bar 
tension and continue to shrink at moisture ten- 
sions above 15 bar. COLE, therefore, is an in- 
dex for swelling and shrinking between limits, 
not a measure of maximum swell and shrink. 
Also, COLE is only a potential index of shrink- 
swell. McCormack and Wilding (1975) found 
that a confinement stress equivalent to 10% of 
the swelling pressure would reduce the percent- 
age total swell by 50%. Hence, actual shrink- 
swell displacement is strongly conditional on 
overburden (soil thickness). 

Discussion 

Diagnostic Criteria for Vertisols, Vertic 
Subgroups, and Non-vertic Soils 

The micromorphology of vertic subgroups 
strongly overlaps with Vertisols and non-vertic 
soils. Some authors working in specific areas 
could differentiate between Vertisols and vertic 
intergrades, but even so the differences found 
were in grade of development and relative abun- 
dance rather than in type of plasmic fabric or 
voids. We found suggestions of correlations (see 
above), the most distinct one being that between 
masepic plasmic fabric and a vertic macrostruc- 
ture. This observation however, is, not of great 
help, as vertic subgroups and non-vertic soils 
may or may not have a vertic macrostructure 
and, hence, may or may not have masepic (and 
vosepic, lattisepic) plasmic fabric. Skelsepic 
plasmic fabric is a less reliable parameter for 
stress, as it is influenced by amount and size of 
sand particles and coarse fragments. 

At first sight the picture gets even more con- 
fused if we turn to non-vertic soils. Nettleton 
and Sleeman (1985) found that some non-vertic 
clayey soils had plasma modifications like Verti- 
sols, viz., masepic, lattisepic, and vosepic plas- 
mic fabric. Blokhuis et al. (1989) gave examples 
of ripening clay sediments that had a stronger 
development of vosepic and masepic plasmic 
fabric than is usually found in Vertisols and in 
vertic subgroups. Unistrial and omnisepic plas- 
mic fabrics are also common in such sediments. 
These authors also found that soil strength fail- 



BLOKHUIS, WILDING, and KOOISTRA: CLASSIFICATION OP VERTIC INTERGRADES 



ure - that obviously occurs mainly along major 
slickensides - is not correlative with microshear 
in the soil matrix: vosepic plasmic fabric ap- 
peared to be most strongly developed along the 
narrowest of planar voids. 

Nettleton et al. (1983) gave the following in- 
terpretation of micromorphological features, 
based on their own work and that of others: 

- Planar voids are an expression of deposi- 
tional and/or stress-strain history of the 
soils; 

- Silasepic plasmic fabric indicates absence of 

strain; omnisepic, skelsepic and lattisepic 
plasmic fabrics indicate a more or less bal- 
anced three-dimensional strain. These 
forms would occur in vertic subgroups; 

- Masepic plasmic fabric is an expression of 

strain; it is typical of Vertisols, especially in 

lower horizons. 

McCormack and Wilding (1974) interpreted 
masepic and lattisepic plasmic fabrics as stress 
failures. Wilding and Tessier (1988) reported on 
the microscopic and submicroscopic basis of 
shear failure. 

The occurrence of vosepic plasmic fabric in 
many Mollisols and Alfisols that show no ma- 
cormorphological evidence of shear (see e.g. 
Stahnke et al., 1983; Ritchie et al., 1974; So- 
becki and Wilding, 1983, and Smith and Wild- 
ing, 1972) is probably the result of plastic defor- 
mation rather than shear failure. 

Brewer (1964) suggested that vosepic plasmic 
fabric is the result of forces just insufficient to 
cause shearing of the soil mass, whereas mase- 
pic plasmic fabric is produced by shearing. Jim 
(1986) found, in experimental studies, that pla- 
nar voids and associated vosepic plasmic fabric 
resulted from shearing, whereas masepic plas- 
mic fabric could be a precursor of vosepic plas- 
mic fabric, prior to shearing and formation of a 
planar void. Field observations by Blokhuis et 
al. (1989) were in support of Jim's hypothesis. 

Another interpretation of strain-related forms 
of plasmic fabric is that they result from tensile 
stress, not from shear stress. Tensile stress 
could explain the strong development of vosepic 
and masepic plasmic fabrics in ripening muds 
(Augustinus and Slager, 1971; Koenigs, 1984). 
Shrinkage of a saturated clay soil on drying can 
be regarded as a compression of the soil fabric by 
an all-round pressure that is quantitatively 
equivalent to the internal soil water suction 
(Towner, 1961). 



Finally, one should realize that some methods 
of thin section preparation create soil stress and 
could enhance the development of shear-related 
forms of plasmic fabric or accentuate plasmic 
fabrics that were present in the natural soil. 
This may be the case when relatively wet clay 
soils (with pF 3 or less) are air-dried, or when 
epoxy resins are used for impregnation, requir- 
ing heating to 60-80 degrees Celsius. 

The conclusion of the above discussion must 
be that the micromorphology does not provide us 
with suitable parameters to distinguish be- 
tween Vertisols, vertic subgroups, and non- ver- 
tic soils. The question now arises: what other 
parameters are available? 

There is, firstly, the macromorphology, with 
two major aspects: vertic structure and crack- 
ing. Vertic structure readily can be ascertained 
in the field. Quantification of aspects of soil 
structure could be considered but is difficult and 
perhaps not necessary. Vertic structure is not 
diagnostic for vertic subgroups according to the 
present Soil Taxonomy definition. 

In the present definitions (Soil Survey Staff, 
1987) the requirements on cracking are largely 
the same for Vertisols and for vertic subgroups. 
Reversible cracking is an obvious feature of Ver- 
tisols, although these cracks can be obscured by 
a surface mulch, or an Ap, or in a moist soil. 

Linear extensibility expressed as COLE or as 
PLE was shown to indicate clearly whether or 
not a soil material had a potential for shrink- 
swell. Vertisols can be considered as soils that 
have realized this potential by developing a ver- 
tic structure. One would expect Vertisols to 
have at least the same COLE as vertic sub- 
groups, and this was confirmed in many obser- 
vations discussed above. The fact that COLE is 
a potential index for shrink-swell and not a 
measure for actual soil displacement - that is 
strongly conditional on overburden, as we have 
stated earlier - does not detract from its value as 
a differentiating characteristic between vertic 
and non-vertic soils. 

Soils that formerly were recognised as Lithic 
or Paralithic Vertic subgroups (Soil Survey 
Staff, 1975) cannot now be named vertic, as they 
have a lithic or paralithic contact at less than 50 
cm depth. One consequence of this is that many 
of the "shallow black soils" of the Deccan plateau 
in India, which have distinct vertic properties, 
cannot be classified either as a Vertisol or as a 
vertic subgroup, although they have distinct 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



vertic properties that are decisive for a specific 
land use. In. fact, these are probably Vertisols 
that have been truncated by soil erosion after 
ages of cultivation. 

If the soil material with high shrink-swell 
potential occurs deep in the profile, it would not 
normally be considered diagnostic in Soil Taxon- 
omy. However, for engineering purposes this 
property would be a major constraint (Cook et 
al., 1988) that might be taken into account if the 
taxonomy is to serve areas outside agriculture. 

Proposals to Amend the Definition of 
Vertic Subgroups 

The present definition of vertic subgroups 
centers around cracking, COLE/PLE, and the 
presence of one or more horizons with a high 
clay percentage. The fact that no provision was 
made for the presence of a vertic structure at 
some depth in soils other than Vertisols has led 
several soil scientists to propose modifications. 

Dasog et al. (1987) and Mermut et al. (1988) 
have suggested that, if the complete micromor- 
phology and, if possible, other soil characteris- 
tics as well are taken into consideration, it will 
be possible to differentiate at least between Ver- 
tisol and vertic subgroup. We think, however, 
that this 'central concept' approach cannot pro- 
vide us with parameters that can be applied 
world-wide. 

Cook et al. (1988) proposed to confine the defi- 
nition of vertic subgroups to the shrink-swell 
potential of the material and the requirement 
that the potential is present within a certain 
depth range. This proposal conforms, apart 
from details, to the present Soil Taxonomy re- 
quirements on COLE and PLE but waives the 
requirements on clay content and cracking. 

Klich et al. (in press) proposed that vertic in- 
tergrades should have, within a defined depth 
zone (i.e. within 1.25m from the soil surface), a 
subhorizon with a minimum thickness (i.e. 30 
cm) that has either a vertic structure or a mini- 
mum PLE (i.e. 5 cm). Cracks may or may not be 
present in soils that exhibit subsurface vertic 
conditions, and therefore these authors propose 
that the cracking requirement be waived. This 
proposal is similar to an earlier one by Graham 
and Southard (1983), who suggested that soils 
with significant shrink-swell activity (i.e. hav- 
ing a vertic structure) in subsoils that lie close to 
the surface probably should be allowed in vertic 
subgroups, regardless of surface cracking. 

In the revised FAO/Unesco system of soil clas- 
sification (FAO-Unesco-ISRIC, 1988) the term 
Vertic properties' is used "in connexion with 



clayey soils which at some period in most years 
show one or more of the following: cracks, slick- 
ensides, wedge-shaped or parallelepiped struc- 
tural aggregates, that are not in a combination, 
or are not sufficiently expressed, for the soils to 
qualify as Vertisols." Vertic soil units in Major 
Soil Groupings other than Vertisols have Vertic 
properties'. This approach, although lacking in 
accuracy of definition, is attractive as it is based 
entirely on field morphology. It does not include 
poorly drained 'potential Vertisols' that do not 
(yet) show surface cracks. 

In the French soil classification system 
(CPCS, 1967), vertic characteristics are recogni- 
sed at a level comparable to subgroups in Soil 
Taxonomy, these may have any - but not all - of 
the characteristics of Vertisols. 

Conclusions 

Considering the above discussion, we propose 
to distinguish between two types of intergrades: 

1. Those that fulfill the present requirements 
on COLE (e.g., over a thickness of at least 
50 cm) and on PLE (e.g., in the upper 100 
cm), that have cracks, but that lack a vertic 
structure. Desiccating clayey muds that 
develop irreversible cracking must be ex- 
cluded from this concept. 

2. Those that have a vertic structure at some 
defined depth (a requirement on COLE/ 
PLE is not required as these values will be 
in the Vertisol range in the section with 
vertic structure). 

Although the categories overlap, the former 
centre around potential Vertisols (cf. the present 
Soil Taxonomy definition of vertic subgroups), 
and the latter covers the soils that exhibit shear 
failure in the subsoil. The first category has the 
cracking features of a Vertisol, and the second 
normally has no cracks or cracks that are too 
small or too shallow. 

Requirements on clay content perhaps could 
be waived: soils with either high COLE or a ver- 
tic structure in a section of the profile usually 
have clay contents over 30 or 35% in that depth 
range. 

One could drop the COLE requirement or 
drop PLE and specify depth and thickness of 
layers that should have a specified mimimum 
COLE. 

Literature Cited 

Augustanus,? GEF, andS.Slager. 1971. Soil formation 
in swamp soils of the ooastelfiiigecf Surinam. Ge- 
oderma 6:203-211. 



BLOKHUIS, WILDING, and KOOISTRA: CLASSIFICATION OF VERTIC INTERGRADES 



Bartelli, L.J., and D.E. McCormack. 1976. Morphology 
and pedologic classification of swelling soils. Transpor- 
tation Research Record 568. National Res. Council, 
Washington D.C.; 8 pp. 

Blokhuis, W.A., M.J. Kooistra, and L.P. Wilding, 1989. 
Micromorphology of cracking clayey soils (Vertisols). 
Int. Working Meeting on Soil Micromorphology, San 
Antonio, Texas, July 10-15, 1988. 

Brewer, R. 1964. Fabric and mineral analysis of soils. 
Wiley, New York etc., 470 pp. 

Bui, E.N., and A.R. Mermut. 1988. Orientation of planar 
voids in Vertisols and soils with vertic properties. Soil 
Sci.Soc.Am.J. 52:171-178. 

Cook, T.D., H. Eswaran, and J. Kimble. 1988. Vertic in- 
tergrades in Soil Taxonomy. Trans. Int. Workshop 
Swell-Shrink Soils, October 24-28, 1988, NBSS & LUP, 
Nagpur, India, pp. 45-47 

CPCS, 1967. Classification des Sols. Travaux Commis- 
sion de Pedologie et de Cartographic des Sols 1963- 
1967. 96 pp. Mimeo. 

Dasog, G.S., D.F. Acton, and A.R. Mermut. 1987. Genesis 
and classification of clay soils with vertic properties in 
Saskatchewan. Soil Sci. Soc. Am. J. 51:1243-1250. 

DeMent, J.A., and L.J. Bartelli. 1969. The role of vertic 
subgroups in the comprehensive soil classification sys- 
tem. Soil Sci. Soc. Am. Proc. 33:129-131. 

FAO-Unesco-ISRIC. 1989. FAO-Unesco Soil Map of the 
World, Revised Legend. World Soil Resources Report 
60, Food and Agriculture Organization of the United 
Nations, Rome, 119 pp. 

Graham, R.C., and A.R. Southard. 1983. Genesis of a Ver- 
tisol and an associated Mollisol in Northern Utah. Soil 
Sci. Soc. Am. J. 47:552-559. 

Hallmark, C.T., L.T. West, L.P. Wilding, and L.R. Drees. 
1986. Characterization on data for selected Texas soils. 
Texas Agr. Expt. Station, Dept. of Soil and Crop Sci- 
ences, Soil Characterization Laboratory, College Sta- 
tion TX 77843. 

Jim, C.Y. 1986. Experimental study of soil microfabrics 
induced by anisotropic stresses of confined swelling 
and shrinking. Geoderma 37:91-112. 

Klich, I., L.P. Wilding, and A.A. Pfordresher (in press). 
Close-interval spatial variability of Udertic Paleustalfs 
in East-Central Texas. 

Koenigs, F.F.R. 1981. Discussion on paper by L.P. Wild- 
ing, Development of structural and microfabric proper- 
ties in shrinking and swelling clays. In J. Bouma and 
P.A.C. Raats (ed) Proceedings of the ISSS Symposium 
on Water and Solute Movement in Heavy Clay Soils. 
Int. Inst. Land Reclamation and Improvement, Publ. 
37, Wageningen, The Netherlands, p. 22. 

Kooistra, M.J. 1982. Micromorphological analysis and- 
characterization of 70 benchmark soils of India. Basic 
reference set. Neth. Soil Survey Inst., 778 pp.; in 4 vol- 
umes. 

McCormack, D.E., and L.P. Wilding. 1974. Proposed ori- 
gin of lattisepic frabic. In: G.K. Rutherford (Editor), 
Soil Microscopy. 4th Int. Working Meeting on Soil 
Micromorphology. The Limestone Press, Kingston, 
Ontario, pp. 761-771. 

McCormack, D.E., and L.P. Wilding. 1975. Soil properties 
influencing swelling in Canfield and Geeburg soils. 
Soil Sci. Soc. Am. Proc. 39:496-502. 

Mermut, A.R., J.L. Sehgal, and G. Stoops. 1988. Micro- 
morphology of swell-shrink soils. Trans. Int.Workshop 
Swell-Shrink soils, October 24-28, 1988. NBSS & LUP, 
Nagpur, India, pp. 127-144. 



Nettleton, W.D., K.W. Flach, and B.R. Brasher. 1969. Ar- 
gillic horizons without clay skins. Soil Sci. Soc. Am. 
Proc. 33:121-125. 

Nettleton, W.D., F.F. Peterson, and G. Borst. 1983. Micro- 
morphological evidence of turbation in Vertisols and 
soils in vertic subgroups. In: P.Bullock and C.P.Murphy 
(Editors), Soil Micromorphology. Vol.2: Soil Genesis. 
AB Academic Publ., Berkhamsted, Hrts., United King- 
dom, pp. 441-458. 

Nettleton, W.D., and J.R. Sleeman. 1985. Micromorphol- 
ogy of Vertisols. In: L.A.Douglas and M.L.Thompson 
(Editors). Soil Micromorphology and Soil Classifica- 
tion. Soil Sci. Soc.Am. Special Publication no. 15, pp. 
165-196. 

Rabenhorst, M.C., and L.P. Wilding. 1986. Pedogenesis on 
theEdwards Plateau, Texas: II. Formation and occur- 
rence of diagnostic subsurface horizons in a climose- 
quence. Soil Sci. Soc. Am. J. 50:687-692. 

Ritchie,A., L.P. Wilding, G.F. Hall, and C.R. Stahnke. 
1974. Genetic implications of B horizons in Aqualfs of 
Northeastern Ohio. Soil Sci. Soc. Am. Proc. 38:351-358. 

Smeck, N.E., A. Ritchie, L.P. Wilding, and L.R. Drees. 
1981. Clay accumulation in sola of poorly drained soils 
of Western Ohio. Soil Sci. Soc. Am. J. 45:95-102. 

Smith, H., and L.P. Wilding. 1972. Genesis of argillic ho- 
rizons in Ochraqualfs derived from fine-textured till 
deposits of Northwestern Ohio and Southeastern 
Michigan. Soil Sci. Soc. Am. Proc. 36:808-815. 

Sobecki, T.M., and L.P. Wilding. 1983. Calcic horizon dis- 
tribution and soil classification in selected soils of the 
Texas coast prairie. Soil Sci. Soc. Am. J. 47:707-715. 

Soil Survey Staff. 1975. Soil Taxonomy: a basic system of 
soil classification for making and interpreting soil sur- 
veys. USDA Agr. Handb. 346. U.S.Govt.Print.Office, 
Washington DC, 754 pp. 

Soil Survey Staff. 1987. Keys to Soil Taxonomy (third 
printing). SMSS Technical Monograph No. 6, Ithaca, 
NY. 

Stahnke, C.R., L.P. Wilding, J.D. Moore, and L.R. Drees. 
1983. Genesis and properties of Paleustalfs of North 
Central Texas: morphological, physical and chemical 
properties. Soil Sci. Soc. Am. J. 47:728-733. 

Towner, G.D. 1961. Influence of soil-water suction on 
some mechanical properties of soils. J. Soil Sci., 12:180- 
187. 

Wilding, L.P., and L.R. Drees, in press. Removal of car- 
bonates from thin-sections for microfabric interpreta- 
tions. Int. Working Meeting on Soil Micromorphology, 
San Antonio, Texas, July 10-15,1988. 

Wilding, L.P., and D. Tessier. 1988. Genesis of Vertisols: 
shrink-swell phenomena. In: Larry P.Wilding and 
Ruben Puentes (ed), Vertisols: their distribution, prop- 
erties, classification and management. 
Techn. Monograph 18, Texas A & M University, Print- 
ing Center, pp. 55-81. 

Yerima, B.P.K., L.P. Wilding, E.G. Calhoun, and C.T. 
Hallmark. 1987. Volcanic ash-influenced Vertisols and 
associated Mollisols of El Salvador: physical, chemical 
and morphological properties. Soil Sci. Soc.Am. J. 
51:699-708. 

Yule, D.F., and J.T. Ritchie. 1980. Soil shrinkage relation- 
ships of Texas Vertisols: I. Small cores. Soil Sci. Soc. 
Am. J. 44:1285-1291. 



SIXTH INTEKNATIONA] Soil CLASSIFICATION WORKSHOP 



Desertification: Concept, Evaluation, Status 

T.G.Boyadgiev 1 

Abstract 

This paper considers desertification as a comprehensive expression of 
natural or induced processes which destroy the equilibrium of the soil, 
vegetation, air, and water. Desertification is a continuous process going 
through several stages before reaching the final one, which is an irrevers- 
ible change. The process occurs in arid, semi-arid, and sub-humid areas. 

The processes leading to desertification are considered to be degradation 
of the vegetative cover, water and wind erosion, salinization, reduction of 
soil organic matter, soil crusting and compaction, and accumulation of sub- 
stances toxic to plants or animals. For each of these processes it is neces- 
sary to quantify the status, rate, inherent risk, and hazards of desertifica- 
tion. 

To assess and map desertification, concrete socio-economic, bio-climatic, 
and physico-geographic data are necessary. If these data are not available, 
the desertification can be assessed by using mathematical models applying 
the interpretative approach. In this case the desertification hazards can be 
assessed as a function of the soil status, vulnarability of land to desertifica- 
tion processes (water action, wind action, salinization), and animal and 
population pressure on the land. 

The data presented in the paper indicate that over 95 percent of the terri- 
tory of Africa is covered with soils having adverse properties for one or 
more cultures and that the negative anthropogenic impact can lead to rapid 
land degradation. 

The processes leading to desertification on the territory of the USSR are 
mainly vegetation degradation and salinization. 



Introduction 

The concept of desertification hazards is 
based on the following formulation (FAO/ 
UNEP, 1983): 

l.Desertification is a result of simultaneous 
action of natural mechanisms and mecha- 
nisms provoked by the influence of man 
and animals, but it can be restricted or 
stopped only through man's activities. 
2. Desertification is a continuous process 
which passes through many stages before 
reaching the final one, which is an irrevers- 
ible change. 

S.Desertification occurs in arid, semi-arid, 
and sub-humid zones, and its intensity is 
greatest in marginal zones. 
The basic processes of desertification are deg- 
radation of the vegetative cover, water erosion, 
wind erosion, salinization and alkanization, re- 
duction of the organic matter, disturbance of soil 
physical properties (crusting and compaction), 
and accumulation of substances toxic for plants, 
animals, and men. It is assumed that the influ- 
ence of climatic elements (drought and intensive 
rainfall, strong winds, high evapotranspiration, 



etc.) and anthropogenic activity leads much 
more quickly to desertification of areas with 
poor and shallow soils than of those having deep 
and fertile soils. 

For an accurate evaluation of desertification 
hazards, as outlined in FAO/UNEP (1983) 
methodology, concrete socioeconomic, bio-cli- 
matic, and physicogeographic data are required 
in order to determine the status, rate, and in- 
herent risk for each basic process. When such 
data are lacking or incomplete (as in developing 
countries) and the evaluation is urgently re- 
quired, then mathematical models are used, 
applying the interpretative approach. In such 
cases, desertification is evaluated according to 
the following categories (FAO/UNEP/ESRI, 
1984): 

l.Anthropogenic influence: It is assumed 

that this influence has the strongest effect 

on destruction of the vegetative cover and 

reduction of soil fertility. 
2.Land vulnerability to the water erosion, 

wind erosion, and salinization. 
3. Soil status at the time of the degradation 

evaluation. 



X N. Poushkarov Institute of Soil Science and Yield Pro- 
gramming. 



BOYADGIEV: DESERTIFICATION 



Anthropogenic Effect 

The continuous growth, of population, as com- 
pared to the static character of the natural re- 
sources (soil, water, vegetation) and the restric- 
tions of land biological productivity, is consid- 
ered by many experts as proof of the growing 
influence of the demographic factor on the land 
and therefore as an indicator of soil degradation 
and desertification hazard. In the FAO/UNEP 
concept this obvious fact takes a quantitative 
expression and allows, even though not fully, a 
substantiation of the anthropogenic influence. 
To that aim, the following relations are used: 

- Number of people/population supporting ca- 

pacity - for assessment of demographic 
pressure. 

-Number of animals/Livestock carrying ca- 
pacity - for assessment of livestock pres- 
sure. 

- The number of people and animals are 
taken from the annual statistics. 

To establish the population supporting capac- 
ity, the following data are used: area by crop 
kinds, yield of each crop, potential annual pro- 
duction, and corresponding value of calories. 
The quantity of calories divided by the popula- 
tion's needs per year determines the amount of 
demographic pressure per unit area, and this 
value divided by the population is used as an 
indicator of the demographic factor effect on the 
environment. It is considered that this effect is 
zero to slight if the ratio people/population sup- 
porting capacity is less than 1,0, and very strong 
when the ratio is greater than 12. 

To calculate the livestock carrying capacity 
for a given territory data on the total quantity of 
vegetation mass and the quantity of consumable 
fodder in kg per unit area are used. This 
amount, divided by the animals' requirements 
per year (2% of live weight per day), provides an 
idea of the degree of livestock pressure per unit 
area. The ratio of the number of animals/live- 
stock carrying capacity is used as a quantitative 
expression of the animal pressure on the envi- 
ronment. This pressure is zero to slight when 
the ratio is less than 0,4, and very strong when 
it is greater than 4,2. These limit values are 2,5 
times lower than those of the demographic fac- 
tor, which shows that the method places greater 
weight on the influence of animals on land deg- 
radation. 

In cases when no specific data on bio-mass are 
available, a formula can be used. Its calculation 
is based on the quantity of rainfall for the re- 



spective bio-climatic zone and on the soil factors 
restricting land productivity. Thus, for ex- 
ample, for the Mediterranean basin, Le 
Houerou (1977) proposes the following formula: 
consumable fodder (kg/ha) = 2,17P - 103,7, 
where P = mean annual precipitation (mm). 
This climatic potential is decreased by 25-50%, 
depending on soil properties unfavorable for 
development of plants (salinity, alkalinity, acid- 
ity, profile thickness, high gypsum and carbon- 
ates content, coarse texture, low fertility, etc.). 

Land Vulnerability 

Here is included the natural vulnerability of a 
given territory to water erosion, wind erosion, 
and salinization. 

A large amount of related information exists, 
particularly about water erosion and saliniza- 
tion. This information, however, does not allow 
us to establish continuous trends for the speed of 
the process or the hazard of reaching irrever- 
sable degradation. It is assumed that the inclu- 
sion of available information in a system of 
mathematical models will allow us to outline the 
hazard zones of desertification. 

The most important elements which the vul- 
nerability assessment models include those dis- 
cussed below. 

For Water Erosion 

For water erosion, the most important ele- 
ments are: 

-Aggressivity of precipitation with regard to 
natural vegetation and cultivated lands; 

-Topographic effect; 

-Coefficient related to the effects of texture, 
water permeability, and humus content of 
the soil; 

-Coefficient related to rock weathering in 
various bio-climatic zones. 

The values obtained from the combination of 
these elements can provide a quantitative ex- 
pression of the water erosion hazard. It is con- 
sidered as a zero to slight hazard if the value is 
lower than 1,9 and very great if this value is 
greater than 9,5. 

Water erosion hazard is greatest for medium 
textured soils developed on a slope under tropi- 
cal conditions. The ratio of texture to land slope 
is shown in Table 1. These data show that more 
than half of the land is medium textured and 
has rolling to mountainous slopes, and conse- 
quently is vulnerable to water erosion. 



10 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



For Wind Erosion 

The data about wind erosion hazard assess- 
ment are very limited and this made it neces- 
sary to rely on the findings of scientists that 
have studied the dependencies existing among 
the following elements: 

- wind velocity and transportation of soil par- 

ticles; 

- mean wind velocity and percentage of active 

winds; 

- movement of soil particles and moisture; 

- soil erodibility and texture; 

- status of soil surface and wind action. 

The establishment of figure values for each 
one of these relationships permits a quantita- 
tive determination of the wind erosion hazard. 
This hazard is zero to slight when the index is 
lower than 3 and very great when it is greater 
than 15. 

Judging by the light texture of the soils, 
which are most susceptible to wind erosion, 
about 18% of the land is subjected to wind ero- 
sion. 

For Salinization 

In world literature salinization is noted as a 
typical example of soil degradation, but precise 
information is not available regarding rate and 
hazard of this process. Even the classical ex- 
amples referred to in the literature, such as the 
Tigris and Euphrates valleys, the Indus valley 
in Pakistan, the Nile and Senegal delta, the 
galodnaya steppe in USSR, etc., do not allow 
precise assessment of the salinization. The rea- 
sons for this are numerous, but fundamentally 
they are the dynamism of the process and man's 
efforts to improve saline lands. 

Many models exist for assessment of saliniza- 
tion hazard. Most of them include the following 
elements: 

-Ratio precipitation/potential evapotranspi- 
ration. It must be noted that one and the 
same ratio can have different effects on 
salts accumulation in the soil. Thus, for 
example, in the sub-tropical zone the ratio 
P/ETP = 0,4 is observed in regions with 250 
mm precipitation, where saline soils occur 
widely. In contrast, with the same ratio in 
the tropical zone, precipitation is 700 mm 
and no saline soils are observed. This 
shows that the ratio P/ETP must be used 
differentially for each zone. 
-Maximum accumulation of salts in the soils 

in the various bio-climatic zones. 
-Capillary rise of the water, depending on the 
soil texture. 



Table 1 - Ratio of Texture to Land Slope 


Textural 
group of soil 
mapping unit 


Percentage of soil mapping 
unit with 


World land surface 
in 


coarse 
texture 


medium 
texture 


fine 
texture 


Sq.Km. 
xlOOO 


% 


coarse textured 
medium textured 
fine textured 
equilibrated 


*50 
<50 


^50 
<50 


2>50 
<50 


23495 
83052 
27908 
1220 


17,3 
61,2 
20,6 
0,9 


Slope 
group of soil 
mapping unit 

level to gently 
undulating 
rolling to hilly 
steeply dissected 
to mountainous 


Percentage of soil mapping 
unit with 


World land surface 
in 


0-8% 
slope 


8-30% 
slope 


>30% 
slope 


Sq. Km. 
xlOOO 


% 


>60 
<80 


0-100 


<20 
<50 
>50 


61600 
53942 
20133 


45,4 
39,8 
14,8 



-Depth and quality of ground water. 
-Topography of the region and status of soil 

cover. 

-Additional elements, such as mineralization 
of surface flowing water, high and low 
tides, composition of the rocks in the water 
catchment basin, microrelief, type of salini- 
zation, irrigation methods, etc. 
Inclusion of these elements (or a part of them) 
in models can provide a most general orienta- 
tion about salinization hazard. These calcula- 
tions, however, have a value only in an overall 
scale, where zones of relatively greater hazard 
can be indicated. Specific regions, however, re- 
quire a differentiated approach for each of the 
indicated elements. The calculations made for 
Africa show that regions with zero to slight haz- 
ard have an index lower than 3,2, while those 
with very great hazard have an index greater 
than 39,0. 

According to recent calculations, some 7,1% of 
the world's land has soils affected by saliniza- 
tion to various degrees. Based on the soil-cli- 
matic condition, it can be assumed that some 
more 8,2% additional areas actually are subject 
to this process. 

Soil Status 

The FAO/UNEP methods (1983) treat the soil 
status separately for each process. For an over- 
all assessment, however, this is why other ele- 
ments are taken into consideration. These ele- 
ments include: 

- soil requirements of the main crops towards 

the soils; 

- soil irrigability; 

- diagnostic horizons and soil properties; 

- variation of the soil cover; 

- land suitability (in three classes) for each 
one of the main crops. 



BOYADGIEV: DESEKTIFICATION 



11 



The soil constraints index obtained 
as a result of combinations of these 
elements provide information about 
unfavorable properties for plant 
growth. The index is used for general 
assessment of desertification haz- 
ards. Soils with an index value of 4,0 
or less are considered very good, and 
those with an index value of 20 or 
more are considered very poor. 

Accumulation of toxic substance 
for plants, animals, and men has a 
local specifity and cannot be treated 
on an equal level with the above cate- 
gories. An assessment of this process requires 
specific studies for each region affected by pollu- 
tion. 

Desertification Hazards 

The data obtained for each of the above cate- 
gories (Anthropogenic Effect, Land Vulnerabil- 
ity, and Soil Status) serve to establish an index 
of the overall desertification hazards. Desertifi- 
cation hazards are small when the overall index 
is lower than 42,2. When this index is greater 
than 127,5 the hazards are very great. By cal- 
culation, the degrees of the overall desertifica- 
tion hazards are presented separately for re- 
gions with the following climatic characteristics: 

- Areas without growing period, the most part 

of which are real deserts and only small 
areas (oases) can be subject of desertifica- 
tion; 

- Areas with 1 to 180 days of growing period, 

which are most sensitive to desertification 
processes; 

- Areas with more than 180 days of growing 
period and mountain territories with low 
temperatures, which are not subject to de- 
sertification. 

Results for Africa and USSR 

Due to the shortage of information for assess- 
ment of desertification in Africa, the interpreta- 
tion method was applied (Boyadgiev T.). The 
results are as shown in Table 2 (FAO/UNEP/ 
ESRI). 

These data show that more than 95% of Af- 
rica has soils with unfavorable properties for 
one or more crops and that the negative anthro- 
pogenic influence can lead to quick land degra- 
dation. In addition, animal pressure (on some 



Table 2 - Results for Africa 


Surface (in %) of Africa by rating class and process 
Rating Soil Water Wind Salinization Population Animal 
Class Constraints Erosion Erosion Pressure Pressure 


None to slight 4,5 
moderate 40,4 
severe 42,8 
very severe 12,3 


84,3 68,1 74,8 54,8 39,2 
12,6 22,8 7,8 31,9 42,4 
2,1 3,8 6,6 10,6 10,4 
1,0 5,3 10,8 2,7 8,0 


Table 3 - Results for USSR 


Degradation Bad land 
of vegatation Water Salinization Water (soil Areas not prone 
cover Erosion erosion status) to desertification 


km 2 770083 
% 43,2 


59036 132081 12228 58584 650483 
3,3 13,0 0,7 3,3 36,5 



60% of the territory) and population pressure 
(on some 45%) have a moderate to very great 
effect. From 15 to 32% of Africa is vulnerable to 
wind action, Salinization, and water action. 

Desertification hazards of the USSR Arid 
Lands (Babaev, A.G., 1988) appear in Table 3. 
These data show that the processes leading to 
desertification of the USSR arid lands are 
mainly degradation of vegetation cover and sal- 
inization. The effect of wind erosion is low and 
that of water erosion is insignificant. The sur- 
face of the bad land is reduced. 

Each one of these processes is manifested 
with different force in the individual countries. 
Specific measures for restricting, stopping, and/ 
or changing the trend of this process must be 
planned, depending on the major degradation 
process. 

References 

Babaev, A.G.(1988) Explanatory Note to the Map on Man- 
made Desertification of the USSR Arid Lands, Scale 1 : 
2,500,000. Ashkhabad, Turkmen SSR Academy of Sci- 
ence. 

Boyadgiev, T.G. (1984) Report on the Modelling for compi- 
lation of the Maps of Desertification Hazards in Africa 
and soil elements used in assessing Desertification and 
Degradation in the world, FAO, Rome. 

FAO/UNEP (1983) Provisional Methodology for Assess- 
ment and Mapping of Desertification, FAO, Rome. 

FAO/UNEP/ESRI (1984) Map of Desertification Hazards, 
Explanatory Note. 

Le Houerou, H. N., and F.H. Hoste. (1977) Rangeland 
Production and Annual Rainfall Relations in the Afri- 
can Sahels - Sudanian Zone. Journal of Range Manage- 
ment, Vol. 30, No. 3, Adsdis Ababa. 



12 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



II. Soil Formation in the Arid Regions of Israel* 

J. Dan 1 

Abstract 

Three climatic subregions i.e. the extremely arid zone, the arid semidesert zone and the 
mildly arid zone, are defined in southern Isreal. These climatic zones are crossed by four 
physiographic subregions that include mountainous areas, large valleys, plains and sandy 
areas. All subregions are characterized by a certain soil and vegetation pattern. 

The mountainslopes on hard rocks in the extremely arid zone are bare of soil cover. On 
soft rocks some physical weathering may be found. Reg soil formation characterizes moun- 
tain plateaus. The valleys are characterized by accumulation of eroded material that ranges 
from stones and gravel in the alluvial fans to fine sand, silt and clays in the inland depres- 
sions (play as). Both dry and wet saline playas are found in these areas. 

Reg soil (Gypsiorthids) that are characterized by a desert pavement cover the large 
plains. Coarse desert Alluvium (Torrifluvents) and some fine desert Alluvial soils are found 
in wadibeds. Seif dunes characterize the desert sandy areas. Fine desert alluvial soils, 
mixed with alluvial sand, are found along ancient and recent wadis that cross these dunes. 

Brown Lithosols (typic Torriorthents), both saline and nonsaline, are found in rocky 
pockets among hard rocks on mountain slopes in the arid, semidesert zone. On chalks and 
marls they are replaced by saline calcareous Lithosols and rendzinic desert Lithosols. 

Young loessial sediments (Torrifluvents, some Xerofluvents) are found in valleys among 
these mountains. On terraces, plains and mountain plateaus, especially in the moister part 
of this region, loessial Serozems (Haplargids) are found. 

Calcic and petrocalcic horizons characterize older gravelly and sandy sediments. Sand 
fields are found on recent sand. 

Alluvial Brown soils and alluvial silty -clayey Serozems are found on alluvial fan material 
in the lower Jordan Valley. Saline and gypsiferous highly calcareous Serozems were formed 
from Lacustrine sediments on the Lissan terrace while marly saline desert Lithosols are 
found on badlands that dissect this terrace. Brown Alluvial soils cover most of the Jordan 
floodplain. Solonchacks are found in depressions and the lowest part of the Jordan 
floodplain. 

Nonsaline brown Lithosols (lithic Torriorthents) are found on mountainslopes in the 
mildly arid parts of the northern Negev. Towards the north and moister parts they merge 
with Brown Rendzinas (lithic Xerorthents) and even with Terra-rossas. 

Loessial soils, mainly loessial Brown soils (calcic Haploxeralfs), cover footslopes and 
small valleys in the northern Negev. Towards the north they are replaced by natric-grumic 
Serozems (Natrargids) and finally by natric Grumusols and reddish brown Grummusols 
(Typic Chromoxererts). 

Loessial Brown soils (calcic Haploxeralfs) usually covering clayey paleosols up to 12 me- 
ters, cover most of the plains of the northwestern Negev. Young loess deposits are found in 
depressions. 

Sandy Regosols characterize the recent sandy deposits of the western Negev. At depth 
they cover mature paleosols. Sand dunes are found near the coast. 

Brown and reddish brown Grumusols (Chromoxererts) were formed from alluvial mate- 
rial on the Lissan terrace in the northern part of the Jordan valley. Hydromorphic 
Grumusols (Pelloxererts) cover impeded drained areas. Highly calcereous inseptic Brown 
soils (calcixerllic Xerochrepts) are found on lacustrine sediments on the Lissan terrace 
while rendzinic desert Lithosols and pale Rendzinas cover the terrace escarpment. Brown 
alluvial soils are found on the Jordan floodplain. 



Introduction 

Most arid regions, especially those of Israel, 
are heterogeneous in many respects. The mean 
precipitation ranges from about 350 mm down 
to less than 50 mm, and consequently the cli- 
mate varies from mildly arid to extremely arid 



1 Inst. of Soils and Water, Agricultural Research Or- 
ganization, The Volcani Center, Bet Dagan, Israel. 

^Published in 1981 as part of the field-guide of the Arid 
Soil Conference tour (Aridic Soils of Israel, Spec. Publ. No. 
190, Agr. Res. Org., Bet Dagan) 



(Rosenan, 1970) (Fig. II.l). Those differences in 
climate affect the vegetation, landscape, and 
soil formation and, as a result, also the land use. 
In order to emphasize these interrelationships, 
three climatic subregions have been defined 
(Dan and Raz, 1970). 

Climatic Subregions 

The first is the extremely arid zone, where the 
annual precipitation is usually less then 80 mm. 
The vegetation in this zone is restricted to favor- 
able ecological sites like dry riverbeds and rock 



DAN: SOIL FORMATION IN THE ARID REGIONS OP ISRAEL 



13 



Fig. II 1. Climate zone map. 




crevices (Zohary, 1955; Waisel et ai, 
1978) (Fig. 112). Due to the absence 
of vegetational cover, erosion - both 
fluvial and aeolian - is severe. 
Hence, the surface deposits are de- 
pleted of fine soil material which is 
redeposited in the moister parts of 
the country (Yaalon and Dan, 1974). 
Soil development is very slow be- 
cause of the extreme dryness. On old 
stable desert surfaces, accumulation 
of airborne salts becomes marked. 

The second zone, the arid semi- 
desert zone, is located wherever the 
annual precipitation ranges from 80 
mm to about 200 mm. The vegeta- 
tion of this zone is diffuse (Danin et 
al., 1975) and cannot prevent severe 



Fig. II 2. A schematic cross section of soils in the extremely 
arid zone. 




UTHOSOUC REG 



Vft PAlE BKPWN LOAM 
BEPOIV'-YmCV I. CAM 



LIMESTONE COARSE 

WITH CREVICES DESERT ALLUVIUM 



TTT 



r~T 



ryr 



TV* 



I T 



STONES ft 
GMVCl 
WITH SOMf 
FINE SOIL 
MATERIAL 




LIMESTONE 



LIME NODULES 



GRASSES & HERBS 




!jfl GRAVEL & 

"'^ >l STONES 



PRISMATIC OR 
BLOCKY STRUCTURE 



LOW SHRUBS 



LOOSE FINE SOIL 
MATERIAL 




ACACIA TREES 



Fig. II 3. A schematic cross section of soils in the arid semi-desert 
zone. 




LIMEY STONY SEROZfM 

MOO BPOW( LI 
MI inf ppfss IOAH^ 

TI^NS 



UGHT nUOwlSH- 
BROV- CP PAif 
PPOVS L0" OR 
SANPT IOAM 



BROWN LITHOSOL ALLUVIAL LOESS 
A 



LIT.HT Y[UPW| e .H- 
BPOWiH OR V(RY 
PAtf BP(>"< LOAM 
OR 5/>HtlY LOAM 




GRASSES &HERBS | ^ ^ ^ j LOW SHRUBS 



SUBANGULAR BIOCKY 
STRUCT OR MASSIVE 



ACACIA TREES 



14 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



T YELLOWISH- 
N L0**t WITH 
f CONCRETIONS 



erosion. In favorable ecological 
sites like footslopes or depressions, 
however, where the runoff water 
concentrates, the vegetation is 
more dense and favors deposition of 
the finely or eroded material 
(Yaalon and Dan, 1974)(Fig. II.3). 
This material consists mainly of 
loess that had been deposited on the 
hillslopes and was re-eroded due to 
the absence of a protective dense 
vegetation cover. Soil development 
is usually slow, due to the dry cli- 
mate, but is somewhat more 
marked than in the drier, extremely 
arid zone. 

The third zone is the mildly arid 
climatic zone, where the annual 
precipitation ranges from about 150 
or 200 mm to about 350 mm. This 
amount of rainfall usually enables 
the formation of a continuous vege- 
tational carpet (Fig. II.4), and most 
of these areas have been character- 
ized in the past by a low steppe. 
The vegetation restricts soil ero- 
sion; as a result, finely or weathered 
material covers the plains and moderate slopes. 
The protective cover of this vegetation also en- 
ables the heavy accumulation of Loess in this 
region (Yaalon and Dan, 1974), as it prevents 
re-erosion of this material. This region receives 
heavy loads of fine aeolian dust that has been 
eroded from the drier desert regions (Yaalon and 
Ganor, 1975). Soil profile differentiation proc- 
esses are significant, and mature, well differen- 
tiated soils are quite widespread. 

Physiographic Units 

Several broad physiographic units can be de- 
fined in the arid areas of Israel. Part of the land 
is mountainous, while other parts consist of 
plains, plateaus, or undulating areas (Fig. II.5). 
Large valleys, especially the large, young rift 
valleys, form another broad physiographic unit. 
Finally, the areas of sand dunes in the desert 
represent a separate unit, due to their specific 
features. 

The landscape, soil formation and pattern, 
and vegetation pattern, etc., are quite different 
in each of these physiographic units, even 
within the same climatic zone (Dan and Raz, 
1970). Erosional processes are severe in the 
mountainous areas, while depositional proc- 
esses characterize the young rift valleys. In the 
center of these valleys, saline marshes and sa- 



Fig. II 4. A schematic cross section of soils in the mildly arid part of 
the northern Negev. 




LOESSIAL LIGHT BROWN 
CLAY SOIL 



BROWN LITHOSOL ALLUVIAL LOESS 



LtOHT TLLCVISH 
BRPWN LOA* OP 
SANDY IOA* 



r~T 



171 



i I 



LIGHT YELICW1SH- 



LIMESTQNE 



LIME NODULES 



GRASSES & HERBS 







PRISMATIC OR 
BLOCKY STRUCTURE 



LOW SHRUBS 




LOOSE FINE SOIL 
MATERIAL 



SUBANGULAR BLOCKY 
STRUCT OR MASSIVE 



ACACIA TREES 



line lakes usually are found, because wadi chan- 
nels do not reach the sea. In the plains, both 
erosional and depositional processes are mini- 
mal; as a result, soil formation processes are 
more marked. In the sand areas, however, wind 
erosion and shifting sand inhibit soil formation. 
Each of the aforementioned physiographic 
units can be found in the three arid climatic 
zones. It was thus thought desirable to define 
climatic-physiographic units, each of which is 
characterized by a unique landscape, soil, and 
vegetation pattern (Dan, 1979a). In the follow- 
ing pages the soil formation of each of these cli- 
matic physiographic units is described. Table 
II. 1 provides a list of the soils of arid lands in 
Israel in relation to climate and physiography, 
and Table II.2 a description of the vegetation. 

Soil Formation in the Mountainous 

Part of the Extremely Arid Desert 

in Israel 

The weathering in the mountains is very slow, 
due to the dry climate. Shallow saline calcare- 
ous desert Lithosols 2 (Lithic Torriorthents) were 
formed on the soft carbonate rocks, like chalk 
and marl, due to the rapid mechanical weather- 
ing of these rocks. These Lithosols consist actu- 
ally of the physically broken down rock mate- 



DAN: SOIL FORMATION IN THE ARID REGIONS OP ISRAEL 



15 



Fig. II 5. Geographic and physigraphio 
lithologic regions of Israel. 




rial, so the chemical feature of these 
soils resembles that of the underlying 
bedrock (Dan and Raz, 1970). These 
Lithosols are very saline, and it 
seems that this salinity is higher 
than that of the underlying rocks. 

This is the result of some salt 
leaching of the uppermost soil layers 
and the concentration of the salts 
beneath the soil surface. The upper- 
most soil layer is later eroded, but 
some of the salts of this layer remain 
in the soil. 



The hard rocks on the slopes are usually bare of soil cover, 
due to their slow breakdown and the severe erosion. On pla- 
teaus some shallow Reg soils (Typic Gypsiorthids and Pet- 
rogypsic Gypsiorthids) were formed on hard rocks (Dan, 
1979b), a feature which characterizes mainly the areas 
where flinty strata are common. In such cases the flint 
gravel accumulates on the surface and protects the soil from 
accelerated erosion. Soil formation in these places is similar 
to the development of the Regs on the plain and is described 
in a later section. 

The erosion in the mountains is severe due to the climate, 
steep slopes, and absence of vegetational cover. As a result, 
most of the weathering products are carried away. The 
coarse alluvial material, which contains many stones and 
gravel (Typic Torriorthent or Typic Torrifluvent), is deposited 
on footslopes, in small valleys, in the streambeds, and along 
the alluvial fans which cover large areas in the valleys be- 
tween the mountains. 

This coarse material in the small valleys is usually young 
and unweathered. The soils that are found in these valleys 
are thus defined, according to the nature of the sediment, as 
coarse Desert Alluvium if they contain mainly stones and 
gravel, or as stony and gravelly sandy Desert Alluvium if 
they contain also large amounts of sand (Committee on Soil 
Classification in Israel, 1979). In some elevated spots like 
terraces or stable dissected alluvial fans and cones, some 
young Reg soils (Typic Gypsiorthids) may be found. The for- 
mation of these soils is similar to that in the plains and is de- 
scribed in a later section. 



2 Great group soil names according to the 
classification of Israel soils are written with 
a capital letter (Committee on Soil Classifi- 
cation in Israel, 1979). They are mostly cor- 
related with the U.S. Classification (Soil 
Survey Staff, 1975). 



Table II.l: Soils of Arid Lands in Israel in Relation to Climate and 
Physiography 


x^Climate 

PhysiographicX. 
unit \ 


Extremely arid 


Arid semi-desert 


Mildly arid 


Mountains 


Bare rocky slopes; 
Lithosolic Regs (on 
mountain plateaus). 
Saline calcareous desert 
Lithosols on chalks and 
marls; in gullies, coarse 
Desert Alluvium and 
stony and gravelly 
Desert Alluvium. 


Brown Lithosols, 
usually saline. Saline 
calcareous desert Litho- 
sols. In gullies, Loess 
and coarse Desert 
Alluvium. 


Brown Lithosols, Brown 
Rendzinas, Rendzinic desert 
Lithosols; various shallow 
Brown soils, some Natric 
Grumusols. In gullies, 
Loessial stony Brown soils 
and various colluvial- 
Alluvial soils. 


Great valleys 
(usually rift 
valleys) 


Coarse Desert Alluvium. 
Gravelly loamy Alluvium. 
Stony -sandy Alluvium 
and gravelly sand. 
Aeoliam and Alluvial 
sand. Fine desert 
Alluvial soils; Takyrs. 
Alluvial Solonchaks, 
Alluvial sterile 
Solonchaks. 


Various Serozems 
(including argil lie 
Serozems), mainly 
loessial and stony 
Serozems (in Negev 
mountains) or alluvial 
silly-clayey Serozems 
(in the Jordan Valley). 
Loess and stony alluviurr 
(in Negev mountains) 
and various calcareous 
Alluvial soils (Jordan 
Valley). 


Grumusols, highly calcareous 
Brown soils and brown silty 
Alluvial soils (in the 
Jordan Valley). Loessial 
and stony Brown soils and 
Alluvial Loess (in the 
Negev). 


Plains, plateaus 
and undulating 
areas 


Regs. In depressions, 
coarse Desert Alluvium 
and some Loamy Desert 
Alluvial soils. 


Stony Serozems; Alluvia 
Loess in depressions. 
Loessial Serozems in 
transition to the midly 
arid zone. 


Loessial light Brown soils 
and some other light Brown 
soils. Alluvial Loess in 
depressions. 


Sand plains 


Sand dunes, some Sand 
fields, some fine 
Desert Alluvial soils. 


Sand dunes, Sand fields 


Sandy Regosols, Sand fields, 
some sand dunes. 



16 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Soil Formation in 

the Large Desert 

Valleys 

Most of the eroded ma- 
terial from the mountains 
is deposited in the large 
desert valleys. The coarse 
stony and gravelly sedi- 
ments were deposited usu- 
ally along the alluvial 
fans, while the fine mate- 
rial - especially fine sand, 
silt, and clays - was car- 
ried farther away toward 
the depression or to the 
lakes and the sea (Dan, 
1979a, b). The soilforma- 
tion on the alluvial fans is 
usually negligible, while 
in the depressions it is 
usually more pronounced 
due to the special hydrol- 
ogy of these areas. The soil formation of these 
two geographic facets will be described sepa- 
rately. 



Soil Formation 

along Alluvial 

Fans 



Table II.2: Vegetation of Arid Lands in Israel in Relation to Climate and 
Physiography 


\Climate 
Physlographfev 
unit N 


Extremely arid 


Arid semi-desert 


Mildly arid 


Mountains 


Scattered low desert shrubs, mainly 
of various Zygophyllum dumosum 


Desert shrubs, mainly of various 
Artemisia herftg-albq plant 


Desert shrubs mainly of 
various Artemisia herba-alba 
plant associations; in many 
parts ateo S^rcopoterium 
spinosum plant associations 


plant associations on mountain 
on hard rocks. 
Low desert shrubs and some 
Acacia trees in rivulets. 


associations on mountain slopes; 
in dry parts ateo Zyoophyllum 
dumosum plant associations; 
quite dense shrub and grass 
vegetation at valley bottoms. 


Great valleys 
(usually rift 
valleys) 


Low desert shrubs mainly of 
/\n a b as i$ articulate! plant 


Scattered low desert shrubs, 
mainly of Hammada scoparia 
and /\riabas/i$ s,yrjaGjji plant 


Steppe vegetation of grasses 
and herbs. 


associations accompained 
by acacia trees in stream- 
beds; in sandy areas 
scattered Haloxylon 
pjBrsjcjun shrubs. Halphy- 
tic shrubs on Solonchaks. 


associations (in the Negev); 
Hatophytfc shrubs in 
Solonchaks (Jordan Valley). 


Plains, plateaus 
and undulating 
areas 


Low desert shrubs mainly of 
Anabasis articulate 


Scattered low desert shrubs, 
mainly of Hammada scoparia 
plant associations on plains. 
Quite dense shrub and grass 
vegetation in streambeds. 


Steppe vegetation of grasses 
and herbs 


associations accompained 
sometimes by acacia trees 
in streambeds; no vegetation 
on Reg soils. 


Sand plains 


Scattered desert shrubs, 
mainly of Haloxvlon persicum. 


Quite dense cover of desert 
shrubs, mainly of various 
Artemisia monosperma plant 


Steppe vegetation of grasses 
and herbs. 





and clay, has been carried farther on to the pla- 
yas and sometimes even to the sea. 

The composition of the alluvial fan sediments 
is related to the rock types which were found in 



Most of the coarse 
material which was 
eroded from the moun- 
tains that border the 
Arava Valley (the rift 
valley south of the 
Dead Sea) has been de- 
posited in the alluvial 
fans. These alluvial 
fans cover large parts 
of the valley, and espe- 
cially the southern 
parts (Ron, 1967). The 
coarse debris, which 
includes mainly large 
stones, has been depos- 
ited in the upper part of 
the fans, while the 
somewhat smaller- 
sized sediment, mainly 
gravel and sand, has 
been deposited in the 
lower part of the fans. 
The still finer sedi- 
ment, like fine sand 



Fig. n 6. A schematic cross section along an alluvial fan and nearby Solonchak 
depression that formed from material which had been eroded from carbonate 
mountains. 








GRAVEL 4 STONES 




DAN: SOIL FORMATION IN THE ARID REGIONS op ISRAEL 



17 



the catchment areas 
(Dan, 1979a,b). These 
include mainly various 
carbonate rocks and 
flint strata (Picard, 
1970). Such rocks do 
not form much sand 
during the mechanical 
weathering processes ; 
as a result, the sedi- 
ments along most of 
the alluvial fans in Is- 
rael range usually 
from coarse Desert Al- 
luvium (Typic Torri- 
orthents or Typic Tor- 
rifluvents) in the upper 
parts of the fans to 
gravelly loamy Allu- 
vium (Typic Torri- 
orthents or Typic Tor- 
rifluvents) in the low- 
est parts of these fans 
(Fig. II.6). 

In the mountain bor- 
dering the southern 
part of the Arava, 
sandstone and granite 
rocks are also quite widespread (Picard, 1970; 
Ron, 1967). In this case the fan sediments from 
these mountains contain much sand. As a re- 
sult, a sandy belt may be recognized in the lower 
part of the alluvial fan (Dan, 1979a,b). Thus, 
the sediments on these alluvial fans range from 
coarse Desert Alluvium (as above) in the upper- 
most parts of the fans, through stony sandy Al- 
luvium (Typic Torriorthents or Typic Torrifluve- 
nts) in the middle part, to gravelly sands (Typic 
Torripsamments) in the lower parts of the fans 
(Fig. II. 7). Some of the sand, in the lower part of 
the fan, may be picked up by wind, due to the 
absence of gravel pavement, and form some 
small dunes (Typic Torripsamments) along the 
lowest part of these fans. 

Sedimentation on the alluvial fans continues 
intermittently also today, and sediments do not 
reveal, as a rule, any soil formation (Dan and 
Raz, 1970; Dan, 1979b). In inactive parts of the 
fans, some very shallow Reg soil formation may 
be seen. 

Soil Formation in Undrained 
Depressions 

Most of the fine alluvial sediments, (Typic 
Torrifluvents) i.e., fine sand, silt, and clay, are 
deposited only when the carrying capacity of 



Fig. H 7. A schematic cross section along an alluvial fan and nearby Solonchak 
depression that formed from material which had been eroded from moun- 
tains where sandstone rock are widespread. 




floodwater is reduced to very low values. This 
occurs only in the depressions among the allu- 
vial fans or in very broad dry riverbeds with a 
negligible gradient, like Wadi el Arish. The 
water table in some of these depressions is found 
at a shallow depth. This feature characterizes 
mainly the depressions near the seacoast, like 
that of Sodom and Elat, or other depressions 
which were found at low elevation, like that of 
Yotvata and Avrona (Amiel and Friedman, 
1971; Dan, 1979a,b). 

Soil Formation in Wet Playas 

The groundwater in these depressions is sa- 
line. The salinity increases toward the center of 
the depressions. The soils in these depressions, 
especially their upper layers, are salinized due 
to capillary rise and evaporation of groundwater 
and occasional floodwater. In peripheral areas 
where the water table exceeds 2 meters, the sal- 
inization may be caused by deep-rooted plants 
that absorb the salts with the water from the 
deeper soil layers and return them to the soil 
surface via the leaves and the plant litter, as 
occurs with the Tamarix plant (Waisel et al., 
1978). As a result of this kind of salinization, 
some typical Solonchaks (Salorthid also some 
Typic Torrifluvents) soils were formed. 



18 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Fig. n 8. A schematic cross section revealing the relationship between soil profile characteristics 
and the various surfaces in the Central Arava Valley. 










A VCHT MU MOWM 

IAMOY LOAM 
MOCNIM-YtUDW 

LOAMY AMD 
C.Ct SAME At Ct 




"'.^-^ GRAvFl , STOHf S 

r-r--ivfsicuLAfl 

L_i._. -^ STRUCTURE 



I 1" 

I lu 



.OOSF SAND OR 
.OOSE SALINE LAYER 



SUBANGUIAR BIOCKY 



-_>". -.[STRATIFIED ALLUVIUM [u^u^| GYPSUM CRUST 

'--I CHALKY MARL 



DESERT mVEMENT 



Soil texture becomes finer toward the center 
of the depressions, as the finer particles are car- 
ried further away during the floods. As a result 
the soil grades from sandy and sandy loam allu- 
vial Solonchaks (Salorthids also Typic Tor- 
rifluvents) in the periphery of the depressions to 
loamy and clay loamy Solonchaks closer to its 
center. The center of the depressions is usually 
occupied by fine-textured alluvial Sterile Solon- 
chaks (Salorthids) which are bare of vegetation 
due to the high salinity of both the soils and the 
groundwater. 

In some places, where the water table is very 
high for most of the year, the various soil layers, 
and especially the deeper ones, suffer severely 
from impeded drainage. As a result, reduction 
processes occur and Gleys (Typic Salorthids) are 
formed. These Gleys are found usually in the 
central part of the depression, where fine-tex- 
tured gley Solonchaks, and especially fine-tex- 
tured sterile gley Solonchaks occur. However, 
gley Solonchaks (Typic Salorthids) may be found 
also in peripheral areas, between the alluvial 
fans where the land surface is much lower than 
in neighboring fan areas. 

Springs are usually present in these areas. 
The vegetation of these peripheral areas is fre- 
quently quite dense as the groundwater is not 
strongly salinized and, as a result, alluvial gley 
Solonchaks occur in these places. Some of these 
soils may also contain quite a high content of 
organic material due to the dense vegetation 
and the impeded drainage conditions; in these 
cases Aquollic Salorthids are recognized. 



Soil Formation in Dry Playas 

Soil formation is usually negligible in depres- 
sions where the water table is deep (Dan, 
1979b). The same may hold true for the soils of 
Wadi el Arish and its tributaries. Thus, these 
soils have characteristics resembling those of 
the alluvial sediment and were designated as 
fine Desert Alluvial soils (Committee on Soil 
Classification in Israel, 1979) (Typic Torri- 
orthent or Typic Torrifluvents). However, in 
some of these depressions, like Qa en Naqb, the 
soils have been severely salinized due to the 
concentration of salts in the ponded floodwater. 
As a result, the sodium concentration in the ex- 
change complex reached high values. The 
ponded floodwater had, to some extent, leached 
the uppermost soil layer and, as a result, a hard 
sodic crust was formed in these soils. These soils 
have been classified as Takyrs, like similar soils 
in Central Asia (Kovda, 1973). 

Soil Formation on the Gravel Plains 

The gravel plains were formed from old desert 
valley fill which was composed mainly of coarse 
alluvium (Horowitz, 1979). The sediments were 
subjected to severe water and wind erosion that 
carried away all the fine material from the sur- 
face; as a result, a cover of gravel and stones re- 
sistant to weathering was gradually formed on 
the soil surface (Evenari et al., 1971; Dan, 
1979b). 

Erosion decreases to very low values after for- 
mation of the gravel cover, due to the protective 



DAN: SOIL FORMATION IN THE ARID REGIONS OP ISRAEL 



19 



feature of the gravel and stones. 
As a result, soil profile differentia- 
tion and soil formation processes 
affect these soils. These are very 
slow, due to the dry climate; they 
include mainly atmospheric salini- 
zation and concentration of gyp- 
sum from airborne salts (Yaalon, 
1963), but some clay formation or 
introduction can also be recog- 
nized. The soils which were 
formed due to the above-men- 
tioned combination of processes 
are defined as Regs (usually Gypsi- 
orthids). 

Stages of Reg Soil Formation 

Several stages of Reg soil forma- 
tion can be recognized. On quite 
young sedimentary surfaces, like 
the Lisan surface near Hazeva and 
inactive parts of alluvial fans, soil 
development is restricted mainly 
to the formation of a thin, 1-2 cm, 
vesicular layer and an underlying 
thin, somewhat reddish layer (Fig. 
IL8). Some salt and gypsum con- 
centration in the deeper layers 
characterizes these soils (Fig. II.9). 
This soil will be still included 
among the Torriorthents. 

The various soil layers become 
thicker in the older geographic sur- 
faces (see Fig. II. 8). This develop- 
ment is accompanied by severe sal- 
inization and concentration of gyp- 
sum (see Fig. II.9) The gypsum 
usually forms several hard pet- 
rogypsic horizons in the deeper soil 
layers. Quite deep Reg soils with 
several petrogypsic horizons (Petrogypsic Gypsi- 
orthids) characterize most of the large gravel 
plains in the Paran Desert and central Sinai. 

The Typical Reg Soil Profile 

The typical mature Reg soil (Petrogypsic Gyp- 
siorthid) consists of several horizons (Dan et al., 
1982; Dan, 1979a). Underneath the desert 
pavement a very pale brown loamy vesicular 
horizon of several centimeters is found. The 
next horizon, of about 10 to 20 cm, consists usu- 
ally of a reddish-yellow dusty and very saline 
loam or clay loam. Soft gypsum or anhydrite 
chunks are usually found in this horizon. 



Fig. II 9. Salinity and gypsum distribution in Reg profiles from various 
surfaces. 

electrical conductivity ( mmho/ 100 g soil) 
o so 100 



OLD REG ( ARAVA CONGLOMERATE SURFACE', 
MATURE REG (MID PLEISTOCENE SURFACE) 

YOUNG REG (LISAN SURFACE) 
COARSE DESERT ALLUVIUM (RECENT SURFACE) 




gypsum 




/ OLD REG (ARAVA CONGLOMERATE SURFACI. 

MATURE REG (MID PIEISTOCFNI SURFACE) 

YOUNG RLG (LISAN SURFACE) 



The deeper layers usually consist of a some- 
what mechanically weathered material. The 
fresh parent material, consisting of a mixture of 
gravel and stones with some finer materials in 
regosolic Regs, or of various carbonate rocks in 
lithosolic Regs, is found usually at a depth of 
about 60 to 90 cm. Hard gypsum crusts - pet- 
rogypsic horizons - are found usually in the 
deeper layers of these soils. The Reg soils are 
bare of vegetation. 

Other Soils in the Reg Plains 

Some of the Reg plains are dissected due to 
the drop of the base level. The gravel and stones 



20 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



that are found on the dissected slopes do not 
reveal any soil differentiation and they may 
thus be defined as gravelly or stony Regosols 
(Typic Torriorthents). (Committee on Soil Clas- 
sification in Israel, 1979). 

The gravel plains are transversed by many 
dry water courses (wadis). Desert alluvium, 
especially coarse desert alluvium (Typic Torri- 
orthents or Torrifluvents), has been deposited in 
these water courses (Dan, 1979a,b). The compo- 
sition of this alluvium depends on the carrying 
capacity of the floodwater; in the large, main 
wadi the floods are, as a rule, heavy, and the 
streamflow is rapid, so that only coarse materi- 
als, like stones, are left behind. In the smaller 
wadis the stream is generally much slower, and 
as a result gravel and even sand and silt are 
deposited. 

The streamflow in the central part of Wadi el 
Arish and its tributaries is generally very slow 
due to the very small gradient, and fine sedi- 
ments - especially fine sand and silt - have been 
deposited in these wadis. A great part of this 
fine sand and silt, as well as similar material 
that was left on the gravelly wadi beds at the 
end of the flood, are carried away by winds and 
form the source of loess deposits (Yaalon, 1978). 

Soil Formation in Sandy Deserts of 
Israel and Northern Sinai 

Soil formation in sandy deserts of northern 
Sinai and the western Negev is negligible. The 
shifting sand is not, as a rule, stable enough for 
soil horizons to be formed. Even in sand fields 
(Typic Torripsamments) where the sand is more 
stable, no soil horizons are detected. 

Most of the sandy region is covered by parallel 
shifting seif dunes (Tsoar, 1978). The seif dunes 
are oriented east- west as a result of the bi-direc- 
tional seasonal winds. 

Depressions between the dunes sometimes 
reach a width of 100 meters or more. The some- 
what coarser lag sand in these depressions is 
usually quite stable and often covered by a 
rather dense vegetation, especially in the north- 
ern and eastern parts of the region. However, 
soil formation is still negligible (Dan, 1979a,b). 

Several dry wadis reach the dune area from 
the south. Most of these water courses get lost 
in the dunes, since they do not carry enough 
floodwater to force their way across the seif 
dune barriers. The fine sandy and silty sedi- 
ments carried by the floodwater are deposited 
between the dunes, which is why these valleys 
are covered by fine Desert Alluvial soils (Typic 



Torrifluvents). These soils are usually covered 
by some sand, and in many of them sandy layers 
are found among the finer deposits. The only 
dry rivers that now cross the dunes are Nahal 
haBesor and Wadi el Arish. Fine Desert Allu- 
vial soils are also found here along the river in 
the recently formed flood plains. 

The altitude of the sand dunes region ranges 
from sea level to about 300 m along the south- 
ern border of the region. As a result of the low 
altitude in the northwestern part of the region, 
groundwater is found close to the surface in the 
interdune areas (Dan, 1979a). This groundwa- 
ter flows slowly from the south, northward to- 
ward the sea. The groundwater becomes highly 
salinized due to strong evapotranspiration, and 
Solonchaks (Salorthids) are formed as the saline 
groundwater rises to the surface. These Solon- 
chaks cover large areas, especially near the sea 
and the lagoons. 

In the southern part of the farther inland sa- 
line depressions where the groundwater is not 
very saline, there is usually some vegetation. 
Many date palms are grown in these places. 
Toward the sea coast, the vegetation gradually 
disappears, as a consequence of the high salinity 
of the groundwater; as a result, typical sterile 
Solonchaks (Salorthids) cover these areas. 

Soil Formation in the Arid Semi- 
desert Parts of the Mountains in 
Israel (in the central Negev and the 
lower parts of the Judean Desert) 

The weathering of the hard rocks in the 
Negev mountains is very slow, due to the dry 
climate. Thus, the soils of these mountains were 
formed mainly from the loessial deposits that 
have been mixed, to some extent, with the me- 
chanical breakdown of the rocks. 

These soils are confined to pockets and crev- 
ices among the rocks where they are protected 
from accelerated erosion. They are designated 
as Brown Lithosols (Lithic Torriorthents). Most 
of these soils are saline, due to the concentration 
of airborne salts, and, as a result, are defined as 
saline Brown Lithosols. However, these Litho- 
sols are usually less saline than the deep soils in 
this region, as the rock pockets and crevices re- 
ceive some runoff water from the rock expo- 
sures. 

This concentration of local runoff water also 
enables the development of relatively rich vege- 
tation (Danin et al., 1975). In very rocky areas 
where rock exposures cover most of the surface, 
the amount of local runoff increases, and in 



DAN: SOIL FORMATION IN THE AKID REGIONS OP ISRAEL 



21 



i ico 
a 

0) 
T3 



130- 



some cases it may even leach the salts; 
thus, in these places, nonsaline Brown 
Lithosols can be found. 

The mechanical breakdown of the soft 
rocks, like chalk or marl, is somewhat 
faster and, as a result, the soils resemble 
the composition of the underlying rocks. 
These soils are included among the cal- 
careous desert Lithosols (Lithic Torri- 
orthents). However, at somewhat moister 
sites, like northfacing slopes, especially in 
the moister part of the region, these soils 
may reveal some formation of an A hori- 
zon (Dan and Nissim, 1976). This is due 
to denser vegetation and, as a result, 
there is also some formation of organic 
material, which may form also a some- 
what better structure. These soils have 
been defined as Rendzinic desert Litho- 
sols (Typic or Lithic Torriorthents) due to 
their resemblance to the pale Rendzinas 
which are found in the moister parts of 
Israel. 

The soils on the soft chalk and marls 
are more saline than those on hard rocks. 
In these places rock outcrops are usually 
absent so that the soil does not receive 
any local runoff water that can leach the soil. 
Moreover, runoff of the soil surface itself is se- 
vere, especially in the desert Lithosols, due to 
the absence of vegetation cover (Evenari et al., 
1971) and the weak structure of the upper soil 
layer. As a result, these soils are severely sal- 
inized. 

Soils of the Valleys and Plains in 

the Arid Semi-desert of the Negev 

and The Judean Desert 

The mountains in the semi-desert areas are 
severely eroded, as no dense protective vegeta- 
tion cover exists in this area (Yaalon and Dan, 
1974). The eroded sediments contain, as a rule, 
loamy loessial materials which are sometimes 
mixed with gravel or stones. The gravelly and 
stony sediments are deposited usually along the 
footslopes and in the large dry wadibeds, while 
the loessial, stone-free sediments are depoisted 
along the small depressions and in the 
flooplains. 

Soil Development from Loess 

These young loessial deposits do not reveal 
any profile differentiation, as their deposition is 
rapid (Dan, 1979a); they are designated as 
Loess or alluvial Loess (Torrifluvents or Xer- 



Fig. II 10. Salinity content of an Alluvial Loess, a Loessial 
Serozem, and a Loessial Argillic Serozem at Sede Boger. 
The differences in salinity are related mainly to soil age. 

electrical conductivity ( mmho/ 100 g soil) 
o to 30 30 *o 30 



so-: 

! 

i 
I 
i 




ARGIIUC LOESSIAL 
SEROZEM 

LOESSIAL SEROZEM 

ALLUVIAL LOESS 



ofluvents, according to the moisture regime). 
On the terraces, however, soil formation may 
already be revealed. Redeposition of lime at a 
shallow depth and slow salinization of the 
deeper soil layers caused by airborne salts char- 
acterize these soils (Dan, 1979a, b) (Fig. II. 10). 
Some clay illuviation may also be revealed. 

These soils are defined as loessial Serozems. 
The B horizon of the young loessial Serozems is 
of a loamy cambic nature and it is designated as 
a loamy loessial Serozem (correlated with the 
Calciorthids) (Committee of Soil Classification 
in Isreal, 1979). With increasing development, 
usually in the higher terraces, the B horizon be- 
comes somewhat finer textured and clay loam 
argillic loessial Serozems (correlated with the 
Typic Haplargids) are formed. Loessial Seroz- 
ems also have been formed from aeolian loess. 
In the Negev mountains they are confined to flat 
palteaus where aeolian loess had been deposited 
during a presumably somewhat mositer period 
in the past (Dan et al., 1979). 

Toward the northern edge of this area, aeo- 
lian loess has been deposited also on plateaus 
and northern slopes, due to somewhat more fa- 
vorable ecological conditions (Yaalon and Dan, 
1974). Loessial Serozems also have been devel- 
oped from these deposits (Dan, 1979b), (Fig. 



22 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



SALINE LIGHT 
YfLLWlSH-BROWH 
SILT LOAH WITH 
Ll CONCRETIONS 
AND GYPSUM 
CRYSTALS AT 
DfPTH 



11.11). These soils are widespread in 
the Be'er Sheva' valley and the Arad 
plateau. The salinity of these Seroz- 
ems decreases gradually toward the 
moister areas in the north until they 
merge with the non-saline loessial 
light Brown soils (Calcic Haploxer- 
alfs, Calcic Palexeralfs or Calcixerol- 
lic Xerochrepts). 

Soil Development from Gravelly 
and Stony Deposits 

Soil formation on the gravelly and 
stony deposits resembles that on the 
loess. Young sediments in the dry 
riverbeds do not reveal any soil for- 
mation and are thus designated as 
coarse Alluvium (Torrifluvents or 
Xerofluvents). On terraces, 

footslopes, and inactive fans, clacic 
and cambic horizons develop and the 
soils are designated as gravelly and 
stony Serozems (Calciorthids) Fig. 
11.12). With increasing time an ar- 
gillic horizon is sometimes formed 
(Argids according to the American 
classification). 

Carbonate content in the calcic 
horizon of the gravelly and stony soil increases 
continuously with time until the lime coatings 
and nodules merge to form a petrocalcic horizon 
(Dan, 1977), and thus petrocalcic stony Seroz- 
ems or petrocalcic Paleargids are formed. Old 
Neogene gravelly and stony sediments are wide- 
spread in the eastern Negev on high dissected 
terraces (Horowitz, 1979). Various stony Seroz- 
ems, especially petrocalcic stony Serozems, 
cover most of these Neogene sediments. 

Soil Development from Sand 

A similar sequence of soil development can be 
seen in the sandy sediments that are wide- 
spread in some synclinal valleys in the eastern 
Negev and in the western part of the Negev. 
The primary stages of soil development include 
Sand fields and sandy Regosols (Typic Torrip- 
samments). These soils are still widespread in 
the western Negev, where such areas have been 
covered by sand during the last few millenia. 
With the passage of time, sandy Serozems were 
formed. 

The primary stage is characterized by a sandy 
A horizon and a shallow sandy cambic Bca hori- 
zon. This soil is designated as sandy quartzic 
Serozem (Camborthids or Calciorthids) (Corn- 



Pig. H 11. A schematic cross section of soils in the northern, 
moister part of the arid, semi-desert zone. 




LOESSIAL SEROZEM 



BROWN LITHOSOL ' ALLUVIAL LOESS 



LIGHT YELLOW I SM- 

BPOWN LOAH Oft 

SANDY LOAM 



BROUN fW VEPY 
PALE IftOWK LOAM 
Of SAMOT LOAM 




LIMESTONE 




PRISMATIC OR 
BIOCKY STRUCTURE 1 



LOOSE FINE SOIL 
MATERIAL 



SUBANGULAR BLOCKY 
STRUCT. OR MASSIVE 



LOW SHRUBS 






ACACIA 



zon is formed and the soil designated as loamy 
quartzic argillic Serozem (Typic Haplargids). 
The concentration of lime in the Bca horizon 
continues here, as in the stony Serozems, until a 
petrocalcic horizon is formed (Dan, 1977), and 
the soils are designated as petrocalcic quartzic 
Serozems (petrocalcic Paleargids). 

These advanced stages of development are 
confined to sites of Neogene sand which are usu- 
ally mixed with some stones, in the vicinity of 
Arad and Aroer. The carbonates of these soils 
originated mainly from airborne sediments 
(Dan, 1977), as the quartzic parent material 
does not usually contain carbonates. 

Soil Formation in the Southern 
Arid Part of the Jordan Valley 

The soils of this region have been developed 
either from alluvial material - mainly along the 
alluvial fans - or from marly-lake deposits (Li- 
san marl) (Dan and Alperovitch, 1971, Dan et 
al., 1981). 

The Soils of the Alluvial Fans 

The large alluvial fans deposited clayey and 
silty soil material that had been eroded from 
soils of the more humid mountains. Various al- 



mittee on Soil Classification in Israel, 1979). i i T -IT. j i ^ r ^- 

With increasing time a loamy argillic Bca hori- luvial Brown S0lls have devel P ed from thls 



DAN: SOIL FORMATION IN THE ARID REGIONS OP ISRAEL 



23 



Fig. II 12. Soil development from gravel in the arid semidesert zone and in the mildly arid 
part of the country (according to Dan, 1977). 



STONY ALLUVIUM 
OR STONY COLLUVIUM 



STONY LIGHT BROWN 
SOIL OR STONY 
SEROZEM 



STONY PETROCALCIC LIGHT 
BROWN SOIL OR STONY 
PETROCALCIC SEROZEM 





GRAVEL AND STONES 



LOAMY SOIL 





STONES WITH LIME COATING 



PETROCALCIC HORIZON 



material. Stony and gravelly alluvial Brown 
soils (Calcixerollic Xerochrepts) are found 
mainly on the upper parts of the fans; the soils 
become more fine-textured and more saline to- 
ward the lower parts of the fans and saline allu- 
vial silty clayey Serozems (Typic Haplargids, 
some Vertic Haplargids) are formed. 

It seems that these soils were salinized in the 
past from a high water table associated with the 
Lisan lake (the ancient Dead Sea). Subse- 
quently, the soils were drained due to the dissec- 
tion of the Lisan terrace, and nowadays they are 
affected mainly by the present dry-climate con- 
ditions. Some small alluvial fans that deposited 
highly calcareous material from the nearby des- 
ert are found among the large alluvial fans. 
Most of the soils in these fans do not reveal any 
profile differentiation and thus they are desig- 
nated as highly calcareous pale brown Alluvial 
soils (Typic Torrifluvents or Typic Torri- 
orthents). 

The water table in the alluvial fans is deep, 
but it becomes shallow toward the lower part of 
these fans. The water is also very shallow in 
narrow strips between the fans; there ground- 
water caused severe salinization and as a result 
Solonchaks (Salorthids) are found along the 
strips. It seems that in the past, before the dis- 



section of the Lisan terrace, or even during the 
existence of the Lisan lake, the extent of the 
areas which suffered from a high water table 
was larger, and it is possible that the saline allu- 
vial silty-clayey Serozems received their salts 
from this source. 

Soil Formation from the Lacustrine Marl 

The soils of the central part of the valley have 
been formed from marly-lake deposits (Dan and 
Alperovitch, 1971). These soils have an AC hori- 
zon sequence and are highly calcareous and sa- 
line. A gypsic horizon is found as a rule at shal- 
low depths. These soils were defined as saline 
gypsiferous highly calcareous Serozems. (Typic 
Torriorthids, sometimes Gypsiorthids or Calci- 
orthids). 

The marly terrace was dissected by the Jor- 
dan River and its tributaries, due to the gradual 
drop of the base level of the Dead Sea that 
started about 12,000 years ago. As a result, 
badlands were formed in the transitional area 
between the terrace and the recent floodplain. 
The soils on these badlands are very shallow 
and include marly saline desert Lithosols (Typic 
Torriorthents). In the depression between the 
badlands, the drainage conditions are very poor 
and the soils are severly salinized. As a result, 



24 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



various Solonchaks (Salorthids), both from Li- 
san marls and alluvial materials, are found in 
these areas. 

Soils of the Jordan Floodplain 

The soils in the recent Jordan floodplain are 
being formed from alluvial material of the Jor- 
dan (Dan and Alperovitch, 1971). Here are 
found various brown Alluvial soils (Torrifluve- 
nts or Xerofluvents) in well-drained positions 
and alluvial Solonchaks (Salorthids) in poorly 
drained areas. It should be pointed out that the 
drainage condition becomes gradually poorer 
toward the Dead Sea, and as a result the Solon- 
chaks cover the whole floodplain in the lowest 
part of this region. In this area, even sterile al- 
luvial Solonchaks that have been formed by ex- 
treme salinization are found. 

Soil Formation in the Mildly Arid 

Parts of the Desert Fringe Areas in 

the Mountains of Judea and 

Samaria 

Soils of Mountainslopes 

The primary stages of soil formation usually 
resemble those of the more arid mountain areas, 
especially in the southern parts of this region. 
Brown Lithosols (Lithic Torriorthents) are 
formed in pockets among hard rocks, while 
Rendzinic desert Lithosols (Lithic or Typic Tor- 
riorthents) are found on chalks and marls. 
However, the soils are somewhat more devel- 
oped and usually also more leached. 

The Brown Lithosols in this region are non- 
saline and contain usually somewhat more or- 
ganic material than in the drier regions. In the 
moister parts of this region these Lithosols 
merge with Brown Rendzinas (Lithic Xero- 
chrepts or Lithic Haploxerolls) that are darker 
colored and contain more organic material. In 
the northern parts of the Judean Desert and 
especially in the Samarian Desert, the soils on 
hard rocks are usually fine textured, as the aeo- 
lian dust that reached this area contained more 
clay. As a result, clayey Brown Rendzinas are 
usually formed in this region. 

The leaching in the Samarian Desert is more 
pronounced because the rate of dust deposition 
that reaches this area is much less than that in 
the Negev (Yaalon and Ganor, 1975). As a re- 
sult these soils are less calcareous, and in favor- 
able spots all the carbonates have been leached 
and Terra rossa soils (Lithic Xerochrepts) 
formed (Zeidenberg and Dan, 1979; Litaur, 



1980). 

On moderate slopes, where the erosion is not 
severe, the aeolian dust cover gradually be- 
comes thicker and loessial light Brown soils 
(Typic Haploxeralfs or Calcic Palexeralfs) are 
formed in the northern Negev and the southern 
part of the Judean Desert (Dan and Raz, 1970). 
Transitional soils are found on somewhat 
steeper slopes. The loessial light Brown soils 
are especially widespread on northern slopes, 
where erosion is less severe due to the denser 
vegetational cover (Yaalon and Dan, 1974). 

Soils of Lower Slopes, Valleys, and 
Depressions 

On the lower parts of the slopes and on 
footslope positions, loessial soil material that is 
mixed with gravel and stones has been depos- 
ited. Soil formation in these places is expressed 
mainly by formation of a cambic and, later on, 
even an argillic B and enrichment of lime at a 
depth of about 50-100 cm. This lime forms a 
typical calcic horizon. The lime concentration in 
this layer increases gradually until petrocalcic 
horizon is formed (see Fig. 11.12). 

In the northern part of the Judean Desert and 
in the Samarian Desert, the deep soils that form 
on moderate slopes, plateaus, and depressions 
are finer textured, due to the clayey nature of 
the aeolian dust in this region (Yaalon and Dan, 
1974; Dan and Alperovitch, 1975). Natric 
grumic Serozems (correlated with typic Natrar- 
gids) are found in the drier parts of this region, 
while toward the moister parts these soils merge 
with Natric Grumusols and finally even with 
typically calcareous reddish Brown Grumusols 
(or typic Chromoxererts). 

The leaching of these soils is somewhat inhib- 
ited due to the fine texture, and as a result some 
of the salt and large amounts of abasorbed so- 
dium remain in the deeper soil layers. The sa- 
line soils are designated as Serozems and, due to 
the high absorbed sodium, they are defined as 
Natric Serozems (Committee on Soil Classifica- 
tion in Israel, 1979). The leaching increases 
gradually toward the moister regions and as a 
result the salts and farther on also the adsorbed 
sodium are leached into the deeper soil layers. 

The difference between the Grumic soils and 
the Grumusols stems from differences in the 
siltrclay ratio of these soils. The proportion of 
silt in the dust that reaches the soils which are 
nearer the desert is relatively higher (Yaalon 
and Ganor, 1975) and, as a result, pedoturba- 
tion is inhibited to some extent and natric 



DAN: SOIL FORMATION IN THE ARID REGIONS OP ISRAEL 



25 



Fig. n 13. Soil development from sand in the mildly arid part of the Northern Negev (according to 
Dan, 1980). 



SANDY QUARTZIC INSEPTIC QUARTZIC 

SAND REGOSOL LIGHT BROWN SAND LIGHT BROWN LOAM 



LOESSIAL LIGHT BROWN 
CLAY LOAM 



.?JJ J | ij* J A^\^ i J..1 < J , -1 * 



;: :V:^-:*^pPi*>>J^^^^^ 

:-\:V:--?;i?;MS^iiSH^ 



MASSIVE OR 
SUBANGULAR BLOCKY 

STRUCTURE : 
LOAMY TEXTURE 



BLOCKY OR PRISMATIC 

STRUCTURE 
CLAY LOAM TEXTURE 




LIME CONCRETIONS 



Grumic Serozems were developed (Zeidenberg 
and Dan, 1978). The clay content increases with 
increasing distance from the desert, and as a 
result Grumusols (Typic Chromoxererts) were 
formed in these areas (Yaalon and Dan, 1974; 
Dan and Alperovitch, 1975). 

Soil Formation in the Mildly Arid 
Plains of the Northern Negev 

The primary topography of this region is gov- 
erned mainly by ancient dune ridges and grav- 
elly coastal sediments. These sediments were 
affected by the slow encroachment of aeolian 
loess. 

The youngest soils have been developed 
mainly from the dune sand; these include sandy 
Regosols (Quartzipsamments) and quartzic in- 
septic light Brown sand (also Quartzipsam- 
ments merging with Calcixerollic Xerochrepts) 
which is chracterized already by a Bca horizon 
(Fig. 11.13). With continuous development, the 
soil and especially the Bca horizon become finer 
textured and loamy quartzic light Brown soil 
(Calcic Haploxeralfs) are formed (Dan and 
Yaalon, 1980). 

These soils are characterized already by an 
argillic Bca horizon. Loess acretion continues 
until the sandy sediments are completely cov- 
ered by the loess (Dan and Yaalon, 1971, 1980). 



Soil formation continues simultaneously, so that 
a typical cumulic loessial soil has been formed. 
This soil is characterized by an ochric loamy A 
horizon and an argillic, clay loamy B horizon. 
This soil is designated as loessial light Brown 
clay loam (correlated with calcic Haploxeralfs). 

Loess deposition affected this area through- 
out several hundred thousand years. As a re- 
sult, a column of about 8-12 meters of clays or 
silty clays with up to six paleosols, underlies the 
recent soil (Dan and Yaalon, 1971; Bruins, 
1976). Rainfall in this area is high enough to 
leach most of the soluble salts from the soil pro- 
file, but some high ESP values are still found in 
the deeper soil layers, indicating restricted 
leaching. 

In depressions and floodplains, fluvial rede- 
position of the loess occurs (Dan et al., 1976; 
Dan and Yaalon, 1980). Soils are as a rule 
young, without profile differentiation, and are 
designated as Loess or alluvial Loess (correlated 
with typic Xerofluvents). In recent to late Pleis- 
tocene terraces, or on footslopes, a cambic loamy 
Bca horizon has been formed, and the soils are 
defined as loessial inseptic light Brown loams 
(correlated with typic or calcixerollic Xero- 
chrepts). 



26 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Soil Formation in the Sandy Areas 
of the Western Negev 

The western Negev has been covered recently 
by aeolian sand that reached the region from the 
drier area in the south (Dan 1979a,b). These 
sands covered older soils, mainly quartzic light 
Brown soils and Loessial light Brown soils (cal- 
cic Haploxeralfs). The deposition of this sand 
has been quite gradual, and therefore the area 
has been covered by quite a dense vegetation. 
The ecological conditions in these sandy areas 
are better than those of the loessial areas far- 
ther to the east, due to the absence of runoff and 
the low waterholding capacity of the sand. 
Penetration of the rainwater in this area is quite 
deep, in a normal year reaching a depth of about 
1 meter or even more. 

Soil development is negligible due to the short 
time that passed from the time of deposition, 
and as a result the soils were defined as sandy 
Regosols (Torripsamments or Quartzipsam- 
ments). However, in several places some lime 
segregation has already been detected, and it 
seems that sandy quartzic inseptic light Brown 
soils (Calcixerollic Xerochrepts) are slowly de- 
veloping in these areas. 

Shifting sand dunes are found along a strip 
several kilometers wide, near the coastline. The 
sand dunes are usually bare of vegetation, but 
in depressions between the dunes some vegeta- 
tion has been found. These depressions are es- 
pecially noticeable near the seacoast, where 
their elevation reaches only several meters 
above sea level and deep-rooted plants can reach 
the groundwater. 

A typical hydromorphic vegetation exists in 
some of these areas. Bedouins took advantage 
of this feature and removed sand from small- 
plots to an elevation of about 1 meter above the 
water table. On these areas they planted vari- 
ous orchards and vegetables, so that the roots 
reach the groundwater. The sand removed is 
piled high on the sides of these plots, thus giving 
the appearance of sunken gardens. 

In conclusion, it should be stressed again that 
all the sands in this area include only recent 
sands. In the future these sands will probably 
form the typical light Brown soils that charac- 
terize this climatic zone, as quite dense vegeta- 
tion usually covers the sand after some time, 
and the loess deposition may be protected by 
this vegetation. A pronounced difference thus 
exists between this zone and the sandy areas in 
the more arid parts, where the vegetation is too 
sparse to stop the movement of the sand and 



where, as a result, soil development will not be 
marked even after a long period. 

Soil Formation in the Mildly Arid 

Part of the Jordan Valley (Bet 
Sfae'an Valley and Biqat Kinarot) 

As in the lower parts of the Jordan Valley, the 
soils have been formed mainly either from allu- 
vial material of the surroundidng mountains or 
from the marly lake deposits (Lisan marl). 

Soil Formation from Clayey Alluvial 
Sediments 

The mountains on both sides of the valley are 
found in the semi-arid and subhumid climatic 
regions (Rosenan, 1970), and the soils of these 
regions consist of typical fine-textured clayey 
Terra rossa, Brown Rendzinas, Protogrumusols, 
(mainly Lithic Xerochrepts) (Chromoxererts and 
Grumusols) (Dan and Raz, 1970; Dan et al., 
1976). As a result, the alluvial material that 
reaches the valley consists also mainly of 
montmorillonite clays, and various Grumusols 
(Vertisols) have been formed from these sedi- 
ments. 

Brown and Reddish brown Grumusols (Chro- 
moxererts) were found in the well-drained ar- 
eas. Many of these Brown Grumusols suffer 
from high ESP values in the deeper soil layers, 
as rainfall is not sufficient to leach the sodium 
from the fine-textured soils. These soils are 
thus defined as Brown and Reddish brown 
natric Grumusols (Committee on Soil Classifica- 
tion in Israel, 1979). The natric Grumusols are 
especially widespread in the drier parts of this 
region, such as in the Bet She'an Valley. 

Various Hydromorphic Grumusols (correlated 
with the Pelloxererts) are also widespread. 
They are found mainly near the springs and the 
formerly swampy areas. Large areas of hydro- 
morphic Grumusols are found on the transi- 
tional strip between the Grumusols and the soils 
that were formed from the Lisan marl. The 
groundwater in this transitional area flowed on 
the Lisan marl sediments, and as a result the 
soils suffered from impeded drainage conditions. 

Many of these soils contain large amounts of 
carbonate, as they were mixed to some extent 
with the underlying highly calcareous sedi- 
ments; these soils are defined as highly calcare- 
ous hydromorphic Grumusols. In some of the 
soils in this transition zone the hydromorphic 
features are no longer evident, as they are now 
well drained. These soils are defined as highly 



DAN: SOIL FORMATION IN THE ARID REGIONS OP ISRAEL 



27 



calcareous marly Grumusols (Committee on Soil 
Classification in Israel, 1979). 

Hydromorphic natric Grumusols are some- 
times found in swampy areas. These soils were 
formed mainly in places where the groundwater 
contained considerable amounts of soluble salts 
(Alperovitch and Dan, 1972). 

On the upper parts of the alluvial fans, where 
the parent material contains large amounts of 
stones and gravel, various Colluvial-alluvial 
soils (mainly Xerofluvents) are found. In the 
lower parts they grade with the Brown and Red- 
dish brown Grumusols. On the transitional 
parts where the amount of gravel and stones is 
intermediate, gravelly Brown and Reddish 
brown Grumusols are formed. 

Soil Formation from the Lacustrine Lisan 
Marl 

Highly calcareous inseptic Brown soils (in- 
cluded among the calcixerollic Xerochrepts) 
have developed from the Lisan marl sediments. 
These soils are characterized by a CA horizon at 
a depth of about 1/2 to 1 meter. They differ from 
the saline gypsiferous Serozems of the lower 
part of the valley by the absence of soluble salts, 
due to the relatively more intensive leaching 
(Committee on Soil Classification in Israel, 
1979). Hydromorphic highly calcareous inseptic 
Brown soils were formed in formerly impeded 
drainage areas that were found among the 
highly calcareous inseptic Brown soils. Some of 
these soils contained, in the past, considerable 
amounts of soluble salts, but most of them have 
recently been removed, due to the construction 
of drainage systems that resulted in their being 
leached by irrigation water. 

Pale Rendzinas (Lithic Xerorthents) and 
Rendzinic desert Lithosols (Lithic and typic Tor- 
riorthents) are found on the terrace scarp be- 
tween the Lisan terrace and the recently formed 
Jordan floodplain. These soils are very young 
and do not reveal any diagnostic horizon except 
for an ochric A horizon, and in the Pale 
Rendzinas sometimes even the beginning of a 
mollic epipedon. 

Soils of the Jordan Floodplain 

The soils of the Jordan floodplain have been 
formed from recent alluvial material. They are 
very young and are still characterized by allu- 
vial layering. The texture of these soils grades 
mostly from loams and silt loams to silty clay 
loams. These soils are calcareous but usually 
not saline and thus they were included among 



the brown Alluvial soils (Typic Xerrofluvents) 
(Committee on Soil Classification in Israel, 
1979). 

References 

Alperovitch, N. and Dan, J. (1972) Sodium affected soils in 
the Jordan Valley. Geoderma 8: 37-57. 

Amiel, A.J. and Friedman, G.F (1971) Continental 
Sabkha in the Arava Valley between the Dean Sea and 
the Red Sea. Bull. Am. Ass. Petrol. Geol. 55: 581-592. 

Bruins, H. J. (1976) The Origin, Nature and Strategraphy 
of Paleosols in the Loessial Deposits of the N.W. Negev 
(Netivot, Israel). M. Sc. thesis, The Hebrew University 
of Jerusalem, Jerusalem. 

Committee on Soil Classification in Israel. (1979) The 
Classification of Israel Soils. Spec. Publ. Agric. Res. 
Orgn, Bet Dagan 137. (Hebrew, with English sum- 
mary) 

Dan, J. (1977) The distribution and origin of Nari and 
other lime crusts in Israel. Israel J. Earth -Sci. 26: 68- 
83. 

Dan, J. (1979a) The soil pattern of the arid regions of Is- 
rael and its effect on present and potential land use. in: 
Golany, G. {Ed.} Arid Zone Settlement Planning, the 
Israel Experience, pp. 214-252. Pergamon Press, New 
York, NY. 

Dan, J. (1979b) {The soils of the Negev.} in: Shmueli, A. 
and Grados, Y. {Eds.} {The Land of the Negev, Man and 
Desert.} pp. 203-218. Publishing House, Israel Minis- 
try of Defense, Tel Aviv, (in Hebrew) 

Dan, J. and Alperovitch, N. (1971) The Soils of the Middle 
and Lower Jordan Valley. Prelim. Rep. Valcani Inst. 
Agric. Res., Bet Dagan 694. (Hebrew, with English 
summary) 

Dan, J. and Alperovitch, N. (1975) The origin, evolution 
and dynamics of deep soils in the Samarian Desert. 
Israel J. Earth-Sci 24. 57-68. 

Dan, J., Gerson, R., Koyumdjisky, H. and Yaalon, D.H. 
(1981) Ardic soils of Isreal, properties, genesis and 
management. Spec. Publ. No. 190 Agric. Res. Orgn. Bet 
Dagan, Isreal. 

Dan, J., Moshe, R. and Alperovitch, N. (1974) The soils of 
Sede Zin. Israel J. Earth-Sci. 22. 211-227. 

Dan, J. Nissim, S. (1976) Soils of the Mountainous Parts of 
the Judean and Samarian Deserts. Spec. Publ. Agric. 
Res. Orgn, Bet Dagan. 60. (Heberew, with English 
summary) 

Dan, J. and Raz, Z. (1970) The Soil Association Map of 
Israel (scale 1:250,000). Volcani Institute of Agricul- 
tural Research, Bet Dagan; and Dept. of Soil Conserva- 
tion, Tel Aviv. (Hebrew, with English summary) 

Dan, J. and Yaalon, D.H. (1971) On the origin and nature 
of the paleopedological formations in the coastal desert 
fringe areas of Israel, in: Yaalon, D.H. {Ed.} Paleo- 
pedology - Origin, Nature and Dating of Paelosols. pp. 
245-260. Israel Universities Press, Jerusalem. 

Dan, J. and Yaalon, D.H. (1980) The evolution of soil and 
landscape in the northern Negev. Studies in the Geog- 
raphy of Israel 11: 31-56. 

Dan, J., Yaalon, D.H., Koyumdjisky, H. and Raz, Z. (1976) 
The Soils of Israel (with map 1:500,000). Pamph. Agric. 
Res. Orgn, Det Dagan 159. 



28 



SIXTH INTERNATIONA! S oil C LASSIPICATION WORKSHOP 



Dan, J., Yaalon, D.H., Moshe, R. and Nissim, S. (1982) 
Evolution of Reg soils in Isreal and Sinai. Geoderma 
28: 173-202. 

Danin, A., Orshan, G. and Zohary, M. (1975) The vegeta- 
tion of the northern Negev and the Judean Desert of 
Israel. IsraelJ. Bot. 24: 118-172. 

Evenari, M., Shanan, L. and Tadmor, N. (1971) The 
Negev: The Challenge of a Desert. Harvard University 
Press, Cambridge, MA. 

Horowitz, A. (1979) The Quaternary of Israel. Academic 
Press, New York, NY. 

Kovda, V.A. (1973) Soils in relation to salinity, irrigation 
and drainage, in: Irrigation, Drainage and Salinity - 
An International Source Book. FAO/UNESCO. pp. 55- 
79. Hutchinson, London. 

Litaur, M. (1980) The soils in a variegated region of 

northeasatern Samaria: their character, means of forma- 
tion and influence on plant associations. M.Sc. thesis, 
Tel Aviv University, Ramat Aviv, Israel. (Hebrew, with 
English summary) 

Picard, L. (1943) Structure and Evolution of Palestine. 
Bull. Dept. of Geology, The Hebrew University of 
Jerusalem, Jerusalem. 

Picard, L. (1970) Geological Map of Israel, 1:500,000. 
Atlas of Israel, III/l. Survey of Israel, Ministry of 
Labour, Jerusalem; and Elsevier Co., Amsterdam. 

Ron, Z. (1967) {The Geomorphology of the Elat Region, 
including a 1:100,000 Map.} Survey of Elat Region, 
Israel, pp. 83-108. The Elat Regional Council, (in He- 
brew) 

Rosenan, N. ( 1970) Rainfall Map of Israel. Atlas of Israel, 
IV/.2. Survey of Israel, Ministry of Labour, Jerusalem, 
and Elsevier Co., Amsterdam. 



Soil Survey Staff (1975) Soil Taxonomy Handbk. U.S. 
Dept. Agric. 436. 

Tsoar, H. (1978) Dynamics of Longitudinal Dunes. PhJ), 
thesis, The Hebrew Univ. of Jerusalem, Jerusalem t 
(Hebrew, with English summary) 

Waisel, Y., Pollak, G. and Cohen, Y. (1978) {The Ecology of 
Vegetation of Israel.} Div. of Ecology, Tel Aviv Univer- 
sity, Ramat Aviv, Israel, (in Hebrew) 

Yaalon, D.H. (1963) On the origin and accumulationof 
salts in groundwater and in soils of Israel. Bull. Res. 
Coun. Israel 11G: 105-131. 

Yaalon, D.H. (1978) Geoderma - Continental Sedimenta- 
tion, Calcrete, Desert Loess and Paelosols, Sand Dunes 
and Eolianites. Guidebook 10th Int. Congr. Sedimen- 
tology, Jerusalem, part II, pp. 195-238. 

Yaalon, D.H. and Dan, J. (1974) Accumulation and distri- 
bution of Loess-derived deposits in the semi-desert and 
desert fringe areas of Israel. Z. Geomorph. (SuppL) 20: 
91-104. 

Yaalon, D.H. and Ganor, E. (1975) Rates of aeolian dust 
accretion in the Mediterranean and desert fringe envi- 
ronments of Israel. Int. Congr. Sedimentology, Nice. 2: 
169-174. 

Zeidenberg, R. and Dan, J. (1981) The Influence of Lithol- 
ogy, Relief and Exposure on the Soil and Vegetation of 
the Arid Region of Eastern Samaria. Israel J. Bot. 30: 
13-31. 

Zohary, M. (1955) {Geobotany.} Sifriat Hapoalim, Mer- 
havya. (in Hebrew) 



Physical Properties Affecting the Productivity and Management 

of Clay Soils in Saskatchewan 

E. de Jong* and J.A. Elliott 1 



Abstract 

In Saskatchewan, clay and heavy clay textured soils historically are rated 
as having a higher productivity than light- and medium- textured soils. Data 
from field experiments are used to illustrate how differences in soil water, 
temperature, aeration, and mechanical properties of the root zone could 
lead to differences in productivity for soils of different texture. The higher 
productivity of clay soils probably is due largely to their higher water hold- 
ing capacity (an important factor in the prevalent crop-fallow rotation), 
possibly higher infiltration of snowmelt water, and higher hydraulic con- 
ductivity at the permanent wilting point. Higher denitrification losses 
could explain why yields on clay soils peak at lower total crop water use 
than yields on medium- textured soils. Problems with wind erosion and lack 
of spring workdays may be more severe on clay soils than on medium-tex- 
tured soils. 



Introduction 

This paper compares the physical properties 
of clays and medium-textured soils, with par- 
ticular reference to those properties that affect 
on the agronomic potential of these soils. Field 
data or data obtained on undisturbed core 
samples are used as much as possible. Unfortu- 
nately, the records often cover only a few years 
and frequently different soils were involved. 

Four soil physical factors are recognized as 
controlling the productivity and management of 
soils: the soil water, the soil aeration and the 
soil temperature regimes, and the mechanical 
properties of the root zone. These four factors 
vary over space and time and to a degree 
depend on each other. In particular, 
variations in water content affect the 
aeration, temperature, and mechanical 
strength of a soil. 



rary flooding of flat and depressional areas occa- 
sionally interferes with farm operations and 
drowns crops. 

Spring wheat is the most commonly grown 
arable crop in Saskatchewan, and the productiv- 
ity rating in Mitchell et al. (1944) is based 
mainly on wheat grown on fields that were fal- 
lowed the previous year. Under these condi- 
tions, the higher water storage capacity of the 
heavier textured soils is a distinct advantage. 
When yield is plotted against crop water use 
(Fig. 1), the difference in maximum yields ob- 
tained on heavy and medium-textured soils is 
small and probably not significant. Figure 1 



Productivity 

Historical yield records indicate that 
the heavier-textured glacial lake depos- 
its represent some of the best agricul- 
tural land in Saskatchewan (see Table 
17 in Mitchell et al., 1944). The high pro- 
ductivity rating of these soils reflects 
their high fertility and drought resis- 
tance and the fact that their smooth and 
undulating topography and absence of 
stones make cultivation easy. Drainage 
is generally adequate, although tempo 

1 E. de Jong and J.A. Elliott, Saskatchewan In- 
stitute of Pedology, University of Saskatchewan, 
Saskatoon, Saskatchewan S7N OWO, Canada. 

"Corresponding author. 



o Medium textured soils 
x Heavy clays 



3000 p 



UJ 



tr 
o 



20OO- 



10OO- 




iO 



20 30 

CROP WATER USE, cm 



40 



50 



Fig. 1 Relationships between spring wheat yields and water use 
for Brown and Dark Brown soils (de Jong and Halstead, 1987). 



29 



30 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Sand Silt Clay Density 


% Mg/m* 


X Sc HvC 6 23 70 1050 


Kp SiCL 18 36 46 138O 




O x 


-1000 


- 


O 




QL 


o 


** 


X 


J" -100 


x x 


*3C 


* 


1- 


XX 


Z 




UJ 




[ -<|Q 


X 


O 




a. 


O w 
X 



a 


X 

o 


H " 1 


X 


5 




-O 1 


1 1 1^1 1 -J 


0.1 0.2 0.3 0.4 0.5 0.6 


WATER CONTENT, g/g 


Fig. 2 Water retention of the subsoils of a Sceptre heavy clay (de Jong 


and Stewart, 1973) and Keppel silty clay loam (Patterson, 1985). 



does illustrate that the heavier soils yield better 
at low water use than the medium-textured 
soils, while at high water use the reverse is true. 
Thus, under semiarid conditions and when prac- 
ticing continuous cropping, the crop uses water 
more efficiently on heavier than lighter textured 
soils. In the following sections, some of the fac- 
tors responsible for this difference are explored. 

Soil Water Regimes 

The amount of water held at a given matric 
potential generally increases as clay content 
increases (Fig. 2). When the water content is 
expressed on a volume basis, the differences are 
less pronounced as the bulk density generally 
increases with decreasing clay content. 
Whether on a volume or a weight basis, avail- 
able water holding capacity, the difference be- 
tween field capacity (assumed to correspond to - 
33 kPa matric potential) and permanent wilting 
point (-1500 kPa matric potential), generally 
increases with increasing clay content. The 
changes in soil volume that accompany changes 
in soil water content result in shrinkage cracks 
that can have a significant effect on transport 
processes in clay soils. 

The saturated hydraulic conductivity of clay 
soils is generally lower than that of medium- 



and light-textured soils. For ex- 
ample, the saturated hydraulic con- 
ductivity of the Sceptre soil in Fig- 
ure 2 is about 30-40 cm/day, while 
the value for the Keppel soil is 
about twice that. However, as the 
soil water content decreases, the 
unsaturated hydraulic conductivity 
of a clay soil decreases less rapidly 
than that of a lighter soil (Fig. 3; see 
also Gardner, 1960; Yang and de 
Jong, 1972). At -33 kPa matric po- 
tential, both soils have a hydraulic 
conductivity of about 10~ 2 cm/day, 
but at -1500 kPa the heavy clay has 
a hydraulic conductivity of 10" 4 cm/ 
day compared to around 10" 5 cm/day 
for the silty clay loam. The higher 
hydraulic conductivity of the clays 
at -1500 kPa may explain their 
higher productivity than lighter 
soils when water is limiting (Fig. 1) 
and lowers the matric potential at 
which permanent wilting occurs 
(Yang and de Jong, 1972). 

Hydraulic conductivity is highly 
variable (Warrick and Nielsen, 
1980). Data from Kirkland (1986) 
shows that the hydraulic conductivity of a 
shrinking clay was more variable than that of a 
non-shrinking medium textured soil (Fig. 4). 
The hydraulic conductivities were measured on 
undisturbed soil cores collected at various times 



>* 
o 

E 
o 



3b 



QD 
>> o 



x Sceptre HvC 

field Test, Keppel SiCL 

o cores, Keppel SiCL 




_L 



JL 



-I .2 .3 .4 .5 .6 

WATER CONTENT, cm 3 /cm 3 

Fig. 3 Dependence of hydraulic conductivity on soil 
water content for Sceptre heavy clay (de Jong and 
Stewart, 1973) and Keppel silty clay loam (Patter- 
son, 1985). 



DE JONG AND ELLIOTT: PROPERTIES/MANAGEMENT OP CLAY SOILS IN SASKATCHEWAN 



31 



during three growing seasons (each value is the 
geometric mean of two or more cores). The 
higher variability in the clay is probably due to 
the presence or absence of shrinkage cracks 
which, even though closed when the soil is satu- 
rated, represent preferential flowpaths (Ritchie 
et al, 1972; Bouma, 1981). At -5 kPa matric 
potential, the variability had significantly de- 
creased, as cracks or pores larger than 60 |im 
would no longer contribute to water flow and, 
hence, the major source of the variability was 
reduced. 

Cracking in clay soils increases infiltration 
rates (Allan and Braud, 1966; Blake et al., 
1973). Large amounts of water more or less in- 
stantaneously disappear into the cracks when 
water is ponded on the surface (Fig. 5). Johnson 
(1962) has suggested that vegetation-induced, 
controlled soil cracking might be an economi- 
cally feasible way to increase infiltration on dry- 



10 3 


Sand Silt Clay 




o/ 




/O 

K * * 




^ Su SiC 12 45 43 X x xx 6 x 


102 


o E SiL 17 56 27 ^g^^ 


o 


JC$f^^ 


TJ 


jfjEf* 


X 


_^r^l<P^^ 


10 

O 


Satura ted^xP^^ 




Jf 


> 1 


rf* 




o oo o ^^ 
x xx^ 


P , 


^ 


D 


x xxx 





5 kPa o o o o o o 


^r 2 


rt o o o o o 


O 10 


- O ooo 





o o , 




o 




x 


o 





M , -3 


X 


_i 10 


" X 


D 


X X 


^rt* 


V * 


tt: 


X X * 


10 4 


X * 


X 




10 5 


X XXX 




MEAN 




\ 




1 1 1 1 f 1 II 


015 20 50 80 95 99 


PROBABILITY, % 


Fig. 4 Hydraulic conductivity of a Sutherland silty clay and an Elstow silty loam 


(Kirkland, 1986). 



farmed wheatlands. In Saskatchewan these 
cracks would be important only for infiltration 
during the rare, intense summer storms which 
might cause surface ponding. At low and moder- 
ate rainfall rates the soil would gradually wet 
up, and rain falling directly in the cracks proba- 
bly would be absorbed by the walls. 

On the Canadian Prairies, infiltration of wa- 
ter from the melting snowpack is often a major 
soil water recharge event, and cracks do contrib- 
ute significantly to the entry of snowmelt water. 
In non-cracked soils, snowmelt infiltration de- 
pends on the amount of snow water and the fro- 
zen water content of the surface soil (Gray et al., 
1984). Snowmelt infiltration in dry, cracked 
clays usually is limited only by the amount of 
water in the snowpack. Gray and his co-workers 
are studying the possibility of increasing snow- 
melt infiltration by creating artificial cracks. 
On heavy clay soils, variation in the spacing and 

orientation of wheat 
rows with respect to 
the slope might natu- 
rally give the desired 
effect. 

Cracks obviously 
increase the possibil- 
ity of evaporation 
losses from the body 
of the soil. For ex- 
ample, Johnston and 
Hill (1945) found de- 
creased water con- 
tents in the soil adja- 
cent to the cracks. 
Selim and Kirkham 
(1970) estimated 
that cracks increased 
evaporation by up to 
30% in a laboratory 
experiment; these in- 
creases are probably 
larger than those 
which occur in the 
field as the soil was 
saturated initially. 
The work of Selim 
and Kirkham sug- 
gests that several 
small cracks proba- 
bly lead to greater 
water losses than a 
single, wide crack. 

The soil water re- 
gime is a function of 



32 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Table I. Maximum and minimum soil water contents 

under different vegetation on Sceptre heavy clay 

(from de Jong and MacDonald, 1975). 

Total water to 135 cm depth 
Vegetation Maximum Minimum 

cm 

Native grassland 55 38 

Wheat on fallow 58 41 



Table 2. Estimated range of available water (50% 

probability) at seeding and heading for two soil types 

(adapted from de Jong and Bootsma, 1988). 

Area NortheaslSaskatchewan SouthwestSaskatchewan 
Rotation Wheat/Fallow ContWheat Wheat/Fallow ContWheat 
mm available water 

At seeding 

Heavy class 1 240-210 210-140 140-120 90-80 

Medium 

texture 2 140-125 135-105 95-80 65-55 



At heading 

Heavy clays 1 150-125 115-95 

Medium 
texture 2 65-55 55-45 

^50 mm water holding capacity 
2 150 mm water holding capacity 



80-55 



35-25 



55-35 
25-15 



vegetation, topography, soil type, and climate. 
The effect of vegetation is illustrated in Table 1. 
Continuous wheat would be expected to have 
water contents between those of the native 
grassland and the wheat grown on summerfal- 
low (Table 1). Kirkland (1986) found little dif- 



Moisture, % volume 
Site 0-15 cm 15- 30cm 
o 3 

18.0 2O.O 

4 




u 




500 



1000 



TIME, minutes 



Fig. 5 Cumulative infiltration (I) in Sceptre heavy clay as a function of time 
at two moisture contents. 



ference in the water extraction pattern of con- 
tinuous wheat grown on Sutherland silty clay 
and Elstow silty loam, under two tillage systems 
(Fig. 6). Note that overwinter-recharge was 
similar on both soils in 1983-84 when the soils 
were wet the previous fall. After the dry sum- 
mer of 1984, the silty clay gained considerably 
more water over the 1984-85 winter than the 
silty loam, presumably due to the role of cracks 
in snowmelt infiltration. Much of the additional 
recharge in the Sutherland silty clay was found 
below 50 cm depth. 

Unfortunately there are no continuous long- 
term records for the soil water regime under dif- 
ferent soil/crop combinations. De Jong and 
Bootsma (1988) have used the Versatile Soil 
Moisture Budget (VSMB) to predict variability 
in soil water on the Canadian Prairies. The 
VSMB provides as good a prediction of growing 
season changes in soil moisture under Sas- 
katchewan conditions as more sophisticated 
models (cf. de Jong and MacDonald [1975] ver- 
sus de Jong and Hay hoe [1984]; de Jong, 1988), 
but problems tend to arise in the predictions of 
overwinter soil water gains (de Jong and 
MacDonald, 1975). Despite this, de Jong and 
Bootsma (1988) ran the model for 60 years using 
existing climatic records. 
Table 2 shows some of their 
predictions for continuous 
wheat and wheat on fallow on 
a heavy and a medium-tex- 
tured soil. At maturity both 
the wheat on fallow and the 
continuous wheat in south- 
west Saskatchewan would 
have depleted most, if not all, 
of the available water. Some 
available water probably 
would still be present in both 
rotations in northeast Sas- 
katchewan, where the climate 
is less arid. Fallow fields 
would have water contents 
similar to fields of continuous 
wheat in spring and would 
stay at that water content un- 
til about September. From 
September to the next spring, 
all fields would recharge to 
the seeding water contents 
represented by wheat on fal- 
low and continuous wheat in 
Table 2. 

Dasog (1986) used the pre- 
dictions of de Jong and 



1500 



DE JONG AND ELLIOTT: PROPERTIES/MANAGEMENT OP CLAY SOILS IN SASKATCHEWAN 



33 



E 

U 



UJ 



UJ 

.J 

U. 

o 

or 
a. 



50 



40 



30 



20 



60 



50 



40 



30 



20 





Conventional 

Zero till 

Elstow 

* Sutherland 




tillage 



10 





100 



-I 



100 190 2OO 23O 30O 

1984 
CALENDAR DAY 



1 


iJlJ 





o 

3 



1985 



Fig. 6 Temporal changes in profile water under zero and conventional tillage 
on Sutherland silty clay and Elstow silty loam. 



50 



40 



rr 3O 



CO 

o 
cc 



o: 



Ul 
I- 



20 
10 

O 
50 

40 




O 

O 30 



a: 
LJ 



2O 



10 





100 150 20O 250 300 100 150 200 250 300 

1983 1984 

CALENDAR DAY 

Fig. 7 Surface soil air-filled porosity and volumetric water con- 
tent for Sutherland silty clay and Elstow silty loam (see Fig. 
6 for legend). 



Bootsma (1988) to estimate 
available water at various 
depths in clay soils and to 
predict crack duration. The 
results are summarized in 
Table 3. Since the model 
does not take into account 
the enhanced snowmelt re- 
charge and evaporation as- 
sociated with cracks, it is 
doubtful whether the data 
in Table 3 is particularly re- 
liable. Dasog (1986) did not 
try to evaluate the cracking 
outside the growing season, 
since no available water 
predictions were tabulated 
between November 1 and 
April 1. Dasog's calcula- 
tions do indicate that crack- 
ing is more severe in the 
subarid southwest than the 
subhumid northeast of Sas- 
katchewan, and they point 
out the increased frequency 
of cracking at depth. The 
latter is borne out by de Jong and 
MacDonald (1975), who found that from 
1968 to 1972 water contents below 75 
cm were virtually steady (at -1500 kPa) 
under native grassland in southwest 
Saskatchewan. Although the model 
may give some indication of the extent 
of cracking, it does not and cannot pre- 
dict the number of cracks (i.e., crack 
spacing). 

Soil Aeration 

Kirkland (1986) showed no signifi- 
cant differences in soil aeration between 
zero- and conventional tillage and only a 
small effect of texture. Surface water 
contents were generally higher on the 
Sutherland silty clay than the Elstow 
silty loam (Fig. 7); therefore air-filled 
porosity showed the opposite trend. For 
much of 1983, air-filled porosity was 
below 20% in the Sutherland soil be- 
cause of the abundant rains (Fig. 6). 
Air-filled porosities of less than 10% 
limit root growth (Wesseling, 1974), al- 
though the critical value may be nearly 
twice as high in Vertisols (Hodgson and 
MacLeod, 1989). Despite the potential 
for restricted aeration, in 1983 the aver- 



34 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



H 1000 



CD 
< 
ill 

2 

o: 
LU 

OL 



100 



10 



*; 



10 



20 



30 



40 



50 



I- 0.25 



0.20 



0.15 



LL 
Q 



h- 0.05 



<oo 



00 



IfoOXx* o 



10 



20 



30 



40 



50 



AIR POROSITY, % v/v 



Fig. 8 Air permeability and relative diffusivity as a function of 
air- filled porosity for Sutherland silty clay (x) and Elstow silty 
loam (o). 



age soil C0 2 concentrations did not exceed 2.5% 
in the Sutherland soil and 1.5% in the Elstow. 
C0 2 levels were much lower in the dry 1984 
growing season. 

To compare pore-efficiency for gaseous trans- 
port, air permeability and relative diffusivity 
were measured several times in the 1982, 1983, 
and 1984 growing seasons on 7.5 cm diam. by 
7.5 cm high soil cores. The relationship between 
relative diffusivity (the ratio of gaseous diffusion 
in the soil to diffusion in air) and air-filled poros- 
ity was similar for both soils, but air permeabil- 



Table 3. Estimated crack duration (days) from 

May 1 to Sept. 1 in clay soils with 200 mm available 

water holding capacity (Dasog, 1986). 1 

Depth NortheasSaskatchewan SouthwesGaskatchewan 

(cm) - days with less than 40% available water - 

0-6 80 80 

6-15 79 86 

15-30 71 78 

30-60 65 83 

60-90 82 105 

90-120 90 115 

1 Assuming 200 mm storage and continuously cropped to wheat 



ity appeared to be slightly higher on the 
Sutherland silty clay than the Elstow 
silty loam (Fig. 8). This would indicate 
that the Sutherland soil has a better 
developed system of continuous, large, 
more-or-less vertical pores than the El- 
stow silty loam. Presumably these 
large, more or less vertical pores are the 
shrinkage cracks in the Sutherland soil, 
even though an effort was made to 
avoid cracks when the samples were 
collected. Thus, the effect of cracks on 
air permeability is probably under-esti- 
mated. 

Although the composition of the air 
in the large pores was similar in the 
silty clay and silty loam, this is not nec- 
essarily true for aeration inside the ag- 
gregates. In non-shrinking soil, air en- 
ters the aggregates as the moisture 
content drops. Aggregates of shrinking 
soils exhibit normal shrinkage (i.e., any 
decrease in water content is accompa- 
nied by an equal decrease in volume), 
as Dasog's data for two heavy clays il- 
lustrate (Fig. 9). 

Using data from undisturbed cores, 
de Jong and Stewart (1973) calculated 
that at saturation the Sceptre subsoil 
had a bulk density of 1250 Mg/m 3 and a 
water content of 0.44 g/g. The cores 
showed structural shrinkage (volume 
loss less than the volume of water lost) 
between 0.44 and 0.42 g/g water con- 



0.8 

10 
O 

* O.6 
U 

i 

o * 
li. 
u 

CL ^ 
CO 


Fig.9P 




ScmBmkl } _ 
I Dasog 

R Bmk2 J (1986) 
/ C Cracks 
S Solid 
SP Sta We inlef aggregate Pores 
' SIP Swelling in tef aggregate Pores 


"T ] 


^4**T^? 
f LS ^^5P 

^ w , 


I ^ 

J s ; 

f 1 . \ 


, IP Interaggregate Pores 
W Water 

\ . 


0.2 0.4 
WATER CONTENT, g/g 

ore space in a Sceptre heavy clay. 



DE JONG AND ELLIOTT: PROPERTIES/MANAGEMENT OP CLAY SOILS IN SASKATCHEWAN 



35 



tent and normal shrinkage from 0.42 to 0.24 g/g 
water content. During structural shrinkage, air 
will enter some stable pores in and between ag- 
gregates, and air-filled pores (or cracks) between 
aggregates should increase in volume. The air- 
filled pores within the aggregates should not in- 
crease until residual shrinkage starts (at about 
0.22 g/g in Fig. 9). The -1500 kPa water content 
for disturbed Sceptre soil is about 0.25 g/g, 
which would suggest that the aggregates re- 
main nearly water-saturated over the range of 
plant available water. 

Because of the difference in degree of water- 
saturation, anaerobic microsites are more likely 
to occur in heavier than lighter textured soils, 
even when the composition of the air in the in- 
teraggregate pores is similar. Denitrification 
measurements support this; e.g., Colaco (1979) 
measured five times as much denitrification on 
a fallowed Sutherland clay than on a medium- 
textured soil similar to the Elstow. The in- 
creased denitrification could perhaps explain 
the lower yields at high water use on heavy 
clays than on medium- textured soils (Fig. 1). 

Soil Temperature 

At the same water content (g/g or cm 3 /cm 3 ), 
lighter textured soils tend to have a higher ther- 
mal conductivity and thermal diffusivity than 
clays (van Wijk and de Vries, 1963). Hence one 
would expect that in the same climatic setting, 
and other factors being similar, temperature 
fluctuations would be slightly larger in the 
sandier soil, but the means would be quite simi- 
lar for soils of different texture (cf. Geiger 1959, 
Table 24). Since heavier textured soils are often 
wetter than lighter soils, they usually warm up 
slower in the spring and cool down slower in the 
fall than lighter soils (Shulgin 1957, Table 20). 

Figure 10 compares 10-day running means of 
soil temperatures in the Sutherland and Elstow 
soils studied by Kirkland (1986). Tillage treat- 
ment only affected soil temperatures in 1984 
when the conventional till plot on the Elstow soil 
was consistently warmer at and 20 cm than its 
zero-till counterpart. This was probably due 
largely to less shading by vegetation as the con- 
ventional till plot yielded 480 kg grain/ha versus 
1135 kg grain/ha for the zero-till plot. The tem- 
perature patterns in 1983 follow the expected 
trend, with the silty clay soil being somewhat 
colder than the silty loam before mid-July and 
the reverse being true in the fall. During the 
very dry 1984 growing season, the soil tempera- 



tures at 20 and 50 cm depths in the silty clay 
were persistently a few degrees lower than in 
the silty loam. Since moisture conditions were 
quite similar for both sites (Fig. 6), the lower 
subsoil temperatures could be due to evapora- 
tion from the walls of cracks (Selim and Kirk- 
ham, 1970). 

Ripley (1973) observed very rapid warming 
(from -2 to +1C) of a Sceptre heavy clay soil 
under native vegetation in spring and suggested 
that infiltration of the snowmelt water down the 
cracks was the likely cause. No such trend was 
observed for the Sutherland silty clay in the 
spring of 1985, even though there was consider- 
able recharge (Fig. 6), but the trend may not be 
visible using 10-day means. Also, the major dif- 
ference in recharge between the two soils was 
below 50 cm depth, where no temperature data 
were collected. 

From the limited data it appears that the 
thermal regimes of mineral soils are not 
strongly affected by texture. It is therefore un- 
likely that different temperature regimes would 
account for the differences in soil productivity 
between heavy and light textured soils. 

Mechanical Properties of the 
Root Zone 

Attempts to define the mechanical properties 
of the root zone that lead to high soil productiv- 
ity generally have been unsuccessful. Several 
authors have reported critical bulk densities 
above which root growth is limited in various 
soil textures (e.g., Pierce et al., 1983), but few of 
Saskatchewan's soils exceed the critical limits 
when sampled at -33 kPa matric potential 
(Ayres et al., 1973). At -1500 kPa matric poten- 
tial, clod densities of shrinking soils may ap- 
proach or exceed the critical values (Dasog, 
1986). Others have concentrated on establish- 
ing critical soil strengths for root growth (e.g., 
Taylor et al., 1966), which decreases as clay 
content increases (Gerard et al., 1982). Ayres et 
al. (1973) measured the strength of several Sas- 
katchewan soils at -33 kPa and found little dif- 
ference between heavy and light-textured soils; 
none of the measured values were close to the 
critical soil strengths reported by Taylor et al. 
(1966). 

Although attempts to relate aggregate size 
distribution to crop production have been 
largely fruitless, aggregate size distribution is 
important in terms of soil erodibility. Water 
erosion risk maps for Canada (Coote et al., 1982) 



36 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



25 r 





150 200 250 300 
1983 



50 



100 150 200 
1984 



CALENDAR DAY 



Fig. 10 Mean daily temperature (10 day averages) for Sutherland silty clay and 
Elstow silty loam (see Fig. 6 for legend). 



show an increase in water erosion risk as the 
clay content of soils increases, even though 
Evans (1980) indicated that medium-textured 
soils were most at risk. In general wind erodibil- 
ity decreases as silt and clay content increase 
(Wilson and Cooke, 1980; Coote et al., 1982); 
however, on the Canadian Prairies medium tex- 
tured soils are considered most resistant to wind 
erosion (Johnson, 1983; Chepil, 1953). 



Although texture 
may be one indicator of 
soil erodibility, aggre- 
gate size distribution is 
the critical factor, as 
both wind- and water- 
erodibility increase as 
the percentage of ag- 
gregates less than 0.5 
to 1 mm increases. Ag- 
gregate size distribu- 
tion changes continu- 
ally in response to 
weather and manage- 
ment, and it is impos- 
sible to make any firm 
statements on differ- 
ences among different 
soil textures. Figure 11 
shows the variation in 
the erodible fraction 
(aggregates <0.83 mm 
diameter as a percent- 
age of the total soil 
weight) of five different 
texture groupings. 
There is a distinct dif- 
ference in the behavior 
of the heavier and 
light- textured soils 
over the winter period, 
with the former show- 
ing an increase in 
erodibility due to frost 
action. The effect of 
frost on aggregate sta- 
bility is strongly af- 
fected by water content 
at freezing, initial ag- 
gregate size, and how 
the soils were dried af- 
ter thawing (Hinman 
and Bisal, 1968). In 
winters with little or no 
snow cover, freeze- 
drying leads to a very 
finely granulated struc- 
ture on clay soils such as the Sceptre and Regina 
associations and makes them highly erodible. 

Going from southwest to northeast Sas- 
katchewan, the length of the frost-free period 
decreases from about 120 to 100 days, and work- 
ability of the soil in the spring becomes critical. 
High soil water contents are more likely to limit 
workability on heavy- than light- textured soils. 
The plastic limit often is used as the maximum 





250 300 



5O 100 
1985 



150 



DE JONG AND ELLIOTT: PROPERTIES/MANAGEMENT OP CLAY SOILS IN SASKATCHEWAN 



37 



Erodible Fraction O.84 mm) 



100 



90 



80 



70 



60 



50 



40 




CL, SiCL 



JULY AUG SEPT 



NOV DEC JAN FEB MAR APR 



Fig. 11 Erodible fraction (% <0.84 mm) for five soil textural groups from July 
to April (Hilliard et aL, 1988). 



water content at which soils should be worked, 
and its relationship to field capacity (-33 kPa 
water content) depends on texture. Preliminary 
analysis of Atterberg limit data for Sas- 
katchewan topsoils suggest that the plastic limit 
is related to the -33 kPa water content (of dis- 
turbed samples) by 

PL = 2.6 + 0.42xH 2 0. 33kPa R* = 0.14 
^ Although the equation has a very low coeffi- 
cient of determination, it suggests that more 
time is required after the spring thaw for clays 
to become workable than for lighter textured 
soils. The importance of texture in determining 
spring field workdays is illustrated in Table 4. 
Especially in the subhumid area, the number of 
working days is substantially less on heavy 
clays than on lighter soils. 

Conclusions 

Past experience has shown that clay soils in 
Saskatchewan are more productive than me- 



Table 4. Expected (50% probability) spring work days 
(Dyer et al., 1978). 

Location Soil April 1 - May 5 April 1 - June 2 

work days 

Regina Heavy 17.7 35.2 

Light 21.9 46.5 

Melfort Heavy 9.2 33.6 

Light 18.3 42.5 



dium- and light- textured 
soils, especially under dry 
conditions and crop-fallow 
rotations. The higher pro- 
ductivity of the clays is at- 
tributed to their higher wa- 
ter holding capacity, which 
results in increased avail- 
able water in the root zone 
of fallow-seeded crops. The 
cracks in dry clays increase 
snowmelt infiltration, thus 
causing higher levels of 
available water than in 
lighter soils, in years when 
abundant snow follows a 
dry autumn. At -1500 kPa 
matric potential, the hy- 
draulic conductivities of the 
clays are higher than those 
of coarser soils, which may 
account for the higher wa- 
ter use efficiency of crops 
grown on clay when water 
is limiting. Hydraulic con- 
ductivity is spatially more 
variable in clays than in medium-textured soils, 
probably due to the contribution of cracks to 
water movement. 

The composition of the soil air in clay soils is 
similar to that in coarser soils, but denitrifica- 
tion data indicate that anaerobic microsites are 
more prevalent in clays. The higher loss of N by 
denitrification could explain why maximum 
yields occur at about 35 cm water use on clays 
and 40 cm water use on medium-textured soils. 
Soil temperature regimes appear to be similar 
for different textures, although infiltration of 
snowmelt water in the cracks might cause a 
more rapid thawing in the subsoils of clays than 
in non-shrinking soils in the spring. 

The number of spring working days is less on 
clays than on medium- and light-textured soils 
and this can become critical in the subhumid 
area of Saskatchewan where the growing sea- 
son is short. Although clays are generally con- 
sidered to be less erodible than lighter soils, 
freeze-thaw cycles under snow-free conditions 
break down the aggregates of clays and make 
them as wind-erodible as fine sandy loams. 

Literature Cited 

Allen, J. B. and H. J. Braud, Jr. 1966. Effect of cracks and 
initial moisture content on the infiltration rate of 
Sharkey clay. Bull. No. 613, Louisiana State Univ. and 
Agric. and Mech. College, Agric. Exp. Station. 



38 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Ayres, K. W., R. G. Button and E. de Jong. 1973. Soil mor- 
phology and physical properties. II. Mechanical imped- 
ance and moisture retention and movement. Can. J. 
Soil Sci. 53:9-19, 

Blake, G., E. Schlichting and U. Zimmermann. 1973. Wa- 
ter recharge in a soil with shrinkage cracks. Soil Sci. 
Soc. Am. Proc. 37:669-672. 

Bouma, J. 1981. Soil morphology and preferential flow 
along macropores. Agric. Water Manage. 3:235-250. 

Chepil, W. S. 1953. Factors that affect clod structure and 
credibility of soil by wind. I. Soil texture. Soil Sci. 
75:473-483. 

Coote, D. R., J. Dumanski and J. F. Ramsey. 1982. An 
assessment of the degradation of agricultural lands in 
Canada. LRRI Contr. #118, Agriculture Canada, Re- 
search Branch, Ottawa. 

Colaco, W. 1979. Denitrification in soils under a wheat- 
fallow rotation. M.Sc. Thesis, Dept. of Soil Science, 
Univ. of Saskatchewan, Saskatoon. 

Dasog, G. S, 1986. Properties, genesis and classification of 
clay soils in Saskatchewan, Ph.D. Thesis, Dept. of Soil 
Science, Univ. of Saskatchewan, Saskatoon. 

de Jong, E. and E. H. Halstead. 1987. Innovative Acres 
1986 Report. Publ. M80, The Saskatchewan Institute of 
Pedology, University of Saskatchewan, Saskatoon. 

de Jong, E. and K. B. MacDonald. 1975. The soil moisture 
regime under native grassland. Geoderma 14:207-221. 

de Jong, E. and J. W. B. Stewart. 1973. Description of site: 
III. Soil characterization. Tech. Rep. 36, Matador Proj- 
ect, Canadian Committee for the International Biologi- 
cal Programme. Univ. of Saskatchewan, Saskatoon. 

de Jong, R. 1988. Comparison of two soil- water models 
under semi-arid growing conditions. Can. J. Soil Sci. 
68:17-27. 

de Jong, R. and A. Bootsma. 1988. Estimated long-term 
soil moisture variability on the Canadian Prairies. Can. 
J. Soil Sci. 68:307-321. 

de Jong, R. and H. N. Hayhoe. 1984. Diffusion-based soil 
water simulation for native grassland. Agric. Water 
Manage. 9:47-60. 

Dyer, J. A., W. Baier, H. N. Hayhoe and G. Fisher. 1978. 
Spring field work day probabilities for selected sites 
across Canada. Tech. Bull. 86, Agriculture Canada, 
Research Branch, Ottawa. 

Evans, R. 1980. Mechanics of water erosion. In M. J. 
Kirkby and R. P. C. Morgan (ed) Soil Erosion. John 
Wiley & Sons, Chichester. 

Gardner, W. R. 1960. Soil water relations in arid and semi- 
arid conditions. In Plant-Water Relationships in Arid 
and Semi-arid Conditions. Arid Zone Research XV:37- 
61. 

Geiger, R. 1959. The climate near the ground. Harvard 
University Press. 

Gerard, C. J., P. Sexton and G. Shaw. 1982. Physical fac- 
tors influencing soil strength and root growth. Agron. J. 
74:875-879. 

Gray, D. M., D. I. Norum and R. J. Granger. 1984. The 
prairie soil moisture regime: Fall to seeding, p. 159-207. 
In G. W. Hass (ed) The Optimum Tillage Challenge. 
Division of Extension and Community Relations, Univ. 
of Saskatchewan, Saskatoon. 

Hilliard, C. R., H. P. W. Rostad and R. D. Stushnoff. 1988. 
Wind credibility of Saskatchewan soils. Report of 1987- 
88 results. Publ. M87, Saskatchewan Institute of Pedol- 
ogy, Univ. of Saskatchewan, Saskatoon. 



Hinman, W. C. and F. Bisal. 1968. Alterations of soil 
structure upon freezing and thawing and subsequent 
drying. Can. J. Soil Sci. 48:193-197. 

Hodgson, A. S. and D. A. MacLeod. 1989. Use of oxygen 
flux density to estimate critical air-filled porosity of a 
Vertisol. Soil Sci. Soc. Am. J. 53:355-361. 

Johnson, W. C. 1962. Controlled soil cracking as a possible 
means of moisture conservation on wheatlands of the 
southwestern Great Plains. Agron. J. 54:323-325. 

Johnson, W. E. 1983. Soil conservation: Canadian prai- 
ries. In H. E. Dregne and W.O. Willis (ed) Dryland Ag- 
riculture. Agron. 23:259-272. 

Johnston, J. R. and H. O. Hill. 1945. A study of the shrink- 
age and swelling properties of Rendzina soils. Soil Sci. 
Soc. Am. Proc. 9:24-29. 

Kirkland, J. A. 1986. The effects of zero and conventional 
tillage systems on soil physical properties. M.Sc. The- 
sis, Dept. of Soil Science, Univ. of Saskatchewan, Sas- 
katoon. 

Mitchell, J., H. C. Moss and J. S. Clayton. 1944. Soil sur- 
vey of southern Saskatchewan from Township 1 to 48 
inclusive. Soil Survey Report No. 12, Fourth Printing, 
Extension Division, Univ. of Saskatchewan, Saskatoon. 

Patterson, G. W. 1985. Effect of fall stubble management 
on overwinter soil moisture storage and crop yield. 
M.Sc. Thesis, Dept. of Soil Science, Univ. of Sas- 
katchewan, Saskatoon. 

Pierce, F. J., W. E. Larson, R. H. Dowdy and W. A. P. 
Graham. 1983. Productivity of soils: Assessing long- 
term changes due to erosion. J. Soil Water Conserv. 
38:39-44. 

Ripley, E. A. 1973. Description of site: II. Climatology of 
the Matador area. Tech. Rep. 12, Matador Project, 
Canadian Committee for the International Biological 
Programme. Univ. of Saskatchewan, Saskatoon. 

Ritchie, J. T., D. E. Kissel and E. Burnett. 1972. Water 
movement in undisturbed swelling clay soil. Soil Sci. 
Soc. Am. Proc. 36:874-879. 

Selim, H. M. and D. Kirkham. 1970. Soil temperature and 
water content changes during drying as influenced by 
cracks: A laboratory experiment. Soil Sci. Soc. Am. 
Proc. 34:565-569. 

Shulgin, A.M. 1957. The temperature regime of soils. Is- 
rael Program for Scientific Translations in 1965. 

Taylor, H. M., G. M. Roberson and J. J. Parker, Jr. 1966. 
Soil strength-root penetration relationships for me- 
dium- to coarse-textured soil materials. Soil Sci. 
102:18-22. 

van Wijk, W. A. and D. A. de Vries. 1963. Periodic tem- 
perature variations in homogeneous soil. In W. R. van 
Wijk (ed) Physics of Plant Environment. North-Holland 
Publ. Co., Amsterdam. 

Warrick, A. W. and D. R. Nielsen. 1980. Spatial variability 
of soil physical properties in the field, p. 319-344. InD. 
Hillel (ed) Applications of Soil Physics. Academic Press, 
New York. 

Wesseling, J. 1974. Crop growth in wet soils. In J. van 
Schilfgaarde (ed) Drainage for Agriculture. Agron, 
17:7-37. 

Wilson, S. J. and R. U. Cooke. 1980. Wind erosion. In M. J. 
Kirkby and R. P. C. Morgan (ed) Soil Erosion. John 
Wiley & Sons, Chichester. 

Yang, S. J. and E. de Jong. 1972. Effect of aerial environ- 
ment and soil water potential on the transpiration and 
energy status of water in wheat plants. Agron. J. 
64:574-578. 



Alternative Classification of Soils with Aridic Soil Moisture Regimes 

R.J. Engel, J.E. Witty, and J.D. Nichols* 



Introduction 

So il Taxonomy, A Basic System of Soil Classi- 
fication for Making and Interpreting Soil Sur- 
veys, 1 has a structure for relating soils to each 
other. According to Dr. Richard Arnold, 2 
"Classes at the order level are separated on the 
basis of properties resulting from the major 
processes of soil formation." Properties used 
are, in general, those that are thought to be the 
result of soil genesis. Soil climate is used as 
order criteria only in the orders of Aridisols and 
Ultisols. 

The objective of Soil Taxonomy is to provide 
hierarchies of classes that permit us to under- 
stand, as fully as existing knowledge permits, 
the relationship between soils and the factors 
responsible for their character (Soil Taxonomy). 

The International Committee on Aridisols, 
ICOMID, issued circular letter version 6.0 on 
April 13, 1989. This is a major revision of the 
Aridisol order. Now, during this second major 
study tour of Aridisols in the United States, is 
an opportune time to examine of some of the 
basic premises of the Aridisol order. 

Soil Taxonomy 

The order of Aridisols is based largely on the 
soil having an aridic moisture regime. In addi- 
tion, the key to soil orders requires Aridisols to 
have one of the following, whose upper bound- 
ary is within 100 cm of the soil surface: petro- 
calcic, calcic, gypsic, petrogypsic, or cambic hori- 
zon or a duripan, or an argillic or natric horizon 
and an epipedon that is not both massive and 
hard or very hard when dry. Some Aridisols do 
not require an aridic moisture regime but have a 
salic horizon within 75 cm of the surface and are 
saturated with water within 100 cm of the sur- 
face for 1 month or more in some years and do 
not have an argillic or natric horizon. 

The Aridisol order is the only order to have 
the soil moisture regime diagnostic at the order 
level. However, the Ultisol order requires a 
temperature regime that is mesic, isomesic, or 
warmer at the order level. 

A proposal to delete the temperature regime 
as criteria at the order level in Ultisols is being 
developed in New York State. 3 



"Classes at the suborder level are separated 
within each order on the basis of soil properties 
that are major controls, or reflect such controls, 
on the current set of soil-forming processes," 
according to Arnold. 2 Moisture regimes are 
commonly used as a criteria for suborders in 
Soil Taxonomy. The moisture regime is the 
basis of 26 suborders and of at least one subor- 
der in all orders except Aridisols. 

Soils with an aridic moisture regime are not 
treated consistently in Soil Taxonomy. With the 
approval of National Soils Taxonomy Handbook 
notice no. 13, 223 subgroups in Soil Taxonomy 
either are required to have or are permitted to 
have an aridic moisture regime. The order of 
Aridisols includes 137, or 61 percent, of these 
subgroups. 

The remaining 86, or 39 percent, of the sub- 
groups are in the orders of Alfisols, Andisols, 
Entisols, Mollisols, Oxisols, and Vertisols. In 
addition, Xeralfs are soils with an aridic mois- 
ture regime bordering on a xeric moisture re- 
gime and having both an argillic or natric hori- 
zon and a surface layer that is both hard or very 
hard and is massive. Xeralfs are not included in 
these numbers. Table 1 lists subgroups in orders 
other than Aridisols that are permitted to have 
or are required to have an aridic moisture re- 
gime. 

Soils with an aridic moisture regime are 
treated, like other moisture regimes, at the sub- 
order level in the orders of Andisols, Oxisols, 
and Vertisols. They are treated, like other mois- 
ture regimes, at the great group level, in the 
order of Entisols. Soils with an aridic moisture 
regime are recognized at the subgroup level in 
the order of Mollisols and, where recognized 
above the series level, in the order of Alfisols. 

This paper examines some of the conse- 
quences of relegating the aridic soil moisture 
regime to a suborder or lower level and of elimi- 
nating the Aridisol order. Such an examination 



"USDA, Soil Conservation Service 



il Survey Staff, "Soil Taxonomy, a Basic System of 
Soil Classification for Making and Interpreting Soil Sur- 
veys." USDA Agric. Handbook No. 436, U.S. Govt. Print. 
Off., Washington, B.C. 1975. 

2 Arnold, Richard W. "Soil Taxonomy, A Tool of Soil 
Survey." 

3 Hanna, Willis E. Personal Communications. July, 
1989. 



39 



40 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Table 1: Subgroups Permitted Or Required To Have An Aridic Moisture Regime In Orders Other Than Aridisols 



ALFISOLS 
Haplustalfs 

Aridic Haplustalfe 
Kandiustalfe 

Aridic Kandiustalfs 
Kanhaplustalfs 

Aridic Kanhaplustalfs 
Paleustalfs 

Aridic Paleustalfs 

Calciorthidic Paleustalfs 
Haploxeralfs 
Palexeralfs 

ANDISOLS 
Vitritorrands 

Lithic Vitritorrands 

Petrocalcic Vitritorrands 

Duric Vitritorrands 

Aquic Vitritorrands 

Calcic Vitritorrands 

Typic Vitritorrands 

ENTISOLS 
Torrifluvents 

Typic Torrifluvents 

Anthropic Torrifluvents 

Durorthidic Torrifluvents 

Durorthidic Xeric Torrifluvents 

Ustertic Torrifluvents 

Ustic Torrifluvents 

Vertic Torrifluvents 

Xeric Torrifluvents 

Torriorthents 

Typic Torriorthents 

Aquic Torriorthents 



Aquic Durorthidic Torriorthents 

Durorthidic Torriorthents 

Durorthidic Xeric Torriorthents 

Lithic Torriorthents 

Lithic Ustic Torriorthents 

Lithic Xeric Torriorthents 

Ustertic Torriorthents 

Ustic Torriorthents 

Vertic Torriorthents 

Xerertic Torriorthents 

Xeric Torriorthents 
Torripsamments 

Typic Torripsamments 

Durorthidic Xeric Torripsam- 
ments 

Lithic Torripsamments 

Ustic Torripsamments 

Xeric Torripsamments 

MOLLISOLS 
Argiborolls 

Abruptic Aridic Argiborolls 

Aridic Argiborolls 
Calciborolls 

Aridic Calciborolls 
Haploborolls 

Aridic Haploborolls 

Torrifluventic Haploborolls 

Torriorthentic Haploborolls 
Natriborolls 

Aridic Natriborolls 
Argiustolls 

Aridic Argiustolls 

Torrertic Argiustolls 



Calciustolls 

Aridic Calciustolls 
Torrertic Calciustolls 

Durustolls 

Aridic Durustolls 
Orthidic Durustolls 

Haplustolls 

Aridic Haplustolls 
Salorthidic Haplustolls 
Torrertic Haplustolls 
Torrifluventic Haplustolls 
Torriorthentic Haplustolls 
Torroxic Haplustolls 

Natrustolls 

Aridic Natrustolls 

Paleustolls 

Aridic Paleustolls 
Calciorthidic Paleustolls 
Torrertic Paleustolls 

Argixerolls 

Aridic Argixerolls 
Aridic Calcic Argixerolls 
Durargidic Argixerolls 

Calcixerolls 

Aridic Calcixerolls 

Durixerolls 

Aridic Durixerolls 
Orthidic Durixerolls 

Haploxerolls 

Aridic Haploxerolls 
Aridic Duric Haploxerolls 
Calciorthidic Haploxerolls 
Torrertic Haploxerolls 
Torrifluventic Haploxerolls 
Torriorthentic Haploxerolls 
Torripsammentic Haploxerolls 



Natrixerolls 

Aridic Natrixerolls 
Palexerolls 

Aridic Palexerolls 

Aridic Petrocaicic Palexerolls 

QXISQLS 
Acrotorrox 

Petroferric Acrotorrox 

Lithic Acrotorrox 

Typic Acrotorrox 
Eutrotorrox 

Petroferric Eutrotorrox 

Lithic Eutrotorrox 

Typic Eutrotorrox 
Haplotorrox 

Petroferric Haplotorrox 

Lithic Haplotorrox 

Typic Haplotorrox 

ULTISOLS 

VERTISOLS 
Torrerts 

Typic Torrerts 

Mollic Torrerts 

Paleustollic Torrerts 



seems prudent and does not necessarily indicate 
that the authors are advocating the change at 
this time. Major changes are necessarily evalu- 
ated, so that any changes made are in the best 
interest of soil science. 

Delete Aridisol Order 

If the moisture regime criteria of Aridisols 
were not considered at the order level, Aridisols 
would key out as Alfisols and Inceptisols. The 
two suborders of the order of Aridisols, Argids 
and Orthids, could be handled at a suborder 
level in the orders of Alfisols and Inceptisols re- 
spectively. The suborder Argids could be consid- 
ered a new suborder of Alfisols, "Torralfs," and 
the suborder, Orthids, could be considered a new 
suborder of Inceptisols, "Torrepts." 

Considering the aridic moisture regime at the 
suborder level in the orders of Alfisols and In- 
ceptisols would make Soil Taxonomy more con- 
sistent. The Aridic moisture regime would then 
be at the same level as it is presently in the or- 
ders of Andisols, Oxisols, and Vertisols. The de- 
letion of the Aridisol order would make all of the 
orders in Soil Taxonomy based on the presence 
or absence of soil properties that are assumed to 
reflect soil forming processes. 



Duripans and calcic, gypsic, natric, and salic 
horizons would remain at the great group level 
in the suborders of "Torralfs" and "Toirepts." 
This is consistent with the level these horizons 
and features are given in other soil orders. 

The great groups of the suborders, Torralfs 
and Torrepts, could remain the same as the 
great groups now in the suborders of Aridisols in 
Soil Taxonomy or could be changed as agreed 
upon by the ICOMID committee. 

The subgroups of the great groups of Torralfs 
and Torrepts, with the exception of those that 
intergrade to the order of Alfisols and Incepti- 
sols, respectively, also require no changes. The 
subgroups also could be changed as agreed 
upon. Table 2 lists the great groups and sub- 
groups of the proposed suborders of Torralfs and 
Torrepts. 

Broaden Aridisol Order 

If the proposal to delete the order of Aridisols 
is not acceptable to the committee, a proposal to 
bring all soils with an aridic moisture regime 
and diagnostic subsurface horizons into the 
Aridisol order could be made. 



ENGEL, WITTY, AND NICHOLS: ALTERNATIVE CLASSIFICATION OP SOILS WITH AHIDIC SOIL MOISTURE REGIMES 



41 



Table 2: Proposed Great Groups and Subgroups of the Proposed Suborders of Torralfs and Torrepts 



TORRALFS 
Duritorralfe 

(aquic) Duritorralfs 
Abruptic Xerollic Duritorralfe 
Abruptic Duritorralfe 
Haploxerollic Duritorralfs 
Haplic Duritorralfs 
Xerollic Duritorralfs 
(Ustic) Duritorralfs 
Typic Duritorralfs 
Haplotorralfe 

Borollic Lithic Haplotorralfs 

Lithic Ruptic-Entic Xerollic Haplotorralfs 

Lithic Xerollic Haplotorralfs 

Lithic Ustollic Haplotorralfs 

Lithic Haplotorralfs 

Borollic Vertic Haplotorralfs 

Borollic Haplotorralfs 

Xerertic Haplotorralfs 

Ustertic Haplotorralfs 

Vertic Haplotorralfs 

Aquic Haplotorralfs 

Arenic Ustollic Haplotorralfe 

Arenic Ustic Haplotorralfs 

Arenic Haplotorralfs 

Durixerollic Haplotorralfs 

Duric Haplotorralfs 

Xerollic Haplotorralfs 

Xeric Haplotorralfs 

Ustollic Haplotorralfs 

Ustic Haplotorralfs 

(Ruptic-Entic) Haplotorralfs 

Typic Haplotorralfs 
Nadurtorralfs 

Aquic Haplic Nadurtorralfs 

Aquic Nadurtorralfs 

Haploxerollic Nadurtorralfs 

Haplic Nadurtorralfs 

Xerollic Nadurtorralfs 

Typic Nadurtorralfs 
Natritorralfs 

Lithic Xerollic Natritorralfs 

Lithic Natritorralfs 

Borollic Glossic Natritorralfs 

Borollic Natritorralfs 

Aquic Natritorralfs 



Durixerollic Natritorralfs 
Duric Natritorralfs 
Glossic Ustollic Natritorralfs 
Haplustollic Natritorralfs 
Haplic Natritorralfs 
Xerollic Natritorralfs 
Ustollic Natritorralfs 
(Glossic) Natritorralfs 
Typic Natritorralfs 
Paletorralfe 

Borollic Vertic Paletorralfs 
Borollic Paletorralfs 
(Vertic) Paletorralfs 
Petrocalcic Xerollic Paletorralfs 
Petrocalcic Ustollic Paletorralfs 
Petrocalcic Ustalfic Paletorralfs 
Petrocalcic Paletorralfs 
(Duric) Paletorralfs 
Xerollic Paletorralfe 
Ustollic Paletorralfs 
Xeric Paletorralfs 
(Ustalfic) Paletorralfs 
Typic Paletorralfs 

TORREPTS 

Calcitorrepts 

Borollic Lithic Calcitorrepts 
Borollic Calcitorrepte 
Lithic Xerollic Calcitorrepts 
Lithic Ustollic Calcitorrepts 
Lithic Calcitorrepts 
Aquic Duric Calcitorrepts 
Aquic Calcitorrepts 
Durixerollic Calcitorrepte 
(Duric) Calcitorrepts 
Xerollic Calcitorrepte 
Ustollic Calcitorrepts 
(Xeric) Calcitorrepts 
Ustic Calcitorrepts 
Argic Calcitorrepts 
Typic Calcitorrepts 

Haplotorrepte 

Borollic Lithic Haplotorrepts 
Durixerollic Lithic Haplotonrepte 
Lithic Xerollic Haplotorrepts 



Lithic Haplotorrepts 
Natric Haplotorrepts 
Borollic Vertic Haplotorrepts 
Borollic Haplotorrepts 
Xerertic Haplotorrepts 
Ustertic Haplotorrepts 
Vertic Haplotorrepts 
Aquic Duric Haplotorrepts 
Aquic Haplotorrepts 
Durixerollic Haplotorrepts 
Duric Haplotorrepts 
Fluventic Haplotorrepte 
Anthropic Haplotorrepts 
Xerollic Haplotorrepts 
Ustollic Haplotorrepte 
(Xeric) Haplotorrepte 
Ustic Haplotorrepts 
Typic Haplotorrepts 

Duritorrepte 

Aquentic Duritorrepte 
Aquic Duritorrepte 
Haploxerollic Duritorrepte 
(Haplustollic) Duritorrepte 
Xerollic Duritorrepte 
(Ustollic) Duritorrepte 
(Ustic) Duritorrepte 
Typic Duritorrepte 

Gypsitorrepte 

Petrogypsic Gypsitorrepte 
Calcic Gypsitorrepts 
Cambic Gypsitorrepte 
Typic Gypsitorrepte 

Paletorrepte 

(Borollic) Paletorrepte 
Aquic Paletorrepte 
Xerollic Paletorrepte 
Ustollic Paletorrepte 
(Xeric) Paletorrepte 
Ustic Paletorrepte 
Typic Paletorrepte 

Salitorrepte 

Aquollic Salitorrepte 
(Haplic) Saltorrepte 
Typic Salitorrepte 



The key to the orders proposed by ICOMID on 
April 13, 1989, allows Aridisols to have a mollic 
epipedon. This change would cause all aridic 
Xerolls,with an argillic, cambic, calcic, petrocal- 
cic, gypsic, petrogypsic, natric, or a salic horizon 
or a duripan within 100cm, and many aridic 
Borolls and Ustolls, to be reclassified as 
Aridisols. This would affect about 16 of the 88 
subgroups with an aridic moisture regime in 
orders other than Aridisols. 

Alfisols with an aridic moisture regime would 
classify as Argids if the requirement that the 
epipedon of an Aridisol, with an argillic or natric 
horizon, can not be both massive and hard, or 
very hard when dry, were removed from the Key 
to Soil Orders of Soil Taxonomy. 

The remaining subgroups with an aridic 
moisture regime also could be brought into the 
order of Aridisols by placing Aridisols above 
Andisols, Oxisols, and Vertisols in the key to the 
orders or by not allowing an aridic moisture re- 



gime in these orders. The suborders "Andids," 
"Oxids," and "Vertids" could be added to the 
Aridisol order for those soils with an aridic mois- 
ture regime now in the orders of Andisols, 
Oxisols, and Vertisols, respectively. 

Summary 

In summary, classes at the order level are, in 
most cases, separated on the bases of properties 
that resulted from the major soil-forming proc- 
esses. The order Aridisols does not follow this 
principle. The order of Aridisols could be treated 
as suborders of Alfisols and Inceptisols with 
very little change to Soil Taxonomy. 
If the order of Aridisols is not deleted, a proposal 
could be made to bring all soils with an aridic 
moisture regime and diagnostic subsurface hori- 
zons into the order of Aridisols. 



42 SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



References 

Arnold, Richard W. "Soil Taxonomy, A Tool of Soil Sur- 
vey." International Committee on Aridisols, "Aridisols" 

Draft: April 13, 1989. Unpublished. 
Soil Survey Staff. "National Soil Taxonomy Handbook," 

430-VI, Issue No. 13. USDA, Soil Conservation Service, 

Washington, D.C., 1989. 
Soil Survey Staff. "Soil Taxonomy, a Basic System of Soil 

Classification for Making Soil Surveys." USDA Agric. 

Handbook No. 436, U.S. Govt. Print. Off., Washington, 

D.C., 1975. 
Witty, John E., and Richard W. Arnold. "Soil Taxonomy, 

an Overview." Outlook on Agriculture, Volume 16, No. 

1, 1987, Copyright 1987. Pergamon Journals Ltd. 

Printed in Great Britain. 



Soil Forming Processes in Soils with Cryic and Frigid Soil 
Temperature Regimes in Idaho 

MA. Fosberg, A.L. Falen, R.R. Blank, and KW. Hippie 1 



Abstract 

Idaho has extensive areas of Aridisols with cryic and frigid soil tempera- 
ture regimes. The pedogenesis of these soils is controlled by landforms and 
parent materials, periglacial activity and patterned ground formation, cli- 
mate, and plant community composition. 

Aridisols with cryic soil temperatures occur at high elevations mostly in 
east-central Idaho and developed principally in limestone alluvium/collu- 
vium with variable amounts of calcareous loess. Periglacial climate and 
patterned ground formations have strongly affected these soils, causing, 
among other changes, pronounced particle size discontinuities in the soil 
profile and spatial dependence of pedon properties at a very short scale. 
Duric horizons characterize pedons older than late-Pliestocene. 

Aridisols with frigid soil temperatures occur at lower elevations in east- 
central and southeastern Idaho and on the O wyhee Plateau of southeastern 
Idaho. Most formed in calcareous loess. Those that occur in eastern Idaho 
typically have A, Bw, Bk, and C genetic horizons and those on the Owyhee 
Plateau have A, Bt, Bk, and Bkqm genetic horizons. 



Introduction 

Idaho has extensive areas of Aridisols with 
frigid and cryic soil temperature regimes. They 
occur at elevations of 1400 to 1700 m and re- 
ceive 170 to 300 mm precipitation. Detailed cli- 
matic information is given by Hippie et al. 
(1987) for the cryic soils. They also occur in 
other western states, such as Wyoming, Utah, 
and Nevada, that have high elevations and 
similar climatic characteristics. In Idaho, they 
occur primarily in the east central part, from 
Salmon to Idaho Falls, and on the Owhyee Pla- 
teau in southeastern Idaho (Fig. 1) (Fosberg and 
McGrath, 1988; USDA-SCS, 1984). The acreage 
summary from the soil moisture and tempera- 
ture map by Fosberg and McGrath (1988) gives 
a total of 142,777 ha (352,803 acres) of Aridisols 
with cryic soil temperatures and 1,495,593 ha 
(3,695,611 acres) of Aridisols with frigid soil 
temperatures in Idaho. 

This paper discusses the factors and processes 
influencing the development and characteristics 
of cold Aridisols in Idaho. Most of the discussion 
concerns soils with cryic soil temperature re- 
gimes with some references to soils with frigid 
soil temperature regimes. The topics of discus- 
sion are: (1) general morphological and physical 
properties of selected soil series; (2) landform 



X M.A. Fosberg and A.L. Falen, Div. of Soils, Dept. of Plant, 
Soil, & Ento., University of Idaho, Moscow, ID; R.R. Blank, 
USDA-ARS, Reno, NV; K.W. Hippie, USDA-SCS State Office, 
Spokane, WA. Contribution from the College of Agriculture, 
University of Idaho, Research Paper No. 89759, Received 19 
Dec. 1989. 



and parent material relationships; (3) effects of 
periglacial climate and patterned ground on soil 
genesis; (4) characteristics and development of 
duripans; (5) chemical and physical properties of 
diagnostic horizons, habitat type communities, 
and soil climate relationships. The conclusions 
proposed are a synthesis of research not only by 
the authors but also by many other scientists. 

Results and Discussion 
General Morphology of Soils 

Aridisols with Cryic Soil Temperature 
Regimes 

Most cryic soils in Idaho formed in carbonate 
rich parent material derived from limestones, 
carbonatic mudstones and siltstones, and 
smaller amounts of calcareous loess. Several 
genetic horizon sequences represent the cryic 
soils (see Table 1). 

These soils all have similar solums with a 
thin calcareous A horizon and weakly developed 
Bk horizons. Differences among pedons occur in 
the degree of cementation, depth to cementa- 
tion, and coarse fragment content of the lower 
Bkq, Bqk, and Bqkm horizons. The texture in 
surface horizons is usually gravelly loam or very 
gravelly loam with increasing coarse fragments 
and sand with depth. Table 1 gives pH, organic 
carbon (O.C.), calcium carbonate equivalent 
(CCE), coarse fragment, and textural class with 
modifier for 5 representative cryic soils. 



43 



44 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Aridisols with Frigid Soil Temperature 
Regimes 

These Aridisols in the upper Snake River 
Plain are weakly developed The soil surveys of 
this area (USDA-SCS, 1973, 1979, 1981) do not 
describe any cemented horizons in these soils. 
Table 2 gives typical genetic horizon sequences, 
plus pH, O.C., CCE, and textural properties rep- 
resenting one soil from this area. The Aridisols 
formed in calcareous loess or calcareous sand 
and gravelly alluvium or in a mixture of these 
materials. They all are relatively weakly devel- 
oped with thin A and B horizons and are calcare- 
ous to the surface. Soils formed in deep loess, 
however, have an A, Bw, and Bk horizonation 
sequence. Extensive areas of these soils in the 
northern part of the Snake River Plain have 
been reworked by wind to form low incipient 
sand dunes aligned in a southwest to northeast 
direction. 

The Aridisols with frigid soil temperature re- 
gimes, on the Owhyee Plateau in southwest 
Idaho, are represented by a typical horizon se- 
quence of A-Bt-Bkq-2Bkqm. These soils are on 
very old landscapes. They have developed 
strong textural B horizons and strongly ce- 
mented silica duripans. Many have been influ- 
enced by periglacial climates and solifluction. 

Landform and Parent Materials 

Soils With Cryic Temperatures 

The rugged basin and range topography in 
east-central Idaho, where Aridisols with cryic 
soil temperature regimes occur (Fig. 1), is a por- 
tion of the Basin and Range Province which 
joins the north side of the Snake River Plain. In 
Fig. 1, the major cryic soil area is designated as 
1. It is composed of four northwest to southeast 
aligned mountain ranges and three broad river 
valleys. Along the east boundary is the Beaver- 
head Mountains of the Bitterroot Mountain 
Range (Continental Divide). The Lemhi and 
Lost River Mountain Range occupy the central 
ranges, where north to south, the Salmon, 
Sheep, and Whiteknob Mountains border on the 
west side of this area. The broad valleys where 
the Aridisols with cryic soil temperature re- 
gimes occur are occupied from east to west by 
the Lemhi, Pahsimeroi, and Salmon Rivers in 
the north and the Birch Creek, Little Lost, and 
Big Lost Rivers in the south. 

Fan terrace formation and glaciation 

These valleys are made up of extensive fan 
terrace systems characterized by transported 



Figure 1: Location Map on a 
soil moisture and tem- 
perature base map. 



1 Cyric Aridisol 

2 Frigid Aridisol 

3 Mesic Aridisol 




sediment eroded from the adjacent mountainous 
uplands. The sediment deposited in the fan ter- 
races is primarily poorly sorted carbonate grav- 
els derived from carboniferous limestones and 
some Devonian and Silurian dolomite rocks 
(Funk, 1977). 

A study by Funk (1977) on the formation of 
fans terraces in Birch Creek Valley and studies 
of glaciation in the adjacent mountains by Knoll 
(1973) and Dort (1969) place fan terrace system 
development during the Quaternary glacial 
chronologies of the area. The two major glacial 
episodes documented for this area are the Bull 
Lake and Pinedale. They have been dated by 
obsidian hydration by Pierce et al. (1976) at 
140,000 to 70,000 years B.P. and 70,000 to 
10,000 years B.P., respectively. The Bull Lake 
correlates with the late Illinoian and Sangamon 
and the Pinedale with the Wisconsin stages of 
the midcontinent glaciations. 

There also were pre-Bull Lake glacial epi- 
sodes (unnamed) and post-Pinedale episodes of 
Temple Lake (10,000 to 3000 yrs. B.P.) and 



FOSBEKG, FALEN, BLANK, AND HIPPLE: SOIL FORMING PROCESSES IN IDAHO 



45 



A 

Bkl 

Bk2 

Bk3 

Bkq 

2Bqk 



Neoglaciation (3000 to 1000 yrs. B.P.). 
Evidence for these glacial episodes is found 
in the moraine deposits at the mouths of 
canyons below glacial cirque formations of 
the mountain ranges. 

Fan terrace characteristics 

The configuration of the fan terrace sys- 
tem usually consists of several inset fan 
segments. This indicates that the overall 
gradient of each fan system has decreased 
with time. Studies by Fosberg, in coopera- 
tion with Funk (1977), show a correlation 
of increasing silica and carbonate cementa- 
tion and development of the soil duripans 
with increasing age of alluvial fan ter- 
races. This cementation ranged from pen- 
dants on coarse fragments, to very thin (1 
mm) laminar caps, to discontinuous ce- 
menting in the younger soils and fans, to 
continuous thick indurated cementation 
up to 30 cm thick in the soils on older fan 
terraces. The degree of soil development 
above the Bqk and Bqkm horizons was 
generally weak regardless of age of the fan 
terraces. 

Soil parent material 

The parent materials for most of these 
soils are the carbonate rich alluvium and 
the carbonate rich loess from the Snake 
River Plain. 

Soils With Frigid Temperatures 

The major areas of Aridisols with frigid 
soil temperature regimes occur in the large area 
joining the south side of Aridisols with cryic soil 
temperature regimes (in Fig. 1, designated as 
2), along the higher elevation fringes of the 
broad extensive area of mesic Xerollic Aridisols 
on the Snake River Plain that are designated by 
3, and on the northern fringes of the Owhyee 
High Plateau (Harkness, 1983). 

Landform and age 

The large area of Aridisols with frigid soil 
temperature regimes, at the east end of the 
Snake River Plain, in the area west and north of 
Idaho Falls and Blackfoot, are on a broad plain 
made up of a series of Pliocene and Pliestocene 
basalt flows, some as young as 2,000 and 4,000 
years B.P. Older basalt flows are capped with 
calcareous loess. On the north half, large areas 
of calcareous alluvium, coming from the Big 
Lost, Little Lost, and Birch Creek river valleys 
to the north, cover the older basalt flows. The 
Owyheee Plateau is dominated by welded tuff 
volcanics of the Miocene age. 



Horizon 


Depth 
cm 


PH 1 


O.C. 2 

% 


CCE S 

% 


>2mm 

% 


Texture 



Table 1. Representative Aridisol soil series with cryic soil 
temperature regimes and their genetic horizon sequence. 



Unnamed (mound) v. gr. loam: Loamy-skeletal, mixed, frigid Xerollic Calciorthids 
-h.t.-A.wyomingensis/Agropyron spicatum 



0-5 

5-15 

15-42 

42-61 

61-100 

100-140 



7.6 
7.9 
8.0 
7,6 
7.8 
8.3 



2.82 
1.96 
1.69 
1.61 
0.90 
0.13 



31 
32 
35 
36 
39 
70 



56 
43 
38 
58 
62 
86 



GRV-L 

GR-L 

GR-L 

GR-L 

GRV-L 

GRX-SL 



Leatherman v. gr. loam: Loamy-skeletal, carbonatic, frigid, shallow Xerollic 

Durorthids - h.t-A.arbuscula /Agropyron spicatum 
A 0-13 6.8 5.00 31 21 GRV-L 

Bk 13-28 7.4 1.74 39 32 GRV-L 

2Bqkm 28-40 7.7 0.54 80 86 Indurated duripan 

2Bqk 40-150 7.9 0.13 82 95 GRX-LS 

Arbus (intermound) gr. loam: Sandy-skeletal, carbonatic, frigid Xerollic Calci- 
orthids - h.t.-A,arbuscula/Agropyron spicatum 
A 0-10 7.6 2.13 2 47 GR-L 

Bk 10-23 7.6 1.63 36 63 GRV-L 

Bkql 23-33 7.9 0.88 66 84 GRV-SL 

Bkq2 33-43 8.4 0.69 70 84 GRV-SL 

Bkq3 43-76 8.6 0.16 76 88 GRX-LCOS 

Bkq4 76-152 8.6 0.11 71 86 GRX-COS 

Bluedome loam: Coarse-loamy, carbonatic, frigid Xerollic Durorthids - 
'h.t.-A.arbuscula/Agropyron spicatum 



A 

Bkl 
Bk2 
Bk3 
2Bqkm 
2Bqkl 
2Bqk2 



0-8 

8-28 

2843 

43-85 

85-115 

115-133 

133-150 



7.1 
7.3 
7.3 
7.2 
7.4 
7.7 
7.8 



3.22 
2.22 
1.48 
1.24 
1.48 
0.69 
0.45 



6 
17 
39 
58 
55 
55 
55 



11 

17 

14 

64 

232 

80 

79 



L 

L 

L 

L 

Indurated duripan 

GRX - COS 

GRX - COS 



Nitchly gr. silt loam: Loamy-skeletal, carbonatic Xerollic Calciorthids - 
\i.t.-A,noua/Agropyron spicatum 



A 


0-10 


7.3 


2.98 


15 


39 


GR - SIL 


Bkl 


10-18 


7.3 


1.83 


80 


45 


GR - SIL 


Bk2 


18-33 


7.2 


0.78 


41 


58 


GRV-L 


Bk3 


33-58 


7.6 


0.37 


73 


62 


GRV-L 


Bkql 


58-84 


7.5 


0.38 


54 


58 


GRV - CL 


Bkq2 


84-107 


7.4 


0.46 


88 


78 


GRX - CL 


Bkq3 


107-152 


7.5 


0.23 


62 


89 


GRX - CL 



*pH determined on saturated paste. 
"O.C. determined by K 2 Cr 2 7 digest and FeS0 4 titration. 
3 CCE determined by acid neutralization and subsequent back titration with a 
standardized base. 



Soil parent material and age 

The loess parent materials in south Idaho are 
extensive (Pierce et al., 1982; Lewis and Fos- 
berg, 1982; Blank, 1987). In southeastern 
Idaho, loess layers were dated by Pierce et al. 
(1976 and 1982) and correspond to the Bull 
Lake and Pinedale glacial episodes described in 
the previous section for soils with cryic tempera- 
tures. Pinedale loess is older than 15,000 yrs. 
B.P. because the Bonneville flood, dated about 
15,000 yrs. B.P., has scoured the loess along the 
Sanke River. The major source of this loess is 
attributed to the glacial outwash flood deposits 
of the Snake and other rivers. 

Effects of Periglacial Climates and 
Patterned Ground Formation 

Patterned ground is a term for the more or 
less symetrical forms such as circles, polygons, 
nets, stripes, garlands, and steps that are char- 
actersitic of, but not confined to, mantles sub- 



46 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



jected to intense frost action in periglacial 
areas (USDA-SCS, 1986). Patterned ground 
has been observed by the authors throughout 
Idaho and other western states and indeed 
occurs throughout the world (Krantz et al. 
1988). It is a relict of former periglacial cli- 
mates. It is best developed in close proximity 
to areas of glaciation at higher elevations. 

The most recent theory on the geophysical 
process in their formation is by Ray et al. 
(1983). Two excellent monographs (Price, 
1972, and Washburn, 1980) review the ex- 
tensive literature on patterned ground. In 
Idaho, three studies (Fosberg and Hironaka, 
1964; Fosberg, 1965; Malde, 1964) show the 
relationship of patterned ground to (1) its ori- 
gin from effects of cold periglacial climates, 
(2) the characteristics and genesis of soils, 
and (3) the effects of these soil properties on 
the distribution of sagebrush habitat types. 

Soil Variability and Periglacial 
Climates 

In the Aridisol area with cryic tempera- 
tures within the boundaries of a common 
landform, it appears that the effects of former 
periglacial climates and patterned ground 
formation account for much of the soil vari- 
ability, as illustrated by the representatives 
in Table 1. Potential variations in pedon 
properties that could result from the effect of 
periglacial climates are: variations of Bk 
verses Bkq or Bqk horizon sequences; vari- 
able degree of cementation; differences in depth 
to indurated duripans; thickness of the soluna; 
and the differences in particle-size distribution 
and gravel content in the A and B horizons. 

Areas with No Periglacial Influence 

In Idaho, Oregon, and Washington a major 
criteria necessary for patterned ground forma- 
tion is a restrictive layer (hardpan, claypan, 
bedrock) at shallow depths. In the major frigid 
area of the eastern Snake River Plain, many 
soils have developed in late Pliestocene or Holo- 
cene deep loess alluvium. Insufficient time for 
the formation of argillic or cemented pans pre- 
cludes the production of patterned ground. 

Areas With Periglacial Influence 

In the Owyhee Plateau these soils are 
strongly developed, having argillic and duripan 
horizons, and are often shallow to bedrock. Pat- 
terned ground is a common feature. 



Table 2. Representative Aridisol soil series with frigid 
temperature regimes and their genetic horizon sequence. 



Horizon 



Depth pH 1 
cm 



O.C. 2 CCE 3 >2mm 



Texture 



Pancheri silt loam: Coarse-silty, mixed, frigid Xerollic Calciorthids - 



h.t.-A wyomingensis JStipa thurberiana 



Al 

A2 

Bw 

Bkl 

Bk2 

Bk3 

Bk4 



0-5 

5-10 

10-30 

30-43 

43-71 

71-96 

96-112 



6.8 



7.7 
8.4 
8.7 
8.7 



112-162 8.5 



2.14 
1.42 
0.98 
0.79 
0.31 
0.19 
0.22 
0.11 









7 

24 

21 

19 

18 



TR 

















SIL 

SIL 

SIL 

SIL 

SIL 

SIL 

SI 

SI 



Grassy Butte loamy sand: Sandy, mixed, frigid Typic Calciorthids - 
h.t - A. wyomingensis /Stipa comata 

A 0-18 LS 

Bkl 18-48 LS 

Bk2 48-88 LS 

Cl 88-128 LS 

C2 128-150 LS 

Malm stony sandy loam: Coarse-loamy, mixed, frigid Xerollic Calciorthids - 
h.t. - A wyomingensis /Stipa comata 

A 0-10 ST-SL 

Bk 10-30 SL 

C 30-60 SL 

2R 60+ Basalt 

Whiteknob gravelly loam: Sandy-skeletal, mixed, frigid Xerollic Calciorthids - 
h.t.-A wyomingensis /Ag ropy ron spicatum 

A 0-13 GR-L 

Bkl 13-23 GR-L 

Bk2 23- 35 GR-L 

Bk3 35- 58 GRV - SL 

2C 58-150 GRV - S 

Heckison silt loam: Fine-loamy, mixed, frigid Xerollic Durargid. - 
h.t. -A wyomingensis / A. spicatum 



A 


0-10 


6.5 


1.22 





9 


SIL 


Btl 


10-23 


6.6 


1.09 





7 


SICL 


Bt2 


23-33 


7.2 


0.66 





13 


SICL 


Bk 


33-48 


8.0 


0.77 


26 


18 


SIL 


Bkq 


48-79 


7.9 


0.60 


33 


61 


GRV -SIL 


2Bkqm 


79-97 


- 


0.56 


66 


92 


GRX-SIL 



*pH determined on saturated paste. 
20.0. determined by K^Cr^ digest and FeSO 4 titration. 
'CCE determined by acid neutralization and subsequent back titration with a 
standardized base. 



Characteristics and Development of 
Duripans 

Duripans Intergrading with Petrocalcics 

General appearance 

Soils with frigid and cryic temperatures oc- 
curring in the horst and graben topography of 
east-central Idaho often contain subsurface 
cemented horizons. These pan-containing soils 
are dominantly located on fan terraces derived 
from limestones. Cemented regions consist of a 
clast-supported framework of rounded gravels 
and cobbles with variable amounts of matrix 
material. Pans are laterally continuous and 
range from 10 to 50 cm in thickness. Maximum 
moisture penetration into the soil profile likely 
controls the depth to the cemented layers as well 
as the parent material discontinuities. 

The appearance of cemented horizons 
changes with age. As time increases, pendants 



FOSBERG, FALEN, BLANK, AND HIPPLE: SOIL FORMING PROCESSES IN IDAHO 



47 



become thicker and begin to bridge individual 
framework clasts eventually coalescing under 
several clasts. The laminar cap becomes thicker 
and more indurated. In incipient and youthful 
pans, there is little matrix material among 
framework clasts and/or the matrix material is 
not cemented. In mature pans, void space 
among framework clasts is filled largely with 
matrix material, and the matrix often is indu- 
rated. 

Duripan criteria 

The large proportion of limestone clasts, and 
the fact that calcium carbonate is the dominant 
cementing agent, requires that these pans be 
considered intergrades with petrocalcics. We 
examined many pans in regard to duripan crite- 
ria as defined by Soil Taxonomy (1975), with 
variable results. Some pan fragments com- 
pletely dissolved in 1 N HC1, yet other frag- 
ments from the same sampling location re- 
mained partially intact. These findings were 
noted regardless of pan maturity. The residue 
after HC1 treatment, however, always contained 
opaline silica. Moreover, in thin sections, opal- 
ine silica was witnessed in laminar caps and in 
some pendants. The silica is undoubtedly caus- 
ing local cementation - bridging of particles - but 
much of it occurs as individual, discrete, silt- 
sized particles that strengthens the pans. The 
presence of this secondary opaline material 
qualifies the cemented horizons as duripans. 

Nature of cementation 

Three distinct regions of cementation occur in 
these duripans: the laminar cap, bridging of 
framework clasts by matrix material, and coa- 
lescing pendants. 

The laminar cap consists of laminations and 
striations, 0.04 to 0.10 mm thick, of alternating 
micritic calcite with brown microgranular cal- 
cite-op aline silica material (an intimate associa- 
tion of calcite with opaline silica). Scanning 
electron microscope observations shows the 
opaline silica occurs in web-like arrangements 
that both bridges particles and occurs as iso- 
lated units. In mature duripans, framework 
clasts often are cemented together through the 
bridging of matrix material. 

Most of the material dissolves with HC1 treat- 
ment, indicating that the dominant cementing 
agent is calcium carbonate. However, in the 
oldest duripans studied, regions containing 
microgranular calcite-opaline silica similar to 
that in the laminar cap were seen. It is not fully 
understood if the calcium carbonate cement was 
emplaced via percolating water or if in situ solu- 



tion-precipitation of limestone occurred in thin 
water films around particles to produce the ce- 
ment. 

It should be mentioned that a portion of the 
limestone is dolomitic and is, therefore, unstable 
in the pedogenic environment. Most dolomite 
clasts have a thin alteration rind of calcite. 
Many framework clasts have a pendant ema- 
nating from beneath (Blank and Fosberg, In 
press, a). At the lower boundary of the duripan 
these pendants often coalesce and, therefore, 
bridge two or more framework clasts. Calcite is 
the dominant component of pendants; however, 
thin concentric layers of opaline silica occur in 
some pendants. The opal creates a brownish- 
green coloration and adamantine luster that is 
evident in the field (Blank and Fosberg, In 
press, a). 

The role of eolian dust 

The eolian dust is an integral factor in the 
genesis of these duripans. The presence of eo- 
lian material is indicated by the large propor- 
tion of allochthonous minerals in the duripans 
(minerals such as certain feldspars, amphiboles, 
and pyroxenes) that could not have been derived 
from the parent fan terrace materials. The eo- 
lian dust serves as a source of silica, the mecha- 
nism of which this paper presents later. Eolian 
dust is a major component of the matrix mate- 
rial filling voids among the framework clasts. 

Eolian dust also may serve as a source of cal- 
cium carbonate for cementation. This state- 
ment may seem odd considering that limestone 
rock is a dominant parent material in the duri- 
pan-containing soils. However, it has been ob- 
served that the crystallinity or inherent chemi- 
cal composition of the limestone material makes 
it resistant to weathering. In dissolution experi- 
ments with HC1, limestones, even silt-sized frag- 
ments, dissolved much slower than dirty gray 
micritic calcite (eolian transported). 

The importance of eolian dust also is indi- 
cated by the mineralogy of the clay-sized frac- 
tion. Mica (illite), kaolinite, and quartz are the 
major clay-sized minerals in the carbonate-free 
residue of limestone. However, clay-sized resi- 
due of HCl-treated whole duripan fragments of- 
ten contain appreciable smectite. The smectite 
is brought in with eolian dust (Lewis and Fos- 
berg, 1982). 

Age and mineralogy relationships 

Incipient duripans (Holocene aged) and duri- 
pans as old as late-Pliestocene dominantly ARE 
smectitic in the clay-sized non-carbonate frac- 
tion. In older duripans (40,000 to 140,000 years 



48 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



B.P.), sepiolite and polygorskite assume impor- 
tance and smectite is altered to these minerals 
over time. 
Duripans of the Owyhee Plateau 

General appearance and micromorphology 

The best developed of these duripans directly 
overlay Miocene to Pliocene basalt on large 
shield volcanos. They also occur over more sili- 
ceous extrusive volcanics and on ignimbrites. 
Several lines of evidence suggest these duripans 
are distinctly different from the aforementioned 
duripans. Rather than occurring as continu- 
ously cemented layer, these duripans occur as 
thin plates 2 to 8 cm thick and 20 to 60 cm in 
diameter. The plates stack atop each other to 
create a thick indurated horizon. 

Soil forming processes have completely 
obsured any indication of an original fabric in 
these duripans. The pre-duripan landscape may 
have consisted of a framework of basalt frag- 
ments in filled with loess. Whatever the initial 
conditions were, the duripans now consist of a 
very complex, densely indurated, laminar-con- 
cretionary fabric whose principle minerals in- 
clude calcite and opaline-silica. 

Polvgenesis 

The micromorphology of these duripans sug- 
gests a complex genetic history. The arrange- 
ment of fabric components suggests these duri- 
pans are polygenetic, the summation of several 
episodes of soil truncation with subsequent 
readdition of loess. 

Three lines of evidence support this conclu- 
sion. First, zones of well-sorted loess agglomer- 
ates occur within the duripans, which could not 
have accumulated by any known subsurface soil 
pedogenic process. Convoluted laminar fabrics 
are seen in most thin sections of duripan plates, 
which are similar to the fabrics produced in arid 
regions by the action of lichens and cyano- 
bacteria at the soil surface. Finally, at one sam- 
pling location, a laterally extensive air-fall 
tephra layer occurs in the middle of the duripan 
plates, which clearly was deposited when the 
duripan or proto-duripan was surficially ex- 
posed. 

Mineralogy 

Calcite is the dominant mineral which occurs 
in these duripans. However, opaline silica may 
account for nearly 20 percent by weight of the 
duripan. The opal is largely opal-A, except for 
one duripan which was dominated by opal-CT 
(Jones and Segnit, 1971). All the duripans ex- 
amined contain sepiolite, which occurs as 



needles emanating from secondary concretions. 
The sepiolite is not well crystallized, and TEM 
micrographs of sepiolite laths often show opal 
spheres attached to the sepiolite with a gel-like 
substance. 

Source of silica 

It generally is agreed that volcanic tephra is 
important in the genesis of duripans in the west- 
ern United States (Flack et al., 1969, 1974). 
The rapid weathering kinetics of amorphous 
volcanic glasses contribute silica to the soil solu- 
tion, which then percolates downward and pre- 
cipitates near the wetting front to create the 
duripan (Chadwick et al, 1987). 

The incorporation of eolian dust into duripans 
and its subsequent alteration may contribute to 
the silification of duripans (Blank and Fosberg, 
In press, b). Dust is incorporated into calcare- 
ous pans via capture of loess agglomerates or by 
fracture fillings to form pedotubules. Once in- 
cased within the duripans, closed-system high 
pH conditions and increased surface Gibbs free 
energy, caused by the force of calcite crystalliza- 
tion, leads to rapid in situ alteration of the loess. 

The stable and end products of alteration in- 
clude opal-A and uncharacterized X-ray amor- 
phous materials. The abundance of composite 
particles and loess pedotublules in these duri- 
pans suggest that their contribution to the silici- 
fication of pans is of equal or greater magnitude 
than the commonly accepted mechanism of ad- 
dition via percolating Si-rich soil solutions from 
overlying soil horizons. 

Chemical and Physical Properties 

of Diagnostic Horizons, Habitat 

Type Communities, and Climatic 

Relationships 

Habitat Type Concept and Soils 

This discussion accepts that the sagebrush 
species, including Artemisia tridentata subspe- 
cies complex, are ecologically significant, and 
their distribution is related primarily to mois- 
ture, temperature, depth, and soil properties 
that are related to soil development (Hironaka 
et al. 1983). The habitat type (h.t.) classification 
concept used here is based on climax vegetation, 
where certain sagebrush and understory grass 
species, related to the environment, are domi- 
nant. 

However, it has been found (Hironaka et al., 
1983) that the requirements for the sagebrush 
and grass species do not always coincide. Each 



FOSBERG, PALEN, BLANK, AND HIPPLE: SOIL FORMING PROCESSES IN IDAHO 



49 



IPERATURE 

S - 

Is- ' 


MOISTURE 

. 2120 1 ftrUic 

:, X 


10 - A. vaseyana/Festuca idahoensis 
, . ^ Pachic Cryoboroll 
teflC O 9 - A. vaseyana/Festuca idahoensis- 
-^ ^^ Agropyron spicatum 
w^ Typic Cryobroll 
H .^ Calcic Cryobroll 
*^r^ 8 - A. tripartita/Agropyron spicatum- 
"j^r Festuca idahoensis 
^^^ Calcic Cryobroll 
i 7 - A. arbuscula/Agropyron spicatum- 
Festuca idahoensis 
Duric Cryoboroll 
6 - A. arbuscula/Agropyron spicatum 
Xerollic Durorthids 
5 - A. nova/Agropyron spicatum 


H 

Figure 2. EC 
habitat ty 


iy^ 1 


Xerollic Calciorthids 
4 - A. wyomingensis/Agropyron spicatum 
Xerollic Calciorthids 
Durixerollic Calciorthids 
3 - A. arbuscula/Agropyron spicatum 
Xerollic Durorthid 

l I l i - . ,. . . 


nic . ' s '- d - A. nova/ Agropyron spicaium 
1CIU Xerollic Calciorthids 
ological gradient showing moisture-temperature relationship to 1 - A wyomingensis/Stipathurberiana 
Ties and soils. Xerollic Camborthids 



will occupy certain positions of the environ- 
mental gradient. Thus, with each sagebrush- 
perennial grass combination, a more homogene- 
ous environment is delineated. It holds that in 
each environment, the potential to produce a 
particular h.t. is different, such as A. ar&^sca/a/ 
Agropyron spicatum versus A. wyomingensis I 
Agropyron spicatum. 

Figure 2 illustrates the entire ecological 
range found along an elevational gradient in the 
cryic and adjacent frigid areas. This ecological 
gradient shows the ecological position of each 
h.t. along with elevation, soil moisture, tem- 
perature, and classification of soils. 

Soil Climate and Habitat Types 

A research study in the Birch Creek Valley by 
Hauxwell (1977), in cooperation with M. Fos- 
berg and M. Hironaka, demonstrated that as 
elevation increased precipitation increased. The 
study dealt with soil characterisitics/sagebrush 
h.t. relationships. Precipitation ranged from 
118 mm for A wyomingensis h.t. to 293 mm for 
A. vaseyana h.t. for the period of September 
1975 to September 1976. Precipitation was 
smallest during winter and largest during sum- 
mer. For all precipitation gauges, 6% of precipi- 
tation was received mid-December to March, 
27% from March to mid-June, 48% from mid- 
June to mid-September, and 19% from mid-Sep- 
tember to mid-December. 



Hauxwell's (1977) study also determined that 
the date of moisture depletion to 1.5 MPa, as 
determined by thermocouple psychrometer 
readings, was, at least, by June 22 for A. wyo- 
mingensis, A. nova, and A. arbuscula, July 6 for 
A. tripartita, and July 20 for A. vaseyana h.t. 
The average summer soil temperature for all 
h.t. (Hauxwell's, 1977; Hippie, 1983; Hippie, 
1987 were within the cryic temperature regime. 

Soil climate studies by Hauxwell (1977) and 
Hippie (1987), ecological habitat type classifica- 
tion by Hironaka (1983) and an abundance of 
unpublished data 2 show a distinct ecological 
sequence for the h.t. as given in Figure 2. 

Genesis of Soil Properties 

Diagnostic soil properties, in addition to cli- 
mate, that interrelate soil genesis and classifica- 
tion and effect the distribution of the plant com- 
munities are: (1) ochric epipedons and the accu- 
mulation of O.C. as it affects the placement of 
soils in Xerollic subgroups, (2) accumulation of 
calcium carbonate and silica compounds to form, 
respectively, calcic horizons and duripans, (3) 
parent material discontinuities that affects 
moisture percolation, and thus, duripan devel- 
opment, (4) particle size distribution, and (5) 



2 Field and laboratory data by M.A. Fosberg and A.L. 
Falen, University of Idaho, Moscow, ID; SCS soil survey 
for Clark, Lemhi, Custer, and Butte Counties, Idaho; SCS, 
National Soil Survey Laboratory, Lincoln, NE. 



50 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



mound-intermound relationship to patterns of 

soil development. 

Ochric epipedon and organic carbon increase 

Progressing in an ecological sequence (Fig. 1) 
from Typic Aridisols supporting desert shrub 
h.t. (southwestern part of unit 3), through the 
sagebrush h.t. area in Aridisols with mesic soil 
temperature regimes (unit 3), through Aridisols 
with frigid soil temperature regimes of the 
Snake River Plain (unit 2), and into the cryic 
soils of Aridisols (unit 3), the 0. C. gradually 
increases with decreasing temperatures (Tables 
1,2,3). Figure 2 shows these relationships. The 
sagebrush-grass h.t.s associated with the soils 
above the Typic Aridisols are represented by 
Xerollic intergrades. 

Many of the Aridisols with frigid soil tempera- 
ture regimes, such as Pancheri (Table 2), have 
formed Bw or cambic horizons because they oc- 
cur outside the area of periglacial influence, 
whereas, Aridisols with cryic soil temperature 
regimes have been mixed by solifluction and 
lack a well defined cambic or Bw horizon. In 
southern Idaho, cambic horizons appear to rep- 
resent a younger stage of soil development be- 
cause similar soils on older landforms often have 
argillic horizons. 

Most of the Aridisols with frigid soil tempera- 
ture regimes, in the upper Snake River Plain, 
are relatively deep and, therefore, support A. 
wyomingensis. In other areas of Aridisols with 
frigid soil temperature regimes, where restrict- 
ing layers exist at a shallow depth, soil profiles 
support A. arbuscula. 

Carbonates 

The most striking feature of the Aridisols 
with cryic soil temperature regimes is the abun- 
dance of carbonate, which contributes to the for- 
mation of calcic horizons, to the cementation in 
the duripans, and the carbonatic family classifi- 
cation (Table 1). High amounts of carbonate in 
the ochric epipedons in these soils come from 
eolian dust and are also related to the high lime- 
stone gravel content and mixing by solifluction 
from periglacial climates. 

In Aridisols with frigid temperatures devel- 
oped in loess, carbonates have leached from the 
A and Bw horizons to 30 cm (Table 2). The ma- 
jor source for these carbonates is calcareous 
loess parent material. Carbonate accumulation 
in mesic soils (Table 3) is very comparable. 
Much of the carbonate source is also from the 
same age of Pinedale calcareous loess or the 
same age of sediments. 



Table 3. Representative Aridisoi soil series with mesic soil 
temperature regimes and their genetic horizon sequence. 



Horizon Depth 
cm 



pH 1 O.C. 2 CCE 3 >2mm Texture 



Greenleaf silt loam: Fine-silty, mixed, mesic Xerollic Haplargids - 
h.t.-A wyomingensis / Stipa thurberiana 



Ap 
Btl 
Bt2 
2Bkl 
2Bk2 
2Bk3 
3Bk 



0-25 

25-31 

31-38 

38-46 

46-61 

61-91 

91-152 



7.8 
7.9 
8.0 
7.5 
7.2 
8.2 
8.2 



0.91 
0.38 
0.32 
0.28 
0.23 
0.15 
0.16 



2 
2 
2 
9 
16 
14 
17 



SIL 
SIL 
SIL 
SIL 
SIL 
SIL 
SIL 



Turbyfill very fine sandy loam: Coarse-loamy, mixed, calcareous, mesic 
Xerollic Calciorthids - h.t-A wyomingensis I Stipa thurberiana 



Ap 0-23 

Bkl 23-51 

Bk2 51-69 

2C 69-89 

3C 89+ 



7.7 
8.0 
8.0 
8.1 
8.0 



0.71 
0.40 
0.26 
0.13 
0.05 



2 

16 
14 
8 
5 






5 
27 



VFSL 

L 

VFSL 

LS 

LS 



Seism loam: Coarse-silty, mixed, mesic Haploxerollic Durorthids - 
h.t.-A wyomingensis / Stipa thurberiana 
Al 0-4 7.4 1.49 L 

A2 0-11 7.3 0.63 L 

Bw 11-23 7.7 0.46 L 

Bkl 23-36 8.0 0.59 13 SIL 

Bkq2 35-64 8.1 0.40 16 SIL 

2Bkqm 64-91 8.5 0.28 11 L 

3C 91+ 8.4 0.18 3 FSL 

Unnamed silt loam: Coarse-silty, mixed, mesic Xerollic Haplargids - 
h.t.-A wyomingensis /Stipa thurberiana 



A 


0-8 


7.0 


0.69 


1 





SIL 


Btl 


8-13 


6.9 


0.64 


1 





L 


Bt2 


13-23 


7.1 


0.60 


1 





SIL 


Bt3 


23-43 


7.1 


0.50 


1 





SIL 


Bkl 


43-64 


8.1 


0.31 


29 





SI 


Bk2 


64-127 


8.4 


0.25 


21 





SIL 


Cl 


127-216 


8.3 


0.16 


15 





SI 


C2 


216-274 


8.0 


0.18 


16 





SI 



l pH determined on saturated paste. 

20.0. determined by K 2 Cr a O 7 digest and FeSO 4 titration. 

3CCE determined by acid neutralization and subsequent back titration 

with a standardized base. 



The exact effect of carbonate concentration 
near the surface on native plant communities is 
not known. However, free carbonates control 
phosphorus at low levels due to calcium phos- 
phate precipitates and low available micronutri- 
ents due to high pH. A. nova h.t. always occurs 
on these carbonate rich soils in Idaho (Table 1). 
If soils from calcareous parent material 
supporting A. nova change to acidic parent 
material, A. nova drops out. 

Pedon discontinuities and duripan formation 

Discontinuities caused by change in coarse 
fragment content and textural contrast affect 
soil hydrology. These textural discontinuities 
occur where strongly developed continuous duri- 
pans form. It is known that porosity difference, 
caused by abrupt textural changes, induced a 
build-up of soil water. With the relatively low 
hydrologic build up, water accumulates here 
and, upon evaporation carbonates and silica 
precipitate and accumulate the cement for duri- 
pan development. This hypothesis is verified by 
Hauxwell's (1977) soil moisture studies of the 
cryic soils. 



FOSBERG, FALEN, BLANK, AND HIPPLE: SOIL FORMING PROCESSES IN IDAHO 



51 



These discontinuities and duripan formations 
affect the distribution of the h.t. When they are 
at shallow depths, the soils support A. arbuscula 
and A. nova. At greater depths or when entirely 
absent, the soils support A. wyomingenesis, A. 
tripartita, and A. vaseyana. 

Periglacial influence on soil properties 

The highly variable coarse fragment content, 
and to a lesser degree, the silts in cryic soils are 
attributed to periglacial activity during the Pi- 
nedale glacial period. Periglacially mixed mate- 
rials are dominantly gravelly and very gravelly 
loams. Substratum textures are commonly a 
cobbly very gravelly loamy sand or sand. As 
noted previously, high percentages of gravels 
and cobbles appear at shallow depths in soils 
supporting A. arbuscula and A. nova. 

The variation in soil properties caused by the 
formation of patterned ground or mounds and 
intermounds causes contrasting complex pat- 
terns of vegetation. In mound-intermound se- 
quences these h.t. are found: Atriplex conferifo- 
lia/A. nova, A. wyomingensis I A. nova, A. wyo~ 
mingenesis I A. arbuscula (Hauxwell, 1977). The 
occurrence of these complex patterns of h.t. re- 
flects the same soil properties found in the 
broader, more extensive vegetation patterns 
associated with specific soils. 

Summary and Conclusions 

In Idaho, Aridisols with cryic soil temperature 
regimes are carbonate rich and show little evi- 
dence of clay translocation. These soils have 
developed on fan terraces and inset fan terraces 
formed from glacial outwash from limestone. 
The terraces range in age from Bull Lake 
(140,000 to 70,000 yrs) to Pinedale (70,000 to 
10,000 yrs) and younger. 

There are several horizon sequences repre- 
sented by several combinations of Bk, Bkq, Bqk, 
Bqkm, and C horizons. Contrasting pedon prop- 
erties, such as variable degree of duripan ce- 
mentation, depth to cementation, thickness of 
solum, variation in particle size distribution, 
and gravel content contribute to different taxa, 
horizon sequences, and distribution of plant 
communities referred to as habitat types (h.t.). 

Periglacial climates have resulted in the for- 
mation of patterned ground, or mound and 
intermound surface relief. These periglacial 
processes have had a major influence in the de- 
velopment of the contrasting pedon properties 
previously mentioned. The dominant soils oc- 
curring on the alluvial fans have duripans grad- 
ing to petrocalcics. 



Cemented regions consist of a clast-supported 
framework of rounded gravels and cobbles with 
variable amounts of matrix material. The ap- 
pearance of cemented horizons changes with age 
and are modified by periglacial effects. The na- 
ture of cementation in these duripans is charac- 
terized by three distinct regions of cementation: 
a laminar cap, bridging of framework clasts by 
matrix material, and coalescing pendants. 

The laminar caps are laminations and stria- 
tions of alternating micritic calcite with brown 
microgranular calcite-opaline silica material. 
Scanning electron microscope observations 
show the opaline silica occurs in web-like ar- 
rangements that bridge particles and occur as 
isolated units. This intimate association of cal- 
cite with opaline silica is also characteristic of 
mature duripans, but the cemented region is 
thicker with the framework clasts cemented to- 
gether. 

The major source for matrix material filling 
voids among the framework clasts is eolian dust. 
This is also a source for silica, as well as some 
carbonate, and is a source for smectite in 
younger pans, which later alters to sepiolite and 
polygorskite in older pans. 

The habitat type plant community classifica- 
tion used here is ecologically significant and is 
based on climax vegetation. The h.t.s are ar- 
ranged along on an elevational moisture and 
temperature gradient. Their distribution is also 
a response to variations in soil properties. The 
ecological distribution of sagebrush occurring in 
the h.t.s and corresponding soil taxa given in 
Fig. 2. are A. wyomingensis, A. nova, and A. ar- 
buscula on Aridisol soils and A. arbusula , A. tri- 
partita, and A. vaseyana are on Mollisol soils. 
The dates of moisture depletion to 1.5 MPa was 
June 22 for the h.t.s in the Aridisols, July 6 for 
the A. tripartita, and July 20 for A. vaseyana. 

Aridisols with frigid soil temperature regimes 
occur at lower elevations below cryic soils and 
above the extensive mesic soils of the Snake 
River Plain. The major area of frigid soils dis- 
cussed here occurs at the upper and east end of 
the Snake River Plain. This area is dominated 
by two ages of calcareous loess having its source 
from glacial outwash during Bull Lake and Pi- 
nedale glacial episodes. The northern edge is 
influenced by high carbonate alluvium from the 
limestone formations in the cryic soil area to the 
north. The loess soils have carbonate leached 
from the A and Bw into the Bk. These soils have 
no duripan development and have not been ef- 
fected by periglacial influences and patterned 



52 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



ground formation. The sagebrush species com- 
posing the h.t.s are the same as for cryic soils. 
However, A. wyomingensis is dominant but with 
different grass components on the loess soils. 

The O.C. in the ochric epipedon gradually in- 
creases when progressing from mesic through 
the frigid and cryic Aridisols. This is not evident 
from relatively uniform color of the ochric epipe- 
don; however, the thickness may increase. Both 
properties are probably masked by high carbon- 
ates but contribute to the xerollic subgroup clas- 
sifications of these Aridisols. 

The abundant carbonate accumulations from 
carbonate rich parent materials in the cryic soils 
is the most striking feature. It contributes to 
the formation of calcic horizons, duripans, and 
carbonatic family classifications. The effect of 
the discontinuities caused by concentration of 
coarse fragments appears to effect the soil's 
hydraulic properties and partially determines 
the position of duripan development. 

The solum depth and variations in particle 
size, distribution, carbonate content, and differ- 
ent degrees of cementation of non-indurated 
horizons was caused by periglacial climates and 
patterned formation. In the frigid areas, not 
influenced by periglacial climates, the formation 
of a cambic horizon represents a young stage of 
soil development correlating to less that 70,000 
years before present. 

Literature Cited 

Blank, R.R. 1987. Genesis of silica and calcite cemented 
soils in Idaho. Univ. of Idaho. Ph.D.Thesis (unpub- 
lished). 

Blank, R.R. and M.A. Fosberg. In press, a. Micromorophol- 
ogy and classification of secondary calcium carbonate 
accumulations that surround or occur on the under- 
sides of coarse fragments in Idaho (U.S.A.). Proceed- 
ings of the Internation Working Meeting on Soil Micro- 
morphology. 

Blank, R.R. and M.A. Fosberg. In press, b. Duripans of 
Idaho, U.S.A.: In situ alteration of eolian dust (loess) to 
an opal-A/X-ray amorphous phase. Geoderma. 

Blank, R.R. and M.A. Fosberg. 1989. Duripan of the 
Owyhee Plateau region of Idaho: Micromorphological 
evidence for polygenesis. In Agronomy Abstracts: An- 
nual Meeting of the Amer. Soc. Agron. 258 Abstract 

Chadwick, O.A., D.M. Hendricks, and W.D. Nettleton, 
1987. Silicification of Holocene soils in northern Moni- 
tor Valley. Nevada Soil Sci. Soc.Am. J. 53:158-164. 

Dort, W. 1969. Multiple glaciation of southern Lemhi 
mountains, Idaho. Preliminary reconnaissance report. 
Idaho St. Univ. Tebiwa, Vol. 5:2-17. 



Flach, K.W., W.D. Nettleton, L.H. Gile and J.G. Cady. 
1969. Pedocementation: induration by silica, carbon- 
ates and sesquoxides in the quaternary. Soil Sci. Vol. 
107:442-453. 

Flach, K.W., W.D. Nettleton, and R.E. Franklin. 1974. 
The micromorphology of silica-cemented soil horizons 
in western North America. In: G.K. Rutherford (Edi- 
tor), Soil Microscopy. Limestone Press, Kingston, On- 
tario, pp. 715-729. 

Fosberg, M.A. and M. Hironaka. 1964. Soil properties af- 
fecting the distribution of big and low sagebrush com- 
munities in southern Idaho. Am. Soc. of Agron., Forage 
Plant Physiology and Soil - Range Relationships Spec. 
Pub. No. 5. 

Fosberg, M.A. 1965. Characteristics and genesis of pat- 
terned ground in Wisconsin time in a chestnut soil zone 
of southern Idaho. Soil Sci. Vol 99:30-37. 

Fosberg, M.A. and C.L. McGrath. 1988. Idaho soil mois- 
ture and temperature map. Univ. of Idaho, Coll. of Agr. 
(unpublished). 

Funk, J.M. 1977. Climatic and tectonic effects on alluvial 
fan systems, Birch Creek Valley, east central Idaho. 
Univ. of Kansas, Ph.D. Dissertation (unpublished). 

Harkness, A.L. 1983. Silica cementation in soils on the 
Owhyee high plateau, Idaho. Soil Survey Horizons, Vol. 
24, No. 2:3-6. 

Hauxwell, D.L. 1977. Sagebrush soils study Birch Creek 
Valley, Idaho. Univ. of Idaho, College of Forestry, 
Range, Wildlife. Range Mgt. Spec. Res. Report (unpub- 
lished). 

Hippie, K.W. and G.P. Butler. 1983. Cryic Aridisols on 
outwash fans and terraces in Idaho. Soil Survey Hori- 
zons, Vol. 24, No. 3:28-33. 

Hippie, K.W., G.H. Logan and M.A. Fosberg. 1987. 
Aridisols with cryic soil temperature regimes. Proceed- 
ings of 1C OM ID IV. 

Hironaka, M., M.A. Fosberg and A. H. Winward. 1983. 
Sagebrush - grass habitat types of southern Idaho. 
Univ. of Idaho, Coll. For. Wildlife, Range Sci. Bull. No. 
35. 

Jones, L.H.P. and E.R. Segnit. 1971. Nature of opal:I, 
Nomenclature and constituent phases. T. Geol. Soc. 
Aust. 18:57-68. 

Knoll, KM. 1973. Chronology of alpine glaciation still- 
stands east-central Lemhi range, Idaho. Univ. of Kan- 
sas, Ph.D. Dissertation. 

Krantz, W.B., K.J. Gleason and N. Craine. 1988. Pat- 
terned ground, A common physical phenomenon shapes 
these uncommon manifestations of natural geometry. 
Scientific Amer. December, p. 68-76. 

Lewis, G.C. and M.A. Fosberg. 1982. Distribution and 
character of loess and loess soils in southeastern Idaho, 
in Bill Bonnichsen and R.M. Breckenridge, editors, 
Cenozoic Geology of Idaho: Idaho Bureau of Mines and 
Geology Bull. 26. 

Malde, H.E. 1964. Patterned ground in the western Snake 
River Plain, Idaho, and its possible cold climate origin. 
Bull. Geol. Soc.Am. 75:191-208. 



FOSBERG, FALEN, BLANK, AND HIPPLE: SOIL FORMING PROCESSES IN IDAHO 



53 



Pierce, K.L., J. D. Obrodovich and I. Friedman. 1976. 
Obsedian hydration dating and correlation of Bull Lake 
and Pinedale glaciations near West Yellowstone, Mon- 
tana. Geol. Soc. Am. Bull., Vol. 87:703-710. 

Pierce, K.L., M.A. Fosberg, W.E. Scott, G.C. Lewis and 
S.M. Colman. 1982. Loess deposits of southeastern 
Idaho: Age and correlation of the upper two loess units, 
in Bill Bonnichsen and R.M. Breckenridge, editors, 
Cenozoic Geology of Idaho:Idaho Bureau of Mines and 
Geology Bull. 26. 

Price, L.W. 1972. The periglacial environment, perma- 
frost and man. Assoc. of Am Geog. Resource Paper No. 
14. 

Ray, R.J., W.B. Rrantz, T.N. Caine and R.D. Gunn. 1983. 
A model for sorted pattern-ground regularity. 
Jour.Glaciology, Vol. 29, No. 102:317-337. 



USDA-SoilConservation Service.1973. Soil survey of Bing- 

ham County area, Idaho. 
USDA-SoilConservation Service. 1975. Soil Taxonomy. 

Ag. Handbook No. 436. 

USDA-Soil Conservation Service. 1979. Soil survey of Jef- 
ferson County, Idaho. 
USDA-Soil Conservation Service. 1981.Soil survey of Bon- 

neville County area, Idaho. 
UDSA-Soil Conservation Service. 1984. General soils map, 

Idaho. USDA, Soil Cons. Service, Boise, ID. 
UDSA-SoilConservation Service. 1986. Part 607, Glossary 

of Landform and geologic terms, pp. 607-1 to 607-45. 
Washburn, A.L. 1980. Geocryology: A survey of periglacial 

processes and environments. John Wiley and Sons, Inc. 



54 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Altocryic Aridisols in China 
Gong Zitong and Gu Guoan 1 

Abstract 

Altocryic aridisol is a suborder of Aridisols in the Chinese SoilTaxonomic 
Classification, which means Aridisols with a frigid or colder soil tempera- 
ture regime. Three great groups, i.e., frost desert soils, cold desert soils, and 
cryic calcic soils, are distinguished, which are concentrated in the 
Qingzhang Plateau of China. 

The Qingzhang Plateau, the especially high and steep Kulakunlen Mts. 
and Kulun Mts., is a virgin land which people yearn for and see as mysteri- 
ous. Until the end of 19th century only some scholars abroad and explorers 
such as Sven Hedin had set foot in the region to investigate geology, geogra- 
phy, and biology. Investigation groups from Germany, Austria, Soviet, Hol- 
land, and Italy engaged in multi-subject survey to Permir, the Karakorum 
Mts., and adjacent regions during the 1920s and 1950s. Liou Shene, a Chi- 
nese botanist, made an on-the-spot observation of flora in the line of the 
Karakorum Mts. and the Arkesai Basin of the Kunlun Mts. in 1932. After 
1949, the Comprehensive Scientific Investigation Group of the Qingzhang 
Plateau, Academia Sinica, carried out overall explorations in Tibet and the 
Hendan Mts. Furthermore, multi-subject surveys began to be undertaken 
in details in the Karakorum Mts. and Kulun Mts. region in 1987. 

This paper discusses the distribution, formation, diagnostics, types, and 
keys of Altocryic aridisols in this area. 



Soil Distribution 

Frost desert soils are mainly located in the 
north Kulun Mts. and the south Beiyangtang 
plateau, at elevations of over 5,000 m in the 
west and over 4,000 m in the east. The cold 
desert soils display on both sides of the moun- 
tains. The east side is the Chaidamu Basin, 
with a height of 3,000 m, and middle and low 
parts of the surrounding mountains, with 
heights of less than 4,000 m. The west side is 
river valley, lakesides, and mountains (<4,500 
m) on the southern Kalakunlun Mts. and on the 
northern Ayila and Gangdesi Mts. 

Cryic calcic soils are concentrated between 
the southern part of the frost desert soil region 
and the Gangdisi Mts., with elevations of 4,300 - 
5,300 m (Fig. 1), and on the north slope of the 
middle and western Himalayas, with elevations 
of 4,100 - 5,100 m. They also are found in the 
vertical belts of the Pamir Plateau within 3,500 
- 4,300 m, on the north slopes of the Kulun Mts. 
and Arjin Mts. at the height of 3,300 - 4,000 and 
3,800 - 4,200 m, respectively, and on the south 
slope of Tian Mts. at the height of 2,400 - 3,000 
m. 

Soil-Forming Characteristics 

The Qingzhang Plateau has lifted since the 
end of the Pliocene, resulting in high and broad 
relief, with an elevation of about 4,500 m and a 

Institute of Soil Science, Academia Sinica, Nanjing. 



total area of 2.4 million square km, amounting 
to 1/4 of the country. Moreover, the plateau is 
the highest, the most complex, and the youngest 
in the world. Above it, are rolling mountains, 
glaciers, frozen soils, and large basins. Because 
the mountains provide a natural defense for 
marine air current, the climate becomes drier 
and is characterized by cool temperatures in 
summer, very cold temperatures in winter, few 
rains, strong wind, and long frost periods. Due 
to relief differentiation the temperature zones 
can be divided into three types: the plateau arid 
cold zone, the plateau arid temperate zone, and 
the plateau semi-arid cold zone (Table 1). 

The Qingzhang Plateau was invaded by gla- 
ciers extensively in the Quaternary, which lead 
to weathering crusts such as infancy, clastic, 
saltbearing and clastic carbonate. 

As to vegetation, Certoides Compacta or 
mated by Carexmoorcroftii appears in the cold 
desert zone, Ceratodes mated by Kalidium 
Schrenkianum in the temperate desert zone, 
and Stipa Purpurea, S. Subsessiliflora Var. Ba- 
siplumosa etc. in the cold grass zone. 

Otherwise the soil is shown with shallow so- 
lum, coarse grain size, low clay content, and low 
biomas. It is infertile. Gypsum-salinization, cal- 
cification, and freezing and thawing play 
main roles in the soil-forming process. 

Gypsum-salinization 

Gypsum-salinization is an important soil- 
forming process of frost desert soils and cold 



GONG ZITONG AND Gu GUOAN: ALTOCRYIC AKIDISOLS IN CraNA 



55 



Table 1. The climate of Altocryic Aridisol region 


soil type 


elevation 
(m) 


temperature f'Cfl 
annul ave. 


rainfall 
>0C accum. 


wind 
(mm) 


(m/s) 


Frost desert 
soils 
Cold desert 
soils 
Cryic calcic 
soils 


4,900 
3,173 
4,415 


-7.7 
1.1 
0.1 


341.2 
1,947 
1,497 


24.0 
82.0 
166.1 


4.5 
2.1 



desert soils. These soils do not leach strongly in 
cold or temperate arid zones. Therefore gypsum 
and soluble salts are left or leached slightly. 
The soluble salts can move down or up over the 
gypsum horizon, because the former is more sol- 
uble to precipitate than the latter. In general, 
distribution of soluble salts in the profile de- 
creases downward in a T-shape, whereas gyp- 
sum concentrates under the vesicular crust hori- 
zon, with the highest concentration in the 
middle of the profile in a rhombus shape, which 
has thickness of 10-20 cm or even more than 30 
cm and a content of more than 15%, good crys- 
talline. 

Calcification 

Calcification also is an important soil-forming 
process of cryic calcic soils. Compared with the 
other two soil types, the leaching and biological 
cycle of these soils are stronger, resulting from 
more rainfall and vegetation. Salts and gypsum 
are mostly leached away and thus carbonate 
leaching and precipitation become prominent. 
Soft and white spots and new formations of hy- 
pha carbonate can be found in the calcic horizon. 
Nevertheless, the calcification is relatively weak 
because of the long frozen period, weak chemical 
weathering, and weak soil-forming process. 

Freezing and thawing 

It indicates all actions that water freezes and 
thaws alternately in soil horizons are bare 
rocks, which gives deep effect on frost desert 
soils forming. 

Marginal topography, such as stone-sea, 
stone-belt, and stone-ring, is an outcome of 
freezing and thawing. Soil in such areas is shal- 
low, more gravelly, and infertile. 

Another outcome is vesicular crust. When 
soils are frozen in the night during the warm 
season, vaporous water and capillary water 
move and condense from the subhorizon to the 
surface, but in the day the surface soil is 
thawed. C(X repeatedly extrudes and escapes in 
a horizontal direction, leading to the formation 
of vesicular crust. 



Diagnostic Horizon and Diagnostic 
Features 

Soil moisture regime is a leading factor gov- 
erning Aridisol formation. It therefore is taken 
as a main criterion in classifying soil groups 
(Table 2). 

In addition, carbonate accumulated horizon 
(calcic horizon), gypsum accumulated horizon 
horizon), and soluble salt accumulated 



Table 2. Identification of altocryic aridisols 



Soil group soil moisture regime soil temperature (C) 
(drying degree) annual average the hottest 



Frost ones 
Cold ones 
Cryic calcic 
ones 



3.5-10 



<0 
<8 

<8 



<5 



horizon (salic horizon), which are deeply related 
to soil moisture and temperature regimes are 
selected as main diagnostic horizons or charac- 
teristics for classifying soil subgroups. 

Calcic horizon 

A horizon accumulation of CaC0 3 (sometimes 
MgCO 3 ) appears in the B or C layer. In semi- 
arid regions with a little more rainfall and 
stronger biological activities, carbonates may be 
dissolved, leached, and precipitated into the 
horizon, which meets the following require- 
ments: 

(1)A thickness of 15 cm or more; 
(2)If carbonates in underside horizon are less 
than calcic horizon, carbonates in calcic 
horizon should be more at least 5% than C 
horizon (in absolute content); or 
(3)If carbonates in underside horizon are 
more than calcic horizon, visible secondary 
carbonates as concretions and powder 
limes in calcic horizon should more than 5% 
or more at least 5% than surface horizon (in 
absolute content). 

If the soil could not fully meet the above re- 
quirements but has carbonates leaching and 
accumulating features, it is considered calcic. 

Gypsic horizon 

A gypsic horizon is rich in secondary sulfate, 

mostly under an arid climate, and its parent 

materials are rich in gypsum. Gypsum crystal is 

coralloid in shape in a thin section (Photo 1). It 

meets following requirements: 

(1)A thickness of 15 cm or more; 

(2)Gypsum content is at least 5% more than 

that of the C horizon (absolute content), or 

the product of gypsum content in percent 

and the thickness of the horizon is not less 



56 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



than 150; if no gypsum is in underside 
layer, gypsum in the 30 cm of solum should 
be more than 5%, and if gypsum in under- 
side is 0-1%, gypsum in the 30 cm of solum 
should be more than 6%. 
If the soil could not fully meet the above re- 
quirements but has gypsum leaching and accu- 
mulating features, it is considered gypsic. 



Photo 1. Gypsic 
frost desert soils. 
(K88-63), depth 6- 
17 cm isochro- 
|matic polarized 
light X 20. 




Salic horizon 

A soil horizon is accumulated with salts more 
soluble than gypsum, resulting from relics of 
rock weathering, precipitation or surface water 
migration. It appears under or in the same hori- 
zon with gypsum, and, the more arid climate, 
the higher position and salt content it displays. 
It meets flowing requirements: 

(1)A thickness of 15 cm or more; 

(2)Salt content is at least 2% or more but less 
than 50%.; 

(3)The product of salt content in percent and 
its thickness in is 60 or more. 

If the soil could not fully meet above require- 
ments but has salts leaching and accumulating 
features, it is considered salic. 

Freezing and thawing feature 

In frost desert regions, especially near ice 
margins, there are disturbed forms such as 
stone-rings, heaving hills, lobate mud flows and 
takirs, etc. A vesicular crust appears in the sur- 
face and a squamose structure in the middle or 
low part of the cryic calcic soils. 

Micromorphologically, aggregates and sorting 
can be seen in Photo 2 and Photo 3 respectively, 
and a lot of fiber clay with optical orientation 
can be seen in Photo 4. 

Types 

Frost desert soils 

Frost desert soils are derived from clastic and 
salt-bearing weathering crusts, above the 
height of 4,000-5,000 m. The climate is cold and 
dry, belonging to the pergelic zone. In summer 





Photo 2 
(above). 
Aggregate in 
cryic calcic 
soils, 

isochromatic 
polarized light 
X20. 

Photo 3 (left). 
Sorting in 
cryic calcic 
soils, 

isochromatic 
polarized light 
X20. 

Photo 4 
(below). 

Gypsic frost 
desert soils, 
right-angled 
polarized 
light X 100. 




the mean air temperature is more than 0C, and 
freezing occurs in the evening. Altocryic desert 
plants are dominant. The coverage is less than 
5% or even bare. With bracteata around an ice 
belt, the land surface is covered by gravel or 
shows stone-ring and takyri of 10 cm. The basic 
soil-forming process is desertification, with the 
influence of freezing and thawing. Soil horizons 
are shallow, generally less than 50 cm, and 
abundant in gravels (more than 60%). Clay 



GONG ZITONG AND Gu GUOAN: ALTOCRYIC ARIDISOLS IN CHINA 



57 




0f 



r >T; 



XV Uj 



7 



.F* 



^ H U 

>. w g 

'^ ''" M , ^ 

O / Q W 5- 

; Q 



U 



? . 



\/i ^-- > V ^ 
.**"s.l\ 1 -' v ~-/ 



</> Uoo 



58 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Soil types 


depth 
(cm) 


Table 3. Physical and chemical properties of frost desert soils 

pH O.M. Tot.N KO gypsum CaCO 5 salts CEC Fe 2 O s (%) 
(H 2 1:2:5) (%) (%) (%) (%) (%) (%) (me/lOOg) free active 


clay 
<2um 


typical 
frost 
desert soil 


0-4 
4-16 
16-30 
30-55 


9.0 
8.7 
8.6 
8.7 


0.41 
0.63 
0.59 
0.50 


0.024 
0.048 
0.043 
0.027 


2.36 0.51 
2.71 0.50 
2.71 0.66 
2.45 0.82 


7.74 
3.01 
1.96 
4.13 


0.019 
0.017 
0.015 
0.011 


1.79 
3.95 
3.29 
2.75 


1.69 
1.80 
2.16 
1.86 


0.20 
0.26 
0.26 
0.21 


6.1 
14.9 
10.7 
7.6 


Gypsic frost 
desert soil 


0-2 
2-6 
6-23 
23-40 
40-60 


8.6 
8.6 
8.6 

8.7 
8.8 


0.31 
0.42 
0.45 
0.37 
0.19 


0.020 
0.025 
0.025 
0.023 
0.013 


2.31 11.52 
2.55 4.80 
1.65 11.85 
2.00 8.70 
2.36 3.88 


7.99 
10.24 
7.99 
11.98 
10.94 


1.008 
1.152 
1.152 
0.988 
0.965 


1.83 
3.22 
2.34 
2.44 
2.19 


0.33 
0.45 
0.34 
0.37 
0.43 


0.07 
0.12 
0.06 
0.09 
0.06 


4.8 
9.1 
1.6 
1.7 
2.8 


Takir Frost 
desert soil 


0-8 
8-20 
20-32 
32-60 


8.7 
9.2 
9.2 
9.1 


1.15 
0.77 
0.45 
0.37 


0.049 
0.035 
0.028 
0.027 


1.88 7.10 
1.94 3.05 
2.15 4.87 
2.23 2.56 


25.13 
26.90 
23.00 
31.12 


0.425 
0.131 
0.070 
0.047 


3.61 
3.90 

4.01 
4.28 


0.63 
0.77 
1.00 
0.96 


0.37 
0.29 
0.75 
0.84 


19.8 
17.6 
17.6 
27.5 




Soil types 
(cm) 


Table 4. Physical and chemical properties of cold desert soils 

depth pH O.M. Tot. N K_0 gypsum CaCO 3 salts CEC Fe 2 O 3 (%) 
(H a 01:2:5) (%) (%) (%) (%) (%) (%) (me/lOOg) free active 


clay 
<2um 


Gypsic cold 
desert soil 




0-1 8.5 
1-6 8.5 
6-17 8.8 
17-28 8.7 
28-50 8.8 
50-70 8.8 


0.22 
0.26 
0.24 
0.28 
0.24 
0.20 


0.019 
0.019 
0.014 
0.010 
0.010 
0.010 


2.17 2.25 
2.13 6.33 
1.56 17.78 
1.48 23.13 
1.83 15.73 
2.41 5.11 


11.28 
11.28 
9.03 
6.25 
8.68 
9.37 


0.487 
0.815 
1.503 
1.627 
2.114 
0.870 


2.37 
3.07 
1.85 
1.93 

1.71 
2.00 


0.48 
0.60 
0.36 
0.41 
0.52 
0.60 


0.09 
0.10 
0.06 
0.06 
0.09 
0.10 


7.5 
15.4 
4.6 
4.2 
2.4 
0.9 


Salic cold 
desert soil 




0-3 8.6 
3-11 8.8 
11-30 8.9 
30-50 8.8 


0.73 
0.49 
0.27 
0.23 


0.021 
0.016 
0.015 
0.013 


2.28 1.76 
2.25 2.64 
2.46 2.09 
2.23 1.17 


12.88 
12.01 
12.53 
10.61 


5.555 

4.697 
1.739 
1.646 




0.51 
0.47 
0.43 
0.37 


0.10 
0.10 
0.08 
0.06 


11.8 
9.7 
8.5 
4.4 



content is less than 10% except for Takyri frost 
desert soils derived from lake-deposit. Capacity 
is 2-4 m,e./100g soil. The surface horizon has a 
vesicular crust of red brown color due to iron 
oxide found in the regions of especially dry and 
ancient, downward, gypsum accumulation hori- 
zons or where salt accumulation occurred. Clay 
and free and active iron contents in the depth of 
2-6 cm of gypsic frost desert soils are especially 
high (Table 3), 

O.M. accumulation is weak in frost desert 
soils, usually less than 0.6 percent dominated by 
fulvic acid. The soil is alkaline in reaction. E^O 
content varies from 1.6 to 2.7% in fine earth 
fraction and is more than 5% in clay fraction. 
Clay minerals are dominated by chlorite and hy- 
drous mica. Boron is abundant in the soil. 

The region of frost desert soils is not suitable 
for grazing because of the low productivity of 
grass and far distance from pastoral areas, but 
it is suitable for the breeding of wildlife such as 
Tibetan antelope and Asiatic wild ass. 

Cold desert soils 

Located on a plateau of temperate desert, cold 
desert soils have intensive desertification and 
salinization, while thaw and freeze reaction is 
relatively weak. The soil is covered by gravel, 
shows white salt, and has a shallow profile gen- 
erally less than 80 cm. Clay content varies de- 
pending on parent materials. CEC changes be- 



tween 2-4 m.e./100 g soil. The surface horizon is 
characterized by a vesicular crust several cm 
thick. The subsurface horizon has a relatively 
small change of moisture and temperature re- 
gimes. Therefore prompt mineral weathering 
has resulted in the highest contents of clay and 
free and active iron oxides (Table 4). Distribu- 
tion of gypsum, salt, calcium carbonate, and 
clay content in the profile are shown in Fig, 2. 

Most of the cold desert soils are used for ani- 
mal husbandry. Some have been used for agri- 
culture when the horizons are thick and irriga- 
tion is possible. 

Cryic calcic soils 

Cryic calcic soils are located in cold and dry 
regions. Compared with froze desert soils and 
cold desert soils, they are characterized by (1) 
relatively high rainfall of 100-300 mm; (2) vege- 
tation composed of perennial plants, having a 
relatively large coverage and (3) loess parent 
material with no stone and gravel in the surface 
horizon. Leaching and biological processes in 
these soils were strengthened. Soluble salts and 
most gypsum were leached out. Calcium car- 
bonate moved down and clayification evidently 
occurred. Semi-decomposed O.M. residual can 
be seen in the surface horizon. Fulvic acid 
slightly dominated in the humus. CEC varies 
from 3 to 7 m.e./100 g soil, depending on clay 
and O.M. content (Table 5). Clay minerals were 



GONG ZITONG AND Gu GUOAN: ALTOCRYIC ARIDISOLS IN CHTNA 



59 



Soil types 



depth 
(cm) 



Greyic cryic 
desert soil 



0-11 

11-23 

23-60 

60-83 

83-110 



8.3 
8.6 
8.8 
8.5 
8.2 



Salic cold 
desert soil 



0-11 9.0 

11-25 9.0 

25-48 9.2 

48-70 9.0 



a 

<L> 

Q 



dominated by hydromi- 
cas, with some chlorite. 

Cryic desert soil re- 
gions are suitable for 
pasture but with two 
constraints. Pasture far 
from a village was un- 
der-used and pasture 
around a village was 
over-used. 

Key to Altocryic 
Aridisols 

Key to suborder 

Aridisols that have a 
frigid or more cold soil tem- 
perature regime 

Altocryic Aridisol 

Key to groups and 
subgroups 

Altocryic Aridisols that 
have a pergelic soil tem- 
perature regime, vesicular 
crust, and freezing-thaw- 
ing disturbed characteris- 
tics 

Frost Desert Soil 

Frost Desert Soils that 
meet following require- 
ments: 

(l)having no salic hori- 
zon, with upper 
boundary within 20 
cm from soil surface; 

(2)having no gypsic horizon or gypsum accu- 
mulation features, with upper boundary 
within 20 cm from soil surface; 
(3)having no takir feature caused by freezing- 
thawing 

- Typical Frost Desert Soil 

Frost Desert Soils that meet above require- 
ments except (1) 

- Salic Frost Desert Soil 

Frost Desert Soils that meet above require- 
ments except (2) 

- Gvpsic Frost Desert Soil 

Frost Desert Soils that meet above require- 
ments except (3) 



Table 5. Physical and chemical properties of cryic calcic desert soils 



P H O.M. 

(H 2 1:2:5) (%) 



TotN 



gypsum CaCO 3 salts 



CEC clay 

(me/lOOg) <2um 



2.44 
1.50 
1.75 
1.49 
1.78 



0.120 2.13 

0.082 2.18 

0.091 2.17 

0.075 2.02 

0.124 2.09 



0.53 
0.63 
0.78 
3.57 
3.90 



12.26 
12.94 
15.79 
13.94 
14.45 



0.397 
0.028 
0.032 
0.114 
0.313 



7.02 11.7 

6.19 11.7 

6.68 12.9 

5.51 13.6 

6.44 13.4 



1.21 
1.00 
0.31 
0.19 



0.070 1.48 

0.069 1.73 

0.027 1.61 

0.021 1.50 



0.27 
0.38 
0.25 
0.53 



11.14 
15.49 
15.31 
13.57 



0.027 
0.034 
0.026 
0.025 



3.22 6.8 

5.95 18.2 

3.41 15.6 

2.73 12.7 




Gipsurn Total Salt CaCO 3 Clay 



Fig. 2. Gipusm distribution in Gypsic frost desert soils 



- Takir Frost Desert Soil 

Other Altocryic Aridisols that have frigid sol 
temperature regime and vesicular crust 

Cold Desert Soil 

Cold Desert Soils that meet following require- 
ments: 

(l)having no salic horizon, with upper bound- 
ary within 20 cm from soil surface; 

(2)having no gypsic horizon or gypsic accumu- 
lation features, with upper boundary within 20 
cm from soil surface 

- Typical Cold Desert Soil 

Cold Desert Soils that meet above require- 
ments except (1) 



60 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



- Salic Cold Desert Soil 

Cold Desert Soils that meet above require- 
ments except (2) 

- Gypsic Cold Desert Soil 

Other Altocryic Aridisols that have cryic or 
more cold soil temperature regime and calcic 
horizon or calcic feature except for soils derived 
from non-carbonated parent material 

Cryic Calcic Soil 

Cold Calcic Soils that meet following require- 
ments: 

(l)having an ochric humus surface horizon 
that have a weighted O.M. content of less 
then 1% within 40 cm from soil surface; 

(2)having no secondary clayification subsur- 
face horizon and having cambic horizon; 

(3)having calcic feature and no calcic horizon; 

(4)having no calcareous reaction from soil 
surface to the bottom of B horizon 

- Tvpic Crvic Calcic Soil 

Cryic Calcic Soils that meet above require- 
ments except (1) 



- Grevic Cryic Calcic Soil 

Cryic Calcic Soils that meet above require- 
ments except (2) 

- Clayic Cryic Calcic Soil 

Cryic Calcic Soils that meet above require- 
ments except (3) 

- Calcic Cryic Calcic Soil 

Cryic Calcic Soils that meet above require- 
ments except (4) 

Calcareous Cryic Calcic Soil 

Main References 

Lei Wenjin, 1989. Classification of Aridisols in China. 
Soils, Vol. 21:2 (in Chinese) 

Gao Yixing et al., 1985. Soils of Tibet. Science Press, Bei- 
jing (in_Chinese) 

Wen Zenwang, 1965. Soi Geography of Xinjiang. Science 
Press, Beijing (in Chinese) 

Gong Zitong and Lei Wenjin, 1989. Aridisols of China. In 
the Proceedings of the Fourth International soil Corre- 
lation Meeting, pp. 111-119. USDA 

Chao Sungchiao, 1984. The sandy deserts and the Gobi of 
China. In Deserts and Arid Lands, pp. 95-114. Martins 
Nizhoff Publishers. 



Aridisols of Spain 

<L Herrero 1 and J. Porta 2 



Abstract 

This paper reviews the distribution and classification of Spanish 

Aridisols and discusses some taxonomic problems and major management 
practices. Little literature is available about Spanish Aridisols. Most of the 
information contained in this paper comes from soil maps and from the 
authors 9 unpublished data. The discussion covers the update of Soil Taxon- 
omy, the unsuitability of NewhalFs method for some Spanish Mediterra- 
nean areas, and the taxonomic status of the gypsiferous soils. 

Many of the soils classified as Aridisois (S.S.S., 1975) are now distributed 
into several Orders because the EC e criterion has been dropped at this hier- 
archic level. Recently, a new method for soil moisture regime calculation 
has been developed in Spain. This method improves the delineation of the 
aridic regions. Micromorphology provides a better knowledge of gypsifer- 
ous soils, and two kinds of gypsic horizons have been identified. Apart from 
marginal uses, the management of Spanish Aridisols is based on water sup- 
ply and/or water saving and on taking advantage of early cropping. 

The following conclusions can be made: (i) the new model of soil moisture 
regime calculations can be applied to the presumed aridic areas of Spain 
and calibrated with soil moisture measurements in the field, (ii) the inclu- 
sion of gypsiferous soils into three Orders disagrees with their distinct mor- 
phology and behavior, (iii) refinement in the gypsic horizon definition is 
needed to link microscopic features and field criteria, and (iv) both old and 
new management practices in Aridisols must be reported in detail for im- 
provement, the adaptation of new technologies, and extension of manage- 
ment practices to new areas. 



Introduction 

In areas with a Mediterranean climate in 
Spain, irrigation is often the only way for sus- 
tainable agriculture. Historical notes show that 

small irrigation canals (acequias) were built 
more than 2000 years ago; and some 31000 Km 2 

were under irrigation in 1984-85 (Leon and 
Delgado, 1988). 

Spain has great expanses subject to a 
semiarid climate, and early soil scientists such 
as Huguet del Villar (1929, 1950) and Kubiena 
(1953) created specific categories for Spanish 
soils associated with aridity. In many Mediter- 
ranean countries, soil moisture is limiting for 
agriculture, and the concept of Aridisol is useful. 
Although the management of arid soils is well 
known in Spain, their classification and map- 
ping according to S.S.S. (1975, 1987) or to FAO 
(1974) still poses significant problems. 

Physiological drought is the essential feature 
used in defining the Aridisols. They must lack 
plant available water for some reason, at least 



. of Soils & Irrigation, SIA-DGA, P.O. Box 727, 50080 
Zaragoza (Spain). Corresponding author. 

2 Dept. of Meterorology & Soil Sci., ETSI Agrdnomos, Rovira 
Roure 177, 25006 L&rida (Spain). 



for defined periods, and have pedogenic hori- 
zons. 

The initial concept of Aridisol included both 
water and salinity stress (S.S.S., 1960). The 
criterion of electrical conductivity (EC ) was in- 
troduced by the 7th Approximation (EC > 1 dS/ 
m at 25C) and was maintained in Soil Taxon- 
omy (S.S.S., 1975) (EC e > 2 dS/m) but has been 
eliminated recently (S.S.S., 1987). At present, 
emphasis is given to the soil moisture regime, 
and now the aridic regime is required for all 
Aridisols, with the exception only of salic hori- 
zon occurrence. 

Moisture as a for 

Definition of Aridisols 

The weakness in the estimation of soil mois- 
ture regime from data collected by meteorologi- 
cal observatories is apparent when the results of 
calculations are compared with the kinds and 
amounts of natural vegetation produced in ma- 
jor soil areas (Guthrie, 1985). The same dis- 
agreements were observed when applying 
NewhalFs model to predict the soil moisture re- 
gime in some Spanish semiarid regions (Lazaro 
et al., 1978; Porta et al., 1983; Alberto et al., 
1984; Jarauta, 1989). Spanish areas having a 
typical Mediterranean climate and vegetation 



62 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



are included in the ustic soil moisture regime 
after Newhall's method (Tavernier and Wam- 
beke, 1976). 

The model of moisture accretion and depletion 
developed for soils in the Great Prairies (S.S.S., 
1975; Newhall, 1976) uses mean monthly pre- 
cipitation and temperature to calculate the soil 
moisture regime, assuming a general available 
water capacity of 200 mm. The need for dichot- 
omy in the diagnosis criteria and the consequent 
changes in the calculations were pointed up in 
Spain (Elias and Ibanez, 1979; Gasco and 
Ibanez, 1979; Ibanez and Gasco, 1983). Very 
few field measurements of soil moisture con- 
tents are available around the world, and the 
same is true for Spain. 

Jarauta (1989) has investigated the soil mois- 
ture contents in some sites of northeastern 
Spain with an ustic soil moisture regime, using 
Newhall's method. These sites have Mediterra- 
nean vegetation, and their crop production is 
affected by soil moisture shortage during the 
plant growing season. A new model of moisture 
accretion and depletion has been proposed, after 
four years of field measurements of soil moisture 
by the gravimetric method. This model has 
been designed to allow for the incorporation of 
local climate and soil characteristics. 

Jarauta's model precisely defines the soil 
moisture regimes and allows different soil con- 
trol sections and water retention capacities as 
well as different rain dates during the month to 
be considered. This model approximates better 
than Newhall's model to field measurements of 
soil moisture. Its results allow for improved 
knowledge of the distribution of aridic soils in 
Spain, separating aridic-xeric from ustic re- 
gimes. In the Ebro basin (NE Spain), this has 
caused the ustic regime to be rejected reference 
to either by soil moisture effects on soil produc- 
tion or four years' measurement of soil moisture 
content in the soil moisture control section. 

The soils having argillic or natric epipedon 
are excluded from Aridisols if their epipedon is 
both massive and hard or very hard when dry 
(S.S.S., 1975, 1987). This is a useful criterion 
because the data about soil moisture regimes 
are often unavailable. 

Areas in Spain with an Aridic Soil 
Moisture Regime 

The application of Jarauta's method will al- 
low a better understanding of the true extent of 
the aridic regime in Spain. It can be hoped that 
this model will be applied to more observatories 



Table I. Diagnostic horizons in the Aridisols of Spain. 

Horizon Frequency 

Epipedons Ochric 
Endopedons Calcic 

Cambic 

Petrocalcic 

Gypsic and 

"Hypergypsic" 

Argillic 

Salic 

Natric 



and to sites where soil moisture data will be 
gathered. So far, the map by Lazaro et al. 
(1978) with some modifications can be used for 
Peninsular Spain. 

Because of the lack of field measurements, 
several authors have proposed pragmatic crite- 
ria in order to attempt to define the extent of the 
aridic soil moisture regime. Neither the selected 
site characteristic s nor their proposed values 
are agreed on by the different authors. The 
most common criterion is elevation (Diaz, 1987; 
Iniguez et al., 1988; Perez et al., 1987a) or ele- 
vation plus slope orientation (Alias et al., 
1987a,b, 1988; Torre and Alias, 1987). In other 
cases the distance from the Mediterranean Sea 
or longitude (Alberto et al., 1984; P6rez et al, 
1987b) also are included. In the future, satellite 
data (Milford, 1987) or hand-held sensors may 
furnish valuable data about the soil moisture 
content of bare or range soils. 

The Aridisols of Spain are distributed into 
three main regions, (i) the Ebro Valley in north- 
eastern Spain, (ii) the southeastern region of 
Spain, and (iii) small areas in the Canary Is- 
lands. The first two regions are the most arid in 
Western Europe. Their vegetation is quite spe- 
cific and contains species whose nearest locali- 
ties are in the Eastern Mediterranean or in 
North Africa. 

Taxa of Aridisols in Spain 

The recent dropping of the electrical conduc- 
tivity criterion (EC e > 2 dS/m at 25 C) forces a 
review of the bibliographical references of 
Aridisols. Soils having an EC e > 2 dS/m must be 
excluded from Aridisols if they do not fit in the 
aridic regime or if they do not have a salic hori- 
zon fitting the requirements for Aridisols. Most 
of the saline Aridisols after S.S.S. (1975) are 
now in several Orders, having their saline char- 
acteristics reflected at the phase level (Porta 
and Boixadera, 1988). 

An accurate interpretation of available data 
about Aridisols of Spain needs a definition of the 
salic horizon based on EC. In practice, salinity 
is measured as EC, and nomograms or approxi- 



HERRERO AND PORTA: ARIDISOLS OF SPAIN 



63 



mate calculations are used for conver- 
sion to salt percentage for classification 
purposes. The updating of the defini- 
tion of salic horizon should state an EC 
threshold in the standard extract at 
the soilrwater ratio of 1:5. 

Table 1 shows the estimated fre- 
quencies of diagnostic horizons in 
Spanish Aridisols, based on a biblio- 
graphical search and the authors' field 
experience. 

Silica cementations producing duri- 
pans do not occur in Spanish aridic ar- 
eas. Calcium carbonate, gypsum, and 
more soluble salts move, producing 
specific categories of soils. 

Salorthids are not common in salt- 
affected areas of Spain. Salorthids are 
associated with a shallow saline water 
table under an ascensional soil mois- 
ture regime; their vegetation is 
Arthrocnemum glaucum and Salicor- 
nia sp. Soils with salic horizon occur in 
small areas; they are mappable only at 
detailed scales and references to them 
are scattered in the literature. 

Table 2 displays the taxa of Spanish 
Aridisols and their location. The table 
was prepared after a critical review of 
the bibliography, and most of the refer- 
ences come from LUCDEME soil maps. 

Gypsiferous Soils: An 
"Erratic" Type of Soil 

In Spain, gypsiferous materials out- 
crop only in the east, having an extent 
of 35487 km 2 (Macau and Riba, 1965). 
Many of these outcrops are in the aridic 
regions. The soils developed from 
gyprock and other gypseous soils are 
well distinguished and easily sepa- 
rated from saline soils, both by farmers 
and by early soil scientists (Huguet del 
Villar, 1929). The Soil Taxonomy ap- 
proach allows reflection of the genetic 
and management specifities of gypsif- 
erous soils. 

Gypsiferous soils in former Soil 
Taxonomy approximations 

The soils enriched with "calcium sul- 
fate" were considered in the Soil Sur- 
vey Manual (S.S.S., 1951) and in the 
5th Approximation of Soil Taxonomy 
(Cline, 1979). 



Table 2. Great Groups of Aridisols cited in Spain. 


Great Group 


Region 


Province 


Area 


Reference 


Haplargids 


S.E. 


Almerta 


Las Negras 


Aguilar & al.,1973 








Campo de 


Martlnez-Raya, 








Dallas 


1987 








Macael 


Aguilar & al.,1987 








Roquetas 


Perez & al. f 1987 








Los Nietos 


authors 


Natrargids 


N.E. 


Huesca 


Fraella 


authors 






Zaragoza 


Bardenas 


Martlnez-Beltran, 










1978 




Canary 


Tenerife 


Tenerife 


Rodrtguez-Hdez.& 




Islands 






al., 1980 


Paleargid 


S.E. 


Almerta 


Rodalquilar 


Aguilar & al.,1973 








Finana 


Aguilar & al.,1987 








Roquetas 


Perez & al., 1987 








Tabernas 


Perez & al., 1987 






Murcia 


C. Cartagena 


Gisbert, 1973 


Calciorthids 


S.E. 


Alicante 


Maigmb 


Alias & Torre, 1987 






Almerta 


Nljar 


Aguilar & al., 1973 










Porta & al., 1980 








Finana 


Aguilar & al., 1987 








Macael 


Aguilar & al., 1987 








Tabernas 


Perez & al., 1987 








Guadix 


Ortega & al., 1988 






Granada 


Cullar 


Simbn & al., 1980 






Murcia 


C.Cartagena 


Gisbert, 1973 








Cehegin 


Alias & al., 1987 








Coy 


Alias & al., 1987 








Lorca 


Alias & al., 1988 








Puerto- 










Lumbreras 


Alias & al, 1988 




Canary 


Tenerife 


Tenerife 


Escobar & al., 1973 




Islands 






Fdez-Caldas & al., 1978 


Camborthids 


N.E. 


Navarra 


Bardenas 


Arricibita, 1987 










Iniguez & al., 1988 




S.E. 


Almerta 


Nljar 


Aguilar & al., 1973 










Porta & al., 1980 








Huercal- 










Overa 


Alonso, 1983 








Campo de 


Martlnez- 








Dallas 


Raya, 1987 








Roquetas 


Perez & al., 1987 






Murcia 


C.Cartagena 


Gisbert, 1973 








Lorca 


Alias & al., 1988 




Canary 










Islands 


Tenerife 


Tenerife 


Escobar & al., 1973 


Gypsiorthids 


N.E. 


Navarra 


Lodosa 


Arricibita, 1987 








Bardenas 


Iniguez & al., 1988 






Teruel 


Hljar 


Porta, 1986 




S.E. 


Alicante 


Maigmb 


Alias & al., 1987 






Almerta 


Tabernas 


Perez & al., 1987 






Murcia 





Sanchez &al.,1982 








Puerto 










Lumbreras 


Alias & al., 1988 








Lorca 


Alias & al., 1988 




Canary 










Islands 


Tenerife 


Gomera 


Jimenez & al., 1988 


Paleorthids 


N.E. 


Huesca 


Lanaja 


authors 






Lerida 


Suner 


Porta & al., 1983 




S.E. 


Almerta 


Nljar 


Porta & al., 1980 








Campo de 


Martinez- 






Dallas 


Raya, 1987 








Murcia 


Cehegin 


Alias & al., 1987 








Coy 


Alias & al., 1987 








Lorca 


Alias & al., 1988 








Puerto 










Lumbreras 


Alias & al., 1988 




Canary 










Islands 


Tenerife 


Tenerife 


Escobar & al., 1973 


Salorthids 


N.E. 


Zaragoza 


Bujaraloz 


Herrero, 1982 




S.E. 


Almerta 


Vega de 










Pulpl 


Alonso, 1983 








Campo de 


Martlnez- 








Dallas 


Raya, 1987 








Roquetas 


Perez & al., 1987 






Granada 


Cullar 


Simbn & al., 1980 






Murcia 





Sanchez &al.,1982 




Balears 


Majorca 


Alcudia 


Porta & al., 1987 




Central 


C. Real 


La Mancha 


Porta, 1975 






Toledo 


Ocana 


Gumuzzio & al., 1984 




South 


Huelva 


Donana 


Ayerbe & al., 1978 






Cadiz 


Puerto de 










Santa Maria 


Gomez & al., 1982 



64 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



The gypsic horizon was defined in the 7th 
Approximation (S.S.S., I960), but soils enriched 
with gypsum were included in the Aridisols, to- 
gether with soils having a calcic horizon, as 
Orthic Calcorthids. This situation was unsatis- 
factory, both from a genetic point of view and for 
land evaluation from large scale maps. 

The Great Group of Gypsiorthids was devel- 
oped in the Galley Proofs of Soil Taxonomy 
(1970-1973). The new Great Group was intro- 
duced in the order of Aridisols (S.S.S.,1975). A 
water solution saturated with gypsum (2.6 g/1 at 
25C) shows an EC > 2dS/m. Accordingly, all 
soils having a gypsic horizon (except those with 
a mollic epipedon) were included in the Gypsi- 
orthids. 

The present status of gypsiferous soils 

In the last revision of Soil Taxonomy (S.S.S., 
1987), the soils with a gypsic horizon can belong 
to three different Orders: those with a xeric 
moisture regime and a mollic epipedon are Cal- 
cixerolls; those without a mollic epipedon are 
Xerochrepts; and those with an aridic moisture 
regime are Gypsiorthids. This situation in- 
creases the heterogeneity of the Great Group of 
the Xerochrepts, because a new Subgroup must 
be created: Gypsic Xerochrepts. Both the cur- 
rent definition of Gypsiorthids and that pro- 
posed for Gypsids (ICOMID, 1989) are based on 
their moisture regime. This results in problems 
in the location of these soils because soil mois- 
ture calculations must be made from scarce field 
measurements of soil moisture. 

Soil moisture regime criteria may not be good 
enough to discriminate soil behavior and land 
use when a well developed, massive gypsic hori- 
zon is present. Most of these horizons may be 
classified as hypergypsic (ICOMID, 1989) and 
under xeric or aridic regimes are impenetrable 
for roots in the dry season (Porta et al., 1977). 
The placement of these gypsiferous soils into 
different Orders (Aridisols and Inceptisols) is 
not fully satisfactory. 

The need for refinement in the gypsic 
horizon definition 

The process of gypsum accumulation in the 
Aridisols can lead to highly gypsum-rich hori- 
zons. Field and micromorphological data about 
gypsic and petrogypsic horizons around the 
world has been reviewed by Herrero (1990). 
Tavernier et al. (1981) used the concept of hy- 
pergypsic, and Witty (1985) discussed this, 
pointing out that some relevant features such as 
subsidence or erosion that are currently associ- 



ated with the gypsic horizon can be explained by 
the gypseous substratum and identified as a 
phase. It can be concluded that there is a need 
for refinement of the gypsic horizon definition, 
and worldwide research is necessary in order to 
give it a broad scope. 

In the xeric and aridic soils studied in Spain 
(Porta and Herrero, 1990), the gypsic horizons 
in Gypsic Xerochrepts and in Gypsiorthids com- 
monly have similar micromorphological charac- 
ters. Moreover, the low water retention capabil- 
ity of gypsum enhances the arid conditions of 
these soils. 

The gyprock often produces a mass of mi- 
crocrystalline gypsum that can be richer in gyp- 
sum than the parent gyprock. So, a gypsum 
enrichment process can be accepted, although 
the causes of the formation of this kind of gyp- 
sum crystals remain unknown. This material 
either stands on the parental gyprock or moves 
along the slope as mud-flow. This weathering 
product with an upper epipedon may be identi- 
fied as a gypsic horizon from the morphology 
and fits the definition of hypergypsic horizon. 

Terms affecting the composition and quantifi- 
cation criteria are misused in some cases. Gyp- 
sum and calcium sulfate are misused as synony- 
mous, and even CaSO 4 (anhydrite) is employed 
instead of CaSO 4 2H 2 O (gypsum). 

Land Use of Aridisols in Spain 
Non irrigated lands 

The Aridisols are an important soil resource 
in Spain, but their moisture is usually too low to 
support rainfed agriculture. In dry farming, 
only winter cereals are possible, barley and 
wheat being the most common. Short duration 
cereals are often preferred because of drought at 
the end of each cycle. In some aridic areas in 
transition with xeric, crops of almond and olive 
trees are possible. In other areas, dry farming 
must be alternated with range for grazing 
sheep, or even only range may be possible. The 
marginal harvesting of plants such as Lygeum 
sparturn for fiber and other plants for soap has 
been abandoned. Recreation and wildlife could 
be other alternative uses for these lands. 

Irrigated lands 

Flooding is the traditional irrigation system 
in Spain, as in most places in the world. In the 
new irrigation districts of Spain, sprinkler irri- 
gation is commonly used for extensive crops, 
and trickle irrigation for fruit trees and vege- 
table crops. 



HERRERO AND PORTA: ARIDISOLS OP SPAIN 



65 



In the northeast aridic region (Ebro basin), 
water of good quality was abundant, and rice 
and drainage were used for salt removal. Not- 
withstanding, salt-affected Aridisols occur in 
some irrigated districts of the Ebro basin 
(Martinez-Beltrn, 1978; Herrero and Aragiifes, 
1988). Severe problems remain in soils whose 
high silt content, sodicity and adverse micro- 
morphological features make pipe drainage dif- 
ficult (Rodriguez et al., 1989). 

In the southeast aridic region, scattered plots 
began to be irrigated from small earth reser- 
voirs built for the prevention of flash floods due 
to storms (Leon et al., 1987). The increase in 
irrigated surface has led to some water shortage 
and quality problems. 

Special management practices 

After the end of the 19th century, a special 
soil management system for vegetable produc- 
tion was developed in the southeast of Spain 
and in the Canary Islands. Shortage of irriga- 
tion water and/or low water quality requires the 
use of this system, which is called "enarenado" 
(sand mulching). Many effects of "enarenado" 
have been cited (Martinez-Raya, 1987): (i) re- 
duction of water loss by evaporation, (ii) atmos- 
pheric water condensation, (iii) soil temperature 
regulation, (iv) advancement of harvest date, 
and (v) saving in fertilizer. Depending on local 
soil and climate circumstances or on market 
considerations, each of these effects can be deci- 
sive for the profitability of a crop. 

In two decades, a great surface area has been 
developed with polyethylene covered crops in 
the aridic zones of southeastern Spain (Leon 
and Delgado, 1988). This technique is often 
combined with "enarenado," and high profitabil- 
ity is being obtained with extra-early horticul- 
tural crops. 

Conclusions 

The broad distribution of Aridisols in Spain is 
well known. A more accurate discrimination 
between aridic and xeric in Mediterranean con- 
ditions has been attained with Jarauta's 
method of soil moisture regime calculation. The 
method was tested in the Ebro basin and could 
be useful in other areas using soil moisture 
measurements for control. 

Some gypsiferous soils, e.g., in central Spain, 
that were classified as Aridisols, belong now to 
Gypsic Xerochrepts. This approach underlines 
the differences in crop production, but field and 
micromorphological studies show a convergence 



in the morphology, behavior, and management 
of gypsiferous soils in Spain, in spite of their 
classification in different Orders. 

Advances in knowledge of the soil moisture 
regime of aridic soils must contribute to the 
planning of soil and water management in wide 
areas of Spain and to the transfer of agricultural 
technology. 

Literature Cited 

Aguilar, J. & al. 1987. Mapa de Suelos 1:100.000. Finana. 

LUCDEME. M.A.P.A. Madrid. 
Aguilar, J. & al. 1987. Mapa de Suelos 1:100.000. Macael. 

LUCDEME. M.A.P.A. Madrid. 
Aguilar, J., J. Fernandez, A. Garcia, and J. Miguez. 1973. 

Estudio edafico de la zona de Cabo de Gata. II. Estudio 

de Mollisuelos y Aridisuelos. An. Edaf. y Agrobiol. 

32:299-340. 
Alberto, F., J.L. Ami e, and J. MacMn. 1984. El clima de 

los suelos de la Cuenca del Ebro. I. Reg imenes de 

Humedad. Anales Estacion Experimental de Aula Dei 

17(l-2):7-20. 
Alias, L.J. & al. 1987a. Mapa de Suelos 1:100.000. 

Cehegin. LUCDEME. M.A.P.A. Madrid. 
Alias, L.J. & al. 1987b. Mapa de Suelos 1:100.000. Coy. 

LUCDEME. M.A.P.A. Madrid. 
Alias, L.J. & al. 1988a. Mapa de Suelos 1:100.000. Lorca. 

LUCDEME. M.A.P.A. Madrid. 
Alias, L.J. & al. 1988b Mapa de Suelos 1:100.000. Puerto 

Lumbreras. LUCDEME. M.A.P.A. Madrid. 
Alias, L.J., and A. de la Torre. 1987. Suelos de la Sierra 

del Maigm6 (Alicante), II: Descripcion de perfiles. 

Datos analiticos, clasificaciony distribucion. An. Edaf. 

Agrobiol. 46:909-930. 
Alonso, A. 1983. Estudio de los suelos de "El Saltador" 

(Huercal-Overa) y "La Vega" (Pulpi), en Aimer ia. Doc. 

Thesis, Univ. Madrid. 
Arricibita, F. J. 1987. Tipologia de suelos salinos en 

Navarra. Doc. Thesis. Univ. Navarra. 
Ayerbe, M. & al. 1978. Guia de las excursiones cientificas. 

VII Reunion Nacional de Suelos. Sevilla. CSIC-SECS. 
Cline, M.G. 1979. Soil Classification in the United States. 

Agron. mimeo N 79-12. Dept. of Agron. Ithaca. 207 p. 
Diaz, J.L. 1987. Estudio edfico y evaluacion forestal en la 

Sierra de los Filabres. I.A.R.A., Sevilla. 
Elias, F., and V. Ibanez. 1979. Comparacibn de dos mode- 

los matem aticos para estimar el regimen de humedad 

del suelo. An. INIA, Ser. General, 6:49-76. 
Escobar, F., and C. Roquero. 1973. Estudios agroedafol 6gi- 

cos de las areas con aptitud para producir tabaco en la 

isla de Tenerife. Min. de Agricultura, Madrid. 
FAO. 1974. Soil map of the world 1:5.000.000. Vol. I. Leg- 
end. UNESCO. Paris. 59 p. 
Fernandez-Caldas, E., and C.M. Rodriguez-Hernandez. 

1978. Suelos formados sobre materiales vole anicos (Is- 

las Canarias). Aridisoles: Natrargids, Calciorthids. An. 

Edaf. Agrobiol. 37:717-731. 
Gasco, J.M., and V. Ibanez. 1979. Criterios para la estima- 

cion del regimen de humedad de los suelos. An. INIA, 

Ser. General, 6:61-76. 

Gisbert, J, 1973. Estudio edafol ogico del Campo de Cart- 
agena. Desarrollo del Orden de los Aridisoles. Doc. 

Thesis. Univ. of Valencia. 



66 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



Gomez, V., J. Perez, and R. Roquero. 1982. The soils and 

water table properties of polders area Castillo de Do fia 

Blanca, Puerto de Santa Maria (Cadiz, Spain). Inter- 

nat. Polders Symp., The Netherlands. 

Gumuzzio, J., J.B. Alvarez, A. Gutierrez, and J. Guijarro. 

1984. Aproximacin a la capacidad de uso de los suelos 

de Ocafia (Toledo). An. Edaf. y Agrob. 43:93-111. 
Guthrie, R.L. 1985. Refinement of taxa based in soil mois- 
ture regimes. Proceed. 5th Int. Soil Class. Workshop: 

37-38. 
Herrero, J. 1982. Salinidad del suelo en salobrares de 

Monegros y Somontano oscense 'como condicionante de 

la vegetacion. Inst. Fernando el Catolico. Zaragoza. 
Herrero, J. 1990. Morfolog ia y genesis de suelos sobre 

yesos. INI A, Madrid, (in press). 
Herrero, J., and R. Aragiies. 1988. Suelos afectados por la 

salinidad en Aragm. Surcos deAragon 9:5-8. 
Huguet del Villar, E. 1929. Geobot anica. Ed. Labor, Barce- 
lona. 
Huguet del Villar, E. 1950. Geoedafolog ia. (Edition of 1983 

by the Univ. of Barcelona). 
Ibafiez, V., and J.M. Gas. 1983. Modificaciones al mod- 

elo matem atico propuesto por Newhall, P. para la esti- 

macion del regimen de humedad del Suelo. Comunica- 

ciones INIA. Serie Recursos Naturales. N. 18. 
ICOMID. 1989. Aridisols, version 6.0. Internat. Commit- 
tee on Aridisols. 
Ifiiguez, J., I. Sanchez-Carpintero, R. Val, F. J. Arricibita, 

M. Vidal, M. Garjon, and G. Vitoria. 1988. Mapa de 

Suelos de Navarra, Hoja 207 y 245. Dep. Edafologia. 

Univ. Navarra. 
Jarauta, E. 1989. Modelos matem aticos del regimen de 

humedad de los suelos. Aplicacion al area meridional 

de Lleida. Doc. Thesis. Univ. Politecnica de Catalufia, 

Barcelona. 
Jimenez Mendoza, C., A Rodriguez and M.L. Tejedor. 

1988. Suelos de la isla de Gomera (Canarias). II. An. 

Edaf. Agrobiol. 47:1159-1169. 
Kubiena, W.L. 1953. The soils of Europe. Thomas Murby 

& Co. London. 
Lazaro, F., F. Elias, and M. Nieves. 1978. Regmenes de 

humedad de los suelos de la Espana peninsular. 

Monograf ias INIA, 20. Madrid. 
Leon, A., F. del Amor, and A. Torrecillas. 1987. El riego en 

la region de Murcia. CSIC, Murcia. 
Leon, A. de, and L.F. Delgado. 1988, Memoria del Mapa de 

cultivos y aprovechamientos de Espana. M.A.P.A. 

Madrid. 
Macau, F., and 0. Riba. 1965. Situacom, caracteristicas y 

extension de los terrenos yesiferos en Espana. Serv. 

Geolgico de O.P., Madrid. 
Martinez-Beltr an, J. 1978. Drainage and reclamation of 

salt-affected soils in the Bardenas Area,Spain. ILRI 

Wageningen. 
Martinez-Raya, A. 1987. Comportamiento del riego bajo 

enarenado en invernadero. Doc. Thesis. Univ. Madrid. 
Milford, J.R. 1987. Problems of deducing the soil water 

balance in dryland regions from Meteosat data. Soil 

Use and Manage. 3(2):51-57. 
Newhall, F. 1976. Calculation of soil moisture regimes 

from the climatic record. USDA-SCS. 
Ortega, E. & al. 1988. Mapa de suelos 1:100.000. Guadix. 

LUCDEME. MAPA. Madrid. 



Perez, A. & al, 1987a. Mapa de Suelos 1:100.000. Roquetas 
de Mar. LUCDEME. MA.PA. Madrid. 

Perez, A. & al. 1987b. Mapa de Suelos 1:100.000. Taber- 
nas. LUCDEME. M.A.P.A. Madrid. 

Porta, J. 1975. Redistribuciones ionicas en suelos salinos: 
influencia sobre la vegetecio n haldfila y recuperaci on de 
suelos con horizonte gypsic 6 y otros suelos halmorfos 
del rio Gigiiela. Doc. Thesis. Univ. of Madrid. 

Porta, J. 1986. Edafogenesis en suelos yesiferos en medio 
semiarido. Unpubl., E.T.S.I.A. Lerida. 

Porta, J. & al. 1983. Los suelos de Catalu fia: 1. Area 
meridional de Lerida. Generalitat de Catalunya. Barce- 
lona. 

Porta, J. & al. 1987. Introducci 6 al coneixement dels s 61s. 
AEAC, Barcelona. 

Porta, J., and J. Boixadera. 1988. Suelos y salinidad en el 
valle del Ebro. In I. Romagosa & al. (eds.) The basis of 
Crop Production. AEAC, Barcelona: 177-201. 

Porta, J., and J. Herrero. 1990. Micromorphology and 
genesis of soils enriched with gypsum. Developments 
in Soil Science 19:321-339. Elsevier. 

Porta J., M. Lopez-Acevedo, and R. Dares. 1980. Los sue- 
los del Campo de Nijar. C.R.P. Aimer ia. 

Porta, J., M. Lopez-Acevedo, and C. Roquero. 1977. Morfo- 
metria y clasificacion de algunos Gypsiorthids en Es- 
pana. Anales INIA, 5:85-111. 

Rodriguez, J., J. Herrero, and J. Porta. 1989. Micromor- 
phological assessment of soil siltation risk indexes in a 
saline-sodic soil in Monegros irrigation district (Spain). 
Proc. Work. Meeting on Soil Micromorphology. San 
Antonio, TX, 1988. 

Rodriguez-Hernandez, C.M., C. Rodriguez -Pascual, andE. 
Fernandez-Caldas. 1980. Aridisoles formados sobre 
materiales vole anicos (Islas Canarias). Natrargids. An. 
Edaf. Agrobiol. 39:1415-1441. 

Sanchez, J.A., and F. Artes. 1982. Genesis, clasificacion y 
cartograf ia de los suelos de la region de Murcia. Caja 
Ahorros Provincial. 72 p. 

Simon, M., J. Aguilar, and C. Sierra. 1980. Los suelos 
halomorfos de la provincia de Grana da. II.- An. Edaf. 
Agrobiol. 39:101-120. Soil Survey Staff. 1951. Soil 
Survey Manual. USDA Handbook no. 18. 

Soil Survey Staff. 1960. Soil classification, a comprehen- 
sive system. Soil Conserv. Serv. U.S. Dep. Agr. U.S. 
Govt. Print. Off. Washington DC. 

Soil Survey Staff. 1975. Soil Taxonomy. USDA-SCS. Agric. 
Handb. 436. 

Soil Survey Staff. 1987. Keys to Soil Taxonomy (3rd print- 
ing). SMSS technical monograph n 6. Ithaca, N.Y. 
Tavernier, R., A. Osman, and M. Ilaiwi. 1981. Soil Taxon- 
omy and the soil map of Syria and Lebanon. Proc. 3rd 
Int. Soil Class. Workshop. ACSAD, Damascus: 83-93. 
Tavernier, R., and A. van Wambeke. 1976. Determinaoan 
del r egimen hidrico de los suelos de Espana segun el 
modelo matem atico de Newhall. Agrochimica 20(4- 
5):406-412. 

Torre, A. de la, and L.J. Alias. 1987. aracteristicas bio- 
climaticas de interes edafologico de la Sierra del 
Maigmo (Alicante, S.E. Espana). Anales de Biologia 
12:29-41. Univ. Murcia. 

Witty, J.E. 1985. Diagnostic horizons of carbonate and 
gypsum accumulations. Proc. 5th Int. Soil Class. Work- 
shop: 269-273. 



Aridisols of New Zealand 

A.E. Hewitt and EG. Beecroft 1 

Abstract 

The Aridisols of New Zealand (NZ) correspond closely with the order of 
Semiarid Soils established in the new NZ Soil Classification. Camborthids 
and Haplargids predominate, and the suborder and great group limits ac- 
cord well with classes established as important soil series. They occur in in- 
land grabens in the southern South Island, where mean annual precipita- 
tion ranges from 350 to 500 mm and the temperature regime is mesic. 

Argillic horizons carry many accessory characteristics but are difficult to 
identify in younger soils. Gamble horizons are also difficult to identify in 
massive, unstratified, non-calcareous loess. The distinction between Ustol- 
lic and Xerollic subgroups depends upon the ambiguous phrase "bordering 
on." 

Permeability is an important soil quality for irrigation interpretations. 
Diagnostic slowly permeable and rapidly permeable layers have proved use- 
ful for defining soil classes in NZ. 

NZ Aridisols are weakly weathered and have a low CEC; the resulting low 
buffering capacity makes them particularly sensitive to management. 
While this is true for NZ Aridisols, it is apparently not true for Aridisols in 
general. This distinction within the order could be made more explicit in 
Soi7 Taxonomy. 



Introduction 

Location and Significance 

Aridisols occur in the South Island of New 
Zealand in the basin and range province of Cen- 
tral Otago (Figure 1). They are restricted to the 
basins and lower hills, where they are sheltered 
by fault block ranges from the rain-bearing 
westerly and southerly winds. Marked precipi- 
tation and temperature gradients occur with 
altitude, and Aridisols pass into Ustochrepts at 
the margins of the basins, which in turn pass to 
Dystrochrepts and Cryochrepts up the moun- 
tain slopes. 

The Aridisols correspond almost exactly with 
the brown-grey earths of the NZ Genetic Soil 
Classification (Taylor and Pohlen, 1962) and the 
Semiarid Soils of the new NZ Soil Classification 
(Hewitt, 1989). 

Although they cover only 2270 km 2 , Aridisols 
are an important soil resource in the South Is- 
land. Large snow-fed rivers flow from the 
Southern Alps, cross the basins, and provide a 
source of high quality water for irrigation. 
Where water is unavailable, the traditional dry- 
land extensive grazing remains. The use of irri- 
gated land is diversifying from irrigated pasture 
to pip- and stone-fruit orcharding and horticul- 
ture. The climate provides the fruit industry 
with late-ripening, off-season products for ex- 
port to northern hemisphere markets. 

Land Resources, Private Bag, Dunedin, New 



HEW ZEALAND 




Figure 1. Location of Aridisols in the South Island of 
New Zealand. 



Climate 

Precipitation (mainly as rainfall) ranges from 
330 to about 500 mm/year. It is evenly distrib- 
uted throughout the year, although a slight 
maximum occurs in the summer. Potential 
evapotranspiration rates exceed 680 mm/year. 
The daily rate of potential evapotranspiration in 
summer typically ranges from 3 to 6 mm but 
may range from 6 to 10 mm during periods of 
desiccating north-west winds. 

All NZ Aridisols have a mesic soil tempera- 
ture regime. The mean annual soil temperature 
(at 30 cm depth) at Alexandra is 11.6 C, ranging 
from a mean monthly 3.0C in July to 19.3C in 
January. 

The Newhall water-balance model was ap- 
plied to data from two climate stations, Alexan- 



Zealand 



67 



68 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



dra and Cromwell (Watt, 1977). These gave 
similar results and comply with the require- 
ments of the aridic soil moisture regime be- 
cause: (1) there was a 75% probability that the 
moisture control section (MCS) is dry in all parts 
more than half the time that the soil tempera- 
ture at 50 cm is above 5C, and (2) there was 
only a 15% probability that the MCS is moist in 
some or all parts for as long as 90 consecutive 
days when the soil temperature at 50 cm is 
above 8C. 

Geology 

The basement rocks are acid quartzo-feldspa- 
thic schist and weakly metamorphosed sand- 
stone (greywacke) (Turnbull, 1987) containing 
no primary calcareous minerals. 

In the late Cretaceous, a deeply weathered 
peneplain was formed on the basement rocks. 
Warping in the Tertiary initiated terrestrial 
sedimentation of coal measures, sands, silts, 
and clays. Block faulting in the late Pliocene 
and Pleistocene formed the present basin and 
range topography. The Tertiary sediments re- 
main in parts of the basins but are mostly over- 
lain by Pleistocene glacial outwash deposits. 
Uplands were strongly modified by periglacial 
action, producing tors, shaved bedrock surfaces, 
and solifluxion debris. Localized areas of deeply 
weathered schist occur on the down- throw sides 
of faults as remnants of profiles weathered in 
the late Cretaceous. 

Soils and Land Units 

The Aridisols occur in 4 major land units. 
Within these units, Aridisol suborders and great 
groups accord well with classes established as 
NZ soil series. 

Terraces and Dissected Terrace Land 

Terraces of rounded gravelly sand alluvium, 
many metres thick, occur adjacent to the major 
rivers. Six distinct levels are recognised, but 
only 4 have significantly different soils (Figure 
2a). The terraces have been cut in Tertiary silts 
or clays. In lateral valleys, not directly affected 
by glacial outwash, the terrace alluvium is thin- 
ner. Irrigation in areas where it is less than 
about 1.5 m to the underlying sediments may 
cause waterlogging and salinization. 

Soils on successively higher terraces increase 
in age and show greater soil development 
(Leamy, 1973). Argillic horizons are absent on 
lower terraces and increase in thickness from 
intermediate to high terraces (Figure 2a). 



Younger argillic horizons are usually brown 
(10YR hue) and older ones are usually reddish 
brown (5YR hue). On high terraces the argillic 
horizons are continuous, but on intermediate 
terraces they are patchy where there has been 
Holocene fluvial activity. 

Early Pleistocene outwash deposits occur as 
finely dissected hills above the level of the high 
non-dissected terraces (Leamy 1972). Bedding 
in the gravels is tilted and reddish-brown argil- 
lic horizons are very thick (60-130 cm), but 
interrupted by younger gully fill deposits in 
which brown argillic horizons are absent or thin 
(20-30 cm). 

Fans 

The most extensive fans are of late Pleisoto- 
cene age and occur in association with the major 
terrace levels (Leamy and Saunders, 1967; 
Orbell, 1974). 

Three distinct parts known as the apex, mid, 
and toe (McCraw 1968) form concentric zones in 
most fans (Figure 2b). In fans from schist or 
greywacke, the three parts are dominantly 
sandy-skeletal, fine-loamy and fine-loamy, or 
fine-silty, respectively. Soils of the fan toes are 
frequently imperfectly or poorly drained, and 
drainage may be aggravated by irrigation. Fans 
from catchments in Tertiary sediments are more 
frequently clayey, alkaline, and saline in the 
toe. The concentric pattern is masked in fans or 
parts of fans that have been overlain by loess. 
Fan soils are more strongly stratified than ter- 
race soils and, where very stony, they are more 
likely to be loamy-skeletal than sandy-skeletal. 

Tertiary Sediments and Schist Saprolite 
Land 

Deeply weathered schist remnants of the for- 
mer Cretaceous peneplain surface occur on 
down-throw sides of major faults (Figure 3a). 
The saprolite materials are of minor extent but 
are important for an understanding of the soils 
and landscape of Central Otago (McCraw, 
1965). The kaolin component in soils from 
younger parent materials is thought to have 
been redistributed in the landscape from the 
weathered schist (Churchman 1978). The 
weathered schist is saline and would appear to 
be the ultimate source of the salts in the derived 
sediments. 

Tertiary sediments outcrop either as smooth 
contoured hills or in terrace scarps. They are 
dominantly clayey but silts, sands, and gravels 
also occur. The sand fraction and coarse frag- 
ments are dominated by quartz, and the clays 



HEWITT AND BEECROFT: ARIDISOLS OP NEW ZEALAND 



69 



CLARS 



L OWSLtfiH 








Figure 2(a) Terrace and dissected terrace land, and (b) Fan land, showing predominant soil series, their land- 
forms, and classification 



(a) Eraser series 
Molyneux series 
Walpuna series 
Lowburn series 
Clare series 

*m.m. mixed mesic 



Ustic Torriorthents, sandy-skeletal, m.m.* 
Ustollic Camborthids, sandy-skeletal, m.m. 
Ustollic Camborthids, sandy-skeletal, m.m. 
Ustollic Haplargids, loamy-skeletal, man. 
Ustollic Haplargids, clayey-skeletal, m.m. 



(b) Ardgour series, Ustollic Haplargid, Loamy- skeletal, m.m.* 
Waenga series, Ustollic Camborthic, Coarse-loamy, m.m. 
Annan series, Ustollic Haplargid, Fine-loamy, m.m. 
Blackmans series, Ustollic Haplargid, Fine-loamy, m.m. 
Galloway series, Typic Haplaquept, Fine-loamy, m.m. 



are predominantly kaolinite and derived from 
the weathered schist. A loess blanket occurs in 
places. The soils are slowly permeable Haplar- 
gids and in many cases are saline. Gypsum oc- 
curs in some soils where it has presumably 
weathered from pyrite (Lindqvist, pers comm.). 
The gypsiferous soils (Blakemore 1968) are the 
only NZ Aridisols which do not respond to 
sulphur fertilizers. 

Basement Rocks Land 

In summit, shoulder, and ridge positions, the 
soils are shallow Camborthids with lithic con- 
tacts. Rock outcrops are frequent with either 
distinctive "fretted" or "tor" landscapes 
(McCraw, 1965; Wood, 1969). On accumulating 
backslopes and footslopes, slowly permeable 
Haplargids occur with brown argillic horizons 
(Figure 3b). 

Review of Taxa 

The subgroups currently recognised in New 
Zealand are listed in table 1. 



Of the Argids, the Ustollic Haplargids are 
predominant. Ustalfic Haplargids occur where 
there has been erosion. Aquic Haplargids occur 
on the lower slopes of fans. Irrigation can in- 
duce the low chroma colours required to meet 
the morphological criteria for Aquic Haplargids. 
The Natragids are common in the Tertiary sedi- 
ment and schist saprolite land but occur only in 
small patches. 

Of the Orthids, the Ustollic Camborthids are 
predominant. Lithic Camborthids occur on 
ridge and crest sites in basement rock land. 
Aquic Camborthids occur on the lower slopes of 
young fans. The extent of the Gypsiorthids is 
unknown. The Salorthids occur in minor areas 
in toes of young fans. 

Taxonomic and Land Use Issues 

Soil Moisture Regime at Subgroup Level 

All NZ Haplargids and Camborthids fail the 
requirement for typic subgroups to be "dry in all 





Figure 3. (a) Tertiary sediments and schist regolith land and (b) Basement rock land, showing soil series, their 
landforms, and classification. * m.m. = mixed mesic and Lm. = illititic, mesic 

(a) Chapman series, Ustollic Camborthid, Fine-loamy, m.m. (b) Alexandra series, Ustic Camborthid, Coarse-loamy, m.m. 
Manor-burn series, Typic Natragid, Fine-loamy, m.m. Sonora series, Ustollic Camborthid, Coarse-silty, m.m. 

Becks series, Ustollic Haplargid, Fine, i.m. Conroy series, Lithic Camborthid, Coarse-loamy, m.m. 

Auripo series, Ustollic Haplargid, Fine, i.m. Hawksburn series, Ustollic Haplargid, Coarse-loamy, m.m. 



70 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



TABLE 1 
Suborder 


. Aridisol subgroups occurring in New Zealand. 

Great Group Subgroup Example 


Argids 


Natrargids 
Haplargids 


Typic Natrargids 
Ustollic Haplargids 
Ustalfic Haplargids 
Aquic Haplargids 


Manorburn 
Waenga 
Hawksburn 
Ranfurly 


Orthids 


Gypsiorthids 
Camborthids 

Salorthids 


Typic Gypsiorthids 
Ustollic Camborthids 
Lithic Camborthids 
Aquic Camborthids 
Aquollic Salorthids 


Becks 
Molyneux 
Conroy 
Linnburn 
Linnburn salty 



parts of the moisture control section for more 
than three-fourths of the time (cumulative) that 
the soil temperature is 5 C or more at a depth of 
50 cm unless the soil is irrigated." The Newhall 
model indicates only 30% probability that soils 
will meet the typic requirement (Hewitt & Watt, 
in prep.). 

Most Haplargids and Camborthids that are 
not severely eroded also fail the organic carbon 
requirement for typic subgroups. This leaves a 
choice between Ustollic and Xerollic subgroups 
which "... have an aridic moisture regime that 
borders on an ustic (or xeric) regime." The most 
appropriate assignment is unclear and hinges 
on the meaning of the words "borders on." If a 
geographic sense is intended, then an Ustollic 
subgroup is appropriate, because the Aridisols 
occur adjacent to soils with ustic moisture re- 
gimes. If a taxonomic sense is intended, how- 
ever, then an objective decision cannot be made 
(a problem also noted by Thomas et al., 1981, 
page F12-9). 

The soil moisture regime calculated for the 
Alexandra climate station is ustic-like to the 
extent that it meets (with 100% probability) the 
requirement for dryness 90 or more days cumu- 
lative each year. It is xeric-like to the extent 
that it meets (with 90% probability) the require- 
ment for dryness in the 4 months following the 
winter solstice (with only 20% probability that 
the MCS will be moist). Although winter rain- 
fall is particularly low, evapotranspiration is 
also very low. Moisture is stored, particularly in 
frozen topsoils, and becomes available for spring 
growth. 

Until the most appropriate subgroup assign- 
ment can be clarified, the soils provisionally 
have been assigned to ustollic subgroups, be- 
cause of their geographic association and their 
original grassland vegetation (a characteristic of 
ustollic subgroups noted by Guy Smith [Forbes, 
1986]). 



Argillic Horizons 

The argillic horizon has proved to be a signifi- 
cant horizon in NZ Aridisols. Soils with argillic 
horizons are slowly permeable, have low air 
capacity (and therefore potential aeration prob- 
lems under irrigation), high bulk density, and 
low macro-porosity. Clay contents vary from 12 
to 37%. They are a significant root barrier with 
high penetration resistance, and, although 
available water capacities (10-1500 kPa) are 
moderate, only a small amount is readily avail- 
able (10-100 kPa). 

Exchangeable Na percent (ESP) values vary 
from 1 to 29%. Horizons with ESP of 15% or 
more fail the structural requirements of Natric 
Horizons. 

Argillic horizons are easily recognized in the 
field in soils on Pleistocene land surfaces, be- 
cause of their texture and clay coatings. A mi- 
cromorphological study by Barrett (1971) con- 
firmed the presence of argillans in two such pro- 
files. 

Field identification of argillic horizons in soils 
on Holocene land surfaces is less certain. Ped or 
pore surfaces have moist colour value of 4 or less 
with olive brown (2.5Y 4/4) being a common 
colour. The cutans, however, are too thin to 
observe with a lOx hand lens and often show a 
dull lustre rather than the waxy lustre expected 
for clay coatings. The micromorphology of these 
horizons has not been described. Particle-size 
measurements usually meet argillic horizon 
requirements, although parent material stratifi- 
cation is indicated in many profiles. 

The soils have been provisionally assigned to 
the Haplargids because they have subsurface 
horizons with the accessory properties of argillic 
horizons. The identification problem has been 
handled in the NZ Soil Classification (Hewitt, 
1989) by grouping together all soils with slowly 
permeable horizons that have cutans (whether 
they be argillans or suspected organic-iron-clay 
complexes). 

Cambic Horizon 

The recognition of cambic horizons is subjec- 
tive in many NZ Aridisols. The soil parent ma- 
terials are non-calcareous, and calcareous dust- 
fall does not occur. Subsurface horizons contain- 
ing calcium carbonate do occur as a product of 
weathering and leaching (Leamy and Rafter, 
1972), and, in such soils, a calcium carbonate 
accumulation horizon establishes the lower 
boundary of the cambic horizon. 



HEWITT AND BEECROFT: AKIDISOLS OF NEW ZEALAND 



71 



TABLE 2. Selected data for three NZ Aridisols 


SOIL 


HOR. 


DEPTH TO 


CLAY 


CEC 


ORG.C 


(LAB NO.) 




BASE (cm) 


% 


me% 


% 


Ixwburn 


A 


11 


8 


6.1 


1.6 


(SB7584) 


Bw 


41 


4 


4.0 


0.3 




Bt 


64 


10 


9.3 


0.2 


Waenga 


A 


28 


14 


8.5 


1.9 


(SB9895) 


Bw 


45 


13 


4.7 


0.4 




Bt 


69 


21 


8.5 


0.3 


Drybread 


A 


13 


18 


6.6 


1.4 


(SB9892) 


AB 


28 


6 


5.3 


0.9 




Bt 


67 


12 


9.5 


0.3 



In many soils on Holocene land surfaces, 
however, calcium carbonate horizons are ab- 
sent, and chroma and hue do not change with 
depth. Cambic horizon recognition then must 
rely on "the presence of soil structure or absence 
of rock structure." Loess in Central Otago is 
non-calcareous, and even very young deposits do 
not show "rock structure" in the form of sedi- 
mentary layering. Furthermore, the soils with- 
out argillic horizons are commonly structure- 
less-massive. 

Permeability 

Permeability, or field-assessed saturated hy- 
draulic conductivity, is a particularly important 
quality for the interpretation of irrigable 
Aridisols. Under orchards using overhead sprin- 
klers for frost fighting, the occurrence of slowly 
permeable layers requires either deep ripping or 
surface recontouring, to off-farm drainage out- 
lets, and ridging of tree rows. The occurrence of 
rapidly permeable layers requires particular 
attention to water scheduling and lateral distri- 
bution of water. 

Instead of reliance upon surrogate properties, 
"slowly permeable" and "rapidly permeable" di- 
agnostic layers have been specified in the NZ 
Soil Classification (Hewitt, 1989). These layers 
are defined either by measured saturated hy- 
draulic conductivity or by morphology, using a 
combination of pedality, particle size, and a 
semi-confined single vane-shear test (Griffiths, 
1985). 

In NZ Aridisols, slowly permeable layers in- 

elude most argillic and natric horizons, as well 

as some cambic horizons. Rapidly permeable 

~ lasers include most cambic horizons in soils with 

sandy-skeletal family particle size class. 



Buffering Capacity 

The clay fractions of NZ Aridisols are domi- 

imated by mica/illite (approx. 50-70%) and ac- 

companied by kaolinite (approx. 10-20%). Minor 

amounts of smectite and interstratified chlorite- 



vermiculite occur in argillic horizons. Organic- 
carbon levels are low, and consequently the 
CEC is low, particularly in the most intensively 
used soils of the fans and terraces. 

Mean oxalate-extractable iron and alumin- 
ium values are very low (0.2 and 0.06% respec- 
tively for all horizons). Consequently, phosphate 
retention (mean of 8% for all horizons) and ab- 
sorbed sulfate (mean of 4 ug/g for A horizons) 
are also very low. 

The soils therefore are poorly buffered and 
many chemical and physical properties are sen- 
sitive to management. Particular care needs to 
be taken in management of fertility and soil 
structure. The following land-use problems 
have resulted. 

1. Inappropriate management advice can lead 
to the application of high rates of ammo- 
nium sulfate fertilizers as ground dressings 
under irrigated orchards. pH(H 2 soil: wa- 
ter = 1:4) levels have been depressed from 
around 7.0 to as low as 4.2, with subse- 
quent toxicities, yield reduction, and in- 
creased disease susceptibility. This can be 
avoided by small and frequent applications 
of fertilizer, taking care to balance and 
monitor pH levels. 

2.Fumigants also may depress pH. For ex- 
ample, one combined application of a fumi- 
gant and a nematicide caused a drop of 1 
pH(H 2 soils water = 1:4) unit in A and B 
horizons. 

3. Overhead sprinklers are used to counter 
spring frosts in stone fruit orchards. Top- 
soils become saturated and, with the im- 
pact of machinery, soil structure deterio- 
rates, infiltration becomes slow, and aera- 
tion is greatly restricted. This also in- 
creases susceptibility to diseases such as 
bacterial blast. 

4.Herbicide strips commonly are used to con- 
trol grass and weed growth under orchards. 
Root studies indicate that herbicides pene- 
trate topsoils and discourage root develop- 
ment in the upper 20 cm. This has particu- 
larly serious implications in sandy-skeletal 
soils where nutrients and available water 
are concentrated largely in topsoils. 
5. Serious erosion has occurred when soils are 
cultivated in windy conditions. Dispersion 
and slaking measurements (McQueen, 
1981) show very low aggregate stabilities 
in most soils. 



74 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



-14* 



-20* 



118' 



124' 



130" 



136' 



142' 



148' 



QUEENSLAND 



20'- 



-26' 



-32' 



38" 



NORTHERN 
TERRITORY 



WESTERN AUSTRALIA 




32'- 



SCALE 



TASMANIA 



112* 
\ 



Fig. LDistribution of Vertisols in Australia (from Hubble et aL, 1983), 



sols occurs in the aridic soil moisture zone, with 
lesser us tic and xeric regimes and only very 
small areas of udic. In terms of median annual 
rainfall, the aridic areas range from about 100 
to 650 mm, and ustic from as low as about 300 
mm in the temperate south (winter dominant) 
to as high as 1800 mm in the tropics. Xeric 
range from about 300 to occasionally as high as 
900 mm, while the few occurrences of Vertisols 
in udic regimes may be as low as about 1200 mm 
and as high as about 2000 mm. Much of the 
aridic and ustic regions are characterised by 
high rainfall variability. 

Soil temperature regimes range from thermic 
in the south to hyperthermic over the greater 
part of the tropics. In spite of the relatively low 



latitudes in northern Australia, the isohyperth- 
ermic area is very small (Murtha, 1986). 

Topography, Vegetation, and Soil Parent 
Materials 

Australia is well known as a land of vast 
plains, and this is exemplified by most occur- 
rences of Vertisols. In general, they occupy gen- 
tly undulating plains with altitudes less than 
500 m. Many occurrences are on wide flood 
plains of inland streams which receive their 
flood waters from higher rainfall regions in their 
headwaters. Flooding frequency varies greatly, 
with periods of up to a decade between major 
floods being not uncommon. 



ISBELL: AUSTRALIAN VEHTISOLS 



75 



L I Arldic 



Meteorological 
stations 



NORTHERN 
TERRITORY 



QUEENSLAND 
SOUTH AUSTRALIA 



WESTERN AUSTRALIA 



SOUTH 
WALES 




The large semi-arid Vertisol regions are 
mainly grasslands, with the truly arid areas 
supporting only sparse low shrublands. In con- 
trast, in eastern Queensland extensive areas of 
Vertisols originally supported Acacia forests. 

As would be expected, the most common par- 
ent materials are alluvial and aeolian clayey 
sediments, sedimentary rocks such as shales, 
mudstones, and impure limestones, and basic 
igneous rocks particularly, basalt. 

Soil Morphology 

Clear color and textural differences in the 
upper metre or so of the profile are not usual in 
most deep Vertisols, apart from the fairly com- 
mon occurrence of a thin (up to 3 cm) surface 



crusty horizon of usually lower clay content In 
Australia, soils with a thicker (> 3 cm) A horizon 
of lighter texture are not thought of as Vertisols 
but as texture-contrast soils, even though the 
top 18 cm may contain more than 30% clay after 
mixing. 

There is a wide range in color, from red (most 
common in the arid and semi-arid regions) 
through shades of brown and grey to black. This 
color range is not necessarily related to rainfall 
or organic matter content, e.g., black soils de- 
rived from basalt may occur in semi-arid re- 
gions. Some grey or brown forms may become 
strongly red-mottled in their deep subsoils. 

The soils have the characteristic Vertisol 
structure profile. Beneath various surface con- 
ditions (see below) is usually a subsurface hori- 



76 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



zon of moderate, medium to coarse blocky or 
polyhedral peds, grading below 30-50 cm into a 
horizon of moderate, medium to coarse lenticu- 
lar peds with prominent diagonal shear planes 
or inclined slickensides, features which extend 
to the full depth of the solum. 

The natural condition of the surface soil var- 
ies widely in Australian Vertisols. These sur- 
face conditions do not necessarily relate to the 
climatic environment or the soil color, but cation 
status is involved and also probably clay content 
and mineralogy. The following classes can be 
distinguished in the virgin state, but insufficient 
evidence is yet available on whether these con- 
ditions always reform following disturbance 
such as cultivation. 

(i)Massive or weakly structured surface 
crusty horizon _ 3 cm thick, often of lighter 
texture (lower clay content). This overlies 
pedal clay (blocky or polyhedral) which is 
not self-mulching. 

(ii)After wetting and drying, a thin 5-10 mm 
surface flake forms which cracks into ir- 
regular polygons (plates) 3-10 cm diameter. 
These may be readily separated and re- 
moved from underlying pedal (blocky or 
polyhedral) clay which is usually not self- 
mulching. 

(iii)Massive or weak, very coarse blocky A ho- 
rizon, no surface flake forms on drying, and 
no surface crusty horizon. 
(iv)Pedal A horizon (blocky or polyhedral) 
which is not self-mulching, no surface flake 
forms on drying, and no surface crusty hori- 
zon. 

(v)Surface soil moderately to strongly self- 
mulching. Initial drying may form a very 
fragile thin (2-3 mm) surface seal which 
readily disintegrates to a self-mulch on fur- 
ther drying. 

Carbonate in the form of discrete nodules or 
diffuse soft masses in the fine earth is a very 
common feature, while variable amounts of gyp- 
sum are almost ubiquitous in the more arid 
soils. Here also, a red-brown hardpan (similar 
to a duripan) may occur within or below the so- 
lum. Depth of solum varies enormously in the 
Australian Vertisols. Shallow forms (50 cm or 
less) are common where formed on underlying 
hard rocks, but also widespread are very deep 
forms with sola ranging up to 6 m or more. In 
between these extremes, depths of 1-2 m are 
usual. 

Australia is unique in the variety and extent 
of gilgai associated with its soils. A summary 



has been given by Hubble et al. (1983). In brief, 
forms range from linear (long narrow parallel 
mounds and broader depressions on slopes, ver- 
tical interval < 30 cm and horizontal interval 5-8 
m), through normal (small irregular mounds 
and subcircular depressions varying in size and 
spacing, vertical interval usually < 30 cm and 
horizontal usually 3-10 m), to melonhole gilgai 
(irregularly distributed large subcircular or ir- 
regular depressions, usually > 3 m greatest di- 
mension, vertical interval 30 cm to as much as 2 
m, horizontal 6-50 m). In most cases differences 
exist between profiles on mounds and depres- 
sions. 

Physical, Chemical, and 
Mineralogical Properties 

Physical Properties 

Clay contents range from about 40 to 80%, 
the higher values usually in black soils derived 
from basic igneous rocks. Contents tend to be 
relatively uniform throughout the profile. 

The other two major physical character- 
istics of Vertisols involve surface soil aggregate 
behaviour, and water infiltration, transmission, 
and storage properties. These are all interre- 
lated and are dependent on clay content, miner- 
alogy, and physico-chemical interactions. Re- 
cent Australian work in this very broad area has 
been reported by Williams (1983), Smith et al. 
(1984), and Coughlan et al. (1987). 

The conceptual and practical difficulties 
of determining soil water storage in Vertisols 
both under rain-fed and irrigated conditions are 
well known. In Australia several approaches 
have been used. The concept of Plant Available 
Water Capacity (PAWC) has been summarised 
by Gardner (1985). PAWC is defined as the dif- 
ference between the wet profile water content 
following irrigation and the dry profile water 
content under a stressed mature crop (sorghum) 
summed over the measured rooting depth. As 
such, PAWC reflects any limitation in the depth 
and degree of subsoil wetting and any physical 
and chemical constraints, including aeration 
and salinity, to subsoil root growth and water 
extraction. The data for a range of Vertisols 
listed by Gardner (1985) show PAWC values 
ranging from 70 to 140 mm over measured root- 
ing depths of 40 to 120 mm (note that these root- 
ing depths were not constrained by solum 
depth). 



ISBELL: AUSTRALIAN VERTISOLS 



77 



Other approaches to determining plant avail- 
able water capacity of Vertisols have been the 
use of surrogate properties, such as regressions 
with cation exchange capacity and -1500 k Pa 
moisture, and the use of the soil chloride profile 
to estimate rooting depth, e.g., Shaw and Yule 
(1978), Gardner and Coughlan (1982), Ahern 
(1988), and Baker and Ahern (1989). 

Williams (1983) points out that the recharge 
of a swelling clay under rainfall is a very differ- 
ent process from that under ponding, which is 
used in obtaining PAWC. *Water entry capacity* 
therefore becomes all important, with surface 
soil conditions (including cracking and cultiva- 
tion) and rainfall intensity dictating the amount 
of water that enters the soil. In short, under 
most dryland conditions the actual capacity of a 
Vertisol to store water is less important than the 
likelihood of a given amount of water actually to 
enter the profile and subsequently be available 
to the plant as stored water. Williams (1983) 
quotes values ranging from 60 to 260 mm of 
water stored under rainfall in Australian Verti- 
sols. 

An alternative, therefore, to the PAWC ap- 
proach in determining soil water storage is to 
make use of the soil water retention properties 
in appropriate soil water balance models. While 
the shrink/swell properties of Vertisols can 
make for complex behaviour, Williams et al. 
(1983) have shown that the texture, structure, 
and mineralogy can be used to estimate the pri- 
mary water retention functions. By using this 
approach, the soil properties are treated inde- 
pendently of the plant. This avoids some of the 
difficulties associated with the PAWC approach, 
which is determined by many interacting fac- 
tors, including the water retention properties, 
the rooting depth and density, and the crop wa- 
ter use. 

Chemical Properties 

No comprehensive review is possible in this 
brief paper, and hence only some aspects will be 
considered. 

As would be expected from the wide climatic 
range, organic carbon (and nitrogen) contents 
vary widely. Data in Spain et al. (1983) show for 
A horizons (0-5/15 cm depth) median organic 
carbon values of 2.2 and 0.8% for the black 
earths and grey, brown, and red clays respec- 
tively. Where the latter group of clays were 
under Acacia forest, the median value was 1.6%. 
Because the samples of the first two groups are 
from both virgin and cultivated sites, it is diffi- 
cult to make meaningful comparisons with Ver- 



tisols elsewhere. For example, the dark Verti- 
sols from India have mean organic carbon con- 
tents of about 0.6% (Sehgal and Sohan Lai, 
1988), but these undoubtedly have been culti- 
vated for much longer periods than soils in Aus- 
tralia. 

The pH profiles of Australian Vertisols are of 
considerable interest and may well be unique. 
Three general classes occur: (i) alkaline - pH 
about 6.5 or more in the surface and increas- 
ingly alkaline with depth, (ii) alkaline/acid - pH 
about 6.5 or more in the surface but becoming 
strongly acid (pH 4.0 - 5.0) below depths of about 
a meter, (iii) acid - pH less than about 6.5 
throughout, and usually strongly acid at depth. 
The alkaline/acid and most occurrences of the 
acid classes are widespread in deep and very 
deep grey and brown clays, often strongly gil- 
gaied, in eastern Queensland under rainfalls 
ranging from about 500 to 700 mm. 

Most soils in the arid and ustic environments 
are saline at depth, with some soils having high 
levels near the surface. Sodium chloride usually 
dominates the total salts (40 - 80%) except in 
soils containing large amounts of gypsum. 

Cation exchange capacity (CEC) and ex- 
changeable cations vary greatly the former 
ranging from 20 to 80 c mol kg" 1 and obvi- 
ously vary with clay content and clay mineral 
type. Calcium is usually the dominant ex- 
changeable cation in the upper horizons and 
magnesium in the deeper subsoils. Many soils 
have appreciable sodium (ESP > 6) in the sur- 
face, rising to high values (30 or more) at depth. 
As a rule, the black earths are less saline and 
sodic than the others. The strongly acid subsoils 
are of interest, with base saturation values 
ranging from 60 to over 90% in spite of pH val- 
ues of 4 to 5. Further study is required on the 
charge characteristics of these acid clays. 

Clay Mineralogy 

While the darker Australian Vertisols par- 
ticularly those derived from basalt are usu- 
ally smectite dominant, data in Norrish and 
Pickering (1983) and other unpublished data 
show that, of about 100 analyses of grey, brown, 
and red clays, over half are dominated by illite 
and kaolin, with some containing little or no 
smectite. In most cases soil parent material 
appears to play a predominant role in determin- 
ing soil clay mineralogy. In the case of gilgai 
microrelief, there seems to be an inverse rela- 
tionship between smectite content and magni- 
tude of gilgai development, particularly exem- 
plified in the case of melonhole gilgai. 



78 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Land Use Aspects 

In much of the arid and semi-arid regions, 
Vertisols are used for grazing of native pastures 
by sheep and cattle. Stocking rates are very low 
in the arid areas. In central eastern Queen- 
sland most of the original Acacia forests have 
been cleared and sown to improved grass pas- 
tures. Vertisols are extensively used for dry 
land agriculture in the east and south, usually 
between rainfall limits of about 300 and 800 
mm. A wide range of winter crops (mainly 
wheat, safflower, barley, and oats) and summer 
crops (chiefly sorghum, maize, cotton, soybeans, 
sunflower, and millets) are grown. Wheat is 
grown as far north as 22S latitude. Vertisols 
also are extensively irrigated, the chief areas 
being central and south Queensland and north- 
ern New South Wales (cotton with some grain 
and fodder crops and small areas of rice in the 
north) and large areas on the Riverine Plain in 
south-eastern Australia, where the main crop is 
rice. 

In the case of irrigation, most Vertisol prob- 
lems involve restricted water entry and trans- 
mission caused by adverse soil physical condi- 
tions, which in turn are at least partly induced 
by high sodium in the upper part of the profile. 
These conditions are very common in the irriga- 
tion areas on the Riverine Plain. In particular, 
gypsum has been used extensively as a soil 
ameliorant, and the recognition of gypsum-re- 
sponsive soils has been widely studied (Loveday 
1974). 

In northern New South Wales and Queen- 
sland, rain-fed cropping on Vertisols is largely 
constrained by soil water, and farmers have 
long used the practice of growing one crop per 
year and fallowing for water storage during the 
off season. Gardner et al. (1988) have pointed 
out that fallowing is but one of three strategies 
to optimize production on Vertisols in a semi- 
arid environment of high rainfall variability. 
The other two methods are (i) using rain when it 
falls by opportunity cropping, provided the soil 
profile water stored equals or exceeds some 
specified amount. (The rationale behind this is 
to reduce evaporation losses from fallow land. 
This strategy involves either winter or summer 
or double cropping each year); (ii) matching 
cropping to plant available water capacity 
(PAWC) and climate by means of cropping mod- 
els. The presence of a potentially large PAWC is 
one thing, but other important questions are 
how often matching can be achieved and what 



are the probabilities of adequate planting and 
growing season rains. Cropping models thus 
can indicate the probability of achieving high 
yields and the chances of avoiding low yields 
resulting in monetary loss. 

Other soil-related land use problems include 
various degradation problems such as erosion, 
structural decline, plough pan formation, and 
organic matter decline, all of which are serious 
to varying degrees. 

Classification 

Soil Taxonomy 

Since the advent of Soil Taxonomy, the term 
Vertisol has wide currency in Australia, al- 
though subdivisions below the order level are 
seldom used. This is largely because the soil 
moisture-based suborders are not realistic in a 
land use sense, nor does the present 'pell' and 
'chrom* separation reflect any meaningful or 
consistent distinction. 

Recent proposed changes to the classification 
of Vertisols in Soil Taxonomy (Comerma et al., 
1988) are mainly at the great group and sub- 
group level, so that apart from the suggested 
addition of Aquerts and Borerts, the present 
suborders remain unchanged. In the Australian 
context this is unsatisfactory for several rea- 
sons. From a land use point of view there are no 
differences between 'wetter' Torrerts and 'drier' 
Usterts and Xererts, and certainly in eastern 
Australia crops are consistently grown in areas 
thought to be aridic. 

A major weakness at the suborder level is the 
present use of cracking patterns as a surrogate 
for moisture regimes. No actual measurements 
of this have ever been undertaken in Australia, 
and it seems unlikely that definitive studies 
have been undertaken elsewhere in the world. 
Further, as Dudal and Eswaran (1988) have 
pointed out, cracking is influenced by a number 
of soil properties other than climatic conditions; 
this is very evident in the wide range of Verti- 
sols occurring in the Australian arid zone. 

Cracking patterns aside, if the concept of soil 
moisture regimes is to be used at the suborder 
level in Vertisols (or any other order for that 
matter), better methods of defining and estimat- 
ing them must be devised. Some of the major 
problems of the present MOREG 3 method have 
been discussed by Isbell and Williams (1981). 
Given the arguments in that paper about the 
shortcomings of the soil moisture control section 
concept, and the somewhat unique hydrological 



ISBELL: AUSTRALIAN VERTISOLS 



79 



properties of Vertisols, it seems that a new ap- 
proach is required. The most obvious is the use 
of improved water balance models which are 
currently under development, a suggestion also 
made recently by Bouma and Loveday (1988). 

Proposed changes at the great group level 
(Comerma et aL, 1988, and unpublished ICOM- 
ERT Circular Letter No. 5) include great groups 
based on the presence of salic, calcic, and gypsic 
horizons in the 'torric' and 'ustic' suborders. 
These would probably represent relatively mi- 
nor soils in Australia in 'ustic' regions, but 
would be more common in the more arid regions, 
particularly gypsic soils. The proposal in Circu- 
lar Letter No. 5 for an acid great group of 
Usterts and Uderts would cater to some but not 
the majority of Australian Usterts which have 
acid subsoils. In summary, the great majority of 
Australian Vertisols would fall into the proposed 
'HapF great groups. 

In an Australian context, the proposals of 
interest at the subgroup level include the use of 
ESP and the suggestion in Circular Letter No. 5 
of a separation of dark forms based on value 3 or 
less and chroma 2 or less. However, the propos- 
als for further subdivision of implied soil mois- 
ture regimes based on cracking frequency is of 
course subject to the same deficiencies as dis- 
cussed earlier in this paper. 

Australian Classification 

The traditional usage in Australia of the 
terms black earths and grey, brown, and red 
clays has some utility in that it tends to sepa- 
rate the generally more fertile and less variable 
dark forms from all the rest. In the Factual Key 
(Northcote, 1979), a similar high level separa- 
tion based on color was used, with some atten- 
tion paid to surface soil condition and solum 
depth at lower hierarchical levels. 

In a proposed new classification currently 
under development, the Vertisol equivalents are 
defined in a similar fashion (but with a lower 
clay limit of 35%, no mixing criterion, and no 
depth restriction). The first suborder to key out 
contains the 'aquic' soils those that are satu- 
rated in at least some part of the upper 0.5m 
continuously for prolonged periods in most 
years. The remaining suborders are based on 
color, using generalized groupings based on the 
Munsell chart giving classes termed red, brown, 
grey, and black. The rationale for this is the 
usual higher clay and smectite content and fer- 
tility levels of the dark forms (mainly due to 
their basic parent materials). In the case of the 
other colors, a tentative separation has been 



made on the basis of their distinctive appear- 
ance. Justification of this subdivision will de- 
pend on the result of searching the extensive 
Australian soil databases. 

Great groups are in general distinguished on 
the nature of the soil surface (some or all of the 
classes listed earlier). Subgroup criteria include 
presence of a red-brown hardpan (duripan), a 
bleached A2 horizon, carbonate and/or gypsum, 
exchangeable sodium percentage, and presence 
of acid horizons. The usual difficulties arise in 
defining subgroups, because few criteria are 
mutually exclusive. At the family level, differ- 
entiae presently used include depth of solum 
and clay content. The use of a CEC/clay ratio as 
a partial surrogate for clay mineralogy is being 
investigated. 

Literature Cited 

Ahern, C.R. 1988. Comparison of models for predicting 
available water capacity of Burdekin soils, Queen- 
sland. Aust. J. Soil Res. 26:409-423. 

Baker, D.E., and C.R. Ahern. 1989. Estimates of effective 
rooting depth for predicting available water capacity of 
Burdekin soils, Queensland. Aust. J. Soil Res. 27:439- 
454. 

Bouma, J., and J. Loveday. (1988). Characterizing soil 
water regimes in swelling clay soils, p. 83-96. In L.P. 
Wilding and R. Puentes (ed.) Vertisols: their distribu- 
tion, properties, classification and management. Tech- 
nical Monograph No. 18, Soil Management Support 
Services. 

Comerma, J.A., D. Williams, and A. Newman. 1988. 
Conceptual changes in the classification of Vertisols. p. 
41-54. In L.P. Wilding and R. Puentes (ed.) Vertisols: 
their distribution, properties, classification and man- 
agement. Technical Monograph No. 18, Soil Manage- 
ment Support Services. 

Coughlan, K.J., D. McGarry, and G.D. Smith. 1987. The 
physical and mechanical characterization of Vertisols. 
p. 89-105. In M. Latham and P. Ahn (ed.) Management 
of Vertisols under semi-arid conditions. IBSRAM Pro- 
ceedings No. 6. 

Dudal, R., and H. Eswaran. 1988. Distribution, proper- 
ties and classification of Vertisols. p. 1-22. In L.P. 
Wilding and R. Puentes (ed.) Vertisols: their distribu- 
tion, properties, classification and management. Tech- 
nical Monograph No. 18, Soil Management Support 
Services. 

Gardner, E.A. 1985. Assessment of the plant available 
water capacity of swelling soils proposed for irrigation 
development, p. 67-79. In Planning and management 
of water for agriculture in the tropics. Proc. Fifth Afro- 
Asian Regional Conference, Townsville. Int. Comm. on 
Irrigation and Drainage. 

Gardner, E.A., and K.J. Coughlan. 1982. Physical factors 
determining soil suitability for irrigation crop produc- 
tion in the Burdekin-Elliot River area. Technical Re- 
port No. 20, Agricultural Chemistry Branch, Dept. Pri- 
mary Industries, Brisbane. 



80 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Gardner, E.A., K.J. Coughlan, and D.M. Silburn. 1988. 
Soil water measurement and management on Vertisols 
in Queensland, Australia, p. 131-165. In S.C.Jutzi, I. 
Haque, J. Mclntire and J.E.S. Stares (ed.) Manage- 
ment of Vertisols in sub-Saharan Africa. Proc. Confer- 
ence held at ILCA, Addis Ababa, Ethiopia 1987. 

Hubble, G.D. 1984. The cracking clay soils: definition, 
distribution, nature, genesis and use. p. 3-13. In J.W. 
McGarity, E.H. Hoult and H.B. So (ed.) The properties 
and utilization of cracking clay soils. Reviews in Rural 
Science No. 5. Proc. Symp. Univ. New England, Armi- 
dale, 1981. 

Hubble, G.D., R.F. Isbell, and K.H. Northcote. 1983. 
Features of Australian Soils, p. 17-47. In Soils: an 
Australian viewpoint. CSIRO, Melbourne/Academic 
Press, London. 

Isbell, R.F., and J. Williams. 1981. Dry soils of Australia: 
characteristics and classification, p. 124-150. In F.H. 
Beinroth and A. Osman (ed.) Proc. Third Int. Soil Clas- 
sification Workshop, ACSAD, Damascus. 

Loveday, J.L. 1974. Recognition of gypsum-responsive 
soils. Aust. J. Soil Res. 25:87-96. 

Murtha, G.G. 1986. Soil temperature regimes - a tropical 
experience, p. 23-27. In A.W. Moore (ed.) The USDA 
soil taxonomy in relation to some soils of eastern 
Queensland. CSIRO Aust. Div. Soils, Divl. Rep. No. 84. 

Norrish, K, and J.G. Pickering. 1983. Clay minerals, p. 
281-308. In Soils: an Australian viewpoint. CSIRO 
Melbourne/Academic Press, London. 

Northcote, K.H. 1979. A Factual Key for the recognition 
of Australian soils. 4th edn. Rellim, Glenside, S.A. 

Probert, M.E., I.F. Fergus, B.J. Bridge, D. McGarry, C.H. 
Thompson, and J.S. Russell. 1987. The properties and 
management of Vertisols. IBSRAM/CAB Interna- 
tional, Wallingford, U.K 



Sehgal, J.H., and J.C. Bhattacharjee. 1988. Typic Verti- 
sols of India and Iraq - their characterization and clas- 
sification. Pedologie XXXVHL67-95. 

Sehgal, J.L., and Sohan Lai (ed.) 1988. Benchmark swell- 
shrink soils of India. National Bureau of Soil Survey 
and Land Use Planning, Nagpur, India. 

Shaw, R.J., and D.F. Yule. 1978. The assessment of soils 
for irrigation, Emerald, Queensland. Technical Report 
No. 13, Agricultural Chemistry Branch, Dept. Primary 
Industries, Brisbane. 

Smith, G.D., D.F. Yule, and K.J. Coughlan. 1984. Soil 
physical factors in crop production on Vertisols in 
Queensland, Australia, p. 87-104. In E.T. Craswell 
and R.F. Isbell (ed.) International Workshop on Soils, 
Townsville 1983. Australian Centre for International 
Agricultural Research. 

Spain, A.V., R.F. Isbell, and M.E. Probert. 1983. Soil 
organic matter, p. 551-563. In Soils: an Australian 
viewpoint. CSIRO, Melbourne/Academic Press, Lon- 
don. 

Stace, H.C.T., G.D. Hubble, R. Brewer, K.H. Northcote, 
J.R. Sleeman, M.J. Mulcahy and E.G. Hallsworth. 
1968. A handbook of Australian soils. Rellim, Glen- 
side, S.A. 

Williams, J. 1983. Soil hydrology, p. 507-530. In Soils: 
an Australian viewpoint. CSIRO, Melbourne/Academic 
Press, London. 

Williams, J., R.E. Prebble, W.T. Williams, and C.T. Hi- 
gnett. (1983). The influence of texture, structure and 
clay mineralogy on the soil moisture characteristic. 
Aust. J. Soil Res. 21:15-32. 



Properties and Classification of Cold Aridisols in Montana 
C. Wang and Tom Keck, Jerry Nielsen, Robert Richardson, and Gordon Decker 1 

Abstract 

Proposed changes in the Aridisol order would have a profound affect on 
the future classification of most Cold Aridisols in Montana. Aridisols are 
largely confined to semi-arid regions of the state which receive between 254 
and 356 mm (10 to 14 inches) of mean annual precipitation. They are sepa- 
rated from Aridic subgroups of Mollisols in these areas because they fail to 
meet the color requirements for a mo Hie epipedon. Organic carbon, crop 
yield, and soil climate data were examined as they relate to the current and 
future classification of Aridisols in Montana. Surface horizons of both 
Aridisols and Mollisols, within the semi-arid region, had similar amounts of 
organic carbon. Crop yield data were limited for Montana soils. The data 
that were available did not support the hypothesis that the darker surface 
color of Mollisols represented a substantially higher yield potential. The 
Montana Agricultural Potentials System (MAPS) was used to evaluate the 
geographic distribution of soil climatic variables across the state. 



Introduction 

Aridisols have been mapped extensively 
throughout semi-arid regions of Montana. They 
account for a major portion of the taxa used in 
the eastern half of the state and in the inter- 
mountain valleys of southwestern Montana. 
Aridic soil moisture regimes are implied by the 
classification; however, the soil moisture re- 
gimes for most of these areas can be more accu- 
rately described as Aridic intergrades of the us- 
tic moisture regime. Water is limiting to the 
growth of mesophytic plants but not to the ex- 
tent typically associated with Aridisols. Suffi- 
cient spring moisture allows for successful crop- 
fallow operations in many of these areas. 

This paper examines the available climatic, 
organic carbon, and crop yield data for cold 
Aridisols as they have been mapped in Mon- 
tana. The underlying question is whether these 
soils are most accurately classified as Aridisols 
or would be more appropriately assigned to In- 
ceptisol and Alfisol soil orders. Proposed 
changes in the Aridisol order, specifically the 
allowance for Aridisols with a mollic epipedon, 
have created a need to more accurately identify 
aridic, ustic, and xeric soil moisture regimes in 
Montana. 

Soil Climate 

The majority of Aridisols in Montana have 
been mapped in areas having a frigid soil tem- 



^raduate Research Assistant and Professor of Soil 
Science, Dep. of Plant and Soil Science, Montana State 
Univ., Bozeman, MT 59717; State Soil Correlator, and 
State Soil Scientist, Soil Conservation Service, USDA, 
Bozeman, Montana, 59715. 



perature regime. Three of the four Montana 
Aridisol sites on the cold Aridisol tour have a 
frigid temperature regime. Mesic Aridisols have 
been mapped in several counties along the Yel- 
lowstone River in the south-central part of the 
state. The Stormitt series is the lone represen- 
tative of a mesic Aridisol in the Montana portion 
of the tour. To date, no cryic Aridisols have been 
mapped. The potential exists for identifying 
such cryic Aridisols in the intermountain valleys 
of southwestern Montana, specifically in Beav- 
erhead county (Nimlos and Tippy, 1981). 

Soil series often are separated by precipita- 
tion zones. In Montana, most soils mapped as 
Aridisols occur within a 254 to 356 mm (10 to 14 
inch) mean annual precipitation range (Fig. 1). 
Areas receiving less than 254 mm of mean an- 
nual precipitation are limited in the state to iso- 
lated pockets in the rainshadow of mountains. 
These occur south of Bridger along the Clark's 
Fork of the Yellowstone River and near Dillon 
along the Beaverhead and Jefferson Rivers (Fig. 
1). Aridisols generally have not been mapped in 
any areas receiving more than 356 mm of mean 
annual precipitation. 

Classification 

The classification of Aridisols in Montana 
would undoubtedly change due to the proposed 
changes in the Aridisol order. As things stand 
today, all Aridisols in the state, having a frigid 
temperature regime and occurring within the 
254 to 356 mm (10 to 14 inch) precipitation 
zone, are in Borollic subgroups. Typic sub- 
groups are restricted to areas receiving less 
than 254 mm of mean annual precipitation. 



81 



82 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 




i 152 - 254 m 
\ 254 - 305 m 
X 305 - 356 m 



ANNUAL PRECIPITATION RANGES 



MONTANA 

ALBERS EQUAL AREA PROJECTION 

PLANT i SOIL SCIENCE DEPARTMENT 

MONTANA STATE UNIVERSITY. BQZEMAN. MT . 



Most 'mesic' Aridisols are classified as Ustollic, 
Ustertic, Haplustollic, and Glossic Ustollic sub- 
groups. 

The Argids currently mapped in Montana are 
Natrargids and Haplargids with some Palear- 
gids. Camborthids and Calciorthids comprise 
most of the Orthids. Two of the Aridisol tour 
sites in Montana have Camborthids and two 
have Calciorthids. To the best of our knowledge, 
only one Gypsiorthid series and two Salorthid 
series have been classified and mapped in Mon- 
tana. 

Changes in Classification 

The proposed classification changes will have 
different effects on taxa as they are currently 
mapped in Montana. For example, most Borol- 
lic Haplargids will remain as Borollic Haplar- 
gids based on Version 6.0 of the revised Aridisol 
keys (International Committee on Aridisols, 
1989). Most Borollic Natrargids and Paleargids 
would change to Haplic and Ustalfic subgroups, 
respectively. This is because in the revised cri- 
teria Borollic subgroups of these soils have a 
color value, when crushed, darker than 3.5 
moist and 5.5 dry. Most Ustollic Haplargids 
would switch to Ustalfic Haplargids and most 



Ustollic Natrargids would probably fit within 
the Haplic Natrargid classification. Above the 
subgroup level, the classification of Paleargids 
and Natrargids would be unchanged. Some 
Haplargids would change to Calciargids. 

The Camborthids would mostly become Hap- 
locambids in the revised classification. The Cal- 
ciorthids would primarily be Haplocalcids with a 
few Petrocalcids. The Borollic subgroup would 
remain with the Haplocalcids but would change 
to Ustocreptic for the Haplocambids. The cur- 
rent Ustollic subgroup of Camborthids would be 
classified as Ustochreptic Haplocambids. 

These projected changes are based on the 
overall revisions in Version 6.0 of the Interna- 
tional Committee report on Aridisols. They are 
not based on any detailed study of specific se- 
ries. In some cases, the proposed revisions 
would inevitably create inconsistencies with the 
way a given series has been mapped in the past. 
Our assumption in the above projected changes 
is that these soils would remain within the 
Aridisol order. 

Relation to Other Soil Orders 

Aridisols have been mapped in close associa- 
tion with VertMols, Entisols, and Mollisols 



WANG, KECK, NIELSEN, RICHARDSON, DECKER: PROPERTIES AND CLASSIFICATION OP COLD ARIDISOLS IN MONTANA 



83 



throughout semi-arid areas of Montana. Verti- 
sols are separated from Aridisols by cracks 1 cm 
or wider to a depth of 50 cm, and by slickensides 
close enough to intersect, or by wedge-shaped 
natural structural aggregates that have their 
long axes tilted 10 to 60 from the horizontal. 
Entisols are separated from Aridisols by their 
lack of diagnostic horizons beyond an ochric 
epipedon. The Aridic subgroups of Mollisols are 
separated from Aridisols on the basis of soil color 
of the surface horizon or horizons. The Aridic 
subgroups of Mollisols would be included within 
the Aridisol order, based on the proposed revi- 
sions. This change is because the revisions al- 
low for mollic surface colors (i.e., mollic epipe- 
dons) in Aridisols and because Aridisols key out 
before Mollisols in the key to soil orders. The 
Scobey series, included at one of the Aridisol 
tour sites, is an example of an Aridic Argiboroll 
that would need to be reclassified as an Aridisol 
if the area was in fact an die. 

Inceptisols and Alfisols have generally been 
restricted in Montana to areas receiving more 
than 356 mm of mean annual precipitation. 
Thus, these soil orders have not generally been 
mapped in close association with Aridisols. A 
decision was made, when Soil Taxonomy was 
implemented, that Alfisols in Montana could not 
occur under grassland vegetation. This decision 
meant that grassland soils, having an argillic 
horizon but not meeting the color requirements 
for a mollic epipedon, had to be classified as 
Aridisols. Also, there were no provisions in Soil 
Taxonomy for aridic intergrades of Inceptisols 
with a frigid temperature regime. For these rea- 
sons, the Aridisol order was used extensively in 
Montana for areas that have aridic intergrades 
of the ustic soil moisture regime. 

As a result, aridic intergrades of Mollisols 
(i.e., Aridic Argiborolls) with an ustic soil mois- 
ture regime are mapped in the same precipita- 
tion zone with ustic intergrades of Aridisols (i.e., 
Borollic Camborthids). Mapping of Aridisols 
today in semi-arid areas of Montana is still in- 
fluenced by earlier decisions. 

The proposed changes in Aridisol taxonomy 
provide Montana with an incentive and an op- 
portunity to re-evaluate the soil moisture re- 
gime for most Aridisols within the state which 
receive between 254 to 356 mm (10 to 14 inches) 
of mean annual precipitation. Potential use and 
management should be of primary concern in 
decisions about the future classification of 
Aridisols in Montana. 



Soil Organic Matter 

Hans Jenny, in 1930, described the basic rela- 
tionship under grassland vegetation between 
soil organic matter and climate. Organic matter 
increases with decreasing temperature and with 
increasing moisture (Jenny, 1930). Mollisols 
occurring in moister and cooler environments 
contain considerably more organic carbon than 
do Aridisols in hotter and drier climates. How- 
ever, this may not be the case when both orders 
occur within the same soil temperature and 
moisture regime. The assumption is that the 
dark surface color of Mollisols represents higher 
organic carbon levels in the soil. The mollic 
epipedon has been viewed as a "fossil record" of 
past root growth and a predictor of future pro- 
duction (Cannon and Nielsen, 1984). Yet if 
Jenny's conceptual model of soil organic matter 
in grasslands is correct, we might expect similar 
ranges in organic carbon for both Aridisols and 
Mollisols when comparing the two within the 
same temperature and moisture zones. 

Individual pedons will vary in organic carbon 
content for a host of site specific reasons: topo- 
graphic position, parent material, soil texture, 
and management history (Nichols, 1984; 
Franzmeier et al, 1985; Sims and Nielsen, 
1986; Aguilar et al, 1988; Yonker et al., 1988). 
Acknowledging this variation, we question how 
well surface color of Mollisols, versus Aridisols, 
in semi-arid areas of Montana, represents differ- 
ences in the amount of organic carbon in surface 
horizons. 

Organic Carbon Data for Montana 
Soils 

We used the data available in the Montana 
Pedon Database (Decker, 1972) to compare or- 
ganic carbon levels between Mollisols and 
Aridisols mapped within the 254 to 356 mm (10 
to 14 inch) precipitation zone in Montana. 
These data originated from two sources, SCS 
soil characterization data from the National 
Soils Lab in Lincoln and a statewide BLM study 
of range and soil characteristics in Montana 
(McDaniel et.al., 1982). The SCS data repre- 
sents both cropland and rangelands sites, while 
the BLM data is for rangeland sites only. Each 
data set was handled separately to compare re- 
sults from different sources. Average organic 
carbon levels were calculated for the top to 18 
cm (0 to 7 inches) and for 18 to 36 cm (7 to 14 
inches) depths for all pedons, and comparisons 
were made based on taxonomic classifications 
(Fig. 2 and 3). 



84 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



SCS Soil Characterization Data 
MOLLISOLS - Frigid Temperature Regime 



Argiborolls* 
n=10 



Haploborolls, Aridic 
n=3 



0.91 



7777771 

HUH 

, V/////1 

.2 .4 .6 .8 1.0 1.2 1.4(2) 



0.84 



1.44 



1.17 



* Aridic, Abruptic Aridic, and Ustertic subgroups 
ARIDISOLS - Frigid Temperature Regime 



Camborthids, Borollic 
n=2 



Paleargids, Borollic 
n=2 



Natrargids, Borollic 
n=7 



1.00 



0.61 



0.67 

. 1/////////////////I 
.2 .4 .6 .8 1.0 1.2 174 1.6 1.8(%) 



1.75 



1.82 



1.46 



ARIDISOLS - Mesic Temperature Regime 



Haplargids, Ustollic 
n=8 



Camborthids * 
n=9 



Natrargids ** 
n=6 



0.66 



0.72 



Illllllll 
/////I/I! 
Illllllll 



0.63 



I///////////// 



1.11 



1.11 



1.26 



.2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8(%) 



* Ustollic and Ustertic subgroups 
** Ustollic, Haplustollic, and Glossic Ustollic subgroups 



TOTALS 



Mollisols, Frigid TR 
n=13 



Aridi sols, Frigid TR 
n=12 



Aridi sols, Mesic TR 
n=23 



0.89 



0.69 



0.68 


Illllllllll 
Illllllllll 
Illllllllll 



1.38 



1.54 



1.15 



(3 TF74 J5 .8 1.0 1.2 1.4 1.6 1.8(%) 



Figure 2. Average organic carbon content (%] in the surface 18 centimeters 
(numbers to right of bar graphs) and in the 18 to 36 cm depth 
(numbers within left side of bar graphs) of Mollisols and Aridisols 
mapped within the 254 to 356 mm precipitation zone in Montana. SCS 
soil characterization data (Decker, 1971). 



WANG, KECK, NIELSEN, RICHARDSON, DECKER: PROPERTIES AND CLASSIFICATION OF COLD ARIDISOLS IN MONTANA 85 



BIN Soil Data 

MOLLISOLS - Frigid Temperature Regime; 254-356 mm Mean Annual Prec. 

Argiborolls, Aridic 



Haploborolls, Aridic 
n=5 



Calciboroll, Aridic 



0.89 



Ullllllllllll 
Ullllllllllll. 
Ullllllllllll 



0.94 



IW////////1 

Ulllllllll 



0.73 



1.48 



1.41 



1.45 



0.84 


niiiniinnii 

Ulllllllllllll 
Ulllllllllllll 



0.90 



UIUUUUIIIIII 



.2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8(%) 
ARIDISOLS - Frigid Temperature Regime; 254-356 mm Mean Annual Prec. 



Haplargids, Borollic 
n=14 



Camborthids, Borollic 
n=6 



Calciorthids, Borollic 
n=2 



Paleargids, Borollic 
n=3 



Natrargids, Borollic 



1.81 



1777777 



1.43 



1.60 



2.12 






75 


///// 



0.63 



llllllll 



0.96 



1.07 



.2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 1.9 2.0(X) 
ARIDISOLS - 127-229 mm Mean Annual Prec. 



Haplargids, Borollic * 
n=l 



Camborthids, Typic 
n=5 



Natrargids, Typic 
n=3 



0.76 


III 
(II 
/// 






h 
46 ft 

h 



0.44 



0.85 



0.62 



0.82 



(5 TF73 .6 .8 1.0 1.2 1.4 1.6 1.8(%) 
* Subgroup classification questionable 

COMPARISON - 254-356 mm Mean Annual Prec. 



Haploborolls and 
Argiborolls 
n=12 

Camborthids and 
Haplargids 
n=20 



0.92 



0.86 



(3 7274 76 



IllllllllllllU 

1.0 1.2 1.4 1.6 1.8(%) 



1.44 



1.48 



Figure 3. Average organic carbon content (%) in the surface 18 centimeters (numbers to 
right of bar graphs) and in the 18 to 36 cm depth (numbers within left side 
of bar graphs) of Mollisols and Aridisols mapped in arid and semi -arid 
regions of Montana. BLM soil characterization data (McDaniel, 1982). 



86 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



SCS Data 

Looking first at the SCS Data, there were 
only two Borollic Camborthid and two Borollic 
Paleargid pedons in the database with organic 
carbon data, but the average organic carbon 
levels in the top 18 cm for these groups were 
1.75% for the Borollic Camborthids and 1.82% 
for the Borollic Paleargids. The sodium affected 
Borollic Natrargids, based on seven samples, 
had an average organic carbon content in the 
top 18 cm of 1.46 percent. Aridisols with a frigid 
temperature as a whole had an average organic 
carbon content of 1.54% in the top 18 cm and 
0.67% in the 18 to 36 cm depth. 

In contrast, Mollisols, within the same pre- 
cipitation zone and with the same frigid soil 
temperature regime, had an average organic 
matter content of 1.38% in the top 18 cm and 
0.89% in the 18 to 36 cm depth. Averages for 
the top 18 cm in different subgroups were 1.50% 
for Aridic Argiborolls, 1.40% for Abruptic Aridic 
Argiborolls, 1.17% for Aridic Haploborolls, and 
1.39% for Ustertic Argiborolls. Admittedly, this 
initial data set is small, but at first glance, the 
data do not support the idea that Mollisols have 
higher levels of organic carbon than associated 
Aridisols in the same climatic zone. 

There was an abundance of SCS Data for 
Aridisols with a mesic soil temperature regime. 
If Jenny's conceptual model is correct, then 
warmer temperatures should result in lower 
organic carbon levels. McDaniel and Munn 
(1985) found that organic matter did decrease in 
both Mollisols and Aridisols as temperature be- 
came warmer. The SCS data substantiate this 
observation. The average for 23 mesic Aridisol 
sites was 1.15 percent for the top 18 cm and 
0.68% for the 18 to 36 cm depth (Fig. 2). 

BLM Data or (Bureau of Land 
Management Data) 

Care must be taken in drawing any conclu- 
sions from such a limited data set. The BLM 
data provided an excellent opportunity to check 
for similar trends in a completely separate data 
set. Again, comparisons were made based on 
subgroups and average organic carbon levels 
calculated for the top 18 cm and then for the 18 
to 36 cm depth. 

Borollic Camborthids in the BLM data had an 
average organic carbon content of 1.6% (n = 6) in 
the top 18 cm. The high for this group was 
2.26%; the low was 1.06%. Borollic Haplargids 
had an average in the top 18 cm of 1.43% (n = 



14) with a high of 2,05% and a low of 0.97%. 
Borollic Calciorthids, based on a sample of only 
two, had an average of 2.12% organic carbon in 
the top 18 cm; the high was 2.28% and the low 
was 1.95%. The corresponding organic carbon 
averages, in the top 18 cm, for the Aridic sub- 
groups of Mollisols were: 1.41% (n = 7) for Aridic 
Haploborolls (high 2.11% and low .71%), 1.48% 
(n = 5) for Aridic Argiborolls (high 1.97% and low 
.99%), 1.45% for the only Aridic Calciboroll, 
Averages for the 18 to 36 cm depth were also 
similar between the above groups of Mollisols 
and Aridisols, except for the unusually high or- 
ganic carbon levels in the two Borollic Calci- 
orthids. 

The BLM data, unlike the SCS characteriza- 
tion data, show consistently low levels of organic 
carbon in the top 18 cm of Borollic Natrargids 
(avg. = 1.07%, n = 13) and Borollic Paleargids 
(avg. = .96%, n = 3). The low average for Natrar- 
gids is expected due to reduced grass production 
on sodium and salt-affected soils. The low aver- 
age for the Paleargids is surprising. It may be 
explained in part by the borderline classifica- 
tion, Borollic Paleargid/Borollic Natrargid, for 
two of the three samples. It seems reasonable to 
conclude that sodium and salt-affected Aridisols 
will have reduced organic carbon levels. 

The BLM data set also included eight Aridisol 
sites from areas receiving 127 to 229 mm (5 to 9 
inches) of mean annual precipitation. As ex- 
pected, the drier Aridisols had consistently 
lower organic carbon levels than either Mollisols 
or Aridisols receiving 254 to 356 mm of mean 
annual precipitation. 

Conclusions from Available Organic 
Carbon Data 

The data available supports Jenny's initial 
idea that soil organic matter varies primarily in 
response to temperature and moisture. Borollic 
Aridisols and Aridic Mollisols have comparable 
levels of organic carbon in their surface horizons 
despite differences in surface colors. Warmer 
soils, as in mesic Aridisols, or drier soils, as in 
Typic Aridisols, have correspondingly lower lev- 
els of organic carbon. Soil factors, such as salin- 
ity or sodicity, which reduce plant production 
will also reduce organic carbon levels. Soil color 
is an attribute useful for separating soil series in 
the field. It does not necessarily, however, rep- 
resent consistent differences in organic carbon 
levels between Mollisols and Aridisols in semi- 
arid areas. 



WANG, KECK, NIELSEN, RICHARDSON, DECKER: PROPERTIES AND CLASSIFICATION OF COLD ARIDISOLS IN MONTANA 



87 



Humic - Fulvic Acids 

Differences in soil color between Aridisols and 
Mollisols in the same precipitation zone could be 
related to differences in their ratios of humic 
and fulvic acids. Extracted humic acid is black 
or dark brown. Fulvic acid, on the other hand, is 
normally tan or lighter brown. The transition 
from wetter to drier grassland soils not only 
reduces the amount of grass and therefore, or- 
ganic carbon production, but also reduces the 
intensity of the humification process (Anderson, 
1979). Not only is there less organic carbon, but 
also a higher proportion of it is the lighter col- 
ored fulvic acid. Shields et al (1968) found 
higher than expected organic carbon contents in 
Gray Wooded Soils (Alfisols) because they con- 
tained a higher proportion of fulvic acids than 
the Chernozemic soils (Mollisols). 

Crop Yield Data 

Site specific crop yield data are extremely 
limited for most of Montana's agricultural soils. 
An on-going cooperative crop yield study by the 
Soil Conservation Service and Montana State 
University is intended to determine how dry- 
land crops respond to different soil, climate, and 
management factors. Table 1 summarizes some 
yield data for Mollisols and Aridisols within the 
254 to 365 mm (10 to 14 inch) precipitation zone 
from the crop-yield study. 

Obviously, specific comparisons are of limited 
value, since widely scattered sites may have 
received different amounts of precipitation and 
different management inputs in a given year. 
The averages are interesting however, in 1986, 
winter wheat produced an average six bushels 
more on Mollisols than on Aridisols in the study 
(254 to 356 mm mean annual precipitation 
zone). There were only slight differences be- 
tween the two groups in spring wheat or barley 



Aridisols 
Yield #ofobs. 
Mgh- 1 



Table 1. Average dryland yields by year for Mollisols 

and Aridisols occurring in the 254 to 356 mm 

precipitation zone in Montana. Data compiled from a 

joint Soil Conservation Service - Montana State 

University crop yield study.> 

Mollisols 

Crop Yield #ofobs. 

Mgh' 1 
1286 

Winter Wheat 2.10 30 

Spring Wheat 2.66 15 

Barley 2.28 12 

1987 

Winter Wheat 2.58 27 

Spring Wheat 3.17 3 

Barley 2.85 4 

1988 - Drought Year 

Winter Wheat 1.92 12 NA 

Spring Wheat 0.87 9 NA 

Barley 0.34 9 0.77 7 

> Unpublished thesis data provided by Linda A. Spencer, graduate 

research assistant, Montana State University, 1989. 



1.69 
2.49 
2.21 

2.60 
2.04 
2.77 



15 

9 

21 

9 
3 
9 



Table 2. SCS cropland and rangeland production 
estimates for selected soil series used in comparisons .> 

Taxonomy/Series Wheat Wheat Hay Ranged 

Mgh- 1 

Aridic Argiborolls, fine 

Scobey 2.35 2.22 3.36 1.46-1.68 

Ethridge 2.35 1.75 3.36 1.46 

Borollic Haplargid/Paleargid, fine 

Pinelli 2.02 1.68 2.24 1.46-1.79 

Phillips 2.15 1.68 2.46 1.23-1.79 

Aridic Haploboroll, fine-loamy 

Kremlin 2.35 2.15 3.36 1.46-1.79 

Borollic Camborthid, fine-loamy 

Yamac 2.35 2.02 2.69 1.23-1.79 

Aridic Haploboroll, fine-silty 

Floweree 2.15 1.68 2.24 1.34-1.68 

Borollic Camborthid, fine-silty 

Lonna 2.29 1.75 3.36 1.23-1.79 

> Data obtained from soil interpretive records (form 5's) of the 
Montana 

Soil Survey, Soil Conservation Service, USDA, 1989. 
Estimated grain and hay yields assume a high level of management. 
# Potential range production for range in a good to excellent range 



production. In 1987, the semi-arid Mollisols 
produced on the average nearly 17 bushels more 
spring wheat than the Aridisols, but this differ- 
ence is based on only three samples for each 
group. There was essentially no difference be- 
tween the two groups in winter wheat or barley 
production. In 1988, a drought year, there was 
insufficient data to make any comparisons. This 
type of crop-yield correlation data is sorely 
needed to assess the productivity potential of 
Montana's agricultural soils. 

It does not appear likely, based on incomplete 
data, that surface color is an important predic- 
tor of potential dryland crop production within 
the semi-arid environment. This agrees with 
potential production estimates recorded on Soil 
Interpretive Records by SCS soil scientists. 
These estimates, based on field experience and 
producer averages, often show only slightly 
lower plant production potentials for Aridisols 
than for corresponding Mollisols (Table 2). 

Crop yield data do not support the idea that 
areas receiving 254 to 356 mm of precipitation 
in Montana meet the criteria for an Aridic mois- 
ture regime, at least not based on the criteria of 
a 70% probability of crop failure in a given year. 

MAPS 

A decision needs to be made on the extent of 
the Aridic soil moisture regime in Montana. The 
proposed changes in the Aridisol order will alter 
the taxonomy of many of Montana's Aridisols in 
any event. Under the proposed changes, soils 
with mollic epipedons and an aridic soil mois- 
ture regimes will be reclassified as Aridisols. 
These proposed changes provide an opportunity 
to re-evaluate soil classification based on soil cli- 
mate. The Montana Agricultural Potentials 
System (MAPS) developed by Caprio and 



88 



SIXTH iNTERNATIONAl Soil CLASSIFICATION WORKSHOP 




\ DRYLAND CROPPING 
. IRRIGATED CROPPING 



FIG. 4. CROPLAND IN MONTANA 



ALBERS EQUAL AREA PROJECTION 

PLANT 8. SOIL SCIENCE DEPARTMENT 
MONTANA STATE UNIVERSITY. BOZEMAN. MT. 




;. '5. AREAS WHERE ARIDISOLS WERE A 
MAJOR COMPONENT OF SOIL MAPS 



MONTANA 

ALBERS EQUAL AREA PROJECTION 



WANG, KECK, NIELSEN, RICHARDSON, DECKER: PROPERTIES AND CLASSIFICATION OF COLD ARIDISOLS IN MONTANA 



89 



Nielsen and others at Montana State University 
provides a valuable tool for evaluating land re- 
sources, including soil climate. 

The MAPS system is a computer-driven map- 
ping system which at this time has over 150 
data layers in it. Data are stored in 18,005 cells 
covering the whole state. Each cell represents 
three minutes of latitude by three minutes of 
longitude or approximately eight square miles. 

Fig. 1 shows the mean annual precipitation 
zones for the arid and semi-arid regions of Mon- 
tana. Areas have been split into 305 to 356 mm 
(12 to 14 inch), 254 to 304 mm (10 to 12 inch), 
and 152 to 254 mm (6 to 10 inch) precipitation 
zones. Large areas of Montana receive between 
254 and 356 mm of mean annual precipitation. 
Only two isolated areas, south of Bridger and 
near Dillon, receive less than 254 mm of mean 
annual precipitation. 

Fig. 4 shows areas of irrigated and non-irri- 
gated cropland in the state. Many of these ar- 
eas, especially in the north central part of the 
state, correspond to areas receiving less than 
356 mm of mean annual precipitation. Fig. 
5 shows areas that have a large component of 
Aridisols mapped within them. This map is 
based on the general soils map for the state. It 
identifies all areas of the state where one or 



another Aridisol taxa is included in the map unit 
name on the general soils map. Areas not in- 
cluded may still have some Aridisols mapped in 
county soil surveys. Overlaying this map on the 
cropland map shows at least a general pattern 
in the eastern half of the state of only small 
amounts of cropland in predominantly Aridisol 
areas. 

Fig, 6 shows areas where potential 
evapotranspiration (ET), as derived by the 
Thornthwaite model, exceeds measured precipi- 
tation. In general, values between 280 and 330 
mm ET deficit correspond to the 254 - 304 mm 
precipitation zone. Values near 150 to 200 mm 
deficit correspond to the 305 to 356 mm precipi- 
tation zone. The dry area south of Bridger has 
ET deficits ranging from 356 to 432 mm. The 
equally low precipitation areas near Dillon have 
ET deficits in a range of only 254 to 304 mm. 
Lower temperatures in the Dillon area should 
be reflected in lower potential evapotranspira- 
tion, possibly indicating areas of cryic Aridisols. 
On the other hand, the short growing season 
and relatively low evapotranspiration demand 
in these areas may make moisture less limiting 
than the sparse precipitation suggests. 

Fig. 7 shows areas of potential mesic soil tem- 
perature regimes based on mean annual soil 




WATER DEFICIT BASED ON THORNTH- 
WAITE POTENTIAL EVAPOTRANSPIRA- 
TION MINUS ANNUAL PRECIPITATION 



ALBERS EQUAL AREA PROJECTION 

PLANT i. SOIL SCIENCE DEPARTMENT 
MONTANA STATE UNIVERSITY. BOZEMAN. MT . 



90 



SIXTH iNTERNATIONAl Soil CLASSIFICATION WORKSHOP 




/ MEAN ANNUAL 

PERATURE > 
GREES C. 



SOIL TEM- 
= 8 DE- 



7. MESiC SOIL TEMPERATURE REGIME 

AREAS ESTIMATED FROM MEAN ANNUAL 
AIR TEMPERATURE PLUS 1 DEGREE F 



MONTANA 

AL8ERS hQUAL AREA PROJECTION 



PLAN! & bOll. SCIbNfL DF.PARTMhNl 
MON1ANA SIAIE UN 1 V b'P'-j 1 I Y . BQ/FMAN. MI. 




/ MEAN ANNUAL SOIL TEM- 
PERATURE >= 8 DE- 
CREES C. 



FIG. '8. MESIC SOIL TEMPERATURE REGIME 

AREAS ESTIMATED FROM MEAN ANNUAL 
AIR TEMPERATURE PLUS 2 DEGREES F 



MONTANA 

ALBERS EQUAL AREA PROJECTION 



WANG, KECK, NIELSEN, RICHARDSON, DECKER: PROPERTIES AND CLASSIFICATION OP COLD ARIDISOLS IN MONTANA 



91 



temperatures approximated by mean annual air 
temperature plus 1F. This is a conservative 
estimate of mesic areas that corresponds rea- 
sonable well with areas mapped as mesic in the 
state or at least perceived as bordering on mesic. 
Fig. 8 is based upon mean annual air tempera- 
ture plus 2F and suggests a substantially 
larger area of mesic soil temperature regime in 
the state. 

Soil climate data are given in Appendix 1 for 
each of the Montana sites on the Aridisol Tour. 
The data are based on the cell values in the 
MAPS system for these sites. Two sites, the 
Lonna and the Cambeth series, occur within the 
same MAPS cell. The Stormitt site occurs in a 
cell with large relief. The presence of higher 
elevation areas in the cell skewed the data for 
this site toward a wetter and cooler environ- 
ment than expected for the Stormitt series. The 
series normally occurs in areas receiving less 
than 254 mm of mean annual precipitation. A 
separate data set was acquired from an adjacent 
drier cell west of the first cell. 

Conclusions 

Aridisols by definition must have an aridic 
moisture regime. They are soils that do not 
have water available to mesophytic plants for 
long periods. We cannot justifiably argue that, 
within the 254 to 356 mm precipitation zone, 
Mollisols occur within an ustic moisture regime 
and Aridisols occur in an aridic moisture regime. 
The proposed changes in Aridisol classification 
allow for a mollic epipedon in Aridisols; thus, a 
decision needs to be made regarding the soil 
moisture regime in these semi-arid areas. 

Our recommendation at this time, is that ar- 
eas receiving between 254 and 356 mm of mean 
annual precipitation be considered to have an 
ustic soil moisture regime bordering on aridic. 
The aridic soil moisture regime should be re- 
stricted largely to areas receiving less than 254 
mm of mean annual precipitation. Many of the 
soils currently classified as Haplargids, Cam- 
borthids, and Calciorthids should be reclassified 
within the Alfisol and Inceptisol soil orders. 
Soils with obvious salinity or sodicity problems, 
which limit either available water or the infil- 
tration of water, would still fit the criteria for an 
aridic soil moisture regime and be classified as 
Aridisols. Most of these soils do not have ade- 
quate internal drainage to be reclaimed and so 
the transient nature of soil salts should not be a 
taxonomic concern. Areas receiving less than 
254 mm of mean annual precipitation represent 
the true aridic soil moisture regime in Montana. 



These recommendations are based in part on 
the apparently similar organic carbon levels and 
yield potentials of most Aridisols and Mollisols 
in these semi-arid areas. The acreage of 
Aridisols would be greatly reduced in Montana. 
Many of the Camborthids and Calciorthids 
would be reclassified as Inceptisols. Haplargids 
and possibly some Paleargids would become 
Alfisols. The classification of Mollisols in semi- 
arid areas would remain unchanged. 

As a final note, cryic Aridisols have not been 
mapped in Montana. The potential exists for 
using cryic Aridisols in the southwest corner of 
the state, but the area involved is small. If cryic 
Aridisols are established in Idaho, Montana 
could probably use them. Their establishment 
does not appear critical for soil mapping in Mon- 
tana. 

References 

Aguilar, R., E.R Kelly, and R.D. Heil. 1988. Effects of cultiva- 
tion on soils in northern Great Plains rangeland. Soil Sci. 
Soc. Am. J. 52:1081-1085. 

Anderson, D.W. 1979. Processes of humus formation and trans- 
formations in soils of the Canadian Great Plains. J. Soil Sci. 
30:79-84. 

Cannon, M.E. and G.A. Nielsen. 1984. Estimated productivity 
of range vegetation from easily measured soil characteris- 
tics. Soil Sci. Soc. Am. J. 48:1393-1397. 

Decker, G.L.1972. Automatic retrieval and analysis of soil 
characterization data. Ph.D. Thesis. Montana State Univer- 
sity, Bozeman, MT. 

Franzmeier, D.P., G.D. Lemme, and R.J. Miles. 1985. Organic 
carbon in soils of North Central United States. Soil Sci. Soc. 
Am. Proc.49:702-708. 

International Committee on Aridisols. 1989. Aridisols: Version 
6.0. Soil Conservation Service, USDA, Washington, D.C. 
Draft, April 13. 

Jenny, H. 1930. A study on the influence of climate upon the 
nitrogen and organic matter content of the soil. Univ. of 
Missouri Agric. Sta. Res. Bull. No. 152. 

McDaniel, P.A., M.E. Cannon, S.J. Harvey, and G.A. Nielsen. 
1982. Characterization of selected soils and vegetation on 
public lands in Montana. BLM Report. Plant and Soil Sci. 
Dept, Montana State Univ., Bozeman. 389 p. 

McDaniel, P.A. and L.C. Munn. 1985. Effect of temperature on 
organic carbon-texture relationships in Aridisols and 
Mollisols.SoilSci. Soc. Am .J.49: 1486- 1489. 

Nichols, J.D. 1984. Relation of organic carbon to soil properties 
and climate in the southern great plains. Soil Sci. Soc. Am. 
J. 48:1382-1384. 

Nimlos, T.J. and D. Tippy. 1981. Soil moisture and tempera- 
ture regimes in Beaverhead County, Montana. Research 
Note. MT For. and Sons. Exp. Sta., School of Forestry, 
Univ. of MT, Missoula, MT. 

Shields, J.A., E.A. Paul, R.J. St. Arnaud, and W.K. Head. 1968. 
Spectrophotometric measurement of soil color and its rela- 
tionship to moisture and organic matter. Can. J. Soil 
Sci.48:271-280. 

Sims, Z.R. and G.A. Nielsen. 1986. Organic carbon in Montana 
soils as related to clay content and climate. Soil 
Sci.Soc.AmJ. 50:1269-1271. 

Yonker, C.M., D.S. Schimel, E. Paroussis and R.D. Heil. 
i 1988.Patterns of organic carbon accumulation in a semi- 
arid shortgrass steppe, Colorado.52:478-482. 



92 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Effects of Minerology and Climate on the Development of Vertic 
Properties in Clayey Soils of Central and Eastern Canada 

C.R. de Kimpe*, C J. Acton, C. Wang, and M.C. Nolin 1 

Abstract 

Clay and heavy clay soils cover an area in excess of 1 million ha in the 
Lowlands of Central and Eastern Canada. These soils contain from 24 to 
40% swelling minerals, the amount of which increases from the northeast to 
the southwest. Bulk density values, at depth, were in the range of 1.42 to 
1.58 and 1.25 to 1.32 Mg.nv 3 in glaciolacustrine and marine sediments, re- 
spectively. Atterberg limits were 50 6% and 26 4% for the liquid and 
plastic limit, respectively. COLE index was generally low, but values higher 
than 0.15 also were recorded, indicating the potential for swell-shrink proc- 
esses. A regular distribution of the precipitation throughout the year does 
not allow the soils to dry to a state during which slickensides may develop. 
The St. Lawrence Lowlands clayey soils therefore did not exhibit vertic 
properties. At best, some profiles could be considered as showing proper- 
ties of incipient vertic development that may qualify them for vertic inter- 
grades to other orders. 



Introduction 

In Central and Eastern Canada, most of the 
clay and heavy clay soils are located in a corri- 
dor extending from the Great Lakes northeast- 
ward within the St. Lawrence Lowland physi- 
ographic region (Bostock, 1970). Smaller areas 
also occur in the Canadian Shield region, espe- 
cially in the Abitibi Uplands, Laurentian High- 
lands, around Lake St. Jean, and in the Appala- 
chian region. From an agricultural point of 
view, the St. Lawrence Lowlands contains by far 
the most important area of clay soils for crop 
production and as such has prompted the inter- 
est of soil scientists in studying the characteris- 
tics and land management problems of these 
soils. They present various properties that are 
similar to the clayey soils of Western Canada, 
but, at the same time, striking differences do 
occur. 

In this paper the general characteristics of 
Central and Eastern Canada clayey soils are 
summarized in relation to Vertisols and Vertic 
subgroups. 

General Description of the St. 
Lawrence Lowlands 

Physiography 

The St. Lawrence Lowlands Region extends 
from the southwestern tip of Lake Erie toward 
the northeast, to the Strait of Belle-Isle (Bos- 
tock, 1970) (Fig. 1). It is subdivided into the 
Western, Central, and Eastern Lowlands. The 
Western Lowlands is separated from the Cen- 



tral Lowlands by the Frontenac axis, an exten- 
sion of the Canadian Shield protruding into New 
York state and crossing the St. Lawrence river 
east of Kingston, Ontario. The Central Low- 
lands extends from Ottawa into the upper part 
of the Gulf of St. Lawrence, east of Quebec city. 
The Eastern Lowlands encompasses the Mingan 
and Anticosti islands and small areas around 
the Gulf and along the north-western coastline 
of Newfoundland. 

The Lowlands cover an area of approximately 
13 million ha, about 8 million ha in the western 
section and 5 million ha in the central and east- 
ern sections (Matthews and Baril, 1960). 
Twenty-five percent of the Lowlands consists of 
silt and clay lacustrine and marine sediments. 
In Ontario, 500,000 ha were surveyed as clay 
and heavy clay soils, and the corresponding area 
for Quebec is 550,000 ha. 

The major parent materials in the western 
section of the Lowlands have a glaciolacustrine 
origin. When the Wisconsinan glacier retreated, 
deglaciation first occurred in the Great Lakes 
area. Melting glaciers caused deposition in sev- 
eral fresh-water lakes (Dyke and Prest, 1987), 
Two of the larger and more recent lakes were 
Lake Whittlesey, a precursor of Lake Erie which 
formed about 13,000 years B.P., and Lake Iro- 
quois, a precursor of Lake Ontario which formed 



. De Kimpe and C. Wang, Agriculture Canada, LRRC, 
Central Experimental Farm, Ottawa, Ontario, Canada K1A 
OC6; C.J. Acton, Agriculture Canada, Ontario Institute of 
Pedology, Guelph, Ontario, Canada, NiH 6N1; and M. Nolin, 
Agriculture Canada, Complexe Scientfique, 2700 Einstein, 
Sainte-Foy, Quebec, Canada G1P 3W8. Contribution No. 89- 
70, LRRC, Ottawa. *Corresponding author. 



DE KIMPE, ACTON, WANG, NOLIN: EFFECTS OF MINEROLOGY AND CLIMATE/SOILS OF CENTRAL AND EASTERN CANADA 93 




Fig. 1 . Physiographic divisions of the St. Lawrence Lowlands 
(After Bostock 1970). 



Prepared by the Information Systems and Cartography Unit, 
Land Resource Research Centre, Agriculture Canada 



about 12,000 years B.P. The large quantities of 
glacial meltwater reworked and sorted much of 
the till, depositing this material as lacustrine 
clays in these early glacial lakes. This section of 
the Lowlands has never been invaded by the 
sea, and sedimentation therefore took place in 
fresh water. 



The Central section of the Lowlands is a large 
uniform plain, characterized by level to gently 
sloping terrain (0-3% slopes) but dissected in 
places by erosion by the St. Lawrence and adja- 
cent rivers (Lajoie, 1975). The elevation within 
this section of the Lowlands does not exceed 40 
m.a.s.l. Clay and heavy clay sediments were 



94 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



deposited during the invasion and the regres- 
sion of the Champlain sea, from approximately 
11,800 to 9,000 years BP (Prest, 1975). The 
marine sediments are generally heavy clay, of- 
ten calcareous with a plastic consistance. They 
were reworked by fluvial and water erosion and 
redeposited as fluviatile sediments. 

The Eastern section of the Lowlands does not 
contain much clayey material and will therefore 
not be discussed further. 

As most sediments in the Lowlands originate 
from the underlying and nearby rock formations 
which, over geological time, were weathered, 
then scoured by continental glaciers during sev- 
eral ice ages, it is appropriate to review briefly 
the bedrock geology of the Lowlands area. 

Bedrock Geology 

The Precambrian rocks belonging to the 
Grenville geological province are the eastern- 
most extension of the Canadian Shield and form 
the basal bedrock formations of Southern On- 
tario and a part of Southern Quebec (Martini et 
aL, 1970). These rocks outcrop north of an un- 
conformity running from Georgian Bay to the 
Thousand Islands in Ontario and northeast- 
ward to Labrador along the north shore of the 
St. Lawrence river. Grenville rocks are strongly 
metamorphosed and contain metasediments 
(quartzites, limestones, amphibolites, parag- 
neiss), metavolanics, and igneous rocks (an- 
orthosites, granites, diorites, syenites, gabbros, 
and pegmatites). 

West of the unconformity, the rock layers dip 
westward gently towards Michigan. Conse- 
quently, younger formations of Ordovician, Silu- 
rian, and Devonian sedimentary limestones and 
dolomites are found successively from east to 
west. Most of the rocks underlying the Low- 
lands in Quebec are also Ordovician dolomites, 
limestones, and shales, with small areas of 
Cambrian quartzites and sandstones. 

To the south, and separated from the Cana- 
dian Shield by the Logan's fault that runs along 
the St. Lawrence Valley, the Appalachian 
Mountains consist of Paleozoic rocks, Cambrian 
to 

Ordovician sedimentary rocks, and volcanics. 
The Appalachian piedmont contains limestone 
formations, whereas slate, schists and chloriti- 
oschists, quartzites, and sandstones are domi- 
nant on the plateaus. The rock formation of the 
Appalachian Mountains were weakly to moder- 
ately metamorphosed. 



Climate 

Temperature and Precipitation 

The mildest winter temperatures are re- 
corded near Lake Erie and St. Glair, and in the 
Niagara Peninsula, where January air tempera- 
ture averages -5"C (Environment Canada, 
1982a and b). This reflects the strong influence 
of the adjacent large water bodies, as considera- 
bly colder winters occur to the north and the 
east. January mean air temperature is -9'C at 
Montreal and Ottawa and -ll'C at Quebec City. 
The mean daily temperature reaches a maxi- 
mum of about 20-22'C in July throughout the 
Lowlands area. Based on limited data avail- 
able, frost penetration in the soil does not exceed 
40 cm and only for periods less than 3 months. 
East of Trois-Rivieres, and depending on the 
snow-cover, frost may penetrate deeper and stay 
for a longer period. 

Precipitation generally is distributed uni- 
formly during the year except in the northeast- 
em part of the St. Lawrence Valley, where a 
tendency towards maximum precipitation in the 
summer and early autumn exists. The mean 
total precipitation is 960 mm at London in the 
southwestern most part of the Western Low- 
lands, decreasing to 780 mm at Toronto, and 
then increasing northeastward throughout the 
Central Lowlands. Average precipitation is 880 
mm at Ottawa, 1060 inm at Montreal, and 1140 
mm at Quebec city (Environment Canada, 
1982a and b). 

The annual potential evapotranspiration in- 
creases from 400 - 500 mm in the northeast to 
500 - 600 mm west of Trois-Rivieres, Quebec, 
and to 600 - 700 mm in a few areas in south- 
western Ontario (Clayton et al., 1977). In the 
May-September period, seasonal water deficit is 
therefore unlikely in the northeastern part of 
the Central Lowlands, but it occurs in the south- 
western part of this section, as well as in the 
Western Lowlands. At the 50% probability 
level, for soils with 100 mm water storage capac- 
ity in the rooting zone, the deficit will be < 200 
mm, but it may be < 300 mm if the storage ca- 
pacity of the soil is reduced to 25 mm (Agricul- 
ture Canada, 1976). For 100 mm water storage 
capacity of the soil, the moisture index is 80-100 
in the northeastern and 60-80 in the southwest- 
ern part of the Lowlands (Ibidem). 

Soil Climate 

The soil temperature regime is cool to moder- 
ately cool boreal in the northeast and changes to 
mild mesic in the area around Montreal and in 



DE KIMPE, ACTON, WANG, NOLIN: EFFECTS OF MINEROLOGY AND CLIMATE/SOILS OF CENTRAL AND EASTERN CANADA 95 



southern Ontario (Clayton et al., 1977). At 
some places near Lake Erie and in the Niagara 
Peninsula, the soil temperature regime is mild 
to moderately warm mesic. 

Two stations will be used to illustrate the 
change in soil temperature over the year (Ibi- 
dem). At Fredericton, a station representing the 
climate of the northeastern part of the Low- 
lands, boreal soil climate is typically character- 
ized, at the 50 cm depth, by a summer tempera- 
ture range of 15-18*C from mid-June to mid- 
September and a winter temperature range of 0- 
2.C from early December to mid- April. Transi- 
tion in the spring and the fall covers periods of 
about 2 months each. 

At Harrow, a station in the southwestern part 
of the Lowlands, mesic soil climate is character 
ized, at the 50 cm depth, by a summer range of 
18-22 C from early July to late September and a 
winter range of 0-2'C from early January to the 
end of March. In the spring, temperature in- 
creases rapidly to reach 8'C in early May, 
whereas in the fall it drops below that tempera- 
ture by the end of November. 

The soil moisture regime is perhumid in the 
northeast from Quebec city to Trois-Rivieres 
(Clayton et al. 1977), which corresponds to peru- 
dic in the Soil Taxonomy (Soil Survey Staff, 
1975). The soil is either moist all year or seldom 
dry. The soil moisture regime is humid from 
Trois-Rivieres to Kingston and subhumid west 
of Kingston. These two regimes correspond to 
the udic moisture regime. Under the humid 
condition, the soil is not dry in any part as long 
as 90 consecutive days in most years, but there 
may be slight moisture deficits. 

In some years under the subhumid condition, 
the soil is dry in some parts when soil tempera- 
ture is > 5'C, which may produce significant 
moisture deficits. An analysis of the frequency 
and distribution of growing season dry spells in 
the years 1957-1979 indicated that short-period 
(10-20 day) dry spells occur at some time every 
year in southern Ontario (Brown and Wyllie 
1984). Dry spells of longer duration (4 weeks or 
more) occur about once every three years. 

Soils in the Lowlands 

Clas sification 

In the western section of the Lowlands, Gray 
Brown Luvisols (Hapludalfs) are the dominant 
soils on most glacial till and some lacustrine 
plains. Gray Brown Luvisols and Eutric Bruni- 
sols (Eutrochrepts) are commonly observed on 
deltaic materials and outwash plains. Luvic 



Gleysols (Aqualfs) and Humic Gleysols (Aquolls) 
are most frequently found on clayey lacustrine 
plains and on the poorly drained areas of the 
clayey textured till plains. 

In the central section of the Lowlands, Eutric 
Brunisols and Gray Brown Luvisols are most 
common on till plains. Melanic Brunisols 
(Hapludolls) and Humic Gleysols (Aquolls) are 
most common on clayey marine sediments, with 
the Brunisols usually present on the better 
drained landscape position. Humo-Ferric Pod- 
zols (Haplorthods) and Dystric Brunisols 
(Dystrochrepts) are the dominant soils on out- 
wash and sandy deltaic materials. 

In the eastern section of the Lowlands, Humic 
Gleysols are most common on marine clay mate- 
rial, whereas Dystric Brunisols and Humo-Fer- 
ric Podzols occur more frequently on all other 
parent materials. 

Soil Pedons 

Five representative pedons will be used to il- 
lustrate the nature, properties, and general 
behaviour of the clay and heavy clay soils of the 
Lowlands. 

Two pedons, of glaciolacustrine origin, were 
selected in the Western Lowlands. The Welland 
soil, sampled at an altitude of 180 m on an aban- 
doned farmland in the Niagara region near Fort 
Erie, Ontario, (42' 58' N and 79* 05' W), is devel- 
oped on reddish lacustrine clay derived from 
glacier reworking of reddish Hal ton clay till. It 
is classified as an Orthic Luvic Gleysol (Agricul- 
ture Canada Expert Committee on Soil Survey, 
1987). 

The Lincoln soil, sampled at an altitude of 
197 m on an improved pasture field in 
Haldimand - Norfolk region near Cayuga, On- 
tario, (43' 02' N and 79 45' W), is developed on 
deep-water glaciolacustrine clay. It is classified 
as an Orthic Humic Gleysol. 

Three pedons were selected in the Central 
Lowlands. The Dalhousie soil, developed on 
marine sediments, was sampled at an altitude 
of 92 m on an improved pasture of the Animal 
Research Centre at Ottawa, Ontario, (45* 17' N 
and 75* 45' W). The Ste Rosalie soil, also devel- 
oped on marine sediments, was sampled at an 
altitude of 33 m on an improved pasture at Ste 
Rosalie, Quebec, (45' 39' N and 72 55' W). The 
Providence soil, developed on fluviatile sedi- 
ments, was sampled at an altitude of 19 m on a 
field cropped for cereals at St. Ours, Quebec, (45' 
52' N and 73* 07 ! W). All three soils in the Cen- 
tral Lowlands are Orthic Humic Gleysols. 



96 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Methods 

Moist soil color was taken on the Munsell 
color charts. After air-drying, the soils were 
passed through a 2 mm-sieve. Most of the ana- 
lytical procedures for the physical and chemical 
analyses conformed to those described in the 
Manual of Methods for soil sampling and analy- 
sis (McKeague, 1978), and they will be referred 
to by their code number in the manual. 

- texture: pi(2.11); - bulk density: core method 
(2.21); - 33 kPa and 1.5 MPa moisture re- 
tention: core method (2.42) and pressure 
plate (2.43); - Atterberg limits: liquid (2.61) 
and plastic (2.62); - coefficient of linear ex- 
pansion COLE: core method (2.31); - pH: in 
0.01 M CaCl (3.11); - organic C: wet oxida- 
tion (3.613); - exchange capacity and ex- 
changeable cations: in 2M NaCl (3.31); - 
extractable sesquioxides: in dithionite-cit- 
rate-bicarbonate (3.51) and ammonium ox- 
alate (3.52) extracts; - surface area: EGME 
method (5.62). 

Results and Discussion 

Soil Properties 

Physical properties 

The soils from the central section, developed 
on sediments that had been, at least at some 
time, under marine influence, had a 5Y hue in 
the deeper horizons (Table 1). Soils from the 
western section, developed on glaciolacustrine 
sediments, were influenced by the nature of the 
parent material, resulting in hues of 10 YR in 
the Lincoln soil and 7.5 YR in the Welland soil. 
Organic matter gave a slightly redder hue, gen- 
erally 7.5 YR, in the surface horizon of most 
soils. 

All soils had a clay to heavy clay texture at 
depth. However, the Providence soil profile also 
had a Cg horizon with silty clay texture. This 
was related to the fluviatile origin of the parent 
material, which may be interspersed with 
coarser material. In the solum, texture was 
somewhat coarser than at depth, ranging from 
silty clay loam to clay, but only the Welland 
pedon showed evidence of clay illuviation (Table 
1). 

Bulk density (Table 1) of the deep horizons 
was greater in the glaciolacustrine (1,42 - 1.58) 
than in the marine sediments (1.25 - 1.32 Mg 
nr 3 ). Values in the same range were recorded in 
several soils developed on marine sediments in 



Quebec (De Kimpe and Mehuys, 1979). The dif- 
ference in bulk density between the two types of 
parent materials probably resulted from the 
sedimentation in quiet versus turbulent waters. 

Larger values for the water retention at 33 
kPa and 1.5 MPa were determined for the soils 
from southern Ontario and Ottawa (>40 and 25- 
35%) than for those located east of Montreal (30- 
40% and 20-35%) (Table 1). The difference is 
thought to be related to the different content of 
swelling minerals and fine clay in the soils of the 
two areas. 

Atterberg limits were comparable for all soils, 
with values of 50 6% for the liquid limit and 26 
+ 4% for the plastic limit (Table 1). It has been 
determined, for a large number of soils with a 
wide range in texture and organic matter con- 
tent, that Atterberg limits were related more to 
organic matter than to the clay content (De 
Kimpe et al., 1982). However, this conclusion 
was not verified in the clay and heavy clay soils 
of this study. 

The K sat values were closely related to soil 
structure and farming practices (Wang et al., 
1985). The soils developed on marine clay, 
Dalhousie and Ste Rosalie, had strongly devel- 
oped blocky structure and numerous biopores in 
the subsoil (Table 1). Therefore, the correspond- 
ing K sat values at depths > 50 cm were high. 
The other three clayey soils developed on gla- 
ciofluviatile, glaciolacustrine material and till 
and had weakly or moderately expressed soil 
structure and few biopores and, in the case of 
the Welland and Lincoln soils, also a high bulk 
density. These soils had low to very low K aat 
values in the subsoil horizons. The surface Ap 
horizons generally had a much greater K 
value than the underlying B horizon because 01 
the frequent plowing and lower bulk density. 

The COLE parameter was generally low in 
all soils of the Lowlands (Table 1). The lowest 
values, < 0.05, were recorded for the Ste Rosalie 
and Providence soils, although a value of 0.15 
was recorded in a Ste Rosalie soil profile (R. 
Asselin, personal communication, 1989). The 
COLE parameter was higher, 0.15, in the 
Dalhousie soil than in the Ste. Rosalie and 
Providence profiles. In the Welland and Lincoln 
soils from southwestern Ontario, values of 
about 0.10 and slightly higher values, respec- 
tively, were obtained. This range of COLE val- 
ues was likely related to the presence and the 
nature of the swelling minerals and to the high 
proportion of fine clay. 



DE KMPE, ACTON, WANG, NOLIN: EFFECTS OF MINEROLOGY AND CLIMATE/SOILS OF CENTRAL AND EASTERN CANADA 97 



Table 1. Selected physical properties of the live pedons. 


Horizon Depth Color Structure^ Clay Silt Bulk Moisture retention Atterberg limits COLE 
density 33 kPa 1.5 MPa liquid plastic index 
(cm) (%) (%) (Mg/m 3 ) (%) (%) (%) (%) 


(cm/h) 


Ste Rosalie (marine origin) Aquoll 
Ap 0-28 10YR3/1 str.f. sub. bl. 43 49 1.26 33 
Bgl 28-42 2.5 Y 5/1 wk.f.sub.bl. 53 44 1.41 34 
Bg2 42-56 5Y 5/1.5 mod.f. sub. bl. 65 35 1.35 40 
Bg3 56-100 5Y5/2 mod.f. sub. bl. 72 28 1.32 45 
Cg 100-140 5Y 5/1.5 str.f. to med.a.bl. 72 27 n.d. + n.d. 
Ckg 140+ 5Y6/1 str.med. to c.a.bl. 68 32 n.d. n.d. 

Dalhousie (marine origin) Aquoll 
Ap 0-17 10YR2/2 str.f. sub. bl. 43 41 n.d. 44 
Bgl 17-24 5GY6/1 wk.f.sub.bl. 31 37 1.37 37 
Bg2 24-53 5Y4/2 mod.med. sub. bl. 62 34 1.37 43 
BCg 53-75 5Y 4.571.5 mod.med. sub. bl. 65 34 n.d. 40 
Cg 75-100 5Y 4/1.5 str.med. to c. bl. 65 34 1.26 47 

Providence (fluviatile origin) Aquoll 
Ap 0-27 2.5 Y 3.572 massive 37 52 1.30 38 
Bgl 27-44 10YR 5.571 massive 44 49 1.35 41 
Bg2 44-70 2.5Y573 wk.f.sub.bl. 52 45 1.33 43 
Bg3 70-94 2.5Y 5.5/3 wk.med. sub. bl. 58 39 1.27 47 
Cgl 94-122 5Y5/2 mod.c. sub. bl. 59 37 1.25 n.d. 
Cg2 122-170 5Y 5.572 n.d. 47 50 n.d. n.d. 

Welland (reworked till) Aqualf 
Ap 0-15 10YR4/2 mod.c. gran. 48 45 1.16 38 
Btgl 15-34 7.5YR4/2 mod.c. sub. bl. 69 28 1.40 51 
Btg2 34-43 7.5YE5/2 mod.c. col. 77 22 1.42 52 
Ckg 43+ 7.5YR5/2 wk.c. col. 73 26 1.42 48 

Lincoln (lacustrine origin) Aquoll 
Ap 0-15 10YR 3.5/3 mod.c. bl. 48 47 1.28 43 
Bgf 15-40 10YR4/1 wk.cbl. 62 34 1.49 36 
BCg 40-50 10YR4/1 wk.c.bl. 57 37 n.d. n.d. 
Ckg 50+ 10YR572 mod.c. bl. 60 38 1.58 33 


21 
25 
30 
31 
n.d. 
n.d. 

18 
15 
29 
27 
26 

23 
25 
26 
28 
n.d. 
n.d. 

20 
31 
36 
35 

27 
25 
n.d. 
21 


43 25 
47 21 
55 24 
55 25 
n.d. n.d. 
n.d. n.d. 

56 40 
n.d. n.d. 
45 27 
n.d. n.d. 
47 23 

44 28 
44 25 
50 25 
54 25 
55 24 
n.d. n.d. 

46 28 
62 26 
n.d. n.d. 
56 28 

43 27 
60 26 
44 22 
51 22 


0.04 
0.04 
0.05 
0.05 
n.d. 
n.d. 

0.11 
0.09 
0.15 
0.15 
0.15 

0.04 
0.05 
0.05 
0.05 
n.d. 
n.d. 

0.07 
0.09 
0.10 
0.08 

0.10 
0.20 
0.10 
n.d. 


12 

7 
23 
29 
n.d. 
n.d. 

15 
6 
12 
20 
35 

58 
2 

9 
n.d. 

33 
n.d. 


+ n.d. ss not determined 
= mod. = moderate, str. = strong, wk. = weak, f. = fine, med. = medium, c. = coarse, sub. = subangular, bl. = blocky, col. = columnar. 


Chemical properties 




Table 2. 


Selected chemical properties of the profiles. 


All soil profiles had pH values ^ 6.8 in the 
deeper horizons (Table 2). In the Welland and 


Horizon 


pH Org C 
(in 0.01 M 


CEC 


Fe, 


Fe o 


Ala 


Al o 


Lincoln soils, values > 7.4 at depth were related 


CaCy (%) 


(cmol/kg) (%) (%) (%) (%) 


to the presence of carbonates in the Paleozoic 


Ste Rosalie 
Ap 


6.2 


2.3 


27 


0.7 


0.4 


0.2 


0.2 


rock formations from which the sediments were 


Bgl 


6.5 


0.4 


27 


0.9 


0.4 


0.2 


0.2 


derived. In the Ste Rosalie and Providence soils, 


Bg2 
Bg3 


6.8 

7.0 


0.3 
0.3 


28 
25 


0.9 
0.9 


0.5 
0.5 


0.3 
0.2 


0.3 
0.3 


pH values of 7.5 were attributed to the presence 
of fossil shells in these marine sediments. 


Cg 
Ckg 


7.5 
7.5 


0.3 
n.d. 


25 
n.d. 


0.9 
n.d. 


0.5 
n.d. 


0.2 
n.d. 


0.3 
n.d. 


Accumulations of organic carbon were mostly 


Dalhousie 
Ap 


6.6 


9.2 


56 


0.6 


n.d. 


0.2 


n.d. 


restricted to the surface horizon, as is commonly 


Bgl 


6.9 


1.0 


23 


0.5 


n.d. 


0.1 


n.d. 


observed in poorly drained soil pedons (De 


Bg2 
BCg 


6.9 
6.9 


0.6 
0.4 


34 
27 


0.9 
0.7 


n.d. 
n.d. 


0.2 
0.2 


n.d. 
n.d. 


Kimpe et aL, 1979). Content ranged from 1.9 to 


Cg 


6.8 


0.2 


27 


0.6 


n.d. 


0.2 


n.d. 


9.2% in the Ap horizon, whereas it decreased to 


Providence 


0.4 to 1% in the Bg horizons. 


Ap 
Bel 


6.0 
5.9 


1.9 
0.9 


28 
27 


1.6 
1.3 


1.1 
0.9 


0.3 
0.3 


0.4 
0.4 


Cation exchange capacity was generally < 30 
cmol/kg" 1 . The high value for the Dalhousie Ap 


**&* 
Bg2 
Bg3 
Cgl 


6.5 

7.1 
7.3 


0.2 
0.1 
0.1 


27 
28 
27 


1.0 
1.0 
1.0 


0.4 
0.5 
0.5 


0.3 
0.3 
0.3 


0.3 
0.4 
0.4 


horizon was related to its high organic matter 


Cg2 


7.5 


0.1 


23 


1.0 


0.4 


0.3 


0.3 


content. The most abundant cation on the ex- 


Welland 
















change sites was Ca 2+ , followed by Mg 2+ . 


Ap 
Btgl 


5.2 
6.5 


2.4 
0.6 


17 
29 


1.7 
1.7 


0.5 
0.3 


0.2 
0.2 


0.2 
0.2 


Dithionite-citrate-bicarbonate extractable Fe 


M-rvfj*. 

Btg2 


7.3 


0.6 


31 


1.5 


0.2 


0.2 


0.2 


content ranged from 0.5 to 3.1%, with the high- 


Ckg 


7.7 


n.d. 


29 


1.2 


0.2 


0.1 


0.2 


est values being found in the Southern Ontario 


Lincoln 


6.4 


3.3 


n.d. 


1.7 


0.5 


0.3 


0.2 


soils. This reflected in the soil color (Table 1). In 


Bgf 


e!o 


0^9 


27 


3.1 


0^7 




0^2 


the Welland and Providence soil pedons, the 


BCg 
Gig 


7.1 
7.9 


0.5 
0.5 


n.d. 
25 


1.7 
1.3 


0.5 
0.2 


oi 

0.1 


0.2 
0.1 


DCB extractable Fe content decreased with 


^Jo 
















depth, whereas there was a maximum at the Bg 



98 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



horizon level in the three other pedons. The 
oxalate-extractable Fe content was <!.!%. The 
percentage of Fe present as crystalline Fe oxides 
( Fe DCB " Fe ) r ang ed fr m 0.4 to 2.4%. The 
amount was highest in the Bg horizons of all 
pedons and also 2 to 4 times larger in the soils 
from southern Ontario than in those from the 
Ottawa-St. Lawrence Lowlands. 

Extractable Al content was <0.4% in all soils 
and generally similar in both extracts. 

Mineralogical properties 

In the northeastern part of the Central Low- 
lands, illite and chlorite were the dominant 
phyllosilicates in the clay fraction and their sum 
amounted to about 40% throughout the profiles 
(De Kimpe et al, 1979). Smectite and vermicu- 
lite contents were about 15 and 9%, respec- 
tively. In the Montreal area of the Central 
Lowlands, illite plus chlorite were the major 
phyllosilicates with an average sum content of 
about 23 and 30% at the surface and at depth, 
respectively. Smectite content was about 18% 
at the surface and 16% at depth, whereas ver- 
miculite was about 11% throughout the pedons 
(Ibidem). 

In both regions, quartz, about 20%, feldspar, 
with 5-10%, and other primary minerals such as 
hornblende, present in minor to low amounts in 
the clay-size fraction, reflected the origin of the 
sediments and underlined the effect of the conti- 
nental glaciers that produced rock flour when 
they scoured the rock formations. The persis- 
tance of weatherable minerals and the small 
changes in the mineralogical composition dem- 
onstrated the low to moderate rate of weather- 
ing in these cool to moderately cool and wet soils 
(Kodama, 1979; De Kimpe and Martel, 1986). 

In the Dalhousie soil of the Central Lowlands, 
smectite content of the clay fraction increased to 
35% at the surface from 22% at depth, whereas 
the micas decreased to 8% at the surface from 
18% at depth indicating pedogenic weathering 
of mica to smectite (Ross et al., 1987). 

Swelling minerals and illite were the major 
phyllosilicates in the Lincoln soil (unpublished 
data) of the Western Lowlands, with the swel- 
ling minerals dominant in the Ap horizon. In 
the lower horizons, illite was associated with 
small to trace amounts of chlorite, quartz and 
feldspar. A similar composition was found in the 
Welland soil (unpublished data). Swelling min- 
erals and illite were dominant throughout the 
pedons and some chlorite was present in the BC 
horizon. Feldspar and quartz were also present 
in small amounts in the clay fraction of this soil. 



Specific surface area was determined for clay 
soils from the central and western sections of 
the Lowlands. Average values of 245 and 210 
m 2 g" x were obtained for the clay fraction of 
Gleysolic soils in the Montreal and Quebec ar- 
eas, respectively (De Kimpe et al., 1979). Mul- 
tiple regression analysis of 34 samples of heavy 
textured Humic Gleysols from southwestern 
Ontario produced a value of 207 m 2 ^ 1 for the 
clay fraction (Evans, 1982). These figures sup- 
ported the mineralogical composition obtained 
from X-ray diffraction. 

Potential for Vertic Properties 
Development 

Ahmad (1983) reviewed the literature on Ver- 
tisols and summarized their outstanding fea- 
tures. Although these soils are found under a 
wide range of climatic conditions, one common 
characteristic is the seasonally of the precipita- 
tion allowing for annual wetting and drying of 
the solum. These soils have a clay content of > 
35%, most often montmorillonitic, that causes 
pronounced changes in volume with changes in 
water content, resulting in deep, wide cracks in 
the dry seasons. The soils are plastic and have a 
sticky consistency when wet. Bulk density is 
high in dry soils and K sat is low in wet soils. 

There is an appreciable increase in microele- 
vation of the soil surface as the soil becomes wet, 
and, when it dries, subsidence occurs and cracks 
develop. In most cases, infilling of the cracks 
with surface material creates pedoturbation and 
tends to restrict strong profile development. 
Internal stresses due to overburden pressure, as 
well as shrinking and swelling of the subsoil, are 
at the origin of slickenside formation, and a gil- 
gai microrelief is commonly observed at the soil 
surface. Surface cracking and slickensides are 
the major features taken into consideration for 
the classification of Vertisols (Soil Survey Staff, 
1975; Soil Survey Staff, 1987). 

The clay and heavy clay soils of the St. Law- 
rence Lowlands of Central and Eastern Canada, 
based on their texture, were well within the 
range of the Vertisols. From the mineralogical 
point of view, although they were not domi- 
nantly montmorillonitic, these soils contained 
significant amounts of swelling minerals which 
are important to the shrink-swell process. They 
also contained illite that, under adequate 
weathering conditions, transforms to swelling 
minerals. 

All soils but Welland had a strong to moder- 
ate blocky structure at depth, which was inher- 



DE KIMPE, ACTON, WANG, NOLIN: EFFECTS OF MINEROLOGY AND CUMATE/SOILS OF CENTRAL AND EASTERN CANADA 



99 



ited from the parent material. Welland had 
weak columnar structure at depth. In the B 
horizons, the structure was affected by pe- 
dogenesis. In the three pedons from the Central 
section, soil structure became finer and less 
strongly expressed than in the C horizons, 
whereas coarse blocky structure was common in 
the B horizons of the two pedons from the West- 
em Lowlands. Blocky structure is conducive to 
the development of vertic properties (McCor- 
mack and Wilding, 1973) through the wetting 
and drying cycle of the peds. 

However, the limiting factor for the develop- 
ment of vertic properties in this region of Can- 
ada is soil climate. As detailed in an earlier sec- 
tion, precipitation is rather uniformly distrib- 
uted throughout the year, but with limited 
probabilities for dry spells. Under the perhumid 
soil moisture regime prevailing in the northeast- 
ern part of the Central section, cracking pat- 
terns developed only during exceptionally dry 
summers, such as in 1988, and the cracks did 
not penetrate very deeply in the soil. 

The humid soil moisture regime in the Ot- 
tawa-Montreal area is characterized by dry 
spells of slightly longer duration so that crack- 
ing may occur more frequently, but again not to 
a sufficient degree to develop permanent vertic 
features. Vertical soil displacement of 10% 
throughout the year has been monitored in a 
Rideau clay soil (K. Wires, personal communica- 
tion, 1989). The cracking pattern was also par- 
ticularly well developed during the dry summer 
of 1988. 

The longest dry spells are regularly observed 
in the subhumid western section of the Low- 
lands. Cracking patterns develop every year 
and in 1988, the cracks developed vertically to a 
depth of 75-90 cm (K. Wires, personal communi- 
cation, 1989). It is not clear, however, whether 
the cracks are permanent from year to year or 
whether sufficiently strong internal stresses 
occur to induce development of slickensides. 
The permanent development of such vertic fea- 
tures has not been reported to date in the clayey 
soils of Eastern Canada. 

Literature Cited 

Agriculture Canada. 976. Agroclimatic atlas. Agrometeor- 
ology Research and Service Section, CBRI, Ottawa, 
Ontario, 17 maps. 

Agriculture Canada Expert Committee on Soil Survey. 
1987. The Canadian system of soil classification. 2nd 
edition. Agriculture Canada Publ.1646, Ottawa, 164 p. 



Ahmad, N. 1983. Vertisols. p.91-123 In: L.P. Wilding, N.E. 
Smeck and G.F. Hall (eds). Pedogenesis and soil taxon- 
omy. Vol II. The Soil orders. Elsevier PubL, New York. 

Bostock, H.S. 1970. Physiographic subdivisions of Can- 
ada, p. 10-30 In: R.J.W. Douglas (ed) Geologic and Eco- 
nomic Minerals of Canada. Geological Survey of Can- 
ada. Economic Geology Report no. 1, Ottawa. 

Brown, D.M. and W.D. Wyllie. 1984. Growing season dry 
spells in southern Ontario. Climatology Bull. 18:15-30. 

Clayton, J.S., W.A. Ehrlich, D.B. Cann, J.H. Day, and I.B. 
Marshall. 1977. Soils of Canada. Vol. I. Soil Report. 
Can. Dept. of Agriculture, Ottawa, 243 p. 

De Kimpe, C.R., M. Bernier-Cardou and P. Jolicoeur. 
1982. Compaction and swelling of Quebec soils in rela- 
tion to their soil-water properties. Can. J. Soil Sci. 
62:165-175. 

De Kimpe, C.R., M.R. Laverdi re and Y.A. Martel. 1979. 
Surface area and exchange capacity of clay in relation 
to the mineralogical composition of gleysolic soils. Can. 
J. Soil Sci. 59:341-347. 

De Kimpe, C.R. and Y.A. Martel. 1986. Retrospective des 
recherches sur la genese et la classification des sols au 
Quebec. Cahiers de 1'ACFAS no 37, 88-114. 

De Kimpe, C.R. and G.R. Mehuys. 1979. Physical proper- 
ties of gleysolic soils in the lowlands of Quebec. Can. J. 
Soil Sci. 59:69-78. 

Dyke, A.S. and Prest, V.K. 1987. The late Wisconsinan 
and Holocene history of the Lauren tide ice sheet. Geo- 
graphie Physique et Quaternaire. 61:237-263. 

Environment Canada. 1982a. Canadian climate normals 
1951-1980. Temperature and Precipitation Ontario. 
Canadian Climate Program, 254 p. 

Environnement Canada. 1982b. Canadian climate nor- 
mals 1951-1980. Temperature and Precipitation Que- 
bec. Canadian Climate Program, 216 p. 

Evans, L.J. 1982. Cation exchange capacities and surface 
areas of Humic Gleysolic Ap horizons from southwest- 
ern Ontario. Can. J. Soil Sci. 62:291-296. 

Kodama, H. 1979. Clay Minerals in Canadian Soils: their 
origin, distribution and alteration. Can. J. Soil Sci. 
59:37-58. 

Lajoie, P. 1975. Agricultural lands in southern Quebec: 
distribution, extent, and quality. Agriculture Canada 
Publ. 1556, Ottawa, 62 p. 

Martini, I.P., R. Protz and W. Chesworth. 1970. The rocks 
and soils of southern Ontario. Dept. of Soil Science, 
Univ. of Guelph, 11 p. 

Matthews, B.C. and R.W. Baril. 1960. The soils of the 
Great Lakes - St. Lawrence Lowlands. Agricultural 
Institute Review, 37-40. 

McCormack, D.E. and L.P. Wilding. 1973. Proposed origin 
of lattisepic fabric. Pages 761-771. In G.K. Rutherford 
(ed) Soil Microscopy 4th Int. Working Meeting on Soil 
Micromorphology. The Limestone Press, Kingston, 
Ontario. 

McKeague, J.A. (ed). 1978. Manual of methods for soil 
sampling Office, Washington D.C. 

Soil Survey Staff. 1987. Keys to Soil Taxonomy (third 
printing) SMSS technical monography no 6, Ithaca, 
New York. 

Wang, C., J.A. McKeague, and K.D. Switzer-Howse. 1985. 
Saturated hydraulic conductivity as an indicator of 
structural degradation in clayey soils of Ottawa area, 
Canada. Soil and Tillage Res. 5:19-31. 



100 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Procedures and Rationale for Development of Adequate and 
Comprehensive Field Soil Data Bases 

R. Langohr 1 
Abstract 

In most pedon data bases, data are missing for an unequivocal classifica- 
tion of the soils or for an exact land evaluation of the soil site. The paper 
describes the procedures and rationale followed by a team of pedoiogists of 
the University of Ghent in order to elaborate new guidelines for the devel- 
opment of more adequate and comprehensive field soil data bases. The sur- 
veyor's attention is particularly asked for features such as, producing more 
numerical data, to draw detailed figures of both vertical and horizontal 
sections, to inform about errors, temporal variabiliiy, activity status, spa- 
tial variability, related distributions and to provide his opinion about the 
genesis of the characteristics he describes. The final goal of the project is to 
raise the quality of the field data base to a level where it would permit to 
understand better the soil-ecosystem dynamics and where it would allow 
more applications than it is today. 



Introduction 

A series of research projects performed in col- 
laboration with M.Sc and Ph.D fellows at the 
International Training Centre for Post Gradu- 
ate Soil Scientists at the University of Ghent 
has allowed intensive evaluation of soil data 
bases for soil classification and land evaluation. 
Published soil pedon data including soil site, soil 
profile and soil analytical data (the whole set of 
these data is called here the "pedon data base" 
or PDB) were rigorously applied to soil classifi- 
cation (mainly the USDA and the FAO-Unesco 
soil taxonomies) and land evaluation diagnostic 
criteria (Adiwiganda, 1986; Lopulisa, 1986; 
Lopulisa et al., 1986a and 1986b; Estoista, 
1988; Vaca, 1988). These investigations showed 
that in most PDBs, data are missing for an un- 
equivocal classification of the soil or for an exact 
land evaluation of the soil site. 

Similar problems, but of a larger magnitude, 
are met when PDBs are consulted for interpre- 
tations and for advise on plant and faunal ecol- 
ogy, ecosystem dynamics and environmental 
management. Furthermore, almost all PDBs 
are particularly inadequate in terms of field 
data needed for soil genesis investigations and 
paleoenvironmental reconstruction. 

As the soil is largely that part of the earth 
where the biosphere and the geosphere meet 
and interact, it represents an essential source of 
information for many disciplines. That is why 
the soil surveyors, among all scientists working 
in the field, should have knowledge of the larg- 
est spectrum of disciplines; soil survey reports 

department of General Pedology, University of Gnent, 
Rrijgslaan 281, B-900-Gnent, Belgium. 



thus can contribute data for studies in ecology, 
environment, geomorphology, Quaternary geol- 
ogy, archaeology etc., if the reports are suffi- 
ciently comprehensive and adequate. 

The above mentioned criticisms can be sum- 
marized in two statements. 

1. The standard guidelines for profile descrip- 
tion and analysis are written in the frame- 
work of traditional soil cartography and as- 
sociated soil classification system(s). If the 
existing guidelines are followed rigorously 
by the surveyors and laboratory staff the 
PDBs would be adequate for these pur- 
poses; however, they are not and therefore 
professiona qualification of the scientists 
can be questioned here. 

2. When it comes to other types of research 
projects, the standard guidelines do not 
provide all the required data. Yet in many 
projects these same standard guidelines 
are followed with as a result numerous 
unsatisfied customers of the PDBs. There- 
fore new guidelines should be developed 
and completed with special training of the 
surveyors. 

Concerning the second statement the group 
of the University of Ghent is trying to improve 
the method of soil data recording in the field. 

Our five main goals are: 

1. To make the field pedon data base (FPDB) 
more adequate for applications known at 
present, 

2. To make the FPDB more comprehensive 
in a way that it may serve in the future as a 
basic source of information for more disci- 
plines than those for which it is used today, 



LANGOHR: PROCEDURES AND RATIONALE FOR DEVELOPMENT OF FIELD SOIL DATA BASES 



101 



3. To raise the quality of the FPDB to a level 
at least equivalent to the present-day soils 
laboratory data bases, 

4. To raise the quality of the FPDB to a level 
where the computer can be used more effi- 
ciently, instead of lowering its quality in 
order to make it more computer-compatible 

5. To make the FPDB more accessible to non- 
pedologists. 

Summary of Procedures and 
Rationale for Field Guidelines 

This paper describes guidelines developed by 
University of Ghent investigators for the pur- 
pose of raising the quality, comprehensiveness 
and adequacy of FPDBs. The term standard 
guidelines used here refers to those guidelines 
for soil profile description elaborated for sustain- 
ing regional or national soil survey activities 
(e.g. Soil Survey Staff 1951, Hodgson 1976, FAO 
1977, FitzPatrick 1977). 

Parent material, groundwater table, soil 
color, structure, etc. are called soil characteris- 
tics. Further distinctions of these characteristics 
are called "aspects" of soil characteristics, such 
as 1) origin-, age-, texture-, and mineralogical 
composition- of parent material, 2) origin-, de- 
gree of stagnation-, chemical composition- and 
color- of groundwater, 3) moist-, dry-, 4) rubbed 
soil color and 5) size-, shape- and grade for soil 
structure. 

As there are numerous different types of ap- 
plications of PDBs, the surveyor should select 
from the guidelines those characteristics which 
are relevant for his investigation(s). It must be 
clear that the goal of the guidelines is that not 
all soil characteristics and aspects of the charac- 
teristics should be described systematically. 

No special effort should be made to avoid over- 
lapping of information. Data about a particular 
characteristic may be asked for in several para- 
graphs of the description; if it is relevant in the 
study, information should be provided each 
time, or reference should be made to the para- 
graph where the information is provided in suffi- 
cient detail. 

Exact and detailed references should be men- 
tioned for all the data extracted from publica- 
tions, reports, maps, photographs, and figures. 
By noting the existence of this information the 
user of the PDB will be able to consult these 
documents whenever more detailed information 
is needed. 



Pre-established classes such as "few-, com- 
mon-, abundant-", "fine-, medium-, coarse-", 
"abrupt-, clear-, gradual-, diffuse-", should be 
avoided as much as possible and should be re- 
placed by direct measurements or estimates 
expressed as numerical data. At a time when 
more and more data are stored in computers, it 
is illogical to force the field surveyor to provide 
part of his data in pre-established classes; this 
represents a serious loss of precision and in fact 
amputates the use of a computer to calculate 
classes for us. It is still possible to derive "class" 
data from numerical data whenever that is use- 
ful or necessary. 

Whenever a feature is estimated or meas- 
ured, an error will be made. In pedology, most 
guidelines for soil description, and even the 
handbooks describing laboratory methods, fo- 
cus little attention on absolute and relative er- 
rors associated with estimates and/or measure- 
ments. The absence of this information reduces 
the data's value. 

Special attention should be given to records of 
the temporal variability (= TV), the activity 
status (= AT) and the spatial variability (= SV) 
of many of the characteristics. 

Some soil characteristics like surface crust- 
ing-, cracking pattern-, mottling- and grade of 
aggregate development- may vary seasonally. 
Other characteristics such as erosion- and inun- 
dation- may have variability over longer time 
spans. Aspects of these characteristics such as 
"duration-" and "frequency-" provide informa- 
tion about temporal variability (= TV). 

The concept of Activity Status (= AS) is to as- 
certain the degree of temporal variability; here 
the soil scientist would note if the observed char- 
acteristic is the result of a still active process or 
not; in latter conditions the characteristic (e.g. 
microrelief-, gullies-, mottling-, biogalleries-) 
will be considered to be a relict, probably associ- 
ated to past and different environmental condi- 
tions. 

Many recent papers focus attention on spatial 
variability (= SV) of soil characteristics. Hand- 
books for recording soil data guide surveyors in 
the collection of spatial variability data for site 
descriptions as well as for profile descriptions. 
For example, information about vegetation-, 
erosion-, and microrelief should be described 
separately for the spot of the profile itself and for 
the surroundings. In many profile descriptions it 
is not clear what information belongs to the 



102 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



exact location of the profile, and what informa- 
tion concerns the surroundings or even the en- 
tire mapping unit; these data are equally impor- 
tant but the spatial position should be specified 
and specific. 

As for variability within the pro file, it is inter- 
esting to quote Soil Taxonomy (Soil Survey 
Staff, 1975, p. 3) : 

"A pedon has the smallest area for which 
we should describe and sample the soil to 
represent the nature and arrangement of 
its horizons and variability in the other 
properties that are preserved in the 
samples... Its lateral dimensions are large 
enough to represent the nature of any hori- 
zons and variability that may be 
present.. .The area of a pedon ranges from 1 
to 10 square metres, depending on the vari- 
ability of the soil." 

It is our experience that soil surveyors very 
seldom check this variability of the soil charac- 
teristics over an area of 10 square metres, or 
even one square metre. Consequently we know 
little about this aspect of the soil characteristics. 
Guidelines should systematically ask for this 
information and not simply mention it in an in- 
troductory chapter. 

Much more effort should go to the explanation 
of the site and profile characteristics beyond the 
usual profile description. This part of the data 
base can eventually be written in a way that 
also non-professionals can read and understand 
the text. Important information about the "ori- 
gin" of site and soil characteristics can be ob- 
tained from the field surveyors. It is an 
enormous loss if surveyors with their huge 
amount of unique professional skills are not 
asked to record their interpretations, thoughts 
and doubts about the genesis, the origin and the 
dynamics of the features described. For many 
characteristics, laboratory data will not provide 
much of this type of information. In a recent call 
for more Field Soil Science R.B. Daniels (1988) 
wrote :"A soil sample in the laboratory is noth- 
ing more than a bag of dirt. That bag of dirt 
becomes a useful research sample only if we 
know the field relations it represents". 

Inclusion of these extra data focuses the sur- 
veyor's attention on additional relevant field 
data. If answers are needed from laboratory 
data, sampling is also focused and better de- 
signed to obtain exactly those laboratory data 
that provide the best or most usable informa- 
tion. 



Particular emphasis should also go to the de- 
tection, description and explanation of related 
distributions (- RD) (e.g. silt capping on stones, 
pseudo-gley along pores, position of clay coat- 
ings-, mangans-, ferrans-, secondary carbon- 
ates-...). Except for a few characteristics like 
coatings, these data are largely absent from 
most handbooks for soil description. It is evident 
that related distributions, if present, should be 
explained. 

In describing soil profiles, as much attention 
should go to the observation and description of 
horizontal sections as to the traditional vertical 
sections. Some characteristics such as cracking-, 
earthworm galleries-, root density-, mottling 
pattern- can only be described and quantified 
correctly by the observation of horizontal sec- 
tions. Numerous studies of horizontal sections of 
soil profiles make us conclude that by adding 
this type of observation the surveyor obtains a 
much more complete and often even a different 
picture of the soil than by only observing the 
vertical profile walls. 

Numerous figures should be part of the FSDB. 
Many characteristics of the soil and its environ- 
ment can be described best with the help of fig- 
ures. Topographical location-, lithostratigraphy 
of the parent material and of the substratum-, 
geomorphology-, microrelief-, associated soils-, 
vegetation-, soil surface form-, soil horizon dis- 
tribution in the vertical profile section-, distribu- 
tion patterns of particular soil characteristics 
such as galleries-, mottling-, carbonate accumu- 
lations- in the horizontal as well as in the verti- 
cal profile sections, are all examples of soil char- 
acteristics for which the description can be dras- 
tically improved and yet at the same time con- 
siderably simplified by adding good quality fig- 
ures. Text and words should be complementary 
to the figures rather than the reverse. Figures 
are of enormous help in the description of the 
sometimes complicated spatial variability and 
related distributions of some soil characteristics. 
They can show the exact location of sampling 
sites. Furthermore, drawing figures forces the 
pedologist to observe very precisely a series of 
selected soil characteristics and to make deci- 
sions regarding such themes as, are the ob- 
served roots dead- or alive-, are the roots concen- 
trated along the ped faces- or not-, what is the 
orientation of the stones-, etc. If figures are im- 
portant, then drawing of profile sections, 
transects and block diagrams should be part of 
the basic training for a field surveyor. 



LANGOHR: PROCEDURES AND RATIONALE FOR DEVELOPMENT OP FIELD SOIL DATA BASES 



103 



Special effort should go into making the data 
base useful to non-specialists to consult the data 
base. Except for a chapter "Brief General De- 
scription of the Profile" no particular effort is 
made for this in the standard descriptions. Simi- 
lar paragraphs should be provided throughout 
the data base, including an explanation for each 
soil horizon. Glossaries also help the non-spe- 
cialist in collecting information from the SDB. 

Acknowledgements 

The author wishes to thank the Belgian Na- 
tional Science Foundation for a grant which al- 
lowed him to assist to the ISCOM VI Meeting. 

Literature Cited 

Adiwiganda, R., 1986. Classification problems of selected 
soil profiles from North-Sumatra (Indonesia) according 
to the USDA soil taxonomy system. Unpubl. M.Sc. 
diss., Univ. Gent. 

Daniels, R.B., 1988. Pedology, a field or laboratory sci- 
ence? Soil Sci. Soc. Am. J. 5215, p. 1518-1519. 

Estoista, R.V., 1988. Adequacy of selected profile data 
from Luzon (Philippines) for classification in the USDA 
(1975) and FAO (1974, 1987) soil taxonomic systems. 
Unpubl. M.Sc. diss., Univ. Gent. 



FAO, 1977. Guidelines for soil profile description (2nd 
Ed.), 66 p, Food and Agriculture Organization of the 
United Nations, Rome. 

FitzPatrick, E.A., 1977. Soil Description. 66p., Dept. Soil 
Sci., Univ. Aberdeen. 

Hodgson, J.M., 1976. Soil Survey Field Handbook, de- 
scribing and sampling soil profiles. Soil Surv. Techn. 
Monograph No. 5, U.K. 

Lopulisa, C., 1986. Critical analysis of the soil characteri- 
zation in soil resource inventories from the tropics, 
with special reference to the use of the U.S. soil taxon- 
omy (1975) system. Unpubl. Ph.D. diss., Univ. Gent. 

Lopulisa, C., Langohr, R., Msanya, B.M., 1986a. Analysis 
of the U.S. Soil Taxonomy (1975) for the classification 
of soils in the Tropics, with special reference to the 
Tropept suborder. Transactions XIII Congres Int. Soc. 
of Soil Science, vol. Ill, 1196-1197, Hamburg. 

Lopulisa, C., Langohr, R., Msanya, B.M., 1986b. Adequacy 
of the existing pedon data of organic soils for the classi- 
fication among the Histosols (USDA Soil taxonomy, 
1975). Proc. Int. Symp. on Peat Soils (in press), 
Yogyakarta, Indonesia. 

Soil Survey Staff, 1951. Soil Survey Manual.U.S. Dep. 
Agr. Handb. 18. U.S. Govt. Printing Office, Washing- 
ton, D.C. 

Vacca, A., 1988. Soil profile data from Italy: their ade- 
quacy for classification in the USDA (1975) and FAO 
(1974, 1987) taxonomies and for land suitability esti- 
mation. Unpubl. M.Sc. diss., Univ. Gent. 



104 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



The Soils of the Desertie and Arid Regions of Chile 

Walter Luzio-Leighton 1 

Abstract 

The anticiclone from the southwest Pacific originates a high pressure 
system that is the main cause of the arid and desertic climate in the north of 
Chile. The soil moisture regime is aridic, with the exception of some close 
basins in the Al tip Ian, where it is aquic. The soil temperature regime is 
isothermic near the coast, thermic in the central part, and cryic in the 
Andes and the Altiplano. 

A distinction is made between the soils from, desertic conditions and 
those from arid and semiarid regions, and the limit in located near the par- 
allel 29* S.L. This arbitrary boundary has been placed in the attempt to 
separate the desertic soils from the arid ones on the basis of the intensity 
and the natur of pedogenic processes that have affected both. 

In the desertic region, the soils with minimum profile development are 
Cryorthents in the Altiplano and Torriorthents in the central "pampas'* and 
coastal area. The soils with structure or color B horizons are Camborthids 
formed on finer sediments than the former. The volcanic materials show a 
very slight development, so that no andic properties have been identified 
yet. Some highly saline soils show a very hard surface crust, which makes 
difficult their placement in Soil Taxonomy. 

In the arid and semiarid regions, the soils present a higher degree of 
development than desert soils. In the coastal area, very well developed illu- 
viation features are found in the soils from marine terraces; thus, argillic 
and natric horizons are frequent. Towards the innerland there is a hilly 
lanscape, and the soils are not well known. Some observations indicate the 
presence of cambic and argillic horizons in soils derived either from lime- 
stone or from granite. 



Introduction 

In a first approach it is necessary to distin- 
guish between desertic and arid or semiarid re- 
gions. In Chile, the limit is located near the 
parallel 29 S.L., where the annual rainfall in- 
creases from less than 100 mm to more than 200 
mm but not more than 270 mm. This is an arbi- 
trary boundary which attempts to separate de- 
sertic from srid soils on the basis of the intensity 
and the nature of the pedogenic processes that 
have affected both. 

The soils of these regions are not very well 
known, because of the difficulties of making soil 
surveys and the minimal interest official and 
private agencies have shown, since the regions 
exhibits very limited farming potential. The 
exception is for the soils of the highly productive 
valleys, transverse the Intermedial Depression 
and approach the Pacific Ocean. Several de- 
tailed soil surveys have been made on those val- 
leys whose principal soils are of aluvial origin; 
however, their significance in the total surface 
of arid lands is little. 



Walter Luzio-Leighton, Facultad de Ciencias Agrarias 
Y Forestales, Universidad de Chile, Casilla 1004, San- 
tiago, Chile. 



This paper provides a general outline of the 
soils from the desertic and arid regions in Chile, 
and in so doing reviews the main researches al- 
ready made, and this author's unedited data. 
The data have been interpreted according to 
author's view of the distribution pattern of the 
soils through the most important physiographic 
units described in the north of Chile, from 
18S.L. to 32S.L. (Figure 1). 

Results and Discussion 

Climatic Factors that Conform the Aridity 
in the North of Chile 

The anticiclone from the south west Pacific 
that originates high pressure systems is located 
between 25 and 30S.L. and 90 West, that is, 
in front of the dryest part of Chile which is nor- 
mally considered from 18 to 32S.L. 

The air masses descending from the tropo- 
sphere are gradually heated, even though near 
the ocean they are cooled, producing a thermic 
inversion below 1,000 m altitude. The thermic 
inversion restricts the vertical movements of the 
air masses, causing an arid climate with limited 
precipitation. 



LUZIO-LEIGHTON: THE SOILS OP THE DESERTIC AND ARID REGIONS OP CHILE 



105 




Figure 1. Arid and desertic regions of 
Chile and distribution area of 
Aridisols. 



1 170 



130 



iOO 



9O 




1900 10 20 30 40 50 iso 70 ~~ 

mobile ovcroge of 30 years 

Figure 2. Rainfall tendency in La Serena (Chile) from 1900 to 1970, us- 
ing the mobile average of 30 years. (Taken from Santibanez, 1985). 



Considering these genetic factors of 
the climate, it is possible to subdivide 
the north region of Chile into four ar- 
eas from west to east: the literal 
(coastal line), the pampa (Intermedial 
Depression), the precordillera (Andes 
piedmont), and the Altiplano (Andes 
plateau). 



Literal Area 

The literal is a narrow strip not more than 15 km wide 
where the thermic conditions are alleviated by the influ- 
ence of the ocean. There is no frost and the temperature 
varies from 7 to 28 C, considering the minimum tempera- 
ture of the coldest months and the maximum temperature 
of the warmest months. The abundant cloudiness makes 
diminish solar radiation and also the thermic fluctuations. 
The average rainfall ranges from less 10 mm per year to 
270 mm in the southern part of the litoral area. The an- 
nual moisture deficit ranges from near 2,000 mm in the 
northern part of the area to 1,400 mm in the southern 
part. 

Pampa 

In the second area, the pampa, the mean annual tem- 
peratures do not vary significantly from those of the litoral 
area, but the daily fluctuations are much greater, from 
33 C during the day to lower than 0C in the coldest 
nights. These conditions are dominant in the so-called 
"extreme desert" that is more or less up to paralel 27S.L. 
Going to the south, the continental features of the climate 
are still present, though without desertic characteristics 
and approaching more mediterranean conditions. Simi- 
larly, precipitation is less than 5 mm in the desert to 250 
mm in the southern part, concentrated only during winter 
months. 

Precordillera 

The third area, the precordillera extends parallel to the 
Andes mountains (north - south direction) as a piedmont 
formation at an altitude more than 2,000 but less than 
2,500 m.o.s.l. The maximum temperatures are around 
25 C and the minimum around 5C. The rainfall fluctu- 
ates from 50 to 100 mm per year. 



106 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Altiplano 

The fourth area, the Altiplano, is a plateau- 
like formation that extends over 4,000 m.o.s.L 
and only up to 23S.L. The mean temperatures 
are always below 6C and frost is frequent all 
the year round. The climate of the Altiplano is 
characterized by a summer rainfall (about 200 
mm) regime that produces a subhumid period 
lasting 1 to 3 months. 

Rainfall regime tendencies 

The cyclic variations of the climate at plane- 
tary scale are very well known, as is their influ- 
ence on the amount and distribution of rainfalls. 
The desertic and arid zones of Chile are sub- 
jected to a process of climatic desertization, ex- 
pressed as a clearly decreasing tendency in the 
amount of annual precipitation. 

Figure 2 shows the rainfall tendency from 
1900 to 1970, using the mobile average of 30 
years technique. 

Soil moisture and soil temperature 
regimes 

Aridic regime 

According to Van Wambeke and Luzio (1982), 
from the border with Peru up to parallel 32S.L. 
there is only an aridic soil moisture regime, 
which goes from the sea coast to the high Andes 
mountains and to the Altiplano (Figure 3). Near 
the coast are many meteorological stations and 
enough data is available to apply the Newhall 
model of calculation, while in the Andes pied- 
mont and the Altiplano the data are scarce and 
isolated. The author therefore had to assume 
that the aridic soil moisture regime extends to 
the Andes, based on data about water deficit 
obtained from the difference between annual 
precipitation and evaporation. 

Aquic Regime 

One of the most interesting features that 
characterize the Altiplano is the presence of 
close basins with restricted drainage where high 
amounts of organic matter have been accumu- 
lated, giving rise to organic soils. Similar situ- 
ations are found in some depressed areas close 
to small streams where the water flows slowly, 
overflowing the surrounding terrains. There 
have been doubts about the use of the concept of 
aquic moisture regime for those cases because 
the water moves permanently with some dis- 
solved oxygen, though some reduction phenom- 
ena are expressed as mottles of high chroma and 
gleyed colors (Luzio, 1985). 




Figure 3. Soil moisture regimes in the desertic and 
arid regions of Chile. A: Aridic; AQ: Aquic; X: 
Xeric. 



Iso Regimes 

With regard to soil temperature regimes the 
iso regimes are associated with the area near 
the ocean. Some scarce data allow us to assume 
the marine influence reaches the inner area of 
pampas. The isothermic regime becomes the 
most important one from the extreme north of 
the country up to parallel 27S.L. From this 
latitude to the south, the isothermic regime is 
restricted to the shore line in a strip no more 
than 15 km wide (Luzio, 1985). 

Crvic Regime 

The cryic temperature regime has been as- 
signed to the Andes mountains and Altiplano 
although, as it was already pointed out, the lim- 
ited or non existent data do not permit us to 
verify this assignment (Figure 4). 



LUZIO-LEIGHTON: THE SOILS OF THE DESERTIC AND ARID REGIONS OP CHILE 



107 



to 




[ 1 isotherm ic 
isomesic 

IIIIIIIIIJII thermic 



Figure 4. Soil temperature regimes in the desertic 
and arid regions of Chile, 



The Soils 

In the desert, the weathering processes have 
been limited to the phyiscal breakdown of the 
parent rocks, producing crumbled materials 
subjected to mass movements such as landslides 
or mudilows. The restricted available water 
produces a very weak organic regime and scarce 
vegetation cover and also does not allow any 
kind of eluviation-illuviation processes (Luzio, 
1985). The conditions prevalent in the arid and 
semiarid regions make possible a more active 
organic regime, a more dense shrub cover, and 
the development of illuviation features enough 
for an argillic horizon (Luzio, 1985). 



The Soils in the Desertic Region 

Group 1 

The first group of soils corresponds to those 
with a minimum profile development, such as 
the shallow soils over rock, stratified soils, and 
skeletal soils of coluvial origin. None of these 
soils are associated with specific landscape or 
area; they are found either in the coastal area, 
in the pampa, or in the high Andes regions and 
the Altiplano (Luzio and Vera, 1982). They are 
coarse textured soils with less than 0.2% organic 
carbon in the surface horizon. Those soils found 
in the Andes and Altiplano regions qualify as 
Cryorthents, and those in the pampa and 
coastal areas qualify as Torriorthents. 

The stratified soils whose origins have to be 
found in aluvial transportation and sedimenta- 
tion in the Andes and Altiplano show an irregu- 
lar distribution of organic carbon with depth, so 
that we have called those soils Cryofluvents 
even though they are never or almost never 
flooded in present conditions. 

Finally, in this first group, skeletal soils must 
be included. Normally they ahve more than 
40% gravel throughout the profile, mostly 
sandy-skeletal or loamy-skeletal, and are associ- 
ated with metastable slopes. 

Group 2 

A second group of soils, also found in all areas 
in the desert region, show some profile develop- 
ment in the form of a structure and/or color B 
horizon. The soils are formed on finer sediments 
than those of the first group and they have been 
considered as Camborthid, Lithic, and Fluventic 
subgroups (Alcayaga and Luzio, 1985). 

Profile I in the Appendix is an example of this 
kind of soil. It is a rather homogeneous soil, 
without stratification and with a cambic horizon 
determined by the presence of soil structure and 
clay increase. It seems the soil is the result of its 
environment, where pedogenic agents have 
been acting according to that medium. The low 
soil temperatures, the limited water availability 
coming as infrequent rains and the almost inex- 
istent vegetation suggest that these weakly ex- 
pressed pedogenic agents have been active for a 
considerable period of time in order to produce a 
slightly to moderately developed soil. This ob- 
servation agrees with Nettleton's and Peterson's 
ideas (1983), in the sense that the diagnostic 
horizons in Aridisols require considerable time 
to form and, consequently, Comborthids have to 
develop on stable surfaces of late Pleistocene or 
greater age. 



108 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Group 3 

A third group corresponds to the soils derived 
from volcanic materials whose distribution is 
restricted only to the Altiplano area. Our expe- 
rience with volcanic soils in arid and desertic 
regions indicates that the pedogenic develop- 
ment is controled completely by the arid envi- 
ronment and that the volcanic materials do not 
play a definitive role in the direction of pe- 
dogenic processes. 

Some data shows that no difference can be 
detected in desertic soils developed from other 
parent materials. The organic regime is very 
restricted and sometimes has an irregular dis- 
tribution with depth. The pH is slightly acid to 
neutral. A dominance of coarse sand and very 
coarse sand exists throughout the profile, and 
the majority of roots are concentrated in the 
second horizon. The percentage of glasses meas- 
ured in all the sand fractions is between 40 to 
60%, the P retention is lower than 15 percent 
and Al + l/2Fe is generally lower than 0,15 per- 
cent and only in some subhorizons is slightly 
higher than this value. These figures indicate 
that the desertic conditions are responsible for 
the very low degree of weathering of the volcanic 
glasses. Depending on the presence or absence 
of a structure B horizon, they have been consid- 
ered as Camborthids or Cryorthents. No one of 
the few profiles analized up to now meet the 
requirements of the andic properties according 
to Circular Letter N 10 (INCOMAND, 1988). 

Moreover, considering our present knowledge 
about these kind of soils, we do not think a Vitric 
subgroup in Orthids would be useful, because 
the behavior of the soils will not differ signifi- 
cantly from the Typic ones (Profile 2 is an ex- 
ample of this kind of soil). 

Group 4 

The fourth group are the soils with high or- 
ganic matter content developed only in closed 
basins and depressed areas of small streams in 
the Altiplano. It seems the moisture regime is a 
reducing one, at least in some periods of the 
year, even though the water table is normally 
moving water. There is no frost and the mean 
annual soil temperature is higher than 0C. 

These areas, called "Bofedales," are not great 
in extent but they are very important as grass- 
land resource that maintains the livestock of the 
region. The classification of these soils has been 
subjected to some discussion because of the 
moving water table. Some of them qualify as 
organic soils (Fibrist and Hemist) and others, 
with smaller organic matter content, as 



Cryaquents and Cryaquepts, based on the pres- 
ence of gley horizons and oxidation-reduction 
mottles throughout the profiles. 

The pedon from Turi (Profile 3) corresponds to 
a stratified soil with high calcium carbonate 
content and alkaline reaction throughout the 
profile. The soil has been formed from medium 
particle size materials in a somewhat poorly 
drained closed basin. During rainy periods the 
ground water table may reach the soil surface 
and the rest of the year is not deeper than 80 
cm. 

Group 5 

The last group of soils in the desert region 
corresponds to those highly saline soils found 
only in the Intermedial Depresion, at altitudes 
near 1,000 m.o.s.l. 

The most striking feature of these soils is the 
presence of a very hard surface crust of salts 
about 25 cm thick. Few analyses from the crust 
and from the different strata have been made, 
and they show the dominant cation is Na+ and 
the anion is C1-; Ca++ and SO 4 are of secon- 
dary importance. The surface crust is composed 
also by 50 to 65 percent clay (Luzio and Vera, 
1982). Our knowledge is very limited about this 
kind of soil, only a small number of profile de- 
scriptions and analyses have been made, indi- 
cating the topsoil is an indurated saline horizon 
composed mainly of NaCl. 

These kinds of horizons have not been de- 
scribed yet in Soil Taxonomy. More studies are 
needed, so that at present we only can ask if we 
should have to consider a "petrosalic" horizon. 

The Soils in the Arid Regions 

The arid regions extend from 27S.L. approxi- 
mately to 32S.L. The soils show a higher de- 
gree of development in comparison to the desert 
soils. The presence of cambic B horizon is com- 
mon, and argillic horizons are not rare on the 
marine terraces. 

Soils from the Coastal Range 

The region is characterized by the presence of 
numerous marine terraces of different levels. In 
those places where there are no marine terraces, 
the Coastal Batholith reaches the shore line and 
the soils are formed "in situ" from the weather- 
ing of the plutonic rocks (Luzio et.a., 1978). 

Significant areas of the terraces are formed 
by stabilized sand dunes covered with xero- 
phytic shrubs and scarces Acacia trees. The lack 
of development of these soils is attributed to the 
very coarse and recent sediments, unstable sur- 



LUZIO-LEIGHTON: THE SOILS OP THE DESERTIC AND ARID REGIONS OF CHILE 



109 



faces due to mass movements or erosion, and the 
lack of enough water to promote the soil proc- 
esses. They have been considered mostly as 
Torripsamments (Alcayaga and Luzio, 1985). 

In higher level terraces from the Pliocene 
(Munoz, 1950; Paskoff, 1970), soils with argillic 
horizons are normally developed. The presence 
of illuviated horizons has been interpreted as 
formed on stable surfaces in a more humid cli- 
mate than present conditions with pronounced 
rainy and dry seasons. Paleargid is the most 
frequent great group. 

The existence of wetter climates during the 
Pliocene and the Quaternary has been empha- 
sized by different authors who give diverse 
kinds of arguments, such as the presence of 
large aluvial terraces and aluvial cones, the 
presence of relic forest, and so on. 

In some medium terraces, argillic horizons 
are replaced by natric horizons whose presence 
is attributed to the numerous marine transgres- 
sions due to epeirogenic movements. The accu- 
mulation of Na+ has taken place along with 
Mg++ so that some of the sampled natric hori- 
zons show 32 percent Na saturation and nearly 
60 percent Mg saturation. They are mostly 
Xerollic Natragids (Alcayaga and Luzio, 1985). 

Finally, in the marine terraces, soils over cal- 
careous sediments are found. The more impor- 
tant are those with a petrocalcic horizon formed 
from a high amount of marine bivalved mollus- 
cans shells clearly recognizable in the lower part 
of the profiles. 

Innerland soils 

The knowledge of the soils of this area is very 
preliminary because no systematic pedogenic 
studies or soil inventories on any scale exist. 

Limestone is one of the significant parent 
rocks found not only in the flat relief of the inter- 
montane valleys but also in the hilly topogra- 
phy. At present there is no a clear idea about 
the origin of the lime in the hilly topography, 
because most of the hills are built up by plutonic 
rocks. Argillic horizons are developed over a 
calcic horizon and no petrocalcic horizons have 
been found yet. 

Other important parent rock is granite. The 
soils developed from this rock are supposed to be 
Camborthids. 

Conclusions 

1. The soils formed from volcanic materials 
are found clearly in the Andes, under a 



cryic temperature regime, and from 18 to 
27 SL, approximately. They are 
Cryorthents. It is thought that the major- 
ity of the other soils have some kind of con- 
tamination with volcanic materials, though 
this is very difficult to evaluate. 

2. In the desertic region there is a clear domi- 
nance of young soils (Torriorthents) with- 
out significant pedogenic processes or diag- 
nostic horizons other than an ochric epipe- 
don. 

3. In the desertic region the most intense soil 
development is a cambic horizon. The soils 
in the Intermedial Depression are Typic 
and Lithic Camborthids and those in the 
Andes, with cryic temperature regime, are 
also Camborthids. No subgroups are now 
recognized. 

4. Paleorthids are found both in the desertic 
and the arid region. 

5. The soils with high organic matter content 
(Histosols or not) are found only in the high 
mountains in close basins with impeded 
drainage. The most feasible explanation 
for their formation is associated with the 
postglacial pond cycle. 

6. Argillic horizons are found only in the arid 
regions and associated with the soils devel- 
oped on marine terraces. They are Palear- 
gids. 

7. Natric horizons are only found in restricted 
areas in the arid region on marine terraces. 
The high exchageable Na+ and the high 
Mg++ have been interpreted as a decisive 
influence of pleistocene marine transgres- 
sions. 

8. The valleys are dominated by non-devel- 
oped soils from aluvial origin: Torri- 
orthents, Torrifluvents, and Torripsam- 
ments. Some valleys have a high salinity 
status. 

Literature Cited 

Alcayaga, S., andW. Luzio. 1985. Clasificaci on taxonomica 
de los suelos de regiones des erticas y aridas del norte 
de Chile. Bolet in N 5 (2): 141 - 144. Sociedad Chilena 
de la Ciencia del Suelo, Santiago, Chile. 

INCOMAND, 1988. Circular Letter N 10. 29 February, 
1988. 80 p. 

Luzio, W. 1985. Genesis y clasificaci on de los suelos de 
regiones kridas y deserticas de Chile. Boletin N 5(1): 
107 - 140. Sociedad Chilena de la Ciencia del Suelo, 
Santiago, Chile. 

Luzio, W., I. Badilla, and W. Vera. 1978. Zonificacion del 
sistema iitoed aiico en el secano costero de la IV Region. 
II Simposio Nacional de la Ciencia del Suelo. Santiago, 
Facultad de Agronom ia, Universidad de Chile: 476- 
495. 



110 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Luzio, W., and W. Vera. 1982. Caracterizacion de suelos en 
ecosistemas de la I y la II Region. SACOR - CODECIA- 
GRO. Informe para la CORFO. 126 p. 

Munzo, C. Jorge. Geolog ia. En: Geografia Economica de 
Chile. Tomo LCorporacion de Fomento de la Produc- 
tion. 55 - 87. 

Paskoff, 1970. Le Chili semi-aride. Recherches geomor- 
phologiques. Viscaye Freres, Paris. 420 p.. 

Santibanez, F. 1985. Rasgos agroclimaticos de la zona 
arida chilena. Boletin de la Sociedad Chilena de la 
Ciencia del Suelo N 5: 1 - 28. Santiago, Chile. 

Van Wambeke, A., and W. Luzio. 1982. Determinacbn de 
los reg imenes de humedad y temperatura para los sue- 
los de Chile. Agricultura T ecnica (Chile) 42(2): 149 - 
159. 

Appendix 

Profile!. 

Location: 30.8 km from San Pedro de Ata- 
cama, in the road from San Pedro de Atacama to 
Tatio. 

Altitude: 3.750 m.o.s.1. 

Geomorphic position: footslope 

Slope: 5 - 10% 

Stoniness: 60% angular gravel, 5% boulders 

Classification: Typic Camborthid 

A - 11 cm. Dark reddish brown (5 YR 3/3) 
and brown (7.5 YR 5/4, dry); loamy sand; single 
grain; loose; non plastic; non sticky; few fine 
roots; 10% gravel; clear smooth boundary. 

A2 11 - 23 cm. Dark reddish brown (5 YR 3/3) 
and brown 7.5 YR 5/4, dry); sandy loam; weak 
fine subangular blocky structure; slightly hard; 
friable; slightly plastic, non sticky; common fine 
roots; few fine pores; 5% gravel; clear smooth 
boundary. 

Bl 23 - 45 cm. Dark reddish brown (5 YR 3/3) 
and brown (7.5 YR 5/4, dry); sandy clay loam; 
weak fine and medium subangular blocky struc- 
ture; slightly hard, friable; plastic, slightly 
sticky; common fine and medium roots, 3% 
gravel; clear smooth boundary. 

B2 45 - 55 cm. Dark reddish brown (5 YR 3/3) 
and brown (7.5 YR 5/4, dry); sandy clay loam; 
weak fine subangular blocky structure; hard; 
friable; plastic, slightly sticky; common fine 
roots; few fine pores; 30% gravel; abrupt wavy 
boundary. 

R 55 - 75 cm Tuff. 



Profile li Typic Camborthid (Tolar S.P 
pedon). 


. Atacama - Tatio 


Depth, 
(cm) 


Hor. 


pH 


O.C. 

% 


Particle size distribution (%) 
sand silt clay 


0-11 
11-23 
23-45 
45-65 


Al 
A2 
Bl 
B2 


5.0 
7.0 
7.0 
6.9 


0.34 
0.02 
0.33 
0.27 


72.0 
66.0 
56.0 
60.0 


16.0 
18.0 
18.0 
20.0 


12.0 
16,0 
26.0 
20.0 


Depth, 
(cm) 


Hor. 


Extractable cations 
Ca Mg Na 

M.A^/ir\T\ rr 


CEC 
K 


B.S. 










0-11 
11-23 
23-45 
45-65 


Al 
A2 
Bl 
B2 


2.4 

4.8 
7.2 
7.2 


0.6 
1.9 
3.0 
3.2 


0.3 
0.4 
0.6 
0.4 


0.7 
0.9 
1.6 
1.7 


_6.3 63 
_8.9 89 
12.599 
13.096 


Depth, 
(cm) 


Hor. 


Water content (%) 
1/3 BAK 15 mmho/cm 


E.G. 




0-11 
11-23 
23-45 
45-65 


Al 
A2 
Bl 
B2 


12.7 
13.6 
17.4 
18.0 


6.8 
7.9 
11.0 
11.5 


0.6 
0.2 
0.5 
0.7 







Profile 2. 

Location: 15 km from Putre town, in the road 
from Putre to Chungara. 
Altitude: 4,300 m.o.s.l. 
Geomorphic position: Altiplanic plateau 
Slope: 5 - 7% 
Coarse fragments: 5% 
Erosion: Probably hydric and eolic 
Classification: Typic Cryorthent. 

Al - 8 cm. Very dark grayish brown (10 YR 
3/2) and dark reddish brown (5 YR 3/2.5, dry); 
coarse sandy loam; weak fine and medium 
subangular blocky structure; friable; plastic, 
non sticky; few fine roots; common fine pores; 
30% gravel; clear smooth boundary. 

A2 8 - 15 cm. Variegated, reddish (5 YR 4/4) 
as dominant; silty loam; with gravels; very fine 
granular structure; friable; slightly plastic, 
sticky; many fine roots; many fine pores; 45% 
gravel (1 cm); clear smooth boundary. 

A/C 15 - 21 cm. Variegated, yellowish red (5 
YR 4/6) as dominant; silty clay loam with grav- 
els; weak fine granular structure; slightly plas- 
tic, slightly sticky; few fine roots; 60% gravel (up 
to 4 cm); clear smooth boundary. 

C 21 - 115 cm. Variegated, light brown and 
brown (10 YR 7/3 and 10 YR 5/3) as dominant; 
gravelly sandy loam; massive; very hard, indu- 
rated, no roots, no pores; 70% pumice gravel (2 - 
4 cm). 

Observations: rhyolitic pumice fragments are 
abundant throughout this profile. They corre- 
spond partially to the gravel described. 



LUZIO-LEIGHTON: THE SOILS OP THE DESERTIC AND AKID REGIONS OP CHILE 



111 



Profile 2: Typic Cryorthent (Putre pedon). 


Depth, 
(cm) 


Hor. 


pH O.C. 

% 


Particle size distribution (%) 
sand silt clay 


0- 8 
8-15 
15-21 
21-115 


All 
A2 
A/C 
C 


7.0 0.29 
7.2 0.63 
7.3 0.46 
7.1 0.1 


56.0 
52.0 
47.2 
60.0 


24.0 
20.0 
24.8 
22.0 


20.0 
28.0 
28.0 
18.0 




Depth. 

(cm) 


Hor. 


Extractable cations 
Ca Mg Na 

mQm/1 r\f\rr 


K 


CEC 


B.S. 

% 










0- 8 
8-15 
15-21 
21-115 


All 
A2 
A/C 
C 


3.6 1.9 
7.5 3.3 
10.4 4.4 
8.6 3.6 


0.3 
0.4 
0.5 
0.6 


0.5 
1.0 
1.2 
1.0 


8.8 
15.6 
20.0 
13.7 


72 
78 
83 
100 


Depth, 
(cm) 


Hor. 


Water content (%) B.C. 
1/3 BAR 15 mmho/cm 


0- 8 
8-15 
15-21 
21-115 


All 
A2 
A/C 
C 


9.5 5.7 
16.5 10.8 
21.6 15.3 
20.5 12.6 


0.3 
0.2 
0.3 
0.3 









2C2 22 - 29 cm. White (10 YR 8/2) with char- 
coal layers, 5 mm thick, of black color (2.5 Y 2/0); 
silty clay loam; massive; firm; plastic and sticky; 
few fine and medium roots; common fine pores; 
violent effervescent (HCl); abrupt smooth 
boundary. 

3Cg 29 - 66 cm. Gray (5 Y 5/1) intermixed 
with black (2.5 Y 2/0) layers of charcoal; gravelly 
clayey loam; massive; plastic and sticky; few 
fine and medium roots; many fine pores; violent 
effervescent (HCl). 

Observations: ground water level: 66 cm 



Profile 3. Typic Cryaquent, calcareous (Turi pedon). 


Depth, 
(cm) 


Hor. 


pH 


O.C. 

% 


CaCO 3 
% 


Particle size distribution (%) 
sand silt clay 


0-15 
15-22 
22-29 
29-66 


A 

Cl 
2C2 
3Cg 


9.0 
9.2 
9.1 
9.1 


6.0 
8.1 
4.1 
6.5 


21.8 
55.6 
70.4 


31.4 
32.6 
67.4 
43.4 


48.0 20.6 
48.8 18.6 
22.0 10.6 
34.0 22.6 


Depth, 
(cm) 


Hor. 


Extractable 
Ca Mg 


cationsCEC 

Na 
meq/100 g 


^ 

"K 


_ 










0-15 
15-22 
22-29 
29-66 


A 
Cl 
2C2 
3Cg 


79.2 
33.1 
28.3 
27.1 


14.2 
13.0 
8.2 
4.0 


121.8 
3.79 
9.0 
6.5 


3.89 
2.81 
1.5 
1.5 


30.0 
32.2 
22.8 
14.1 


Depth, 
(cm) 


Hor. 


Water content (%) 
1/3 BAR 15 


ESP 

_% 


E.G. Soluble cations 
mmho/cm Ca Mg Na K 


0-15 
15-22 
22-29 
29-66 


A 
C16 
2C2 
3Cg 


56.5 
78.5 
57.9 
52.8 




25.2 
36.1 
25.9 
28.7 


6.6 

10.1 


150 6.6 15.7 22.2 7.6 
14 1.4 1.4 12.0 1.8 
6 . _. ._ _. 

4 . . _. 



Profile 3. 

Location: "Vega de Turi", between Toconce 
and Aiquina villages 
Altitude: 2.900 m.o.s.l. 
Geomorphic position: extended closed basin 
Slope: 2% 

Water table: 66 cm depth 
Calcareous surface crust 
Classification: Typic Cryaquent, calcareous 

A - 15 cm. Brown to dark brown (7.5 YR 4/ 
4); silty clay loam; weak medium subangular 
blocky structure; friable; plastic, slightly sticky; 
many fine and medium roots; many fine pores; 
oxidations about 30% of the horizon; violent ef- 
fervescent (HCl); clear smooth boundary. 

Cl 15 - 22 cm. Reddish yellow (7.5 YR 7/8); 
silty loam; massive; friable; plastic and sticky; 
common fine and medium roots; many fine 
pores; violent effervescent (HCl); abrupt smooth 
boundary. 



112 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



A Review of Recent Research on Swelling Clay Soils in Canada 
A.R. Mermut, 1 D.F. Acton, 1 and C. Tarnocai 2 

Abstract 

This paper reviews recent studies of the characteristics, behavior, gene- 
sis, and classification of swell-shrink soils in Saskatchewan. All the soils 
studied crack to varying degrees and have distinct slickensides in the sub- 
soil. The solum of the soils is much deeper than originally thought. Horizon 
designation is a problem in these soils. A case for B horizon can be argued, 
but this B horizon should be recognized as Bw. The identification of C hori- 
zons may be based on structure and the depth of gypsum and salts. 

Physical, chemical, and mineralogical properties of the soils vary along 
an environmental gradient from south to north. Micromorphological fea- 
tures reflect the magnitude of the stress and soil displacement. Stress fea- 
tures were minimal or nonexistent in the C horizons. Genesis of the soils is 
governed by the smectite content (extent of swelling), climate, and vegeta- 
tion. While the majority of swell-shrink soils may be considered as Verti- 
sols, these soils would not fit into any of the Vertisol suborders of Soil Tax- 
onomy. Therefore, a new suborder, Borerts, is appropriate for the Vertisols 
occurring in frigid and cryic soil temperature regimes. Currently, swell- 
shrink soils are classified at the family level in the Canadian system. It 
seems inconsistent with the differentiating criteria used at higher catego- 
ries in the Canadian system to relegate these criteria to a lower categorical 
level. The Canadian system could be improved by developing a separate 
order for swelling soils. 



Introduction 

Heavy clay soils with swell-shrink properties 
occur in glacial lake sediments that extend from 
sub-arid grassland to subhumid grassland-for- 
est transitional zone in the Canadian Prairies. 
Early soil survey work in Saskatchewan (Mitch- 
ell et aL, 1944) considered clay soils as a sepa- 
rate type from other regional soils because of 
their high fertility status and their resistance to 
drought. At the national level, these soils were 
variously referred to as Argillaceous Regosols or 
Gruraic soils, and they have been classified at 
various categorical levels. Clay soils with 
marked swell-shrink potential were not sepa- 
rated from others at the order level in the Cana- 
dian System of Soil Classification (CSSC). 
However, grumic subgroups were created 
within the Chernozemic order (Clayton, 1963) to 
accommodate these soils. These subgroups re- 
mained in the system until 1974 (Canada De- 
partment of Agriculture, 1974). However, in the 
final version of the CSSC (Canada Soil Survey 
Committee, 1978), "grumic" property has been 
used to separate the soils at the family level. In 
the CSSC the term "grumic" is intended to indi- 
cate those soils that have fine texture, smectitic 
mineralogy, and self-mulching properties. It 

Saskatchewan Institute of Pedology, University of Sas- 
katchewan, Saskatoon, SK S7N OWO Canada. Contribution 
No. R627. 

2 Agriculture Canada Land Resource Research Centre, Ot- 
tawa, Ontario K1AOC6, Canada. 



can be argued that self-mulching properties are 
related to soil genesis; therefore, these soils 
should be considered above the family level. 
The lack of data on swell-shrink properties of 
clay soils likely hindered the use of these criteria 
in soil classification in Canada. 

While it has been long recognized that clay 
soils differ from other Chernozemic soils in the 
region, no attempt was made to study them in 
sufficient detail to understand their genesis, 
properties, and relations to Vertisols described 
in Soil Taxonomy (Soil Survey Staff, 1975). 
Soils that have morphological characteristics 
similar to Vertisols but occur in colder regions 
were excluded from this order. Thus, until very 
recently, the soils in the north central U.S.A. 
and the Canadian Prairies, with frigid and cryic 
soil temperature regimes, were not considered 
within the Vertisol order. In September 1980 
approval was given by USDA Soil Conservation 
Service to drop the temperature requirement for 
this soil order (Soil Survey Staff, 1982). 

Several research studies were initiated to 
examine the characteristics of these soils in Sas- 
katchewan. The objective of the present paper 
is to review the recent information regarding 
the morphology, mineralogy, micromorphology, 
and soil forming processes that are operative in 
swell-shrink soils in Saskatchewan and to 
evaluate these in terms of their genesis and 
classification at the international levels. 



MERMUT, ACTON, AND TARNOCAI: A REVIEW OF RECENT RESEARCH ON SWELLING CLAY SOILS IN CANADA 



113 



Morphology and Micromorphology 

Horizon Designation 

The lack of color contrast, common occurrence 
of carbonates throughout the profile, and poor 
expression of structure when moist, make hori- 
zon designation difficult in the heavy clay soils 
in Saskatchewan (Mermut and St. Arnaud, 
1983; Dasog et al. 1987). However, coarse pris- 
matic, breaking to angular blocky structure, 
may be observed when the soils are dry. In 
Canada, if the horizon below the Ap is calcare- 
ous or massive, such a horizon is designated as a 
C horizon (Ayres et al., 1985). Therefore, desig- 
nating the solum and differentiating this from 
the C horizons is often difficult (Mermut and St. 
Arnaud, 1983). The boundary between the so- 
lum, which may be more than one meter thick, 
and the C horizon is often sharp, although it 
may be diffuse and a transitional horizon may 
be recognized. The upper part of the C horizon 
is physically disrupted and, at depth, prominent 
varying with partial disruption may be ob- 
served. However, at the contact zone between 
the solum and C horizon, more soluble salts and 
associated gypsum often are concentrated, 
which may be helpful in determining the depth 
of solum. As will be discussed later, micromor- 
phological studies are helpful in determining the 
horizon designation as well as in understanding 
the genesis of these soils. 

Another question is how to designate the hori- 
zon within the solum. Subsoil horizons that 
show compound prismatic-blocky structure 
when dry are considered to be a B horizon (Can- 
ada Soil Survey Committee, 1978; Soil Manage- 
ment Support Services, 1982). Dasog et al. 
(1987) suggested that a case for B horizon can be 
argued on structural consideration alone and 
that these B horizons should be recognized as 
Bw (Bmk in Canadian System). Wilding and 
Tessier (1988) indicated that, based on morpho- 
logical and other characteristics, it is fully ap- 
propriate to recognize the presence of a "cambic 
horizon" in Vertisols. 

Cracking and Slickensid.es 

All heavy clay soils in Saskatchewan show 
cracking. Despite its importance for classifying 
Vertisols, information on direct measurement of 
crack parameters is rare in the literature. Lim- 
ited studies by Dasog (1986) showed that crack- 
ing in the subarid regions of Saskatchewan is 
less than one half as intense as in Vertisols in 
the xeric moisture regime of Israel (Yaalon and 



Kalmar, 1978) and even lower than in the Verti- 
sols of Sudan (El Abedine and Robinson, 1971). 
In both Israel and Sudan, summers are much 
drier than in Saskatchewan and precipitation 
during the summers in Saskatchewan effec- 
tively decreases the intensity of cracks. 

Under native grassland, cracks are shallower 
and narrower. The moisture regime of Sas- 
katchewan soils influences the duration and in- 
tensity of cracking and limited information sug- 
gests that cracks may remain open for more 
than 90 cumulative days. Considering the mois- 
ture distribution gradient, one would expect a 
decreasing intensity from south-west to north- 
east. Furthermore, days are much longer dur- 
ing the summer and, therefore, more effective in 
crack development. This fact needs to be consid- 
ered while comparing these soils with other 
swell-shrink soils that occur in southern lati- 
tudes. The degree and frequency of changes in 
moisture content of the soil are perhaps the 
most important parameter that controls the in- 
tensity of cracking and movement within the 
soil, provided that soils have sufficient COLE 
(enough swelling clays) to produce movement. 
The degree of swelling and shrinking is de- 
creased with organic matter, carbonates, gyp- 
sum, high electrolyte concentration, sesquiox- 
ides, and low activity clays which bind and ce- 
ment the soil fabric.Dasog et al. (1988) found 
the following relationship between COLE and 
the content of expandable clay in Saskatchewan 
soils (1): 

COLE = -0.0026 + 0.0033 x % expandable 
clays (1). 

This would mean that, to produce COLE 
value of 0.1 > 30%, swelling clay is needed. This 
equation agrees well with that reported by 
Schafer and Singer (1976) for some swell-shrink 
soils in California (2): 

COLE = -0.00123 + 0.00336 x % expandable 
clays (2). 

Another equation suggested by Dasog (1986) 
indicates the relationships between fine clay 
and COLE (3): 

COLE = 0.042 + 0.0037 x % fine clay (3). 

According to Grossman et al. (1985), a soil 
with an intermediate COLE value of 0.05, at 
1,500 kPa moisture content, would produce 
cracks of about 1 cm wide and repeat distance 
between the cracks of about 20 cm. If we apply 
an intermediate COLE value of 0.05 to the 
above equations, a minimum of 15% swelling 
clay or 17% fine clay would be required. These 
will approximately equal a minimum of 30% 



114 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 




Fig. 1 Photomicrographs from the Melfort soil. A. mullgranoidic (Alpl horizon 2-10 cm). B. ferriargillans (Bnt 
horizon. 23-31 cm). C. parallel striated b-fabric in very dense mass (Bwl horizon, 42-50 cm). D broken peds with 
striated fabric and planar voids. 



total clay, which is also equal to the clay content 
criterion used by Soil Survey Staff (1975) to dif- 
ferentiate Vertisols. We therefore suggest that 
there is no need to mention clay content in the 
definition, but a definition of diagnostic vertic 
features is now desirable. 

Slickensides occur in the subsoil of all swell- 
shrink soils studied in the four major soil zones 
of Saskatchewan, where the difference between 
horizontal and vertical stresses is large. It ap- 
pears that, because of the lower overburden 
pressure, the upper part of the solum remains 
relatively stable. The intact columnar and pris- 
matic structures of the shallow Solonetzic soil 
formed on the heavy Melfort and Tisdale soils 
and the presence of slickensides in the heavy 
subsoil below the Solonetzic B horizon testify to 
this view. 



Micromorphology 

Swelling heavy clay soils in Saskatchewan 
have distinct micromorphological features (Mer- 
mut and St. Arnaud, 1983; Dasog et al. 1987). 
Because of cracking and the formation of slick- 
ensides, these soils have produced entirely dif- 
ferent void patterns (Figs. 1, 2). Subparallel 
joint planes, slickensides, meta-skew planes, 
and craze planes with smoothed surface confor- 
mation are the types encountered (Mermut and 
St. Arnaud, 1983). Both cracking and soil dis- 
placement produces planar voids. Micromor- 
phology provides an excellent opportunity to dif- 
ferentiate the voids that are produced by stress 
from those formed by simple desiccation. At 
depth, surfaces of planes are generally slicken- 
sides (Fig. 2C), indicating the high stress and 
displacement resulting from the swelling clays. 



MERMUT, ACTON, AND TAKNOCAI: A REVIEW OP RECENT RESEARCH ON SWELLING CLAY SOILS IN CANADA 



115 




Fig. 2 Photomicrographs from the Kelvington soil. A. surface granular structure (Ap horizon, 3-13 cm). B reticulate 
striated b- fabric (AB horizon 18-28 cm) and planar voids. C. porostriated b-fabric (Bw horizon, 38-48 cm). D. physi- 
cally broken parent material without any orientation (argillasepic, porphyroskelic) (BC horizon, 110-120 cm). 



Because of shrinkage tension in the soil mate- 
rial, a decrease of plasma volume occurs and 
this produces the system of cracks and very 
dense ground mass (Figs. 1C, 2B) (Porphyroske- 
lic related distribution pattern in Brewer, 1976 
terminology). 

The clay soils have stress, related masepic 
fabric (striated b-fabric according to Bullock et 
al., 1985) below the surface soil, and this typical 
plasma separation (Figs. 1C, 2B) extends down 
to about 1 m, where the original glacio-lacus- 
trine sediments are found to be highly broken, 
maintaining their original unistrial fabric (sedi- 
mentary fabric) (Fig. ID). The proportion of 
unistrial fabric as measured by Dasog et al. 
(1987) increases with depth (Fig. 3). In the 
Sceptre, Regina, Melfort, and Tisdale soils, all 
horizons designated as C contained more than 
60% unistrial fabric, suggesting that moisture 



change frequency is minimal at this depth. 
Moisture measurement by de Jong and 
McDonald (1975) showed that little change in 
moisture occurred at a depth of 105-135 cm on a 
Sceptre clay. The transitional BC horizons have 
30-60% unistrial fabric that can be differenti- 
ated from the rest of the solum. This transi : 
tional zone also contains gypsum, which can be 
used as an additional characteristic in identifi- 
cation of this zone. 

Although micromorphology helps in precisely 
differentiating the transitional and C horizons, 
it may not be the most convenient and practical 
technique for horizon designation. A large sea- 
sonal moisture deficit may be responsible for 
better expression of masepic fabric as well as for 
the formation of slickensides. The higher 
amounts of iron-manganese nodules observed 
also suggest that masepic fabric is better ex- 



116 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Unistriol fabric ( */. ) - 

20 60 100 20 60 100 20 60 100 20 60 1 00 




I 



Fig. 3 Occurrence of unistrial fabric in the soils stud- 
ied, estimated by point counts (from Dasog et aL 
1987). 



pressed in the Rouleau and Kelvington soils. 
These two soils remain wet for a certain period 
of time. Desiccation of the soil, during an un- 
usually dry summer, likely creates very high 
tension, causing the formation of large cracks. 

Ap horizons display dominantly matrigranic- 
metamatrigranoidic fabric (associated with 
granular structure, Fig. 2A). As this type of 
fabric also is found in other soils, such as Cher- 
nozemic and Luvisolic soils, this characteristic 
alone cannot be used to differentiate Vertisols 
from other soils. 

Soil nodules, with sharp and diffuse bounda- 
ries, embedded in the soil matrix are other strik- 
ing features found in the clay soils. The number 
of nodules decreases with increasing depth. As 
will be discussed later, they possess strong evi- 
dence of the churning process. 

Mineralogy 

Several investigators (Warder and Dion, 
1952; Rice et aL, 1959; Kodama and Brydon, 
1965; Mills and Zwarich, 1972; Dudas and 
Pawluk, 1982) have reported the occurrence of 
smectite and illite as the predominant clay min- 
erals in the soils of the prairie provinces of west- 
ern Canada. Detailed characterization of some 
swelling lacustrine parent material from south- 
ern Saskatchewan (Mermut et aL, 1984) con- 
firmed that they are dominated by smectite and 
have a similar mineral composition. This sug- 
gests that the clays originate from the same 
source. The general order of clay mineral abun- 



dance, which remains almost constant through- 
out the profile, is smectite (> 50%), illite (15- 
30%), kaolinite (10%), vermiculite (< 10%), and 
quartz (< 10%). Smectites in Saskatchewan 
have high iron and their chemical composition 
are similar to smectites of Indian Vertisols 
(Mermut et aL, 1984). 

Dasog (1986) reported that vermiculite con- 
tents may be significant in swelling soils occur- 
ring in the northern subhumid grassland-forest 
transition and forest regions and, therefore, 
these soils are expected to have a higher layer 
charge. The degree of stress-induced plasma 
separation in the soils of the transitional zone is 
much higher than other swelling soils in Sas- 
katchewan. Several factors play a role in the 
development of soil plasma. Because of the low 
swelling abilities of vermiculites (due to their 
high charge), they form very distinctly oriented 
clay domains under stress. Southern Sas- 
katchewan soils have more organic matter in 
the surface horizons, which may have a masking 
effect on plasma separation. Contrarily, due to 
low organic matter in the subsoil, swell-shrink 
potential is likely higher in the northern soils. 
Furthermore, large fluctuations in the soil mois- 
ture state of northern soils may be responsible 
for a high magnitude of volume change that ulti- 
mately produces a high degree of plasma sepa- 
ration. 

The soils have up to 50% fine clay and have a 
high surface area (600-800 m2g- 1 ), important 
factors when considering the swell-shrink po- 
tential of these soils, cole values range from 0.1 
to 0.168, except for the surface horizons of the 
Melfort and Tisdale soils that have high con- 
tents of organic matter. Southern clays contain 
more than 15% exchangeable Na+ and high 
amounts of electrolytes at depth and the 
amounts of exchangeable Na and electrolytes 
decrease gradually towards the north. It is clear 
from the foregoing that, at this stage, it is very 
difficult to predict which soils are more prone to 
soil displacement. 

As indicated by Wilding and Tessier (1988), in 
the presence of high electrolyte, only interpar- 
ticle pore water contributes to field volume 
change. If the electrolyte concentration is low 
and smectites are saturated with Na, both inter- 
particle and a portion of the interlayer water 
contribute to field volume changes with chang- 
ing soil moisture content. High electrolyte con- 
centration in this regard may be a factor that 
reduces the swelling pressure in the BC and C 
horizon. Furthermore, this zone also contains 



MERMUT, ACTON, AND TARNOCAI: A REVIEW OP RECENT RESEARCH ON SWELLING CLAY SOILS IN CANADA 



117 



o 
o 



E 
o 



E 



d 
en 



O Apk (0-20 cm) 
Bmk 1(20- 50cm) 
BC (70-IIOcm) 




O 
O 



O 
O 



E 

o 



90 

60 
70, 
60 
50 
4O 



O 5 IO 15 20 25 30 35 40 

Water content (cm 3 g" 1 )xlOO 
MELFORT 



I 



a. 
to 



90 
85 
80 
75 
70 
65 1 
60 
55 

i 

50 



REGINA 




o Apk ( -20cm) 

O Bmk2 (62-88cm) 

BC2 (II3H35cm) 

' ' _ 



IO 15 20 25 30 35 40 45 



Water content (cm 3 gr 1 )xiOO 




a Ap (0-l7cm) 
O BC (65-IO5cm) 
A Ctk (105-KOcm) 



O 
O 



E 
o 



"3 



O 5 10 15 20 25 30 35 40 

Water content (cm 3 g" l )xlOO 




O Ap (0- 10cm ) 
A Bnt| (35-60cm} 
A Ck (I30-<63cm) 



5 IO 15 20 23 3O 35 4O 

Water content (cm 3 g~')xlOO 



Fig. 4 Shrinkage curves of selected horizons from four clay soils in Saskatchewan (from Dasog, 1986). 



appreciable amounts of gypsum. It is known 
that gypsum independently promotes floccula- 
tion of clays and lowers swell-shrink potential. 

Shrinkage curves of some selected horizons of 
the Sceptre, Regina, Melfort, and Tisdale (Fig. 
4) show that there is a shrinkage in the range of 
field moisture ;therefore cracking is expected in 
these soils. 

Soil Genesis and Classification 

Soil Genesis 

Swelling clays are found in three climatic re- 
gions in Saskatchewan: sub-arid, semi-arid, and 
sub-humid. However, the lack of information 
does not allow us to discuss the genesis of the 
soils in different climatic regions. The discus- 
sion provided here is more general and reflects 
our current knowledge. 



As indicated above, all clay soils in Sas- 
katchewan crack as a result of seasonal mois- 
ture fluctuation. It is postulated that the degree 
of the moisture deficit is higher for the sub-arid 
and semi-arid Sceptre and Regina than for the 
sub-humid Melfort, Tisdale, and Kelvington 
soils. As indicated by Soil Survey Staff (1975), 
the soils that are in vertic subgroups have po- 
tential for movement but may not become dry 
enough or moist enough to produce sufficient 
movement. 

In a laboratory study, Fredlund (1975) ob- 
served swelling pressures of about 400 to 1000 
kPa and shear strengths of 20 to 40 kPa when 
Regina soil was moistened. This shows that the 
smectite containing clays have high swelling 
pressure and low shear strength and, therefore, 
the potential for soil displacement and forma- 
tion of slickensides exists in these soils. As indi- 
cated above, slickensides are observed in the 



118 



SIXTH INTEKNATIONA! Soil CLASSIFICATION WORKSHOP 



subsoil in all four major soil zones in Sas- 
katchewan. This, according to Mermut and Ac- 
ton (1985), shows that heavy clay soils undergo 
sufficient cyclic cracking and swelling in most 
years to produce appreciable soil movement; 
hence the slickensides. 

Several mechanisms were suggested regard- 
ing the formation of slickensides (Ahmad, 1985; 
Wilding and Tessier, 1988). It is not the intent 
of this paper to elaborate on these mechanisms, 
but rather to suggest some ideas about the pos- 
sible pathways that result in the formation of 
slickensides in Saskatchewan soils. More than 
one cycle of wetting and drying is possible each 
year, once during the spring melt and once or 
twice during the summer. It is known that 
cracks promote bypass flow of water, which in- 
duces differential wetting in the subsoil. Dasog 
et al. (1987) suggest that such wetting occurs 
during the periods of heavy rain in the late 
summer or fall. In a year when fall precipitation 
is low, the soil may remain cracked during win- 
ter. In such years, differential wetting occurs 
due to bypass flow of meltwater. During this 
flow of water, in addition to crack infill by grav- 
ity, some surface fine granular structural aggre- 
gates also are transported. Tonguing of the sur- 
face soils and the presence of soil nodules, de- 
creasing with increasing depth, are strong evi- 
dence of downward movement of surficial mate- 
rial. Such a movement also was considered as 
evidence of the churning process. While recent 
reports (Wilding and Tessier, 1988) suggest that 
the crack infilling process is only partially func- 
tional in the genesis of Vertisols, it is a process 
that makes the Vertisols distinctly different 
from other soils. 

Our limited observation revealed that the 
slickensides in the subsoil are less frequent and 
they all have a similar nature. They are deepest 
and most frequent in the Rouleau soil. This is 
expected, as this soil is much deeper than other 
swell-shrink soils and it has a soil moisture re- 
gime that can be classified as aquic. Despite the 
presence of well preserved columnar and pris- 
matic structure and clay cutans in the B horizon 
of the Melfort and Tisdale soils (Fig. 1A, B), the 
occurrence of the slickensides in the subsoil 
raises several serious questions. It appears that 
slickensides form in the subsoils of these two 
soils by vertical overburden and horizontal swel- 
ling pressures. Displacement forces and associ- 
ated features generated in the subsoil do not 
cause any major disturbance in the upper part 
of the solum, and this part remains relatively 



stable. Yaalon and Kalmar (1978) reported that 
low overburden pressure and cracks would pre- 
vent high horizontal stress. Dudal and Eswaran 
(1988) identified several distinct zones or hori- 
zons in Vertisols. They recognize that surface 
soils, subject to cracking, have large prisms 
which may part to coarse, angular blocky ele- 
ments. Slickensides were never found within 
the upper 20 cm of the Vertisols. 

While the above discussion may confirm that 
pedoturbation is an unimportant process in the 
Melfort and Tisdale soils, it suggests that the 
influence of subsoil displacement on the upper 
solum may be lesser than in the soils of sub-arid 
and semi-arid regions (Sceptre and Regina), 
Relatively thicker soils observed in depressional 
areas in the Regina clay plain may explain the 
surficial rearrangement of material in these 
swell-shrink soils (Mermut and Acton, 1985). In 
addition, these studies on Saskatchewan soils 
show that the features related to swelling and 
shrinking are not as distinct as in comparable 
subtropical and tropical soils. 

Soil Classification According to Soil 
Taxonomy 

The swell-shrink soils in Saskatchewan, de- 
scribed above, meet the requirement of the new 
concept of Vertisol as described in the 5th circu- 
lar letter by Comerma (1989) and can, therefore, 
be classified within the Vertisol soil order. Ear- 
lier, Vertisols that have frigid or colder soil tem- 
perature regimes were excluded in SMSS Soil 
Taxonomy for unknown reasons (Comerma et 
al,, 1988). The soils do not meet the criteria of 
any suborders of the Vertisol described by Soil 
Survey Staff (1975). The only suborder that is 
closest to this group of soils is Xererts. However, 
clay soils in Saskatchewan do not have xeric 
moisture regime for two reasons: 1) cracks may 
open and close more than once each year, and 2) 
the xeric moisture regime is that typified in 
Mediterranean climates. Therefore, based on 
soil temperature regime, a new suborder, "Bor- 
erts," equivalent to Borolls and Boralfs, is pro- 
posed (Dasog et al., 1987). By definition, Bor- 
erts are Vertisols that have frigid or cryic soil 
temperature regimes. This is slightly different 
than that proposed by Comerma et al. (1988) in 
which they exclude frigid temperature regime 
from the definition. 

In the 5th circular letter by Comerma (1989) 
it is recommended ( J. Witty) that the name Bor- 
ert be changed to Cryert. Temperatures re- 
ported by Treidl (1979) suggest that the tern- 



MERMUT, ACTON, AND TARNOCAI: A REVIEW OF RECENT RESEAKCH ON SWELLING CLAY SOILS IN CANADA 



119 



perature regime of the southern Saskatchewan 
soils is frigid and that of northern soils is cryic 
(Boreal and Cryoboreal as defined in Canada). 
By definition, the soil temperature regime of 
Boreal environment is frigid or cryic, but some 
have a pergelic temperature regime (Soil Survey 
Staff, 1975). This means that a range of cold 
temperature regimes is included in Boreal. 
Therefore, it seems more appropriate to keep 
the term Borert so that the soils that have both 
frigid and cryic soil temperature regimes can be 
recognized under one suborder. 

There are an estimated 14560 km2 of the 
Sceptre, Regina, and similar soils in the Brown 
and Dark Brown soil zones alone in Canada 
(Clayton et al., 1977) and appreciable areas 
within the Black (Mermut and St. Arnaud, 
1983) and Gray soil zones of Saskatchewan and 
Manitoba that may meet the definition of Bor- 
erts. The swelling clays in the Red River Valley 
of North Dakota, the Grumusols of Montana 
(Hogan et al., 1967), and gilgaied soils in South 
Dakota (White and Bonestell, 1960) also may be 
included in this suborder. 

Perhaps two great groups, Haploborerts and 
Humiborerts, can be recognized in Borerts. 
Soils such as Kelvington have a high content of 
organic matter, and should be recognized and 
classified as Humiborerts. We have made this 
proposal to the International Committee and we 
hope that the meeting will enable us to clarify 
this point. By definition, Humiborerts have 12 
kg or more organic C in a unit volume of I 
square meter to a depth of 1 meter below the top 
of the mineral soil surface, exclusive of any 
horizon that may be present. 

Several subgroups have been suggested in 
Comerma's 5th circular letter (1989). Our ob- 
servations have shown that soils in sub-arid and 
semiarid regions of Saskatchewan have one or 
more horizons with an ESP of >15 or a SAR of 
>13 within a depth of 100 cm of the soil surface 
and meet the requirement of "sodic" subgroup 
(Table 1). According to Wilding and Tessier 
(1988), clays may be dispersed at relatively low 
SAR and ESP (5 to 8% or less) 
under low electrolyte systems and 
an ESP of 15% or more may be 
required in high electrolyte sys- 
tem. Poorly drained soils occur- 
ring in depressional areas, that 
are saturated with water for some 
time, can be classified as "aquic." 
Soils in the Kelvington area have 
a moist color value of 4 or a dry 



value of 6 or more in one or more subhorizons 
within the upper 30 cm of the soil surface, re- 
flecting the influence of parent material. 

These soils have low organic matter and very 
high clay contents. The dark soils that have a 
moist color value of 3 or less, and chroma of 2 or 
less, and a dry color value of 5 or less in all sub- 
horizons within the upper 30 cm of the surface, 
previously grouped in the "pell" categories, will 
be called "typic." Considering the 5th circular 
letter by Comerma (1989), the following three 
subgroups are suggested within Haplo and 
Humiborerts: 

Sodic Haploborerts are Borerts that have 
within a depth of 100 cm of the soil surface 
one or more subhorizons with an ESP of > 
15 or SAR of ^ 13. 

Chromic Haploborerts are Borerts that have 
a dominant value, moist, of 4 or more and a 
chroma of 3 or more or a value, dry of 6 or 
more in some part of the upper 30 cm of the 
soil surface. 
Typic Haploborerts, other Haploborerts. 

Definition of Vertic Horizon 

As indicated above, a definition of a diagnos- 
tic vertic subsurface horizon, such as Bw, is 
needed. Slickenside formation, as a result of soil 
displacement, is an essential process for Verti- 
sols. The definition, therefore, should be cen- 
tered around the features related to soil dis- 
placement. Soils that show slickensides and 
wedge shaped aggregates generally are deeper 
and have enough swelling clay to promote crack- 
ing during the dry season. It seems that slicken- 
side and wedge shaped aggregates are the most 
essential features for Vertisols. Comerma et al. 
(1988) suggest that gilgai should no longer be a 
required criterion at the order level to classify a 
soil as a Vertisol. 

Using the existing criteria for Vertisol (Soil 
Survey Staff, 1982), the following simple defini- 
tion can be made. A vertic horizon is a subsur- 
face horizon or horizons with slickensides close 
enough to intersect, or wedge shaped peds that 



Table 1. 


SAR and exchangeable Na% in the Rouleau and Kelvington soils. 


Rouleau 
Depth cm 


Exchangeable 
SAR Na% 


Organic 
carbon% 


Kelvington 
Depth cm SAR 


Exchangeable Organic 
Na% carbon % 


0-15 
15-29 
29-46 
46-71 
71-97 
97-133 
133-163 
163-200 


1 
1 
4 
11 
15 
21 
14 
1 


1 
1 
3 
9 
10 
8 
7 
1 


1.59 
1.18 
1.07 
0.84 
0.51 
0.42 
0.49 
0.57 


0-10 
10-31 
31-65 
65-99 
99-131 
131-183 
183-210 


RR 
1 
1 
2 
2 
1 


1 
1 
2 
2 
3 
3 
3 


3.00 
0.80 
0.48 
0.40 
0.35 
0.33 
0.23 



120 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



have their long axes 10 to 60 from the horizon- 
tal. A vertic horizon is 25 cm or more thick, with 
its upper boundary within 100 cm of the soil 
surface. 

Proposed Revisions in the Canadian 
System 

Although the number of representative soils 
observed and analyzed in Saskatchewan may 
not be warrant to propose changes in the Cana- 
dian classification system. As it stands, the fol- 
lowing proposal is an outline to form the basis 
for establishing the need for changes in the sys- 
tem. More work is needed to determine the ex- 
tent and distribution of the swelling clay soils in 
Alberta and Manitoba. 

The soils that have diagnostic vertic subsur- 
face horizons may be classified within a new 
"Vertisolic" order in the Canadian System of Soil 
Classification. There are presently two funda- 
mental problems within the classification of clay 
soils: 1) there are examples of these soils occurr 
in sub- and semi-arid regions which do not meet 
color criteria of a Chernozemic A horizon and 2) 
considering the evidence for a B horizon, be- 
cause of the presence of carbonates, these soils 
also should not be classified in the Regosolic 
order. It is also difficult to reconcile "grumic" as 
a family criterion among other recognized crite- 
ria, as this criterion is related to genesis but not 
properties and has to be recognized at a higher 
level. This suggestion for a revision is consis- 
tent with the philosophy of the system, whereby 
classes at higher categorical levels reflect the 
broad differences in the soil environments that 
are related to the differences in soil genesis 
(Canada Soil Survey Committee, 1978). 

Using the same principles of the CSSC Sys- 
tem, four great groups can be established: 

Brown, Dark Brown, Black, and Dark Gray 
(including Gray). 

Within each group the following subgroups 
can be recognized: 

Orthic, Rego, Calcareous, Solonetzic, Gleyed, 
and combination of these. 

Most soils in Saskatchewan that have high 
shrink-swell potential have cracks for some pe- 
riod in most years and a potential LE of 6 cm in 
the upper 1 m. These soils that have potential 
but do not exhibit the required characteristics 
should be recognized within the Vertic sub- 
group, such as Vertic Dark Brown Chermozem. 



References 

Ahmad, N. 1985. Acid Vertisols of Trinidad, p. 141-151. 
In Proceedings of 5th Int. Soil Correlation Workshop, 
Soil Survey Administration, Ministry of Agriculture 
and Irrigation, Khartoum, Sudan. 

Ayres, K.W., D.F. Acton and J.G.Ellis. 1985. The soils of 
the Swift Current map area, 72J Saskatchewan. Ex- 
tension Publ. No. 481 University of Saskatchewan, 
Saskatoon. 

Brewer, R. 1976. Fabric and mineral analysis of soils. 
Robert E. Kriger Publishing Co., Huntington, N.Y. 

Bullock, P., N. Fedoroff, A. Jongerius, G. Stoops and T. 
Tursina. 1985. Handbook for soil thin section descrip- 
tion, Waine Research Publ., Wolverhampton, England. 

Canada Department of Agriculture. 1974. The system of 
soil classification for Canada, Publ. 1455, Ottawa, 
Canada. 

Canada Soil Survey Committee. 1978. The Canadian 
system of soil classification. Can. Dep. Agric. Pub. 
1646, Supply and Services, Canada, Ottawa. 

Clayton, J.S. 1963. Report on classification of Chernoz- 
emic soils. In report on the fifth meeting of the Na- 
tional Soil Survey Committee of Canada, 4 - 8 March. 
Winnipeg, Manitoba. National Soil Survey Committee, 
Agriculture Canada, Ottawa. 

Clayton, J.S., W.A. Ehrlich, D.B. Cann, J.H. Day, and LB. 
Marshall. 1977. Soils of Canada, Vol. II Soil inventory. 
Research Branch, Can. Dept. Agric. Ministry of Supply 
and Services Canada, Ottawa. 

Comerma, J.A. 1989. Fifth circular letter International 
Committee on Vertisols. USDA-SCS, Washington, 
B.C. 

Comerma, J.A, D. Williams and A. Newman. 1988. Con- 
ceptual changes in the classification of Vertisols. p. 41- 
54. In L.P. Wilding and R. Puentes (ed.) Vertisols: 
Their distribution, properties, classification and man- 
agement. Texas A and M University Printing Center, 
College Station, TX 77843. 

Grossman, R.B., W.D. Nettleton, B.R. Brasher. 1985. 
Application of pedology to plant response prediction for 
tropical Vertisol. p. 97-116. In Soil Survey Administra- 
tion Khartoum Sudan (ed.), Proceedings of the 5th In- 
ternational Soil Classification Workshop - Taxonomy 
and Management of Vertisols and Aridisols, Part I 
papers 2-11 November, Khartoum, Khartoum Univer- 
sity Press, Khartoum, Sudan. 

Dasog, G.S. 1986. Properties, genesis and classification 
of clay soils in Saskatchewan. Ph.D. Thesis, University 
of Saskatchewan, Saskatoon, Canada. 

Dasog, G.S., D.F. Acton and A.R. Mennut. 1987. Genesis 
and classification of clay soils with vertic properties in 
Saskatchewan. Soil Sci. Soc. Am. J. 51: 1243-1250. 

de Jong, E. and K.B. McDonald. 1975. The soil moisture 
regime under native grassland. Geoderma 14: 207-221. 

Dudal, R. andH. Eswaran. 1988. Distribution, properties 
and classification of Vertisols. p. 1-22. In L.P. Wilding 
and R. Puentes (ed.) Vertisols: their distribution, prop- 
erties, classification and management. Texas A and M 
University Printing Center, College Station, TX 77843. 



MERMUT, ACTON, AND TARNOCAI: A REVIEW OP RECENT RESEARCH ON SWELLING CLAY SOILS IN CANADA 



121 



Dudas, M.J. and S. Pawluk. 1982. Reevaluation of the 
occurrence of interstratified clays and other phyllosili- 
cates in southern Alberta soils. Can. J. Soil Sci. 62: 61- 
69. 

El Abedine,Z. and G.H.Robinson. 1971. A study of crack- 
ing in some Vertisols of Sudan. Geoderma 5: 229-241. 

Fredlund, D.J. 1975. Engineering properties of expansive 
clays. Internal. Research Report (IR-7), Transporta- 
tion and Geotechnical Group, Department of Civil En- 
gineering, University of Saskatchewan. 

Hogan, E.K, J.L. Parker, V.K. Haderlie, R.C. McConnell, 
and W.W. Janssen. 1967. Soil Survey of Judith Basin 
area, Montana-series 1959, no. 42. USDA-SCS. U.S. 
Government Printing Office, Washington, B.C. 

Kodama, H. and J. E. Brydon. 1965. Interstratified 
montmorillonite-mica clays from subsoil of the Prairie 
Provinces, Western Canada. Clays Clay Miner. 13: 
151-173. 

Mermut, A.R. and D.P. Acton. 1985. Surficial rearrange- 
ment and cracking in swelling clay soils of the glacial 
lake Regina basin in Saskatchewan. Can. J. Soil Sci. 
65: 317-327. 

Mermut, A.R. and R.J. St. Arnaud. 1983. Micromorphol- 
ogy of some Chernozemic soils with grumic properties 
in Saskatchewan, Canada. Soil Sci. Soc. Am. J. 47:536- 
541. 

Mermut, A.R., K. Ghebre-Egziabhier, and R. J. St. Arnaud. 
1984. The nature of smectites in some fine textured 
lacustrine parent materials in southern Saskatchewan. 
Can. J. Soil Sci. 64: 481-494. 

Mills, J.G. and M.A. Zwarich. 1972. Recognition of inter- 
stratified clays. Clays Clay Miner. 20: 169-174. 

Mitchell, J., H.C. Moss and J.S. Clayton. 1944. Soil sur- 
vey of southern Saskatchewan. Soil Survey Rep. No. 
12, University of Saskatchewan, Saskatoon, Canada. 

Rice, H.M., S.A. Forman and L.M. Party. 1959. A study of 
some profiles from major soil zones in Saskatchewan 
and Alberta. Can. J. Soil Sci. 39: 165-177. 



Schafer, W.M. and M.J. Singer. 1976. Influence of physi- 
cal and miner alogical properties on swelling of soils in 
Yolo county, California. Soil Sci. Soc. Am. J. 40: 557- 
562. 

Soil Management Support Services. 1986. Designation 
for master horizons and layers in soils. Department of 
Agronomy, College of Agriculture and Life Sciences, 
Cornell University. 

Soil Survey Staff. 1975. Soil Taxonomy: A basic system of 
soil classification for making and interpreting soil sur- 
veys. USDA-SCS. Agric. Handb. 436. U.S. Govern- 
ment Printing Office, Washington, D.C. 

Soil Survey Staff. 1982. Amendment to Soil Taxonomy, 
part 615. In National Soil Taxonomy Handbook issue 
No. 1. 430-VI-NSTH, 1982, USDA-SCS, Washington, 
D.C. 

Treidl, R.A. 1979. Handbook on agricultural and forest 
meteorology, Part II. Fisheries and Environment Can- 
ada, Ministry of Supply and Services, Ottawa, Canada. 

Warder, F.G. and H.G.Dion. 1952. The nature of the clay 
minerals in some Saskatchewan soils. Sci. Agric. 32: 
535-547. 

White, E.M. and R.G. Bonestell. 1960. Some gilgaied soils 
in South Dakota. Soil Sci. Soc. Am. Proc. 24: 305-309. 

Wilding, L.P. and D. Tessier. 1988. Genesis of Vertisols: 
Shrink-swell phenomena, p. 47-62. In L.P. Wilding 
and R. Puentes (ed.) Vertisols: Their distribution, 
properties, classification and management. Texas A 
and M University Printing Center, College Station, TX 
77843. 

Yaalon, D.H. and D. Kalmar. 1978. Dynamics of cracking 
and swelling clay soils: displacement of skeletal grains, 
optimum depth of slickensides, and rate of intrape- 
donic turbation. Earth Surface Processes, 3: 31-42. 



122 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Thermal Regime and Morphology of Clay Soils in Manitoba, Canada 

G.R Mills, 1 E.G. Eilers, 2 and EL Veldhuis 2 

Abstract 

Soil thermal regime currently takes precedence over physical features 
such as continuous cracking and slickensides in the classification of clayey 
soils in the Vertisol order in Soil Taxonomy. Thermal regime data from a 
transect of nine clayey soils from latitude 49*N to 5617'N were used to 
evaluate the extension of Cryerts (cold Vertisols) to northern latitudes. Soil 
and vegetation conditions range from Black and Dark Gray Chernozems 
(Typic Cryoboroll and Argic Cryoboroll) under subhumid grassland and 
grassland-forest transition to Gray Luvisols (Cryoboralf) in the cool, more 
humid, Boreal forest. Soil thermal regimes vary from cold to very cold. The 
most northerly Luvisols have permafrost within 1.5 m, resulting in a very 
cold thermal regime. Mean annual and mean summer soil temperatures at 
50 cm decrease northward at a rate of 0.9*C and 1.5*C per degree of latitude, 
respectively. 

These soils have the clay content and mineralogy with potential to de- 
velop vertic soil properties. Although vertic features were not noted on the 
transect, observations of closely comparable soils throughout the same re- 
gions indicate a decreasing presence of vertic features from southern Black 
soils to the northern Luvisols. Vertic properties are sufficiently common in 
the Black and Dark Gray soils for their classification as Cryerts if the tem- 
perature limit is altered to include colder soils. Cold to very cold Luvisols 
and very cold Cryosols in combination with a generally higher micaceous 
clay content, more humid soil moisture regime, and cryoturbation effects 
have less potential for development of vertic soil features. Consequently, 
vertic properties are not sufficiently common or well developed in Luvisol 
or Cryosol soils to warrant classification in the proposed Cryert suborder. 



Introduction 

Manitoba is centrally located in the mid lati- 
tudes (49oN to 60oN) of North America. The 
ecological zonation across this latitude ranges 
from grassland and grassland-forest transition 
vegetation with Chernozemic Black (Cryoboroll) 
soils in the south, to Boreal forest associated 
with Luvisolic (Cryoboralf) soils in central re- 
gions, and open Subarctic forest and Tundra 
associated with Cryosolic (Pergelic Ruptic 
Cryochrept) soils in the north. Clay soils cover 
some 7,000 kha (Canada-Manitoba Soil Survey, 
1989) between latitude 49N and 58N (Figure 
1). 

The objective of this paper is twofold: 1) to 
describe the thermal regime of well to moder- 
ately well drained clay soils under forest cover 
as determined from nine benchmark sites along 
a south-north transect, and 2) to evaluate the 
possible extension of Cryerts (cold Vertisols) in 
northern latitudes. 

Geological Setting 

Physiographically and geologically the study 
area transects two large, distinctly different 

1 Soils and Crops Branch, Manitoba Agriculture. 

2 Land Resource Research Center, Agriculture Canada, 362 
Ellis Building, University of Manitoba, Winnipeg, MB, 
R3T2N2. 



areas: old, massive, Precambrian crystalline 
rock forming the Canadian Shield in the north 
and younger, mainly sedimentary rock of the 
Manitoba Plain and the Saskatchewan Plain in 
the south (Bostock, 1970). The entire area was 
subjected to multiple glaciation during the 
Pleistocene Ice Age (Prest, 1970). 

During the final recession of the continental 
ice sheet, extensive areas were inundated by 
glacial Lake Agassiz ponded between higher 
lands to the south and the receding ice front. 
Extensive areas of clayey- textured sediments in 
the Shield region occur as level lacustrine basins 
and as undulating and hummocky plains where 
the bedrock is closer to the surface. The soils 
range in drainage from well to poor in undulat- 
ing to hummocky terrain and poorly to very 
poorly drained in flat lying terrain. In the Mani- 
toba Plain, lacustrine clay deposits blanket the 
level to gently undulating morainal deposits in 
which low-lying areas are dominantly imper- 
fectly and poorly drained. 

The lacustrine clays of both regions are fine 
clayey and dominantly moderately calcareous. 
Sediments in the Manitoba Plain are derived 
primarily from Cretaceous shale bedrock which 
underlies the Saskatchewan Plain to the west. 



MILLS, EILERS, AND VELDHUIS: THERMAL REGIME AND MORPHOLOGY OF CLAY SOILS IN MANITOBA, CANADA 



123 




AREA OF CLAY SOILS IN POLYGONS 

j ""{ DOMINANT! 40%) 

l\~ I 5U8DOMlNarJT(l5-4O%) 

PHYSIOGRAPHIC DIVISIONS 

[ 1 1 HUDSON BAY LOWLAND 

[ 2 ] CANADIAN PLAIN 

| 3 1 MANITOBA PLAIN) 

| 4 ] SASKATCHEWAN PLAIN 



Figure 1. Physiographic divisions and extent of clayey soils in Manitoba. 



In contrast, sediments in the Shield consist of 
clays derived from both Cretaceous rock and the 
local Shield rock. Although clay mineralogy is 
mixed in both regions, high shrink-swell clays 
such as smectite are co-dominant with mica (il- 
lite) in the south, and illitic clay derived from 
the Shield rock is dominant to the north of Lake 
Winnipeg. (J.G. Madden, personal communica- 
tion). 

Climatic Conditions 

Manitoba, due to its location in interior North 
America at mid latitudes, is characterized by a 
continental climate with relatively short, warm 
summers and long, cold winters, with continu- 
ous snow cover from about November 20 to 
March 30 in the south and from about October 
30 to May 1 in the north. The climatic data 
along a south-north transect through the study 
area show a gradual cooling and a decrease in 
length of frost-free season with increasing lati- 
tude (Table 1). At the north end of Lake Win- 
nipeg, climatic conditions result in the first oc- 



currence of permafrost in organic deposits. Per- 
mafrost is characterized by some depth in the 
soil where the temperature will remain below 
0C over at least two consecutive winters and 
the intervening summer (Brown and Kupsch, 
1974). 

Although total precipitation generally de- 
creases to the north, the atmospheric moisture 
balance along the transect shifts from seasonal 
summer deficit in the south to very short periods 
of deficit or slight surplus in the north. The 
change in moisture deficit results from reduced 
evapotranspiration due to the generally cooler 
and shorter growing season in the north. 

Ecoclimatic Regions 

Ecoclimatic regions are relatively homoge- 
nous areas of the Earth's surface characterized 
by distinctive ecological responses to macrocli- 
mate as expressed by vegetation, soils, fauna, 
and aquatic systems (Ecoregion Working Group, 
1989). Broad climate-soil-vegetation relation- 
ships are described in terms of seven major 



124 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Table 1. Climatic data from selected Manitoba weather stations (Atmospheric Environment Service, 1980). 



Station Latitude Longitude 

north west 



Mean air temperature CC\ Mean Frost-free Precipitation (mm) Period of continuous 

period (days) snow cover* 
Annual Jan July Annual May-Sept. 



Grassland transition Region, Gt 

Emerso 4900' 97'16' 3.1 -18.2 17.5 125 

Winnipeg 4954' 9714' 2.1 -19.3 19.6 122 

Stonewall 50"07' 97"20' 1.1 -20.3 19.0 118 

Low Boreal subhumid Region, LBs 

Gypsumville 51'40' 9844' 0.7 -20.7 18.0 101 

Mid Boreal subhumid Region, MBs 
Grand Rapids 5311' 9916' -0.7 -21.8 17.7 101 

High Boreal subhumid Region, HBs 

Wabowden 54'55" 98'38' -2.2 -24.6 16.9 105 

Thompson 5548' 97'52' -3.9 -26.6 15.6 63 

Low Subarctic Region, LS 
Gillam 5621' 94'42' -4.6 -26.9 15.1 60 



515.7 
525.5 
538.23 

418.4 
442.7 

464.2 
542.4 

422.3 



343.6 
350.2 
38.9 

271.2 
267.9 

297.6 
3303 

278.8 



Nov. 21 -March 30 

Nov. 23 - April 12 
Nov. 15 - April 22 

Oct. 31 -April 28 
Oct. 29 -April 21 

Oct. 30 - May 1 



* Average date of persistent snow cover based on first date for 2 cm of snow cover lasting 7 days and last date of 2 cm 
of snow cover for last continuous 7 days. Winnipeg Climate Centre, Unpublished data from National Climatological Archive. 



ecoclimatic regions. Figure 2 shows the loca- 
tions of these regions and Table 2 provides a 
brief summary of their characteristics. (Mani- 
toba Ecoclimatic Region Working Group, 1985). 

Materials and Methods 

Location and Description of Study 
Transect 

Figure 2 shows the locations of the nine 
benchmark sites on the south-north transect. 
Table 1 provides the climatic data representa- 
tive of the four ecoclimatic regions along the 
transect, and Table 2 gives soil classification, 
site, and vegetation characteristics. The soils 
included in this study are classified as Black 
Chernozem, Dark Gray Chernozem, and Gray 
Luvisol in the Canadian system of soil classifica- 
tion (Agriculture Canada Expert Committee on 
Soil Survey, 1987). The equivalent classifica- 
tion in Soil Taxonomy (Soil Survey Staff, 1975) 
is Typic Cryoboroll, Argic Cryoboroll, and Cry- 
oboralf, respectively. 

All soils are developed on lacustrine clay, with 
the exception of the soil at site 62O6, which de- 
veloped on mixed clay and loamy morainal ma- 
terial. Table 3 describes the morphology of each 
soil along the transect. Table 4 provides addi- 
tional physical and chemical characteristics for 
representative Black and Dark Gray soils and 
two Gray Luvisol soils (one from southern Mani- 
toba and one from the north) which were se- 
lected to closely approximate the clay soils en- 
countered in the transect. 

Soil Thermal Regime 

Soil Temperature 

Soil temperatures were measured at 6 depths 
(5,10,20,50,100, and 150 cm) using thermistors 



(1986-1989). Sampling frequency averaged 8 
times per year in a random manner. Sampling 
included measurements in January and Sep- 
tember of each year, to represent a winter and a 
late summer measurement, respectively. How- 
ever, minimum soil temperatures have not been 
reached at lower depths in January, and, in 
September, soil temperatures near the surface 
have started to cool, particularly at the northern 
sites. 

The soil thermal regime at each site was 
evaluated using the three years of thermistor 
data from the 20, 50, 100, and 150 cm depths 
(Table 5). The observed soil temperatures from 
June 1986 to June 1989 were fitted to an annual 
sine curve to mathematically calculate the best 
fitting line to the data and to derive a daily nor- 
mal temperature for each day of the year (Re- 
imer and Shaykewich, 1980). Then the mean 
annual soil temperature (MAST) and mean 
summer soil temperature (MSST) for the period 
June 1 to September I were calculated. Mean 
January and mean September soil temperature 
at each depth were calculated from the 1986 to 
1989 mid-monthly data. 

Soil Thermal Gradients 

The latitudinal gradient in soil thermal re- 
gime is based on MAST and MSST values calcu- 
lated from soil temperature measured at 50 cm 
for the soils at each benchmark site. The data 
for the nine benchmark sites were grouped and 
averaged to provide an estimate of the rate of 
change with increasing northern latitude. Fig- 
ure 3 includes mean annual air temperatures 
along the transect. Figure 4 shows the vertical 
gradient for 1988 for each benchmark site as 
representative of a winter (January) and late 
summer (September) regime. 



MILLS, EILERS, AND VELDHUIS: THERMAL REGIME AND MORPHOLOGY OF CLAY SOILS IN MANITOBA, CANADA 125 



58' 



Results and 
Discussion 

Soil 

Morphology 
and Properties 

Morphological 
di f f e r e n c e s 
among the soils 
in the transect 
reflect prevailing 
climate and vege- 
tation at each 
site. Black Cher- 
nozem soils in 
the south (sites 
62H13 and 

62H4) are char- 
acterized by dark 
colored organic- 
rich Ah horizons 
(Table 3) due to 
relatively high 
levels of organic 
carbon (Soil 1, 
Table 4). 

The Dark Gray 
soil (site 62O6) 
occurs in the 
transitional zone 
of mixed grass- 
land and forest 
vegetation be- 
tween the Black 
soils and the 
Gray Luvisols to 
the north. Dark 
Gray soils are 
characterized by 
dark colored sur- 
face horizons 
high in organic 
carbon (Soil 2, 
Table 4) but 
which tend to dry 
to lighter colors 
and show struc- 
tural modifica- 
tion due to in- 
creased eluvia- 
tion. The underlying B horizon contains accu- 
mulation of clay resulting in coarse granular to 
fine subangular blocky structure. 

Gray Luvisol soils (sites 63B2, 63J3, 63J2, 



60 



50- 




52- 



49 JL 



A Benchmark site 

AES Meteorological station 



Area Of Clay Soils In Polygons 
Dominant (> 40%) 

Subdorninant (15-40%) 



Ecoc lima tic Regions 

Grassland Transition 
Low Boreal , subhumid 

Mid Boreal ,subhumid 
High Boreal, subhumid (South) 
-High Boreal, subhumid (North) 

Low Subarctic 

High Subarctic 

Low Arctic 



HBs 2 

LS 
HS 

LA 



IOO 



49 



Permafrost Regime 

Southern limit of continuous 
permafrost 

Southern limit of permafrost 



Figure 2. Ecoclimatic regions, permafrost regime, 
and location of benchmark sites. 



63O2, 63P19 and 64A4) from the Mid and High 
Boreal Ecoclimatic regions show stronger leach- 
ing as a result of the growth and decomposition 
of forest vegetation. Their main characteristics 



126 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Table 2. Ecoclimatic Region and Benchmark Site Characteristics 


Ecoclimatic Regions 


Benchmark Site 


Soil Order or Vegetation Type 


Site 


Soil Subgroup* Elev. 


Vegetation Type 


Regional Occurrence 


Great Group (natural) 




No. (m) 


(actual) 


ofVertic 










Characteristics** 


Grassland transition region, Gt 










Black Chernozem Grassland- As pen 


62H13 


Rego Black 234 


Oak, aspen 


Surface cracking to 


Parkland 


62H14 


Orthic Black 227 


Oak, aspen 


1.5 m; slickensides 


Low Boreal subhumid region, LBs 










Dark Gray Mixed Deciduous- 


62O6 


Dark 246 


Aspen, oak, white 


Surface cracking to 


Chernozem, Organic Coniferous Forest 




Gray 


spruce 


0.5 m; slickensides 


Mid Boreal subhumid region, MBs 










GrayLuvisol, Mixed Forest 


63B2 


Orthic Gray 248 


Mixed white spruce- 


Shallow cracks upper 


Organic 




Luvisol 


aspen 


solum; slickensides 


High Boreal subhumid region, HBsl 










Gray Luvisol, Coniferous and 


63J3 


Orthic Gray 237 


Black spruce, aspen, 


Shallow cracks upper 


Brunisolic, Mixed Forest 




Luvisol 


black poplar, balsam fir 


solum 


Organic 


63J2 


Orthic Gray 225 


Black spruce, black 








Luvisol 


poplar 






6302 


Orthic Gray 220 


Black spruce, jack pine 








Luvisol 






High Boreal subhumid region, HBs2 










GrayLuvisol, Coniferous 


63P19 


Solonetzic Gray 210 


Jack pine, black spruce 


Cracking in upper 


Brunisolic, Forest 




Luvisol 


aspen 


solum; stress 


Organic, Cryosolic 


64A4 


Orthic Gray 180 


Black spruce, few white 


Surface cracking and 






Luvisol (cryic) 


spruce 


slickensides not 


observed 










Low Subarctic region, LS 










Brunisolic, Open Coniferous 




No Sites 












Cryosolic Forest 










High Subarctic region, HS 










Crvosolic Ooen Coniferous 




No Sites 






Brunisolic Forest & Tundra 










(shrub) 










Low Arctic region, LA 










Cryosolic Tundra (shrub) 




No Sites 
















*Soil Classification according to The Canadian System of Soil Classification. Agric. Canada Expert Committee on Soil Survey 


, 1987. 


**Vertic characteristics observed on a regional basis and not necessarily at the benchmark soil site in the transect. 



are A horizons with platy or granular structure, 
which are light colored on drying due to the elu- 
viation of clay, organic matter and iron and alu- 
minum sesquioxides. The underlying illuvial B 
horizons are characterized by subangular blocky 
structure (Table 3) and accumulation of clay 
(Table 4). 

Leaching of CaCOS and reduced pH in the 
sola accompanied by downward movement of 
clay and other associated colloidal materials is 
characteristic of Luvisolic soils (Soil 3 and 4, 
Table 4). The clayey parent materials of these 
soils are moderately to strongly calcareous with 
clay content usually in excess of 60 percent 
(Table 4). The six Luvisolic soils in the transect 
occur over a latitudinal distance that results in 
decreasing soil temperature from south to north 
and the occurrence of permafrost in the most 
northerly Luvisol (site 63A4). 

The soils in this transect traverse two distinct 
regions of clay mineralogy. The fine clay frac- 
tion in the parent materials of both regions is 
characterized by mixed mineralogy dominated 
by illite north of Lake Winnipeg and smectite 
codominant with illite to the south. Shrink- 
swell phenomena and development of Vertisols 



have been observed in such materials but are 
more common in the south. 

Although continuous cracking and slicken- 
sides were not noted in the benchmark soils in 
the transect, observations on similar clayey 
materials and landscapes in the same regions 
indicate that these vertic features do occur in 
Black and Dark Gray soils and occasionally in 
Gray Luvisols. Cracking between 1 and 1.5 m 
deep has been observed in Black Chernozems in 
most years. The cracks are less continuous and 
more shallow (about 0.5 m in depth) in Dark 
Gray soils and generally are restricted to shal- 
low cracking in the upper solum of Gray Luvi- 
sols (0.3 to 0.5 m). 

Slickensides are most common in well and 
imperfectly drained Black and Dark Gray soils. 
Prominent slickensides are described in clayey 
Luvisols in east-central Saskatchewan (Acton et 
al., 1989) but have not been observed in the 
colder Luvisols in Manitoba north of Lake Win- 
nipeg. The potential for differential swelling 
pressures and slickenside formation is reduced 
in these soils due to cold (cryic) and very cold 
(pergelic) thermal regimes and the dominance of 
illitic clay mineralogy in the parent material. 



MILLS, EILERS, AND VELDHUIS: THERMAL REGIME AND MORPHOLOGY OP CLAY SOILS IN MANITOBA, CANADA 



127 



Soils in which vertic features 
are most common are usually af- 
fected by pronounced oscillation 
between the wet and dry moisture 
state (Wilding and Tessier, 1988). 
Northern soils in the transect usu- 
ally do not dry out to the same 
degree as the southern soils, due 
to reduced evapotranspiration re- 
sulting from shorter growing sea- 
son, increased proportion of conif- 
erous vegetation, and lower air 
temperature causing reduced 
demand for soil moisture. As a 
result, soil moisture fluctuation in 
the clay Luvisol soils, particularly 
deeper in the profile, is less pro- 
nounced than that found in the 
Black and Dark Gray soils in the 
south. In addition, in the most 
northerly soils containing perma- 
frost, gradual melting of the upper 
layer of frozen material during the 
thaw season helps to maintain soil 
moisture levels in the surface soil. 
Potential for physical movement 
in soils north of Lake Winnipeg is 
further reduced by a lower content 
of high shrink-swell clay, due to 
the dominance of illitic clay. 

Thermal Regime of Clay Soils 

The thermal regime of well 
drained clay soils under forest 
vegetation shows regional vari- 
ability along the south-north 
transect and seasonal variation 
within the soil profile at each of 
the nine benchmark sites. 

Regional Thermal Regimes in 
Manitoba 

Mean Annual Soil Temperature 
(MAST) values at 50 cm range 
from a high of 6.6oC in a Black 
Chernozem soil to a low of 0.2 C 
for the northern-most Luvisolic 
soil affected by permafrost (site 
64A4, Table 5). Mean Summer 
Soil Temperature (MSST) values 
vary from 12.6C in the south to 
1.1 C at the most northerly Luvi- 
sol soil (site 64A4). Soil thermal regime of the 
Chernozemic Black soils in the Grassland tran- 
sition region and the Dark Gray Chernozem soil 
in the Low Boreal region is cold (frigid, in U.S. 



Horizon 


Depth 
(cm) 


Colour 
(moist) 


Texture 


Structure 


Special Features 



Table 3. Morphological description of benchmark soils. 



L-H 

Ahl 

Ah2 

AC 

Ck 

2Ck 

L-H 

Ah 

Bra 

BC 

Ck 

2Ck 

L-H 

Ahe 

Bt 

BC 

2Ckl 

2Ck2 

L-H 
Ae 
Bt 
Ck 

LFH 

Ahe 

Ae 

Bt 

BC 

Ck 

LFH 

Ae 

Btnj 

Bt 

Cca 

Ckl 

Ck2 

LFH 

Ae 

AB 

Btnj 

Bt 

BC 

Ck 

LFH 

Ae 

AB 

Btnj 

Bt 

BC 

Ckl 

Ck2 



LFH 

Ae 

Bt 

Ckgj 

Ckz 



Site 62H13 Rego Black" (Typic Cryoboroll)" 4908'N, 9715'W 

3-0 - 

0-16 10YE2/1 C gr weak tonguing of 

16-36 10YR2/1 C gr A horizons through 

36-50 5Y2.5/1 C gr AC into Ck 

50-65 2.5Y4/2 SiC gr 

65-110 10YR4.5/3 SiCL gr 

Site 62H4 Orthic Black" (Typic Cryoboroll)" 49 Q 4TN, 9708'W 
2-0 

10YR2/1 

10YR3/1.5 

2.5Y3/2 

2.5Y4/2 



0-24 
24-36 
36-46 
46-80 
80-100 



10YR6/3 



C 
C 
C 

C-SiC 
SiCL 



sbk 
gr 
gr 
gr 



silt content increases 
below 1 m 



Site 6206 Orthic Dark Gray" (Argic. Cryoboroll)" 5144'N, 9846'W 
10-0 _ 

" gr 

gr 

gr 

gr 

gr 



0-6 

6-14 

14-22 

22-100 

100-120 



10YR2/2 

10YR3.5/2 

10YR3.5/2 

2.5Y6/2 

2.5Y6/2 



mixed clay and silty till 



C 
C 

C-SiC 

SiL 
C-SiL 

Site 63B2 Orthic Gray Luvisol" (Cryoboralf)" 52'39'N, 98 e 56'W 
10-0 .... 

0-8 10YE4.5/2 L pit 

8-20 10YR3/3 C sbk 

20-140 10YR5.5/2.5 SiC gr mixed clay and silty till 

Site 63J3 Orthic Gray Luvisol" (CryoboralfT 5412'N, 991 1'W 

10-0 .... 

0-5 10YR2/1.5 L gr some tonguing of A Ae 

5-13 10YR6/3 FSL pit horizon into B horizon 

13-32 10YR3/2 C sbk at 25 cm 

32-45 10YR4/3 C sbk 

45-100 10YR5/2 SiC gr 

Site 63J2 Orthic Gray Luvisol" (CryoboralQ" 54"42'N, 98;W 

12-0 .... 

0-7 10YR6/3 SiL gr 

7-12 10YR4/4 HC sbk 

12-23 10YR4/3 HC sbk 

23-30 10YR5/4 SiC gr 

30-105 10YR4/3 HC abk 

105-115 2.5Y7/2 HC abk 

Site 63O2 Othric Gray Luvisol" (CryoboralQ" 55'32'N, 9803'W 
6-0 5YR2.5/2 

C gr 

C sbk 

HC sbk 

HC gr 

HC gr 



weak columns with 
shallow cracking in 
upper solum 

silty varves at 1.5 m 



0-7 

7-10 

10-17 

17-30 

30-36 



10YR4/2 

10YR5/3 

10YR4/2.5 

10YR4/2 

10YR5/2 



shallow cracking in 
upper solum 



36-90 



10YR4.5/3 HC 



Site 63P19 Solonetzic Gray Luvisol" (CryoboralQ" 5555'N, 97'42'W 



10-0 

0-10 

10-14 

14-24 

24-50 

50-60 

60-100 

100-120 



10YR4/3 
10YR5/3 
10YR4/3 
10YR3/4 
10YR4/3 
10YK3/3 
7.5YR5/4 



SiCL 

SiC 

SiC 

HC 

SiC 

HC 

HC 



pit 
pit 
col 
gr 

gr 
mss 



pronounced vertical 
cracking in upper solum 



Site 64A4 Orthic Gray Luvisol" (CryoboralfT 5617'N, 9603'W 
cryic phase 

25-0 .... 

0-7 10YR3.5/2 SiC gr cracking absent 

7-18 7.5YR3/2 C gr cracking absent 

18-28 10YR4.5/4 C mss weakly cryoturbated 

28-50 10YR6/3 SiC mss vein ice, cryoturbed 



Texture: CrClay; HC:Heavy Clay; SiC:Silty Clay; SiCL:Silty Clay Loam; SilrSilt Loam 

Structure: plt:platy; abkiangular blocky; sbk:subangular blocky; gngranular; mss:massive; 

col:columnar 

*Soil classification according to System of Soil Classification for Canada, (Agriculture Canada 

Expert Committee on Soil Survey, 1987). 

**Soil classification according to U.S. Soil Taxonomy (Soil Survey Staff, 1975). 



Soil Taxonomy criteria for soils with LFH hori- 
zons). Thermal regimes characterizing the Luvi- 
solic soils in the Mid and High Boreal region 
vary from cold (cryic, in Soil Taxonomy) at sites 



128 



SIXTH iNTERNATIONAl Soil CLASSIFICATION WORKSHOP 





January soil temperatures are above 0C al 
50 cm throughout the Black Chernozemic clays 
(sites 62H13 and 62H4) and fall below 0*C ir 
the Dark Gray (site 62O6) and Luvisohc soils 
(Table 5). Minimum temperature in January al 
150 cm remains above 0C at all sites except ir 
the most northerly Luvisol (64A4). This soil has 
permafrost at 1.5 m and has a very cold (per 
gelic) thermal regime. The temperature of the 
permafrost at this latitude of northern Mani 
toba remains relatively constant at only a fev\ 
tenths of a degree below 0C. The maximun 
thickness of permafrost at this latitude has beer 
reported to be about 15 m (Brown, 1970), but is 
much thinner at the benchmark site. 

Latitudinal Gradient 
Figure 3 shows temperature at 50 cm and the 
mean annual air temperature along the 
transect. The regression equations for MAS! 
AND MSST at the 50 cm depth and the mear 
annual air temperature along the transect shov 
that the rate of temperature change per degret 
of latitude ranges from a minimum of 0.9C fo] 
MAST to a maximum of 1.5C for MSST. Thesi 
estimated rates of decrease in MAST and MSS1 
follow the northward trend of mean annual aii 


Table 4. Physical and chemical characteristics of soils 
selected to closely approximate soil conditions in the 
south-north transect 


Horizon Depth Total Silt Clay pH CaCO s Org. C.E.C. 
(cm) Sand Equiv. C m.e. 
% % % % % /lOOg 


1. Orthic Black (Typic Cryoboroll) 
Apl 0-9 30 27 43 6.7 - 4.5 46 
Ap2 9-24 22 32 46 6.7 - 4.4 49 
BM 24-42 12 26 62 7.2 - 1.9 48 
Bmk 42-65 9 29 62 7.6 8.8 1.1 43 
BC 65-78 7 31 62 7.7 14.1 1.0 41 
Ckl 78-95 4 31 65 7.8 10.6 - 40 
Ck2 95-110 1 32 67 7.8 12.8 - 39 
2. Orthic Dark Gray (Argic Cryoboroll) 
LH 4-0 6.1 2.2 29.7 
Ahe 0-5 19 26 55 6.1 - 3.6 39 
Bt 5-20 15 18 67 7.1 1.6 1.4 37 
BC 20-41 15 20 65 7.9 15.3 0.8 24 
Ckl 41-76 4 15 81 8.1 16.4 0.1 22 
Ck2 76-91 2 8 90 8.1 4.2 - 24 
3. Solonetzic Gray Luvisol-south Cryoboralf) 
LF 5.0 ... 5.0 - 39.8 121 
Ae 0-5 9 30 61 5.1 - 2.6 40 
AB 5-10 6 29 65 4.7 - 1.3 38 
Btnj 10-25 4 21 75 5.0 - 0.8 40 
Bt 25-45 1 14 85 6.1 - 0.8 46 
BC 45-60 11 89 7.5 6.8 0.4 34 
Ckl 60-100 1 12 87 7.6 9.2 - 32 
Ck2 100-130 1 21 78 7.5 6.0 - 28 
4. Solonetzic Gray Luvisol-north (Cryoboralf) 
LH 2-0 - - 5.9 - 32.4 87 
Ae 0-6 3 39 58 5.5 - 4.0 38 
AB 6-16 1 38 61 5.6 - 2.5 31 
Btnj 16-34 1 46 53 5.9 - 1.0 35 
Bt 34-50 54 46 7.0 - 0.5 34 
Ckl 50-98 40 60 7.6 8.3 - 31 
Ck2 98-138 31 69 7.7 7.7 - 31 


63B2, 63J3, and 63J2 to very cold or cryic at 


Table 6. Mean soil temperatures at benclxmark sites. 


sites 63O2 and 63P19. A very cold (pergelic in 

.Q/07/ TV/TO 770 777V f*T*itpTipO tTiPTmpil VPCHTYIP nppnvci 


Site No. Depth Mean Soil Temperature" Mean Soil Temperature" 
(cm) Annual Summer n January September 


at 150 cm at site 64A4, where the Luvisol soil 
contains permafrost. 
The difference between MAST and MSST 
decreases with increasing latitude (Figure 3). 
Differences at 50 cm are greatest in the Black 
soils in the south (5.5C at site 62H13 and 
6.0C at site 62H4) and least in the northern 
Luvisols (0.9C at site 64A4) affected by the 
permafrost. These differences are somewhat 
smaller at the 150 cm depth, ranging from 1.4 
to 2.6C in the 4 southern sites to no difference 
in northern soils affected by presence of per- 
mafrost (Table 5). 
As mean annual air temperature decreases, 
the difference between soil and air tempera- 
ture tends to increase (Smith et al., 1964). 
MAST is generally about 3 to 4C warmer than 
the corresponding air temperature at latitude 
49N, increasing to about 5C warmer than 
the air temperature north of latitude 55 N 
(Figure 3). At these latitudes, summer soil 
temperatures are appreciably lower than the 
air temperature (Smith et al., 1964). In this 
study, the MSST at all sites (Table 5) is lower 
than summer air temperature recorded at 
nearby climatological stations (Table 1). 


62H13 20 5.7 13.3 23 -1.7 12.5 
4908'N 50 6.0 11.5 0.3 12.2 
97'15'W 100 6.0 9.0 2.4 11.2 
150 6.0 7.4 3.9 10.1 
62H4 20 4.1 13.1 24 - 11.5 
49'47'N 50 6.6 12.6 0.7 13.6 
97'08'W 100 6.2 9.5 2.3 12.2 
150 6.3 7.9 4.5 11.4 
6206 20 4.8 10.9 19 -1.7 10.8 
5144'N 50 4.6 9.5 -0.3 10.4 
9846'W 100 4.3 7.6 1.2 9.4 
150 4.6 6.9 2.7 9.5 
63B2 20 3.8 9.1 19 -2.0 8.8 
5239'N 50 3.8 7.8 -0.3 8.6 
9856'W 100 4.6 8.5 0.8 8.0 
150 4.2 6.8 1.7 7.4 
63J3 20 2.7 7.3 18 -2.3 8.2 
5412'N 50 2.3 5.8 -0.9 7.4 
99'11'W 100 2.8 5.0 O.7 7.5 
150 2.7 4.1 1.4 6.6 
63J2 20 2.3 7.2 18 -3.3 7.9 
5442'N 50 2.1 5.0 -1.3 7.6 
98'58'W 100 1.8 3.0 O.3 5.9 
150 1.7 2.2 O.9 4.9 
6202 20 1.6 5.1 29 -2.4 7.0 
55'32'N 50 1.1 3.2 -1.3 5.5 
9803'W 100 1.3 2.4 -O.2 4.9 
150 -0.7 0.9 -O.7""" 3.2 
63P19 20 1.1 5.0 40 -2.8 6.3 
55-55^ 50 1.1 3.4 -O.7 5.3 
97'42'W 100 1.1 1.8 0.4 3.8 
150 0.9 ,1.0 O.6 2.3 
^* T 20 0.6 3.7 15 -3.4 5.1 
56'17'N 50 0.2 l.l -1.0 19 
96"09'W 100 0.1 0.1 -0.0 <X5 
150 -0.1 -0.1 -0.1 -0.0 


"Mean soil temperature calculated from thermistor observations, 1986-1989. 
. , _ n=number of observations at each site. 
Mean January and September soil temperature 1986-89, 3 observations. 
January temperature derived from one observation due to sensor failure. 



MILLS, EILERS, AND VELDHUIS: THERMAL REGIME AND MORPHOLOGY OP CLAY SOILS IN MANITOBA, CANADA 



129 



temperature which de- 
creases at a rate of about 1C 
per degree of latitude. 

Vertical Thermal Regime 

Vertical gradients in soil 
temperature vary with sea- 
son and latitude. The soil 
thermal regimes for January 
and September 1988 are 
plotted as a function of depth 
for each site in the transect 
(Figure 4). These soil tem- 
perature gradients are in- 
dicative of the maximum 
seasonal variation at each 
site. As would be expected, 
the vertical temperature gra- 
dient to the surface is nega- 
tive in January and positive 
in September. The mean 
annual soil thermal gradi- 
ents (1986-1989 data) also 
plotted for each site in the 
transect are nearly isother- 
mal and shift toward 0C 
with increasing latitude. 

The difference between 
September and January 
temperature decreases with 
increasing latitude as well as 
with increasing depth in the 
soil. Greatest seasonal tem- 
perature range in the upper 
50 cm occurs in the Black 
soils at the two southern- 
most sites (62H13 and 62H4) 
and decreases only slightly 
to the north along the 
transect. The range in sea- 
sonal temperature variation 
below 50 cm decreases more 
rapidly with increasing lati- 
tude. 

The seasonal temperature 
difference is least at the 
northern-most Gray Luvisol 
(site 64A4). This soil has permafrost at lower 
depths in the profile and in some summers re- 
mains frozen between 50 and 100 cm. The tem- 
perature of permafrost at this latitude is only 
slightly below freezing (Brown 1970) and is vir- 
tually isothermal with depth. The occurrence of 
the permafrost acts as a heat sink and provides 
a buffer against any large variation in soil tem- 
perature near the frost table. 



+ Mean Annual Soil Temperature (50cm.) 
O Mean Summer Soil Temperature (50cm.) 

Mean Annual Air Temperature 
62H4 Benchmark Site 




49 50 



52 54 56 58 

LATITUDE , DEGREES NORTH 



Figure 3. Scatter diagram and fitted lines of mean soil temperature at 50 cm 
and mean air temperature at 1.2 m. 



Summary 

Clay soils cover some 7 000 kha in Mani- 
toba between latitude 49N and 58'N. Soil de- 
velopment and soil thermal regime were exam- 
ined in relation to climate and vegetation at 
nine benchmark sites currently under forest in a 
south-north transect extending from latitude 
49N to 56'17'N. Warmer climate and subhu- 



130 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



TEMPERATURE,C TEMPERATURE/C 

n -6 -4 -2 2 4 6 8 10 12 14 16 16 .,.--4-20 2 4 6 8 10 12 14 16 18 






n m"T "r 


-T i i i [ T" r T r i i i i i i i i i T 


in 




10 


\ 


/ 


9O 


\ 7 


20 

K(\ 


\ 
\ 




i 


f 


K.f) 


\ 1 
\ 


i 
i 


O_ 

r" 

h- 
O. 

CtlfVt 


\ 


V 
\ 


/ 

/ 
/ 
/ 
i 
t 


E 5 
o 

r 

V- 
CL 
Ul 

^inn 


\ 
\ 
\ 

\ 


t 
f 
i 
/ 
i 
t 


ISO 




\ 

\ 
\ 

\ 


i 
t 62H13 
/ La1.4908'N 
i Elev. 234m. 


i^n 


\ 

1 


i 63J2 
. / La1.5442'N 
* / Elev. 225m. 


o 


TEMPERATURE/C 
-6 -4 -2 2 4 6 8 10 12 14 16 18 n 


TEMPERATURE *C 
-6 -4 -2 2 4 6 8 IO 12 4 >6 18 


10 


111111 








i 


on 


V 


/ ; 


on 


X 

- S 


/ * 


*5O 


\ 
\ 


\ 1 


6 50 

O 

r" 

0. 

^100 

150 


10 
20 

50 
E 

i" 
i- 

0. 

u 
IOO 

150 


10 
20 

50 

X 
K 
CL 

Q , no 


\ 
\ 
\ 


/ 
/ 
/ 

/ 


s. 

I 

h- 
Q. 

S IOO 

150 


10 
20 

50 
o^ 

X 

t 

CL 

LJ 

ioo 

150 


10 
20 

50 
o 

i" 

CL 

UJ 

!00 

150 


10 
20 

E 5 

O 

r" 
f- 

Q. 

^100 
150 


\ 

\ 


\ 
\ 
\ 
\ 
\ 
\ 


/ 
/ 

/ 
/ 
/ 
/ 

i / 


\ 
\ 
\ 
\ 

I 

V 


i 

i 
i 
j 





' t 

t 62H4 
/ Lot.4947'N 

/ Elev. 227m. 
/ 


f ! 6302 
/ Lai. 5532 N 
/ Elev. 220m. 

/ 


TEMPERATURE,C 
-6 -4 -2 2 4 6 8 10 12 14 16 18 


TEMPERATURE ,C 
-6 -4 -2 2 4 6 8 10 12 14 16 18 


\ - 


1 F T T I f I T I 1 T 1 | 1 I T T 


i i i i r 


i i i i i i i i i i > . i ( 




v 


N, 

\ 
\ 


/ 


\ 
\ 

\ 




i 
i 
/ 
/ 




\ 
\ 

\ 


/ 

/ 

/ 

/ 


\ 
\ 
\ 

^ 


/ 

/ 
/ 
/ 
t 
i 




\ 

\ 
\ 

\ 


6206 
Lat.5l44N 
Elev. 246m. 




i 

1 63P19 
/ Lot. 5555'N 

/ Eiev.210m. 
i 


TEMPERATURE ,C 
-6 -4 -2 2 4 6 8 10 12 14 16 18 


-6 -4 -2 


TEMPERATURE ,C 
2 4 6 8 10 12 14 16 18 


i r i i i i 


i i i i i i i r r r 


111111 


i i i i i i i i i i i i < i i i 




/ 




/ 


^ ^ 


\ ^" 




\ 
\ 
\ 




\ 
\ 
\ 


\ 
\ 


) / 






\ 
\ 

V 


1 

J 

; 


^ 

\ 

\ 
\ 


i 

i 
i 
t 






\ 

\ 


/ 63 B2 

Lot.5239'N 
i Elev. 248m. 


icn 




64A4 
Lat.56l7'N 
Elev. 180m. 




TEMPERATURE C 
-6 -4 -2 Z 4 6 8 10 12 14 16 IB 


LEGEND 
MAST 
JST- JANUARY ,1988 
XST-SEPTEMBERJ988 








V 




/ 




\ 
\ 
\ 




/ 

/ 

/ 




\ 
\ 
\ 


\ 








I 

\ 
\ 


/ 63J3 

/ Lo1.54*l2'N 
Elev. 237m. 




Figure 4. Vertical thermal gradients along south-north transect. 



MILLS, EILERS, AND VELDHUIS: THERMAL REGIME AND MORPHOLOGY OP CLAY SOILS IN MANITOBA, CANADA 



131 



mid conditions in the south result in grassland 
and grassland-forest transition vegetation char- 
acterized by Black and Dark Gray Chernozem 
soils. Cooler, more humid conditions associated 
with the Boreal forest extend northward from 
the grasslands to the southern edge of the 
Subarctic forest. Soil development in this region 
results in Gray Luvisol soils; the more severe 
climate associated with the northern-most Luvi- 
sols results in the occurrence of permafrost be- 
tween 1 and 2 m and cryoturbation of the 
ground surface. Although vertic soil features 
were not noted in the soils of the transect, they 
have been observed in similar clayey materials 
in the regions covered by this study. 

The thermal regimes of clay soils vary from 
cold (MAST 2-8'C, MSST 8-15C) for Black and 
Dark Gray Chernozems to cold and very cold 
(MAST -7 to 2C, MSST 5 to 8C) for the Luvi- 
sols. The most northerly clay soils have perma- 
frost within 1.5m, resulting in a very cold ther- 
mal regime at this depth. The calculated rate of 
decrease in MAST (0.9C/degree of latitude) and 
MSST (1.5C/degree of latitude) shows a north- 
ward trend following the mean annual air tem- 
perature, which decreases at a rate of about 
1.0C/degree of latitude. 

Based on clay content and mineralogy, the 
soils in the transect have the potential to de- 
velop vertic properties (continuous cracking and 
slickensides). Vertic properties are common in 
Black and Dark Gray soils which are subject to 
relatively long periods of drying to depths be- 
tween 1 and 2 m. These soils have a frigid ther- 
mal regime according to U.S. Soil Taxonomy cri- 
teria and could be classified as Vertisols if the 
temperature limit is altered to include colder 
soils. 

Although vertic soil properties have been ob- 
served in warmer Luvisols, they are not com- 
mon in clayey Luvisols and Cryosols with colder 
cryic and pergelic thermal regimes. The colder 
thermal regime and the occurrence of perma- 
frost in the most northern Luvisols combine 
with a significantly higher micaceous clay min- 
eralogy, more humid soil moisture regimes, and 
cryoturbation effects to reduce the potential for 
development of vertic soil features. As a result, 
vertic properties are not sufficiently common or 
well developed in northern Luvisol and Cryosol 
soils in Manitoba to warrant classification in the 
proposed Cryert suborder. 



Acknowledgements 

The authors acknowledge the assistance and 
cooperation of the following people: Mr. C. Aglu- 
gub and Mr. N. Lindberg for data compilation 
and analysis; Mr. J. Griffiths for technical assis- 
tance in preparation of graphics; Mrs. D. Sand- 
berg for typing the manuscript; and staff of the 
Department of Soil Science, University of Mani- 
toba, and the Canada-Manitoba Soil Survey for 
consultation and review of the manuscript 

References 

Acton, D.F., G. Coen and R.E. Smith, 1989. Distribution 
and properties of clay soils of subarid to subhumid re- 
gions of the Interior Plains in western Canada. In Pro- 
ceedings of the Sixth International Soil Correlation 
Meeting (ISCOM VI) Saskatchewan, Canada, Mon- 
tana, Idaho and Wyoming, U.S.A. In Press. 

Agriculture Canada Expert Committee on Soil Sur- 
vey. 1987. The Canadian System of soil classification, 
2nd ed. Agric. Can. Publ. 1646, 164 pp. 

Atmospheric Environment Service. 1980. Canadian Cli- 
mate Normals. 1957-1980. Vol. 2, Temperature; Vol. 3, 
Precipitation; Vol 6, Frost. 

Bostock, H.S. 1970. Physiographic subdivisions of Can- 
ada, pp. 9-30 in Geology and Economic Minerals of 
Canada, Dept. of Energy, Mines and Resources Can- 
ada. Geological Survey of Canada, Economic Geology 
Report No. 1. 

Brown, R.J.E. 1970. Permafrost in Canada. University of 
Toronto Press. 234 pp. 

Brown, R.J.E. and W.O. Kupsch, 1974. Permafrost Termi- 
nology. National Research Council of Canada, Techni- 
cal Memorandum No. Ill, 62 pp. 

Canada-Manitoba Soil Survey. 1989. Soil Landscapes of 
Canada-Manitoba, Land Resource Research Centre, 
Res. Br., Agriculture Canada, Ottawa, Ontario. Agric. 
Can. Publ. 5242/B (LRRC Cont. No. 87-16)Report and 
map. 

Ecoregion Working Group. 1989. Ecoclimatic regions of 
Canada, first approximation. Ecoregion Working 
Group of the Canada Committee on Ecological Land 
Classification, Ecological Land Classification Series 23. 
Sustainable Development Branch, Canadian Wildlife 
Service, Conservation and Protection, Environment 
Canada, Ottawa, Canada. 119 pp. and map. 

Madden, J.G., personal communication. Clay mineralogy 
of some Manitoba parent materials. Canada-Manitoba 
Soil Survey, 362 Ellis Building, University of Manitoba, 
Winnipeg, Manitoba. R3T 2N2. 

Manitoba Ecoclimatic Region Working Group. 1985. 
Ecoclimatic Regions of Manitoba, Canada-Manitoba 
Soil Survey, 362 Ellis Building, University of Manitoba, 
Winnipeg, Manitoba. R3T 2N2. 

Prest, V.K. 1970. Quaternary Geology of Canada, pp. 77- 
764 in Geology and Economic Minerals of Canada, 
Dept. of Energy, Mines and Resources, Canada. Geolo- 
giucal Survey of Canada, Economic Geology Report 
No.l. 



132 SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 

Reimer, A. and C.F. Shaykewich. 1980. Estimation of Soil Survey Staff. 1975. Soil Taxonomy. Soil Conservation 
Manitoba soil temperatures from atmospheric mete- Service. U.S. Department of Agriculture. Agricultureal 

orological measurements. Can. J. Soil Sci. 60:299-309. Handbook No. 436. Washington B.C., 754 pp. 

Smith, G. D., L.H. Robinson, and D. Swanson, 1964. Soil Wilding, L.P. and D. Tessier. 1988. Genesis of Vertisols: 
temperature regimes-their characteristics and pre- Shrink-Swell Phenomena, pp.55. In L.P. Wilding and 

dictability. SCS-TP-144 Soil Conserv. Serv., U.S. Dep. R. Puentes (eds). Vertisols: Their Distribution, proper- 

Agr., Washington, D.C. ties classification and management, Texas A & M 

Press. College Station. 



Periglaclal Features as Sources of Variability 
in Wyoming Aridisols 

L.C. Munn 1 



Abstract 

Periglacial features including ice wedge casts, ground wedges and low- 
relief soil mounds are common in the basins of Wyoming. These features 
represent former permafrost environments, during glacial episodes. Soil 
formed on this patterned ground are younger and show different morphol- 
ogy and chemistry compared to older, undisturbed soils on the same land- 
scape. The features thus contribute to soil variability; this variability influ- 
ences present-day vegetation distribution and yield. 



Introduction 

The high, nonglaciated basins of Wyo- 
ming contain many relicts of previous perigla- 
cial environments casts of ice wedges, ground 
wedges, and mounded topography (Mears, 1987; 
Nisson, 1985; Spackman and Munn, 1984). In 
the surrounding mountains, glacial tills and 
sorted patterned ground features attest to the 
extent of the glacial climates (Richmond and 
Fullerton, 1986). The purpose of this paper is to 
discuss the contribution of these relict features 
to the spatial variability in morphology and 
chemistry on Aridisol landscapes. 

Methods 

Data for this paper are taken largely from 
previously published work by the author, from 
Wyoming soil survey reports, or from data col- 
lected by the National Soil Survey Laboratory 
(1988). Methods of data collection and labora- 
tory analyses are given in the documents cited 
and generally represent standard methods 
specified in Soil Taxonomy (Soil Survey Staff, 
1975) and the Soil Survey Manual (Soil Survey 
Staff, 1981). 

Periglacial Features: Mode of 
Formation 

Black (1976) details the formation of both ice 
wedges and ground wedges in modern perma- 
frost environments. Mears (1981, 1987) has 
documented multiple occurrances of both fea- 
tures in Wyoming's cold basins. Both types be- 
gin with thermal cracking-contraction cracks of 
1 to 5 mm wide that form during the cold period 
of the annual climatic cycle. In moist perma- 
frost environments, water from the thawing ac- 



^epartment of Plant, Soil and Insect Sciences, Univer- 
sity of Wyoming, Laramie, WY 82071. 



tive layer fills the cracks in the spring. In dry 
permafrost environments, blowing soil particles 
fill the cracks during the cold period. In both 
cases, the initial filled cracks in the permafrost 
layer act as a "memory" and subsequent cracks 
open each successive season in nearly the same 
location until ice or ground wedges of up to sev- 
eral meters in cross section have built up (Fig- 
ure 1). In three dimensions, the cracks form 
polygons with diameters of 1 to 40 m. 

Washburn (1980) proposed that a mean an- 
nual air temperature of -5C was required to 
form wedges. If this threshold temperature is 
valid, then mean annual temperature in Wyo- 
ming during the Pleistocene must have veen at 
least 10C (Laramie Basin) to 13C (Bighorn 
Basin) colder than the present climate. In addi- 
tion to a subzero mean annual temperature, a 
rapid drop in temperature, variously estimated 
at from 4"C to 10C (Washburn, 1980), is re- 
quired to initiate cracking of the frozen soil, 
sediment, or rock. The presence of frozen water 
is required for the formation of continuous 
cracks in loose soils or sediments. Walker 
(1987) describes a fossil animal assemblege for 
northwestern Wyoming that agrees with the 
concept of a permafrost tundra. 

Another feature, thought to be an indicator of 
a periglacial environment, is large (8-10 m di- 
ameter), low relief (0.5-1 m) mounds. These 
mounds are spaced regularly across large areas 
of Pleistocene-aged surfaces in Wyoming 
(Spackman and Munn, 1984). Spackman and 
Munn (1984) hypothesized that the mounds 
were formed by cryoturbation as a result of pres- 
sure engendered by water being trapped be- 
tween a downward freezing front at the top of 
the active layer and an upward freezing zone 
from the top of the permafrost. In coarse-tex- 
tured materials, water migration to the freezing 



133 



134 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



PERIGLACIAL 
ENVIRONMENT 



MODERN 
ENVIRONMENT 



Pre -Wisconsin 
Soil (Host) 




Permafrost 




Late-Wisconsin, 
Early Hoiocene Soil 



Pre- Wisconsin 
Paleosol 



Ice Wedge 



Wind Transported Material 



Ice Wedge Cast 



Pre- Wisconsin 
Soil (Host) 




Permafrost 




Late- Wisconsin, 
Early Hoiocene Soil 



Pre-Wisconsin 
Paleosoi 



Sand Wedqe ReRc < Sand ) Wedge 

Figure 1* Schematic representation of mode of formation of ice and sand (ground) wedges (Munn, 1987). 



fronts might be slow enough for this to occur 
during rapid freezing. 

Another hypothesis is that the mounds repre- 
sent "mudboils" comparable to those in the mod- 
ern arctic on materials with low plastic and liq- 
uid limits (Shilts, 1978). Mudboils are a form of 
nonsorted circles (Washburn, 1980) that form 
during the annual thawing cycle of the active 
layer. Positive pore water pressure is developed 
because the permafrost layer prevents drainage 
of meltwater and supersaturation results as ice 
lenses melt. Pressure built up under a confining 
surface layer, or carapace (in this case, the Hap- 
largid Bt horizon), is released through an 
upwelling of a soil and water slurry. This pres- 
sure release mechanism accounts for the statis- 
tically regular distribution of the mounds 
(Spackman and Munn, 1984). 



Discussion 

Both types of wedge features and the cryotur- 
bation mounds (Figure 2) are important sources 
of variability on Pleistocene-age landscapes. 
Table I summarizes variations in textural prop- 
erties, morphology, and carbonate accumula- 
tions for several sites in Wyoming. Note that 
the wedge or mound soils are typically in a dif- 
ferent (coarser) textural family than the host 
soil. The mounded landscapes are typically 
Camborthid (mound)/Haplargid (intermound) 
complexes. In the Laramie Basin, the mound 
soil provides a sandy range site, while the inter- 
mound soil is in a loamy range site. Depth of 
penetration of precipitation is typically greater 
in the coarser-textured wedge or mound soils. 



Table 1: Comparison of Texture and Related Properties in Former-Permafrost Landscapes, Wyoming 


Feature Percent of 
Landscape 


Soil Family 


Maximum 
Percentage 
Clay CaCO, 


Gravel in 
B Horizon 
% by volume 


Electrical Conductivity 
at 100 cm depth 
(ds-m- 1 ) 


Carbonate 
Accumulation 1 
(kg-nr 8 ) 


Ice Wedge Casts 30 (Wedge) 
Rawlins Area 
(Munn, 1987) 


(W) a Typic Haplargids, coarse-loamy, mixed, frigid. 
(H) Typic Haplargids, fine-loamy, mixed, frigid. 


16 
31 


9 
45 


5 
25 


1.9 
>15.0 


92 
250 


Ground Wedges, 70 (Wedge) 
Laramie Area 
(Munn and Spackman, in review) 


(W) Borollic Natrargids, fine-loamy, mixed. 
(H) Borollic Natrargids, fine, mixed. 


33 
39 


7 
10 


<1 
3-5 


8.5 
12.3 


171 
419 


Ice Wedge Casts, 35 (Wedge) 
Laramie Area 
(Spackman and Munn, 1984) 


(W) Borollic Haplargids, coarse-loamy, mixed. 
(H) Borollic Haplargids, fine-loamy, mixed. 


15 
33 


11 
25 


6 
20 


0.6 
0.7 


113 
250 


Cryoturbation 35 (Mound) 
Mound, 
(Spackman, 1982) 


(M) Borollic Camborthids, coarse-loamy, mixed. 
(H) Borollic Haplargids, fine-loamy, mixed. 


10 
33 


7 
25 


32 
20 


10.9 
0.7 


169* 
250 


Ground Wedge, 5-10 (Wedge) 
Kernmerer Area 
(National Soil Survey Laboratory, 1988) 


(W) Typic Haplargid, fine-loamy, mixed, frigid. 
(H) Borollic Haplargid, fine, mixed. 


24 
47 


18 
29 


TR 

1 


3.3 
4.3 


247 
373 


Carbonate accumulation to varying depths to include entire solum. 
*W - Wedge, H - Host, M - Mound. 
Includes primary carbonate from parent material (ruptured host Bk horizons). 



MUNN: PERIGLACIAL FEATURES AS SOURCES OP VARIABILITY IN WYOMING ARIDISOLS 



135 









w - \ r/i 

f :!%r! 
. .^''-^ufe 
1- , . ''.*> 








Figure 2. Composite of (a) ground wedges near Laramie (Munn and Spackman, in review), (b) ice wedge cast 
near Rawlins (Munn, 1987), (c) mounded topography near Laramie (Spackman and Munn, 1984), and (d) 
ground wedge showing vertical fabric (Munn and Spackman, in review). Both wedge features show contor- 
tion of the Bk horizon of the wedge host soil adjacent to the wedge. 



Plant Response 

Catt (1986) describes crop response to ice 
wedge casts in Europe. In England, yields of 
both barley and wheat were approximately 
twices as high on fine-textured wedge fills as on 
the coarse-textured polygon interiors (e.g. 5.06 
tonne ha" 1 vs. 2.11 tonne ha' 1 for wheat). When 
the wedge fill material was coarser-textured 
than the material of the wedge host, grain yields 
were lower on the fills at sites in England, Den- 
mark, and Germany. 



In arid climates, higher biomass production 
typically occurs on coarse-textured soils (and 
wedge fills) where storage of plant-available 
water is greater. At the Kemmerer site, plant 
available water holding capacity in the wedge 
fill is .31 cm cm' 1 , compared to .20 cm cm' 1 in the 
host soil (Natinal Soil Survey Laboratory, 
1988). Water held at fifteen bars is 9% in the 
wedge, in the host soil it is 15%. Where the 
wedges penetrate dense Bk horizons, as at this 
site, improved grass growth over the wedge fill 



136 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



SOIL PROFILE (SOUTH-FACING WALL) 




30IL PROFILE (SOUTH-FACING WALL) 




Figure 3. Cross section of mounded landscape near Laramie (Spackman, 1982; Spackman and Munn, 1984). 
The mounds (Camborthids) occupied approximately 35% of the transect. Fossil ice wedge casts were pres- 
ent at several places along the transect, e.g., 11 to 12 m and at 13 m and 18 m. Intermound soils are Haplar- 
gids with calcic horizons. 



often results in a clear outline of the polygonal 
wedge pattern. 

Soil Morphology and Genesis 

Soil mounds produced by cryoturbation show 
a definite admixture of materials from pre-exist- 
ing soil horizons, including carbonates and sol- 
uble salts. In contrast, material in ice wedge 
casts originated in the active layer (annual thaw 
zone above the permafrost). Material from this 
layer slumped into the depression formed as the 
ice wedge melted with the final thawing of the 
permafrost. For ground wedges, the wedge fill 
originated as aeolian material locally derived 
from the surface of the active layer. 

Soils developed on the mounds contain diffuse 
"primary" carbonates inherited from the parent 
material, in addition to secondary carbonates, 
which occur as threads, rock-rinds, and nodules. 
Carbonates in ice wedge casts and in the ground 
wedges represent "secondary" carbonates. In 
both mound and wedge soils, the accumulations 
of clay, carbonate, and gypsum are less than in 
the host soils, as would be predicted by their 
younger age. Rates of accumulation of clay and 
carbonates seem to be very similar to those de- 
scribed for the Las Cruces, New Mexico area by 
Gile and Grossman (1979). 

The Laramie Basin ground wedges and the 
ground wedges at Kemmerer are interpreted as 
being of Bull Lake (Illinoian) age, while the 
other features are thought to be of Pinedale 
(Wisconsin) age. Rodents (Cynomys spp, Sper- 
mophilus richardsmi) prefer the mounds for 



burrow sites, and excavations of the mounds 
reveal active burrows as well as krotovinas. 
This activity tends to retard leaching of carbon- 
ates and salts from the mounds, despite coarser 
texture and gravel content (Table 1). Figure 3 
illustrates the complexity produced on a land- 
scape with wedge features, mounds, and rodent 
activity in the Laramie area (Spackman, 1982). 
The mound-forming event disrupted the original 
loamy over gravelly sand stratification in the 
alluvial parent material, mixing gravel through- 
out the profile of the mound soil. 

Conclusions 

The ubiquitousness of the wedges and 
mounds in Wyoming's high basins (Mears, 1987; 
Nisson, 1985) and the marked differences in soil 
morphology and chemistry between host and 
wedge or mound soils make their features im- 
portant contributors to the complexity of Wyo- 
ming's desert landscapes. Most stable, older- 
than-Holocene surfaces in western Wyoming's 
basins have polygenetic soils related to former 
permafrost climates. 

Literature Cited 

Black, R.F. 1976. Periglacial features indicative of perma- 
frost: ice and soil wedges. Quat. Res. 6:3-26. 

Catt, J.A. 1986. Effects of the Quaternary on soils of pres- 
ent land surfaces, p. 206-232. In Soils and Quaternary 
Geology. Clarendon, Oxford. 

Gile, L. H. and R. B. Grossman. 1979. The desert project 
soil monograph. USD A. U.S. Gov't. Print. Office, Wash- 
ington DC. 



MUNN: PERIGLACIAL FEATURES AS SOURCES OF VARIABILITY IN WYOMING ARIDISOLS 



137 



Hears, B., Jr. 1981. Periglacial wedges and the late Pleis- 
tocene environment of Wyoming's intermontane ba- 
sins. Quat. Research 15:171-198. 

Mears, B., Jr. 1987. Late Pleistocene periglacial wedge 
sites in Wyoming: an illustrated compendium. Memoir 
No. 3. The Geological Survey of Wyoming, Laramie. 

Munn, L.C. 1987. Soil genesis associated with periglacial 
ice wedge casts, southcentral Wyoming. Soil Sci. Soc. 
Am. J. 51:1000-1004. 

Munn, L.C. and L.K. Spackman. In review. Soil genesis 
associated with periglacial ground wedges, Laramie 
Basin, Wyoming. Quaternary Res. 

National Soil Survey Laboratory. 1988. Pedon 88P0858 
Tresano Varient. USD A- SCS, Lincoln, NE. 

Nisson, T.C. 1985. Field and laboratory studies of selected 
periglacial wedge-polygons in southern Wyoming. M.S. 
thesis. Univ. of Wyoming, Laramie. 

Richmond, G.M., and Fullerton, D.S. 1986. Summation of 
quaternary glaciations in the United States of America. 
Quaternary Science Reviews, Vol. 5. Pergamon, New 
York. 

Shilts, W.W. 1978. Nature and genesis of mudboils, cen- 
tral Keewatin, Canada. Can. J. Earth Sci. 15:1053- 
1068. 



Soil Survey Staff. 1975. Soil taxonomy. Agric. Handbk. 
No. 436. USDA. U.S. Govt. Print. Office, Washington 
DC. 

Soil Survey Staff. 1981. Soil survey manual. Chapter 4. 
Working Draft (430-V-SSm). U.S. Govt. Print. Office, 
Washington DC. 

Spackman, L.K. 1982. Genesis and morphology of soils 
associated with formation of Laramie Basin (Mima- 
like) mounds. M.S. thesis. Dept. of Plant Science. Univ. 
of Wyoming, Laramie. 

Spackman L.K. and L.C. Munn. 1984. Genesis and mor- 
phology of soils associated with formation of Laramie 
Basin (Mima-like) mounds in Wyoming. Soil Sci. Soc. 
Am. J. 48:1384-1392. 

Walker, D.N. 1987. Late Pleistocene/Holocene environ- 
mental changes in the northwestern plains of Wyo- 
ming: the mammalian record. Illinois State Museum 
Scientific Paper 22:334-392. 

Washburn, A.L. 1980. Geocryology: a survey of periglacial 
processes and environments. London/Halstead, New 
York. 



138 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



A Comparison of Land Use and Productivity of Clay and Loam Soils 
within the Interior Plains of Western Canada 

C. Onofrei 1 , J. Dumanski 2 , E.G. Eilers 1 , and RE. Smith 1 



Abstract 

Clay and loam soils constitute major land resources for crop production 
in western Canada. In the present study, two separate investigations 
within the Chernozemic soil zone (Brown, Dark Brown, and Black) were 
completed. The first was a comparative land use analysis for farms on clay 
and loam soils in Saskatchewan, based on data from Statistics Canada; the 
second was a crop production assessment for clay and loam soils in southern 
Manitoba, using a wheat yield modeling simulation technique. 

Spring wheat is the dominant crop grown in each soil zone. Results 
of the land use analysis showed that farms in the Brown zone operate mar- 
ginally more efficiently than in the other areas, with slightly better effi- 
ciency on clay than on loam soils. Estimates of efficiency were based on 
ratios of sales to expenses. 

The crop productivity analysis indicated that wheat yield frequency 
distributions varied not only between the two textural groups, but also 
within the same textural groups within a given soil zone. The results of this 
analysis also indicated that certain soil characteristics, agronomic land at- 
tributes, and weather elements strongly affect productivity in the interior 
plains of western Canada. Progress in yield prediction research can be 
made by focussing on these key controling factors. 



Introduction 

Production of grains and oilseeds is the main 
agricultural activity in the interior plains of 
western Canada. Generally these commodities 
are produced on land resources consisting of 
loam and clay soils. The term "land" as used in 
this paper includes elements of soil as well as 
elements of geology, the atmosphere, land man- 
agement, and land use (FAO, 1984). 

A system's efficiency is determined as the ra- 
tio of its performance to the costs involved. The 
general indicator used to measure the perform- 
ance of a crop-oriented agricultural system is 
annual yield. Due to the stochastic character of 
weather, however, annual crop yields are often 
highly variable, particularly these obtained 
within the Canadian Prairie region. 

In general terms, the climate of the southern 
Prairies is characterized by south to north gra- 
dients of energy and moisture; the temper ature 
decreases while the moisture increases with in- 
creasing latitude. However, crop yield variabil- 
ity on scales of space and time exhibits a more 
complicated and complex pattern than can be 
explained by simple correlation of yield with 
these two factors. 



: Land Resource Research Centre (LRRC), Research 
Branch, Agriculture Canada, University of Manitoba, 
Winnipeg, Manitoba. 

2 LRRC, Research Branch, Agriculture Canada, Ot- 
tawa, Ontario. 



The objectives of this study were: (I) to com- 
pare the general socialeconomic characteristics 
of farms that operate on clay and adjacent loam 
soils within the three major soil zones, and (2) to 
illustrate the variability in crop yield character- 
istic of clay and loam soils within the Black zone 
of southern Manitoba. These latter characteris- 
tics illustrate the relative levels of production 
risk between these two types of soil as calcu- 
lated by yield probability functions. 

Clay soils occupy a significant area in western 
Canada although they are less extensive than 
loam soils. About 5.6 million ha of clay soils are 
located in the southern Prairies in the Brown 
(1.2 million ha), Dark Brown (1.9 million ha), 
and Black (2.5 million ha) soil zones. These 
zonal Great Groups are distributed approxi- 
mately concentrically from the south-central 
portion of the region. Figure 1 schematically 
presents the location of study sites 

Data Used and Methods of Analysis 

Land Use Investigation 

For the land use investigation two adjacent 
polygons were selected from each of the three 
Chernozemic soil zones of southern Sas- 
katchewan. The polygons were selected from 
the 1:5 M Soils of Canada map (Clayton et aL, 
1977) and data for these polygons was obtained 



ONOFREI, DUMANSKI, EILERS, AND SMITH: A COMPARISON OF LAND USE AND PRODUCTIVITY IN WESTERN CANADA 139 



from the Land Potential Data 
Base (Kirkwood et aL, 1989). 

Data for the land use investi- 
gation were derived from the 
1981 Census of Agriculture. 
Methods employed were descrip- 
tive statistics and simple eco- 
nomic analysis on an "average" 
farm data set for each polygon 
(Huffman, 1988). The represen- 
tative data set was obtained by 
overlaying data from Census 
Enumeration Area (EA) maps 
on the Soils of Canada map. 

The arithmetic mean was con- 
sidered an appropriate statistic 
for describing the central ten- 
dency of physical variables and 
crop distribution characteristics 
such as farm size and crop ar- 
eas. Since some farms on the 
Prairies are mixed farms, the 
distribution of economic vari- 
ables can be highly skewed by 
extremely small or extremely 
large data. To avoid the effect of such data on 
the measure of location, the median was se- 
lected as the most appropriate statistic for de- 
scribing economic variables such as capital 
value, expenses, and sales. 

Crop Productivity Assessment 

Modeling simulation techniques were em- 
ployed in the crop yield investigation. The study 
focused on the combined effects of soil proper- 
ties, management, and climate/weather ele- 
ments on crop production. Soil series were used 
as the soil component in the model, but since a 
soil series may occur in different polygons (geo- 
graphical locations), crop yields obtained on 
similar soils in different locations could be quite 
different. 

The modeling procedure was based on a simu- 
lation technique developed for land evaluation 
(Onofrei, 1986). Spring wheat is the most im- 
portant commodity in the Prairie region, and it 
was used as the indicator crop. Since yield is 
controlled primarily by the weather pattern spe- 
cific to each growing season, PIXMOD, a physi- 
cal model, was run repeatedly for each polygon/ 
soil, using historic daily weather records for the 
period 1964 to 1983. Overall, this provided yield 
frequency distributions, which give insights 
into natural risk for crop production. 

PIXMOD differs from other simulation mod- 
els in the sense that it operates with readily 



SOIL ZONES 
Brown E:lVJ Dork Groy 



Dork Brown I 1 Gray Luvisol 
VTA Block S *? 



SASKATCHEWAN 

i 

CANADIAN SHIELD 




,J-*9- 



LjLoam O Clfl y 

Figure 1. Schematic representation of the location of sites included in the 
land use investigation (D O) and in the crop productivity assessment 

(0 ). 



available data sets (soil, weather, and manage- 
ment) from soil surveys, soil testing laborato- 
ries, and climatic stations. It uses potential 
wheat yield estimates as calculated by the FAO 
model (FAO, 1978; Dumanski and Stewart, 
1981). The potential wheat yield is considered a 
theoretical limit for photosynthesis, and it is 
used in this study as the agronomic ceiling. 

It should be noted that the term "potential" is 
used in modeling and in this study in a manner 
similar to the use of the potential evapotranspi- 
ration term in agrometeorology; i.e., the state of 
the system in a standard condition. The poten- 
tial yield is defined as the wheat yield that can 
be obtained on a given land area if all matter- 
energy forms required by the crop, with the ex- 
ception of photosynthetically active radiation, 
are maintained at optimum over the entire 
growing season. In real life, such conditions vir- 
tually never exist. However, based on potential 
yield and daily calculation of phenological crop 
development, soil moisture content, nitrate- 
nitrogen, and soil temperature, PIXMOD simu- 
lated the above-ground dry matter for a growing 
season, and this was converted into grain yield, 
using the harvest index approach. 

The productivity model was run for seven 
areas (polygons) in southern Manitoba (Table 
1), as identified on the Generalized Soil Land- 
scape map of Manitobal. These were selected to 
reflect difference in productivity between loam 



140 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Polygon Coordinate of 
No. 1 centroid polygon 

longitude latitude 



Soil series 

name 
CanSis code 2 



90 

75 

73 

125 

53 

1 



97'17' 

96*51' 

97"17' 

9800' 

lOO'OO' 

10035' 

ions 1 



4944' 
49*41' 
4905' 
4912' 
51 "08' 
50 '27' 
4903 



RIV 
OBO 
DCS 
RLD 
DPH 
NDL 
'RYS 



: Dencross, RLD = Reinland, DPH = Dauphin, NDL = 



Note: 

Polygon number from GSLM (1:1M); 

2 Soil Series name: RIV = Red River, OBO = Osborne, DCS : 

Nwdale, RYS = Ryerson; 

'Subgroup: GLRBL - Gleyed Rego Black (Chernozem), RHG - Rego Humic Gleysol (Gleysol), OBL - Orthic Black 

(Chernozem); 

4 U.S.A. Taxonomic equivalent: ACB - Aquic (Vertic) Cryoboroll, TCA - Typic Cryaquoll, AHB - Aquic Haploboroll, TCB - 

IVpic Cryoboroll, UHB - Udic Haploboroll. 



and clay soils in ad- 
jacent areas. Four 
polygons have 
dominantly clay 
soils, 89-Red River 
(89-RIV), 90- 

Osborne (90-OBO), 
75-Dencross (75- 
DCS) and 125-Dau- 
phin (125-DPH), 
and three polygons 
have dominantly 
loam soils, 91-Rein- 
land (73-RLD), 53- 
Newdale (53-NDL), 
and 1-Ryerson (1- 
RYS). All polygons are located in the Black soil 
zone. 

Three polygons with clay soils (89-RIV, 90- 
OBO, and 75-DCS) and one polygon with loam 
soil (73-RLD) occur in close proximity to each 
other and thus have similar weather conditions. 
Therefore, considering that management input 
was kept constant in each model run, the differ- 
ences between the yields in these polygons re- 
flect mainly the impact of different soil attrib- 
utes on productivity. The 125-DPH clay soil and 
the 53-NDL loam soil are located near the 
northern limit of the Black Chernozemic zone in 
Manitoba, while the 1-RYS loam soil is located 
in the southwest corner of Manitoba which is 
characteristically the driest area in the prov- 
ince. The Red River (89-RIV) and Dauphin 
(125-DPH) clay soils are closely related to each 
other, as are the Newdale (53-NDL) and Ryer- 
son (1-RYS) loam soils, in terms of their major 
soil characteristics and agronomic attributes. 
The differences between yields on these soils, 
therefore, reflect mainly the impact of regional 
weather on land productivity. 

Three categories of data were compiled for 
the crop production investigation, namely soil, 
daily weather, and management data. The soil 
data included: rooting depth, surface texture, % 
of coarse fragments (gravel), drainage class, in- 
filtration rate, water table depth, incoming run- 
off, shape and frequencies of unconnected mi- 
crodepressions, slope class of depressions, and 
diagnostic soil profile horizons. For each identi- 
fied master horizon, the centre point, bulk den- 
sity, particle size distribution (clay, silt, very 
fine sand, fine sand), organic carbon, volumetric 
field capacity, volumetric wilting point, and 
volumetric water content at seeding time also 
were considered in the model. All soil proper- 



Table 1. Location and general characteristics of soil series included 
in the crop productivity analysis. 



Profile 
texture 



Drainage 
class 



Water table 

depth 

(cm.) 



Classification 



Canada 3 



U.S.A. 4 



Clay 
Clay 
Clay 
Loam 
Clay 
Loam 
Loam 



Imperfect 

Poor 

Imperfect 

Imperfect 

Imperfect 

Well 

Well 



>150 
120 
>150 
>150 
>150 
>150 
>150 



GLRBL 

RHG 

GLRBL 

GLRBL 

GLRBL 

OBL 

OBL 



ACB 
TCA 
ACB 
AHB 
ACB 
TCB 
UHB 



ties/attributes were derived from the Canadian 
Soil Information System (CanSIS) and from dif- 
ferent soil survey reports. 

Weather data included: precipitation, maxi- 
mum air temperature, minim rim air tempera- 
ture, solar radiation at the top of the atmos- 
phere, and photoperiod. A weather file was cre- 
ated for each of the selected polygons, using the 
Thiessen weighting procedure (Williams and 
Hayhoe, 1982) to extrapolate from observed 
data recorded by the Atmospheric Environment 
Service (AES). 

Of the many management data which influ- 
ence crop yields, the following were considered 
the most relevant for modeling wheat produc- 
tion in the Prairie region: seeding date, soil ni- 
trate-nitrogen content at seeding time, nitrogen 
fertilizer applied (amount and date of applica- 
tion), harvest date, and the agronomic crop yield 
potential. The latter parameter, although not 
fully controlled by the management factors, is 
used as an indication of the agronomic ceiling 
for a given crop for given management inputs 
and technology (applied knowledge). 

The simulated yield series for each soil in 
each location (polygon) were analyzed statisti- 
cally, employing the following tests: 

a) the Quade test and the Friedman test 
(Conover, 1980) to determine if all soils 
have identical effect on yield 

b) the Shapiro- Wilk test of normality for each 
yield series 

c) transformation of nonnormally distributed 
data series to yield a family of normal dis- 
tributions 

d) calculation of location and scale parame- 
ters (ILL and a) for each distribution function. 

The details of these statistical analyses are 
presented in a more comprehensive land evalu- 
ation study (Onofrei, 1987). 



ONOFREI, DUMANSKI, EILERS, AND SMITH: A COMPARISON OP LAND USE AND PRODUCTIVITY IN WESTERN CANADA 141 



Results and Discussion 

The data presented in this study pertain to 
the dominant conditions in each polygon. Al- 
though local soil properties, weather, or man- 
agement can have significant impact in a lim- 
ited area, it is considered that these effects are 
small overall in comparison to the geographic 
location of the area. In the same vein, the data 
pertain to specific polygons, which, although 
suitable for comparative analysis, may not nec- 
essarily describe accurately a larger zone. 

Land Use Characteristics 

In general, farm size and the proportion of 
summerfallow land decrease on both clay and 
loam soils from the Brown to the Black soil zone 
(Table 2). Summerfallow accounts for about 
40% of the use of arable land on farms in the 
Brown zone and 37% on Dark Brown, but this 
decreases to about 16-20% in the Black zone. 
Along with decrease in farm size, there is a pro- 
gressive increase in absolute value of land and 
in the marketable production per hectare. 
These characteristics reflect the fact that large 
farms tend to concentrate in the Brown and 
Dark Brown zones where productivity per unit 
area is low, the risks in crop production are 
higher, and crop rotations that include fallow 
once in 2 or 3 years are standard farming prac- 
tices. However, with decrease in farm size and 
larger proportion of land used for continuous 
cropping, as in the Black soil zone, there is a 
progressive increase in total capital investment 
and operating expenses. 

Farms on clay soils tend to be smaller than 
those on loam soils in the Brown, but conversely 
they are larger in the Black zone. The propor- 
tion of area cultivated is 5-7% larger on clay 
soils in each zone. In general there are no major 
differences between the two textural groups in 
proportions of field crops grown, but clay soils 
are used more 
commonly for 
cereal crops. 
On loam soils 
the proportion 
of cereal crops 
decreases in 
favor of forage 
crops. 

A noticeable 
difference ex- 
ists, in terms of 
economic char- 
acteristics, be- 



tween farms on clay soils and those on loam soils 
in all zones. For example, total capital invest- 
ment on clay vs. loam soils is 23% higher on 
Brown soils, 56% higher on the Dark Brown 
soils, and 21% higher on the Black soils. Also, 
the operating expenses on clay soils are 11-28% 
greater than on the loam soils. Overall, farms 
on clay soils outperform these on loam soils in 
terms of total sales per unit area and provide 
better returns to investment (Huffman, 1988). 
Based on 1981 data, the value of production per 
cost of inputs ranged from 2.82 on clay soils in 
the Brown to 1.99 on clay soils in the Black. 

Crop Productivity 

Running PIXMOD for all polygons considered 
in the study, using 20 years of historic weather 
data, resulted in seven series of annual wheat 
yield values. Assuming normal distribution of 
wheat yield on every soil/polygon included in the 
analysis, and pooling the yields on clay soils and 
the yields on loam soils, the overall average 
yield was 2025 kg/ha on clay and 2046 kg/ha on 
loam. This might suggest at first glance that 
wheat yields are about uniform across the Black 
Chernozem soil zone in Manitoba, with no sig- 
nificant differences between yields obtained on 
clay soils and those obtained on loam soils. In 
reality, this is not the case. 

Statistical analysis of yield series, employing 
the Quade and Friedman test, suggested that 
the yields simulated on different soils were sig- 
nificantly different at a = 0.01. All except Den- 
cross clay (75-DCS) and the Ryerson loam (1- 
RYS) yield series passed the normality test. 
Since many decisions of land use planners and 
land use managers are based on the assumption 
that yield is normally distributed, it was consid- 
ered advantageous to describe wheat yields by a 
family of normal probability functions. The two 
heterogeneous yield populations (75-DCS and 1- 
RYS) were dissected into components according 



Table 2. Land Use characteristics on clay and loam soils with the Chernozemic Great Groups. 






Brown 




Dark 


Brown 


Black 




Clay 


Loam 


Clay/ 


Clay 


Loam 


Clay/ 


Clay 


Loam 


Clay/ 


Characteristic 


A1013 


A1019 


Loam 


A2055 


A2039 


Loam 


A3098 


A3086 


Loam 




-KIN 


-ALS 


ratio 


-REG 


-DIV 


ratio 


-MEL 


-PRA 


ratio 


Farm Size (ha) 


483.54 


516.19 


0.94 


384.94 


397.19 


0.97 


280.60 


251.20 


1.12 


Cultivated/Farm (%) 


93.42 


88.67 


1.05 


94.99 


88.78 


1.07 


90.41 


83.18 


1.09 


Summerfallow (%) 


41.50 


41.36 


1.00 


36.26 


37.45 


0.97 


16.07 


19.23 


0.84 


Wheat (%) 


44.66 


46.25 


0.97 


50.37 


41.31 


1.22 


33.10 


30.26 


1.09 


Oilseed (%) 


0.65 


0.28 


2.32 


0.95 


0.97 


0.98 


13.08 


10.59 


1.24 


Other Grain (%) 


9.54 


5.11 


1.87 


4.99 


8,13 


0.61 


28.48 


20.26 


1.41 


Hay(%) 


0.54 


1.54 


0.35 


2.63 


5.45 


0.48 


3.29 


9.75 


0.34 


Improved Pasture (%) 


1.54 


3.91 


0.39 


2.30 


3.55 


0.65 


2.11 


4.44 


0.48 


Total Capital Investment ($/ha) 


1926.72 


1563.69 


1.23 


2094.71 


1338.71 


1.56 


2423.31 


2008.88 


1.21 


Operating Expenses ($/ha) 


50.64 


45.49 


1.11 


53.43 


46.49 


1.15 


99.54 


77.97 


1.28 


Total Sales ($/ha) 


151.57 


129.29 


1.17 


143.66 


121.83 


1.18 


203.66 


155.48 


1.31 


Sales/Investment 


0.08 


0.08 


1.00 


0.07 


0.09 


0.78 


0.09 


0.08 


1.13 


Sales/Expenses 


2.82 


2.69 


1.05 


2.56 


2.54 


1.01 


1.99 


2.01 


0.99 



142 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Polygon 

89 - RIV 

90 -OBO" 
75 - DCS 
73 - RLD 
125 - DPH" 
53 - NDL' 
1 - RYS 



2341.00 
1972.00 

2206.00 
2189,00 
2692.00 



to a procedure described 
by Hald (1952), and their 
distributions were ap- 
proximated by two nor- 
mal distribution func- 
tions with parameters 
((i p a x ) and (|H 2 , o 2 ) com- 
bined in the ratio A,^ 

^(V^D- 

Table 3 presents the 

parameters of the den- 
sity functions as calculated from the wheat yield 
series. It should be noted that, in some polygons 
and in some years, killing frost (T <- 2C) oc- 
curred before the wheat crop reached maturity. 
Such frost events were not considered in the cal- 
culation of the parameters of yield probability 
density functions. However, these events were 
included in the risk analysis. 

Analyses show that wheat yield distributions 
differ from each other in terms of at least one of 
the parameters of the probability density func- 
tion, either in location (\i) y scale (a), or shape (A, ). 
The probability density curves differ not only 
between the two textrual classes but also within 
textural classes (Fig, 2a and 2b). 

The Red River clay (89-RIV), an imperfectly 
drained soil with an artificial surface drainage 
network, was the most productive among the 
clay soils considered in the analysis. Dauphin 
clay (125-DPH), a soil very similar to Red River 
but approximately 350 km northeast, ranked 
second in productivity. The main difference be- 



Table 3. Parameters of wheat yield distribution functions 
for repesentative polygons/soils (kg/ha). 



Normal distribution 



Parameters 



Heterogeneous distribution 



311.00 
431.00 

377.00 
188.00 
345.00 



0.50 2296.00 214.00 0.50 1531.00 405.00 



0.65 2368.00 217.00 0.35 1228.00 393.00 



Note: * Killing frost occured before the wheat crop reached the maturity stage in one year out of 20 years. 



tween these two soils is amount and reliability 
of available solar energy to the crop. Soils in the 
Dauphin area warm up more slowly in the 
spring, delaying the seeding date and depress- 
ing crop growth. In addition, a killing frost oc- 
curred in 125-DPH in 1972 at a time when 
wheat development was half-way between soft 
dough and maturity. 

On the other hand, Osborne and Dencross 
clay soils, which occur in the same geographic 
area as the Red River soil, have quite different 
wheat yield distributions. The yields on these 
two soils are generally depressed and highly 
variable. 

On Osborne clay (90-OBO), a poorly drained 
soil, low yields are induced by a combination of 
low topographical position in the landscape, low 
infiltration rate (approximately 1 cm/day), pres- 
ence of a water table at shallow depth, and lack 
of adequate surface drainage. Seeding fre- 
quently is delayed on Osborne soils because of 
their poor drainage and cooler spring tempera- 



a) 



P(x) 




125-DPH" 
89-RIV 
90-OBO' 
75-DCS 



1-RYS 
53-NDL* 
73-RLD 



1 000 2000 3000 

GRAIN YIELD (kg/ha) 



4000 




1000 2000 3000 ' *' lf 4000 

GRAIN YIELD (kg/ho) 



Figure 2. Probability density curves of wheat yield: a) on clay soils, and b) on loam soils. 



ONOFREI, DUMANSKI, EILERS, AND SMITH: A COMPARISON OF LAND USE AND PRODUCTIVITY IN WESTERN CANADA 143 



P(x) 



a) 



P(x) 



b) 



89-RlV (CLAY) 
73-RLD (LOAM) 




1000 2000 3000 

GRAIN YIELD (kg/ha) 



4000 




125-DPHYCLAY) 
53-NDL'(LOAM) 



1000 2000 3000 

GRAIN YIELD (kg/ho) 



4000 



Figure 3. Probability density curves of wheat yield on adjacent clay and loam soils: a)South- 
Central and b)North of Black zone. 



tures. Often, portions of cropped fields are 
flooded and wheat yields are partially lost. The 
likelihood of excess of water, or more precisely 
the lack of oxygen in the soil and consequently 
the depression of crop yield, is very high on 
Osborne soils. Also, a killing frost (T = - 4C) 
occurred before the wheat crop reached matur- 
ity in 1974. 

The Dencross clay (75-DCS), an imperfectly 
drained soil, presents somewhat better charac- 
teristics than 90-OBO. The topographical posi- 
tion in the landscape is higher and the infiltra- 
tion rate >2 cm/day. The yield on Dencross soils, 
however, is depressed about 50% of the time, 
due to high precipitation during the growing 
season. In the remainder of the time the wheat 
yields can be as high as those obtained on the 
Red River clay soils. Overall, however, yield 
variability on Dencross clay is considerably 
higher than on Red River clay. 

The loam soils are very similar in terms of 
their characteristics, but variation in yield can 
still be considerable (Fig. 2b). If the risk of frost 
is disregarded, then Newdale soil (53-NDL) 
outperforms the Reinland soils (73-RLD). How- 
ever, in the 53-NDL polygon, a killing frost oc- 
curred before the crop reached maturity in 1982. 

The yield distribution on the Ryerson loam (1- 
RYS) exhibits heterogeneity similar to that ob- 
served on the Dencross clay (75-DCS) but for 
entirely different reasons. The 1-RYS polygon is 
located in the driest area of Manitoba. High soil 
moisture deficits during the growing season on 



these soils resulted in depressed wheat produc- 
tion in 7 years out of 20 or about 35% of the 
time. Nevertheless, during years of high pre- 
cipitation, yields on Ryerson soils can be almost 
as high as those on Newdale soils. 

Figures 3 a and 3b illustrate an overall com- 
parison of crop production on clay soils versus 
loam soils for selected polygons. In south cen- 
tral Manitoba, the Red River clay outperformed 
the Reinland loam soil (Fig. 3a), whereas, near 
the northern edge of the Black Chernozems, the 
Newdale loam soil outperformed the Dauphin 
clay (Fig. 3b). Of particular significance was 
that the yield variability on both these clay soils 
was less than that on the loam soils included in 
the comparison. 

Information on the long-term mean yield lev- 
els and on the shape of yield distribution curves 
can provide important input for risk analysis. 
The yield density functions can be used to quan- 
titatively asses the probability of obtaining a 
given threshold yield, and this can be correlated 

Table 4. The rank of land units (Polygons) based on risk 
of obtaining a wheat yield equal to or lower than 1783 



Rank 



Polygon 



P(X<:1783) 



1 


89 


- RIV 


0.037 


2 


53 


- NDL" 


0.054" 


3 


125 


- DPH" 


0.064" 


4 


73 


- RLD 


0.131 


5 


1 


- RYS 


0.322 


6 


75 


- DCS 


0.366 


7 


90 


- OBO' 


0.380" 



"NoTeT 

"Killing frost (T^ 2*C) occurred before the wheat crop reached 
maturity in one out of 20 years 

" Probability was calculated assuming that the yield was lost (X=0) in 
the years when the killing frost occurred. 



144 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



with risk. Comparative production risk, which 
considered all clay and loam soils included in 
this study, was performed based on the eco- 
nomic break-even yield of 1783 kg/ha, calcu- 
lated for Manitoba in 1987 (break-even yield is 
the yield necessary to cover input costs only). In 
this analysis it was assumed that the yield was 
zero in the years when a killing frost occurred 
before wheat crop reached maturity. Table 4 
presents the probability of obtaining a yield 
equal to or lower than 1783 kg/ha on each soil/ 
polygon combination. 

The lowest risk of obtaining such an undesir- 
able yield was on the Red River clay (98-RIV). 
On these soils one can expect yields equal to or 
lower than 1783 kg/ha in only about four out of 
100 years; P(X=Grain yield <1783 kg/ha) = 
0.037. The highest risk of obtaining this low 
level yield was on the Osborne clay (90-OBO), a 
catenary member with Red River clay. Within 
the agroecosystems that operate on Osborne 
clay, one could expect yields to be equal to or 
lower than 1783 kg/ha in 38 out of 100 years. 
The risk of obtaining this threshold yield was 
higher on the Reinland loam (73-RLD) than on 
the Red River clay but lower than on the 
Osborne or Dencross clay. Only a minor differ- 
ence in probability was detected between the 
Newdale loam (53-NDL) and the Dauphin clay 
(125-DPH), located at the northern fringe of the 
Black zone. It is important to note that individ- 
ual rankings are not absolute values, but they 
are relative to the soils included in the compari- 
son and to the threshold yield selected as the cri- 
teria for comparison. For example, if 1168 kg/ha 
is taken as the threshold yield (the break-even 
yield for wheat calculated for Manitoba in 1985), 
then there are no differences in terms of risk 
between the Red River and Dauphin clay soils 
and the Newdale loam soil. 

Conclusions 

In general, farm type and cropping systems in 
the interior plains of western Canada changed 
with gradients in temperature and precipitation 
but not with soil textural classes within a re- 
gion. A large proportion of available arable land 
in the Brown and Dark Brown zones was used 
for summerfallow, whereas in the Black zone 
continuous cropping is much more common. 
Spring wheat was the dominant crop cultivated 
in all soil zones. The largest values of total capi- 



tal investment and operating expenses per unit 
area were found on clay soils in the Black zone, 
but these farms also had the highest total sales 
per hectare. If the ratio of sales/expenses is 
taken as an overall indicator of efficiency, then 
farms in the Brown zone were the most efficient, 
with those on clay soils being slightly more effi- 
cient than those on loam soils. 

The wheat yield analysis showed that crop 
productivity varied widely not only between tex- 
tural groups but also within the same soil tex- 
tural class. The risk of obtaining a given thresh- 
old yield varied within catenary members of soil. 
Some clay soils outperformed adjacent loam 
soils (Red River vs. Reinland), whereas others 
ranked behind the loam soils (Osborne vs. Rein- 
land, Dencross vs. Reinland, Dauphin vs. 
Newdale). 

Defining the variability of yield based on yield 
frequency distributions provides a new dimen- 
sion for crop yield interpretation. This informa- 
tion provides a more complete and realistic base 
for evaluating relative land quality for crop pro- 
ductivity. Such information can be used to as- 
sist with planning decisions and to facilitate the 
development of fair and equitable agricultural 
support programs. Results of this study indicate 
that only a certain number of fundamental soil 
characteristics, agronomic land attributes, and 
weather elements strongly affect productivity in 
the interior plains of western Canada. Consid- 
erable progress in reserch can be made by cen- 
tering efforts on these key, controlling factors. 

Literature Cited 

Clayton, J. S., A.W. Ehrlich, B.D. Cann, H J. Day andBJ. 
Marshal. 1977. Soils of Canada. Research Branch, 
Canada Department of Agriculture, Ottawa, Ontario. 
482 p. 

Conover, W.J. 1980. Practical Nonparametric Statistics. 
John Wiley and Sons, New York. 493 p. 

Dumanski, J. and R.B. Stewart 1981. Crop production 
potentials for land evaluation in Canada. LRRI, Agr. 
Canada. Ottawa, Ontario. 80 p. 

FAO, 1978. Report on the agro-ecological zones project. 
Vol. 1. Methodology and results for Africa. World Soil 
Resour. Rep. 48 Rome. 158 p. 

FAO, 1984. Guidelines: land evaluation for rainfed agri- 
culture. Soil Bull. 52. Rome. 237 p. 

Hald, A. 1952. Statistical Theory with Engineering Appli- 
cations. Applied Statistics, John Wiley and Sons, Inc., 
New York. 783 p. 



ONOPREI, DUMANSKI, EILERS, AND SMITH: A COMPARISON OP LAND USE AND PRODUCTIVITY IN WESTERN CANADA 145 



Huffman, E. 1988. A description of physical and economic 
strategies of farming in the major soil zones of the 
Canadian Prairies. In Crop production risks in the 
Canadian prairie region in relation to climate and land 
resources. Dumanski, J. and V. Kirkwood (eds.) Tech- 
nical Bull. 1988-5E, LRRC, Research Branch, Agricul- 
ture Canada, Ottawa, Ontario, pp 17-30. 

Kirkwood, V., J. Dumanski, A. Bootsma, R.B. Stewart and 
R. Muma. 1989. The land potential data base for Can- 
ada. User's handbook. Res. Br., Agric. Canada, Ottawa. 
53 p. 

Onofrei, C. 1986. A method of Land Evaluation Using 
Crop Simulation Techniques. Unpubl. Ph.D. Thesis, 
University of Manitoba, Winnipeg. 314 p. 



Onofrei, C. 1987. Land Evaluation based on wheat yield 
probability density functions. Technical Report, De- 
partment of Soil Science, University of Manitoba and 
LRRC, Research Branch, Agriculture Canada, Ottawa, 
Ontario. 197 p. 

Williams, G.D.W. and H.N. Hayhoe. 1982. Procedures for 
computing and mapping Thiessen weighting factors for 
digitized district boundaries and climatological station 
latitude and longitude. Technical Bull. 1982-3E, LRRI, 
Research Branch, Agriculture Canada, Ottawa, On- 
tario. 22 p. 



146 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Reclamation Management and Techniques for Cold Entisols in 

Southwestern Wyoming 

RE. Parady 1 and Norman E. Hargis 2 



Abstract 

Bridger Coal Company operates a 5.4 tonne year' 1 (6.0 million tpy) surface 
coal mine 56 km (thirty-five miles) northeast of Rock Springs, Wyoming. 
Approximately 8,100 ha (20,000 acres) are under permit, with disturbance 
over the life of the mine projected to reach 4,050 ha (10,000 acres). Located 
on the western rim of the continental divide, the mine receives less than 23 
cm (9 inches) of precipitation annually. Soils in the area are coarse-tex- 
tured, and problems associated with elevated salinity and sodicity are en- 
countered. 

A variety of common reclamation techniques have been modified to re- 
flect these conditions. Soil horizons are segregated during salvage opera- 
tions (the surface six inches as topsoil and the balance as subsoil). Unsuit- 
able materials are not salvaged. Direct application of soil is used to maxi- 
mize native plant regeneration and conserve soil fertility. Interseeding of 
seeding failures has proven to be significantly more successful then chisel 
plowing and reseeding. Broadcast seeding has been ineffective because of 
strong winds, and a no till drill has been modified to handle diverse seed 
mixes and rock conditions. The utility of fertilization under typically xeric 
moisture regimes is being evaluated. 



Introduction 

Bridger Coal Company mines from 3.9 to 6.5 
million tonne year" 1 (4.4 to 6.5 million tpy) of 
subbituminous steam coal at its surface strip 
mine located 56 km (thirty-five miles) northeast 
of Rock Springs in southwestern Wyoming. The 
mine is adjacent to the continental divide at ele- 
vations ranging from 2,073 to 2,150 m (6,800 to 
over 7,100 feet.) 

Mean annual precipitation is 0.22 m (8.8 
inches), and the average number of frost free 
days is 100 (Bridger Coal Company, 1980). Br- 
idger's permit to mine encompasses nearly 
8,100 ha (20,000 acres), with 2,000 ha (4,950 
acres) disturbed and over 520 ha (1,280 acres) or 
26% reclaimed to date (Bridger Coal Company, 
1988). Life of mine disturbance is projected to 
reach approximately 4,050 ha (10,000 acres). 

Reclamation feasibility on arid lands has 
been questioned since the resurgence of the 
western coal mining industry in the early 1970s. 
The National Academy of Sciences (1974) sug- 
gested that 0.25 m (ten inches) of precipitation 
was necessary to sustain revegetation. Bridger 
Coal has developed or modified a variety of rec- 
lamation equipment and techniques to reflect 
local conditions and provide the foundation for 



Manager, Safety and Reclamation, Bridger Coal Com- 
pany, P.O. Box 2068, Rock Springs, Wyoming, 82901. 

2 Reclamation Coordinator, Bridger Coal Company, P.O. 
Box 2068, Rock Springs, Wyoming, 82901. 



successful reclamation. The following informa- 
tion is based upon observations during the past 
8 growing seasons. 

Soil Environment 

An Order One soil survey was conducted in 
1978 which identified 25 series. Soils are pre- 
dominantly Entisols (Table 1). Torrifluvents 
occur in alluvium along major ephemeral drain- 
ages. Torripsarnments occur on sand dunes. 
Torriorthents occur in smaller ephemeral drain- 
ages and on slopes and uplands. The typical 
thickness of A horizons is 0.05-0.13 m (2-5 
inches). Normally C horizons are immediately 
below the A horizon; AC horizons are rare. 

Soil formation has been limited and is closely 
related to local geology. Soils are derived from 
Cretaceous calcareous shale, Tertiary sandstone 
and shale, aeolian sand, and alluvium (Bridger 
Coal Company, 1980). 

The temperature regime is frigid. The 25 soil 
series support 7 range sites: saline upland, 
shallow loamy, loamy, sands, saline lowland, 
impervious clay, and saline subirrigated. Prin- 
cipal plant communities are gardner saltbush, 
big sagebrush- wheatgrass, grease wood, and big 
sagebrush-rubber rabbitbrush. Vegetative cover 
is less than 25%. Rock outcrops with little or no 
soil cover are common. Slickspots, which are 
saline-sodic and unsuitable as soil, occur in 
some major drainages. 



PARADY AND HARGIS: RECLAMATION MANAGEMENT AND TECHNIQUES FOR COLD ENTISOLS IN SOUTHWESTERN WYOMING 147 



Table I. Major taxonomic groupings of soil series in 

Bridger Coal Company mining permit, Sweetwater 
County, Wyoming. 



Order 



Great Group 



Series 



Aridisols 



Entisols 



Natrargids 
Calciorthids 

Camborthids 
Torrifluvents 



Torriorthente 1 



Torripsam ments 



Westvaco 

Cambarge 

Pepal 

Sage Creek 

Quealman 

Laney 

Chrisman 

Haterton 

Horsley 

Huguston 

Leckman 

Rock Springs 

Terada 

Thayer 

Garsid 

Monte 

Dinco 

Dines 

Boltus 

Tasselman 

Winton 

Corlett 

Fandaly 



Includes one unnamed series. 



Specifications for soil management are based 
on soil chemical and physical properties. Soils 
are salvaged for use in reclamation to the paral- 
ithic contact or to the depth of elevated salt, 
sodium, or other unsuitable property. This in- 
sures a suitable soil environment for seedlings 
and growing plants. 

Soil Management 

Soils on the mine site are typically Entisols, 
coarse textured, with an average pH of 7.5 to 8.0 
and electrical conductivity in the 4.0 to 6.0 ds 
nr 1 range. Problems associated with elevated 
salinity, sodicity, and boron levels are encoun- 
tered. Bridger Coal has implemented a soil 
management program to assure proper use of 
soil resources. 

First, a staking program is used on the high- 
wall to identify unsuitable native soils. Unsuit- 
able materials are not be moved onto recon- 
toured spoil unless soil heterogenity is specifi- 
cally desired. Suitability is defined in accor- 
dance with state guidelines (Table 2), 

During soil stripping operations, soil horizons 
are segregated, the surface 15 cm (six inches) as 
topsoil and the balance as subsoil. Soils range 
from 15-60 cm thick (six inches to sixty inches), 
with a mine wide average of 38 cm (fifteen 
inches). Revised permit language has been ap- 
proved by the state to allow for variable soil 
application thicknesses from 30 cm (twelve 
inches) on the ridge positions to 76 cm (thirty 
inches) in lowland positions. Variable soil 
depths provide a foundation for vegetative di- 
versity. 



Stripped soil is either stockpiled or hauled 
directly onto a completed regraded area. Stock- 
piles are contoured to allow farming operations 
on the side slopes and to minimize erosion. 

Direct application of soil is a key element in 
achieving diversity elements of bond release cri- 
teria. Soil is picked up with scrapers and trans- 
ported across or through the pit, and then placed 
directly on ripped recontoured spoil. Bridger 
Coal initiated direct application of soil in 1976, 
and has succeeded in using this preferable 
method on over 240 reclaimed hectares (600 
acres) to date. The increased cost of longer 
hauls associated with direct application is offset 
by elimination of double handling incurred by 
stockpiling. This technique maximizes native 
plant regeneration and conserves soil fertility. 
Bridger Coal has also completed soil application 
to 70 ha (175 acres) using stockpiled material as 
a subsoil covered with a 0.3 m (six inch) surface 
application hauled directly form the highwall. 

Specifically, ten species have volunteered 
from direct applied soil. These species include 
big sagebrush (Artemisia tridentata), Sand- 
berg's bluegrass (Poa sandbergu), Fendler's 
bluegrass (Poa fendleriana), greasewood (Sarco- 
batus vermiculatus), rubber rabbitbrush 
(Chrysothamnus nauseosus), Plains wallflower 
(Erysimum asperum), scarlet globemallow 
(Sphaeralcea coccinea), and scarlet gilia (Ipo- 
mopsis aggregata). Direct application of soil 
aids in returning the shrub and forb components 
of the plant community, as well as in establish- 
ing understory species of the grass component. 

Overburden/Parting Management 

The objective of the overburden characteriza- 
tion and handling plan is two fold. First, to in- 
sure that the surface 1.2 m (four feet) of mate- 



Table 2. Selected Criteria for Assessing Suitability of 
Materials for Use Within the Root Zone. 


Parameter 


Suitable 


Unsuitable 


Acid-Base Potential 


>-5.0 


<-5.0 


(A/B) (Tons CaCO 3 






per 1,000 tons) 






Boron (ppm) 


< 5.0 


> 5.0 


Electrical Conductivity 


< 8.0 


> 8.0 


(ds m' 1 ) 






Exchangeable Sodium 


<20.0 


>20.0 


percentage (%) 






Molybdenum (ppm) 


< 1.0 


> 1.0 


pH 


5.0 - 9.0 


< 5.0, > 9.0 


Saturation 


25 -80% 


< 25, >80% 


percentage (%) 






Sodium Adsorption Ratio 


< 10.0 


>10.0 


Texture 


< 50% clay, 


> 50% clay 




< 85% sand, 


> 85% sand 


Total Organic 


< 10.0 


>10.0 


carbon (%) 







148 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



rial is suitable for plant growth. Second, to in- 
sure that unsuitable materials are not placed in 
surface drainages, where it could be eroded To 
meet these objectives, we employ the following 
techniques: 

LDrilling on a 65 ha (160 acre) spacing. 
Laboratory analysis has been completed on 
the entire area. This program defined the 
specific problems (salinity, sodicity, and 
boron) and zones of unsuitable materials. 
2.Unsuitable materials comprising less than 
15% of the total volume are mixed during 
mining with draglines. 
3. Supplemental stripping equipment is used 
to extract unsuitable materials, primarily 
interburden, and place them on the pit 
floor, below the regraded surface but above 
the anticipated potentiometric surface. 
4.The dragline swing may be modified to 
place material lower within the spoil peak 
for subsequent burial. 

S.Following recontouring, regraded spoils are 
sampled on 120 m (400 foot) grid in 0.6 m 
(two foot) intervals to a 1.2 m (four foot) 
depth. Samples are analyzed for pH, EC, 
ESP or SAR, molybdenum, boron, and acid/ 
base potential. Results are evaluated using 
the criteria in Table 1. 
6. If unsuitable materials are identified, they 

are either covered or relocated. 
Since the fall of 1981, samples have been 
taken and analyzed to provide a basis for evalu- 
ating the suitability of root zone materials prior 
to soil application. Results have been reported 
in detail in the 1982 through 1988 Annual Re- 
ports. To date, 24 areas comprising nearly 220 
ha (540 acres) have been evaluated. A total of 
516 samples have been taken. Less than seven 
percent of these samples have had unsuitable 
value for any parameter, and less than 0.8 ha 
(two acres) have been withheld from soil appli- 
cation on the basis of spoil unsuitability. These 
results clearly demonstrate the guaranteed 
cover plan has been successful in accomplishing 
proper materials placement. 

Farming Operations 

The goal of reclamation is set by statute un- 
der the Surface Mining and Reclamation Act of 
1977 to establish a diverse, native plant commu- 
nity capable of regenerating itself. Diverse tech- 
niques must be employed to achieve diversity of 
species in a reclaimed plant community. Seed 
drills used initially in reclamation, specifically 



the Laird Rangeland drill, were not capable of 
handling fluffy, trashy native seeds such as 
winterfat (Ceratoides lanatd). Consequently, 
only 5 or 6 species were seeded in early reclama- 
tion efforts, generally wheatgrasses and 
fourwing saltbush. 

To remedy this, Bridger Coal Company pur- 
chased a Tye "Pasture Pleaser" no till seed drill. 
The drill has three seed boxes, with fairly stan- 
dard wheatgrass and legume boxes. The third, 
a shrub box, is specifically equipped with agita- 
tor discs and larger picker wheels to handle 
trashy seed. In addition, the large seed tube 
from this box can distribute seed across the en- 
tire furrow. The result is variable planting 
depths, including shallow planting depths that 
are desirable for most of these native species. 
Bridger Coal has therefore been able to use 18 to 
20 species in each of its seed mixes, correspond- 
ing to four range sites: shallow loamy, sands, 
saline upland, and saline lowland. Big sage- 
brush (Artemesia tridentata) was successfully 
established with this technique in several fall 
seedings. Studies by DePuit and Coenenberg 
(1979) have indicated that increasing the num- 
ber of species in a seed mix increases the diver- 
sity of the resulting plant community. 

An additional technique that has proven suc- 
cessful is interseeding. Interseeding involves 
seeding with a no till directly into an existing 
reclaimed surface, rather than chisel plowing 
and reseeding. The advantage lies in minimiz- 
ing disturbance to the soil and in keeping exist- 
ing vegetation intact. During the fall of 1981, 
portions of a reclaimed area were either in- 
terseeded or chisel plowed and reseeded. By 
1984, the interseeded area showed 180 desir- 
able plants m' 2 , compared with 53 desirable 
plants nr 2 on the area chisel plowed and then 
reseeded (Bridger Coal Company, 1984). In- 
terseeding can also be useful as it provides a 
second age group of plants within the commu- 
nity. 

Broadcast seeding has had limited success at 
Bridger Coal, primarily because of wind. Broad- 
cast seeding is intended to provide the shallow 
planting depth necessary for native species, as 
well as improving reclamation aesthetically by 
eliminating the appearance of drill rows. A 
modified broadcast seeder was used in 1981 on 
150 acres. Average first year seedling density 
resulting from the broadcast seeder was 15.2 
seedlings nr 2 . Average first year seedling den- 
sity on 53 ha (130 acres) seeded with a drill in 



PAKADY AND HARGIS: RECLAMATION MANAGEMENT AND TECHNIQUES FOR COLD ENTISOLS IN SOUTHWESTERN WYOMING 149 



1981 was 47.5 seedlings nr 2 (Bridger Coal Com- 
pany, 1984), a 300% difference. 

Several changes in mulching operations have 
significantly improved productivity. First, a 
round bale buster was purchased to replace both 
a blower type mulcher and a tub mulcher. The 
blower was labor intensive, requiring a tractor 
operator, two hay handlers, and blower opera- 
tor. The tractor operator can self-load the bale 
buster from the cab, eliminating the need for 
three people. Second, large (500 - 725 kg) (1100 
- 1600 pound) round bales are used, eliminating 
the handling involved with small bales. Third, a 
4.6 m (15 foot) working width flexible crimper 
with hydraulically operated gangs and trans- 
port wheels was put into service, replacing a 
small 1.8 m (six foot) crimper. A final improve- 
ment in the mulching operation was replacing 
straw with grass hay. Hay adheres to the sur- 
face better. 

Shrub Establishment 

Proposed state regulations require establish- 
ment of one shrub per square meter in a mosaic 
pattern on 10% of the mine's reclaimed area. 
This standard has been met on 20 ha (48 acres) 
or 6% of reclamation to date and has nearly 
been met on several additional areas. 

Three species of sagebrush (Artemisia triden- 
tata subsp. wyomingensis, Artemisia cana, Ar- 
temisia frigida), fourwing saltbush (Atriplex 
canescens), Gardner's saltbush (Atriplex gard- 
neri), winterfat, rubber rabbitbrush, grease- 
wood, and spiny hopsage (Grayia spinosa) are 
currently used in different seed mixes to pro- 
mote shrub establishment for wildlife habitat. 
Fourwing saltbush, Gardner's saltbush, and 
winterfat have been especially successful in rec- 
lamation seedings. 

Direct application of soil also maximizes 
shrub establishment by increasing the survival 
of propagules and vesicular arbuscular mycor- 
rhyzae that remain in the soil at the surface of 
reclamation. Wyoming big sagebrush, rubber 
rabbitbrush, and greasewood have been estab- 
lished with the technique. 

Irrigation 

A cooperative research project with the Uni- 
versity of Wyoming has been completed to as- 
sess the establishment of a predominantly na- 
tive, diverse seed mix under irrigation. The ob- 
jectives of the research included determining 
optimum irrigation rates for initial vegetation 



establishment; determining optimum seasonal 
scheduling and duration; and defining interac- 
tive effects of varied treatments on initial and 
ultimate vegetation density, productivity, spe- 
cies composition, and diversity (Vincent et al, 
1986). 

Fertilization 

Analysis of a poor reclamation area first 
seeded in 1981 revealed total nitrogen and 
available phosphorous levels (.03% N and 2.1 
ppm P) below desirable plant available levels 
(Bridger Coal Company, 1981). Fertilizer at 170 
kg/ha (150 Ibs/acre) of 18-46-0 was applied in 
the spring of 1984 and the area was then in- 
terseeded. Seedling establishment from the in- 
terseeding appears satisfactory and established 
plant vigor appears improved. However, the 
utility of fertilizer in the region is probably lim- 
ited to average or above average precipitation 
years. 

Summary 

Many of the initial concerns over reclamation 
feasibility in a semi-arid desert environment 
have been laid to rest. Improvements have oc- 
curred in soil management, shrub establish- 
ment, and farming operations. Experiments are 
underway with various techniques such as irri- 
gation and fertilization. 

Wyoming Department of Environmental 
Quality personnel have ocularly evaluated all 
reclaimed areas at Bridger Coal annually since 
1982. The percentage of reclamation rated good 
or fair increased from 38.2% in 1982 to over 90% 
in 1988. The percentage of disturbance re- 
claimed has doubled in less than five years, from 
13.3% in 1979 to 26% in December of 1988. 
Reclamation has been successfully achieved in 
areas receiving less than 0.25 m (ten inches) of 
precipitation. 

Literature Cited 

Bridger Coal Company, 1980. Jim Bridger Mine Base 
Document, Wyoming Permit to MineNo.338-C. Volume 
2: Mine Plan and Volume 17: Soil Survey Report. P.O. 
Box 2068, Rock Springs, Wyoming. 

Bridger Coal Company, 1981. Jim Bridger Mine 1981 
Annual Report, Permit to Mine No. 338-C. Volume 3: 
Irrigation. P.O. Box 2068, Rock Springs, Wyoming. 

Bridger Coal Company, 1983. Jim Bridger Mine 1983 
Annual Report, Permit to Mine No. 338-T1. Volume 1. 
P.O. Box 2068, Rock Springs, Wyoming. 



150 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Bridger Coal Company, 1984. Internal correspondence- 
Vegetation Monitoring. P.O. Box 2068, Rock Springs, 
Wyoming. 

DePuit, E.J. and J.G. Conenenberg, 1979. Methods for 
establishment of native plant communities on topsoiled 
coal stripe mine spoils in the Northern great Plans. 
Reclamation Review 2: 75-83. 

National Academy of Sciences, 1974. Rehabilitation poten- 
tial of western coal lands. Ballinger Publishing Co., 
Cambridge, Massachusetts. 



Vincent, R.V., E.J. DePuit, J.L. Smith, J.A. White, and 
F.E. Parady. 1986. Design of temporary irrigation for 
plant establishment on arid coal mined lanfds in Wyo- 
ming. Pages 45-47 in 39th Annual Report, Vegetative 
Rehabilitation and Equipment Workshop, U.S. Depart- 
ment of Agriculture, forest Service, Misoula, Montana. 

Wyoming Department of Environmental Quality, 1984. 
Guideline No.l-Soils and Overburden. Cheyenne, Wyo- 
ming. 



Vertisols of New Caledonia: Distribution; Morphological, Chemical, 
and Physical Properties; and Classification 

R Podwojewski and A.G. Beaudou 1 
Abstract 

The Vertisols of New Caledonia are found on the west coast of the main 
island, where a dry and varied climate prevails. They occur in the lower 
parts of soil catenas, overlying basic rocks, or on old old alluvial terraces. 
The chemical composition of the parent material determines the chemical 
properties of the soils. Three kinds of vertisol have been identified: calci- 
magnesic vertisols, hypermagnesic vertisols, and sodic and acid vertisols. 
Some forms, however, show characteristics intermediate between the three 
categories. This paper describes the vertisols in terms of their morphologi- 
cal and analytical features and proposes a means of classification. 



Introduction 

New Caledonia is situated slightly north of 
the Tropic of Capricorn (Fig. 1). Vertisols there 
have been described as "tropical black clays" in 
the past (Tercinier, 1953). They cover an ap- 
proximate total area of 100,000 ha (Latham et 
al., 1978). Compared to the surface area of the 
Territory as a whole (16,000 km 2 ), this area is 
relatively small, but vertisols represent nearly 
50% of the land suitable for mechanized agricul- 
ture. These deep soils are present only on the 
west coast of New Caledonia (Podwojewski, 
1984; Podwojewski, Beaudou, 1987) (Fig. 2). 

Vertisols have a succession of three horizons 
(A.F.E.S., 1988). They include a surface horizon 
(Vs), an intermediate horizon (V) with a polyhe- 
dral or prismatic structure and a number of 
shiny or furrowed surfaces, and a vertic horizon 
(Vv) with a wedge-shaped, rhombohedral or 

Figure 2. Location of Vertisols in New Caledonia 



CORAL 



/AUSTRALIA 



o 




P A C I F I C 



OCEAN 



Auckland 
NEW ZEALAND 




Figure 1. Location of New Caledonia 




Hyprmagnal vortitol 
Sodlo and acid vrtial 



. IOOO Uohytt(mM) 



28 SO 75 100 Km 



'Podwojewski, Orstom, 70-74 route d'Aulnay, 93 140-Bondy (France). A.G. Beaudou, Jalan Tampak Siring, 10, Cipete, Jakarta 
Selatan (Indonesia). 



151 



152 



SIXTH INTERNATIONAL SOIL CLASSIFICATION WORKSHOP 



sphenoidal structure. The high content of smec- 
tites in these horizons is responsable for the 
vertic structure. 

Formation 

Parent material, landscape position, and cli- 
matic conditions influence the formation of ver- 
tisols in New Caledonia. Most vertisols gener- 
ally developed from three major groups of par- 
ent material. These are relatively basic rocks 
that are fairly rich in alkali/earth cations (cal- 
cium or magnesium). The weathering of these 
rocks produces smectite type clays. 

The first of the two petrographic groups in- 
cludes flysches, pelites, limestones and basalts, 
dolerites, and gabbros, basic rocks in which 
%SiO varyies from 50 to 55%. The dominant 
base is calcium, closely followed by magnesium. 
In some cases, the magnesium may , however, 
be slightly dominant. These rocks contain little 
potassium or sodium. 

Serpentinites (associated with peridotites) 
constitute the second group. These are ultraba- 
sic rocks containing less than 50% SiO 2 . The 
only base present is magnesium. 

The third group is composed of graywackes 
and schists. These rocks are more acid than the 
other groups and contain more than 55% SiO 2 . 
The dominant bases are magnesium and so- 
dium. Little calcium or potassium is in these 
rocks. 

Location 

Vertisols are on the plains of western New 
Caledonia at an elevation less than 100 m. (Fig. 
3). They always are located in the lower portions 
of the landscape (Beaudou et al., 1983). They 
occur on the lower part of toposequences formed 
on materials derived directly from the subjacent 
rock or on colluvia. The soil sequence is as fol- 
lows: an immature erosion soil, a vertic brown 
soil, and a vertisol. The transition to vertisol is 
most evident in places where the slope is gentle 
and where external drainage is slow. 

Vertisols also occur on the old alluvial ter- 
races of the main water courses on the west 
coast of New Caledonia. These are commonly 
the highest terraces. External drainage is very 
limited or non-existent. There is a continuum 
between vertisols formed on the substratum 
(lithomorphous vertisols) and vertisols forming 
in old alluvia (topomorphic vertisols). Gilgai 
microrelief is not observed in New Caledonia. 



Climate 

New Caledonia is characterized by a semi-hot 
oceanic tropical climate. Vertisols occur only on 
the plains of the west coast, leeward of the 
south-easterly tradewinds. In these areas, mean 
annual rainfall is usually less than 1,000 mm, 
with minimum values below 800 mm. 

There is a great deal of annual and seasonal 
variability rainfall. The wet season, however, 
spans mid-December to the end of March, and 
the dry season runs from September to Novem- 
ber. 

The mean annual temperature is approxi- 
mately 23C. In the hot season (February), it 
rises to 26C. In August, during the cool season, 
the mean temperature drops to 19C. 

The soil moisture balance is negative. Only in 
July and August (the southern winter) is there a 
positive moisture balance. From September to 
November, the balance is distinctly negative. 

Types of Vertisols in 
New Caledonia 

Three groups of vertisols may be distin- 
guished by their morphological and chemical 
properties. The calcimagnesic vertisols are 
formed from limestone, flysch, or basalt. Their 
extent is more than 55,000 ha. The hypermag- 
nesic vertisols are formed from serpentinites or 
serpentinite colluvia associated with colluvia 
derived from peridotites. They cover about 
35,000 ha. The sodic and acid vertisols are 
formed in siliceous pelites or graywackes. Their 
extent is less than 10,000 ha. 

Morphological Features 

Tables 1 and 2 summarize the morphological 
features of these soils. 

Calcimagnesic Vertisols 

Salts accumulation generally occurs at depth 
of approximately 100 cm, within the vertic hori- 
zon (Vv). Commonly a substantial accumulation 
of gypsum crystals (Vgy) is observed. In places, 
gypsum accounts for more than 25% of the dry 
weight of the horizon. These crystals are lenticu- 
lar, commonly between 0.5 and 2 cm in size but 
sometimes larger (Podwojewski, 1984). 

In the vertisols located on flat areas, the Vgy 
horizon may be underlain by substantial dark 
gray or black accumulations of manganese 
(VMn). The thickness of this layer is commonly 



PODWOJEWSKI and BEAUDOU: VERTISOLS OF NEW C 



ALEDONIA 



153 



Exchangeable 



40 
m 



30 



20- 




, u k t 

or V horizons in the *,,qunce( in me%) 



Mill 



Omi% 

20 

-40 




High, alluvial 
terrace 



Recent 
alluvial 
terrace 



Presence of gypsum criatal* 



I Presence of carbonates 



I 



1 TAMO A 



river 




200m Total length of the sequence : 800m 



WtM. 



0,5- < 



1,0-J 



1,0- 



1,3-4 



vj>>^^^ 

sV^x^N 

>vN \v\vx\ 



w//////?, 



0,5 



1,5- 



2.0- 



2,3-1 



\ \ \ 



\\\ 



0.3 



X 



1,0- 
1,5 



2.0- 



2,3- 




A 



x \ \ 

X \ \ 



0.3- 



1,0- 



1,5- 



2,0 - 



03- 




2,0-1 



3A 



3B 



LEGEND: 



i 
Humiferous horizon - A 

Colored and structured 
horizon - 8 

Vertic horizon - 8v 



7^.:V7] Alluvial deposits 
' ' " Sandy-loam 



[\\\| Horizon of rock weathering -C 




Accumulation of gypsum 

Accumulation of carbonates 
B Ca - C Co \ 
8Mg - C Mg / 



J Accumulation of Mn oxydes 



3C 



2 - Vertic brown soil 

3A - Calcimagnesic vertisoi 
38 - Gypsic colcimagnesic 
3C- Hypermagnesic vertisol 
4 - Alluvial soil 



154 



SIXTH INTERNATIONAL SOIL CLASSIFICATION WORKSHOP 



Table 1 - Morphological Features in the Upper Part of the Profile 
of the Calcimagnesic (A), Hypermagnesic (B), and Sodic and Acid (C) Vertisol Types 


HORIZON 


TYPE DEPTH 
(cm) 


COLORS 


TEXTURE 


STRUCTURE 


PEDOLOGICAL 
ACCUMULATIONS 


A ! 

Surface 
horizon 

Surface 
degraded 
horizon 


A 0-30 

B 0-30 
C 0-10 


10YE 3/1 
very dark gray 

10YR 3/1 to 
10 YR 2/0 black 

10YR4/1 
dark gray 


Clay 

Clay 

Silt 
loam 


fine to medium 
angular blocky 

fine to medium 
angular blocky 
fine 
angular blocky 


few manganese concretions 
diameter < 0.5 cm 

few manganese concretions 
diameter < 0.5 cm 

many manganese concretions 
diameter 0.1 to 1.0 cm 
sometimes 10% of the volume 


A M 
Surface 
degraded 
horizon 


C 10-25 
only 


10 YR 3/1 
very dark gray 


Clay 
loam 


angular blocky 
medium 


few manganese concretions 
diameter 0.1 to 1.0 cm 


AB 

Intergraded 
horizon 


A 25-50 

B 30-50 
C 30-50 


often mixed 
colors 
(infilled cracks) 
10YR 3/2 very 
dark grayish 
brown 

2,5Y 3/2 very 
dark grayish 
brown 
10YR 3/2 to 4/2 
dark grayish 
brown 


Clay 

Clay 
Clay 


coarse angular 
blocky to 
coarse prismatic 

few pressure faces, 
few slickensides 
< 10 cm long 


sometimes nodules of CaCO3 
<2cm 
rare manganese concretions 
< 0.5 cm 

rare manganese concretions 
< 0.5 cm 

rare manganese concretions 
< 0.5 cm 


BV 

Vertic horizon 
low organic 
matter: 
infilled cracks 
or infilled 
worm channels 


A 50-70 
B 50-70 
C 50-70 


10YR 3/3 to 4/4 
dark brown 

2,5Y4/2to4/4 
dark grayish 
brown 
7,5YR 3/4 to 4/4 
dark brown 


Heavy 
Clay 


parallelepiped 
structure, 
intersecting 
slickensides: 
20-40 from 
horizontal 
> 15 cm long 


rare manganese concretions 
< 0.5 cm 

rare manganese concretions 
< 0.5 cm 

rare manganese concretions 
< 0.5 cm 



between 10 and 
15 cm. The man- 
ganese oxides 
occur in the form 
of dendrites and 
cutans, and also 
in the form of 
nodules of vary- 
ing degrees of 
cementation. 
The manganese 
oxide contents do 
not exceed 5% of 
the weight of the 
dry soil. It is 
likely that the 
presence of these 
elements is due 
to the existence 
of a former water 
table. The gyp- 
sum crystals in 
and under this 
VMn horizon are 
agglomerated 
and form "sand 
rose" type forma- 
tions. The gyp- 
sum crystals then disappear fairly rapidly with 
depth. 

Calcium carbonates occur below the gypsum. 
They gradually appear in the form of "powdery 
volumes," very small pockets, nodules, and 
crusts of varying degrees of cementation. Car- 
bonate contents increase with depth. 

Hypermagnesic Vertisols 

These vertisols are similar to calcimagnesic 
vertisols with respect to: 

- the occurence of carbonates below the VMn 

horizons in varying forms, from very small 
pockets to crusts, depending on depth 

- the presence, at depth, of a horizon where 

manganese oxides have accumulated. 
They differ in these ways: 

- total absence of gypsum crystals 

- presence of accumulations of magnesium 

carbonates (magnesite or giobertite), 
which are rarely present in continuous 
form (crusts). 

Other differences may also be noted but these 
are often much less marked. Hypermagnesic 
vertisols appear blacker at the surface (10YR2/1 



to 2.5Y2/0) than the calcimagnesic vertisols and 
the sodic and acid vertisols. The vertic horizons 
of the hypermagnesic vertisols are more olive 
brown in color (2.5Y4/4 to 5/6) than the calci- 
magnesic vertisols (10YR4/4 to 5/6). 

Sodic and Acid Vertisols 

Their essential features occur in the surface 
horizon, which is grey in color (10YR4/1 or 5/1) 
and in texture is much less clayey (silt loam or 
loam). It also often includes abundant concre- 
tions of manganese oxide ("pellets"). 

The vertic horizons possess a less distinct 
wedge-shaped structure. The cohesion is very 
strong. The coloring is usually brighter (7.5YR 
to 5YR4/4 to 5/6). 

The sodic and acid vertisols lack sulfate con- 
centrations. Carbonate accumulations are few. 
They are calcic and magnesic, occurring in the 
form of highly friable nodules and pockets. 

The accumulation of manganese oxides at 
depth is less marked. These oxides occur in the 
form of nodules, dendrites, and very small pock- 
ets. 



PODWOJEWSKI and BEAUDOU: VERTISOLS OP NEW CALEDONIA 



155 



Table 2 - Morphological Features in the Lower Part of the Profile 
of the Calcimagnesic (A), Hyperznagnesic (B), and Sodic and Acid (C) Vertisol Types 


HORIZON 


TYPE DEPTH 
(cm) 


COLORS 


TEXTURE 


STRUCTURE 


PEDOLOGICAL 
ACCUMULATIONS 


CONSISTENCE 


Vvor(B) 


A 90 -150+ 


10 YR 4/4, 5/4, 
5/6 yellowish 
brown 


Heavy 
Clay 


parallelepiped 
structure, 


rare manganese concretions 
< 0.5 cm 
rare soft carbonate 


plastic (wet) 
sticky 
firm (moist) 


VGyor(B)Gy 
NOT ALWAYS 

PRESENT 


A 90 -150+ 
only 


many intersecting 
slickensides: 
20-40 from 
horizontal 
> 15 cm long 


lenticular gypsum crystals 0.5 to 2.0 cm 
long, rare 5.0 cm sometimes > 20% weigh 
of air dried soil 
In VMn horizon and below: agglomerate 
crystals into spheric accumulations 2 to 
15 cm diameter 


VCaor(B)Ca 
Under V Gy 
When V Gy 
is present 


A 90 -150+ 
frequent 

C 

rare 


In the upper part of the horizon: soft 
carbonate in spheroid volumes 2-10 cm 
diameter. In the lower part: little soft 
concretionary carbonate; many nodules 
1-3 cm diameter. 


Vvor(B) 


B 90-150+ 


2,5 Y 4/4, 5/4, 
5/6 light 
olive brown 


Heavy 
Clay 


parallelepiped 
structure, 
many intersecting 
slickensides: 
0-40' from 
horizontal 
> 15 cm long 


are manganese concretions 
< 0.5 cm 
rare soft Mg carbonate 


plastic(wet) 
very sticky 
firm (moist) 


VMgor(B)Mg 


B 90-150+ 
frequent 


In the upper part of the horizon: soft 
Mg carbonate in spheroid volumes 2-10 cm 
diameter. In the lower part: little soft 
concretionary carbonate; many white 
nodules 1-3 cm diameter of magnesite 


V v or (B) 


C 90-150+ 


7,5 YR 4/4, 4/6 
5 YR 4/4, 4/6 
strong brown 
to yellowish 
red 


Heavy 
Clay 


parallelepiped 
structure, < 0.5 cm 
intersecting 
slickensides: 
20-40 from 
horizontal 
> 15 cm long 


rare manganese concretions 
rare soft Ca carbonatee 


slightly 
plastic (wet) 
slightly 
sticky 
very firm 
(moist) 


VMn 
only in old 
alluvial 
terraces 


AorB 
depth varies, 
10 to 15 cm 
thick 


black 


Heavy 
Clay 


same structure 
as in Vv for type 
A or B soils 


manganese concretions and manganans 
on the slickensides and around the pores 
(on tubular and planar voids) 



Analytical Features 

Table 3 presents the chemical analysis of 
these soils. 

Mineralogy 

The calcimagnesic vertisols consist of well 
crystallized montmorilionite (with a little 
quartz).The hypermagnesic vertisols feature the 
presence of ferriferous smectite of the bowlingite 
variety (Latham & al, 1978), lacking alumin- 
ium. The degree of crystallinity appears to be 
higher (clearer peaks) than in other vertisols. 

The sodic and acid vertisols consist of 
montmorilionite associated with interstratified 
illite and montmorilionite. 

Texture 

All the vertic horizons have a clay content 
over 50%. However, the presence of carbonates 
and manganese oxides brings about a relative 
decrease in the clay content of these horizons. 

The surface horizons are generally less 
clayey. In the sodic and acid vertisols, the im- 
poverishment in the clay content (or leaching) at 
the surface is very evident. However, this clay 
does not redistribute itself at depth in the form 



of argillans in an accumulation horizon (Denis, 
Mercky, 1979). 

pH 

The surface horizons of all vertisols in the 
study area have a pH between 5.5 and 6.0. At 
depth, the values vary, depending on the type of 
vertisol. 

Calcimagnesic vertisols with high salt accu- 
mulations have a pH which varies with the na- 
ture of these accumulations: 5.5 where gypsum 
is present, 7 to 8 where carbonates are present. 

The hypermagnesic vertisols, lacking sul- 
fates, but rich in magnesium carbonates at 
depth, feature regularly increasing pH values 
which may reach or exceed 8 in the giobertite 
(magnesite) horizons. 

The sodic and acid vertisols have a pH lower 
than 6.0 and in some places even less than 5.0. 
Under such conditions, exchangeable alumin- 
ium may be present. 

Exchangeable Bases and Exchange Capacity 

The method for extracting exchangeable 
bases recommended by Tucker (1985 b.) using 
ammonium chloride (at pH 7 in Noumea) has 



156 



SIXTH INTERNATIONAL SOIL CLASSIFICATION WORKSHOP 



Table 3 - Chemical Analysis of the Calcimagnesic (A), Hypermagnesic (B), 
and Sodic and Acid (C) Vertisol Types 


TYPE OF VERTISOL A B C 
HORIZON Aj Aj A u 
DEPTH (cm) 0-30 0-30 0-12 


C 

A. 
12-35 


ABC 
AB AB AB 
40-60 40-60 40-60 


A B C 
BGy BMg B 
100-140 80-120 80-100 


C 

BCa 
195-240 


GRANULOMETRY% 
Clay 46.5 44.6 24.0 
Fine silt (0.002-0.02 mm) 21.8 17.6 30.0 
Coarse silt (0.02-0.05mm) 11.0 9.2 22.5 
Fine sand (0.05-0.2mm) 10.4 16.7 16.5 
Coarse sand (0.2-2mm) 5.6 7.4 3.5 


37.6 
20.5 
11.2 
17.3 
12.7 


47.6 59.2 53.9 
19.5 10.6 18.5 
9.7 6.1 9.9 
11.4 11.8 16.3 
11.0 9.3 2.0 


70.9 54.4 49.3 
12.1 11,4 17.2 
7.6 4.0 10.0 
4.9 14.7 21.8 
5.8 13.8 2.0 


64.4 
11.8 
5.7 
5.9 
10.5 


MOISTURE (g/lOOg) at pF 
2,5 41.0 47.4 33.7 
4,2 23.9 29.8 12.2 


31.7 
16.4 


42.7 62.5 39.2 
23.3 40.3 22.3 


46.2 54.3 39.2 
26.3 35.9 18.5 


44.0 
28.2 


ORGANIC MATTER g/lOOg * 
C 3.04 2.13 2.337 
N 0.184 0.182 0.141 
Organic Matter 5.2 3.7 4.0 
C/N 16.5 11.7 16.6 


0.847 
0.100 
1.5 
8.5 


0.84 0.85 0.489 
0.073 0.070 0.074 
1.4 1.5 0.8 
11.5 12.1 6.6 






pHH 2 1:2.5 5.9 6.2 5.3 
pHKCl 4.8 5.0 4.7 


5.1 
4.2 


6.2 7.6 4.5 
5.1 6.2 3.6 


5.8 8.4 4.6 
4.8 6.8 3.7 


8.2 

7.1 


EXCHANGE COMPLEX cmol(p+)/kg * 
Ca~ 12.1 1.8 4.4 
Mg* 15.0 40.20 3.8 
K* 0.84 0.08 0.2 
Na+ 0.65 0.48 0.34 
I Cations 28.59 42.56 8.74 
C.E.C. 36.2 49.0 17.7 
Base saturation % 78.9 86.9 49.4 


4.3 
7.3 
0.1 
2.9 
14.5 
21.0 
69.0 


11.2 0.23 3.6 
16.2 66.90 10.5 
0.11 0.04 0.14 
2.8 1.30 4.7 
30.31 68.47 18.94 
34.4 61.0 23.4 
88.1 SAT. 80.9 


11.9 0.34 3.5 
23.6 59.20 10.6 
0.14 0.03 0.12 
4.69 1.40 5.1 
40.33 60.97 19.32 
41.6 57.9 22.5 
96.9 SAT. 85.9 


19.4 
28.2 
0.12 
6.16 
53.88 
46.2 
SAT. 


Total phosphorus 
P 2 B mg/kg 480 430 


420 


130 150 






PERCHLORIC TOTAL ANALYSIS g/lOOg # 
Loss on ignition 10.44 10.57 6.6 
Residue 38.96 54.98 59.4 
Si0 2 32.68 28.86 25.0 
A1 2 S 6.31 2.87 3.6 
Fe 2 0, 6.35 22.88 2.4 
Mn0 2 0.58 1.00 0.77 
TiO a 0.87 0.18 0.43 
CaO 0.48 0.19 0.15 
MgO 0.76 4.22 0.23 
KjO 0.08 0.02 0.13 
Na 2 0.08 0.04 0.10 
SiO/Al 2 O 3 mol. 8.8 17 11.9 
Cr 2 3 2.32 


6.9 
47.9 
26.6 
9.1 
6.3 
1.5 
0.60 
0.09 
0.47 
0.20 
0.16 
5.0 


7.72 9.47 6.6 
41.30 56.46 45.4 
29.60 38.26 28.7 
9.07 2.65 11.7 
7.35 22.02 6.0 
1.24 1.23 0.05 
0.84 0.16 0.72 
0.39 0.10 0.07 
0.84 5.34 0.63 
0.06 0.01 0.30 
0.16 0.06 0.29 
5.5 24.5 4.2 
1.11 


7.74 7.94 5.5 
28.40 61.94 49.6 
34.00 39.80 26,9 
11.15 1.97 9.8 
7.58 17.16 5.6 
0.04 0.42 0.03 
0.86 0.10 0.67 
4.20 0.13 0.07 
1.22 9.07 0.62 
0.11 0.01 0.36 
0.32 0.07 0.34 
5.2 34.2 4.7 
0.52 


8.43 
29.14 
35.42 
10.58 
8.72 
0.22 
0.98 
2.94 
2.58 
0.15 
0.45 
5.7 


Mg/Ca 1.2 22.3 0.86 
Na/CEC % (ESP) 1.8 1.0 1.9 


1.7 
13.8 


1.4 91 2.9 
8.1 2.1 20.1 


2.0 174 3.0 
11.3 2.4 22.6 


1.4 
13.3 


* Exchangeable bases on vertisols of type A have been extracted by B.M. Tucker's methodology: Ammonium chloride at pH 7.0 
# g/lOOg of dried soil at 105C 



proved the most suitable for the vertisols of New 
Caledonia. This method restricts the influence 
of soluble salts in the extraction of exchangeable 
bases, mainly in the gypsum and carbonate ho- 
rizons. 

Calcimagnesic vertisols have an exchange ca- 
pacity that is approximately 40 me/lOOg soil 
(see figure 4). They are base saturated in the 
vertic horizons. The Mg/Ca ratio is higher than 
1 but rarely exceeds 2. Magnesium remains the 
dominant cation even when gypsum is present. 
The deep horizons are relatively richer in ex- 
changeable sodium. 

Hypermagnesic vertisols have a higher ex- 
change capacity (50 and 60 me/lOOg soil), which 
might well be due to the higher clay contents 



and better crystallization of the smectites. The 
vertic horizons are base saturated. The Mg/Ca 
ratio is higher than 4.5 and increases with 
depth. The proportion of Mg++ in relation to the 
sum of exchangeable cations is higher than 80%. 
The deep horizons are also high in exchangeable 
sodium. 

Sodic and acid vertisols have an exchange ca- 
pacity which is relatively low in the surface hori- 
zons (15 to 20 me/lOOg soil) but increases a little 
with depth at the same time as the clay content, 
reaching 30 to 35 me/lOOg. The base saturation 
of the vertic horizon is slightly lower than that 
for other vertisols (80 to 90%). The exchange- 
able sodium content is often greater than 10 or 
more. The Mg/Ca ratio increases considerably 



PODWOJEWSKI and BEAUDOU: VERTISOLS OP NEW CALEDONIA 



157 



Figure 3. Amounts of exchangeable Ca++, MG++, 
and Na+ in the three main groups of Vertisols 



10% No 




LEGEND: 

Calcimagnesic vrtisolf 
Hyprmagngic vtrtiol* 
Sodic and acid vtrtisolt 




from the surface (mean value of 2) toward the 
deeper horizons, where it often exceeds 5. 

In summary, the following are very important 
in the characterization of the various types of 
vertisols in New Caledonia: 

1. Magnesium is the dominant cation in the 
vertic horizons, including calcimagnesic 
vertisols with accumulations of calcium 
salts (gypsum, lime). 

2. The exchangeable calcium content is rela- 
tively higher in the surface horizons than 
in the vertic horizons. 

3. The exchangeable sodium content in- 
creases rapidly at depth in the vertic hori- 
zons. A considerable proportion of the so- 
dium, in all probability, can be attributed 
to sea spray and aerosols. 

4. The exchangeable potassium content is 
very low. Only rarely, however, is it com- 
pletely absent. 

Other Elements 

The total phosphorus contents are very low, 
usually less than 800 ppm. The available phos- 
phorus contents (Olsen method, modified by 
Dabin, 1967) are less than 20ppm. (Hypermag- 
nesic vertisols developing from ultrabasic rocks 
are very rich in chromite, an element which 
masks phosphorus when the colorimetric 
method is applied. Use of the X-ray fluorescence 



method, however, also reveals very low phos- 
phorus contents [Latham, 1986].) 

Total analyses show a SiOJAlJ) 3 molecular 
ratio which is always above 4. This ratio may 
reach much higher levels in vertisols derived 
from serpentinites, which are rocks totally lack- 
ing aluminium (hypermagnesic vertisols). The 
mean total iron contents mostly vary between 8 
and 10% but reach 20% in the hypermagnesic 
vertisols. Of the alkaline and alkaline-earth ele- 
ments, magnesium is in most cases the most 
abundant (with the exception of soils having de- 
veloped from carbonate-rich flysches and lime- 
stones). Sodium is often more abundant than 
calcium. It should be noted that, in the horizons 
which show traces of manganese accumulation, 
this element, on average, reaches contents of 2 
to 3% and even, in infrequent cases, 10%. The 
vertic horizons of the hypermagnesic vertisols 
which derive from peridotites and serpentinites 
are enriched in chromium, nickel, and cobalt. 

Discussion and Conclusion 

The data relating to exchangeable bases and 
the exchange capacity suggest two means of 
classification for vertisols in New Caledonia. 

The first would be based on the exchangeable 
Mg+VCa""" ratio. Using this method we may dis- 
tinguish between: 

- calcic vertisols: the Mg+VCa 4 * ratio is less 
than 0.5 in the vertic horizons. 

- calcimagnesic vertisols: the Mg+VCa^ ratio 
is between 0.5 and 2.0 in the surface hori- 
zon and rises to 3.0 in the vertic horizons. 

- magnesic vertisols: the Mg^/Ca** ratio is be- 

tween 3.0 and 5.0 in the surface horizon 
and rises to 10.0 in the vertic horizons. 

- hypermagnesic vertisols: the Mg^/Ca^ ratio 

is higher than 5.0 in the surface horizon 
and exceeds 10.0 in the vertic horizon. 
The second method of classification is based 
on the ratio of exchangeable Na + percentage 
(ESP) measured in the vertic horizons (depth 
greater than 60 cm). It is thus possible to distin- 
guish among: 

- non sodic vertisols: ESP is less than 8.0 

- moderately sodic vertisols: ESP is between 

8.0 and 15.0. 

- sodic vertisols: ESP is greater than 15.0. 

It is thought that these vertisols were proba- 
bly formed under the influence of a paleoclimate 
much dryer than the current climate, because 
the accumulations of gypsum are unlikely to 
have developed in the currently prevailing 



158 



SIXTH INTERNATIONAL SOIL CLASSIFICATION WORKSHOP 



weather conditions in New Caledonia. 

The hypermagnesic vertisols are subject to 
the standard use constraints typical of all verti- 
sols and also to those due to the major cationic 
imbalance of the exchange complex which is 
marked by the distinct predominance of magne- 
sium. This magnesium surplus leads to calcium 
deficiencies. Amendments in the form of gyp- 
sum make it possible to restore the balance in 
these vertisols (Beaudou et aL, 1984) and sub- 
stantially increase yields (Bonzon, 1986, 1988). 

Literature Cited 

A.F.E.S., 1988. Referential Pedologique Francais, 2nd Ed., 
INRA, 251 p. 

Beaudou, A.G., M. Fromaget, P. Podowojewski, E. Bour- 
don, H. Le Martret, and D. Blavet. 1983. Cartographic 
typologique des sols: methodologie, Orstom, Noumea, 
31 p. 

Beaudou, A.G., H. Le Martret, and B. Denis, 1984. Magne- 
sian soils of New Caledonia. Experiment to restore the 
balance of the base exchange complex. South Pacific J. 
of Natural Sc., vol.7, 78-99. 

Bonzon, B., et al. , 1986. Effet des amendements calciques 
sur un sol sodique acide et sur un vertisol hyper- 
magnesien, Orstom, Noumea, 36 p. 



Bonzon, B. et al. , 1988. Effet des amendements calciques 

sur vertisol hypermagnesien, Orstom, Noumea, 69 p. 
Dabin, B., 1967. Application des dosages automatiques a 

1'analyse des sols. Cah. Orstom, ser. Pedol., vol. V, no. 

3. 
Denis, B., and P. Mercky, 1982. Etude pedologique de la 

region de Pouembout, Orstom, Noumea, Tome 1:150 p., 

Tome 2:109 p. 
Latham, M., P. Quantin, and G. Aubert, 1978. Etude des 

sols de Nouvelle Caledonie, Orstom, Paris, Not. Expl. 

no. 78, 138 p. 
Latham, M. 1986. Alteration et pedogenese sur roches ul- 

trabasiques en Nouvelle Caledonie, Orstom, Paris, Et. 

& Theses, 331 p. 
Podowojewski, P. 1984. Les sols de Nouvelle Caledonie a 

accumulation de Gypse, Orstom, Noumea, 23 p. 
Podowojewski, P. and A.G. Beaudou, 1987. Carte mor- 

phopedologique de la Nouvelle Caledonie a 1/200.000, 

Orstom, Noumea, 3t. 
Tercinier, G. 1953. Sols et Terres de la Nouvelle Caledonie 

et autres territoires Francais du Pacifique. Seme 

congres Pac. Sci., Manille, 1953. 
Tucker, B.M. 1985 a. Active and exchangeable cations in 

soils, Aust. J. Soil Res. 23, 195-201. 
Tucker, B.M. 1985 b. Laboratory procedures for soluble 

salts and exchangeable cations in soils, CSIRO, Soils 

Techn. Pap., 47 p. 



Salinity Development, Recognition, and Management 

in North Dakota Soils 

J.L. Richardson 1 , D.G. Hopkins, B.E. Seelig, and M.D. Sweeney 

Abstract 

The field recognition of salinity is complicated because soluble salts are 
expressed morphologically only when the soil is very dry. Salinity can be as- 
sessed directly by morphological observation. Vegetation indicators are 
used, but they introduce a degree of unreliability. Plant indicators vary be- 
cause of moisture conditions, time of year, and plant species. A significant 
problem in both soil survey and subsequent management of these saline and 
sodic areas is recognizing the intensity of the problem. In sodic map units, 
recognition of soil variability is problematic because of the scarcity of soil 
chemical data and variations in landscape expression in different sodic map 
units. In western ND more saline and sodic soils occur on terraces and 
fioodplains. Also saline seeps on erosional footslopes are important. In 
eastern ND the salinity is associated with pond edges in Calciaquolls. A 
district system of evaporite sequences occur from calcite, then gypsum and 
lastly efflorescent crusts. Freezing also concentrates salt and changes the 
salt type. Most salts in ND are sulfates; on freezing Na 2 SO 4 H 2 O precipitates 
more easily than MgSO 4 H 2 O. 



Introduction 

North Dakota has two distinct geologic re- 
gions: 1) Quaternary terrains of till, outwash, 
and lacustrine sediments; and 2) Tertiary sedi- 
ments over Cretaceous sands and shales. The 
Quaternary deposits occur often as closed drain- 
age systems. They do not have outlets; we often 
call these areas "prairie potholes." Salinity as- 
sociated with closed or partially closed drainage 
systems occur mostly in three major physiogra- 
phic provinces in North Dakota: 1) Drift Prairie; 
2) Missouri du Coteau; and 3) Glacial Lake 
Agassiz. Prairie potholes have been drained 
extensively throughout the region with open 
ditches constructed as outlets for waterflow. 
Salinity problems are abundant because of the 
youthful nature of the landscape and shallow 
water tables. Many of the saline soils are Cal- 
ciaquolls that are essentially soils with eva- 
porite deposits that have formed Bk horizons. 

In the southwestern part of the state, the re- 
sidual soils formed in bedrock sediments that 
have open-erosional landscapes with abundant 
streams and few natural wetlands. Salinity is 
associated with sodicity in footslopes and 
toeslopes. Due to the practice of summer fallow- 
ing and the stratified nature of the Tertiary 
sediments, saline seeps are common. 



. Soil Science, North Dakota State Univ., Fargo, 
ND 58105. Published with the permission of the Director 
of the North Dakota Agric. Exp. Sta. Corresponding au- 
thor. 



This paper outlines salinity occurrence in 
North Dakota by region and discusses manage- 
ment and recognition of these soils. 

Recognition of Salinity for Soil 
Survey 

The exception is gypsum, which precipitates 
from solution at concentrations of about 2 g/L' 1 
or 2 dS/nr 1 . At 20 C sodium sulfate precipitates 
approximately at 200 g/L- 1 or > 200 dS/nr 1 , 
magnesium sulfate at > 300 g/L" 1 or > 300 dS/ 
nr 1 , and sodium chloride at > 375 g/Lr 1 or > 375 
dS/nr 1 (Timpson et al, 1986). Such extreme sa- 
linity levels are reached only in surface crusts or 
in rare circumstances. 

In recent years, electromagnetic induction 
devices have improved our confidence in deline- 
ating saline soils (Richardson and Patterson, 
1986; Wollenhaupt et al., 1986). 

In North Dakota, strong (> 16 dS/nr 1 ) salinity 
is recognized at the series level. Slight and 
moderate salinities (4 to 16 dS/nr 1 ) are recog- 
nized as saline phases. For example, in Grand 
Forks County (Doolittle et al, 1981), Bearden, 
Bearden saline, and Ojata represent increasing 
salinity from very slight to strong. The mem- 
bers of the soil survey team in this county were 
uncertain about saline soil mapping unit deline- 
ations, but examination of the field delineations 
demonstrated that the units were accurate 
(Richardson and Patterson, 1986). The deline- 
ations were tested with an electromagnetic in- 
duction meter, the Geonics EM-58 that has been 



159 



160 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



used throughout the northern Great Plains 
(Wollenhaupt et al., 1988). 

Hopkins et al. (1987) found forage yield esti- 
mates on sodic rangeland were vastly underesti- 
mated because a Leptic Natriboroll was over- 
mapped due to the striking visual impact of bar- 
ren panspots on the soil landscape. 

Recent work in North Dakota has shown that 
saline seep formation resulting from agricul- 
tural management is actually a form of deserti- 
fication (Timpson and Richardson, 1986). 

Classification of Saline and Sodic 
Soils 

Sodic soils with significant accumulations of 
soluble salts are classified as saline-sodic (Salin- 
ity Lab Staff, 1954). To meet the saline-sodic 
criteria, the soil saturation extract must have 
an electrical conductivity (EC) > 4 dS/nr 1 and 
soil exchangeable sodium percentage (ESP) > 
15. Saline-sodic soils are often flocculated due to 
high soil solution EC and may physically re- 
semble saline soils. Some saline Calciaquolls 
meet the requirements for saline-sodic soils but 
unfortunately have not been separated from sa- 
line Calciaquolls. Sodic soils by contrast have 
morphological differences due to clay dispersion 
by sodium. Some sodic soils with high amounts 
of dispersible clay, such as Leptic Natriborolls 
and Typic Natraquolls, also meet the criteria for 
saline-sodic soils. In North Dakota a different 
definition for sodic-saline soils has been used. 
Saline-sodic soils must have a dense subsoil ac- 
companied by a saturation extract EC greater 
than 2 dS/nr 1 and ESP greater than 15. 

Recognition and delineation of saline-sodic 
soils is complicated by variability in sodicity and 
salinity. Seelig and Richardson (1989) found 
dispersible clay differences between sodic soils 
to be related to landscape position. We found 
that identification of landscape position and soil 
moisture regime was necessary to predict sa- 
line-sodic properties. 

Landscape and Saline 
Relationships 

Salinity In Erosional Topographies Of 
Western North Dakota 

Of fundamental concern is the recognition 
that distribution of saline soils is governed by 
landscape position. Saline or sodic soils develop 
in open or closed geomorphic systems wherever 
convergent water flow concentrates salts. Thus 



soils in a drainage basin are related by the 
transfer of chemical constituents and can be 
thought of as representing a "single geochemical 
landscape" as pointed out by Huggett (1975). 
Recently, research emphasis has shifted to the 
accelerated rate of secondary soil salinization 
caused by cultural management (Halvorson and 
Black, 1974; Doering and Sandoval, 1978; 
Ferguson and Bateridge, 1982; Timpson and 
Richardson, 1986). 

In western North Dakota, saline and sodic 
soils occur in three distinct landscape positions. 
The largest areas of saline and sodic soils are 
found in broad flats of floodplains and terraces. 
Saline groundwater is often shallow and moves 
upward into the soil root zone by capillary rise. 
Saline phases of Torrifluvents and Ustifluvents 
and the Natrargids, Natraquolls, and Natribor- 
olls are found in these landscapes. 

Saline seeps frequently occur on upper pedi- 
ment footslopes. Here, small scale stratigraphic 
controls result in seepage at the slope break 
associated with the backslope. Panspots, a 
ubiquitous surface feature found in sodic land- 
scapes, are related by a similar process. They 
generally merge with adjacent upland soils at a 
scarp face resembling a micro-pediment back- 
slope. The higher salinity in the panspot soils is 
likely the result of throughflow seepage caused 
by structural and/or textural control on permea- 
bility in the adjacent upland soils (Murphey and 
Daniel, 1935; Bowser et al., 1962; Hopkins et 
al., 1987). 

The most extensive landscape position is on 
lower erosional footslope positions. The major- 
ity of Natrargid and Natriboroll soils are found 
in these landscape positions (Rosek and 
Richardson, 1989) over extensive areas. Sea- 
sonal throughflow is vertically controlled above 
relatively impermeable substratum and con- 
verges in lower footslope positions either as 
perched shallow groundwater or as seepage 
(Hadley and Rolfe, 1955). 

Salinity In Till And Lacustrine 
Landscapes 

Numerous wetlands and wet soils result from 
the lack of an integrated stream flow network in 
the Prairie Pothole Region and in the Lake 
Agassiz Physiographic Province. Groundwater 
movement (Miller et al., 1985; Arndt and 
Richardson, 1988) controls many of the proc- 
esses of soil development. Therefore, recogni- 
tion of the pathways of water movement are 
important in understanding soil development 



RICHARDSON, HOPKINS, SEELIG, AND SWEENEY: SALINITY/IN NORTH DAKOTA SOILS 



161 



and in particular salinity (Bigler 
and Richardson, 1984; 

Richardson and Bigler, 1984). 

Several studies have helped 
our understanding of the ground- 
water movement in these sys- 
tems. Lissey (1971), in particu- 
lar, introduced the notion of de- 
pression-focused recharge and 
discharge (Fig. 1). In these areas, 
water flows to the groundwater 
only in the smaller depressions 
(recharge). Water flows from 
groundwater to the surface in 
lower wetlands (discharge). The 
wetlands or prairie potholes are 
"windows" of the groundwater at 
the earth's surface. The dis- 
charge points become progressively more saline. 
Note that the water tends to flow laterally in the 
groundwater system as suggested by Lissey 
(1971). Such water flow systems are expected in 
the subhumid and drier climate zones. 

MacLean and Pawluk (1975) observed lateral 
flow in relatively open systems and correlated 
salinity and groundwater. They also observed 
desalinized-sodic systems that were apparently 
due to lateral flow above the water table. Mills 
and Zwarich (1986) and Winter (1986) observed 
lateral flow in a variety of prairie pothole land- 
scapes. They noted many seasonal flow rever- 
sals complicating the flow system. 

Salinity As Evaporite Deposits Associated 
With Wetlands 

Knuteson et al. (1989) noted that depression- 
focused recharge affected soil development in 
lacustrine landscapes. A Calciaquoll is expected 
to form around ephemeral wetlands, but leached 
soils occur in them. Upward movement of capil- 
lary is dominant water at the pond edge, even 
though saturated flow may reverse around 
many ponds. Unless a groundwater discharge 
area is encountered, the salt content remains 
low. Calcite is the main evaporite concentrated 
in the Calciaquoll soils on the margin around 
recharge wetlands. Saline soils occur on the 
margins of discharge wetlands that have con- 
centrations of evaporites more soluble than cal- 
cite (Stein wand and Richardson, 1989; Arndt 
and Richardson, 1989). 

The combination of discharge and evapo- tran- 
spiration create an "edge effect" in which eva- 
porites accumulate (Fig. 2). This observation 
was earlier noted by Whittig and Janitzky 



SIMPLIFIED/ / : - 




PRAIRIE 

POTHOLE 

SYSTEM 



LOCAL GROUNDWATER FLOW 

In 

FRACTURED X *"* 
TILL 



Figure 1. A simplified flownet with equipotential lines in meters for the 
Prairie Pothole region* Note that water table is mounded under 
wetlands and does not follow the topography well (Modified from 
Lissez, 1971). 



(1963) in California. They found saline and 
sodic conditions similar to those in North Da- 
kota wetlands, but their warmer environment 
with low sulfates produced an evaporite se- 
quence dominated by Na^CO^. In North Dakota, 
the observation of Na 2 C0 3 is unusual. Abun- 
dant sulfate salts probably result from oxidation 
of reduced sulfur in organic matter or pyrites 
that were originally in marine shales (Hendry et 
al., 1986; Mermut and Arshad, 1987). 

The Hardie and Eugster (1970) system of 
closed basin brine evolution works well to ex- 
plain the distribution and type of salts in saline 
soils from North Dakota (Skarie et al., 1986, 
1987; Arndt and Richardson, 1989) (Fig. 3). In 
North Dakota, sulfatic evaporites are the domi- 
nant source of salinity. The first evaporites to 
form are magnesium calcites in the northern 
Great Plains (St. Arnaud and Herbillon, 1973; 
Knuteson, 1985). 

With a binary salt, such as calcite, either the 
anion or the cation phase will increase in the soil 
solution by the precipitation of the salt. In the 
till soils, continued precipitation of calcite 
caused by evapotranspiration concentrates dis- 
solved Ca 2 * and depletes soluble C0 3 2 ~. This is 
an illustration of Hardie and Eugster's (1970) 
chemical divides. Note that in outwash areas 
HCO 3 2 - was observed to be more abundant than 
Ca 2+ , and the pH usually exceeded 8.4. In 
warmer climates soil solutions would not hold 
the CO 2 in solution as well as in North Dakota. 
This may be another reason for the differences 
between North Dakota wetland edges and the 
edges observed by Whittig and Janitzky (1963) 
in California. 



162 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



FLOW-THROUGH WETLAMD 



Last and Schweyen (1983) 
noted several alkaline lakes with 
high carbonate. In these lakes 
the evaporites contained proto- 
dolomite. In one alkaline soil in 
North Dakota (maximum pH 
9.3), dolomite was observed as 
well Two other cases of dolomite 
in soils of the northern Great 
Plains have been reported (Sher- 
man et al., 1962; Rostad, 1975). 
Therefore, it appears that, in 
soils with more CO 3 2 " than Ca 2+ , 
the precipitation of calcite leads 
to high pH levels and dolomite. 
The conditions of dolomite forma- 
tion require very high levels of 
Mg 2+ with respect to Ca 2+ as well 
as elevated pH levels. This is one 
of the "chemical divides" of Har- 
die and Eugster (1970). 

Keller et al. (1986) observed 
that a few salt efflorescent eva- 
porites contained burkeite (Na 6 (CO 3 ) 2 S0 4 ) and wand and Richardson, 1989). Gypsum is the 
tychite (Na 6 Mg 2 (CO J SO 4 ). These are "alkaline" mineral that has been observed to precipitate 




HAPLAQUOLL 



CALCIAQUOLL 



8 C A L S 



Figure 2. An example of the evaporite deposition at pond edges in the 
Prairie Pothole Region, Numbers are for gypsum by weight. Average 
EC was > 16 mmhos/cm (after Steinwand and Richardson, 1989). 



evaporites and would be the equivalent of the 
sodium carbonates of Whittig and Janitzky 
(1963). The alkaline evaporites have been ob- 
served to occur only in coarse textured soils; 
rapid degassing of CO 2 may contribute to their 
formation. 

The more typical evaporite condition, how- 
ever, contains more Ca 2+ than HCO (Skarie et 



under these conditions; the Hardie and Eugster 
model predicts this observation. The ubiquitous 
nature of gypsum in saline and sodic soils of 
North Dakota indicates that the soil solution is 
saturated with respect to gypsum. It also indi- 
cates that gypsum is an ineffectual reclamation 
material in high sodic saline soils. As gypsum 
forms, the Ca 2+ is depleted from the soil solution 



Na, Ca, Mg, HCO 3| SO-., Cl 







2/77 c.*> Alkalinity 



alcite Precipitates 



al. 1987; Arndt and Richardson, 1989; Stein- (Skarie et al., 1987 and Arndt and Richardson, 

1989). The net effect is that sodium 
and magnesium sulfates concen- 
trate in the soil solution. These 
salts are generally the most abun- 
dant saline soils from North Da- 
kota. They usually precipitate only 
in soil surfaces in highly concen- 
trated forms as efflorescent crusts 
(Keller et al., 1986). 

Most saline soils from North Da- 
kota do not have precipitates of sol- 
uble "salts." Frequently soil sur- 
veyors observe "salts" in the soil as 
opposed to on the soil. These salts 
are most likely gypsum and may or 
may not indicate saline conditions. 



Na, Ca, Mg, SO,, Cl 





Gypsum Precipitates 

"so-:> " 2/J7 M 8 3 *> Alkalinity 




^epiolite Precipitates 

Alkalinity>2m 
Ilf 



'MO** 



After Drever, 1982 

Figure 3. Hardie and Eugster (1970) chemical divides model for closed 
basin brine evolution. The evaporites of the Prairie Pothole con- 
form well to the model even though our system is open. 



Salinity and Freezing 

Another factor in northern soils 
is the effect of freezing on salt con- 
centration. Timpson et al (1986) 



RICHARDSON, HOPKINS, SEELIG, AND SWEENEY: SALINITY/IN NORTH DAKOTA SOILS 



163 



and Keller et al (1986) noted the temperature 
change effects on sulfate minerals. Arndt and 
Richardson (1986, 1989) observed that ice on 
lakes contained the sodium sulfate mineral, 
mirabilite, and the lake water was concentrated 
with magnesium sulfate. Beke and Palmer 
(1989) observed mirabilite in frozen soils of 
southern Alberta. 

Freezing can segregate sodium from magne- 
sium in sulfate rich saline soils. This may in- 
crease Mg 2 * on the exchange complex in the fall 
and winter. On warming, a flush of ice-freed 
sodium will dissolve into the soil solution and 
can temporarily create a mass action exchange 
with soils. The freezing action in combination 
with ion pairing (Alzubaidi and Webster, 1983) 
should maximize the influence of sodium. In 
such conditions "magnesium solonetzs" can be 
created with a minimum, of sodium. 

Management 

Management Of Dryland Saline Soils 

Dryland saline soils are the best managed as 
range land (pasture land) or hay land. Tillage 
concentrates salts and creates surface efflores- 
cences that only partially are revegetated, if at 
all. Caution is advised with regard to drainage. 
Skarie et al. (1987) noted that drainage in many 
flat landscapes increased salinity along the 
ditches. Griffin et al. (1985) also found that la- 
goons or any other source of water in these land- 
scapes raised the local water table and caused 
saline soils. 

Richardson and Amdt (1989) have noted that 
any wetland that receives groundwater would 
likely be saline if drained. It is suggested that 
any wetland with a component of discharge 
should not be drained; this includes all semiper- 
manent ponds. Further, a substantial zone of 
native grasses should be left around the wet- 
land. Salinization of soils adjacent to wetlands 
may occur by plowing too close to a pond edge. 

It is clear that selected grazing systems can 
increase total range production for livestock and 
wildlife on soils that are prone to salinization. 
Sedivec (1989) noted that grazing is a compat- 
ible land use with waterfowl production. 

Saline seeps are common in the southwestern 
part of the state and on the Missouri Coteau. 
These are emphemeral springs of saline water 
that appear during periods of higher than usual 
precipitation. Ferguson and Bateridge (1982) 
found that crop-fallow management increased 
deep percolation of water that contributed to 



lateral subsurface flow to saline seeps. Saline 
seeps are best managed by reducing deep perco- 
lation of water in the recharge areas by planting 
crops. We recommend a management system 
that includes deep rooted crops, such as alfalfa, 
and no summer fallow. Division ditching and 
tiling removes water from the seep directly. 

Management Of Saline-Sodic Soils 

The first rule to note in the northern Great 
Plains is that reclamation of a natric or sodic 
condition with gypsum will not be successful in 
most cases. The soil solution is saturated with 
respect to gypsum; and adding more gypsum 
will not make it any more soluble. Calcium chlo- 
ride is suggested as the amendment of choice in 
sulfatic soils. 

Management of saline-sodic soils depends not 
only on chemical and physical soil properties, 
but also on soil moisture regime. Soluble salt 
removal and replacement of exchangeable so- 
dium by calcium will reclaim these soils (Salin- 
ity Lab Staff, 1954), but the moisture regime 
must allow adequate leaching. Often this is not 
the case, because these soils have formed due to 
high water tables. 

Seelig and Richardson (1989) found that the 
most severely affected sodic soils had the high- 
est salinity. Decisions to improve the moisture 
regime by drainage should be made judiciously. 
Attempts to drain and leach saline-sodic Cal- 
ciaquolls are likely to lead to a puddled soil due 
to clay dispersion upon removal of the soluble 
salts (Salinity Lab Staff, 1954). Sandoval 
(1978) reported that deep plowing improved a 
saline-sodic Leptic Natriboroll in western North 
Dakota. This technique is most successful on 
the drier sodic soils. 

The key to proper management of saline-sodic 
soils is accurate determination of soil moisture 
regime. Saline-sodic soils with drier moisture 
regimes are less restrictive to use and manage- 
ment than their wetter counterparts and have 
the greatest reclamation potential. 

Management Of Irrigated Soils 

Salinity problems in the western states occur 
from the improper application of irrigation wa- 
ter. Numerous soils have become saline through 
the use of poor quality irrigation water, and, in 
many cases, lack of adequate leaching has com- 
plicated the management of certain soils 
(Sweeney, 1973). 

Drainage is an important requisite, because 
excess water in the root zone restricts root 



164 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



growth, delays warming of the soil in the spring, 
and induces the accumulation of salts. Subsur- 
face drains often are required to maintain the 
water table below the rooting depth of the crop 
and to provide for the removal of additional 
water necessary for leaching of excess soluble 
salts. 

High concentrations of salt in the soil increase 
the energy expended by the plant to obtain wa- 
ter and reduce the evapotranspiration and 
growth rates of the plant. It increases the irri- 
gation requirement, because additional water in 
excess of that needed for evapotranspiration 
must be applied to leach the salts from the root 
zone. Most saline soils can be improved by 
leaching with good quality water because many 
of the salts are very soluble. Soil physical condi- 
tions for crop production are generally good. 

To prevent secondary salinization by irriga- 
tion, an inventory of soils' water quality should 
be made. The need for surface and/or subsur- 
face drains, soil amendments, deep tillage, and 
salt tolerant crops can be assessed. Proper 
management based on the needs assessment 
will minimize the potential for salinization. 

References 

Alzubaidi, A., and G.R. Webster. 1983. Ion pairs in a solon- 
etzic soil Can. J. Soil Sci. 63:479-484. 

Arndt, J.L., and J.L. Richardson. 1986. The effects of 
groundwater hydrology on salinity in a recharge- 
flowthrough-discharge wetland system in North Da- 
kota, p. 269-277. In G. Van der Kamp and M. 
Madunicky (eds.) Proceedings of the Third Canadian 
Hydrogeological Conference, Saskatoon Sask., 21-23 
April 1986. 

Arndt, J.L., and J.L. Richardson. 1988. Hydrology, salin- 
ity and hydric soil development in a North Dakota prai- 
rie-pothole wetland system. Wetlands 8:93-108. 

Arndt, J.L., and J.L. Richardson. 1989. Geochemical de- 
velopment of hydrogen soil salinity in a North Dakota 
prairie-pothole wetland system. Soil Sci. Soc. Am. J. 
53:848-855. 

Beke, G. J. and C.J. Palmer. 1989. Subsurface occurrence 
of mirabilite in a Mollisol of southern Alberts, Canada: 
A case study. Soil Sci. Soc. Am. J. 53:1611-1614. 

Bigler, R.J., and J.L. Richardson. 1984. Classification of 
soils in prairie pothole wetlands with deep marsh plant 
species in North Dakota. Soil Survey Horiz. 25:16-24. 

Bowser, W.E., R.A. Milne, and R.R. Cairns. 1962. Charac- 
teristics of the major soil groups in an area dominated 
by Solonetizic soils. Can. J. Soil Sci. 42:165-179. 

Doering, E.J. and P.M. Sandoval. 1978. Chemistry of seep 
drainage in southwestern North Dakota. Proc. Subcom- 
mission on Salt Affected Soils, llth Intern. Congr. Soil 
Sci., Edmonton, Alberta., Vol 3 pp 1-15. 

Doolittle, J.A., C.J. Heidt, S.J. Larson, T.P. Ryterske, 
M.G. Ulmer, and P.E. Wellman. 1981. Soil survey of 
Grand Forks County, North Dakota. U.S. Department 
of Agriculture-Soil Conservation Service, U.S. Govern- 
ment Printing Office, Washington DC, U.S.A. 



Eugster, H.P., and B.F. Jones. 1979. Behavior of major 
solutes during closed-basin brine evolution. Am. J. Sci. 
279:609-631. 

Ferguson, H., and T. Bateridge. 1982. Salt status of glacial 
till soils of north central Montana as affected by the 
crop-fallow system of dryland farming. Soil Sci. Soc. 
Am. J. 46:807-810. 

Griffin, D.M. Jr., R.L. Skarie, A. Maianu, and J.L. 
Richardson. 1985. Effects of prolonged lagoon leakage 
on agricultural land. J. Civil Engr. 4:797-806. 

Hadley, R.F. and B.N. Rolfe. 1955. Development and sig- 
nificance of seepage steps in slope erosion. Trans. Am. 
Geophys. Union 36:792-804. 

Halvorson, A.D. and A. L. Black. 1974. Saline seep devel- 
opment in dryland soils of northwestern Montana, Soil 
and Water Cons. J. 29:77-81. 

Hardie, L.A., and H.P. Eugster. 1970. The evolution of 
closed-basin brines. Miner. Soc. Am. Spec. Paper 3:273- 
290. 

Hendry, M. J., J.A. Cherry, and E.I. Wallick. 1986. Origin 
and distribution of sulfate in a fractured till in south- 
ern Alberta, Canada. Water Resources Research 22:45- 
61. 

Hopkins, D.G., J.L. Richardson, and M.D. Sweeney. 1987. 
Composition comparisons in sodic map unit deline- 
ations on the Dickinson Experimental Station Ranch 
Headquarters, North Dakota. Soil Survey Horiz. 
28(2):46-50. 

Huggett, R.J. 1975. Soil landscape systems: a model of soil 
genesis. Geoderma 13:1-22. 

Keller, L.P., G.J. McCarthy, and J.L. Richardson. 1986. 
Mineralogy and stability of soil evaporites in North 
Dakota. Soil Sci. Soc. Am. J. 50:1069-1071. 

Knuteson, J.A. 1985. Microrelief and pedogenesis of soils 
of the Lake Agassiz plain. Ph.D. Diss., North Dakota 
State Univ., Fargo. (Diss. Abstr. 86-06139). 

Knuteson, J.A., J.L. Richardson, D.D. Patterson, and Lyle 
Prunty. 1989. Pedogenic carbonates in a Calciaquoll 
associated with a recharge wetland. Soil Sci. Soc. Am. 
J. 53:495-499. 

Last, W.M., and T.H. Schweyan. 1983. Sedimentology and 
geochemistry of saline lakes of the Great Plains. Hy- 
drobiologia. 105:245-263. 

Lissey, A. 1971. Depression-focused transient ground 
water flow patterns in Manitoba. Geological Associa- 
tion of Canada Special Paper 9:333-341. 

MacLean, A.H., and S. Pawluk. 1975. Soil genesis in rela- 
tion to groundwater and soil moisture regimes near 
Vegreville, Alberta. J. Soil Sci. 26:278-293. 

Miller, J.J., D.F. Acton, and R.J. St. Arnaud. 1985. The 
effect of ground water on soil formation in a morainal 
landscape in Saskatchewan. Can. J. Soil Sci. 65:293- 
307. 

Mills, J.G., and Zwarich. 1986. Transient ground-water 
flow surrounding a recharge slough in a till plain. Can. 
J. Soil Sci. 66:121-134. 

Mermut, A.R., and M.A. Arshad. 1987. Significance of 
sulfide oxidation in soil salinization in southeastern 
Saskatchewan, Canada. Soil Sci. Soc. Am. J. 51:247- 
251. 

Murphy, H.F. and D.A. Daniel. 1935. Some chemical and 
physical properties of normal and Solonetz soils and 
their relation to erosion. Soil Sci. 39:453-461. 



RICHARDSON, HOPKINS, SEELIG, AND SWEENEY: SALINITY/IN NORTH DAKOTA SOILS 



165 



Richardson, J.L., and R.J. Bigler. 1984. Principal compo- 
nent analysis of prairie pothole soils in North Dakota. 
Soil Sci. Soc. Am. J. 48:1350-1355. 

Richardson, J.L., and D.D. Patterson. 1986. Verification of 
saline mapping units for a soil survey. Soil Survey 
Horiz. 27:22-27. 

Richardson, J.L., and J.L. Arndt. 1989. What use prairie 
potholes. Soil and Water Conserv. J. 44:No. 3:196-198. 

Rosek, M.J., and J.L. Richardson. 1989. Comparison of 
erosional footslope soils in high and low relief drainage 
basins of western North Dakota. USA. Geoderma (Ac- 
cepted for Publication). 

Rostad, H.P.W. 1975. Diagenesis of postglacial carbonate 
deposits in Saskatchewan. Can. J. Earth Sci. 12:798- 
806. 

St. Arnaud, R. J. and A.J. Herbillon. 1973. Occurrence and 
genesis of secondary magnesium-bearing calcites in 
soils. Geoderma 9:279-298. 

Salinity Laboratory Staff. 1954. Diagnosis and improve- 
ment of saline and alkali soils. Agric. Handb. No. 60, 
USDA. U. S. Gov. Print. Off. Washington, D. C. 

Sandoval, F.M. 1978. Deep plowing improves sodic 
claypan soils. North Dakota Farm Res. 35:4:15-18. 

Sedivec, Kevin. 1989. Effects of specialized grazing sys- 
tems on upland nesting waterfowl production in 
southcentral North Dakota. Master of Science Thesis. 
Animal and Range Sci. Dep., North Dakota State Univ., 
Fargo. 

Seelig, B. D, and J.L. Richardson. 1989. Soil profile char- 
acteristics on a sodium affected landscape in central 
North Dakota. Papers presented at the 32nd Annual 
Manitoba Society of Soil Science Meeting. Dep Soil Sci., 
Univ. Manitoba, Winnipeg, pp 46-61. 

Sherman, G.D., F. Schultz, and F.J. Alway. 1962. Dolomi- 
tization in soils of the Red River Valley, Minnesota. Soil 
Sci. 94:304-3 13. 



Skarie, R.L., J.L. Richardson, A. Maianu, and G.K. Clam- 
bey. 1986. Soil and groundwater salinity along drain- 
age ditches in eastern North Dakota. J. Environ. Qual. 
15:335-340. 

Skarie, R.L., J.L. Richardson, G.J. McCarthy, and A. 
Maianu. 1987. Evaporite mineralogy and groundwater 
chemistry associated with saline soils in eastern North 
Dakota. Soil Sci. Soc. Am. J. 51:1372-1377. 

Steinwand, A.L., and J.L. Richardson. 1989. Gypsum oc- 
currence in soils on the margin of semipermanent prai- 
rie pothole wetlands. Soil Sci. Soc. Am. J. 53:838-842. 

Sweeney, M.D. 1973. Soil and water characteristics im- 
portant in irrigation. Circular S&F-573, Extension 
Service, North Dakota State University, Fargo. 

Timpson, M.E., J.L. Richardson, L.D. Keller, and G.J. 
McCarthy. 1986. Evaporite mineralogy associated with 
saline seeps in southwestern North Dakota. Soil Sci- 
ence Society of America Journal 50:490-493. 

Timpson, M.E., and J.L. Richardson. 1986. Ionic composi- 
tion and distribution in saline seeps of southwestern 
North Dakota, USA. Geoderma 25:295-305. 

Whittig, L.D., and P. Janitzky. 1963. Mechanism of forma- 
tion of sodium carbonate in soil: I. Manifestation of 
biological conversion. J. Soil Sci. 14:322-333. 

Winter, T.C. 1986. Effect of ground-water recharge on 
configuration of the water table beneath sand dunes 
and on seepage in lakes in the sandhills of Nebraska, 
U.S.A. J. Hydrology 86:221-237. 

Wollenhaupt, N.C., J.L. Richardson, J.E. Foss, and E.C. 
Doll. 1986. A rapid method for estimating soil salinity 
from apparent soil electrical conductivity measured 
with an above-ground electromagnetic induction meter. 
Can. J. Soil Sci. 66:315-321. 



166 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Aridisols of Argentina 
C.O. Scoppa and RJVL di Giacomo 1 

Abstract 

The aridic soil moisture regime (SMR) covers some 1,670,000 km2, which 
is 59% of the total surface of Argentina (2,800,000 km2). It extends on a line 
crossing the country in a north-south direction and on which some 502,000 
km2 of Aridisols (most of them Argids) have developed, representing 32% of 
the surface with an aridic SMR. The remaining 68% is covered by Entisols 
and some ustic, xeric, and aridic Mollisols, ustic Alfisols, and Inceptisols in 
the boundary areas with the Aridic regime. 

The parent material of the Aridisols corresponds to sandy aeolian and 
fluvial sediments of Holocene age which are burying, at different depths, 
clayey sedimentary strata of Pliopleistocene age laying on different Pre- 
quaternarian rocks. This pattern of sedimentological stratification is re- 
peated along the whole country, showing evidence of the action of exogen 
agents closely related to the relief. 

The geographical distribution of the Aridisols, at a regional level, shows 
that they are rather common in the north (between 22 and 28 degrees south 
latitude) in a mountainous landscape with closed depressions where the 
Argids (Paleargids) and Orthids (Camborthids) are found in similar propor- 
tions. They are scarce in the planes of the great central sedimentary basin 
(between 28 and 39 degrees south latitude) where Orthids (Paleorthids, 
Calciorthids, and Camborthids) are dominant. 

The most representative area is the Patagonia region, located in the 
southern portion of the country (between 39 and 52 degrees south latitude). 
The landscape is one of dissected tables where Argids and the great groups 
Natrargids, Paleargids, and Haplargids dominate over Calciorthids, Cam- 
borthids, and Paleorthids. 

The areal distribution and the aforementioned taxonomic difference is 
due to the distribution and thickness of the Holocene sediments. This is a re- 
sult of the prevailing winds and is closely related to relief and vegetation. 
These cold, dry, and intense winds originate at the South Pacific anticy- 
clone. They have irregularly distributed the Holocene sedimentary layer, 
leaving at different depths the clayey Pliopleistocene materials, which are 
parent material of the argillic horizon of the Argids. 

These Pliopleistocene materials had their origin under climatic and envi- 
ronmental conditions quite different from the present. 

The difference at the great group and subgroup levels is given by the 
textural, physical, chemical, and physico-chemical features of the parent 
materials in combination or not with the present pedogenetic conditions. 



Environmental Conditions of 

Argentina 

Argentina is located between 66 degrees 57 
minutes and 73 degrees 29 minutes W longitude 
and 21 degrees 46 minutes and 55 degrees 21 
minutes S latitude. The surface is 2,800,000 
km2. 

The country offers different and contrasting 
landscapes of different origin, nature, and mor- 
phology: mountains and basins in the west, 
plains in the northeast and southeast, and ta- 
blelands in the south and northeast. The Cordil- 
lera de los Andes is the axis of these environ- 

^entro de Investigaciones de Recursos Naturales 
(CIRN). Institute Nacional de Tecnologia Agropecuaria 
(INTA). Las Cabanas y Los Reseros s/n - (1712) - Castelar 
- Argentina. Phone 01-621-0281. Telex 17519 INTA AR 
FAX 54-111-1917. 



mentally contrasting conditions, and the cli- 
mate - in a NE-SW direction - varies from sub- 
tropical (humid) to temperate warm (humid, 
semi-arid, and arid) and temperate cold (arid 
and humid). According to the climate, the vege- 
tation varies from subtropical forest, to park 
and subtropical savanna, pasture, semi-arid 
woodland, shrub desertic steppe, and desert of 
altitude. 

Some 59% of the total surface (1,670,000 
km2) shows an arid condition. The arid and 
semi-arid regions are extended across a longitu- 
dinal line approximately limited by two mean 
annual rainfall lines of 300 mm, one located to 
the east and the other one to the west, along the 
Cordillera de los Andes. In this frame, Aridisols - 
representing 32% of the total soils with and 
aridic SMR. of the country - have developed. 



SCOPPA AND DI GlACOMO: ARIDISOLS OF ARGENTINA 



167 



REGION 



N R TE 



29 



10* 



REGION 



CENTRAL 



- REGION 



Pedoenvironmental 

Conditions of the 
Argentine Arid Region 

Pedoclimate 

The soil moisture and tempera- 
ture regimes (SMR and STR) of 
Argentina are shown in Figure 1, 
taken from Van Wambeke and 
Scoppa (1976). 

The distribution and extent of 
the aridic regime correspond to 
the general conditions of the exist- 
ing climate. This SMR coincides in 
general with the arid and semi- 
arid regions delimited by the 300 
mm mean annual rainfall line and 
is only interrupted in the north, 
by an inclusion of udic and ustic 
regimes due to specific conditions 
of the relief. The STR are mesic 
and thermic in the north, thermic 
and hyperthermic in the center, 
and mesic and thermic in the 
south. 

Parent Material 

In the region of aridic SMR, 
most of the Prequaternarian rocks 
are covered by semi-consolidated 
sediments more or less clayey and 
presenting different concentra- 
tions and shapes of calcium car- 
bonate, gypsum, and sodium of 
the Low Quaternary (Pliopleisto- 
cene?). Overlaying these sedi- 
ments there is a sandy aeolian 
layer (Holocene?) more or less rich 
in pyroclastic material. The thick- 
ness and distribution of this layer 
varies with latitude. In some sec- 
tors this layer has been and is 
being worked by wind action or is 
covered by modern sandy sedi- 
ments (Recent) of aeolian or fluvial origin. 

The sedimentary formations of the Pliopleis- 
tocene and Holocene as a whole constitute the 
parent material of the Aridisols. 

Relief 

The relief is composed of three main elements: 
the mountains located in the west, following a 
north-south line along the country; the plains in 
the central and western parts; and the table- 
lands in the south (Figure 1). Most of the moun- 



Soil Moisture and 

Temperature 
Regimes of Argentina 



PATAGON 1C A 



CA: Cotgmorca 

: Corr 
CHU: Cbubut 
PR: Pras 

: La 
ME: Wn4ftz 
NO: Maunn 
8Ct 
8J: San Juan 
IT: a!t 




90' 



Figure 1 



tains correspond to the Cordillera de los Andes 
and other associated morphostructural ele- 
ments. The height ranges from 4,000 to 7,000 m 
with peaks and valleys. In general the moun- 
tains follow a meridian line. 

The plains extend east from the piedmont of 
the Cordillera, showing smooth shapes. In many 
cases they are of aeolian origin (dunes). The 
drainage pattern is very poor. 

The tablelands correspond to dissected plains 
with stepped relief located east of the Patagonic 



168 



SIXTH INTERNATIONA! Soil CLASSIFICATION WOKKSHOP 



Cordillera and extending down to the Atlantic 
Ocean. The general characteristic of this re- 
gion is the presence of a tabular relief falling 
to the river valleys or the sea. In some cases it 
is cut by low sierras, depressions, or basaltic 
plains. 

Vegetation 

The following vegetation patterns (modi- 
fied from Cabrera, 1976) are found in the 
aridic moisture regime. 

Woodland semideciduous xerophytic 

It is found in the northeast and east cen- 
tral. The most common community is quebra- 
cho Colorado (Schinapsis balansae), quebra- 
cho bianco (Aspidosperma quebracho bianco), 
algarrobo, and calden (Prosopis sp.). Gram- 
ineous savannas and halophilous steppes con- 
stitute secondary communities. 

Shrub and gramineous steppe (altitude 2.000 
- 4,500 m) 

Located west of the preceeding. The char- 
acteristic feature is the absence of trees and 
the dominance of cactus, leguminous and 
composed shrubs. 

Shrub and gramineous steppe (altitude 2,000 
ml 

A central strip extends from the boundary of 
the preceeding formation down to the Atlantic 
Ocean. Small tree species and zigophilaceous 
shrub of the Larrea sp. kind and Prosopis 
shrubs are dominant. 

Shrub and Herbaceous steppe 

Characteristic of the Patagonic region, it ex- 
tends from the limit of the preceedingly men- 
tioned formation down to the southern extreme 
of Argentina. The dominant species are neneo 
(Molinum spinosum), coiron amargo (Stipa sp.), 
coiron bianco (Festuca palescens), and Larrea 
sp. 

Taxonomy and Inventory 

The Suborders, Great Groups, and Subgroups 
of Aridisols known in Argentina, as well as the 
surface they occupy and the percentage they 
represent up to the Great Group level, taken 
from Atlas de Suelos de la Rep#blica Argentina 
(1989), are shown in Table 1. These taxa are 
mostly associated with Entisols. Only in the 
border sectors of the Aridic regime, in transition 
to us tic and xeric regimes, can an association 
with Suborders, Great Groups, and Subgroups 
of ustic, xeric and aridic Mollisols, some Incepti- 
sols, and ustic Alfisols be found. 



TABLE 1: Taxonomy and inventory of Argentina Aridisols 



SUBORDERS 



GREAT GROUPS SUBGROUPS 



ARGIDS 
306,396 Km2 
61% 



NATRARGIDS 
114,500 Km2 
37% 



typic, aquic, ustollic, 
haplic, lithic,lithic- 
xerollic, borollic, 
xerollic and haploxerollic. 



PALEARGIDS typic, petrocalcic, 
102,270 Km2 petrocalcic-xerollic, 

33 % ustollic, xerollic and 

borollic. 



ORTHIDS 
195,000 Km2 
39% 



HAPLARGIDS 
89,600 Km2 
30% 

CALCIORTHIDS 
79,600 Km2 
41% 

CAMBORTHIDS 
69,300 Km2 
36% 

PALEORTHIDS 

32,400Km2 

16% 

SALORTHIDS 

ll,OOOKm2 

6% 

GYPSIORTHIDS 

2,OOOKm2 

0,7% 

DURORTHIDS 

760Km2 

0,3% 



typic, borollic, aquic, 
arenic, lithic, lithic- 
xerollic, ustollic, 
xerollic,lithic-ustollic. 
typic, ustollic, lithic, 
lithic-ustollic, xerollic 
and borollic -lit hie. 

typic, borollic,fluventic, 
lithic, ustollic, lithic- 
xerollic and natric. 

typic, ustollic and 
xerollic. 



typic and acuolic. 



typic, calcic and 
petrogypsic. 



typic. 



Distribution 

Within the great region of aridic SMR 
(1,670,000 km2), only 32% of this surface 
(502,000 km2) corresponds to Aridisols, distrib- 
uted as follows: 306,389 km2 (61%) is occupied 
by Argids and 195,889 km2 (39%) by Orthids 
(Table 1). 

The geographic distribution at the Suborder 
and Great Group levels is shown in Graphics 1, 
2, and 3, disposed by provinces and regions from 
north to south, extended between 22 and 52 
degrees south latitude. 

If we analyze this distribution, it appears that 
most of the Aridisols (Argids + Orthids) of Ar- 
gentina are in the Patagonic region (between 39 
and 52 degrees South latitude) and that they 
are more or less equally represented in the cen- 
tral (between 28 and 39 degrees south latitude) 
and northern (between 22 and 28 degrees south 
latitude) regions. 

At the Suborder level, Argids dominate in the 
Patagonic and northern regions, while in the 
central region Orthids are almost exclusive. 

As for the Great Groups of the Argids, Palear- 
gids are practically the only existing in the 
north. There are only a few Haplargids in the 
center, and Natrargids, Paleargids, and Haplar- 



SCOPPA AND DI GlACOMO: ARIDISOLS OF ARGENTINA 



169 



SUBORDERS ARID I SOLS 




ARGIDS DORTHIDS 



120.000 

100.000 

K 80.000 

m 60.000 

2 40.000 

20.000 





_T-I n. _. n. ,. n_n 




|Juj Sal Cat Tu .E La Co Sa Me Sa La } Ne Rio B. Ch Sa j 
Province* Jljy ta am cu pte RJ O rc | n nd n Pa ,uq Ne Air ub nta | 

I arc ma ;ro ja ob Ju oza Lui mp|Ue gr es ut Or j 

Regions and an a an sano uz 

Latitude S 22 NORTH 2S CENTRAL 39 PATAGONIA 52 

Graphic 1 



gids are found in the south, with different pro- 
portions. 

Within Orthids, Camborthids and Calci- 
orthids along with very few Salorthids are the 
most common in the northern and central re- 
gions, whereas Calciorthids and Gypsiorthids 
are dominant in the Patagonia. 

A summary of the above can be seen in Table 
2. The percentage of each of the Suborders and 
Great Groups is given for every one of the three 
considered regions. 

Characteristics of the Taxa in the 
Different Regions 

A brief description of the associations, parent 
material, and landscape and some characteris- 
tics of the morphology of the Great Groups in 
the different regions considered is given below. 
In Table 3 a summary of the physical and 
chemical properties of these taxa is displayed. 



so.ooo 








GREAT GROUPS ARGOS | 


Paleargids ONatrargids Haplargids 








40.000 
* 30.000 




Jj 


[UL 


, 20.000 { 




Jl 


Jill 


* 10.000 L w I 

m ,B JL. 

.Juj Sa! Ct 
Provinces j uy ta ar 

Regions and ' a 
Latitude S 22 NORT 


, , 




Jill 


it Tu ;S.E La Co Sa Me 
n cu jste Rio rd n nd 
c ma jro ja ob Ju oza 

n a an 
H 28 CENTRAL 


Sa La | Ne Rio B. Ch Sa 
n Pa juq Ne Air ub nta 
Lui mp | ue gr es ut Or j 

sano uz 
39 PATAGONIA 52 




Graphic 2 







Northern Region 

Paleargids 

These soils appear in associations of 
the subgroups: typic, ustollic, and petro- 
calcic and/or in association with Enti- 
sols. Paleargids have developed on aeo- 
lian friable materials - mixed with allu- 
vial and colluvial materials - laying on 
rubble and gravel, on ancient stabilized 
forms: piedmonts, alluvial cones, and 
plains. The red argillic horizon often 
appears in the surface as a consequence 
of the erosion of the Al horizon. A petro- 
calcic horizon can be present, separat- 
ing the profile either from the Paleozoic 
rock or from Cenozoic sediments. 

Haplargids 

Associated with Entisols (Torriorthents) they 
also have developed on colluvial, alluvial, and 
aeolian sediments within natural drainage 
ways. The typic subgroup is found on fluvial ter- 
races. The borollic subgroup appears in smooth 
relief areas with gentle slopes (Puna and Pre- 
puna) and is above 4,000 m in elevation. 

Natrargids 

Associated with Entisols and other Argids, 
they have developed mainly in alluvial on flu- 
vial plains and on flood plains with depressions. 

Camborthids 

These appear in associations of different sub- 
groups or are associated with Entisols. Alluvial 
or aeolian sediments from the foothills of high 
mountains are the parent materials for these 
soils. Lithic subgroups appear below 4,000 m, 
whereas, above this elevation (Puna and Pre- 
puna), borollic subgroups are frequent. The 
typic and fluventic subgroups are found on allu- 
vial fans and in the flat sur- 
roundings of drainage ways. 
The ustolic subgroup appears 
in higher, more moist places. 

Salorthids 

These appear associated 
with ustic Inceptisols in areas 
bordering salt deposits and 
fluvial surroundings, on 
plains, and in poorly drained 
depressions. The conductivity 
is higher than 60 mmhosXcm 
(Vargas Gil and Culot, 1980). 



170 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 























1 GREAT GROUPS OR1HIDS g 
1 lim^mmsmsy^m^ 


40.000 - 

K 30.0007 

2 20.000] 


' ^ camborth i 

ds 

i 

i H S a 1 o r t h i d 
s 


LsCalciorth ESPaleorthi 
ids ds 

ElDurorthid BGypsorthi 
s ds 


i 

ii B 
















n 

di 


10.000 ] 


j3 




R 


R . 




n 


n 


ji IB 

Hfl .j^lL 


J V 

Bt JL , 


o ^ - 

,JU 


Sa Ca 


Tu S. 


La 


Co 


Sa 


Me 


Sa La 


Me Ri B. 


Ch Sa i 


Provinces Jju 


1 t ta 


cu 'Es 


Ri 


rd 


n 


nd 


n Pa 


)u q o A i 


ub nt . 


! y 


a ma 


ma 't e 


o 1 


ob 


Ju 


02 


Lu mp 


!ue Ne re 


ut a . 


Reg I ons and 1 


re 


n 'ro 


a 


a 


an 


a 


1 s a 


n gr s 


Cr 


Lat i tude S 22 


NORTH 


28 


CENTRAL 


39 PATAGONIA 52 


Graphic 3 



Central Region 

Haplargids 

These are subordinated components of asso- 
ciations with Entisols and developed from allu- 
vial sediments mixed with aeolian sediments in 
flat or concave-flat areas. In the particular case 
of the "barreales" (mudholes), the alluvial sedi- 
ments brought coarser materials produced by 
erosion processes which occurred in higher ar- 
eas. 

Natrargids 

These are associated with other Argids and/or 
Entisols and are developed in aeolian sediments 
in low areas with impeded drainage or in areas 
around salt deposits. These soils contain high 
values of soluble salts. 

Salorthids 

These are in the typic and aquollic subgroups 
and are associated with other Orthids. They 
have developed in fluvial-lacustrian 
materials overlaying sandy-loam and 
clayey sediments located in low areas of 
basaltic plains and in areas surrounding 
salt deposits and lagoons. Conductivity 
is higher than 35 mmhos/cm. 

Paleorthids 

These are associated with other 
Orthids and with Entisols. They are 
formed in loessal materials and are lo- 
cated in the higher portions of more or 
less defined hills with dendritic drainage 
patterns and usuallly contain lime lay- 
ers. A petrocalcic horizon can be found at 
different depths. 



Qamborthids 

These appear as minor com- 
ponents in associations with 
Entisols and some Mollisols 
and also are found in associa- 
tion with other Aridisols. The 
parent materials are aeolian/ 
alluvial deposits in a plain 
landscape. Soft powdery lime 
appears under the diagnostic 
horizon. 

Calciorthids 

These are associated with 
Entisols and developed from 
aeolian sands recently depos- 
ited on silicified lime. These 
aeolian materials sometimes 
are mixed with other materi- 
als of a fluvial type (redeposited loess). These 
soils are found in distal areas of piedmont fans 
and relict areas of river beds and alluvial ter- 
races. 

Durorthids 

These are associated with Entisols (Torri- 
orthents) and developed in aeolian quartzitic 
sands in areas where the water table is close to 
the surface producing the saline characteristics 
and cemented horizons. 

Gypsiorthids 

These are primarily in the petrogypsic and 
typic subgroups. The parent materials are flu- 
vial-lacustrian materials with a high content of 
calcium sulfate and soluble salts. They are com- 
mon in areas near salt deposits. 

Patagonic Region 

A common characteristic of this region is the 
so - called "desert pavement," detritic cover due 



TABLE 2: Geographic distribution of Argentina Aridisols. 


1. SUBORDERS (%) 


REGIONS 


ARGIDS 


ORTHIDS 




Northern 


11 


10 




Central 


3 


22 




Patagonia 


86 


68 




2. ARGIDS GREAT GROUPS (%). 


REGIONS 


PALEARGID 


NATRARGIDS 


HAPLARGIDS 


Northern 


35 


1 


3 


Central 





1 


2 


Patagonia 


65 


98 


95 


3. ORTHIDS GREAT GROUPS (%). 


REGIONS 


CAMBORTHIDS 


CALCIORTHIDS 


PALEORTHIDS 


Northern 


26 








Central 


16 


14 


39 


Patagonia 


58 


86 


61 




SALORTHIDS 


DURORTHIDS 


GYPSIORTHIDS 


Northern 


3 








Central 


58 


100 


12 


Patagonia 


39 





88 



SCOPPA AND DI GlACOMO: AE1DISOLS OP ARGENTINA 



171 



TAILE 3 


AII1EVIATED MORPHOLOGICAL, PHYSICAL AND CHEMICAL PROPERTIES Of SOME PCDONS OF Till GREAT CROUPS IM THE DiffEREX! REGIONS 












CAtClOITHIDS 








Oft )fc 


tdw 


*M*iitUy 


Organic 






Ktrt* MI*** 




(acton* 


C*w 




AflftAS 


&** 


w *"*** i 






fefttf 


Sill 


ai 


Corboa 


Nitrgft 


C4CO t 


C 


b0 


Ma 


X * 


ccc 


So4Kra 


ty 


>M 








* 


*il 






* 


t 




a4/OOi'. 




% 


Sffi.1 


l 


CEXTRAg, 


TYPIC 


Al 
AC 
CCA 


0-30 
30-30 
30-100 


IOYR 4/3 
IOYR 4/3 
IOYR 5/3 


67.0 
65,1 
60,9 


23,7 

26,4 
26,0 


11.5 
10,7 
11,7 


0,73 
0,40 
0,18 


0,060 
0,045 


5.8 

18.2 
10.0 


: 


- 


0,8 
0,9 
2,2 


1.3 
0,9 

0,8 


21.2 
18,8 
21.2 


4,0 
3.0 
10,0 


6,4 
7.5 


7,7 
7,3 
7,4 






Al 


0-18 


IOYR 3/5 


82,1 


14,8 


8.1 


0,28 


_ 


^ 


13,6 


3,9 


0.7 


0,3 


18,4 






7 } 


PATAGONIA 


TYPIC 


AC 
2CI 
2C2CA 


18-32 
32-43 
45-100 


IOYR 3/3 
IOYK 3/2 
IOYR 5/3 


82,8 
70,1 
37.7 


6.0 
21, 
6. 


12.5 
17.2 
23.6 


0.27 
0,66 
0,68 


0.076 
0,064 


o7$ 

33.6 


12,3 
25,6 


2,6 

5,6 


0.4 
2,5 
U7 


0.5 
0.4 
0.4 


16.5 
32,2 
22.8 


7,0 
7.0 


2,3 


7,3 
7,6 
7,6 










CAHBORTHTDS 


NORTHERN 


TYPIC 


Al 
82 
C 


0-20 
20-30 
50-30 


IOYR 6/2 
7.5YR 5/4 
7,3YR 7/2 


80,8 
77,2 
81,2 


11, 
12. 
10, 


8.8 
flU 


0,32 
0.09 
0,09 


0.040 
0.028 





3.03 


0,76 


0,24 
0.25 
0,33 


0.36 
0.38 
0,29 


3,36 

4.34 
3,42 


7,14 
3.31 
9,63 


0.25 
0,28 
0,26 


7,0 
8.1 
8,3 










CENTRAL 


urrouic 


Al 
12 
C 


0-22 
22-60 
60-100 


IOYR 3/2 
IOY* 3/2 
10YR 3/4 


5,0 
10,0 
26,6 


68. 

61. 
53, 


28.8 
30,9 
18.0 


2,67 
1.70 
0.98 


0.299 
0,178 


- 


















29,3 

24,1 


2,4 
1,9 


0,1 
0.2 


U5 
0.9 


30,1 
17,3 


- 


. 


l',t 


PATAGONIA 


TYPIC 


Al 
82 
B3CA 
CCA 


0-27 
27-40 

40-30 
50-60 


IOY1 4/3 
^,5YR 4/4 
7.5YR 5/4 
7.5YR 6/4 


62,7 
65,7 
54.1 
67,2 


22. 
19. 
12. 
19. 


14.3 
15.1 
1,2 
12.9 


0.34 
0.29 
O.U 
0.24 


0.048 
0,038 
0.025 
0,019 


3.9 

10,71 


13,8 
23.0 


3,18 
3.06 


0,56 
0.85 
0,63 
0,86 


1,38 
0,92 

0.62 
0,26 


16,33 
22,20 
17,18 
17,47 


3,40 
3, S3 
3,68 
4,89 


- 


7,5 
7,6 

8,0 
7,8 














r 


ALEORTi 


HI?? 






















CENTRAL 


TYPIC 


Al" 

2C 

tOflCA 


0-13 
13-42 


IOYR 4/3 
IOYR 6/4 


57,0 
58,0 


33, 
26, 


11,5 
17,6 


0,64 
0.73 


0,086 
0,065 


4,3 

2,2 








0,6 
1,2 


1,2 

0,3 


11,4 
12,7 


9,0 


55,5 


7,3 
7,2 






AC 


0-18 


10YR3/2 


86,1 


14. 


.4,2 


0,54 


0,036 


1.3 


^ 





0,5 


1*4 


13,7 






7 & 


PATACOMIA 


XEROLLIC 


Cl 


18-43 


IOYR 3/2 


84.0 


9. 


6,4 


0,4 


3,052 


2,3 





< 


0.6 


0.2 


U,5 






7.9 






C2 


43-50 


IOYR 3/3 


81,9 


5, 


9,0 


0,36 


3,045 


3,9 


. 


. 


0,7 


0.2 


12.9 


_ 




8,1 






E00CA 


































ULORTHIDI 






Al 


0-20 


10YE 8/2 


63.0 


21.0 


9.0 


0.42 


0,038 


1,1 


_ 





0,7 


0.13 


4.6 


13,2 


66,0 


7,f 


WORTHED 


USTOLLIC 


AC 


20-40 


IOYR 3/2 


54.0 


31,0 


13,0 


1,37 


0,085 







_ 


2,2 


1.59 


16,0 


13. a 


67,0 


7,8 






C 


40-70 


IOYR 3/2 


67.0 


23,6 


9,4 


0.65 


0,052 


6,5 


- 


- 


6,03 


1.68 


10.4 


58,0 


32,0 


8,0 






I 


0-25 


7,5YR 4/4 


14,3 


86,5 


15,1 


0,31 


0.033 


1,3 





* 


5,8 


1,3 


15,0 


39,0 


83,3 


1,0 


CEVTRAI. 


TYPIC 


II 


23-40 


7, SYR 4/4 


10,3 


15,9- 


22.0 


0.27 


- 


3,7 


- 


- 


21,6 


19 


25,2 


50,0 


38,8 


8,2 






I 


0-20 


5* 6/3 


7,4 


18.8 


10,8 


0.06 


0.005 


1,9 


w 


~ 


7.2 


0,6 


15,2 


47,0 


8,3 


9,2 


PATACOMIA 


AQUOUIC 


II 


20-45 


3Y 5/3 


43,5 


36,2 


20,3 


0,1 


0,011 


1,4 








15,6 


1,2 


22.3 


70,0 


6,0 


8.7 






III 


45-100 


5Y 5/4 


51,4 


35.6 


12,9 


0.28 


0,028 


19,6 


- 


- 


25,0 


1,4 


29.9 


14,0 


32,2 


9,4 












DURORTHIDS 


















AC 


0-20 


IOYR 3/4 


83,37 


6,4 


7.23 


0,24 





_ 


3,20 


2.39 


0.32 


1.81 


7.47 




1,3 


6.6 


CkMTRAL 


TYPIC 


C 


20-40 


IOYR 3/4 


86, 14 


6,97 


7.59 


0,14 


- 


tracts 







0.69 




8.4S 





3.34 


7.6 






2C28 


40-50 


- 


75,57 


11,0 


14,04 


O.U 


- 


tr&caa 


- 


- 


3,02 


- 


16,64 


- 


9,24 


8,0 







to the selective removal and the thin discontinu- 
ous sandy accumulations. 

Natrargids 

These are associated with Entisols and with 
other Argids. They developed in aeolian sedi- 
ments overlaying clayey sediments wich are 
more or less consolidated and can be found on 
various landscape positions: slopes, plains, 
tables, flats, and microdepressions. 

Haplargids 

These are associated with Entisols, Orthids, 
and xeric Mollisols. They have developed in aeo- 
lian sediments overlaying a clayey layer. They 
are found on plains, undulating tables, short 
steep slopes, and fluvial terraces. 



Paleargids 

They are associated with Entisols, other Ar- 
gids, and xeric Mollisols. The parent materials 
are aeolian sediments of different thickness 
which overlay ancient clayey sediments. The 
concentration and hardness of calcium carbon- 
ate is variable. The position on the landscape is 
variable. A petrocalcic horizons occurs in some 
perfiles. 
Galciorthids 

They occur in association with other Orthids, 
Argids, and Entisols. They have developed in 
aeolian materials, on extended flat areas within 
the basaltic tables, on dissected smooth hills, 
and on marginal depressions of alluvial fans. 



172 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



TABLE3.CONT. 


AWtVIATKD NORPUOLOCICAL. PHYSICAL AKO CttEXICAL PXOPERTZES OP SOME PEOOKS Of THE GREAT CROUPS IK THE DIFFERENT REGIOHS 


JUTRARCIDS 












Oto*ilntf7 


QtVrts 






CatroctoMi ftlAt 




Catgut 








A*iM 


tofefefeJH 


4wU 






*< 


Sill 


C*y 


CttflMM 


****** 


C.CO, 


<* 


Ut 


Ntt 


* 


cec 


Sftdfw* 


Nrfly 


M 













JMitl 






* 


k 








*/ 








% 


SJjftt 


Mil* 




WITHERS 


AQUIC 


Al 


0-18 
18-70 


IOYR 2/2 
IOYR 3/2 


82,2 

41,0 


14,00 
35.00 


3.8 
24,0 


0.46 
0.40 


0.038 
0.05 


1,0 
3.6 


- 





0,36 
1,80 


1,12 
1,60 


8,1 
16,0 


4.6 
15.0 


1,15 
7,20 


06 
I 








Ap 


0-18 


IOYR 3/3 


58,8 


43,00 


13,8 


1,03 


0.043 


m 


11,8 


3,4 


1,30 


1,70 


16.1 


8.00 




5 








82t 


18-46 


7. SYR 3/2 


47.2 


35,40 


30.80 


1.07 


0.043 


0.4 






10,00 


2.60 


33.9 


28.00 


1,8 


7 




CENTRAL 


TYPIC 


83ca 


46-47 


7, SYR 5/6 


42.3 


37,20 


27,30 


1.07 


0.060 


7,6 








12,90 


2.30 


36.8 


35.00 


3.9 


8 








Ce 


67-80 


- 


40. a 


16,80 


24,30 


1.05 


- 


20,4 


- 


- 


12,90 


1.9 


30.2 


43.00 


10,9 


6 








Al 


0-9 


IOYR 3/4 


56.7 


53.40 


9, 


0,9 


0.071 


l.S 


m 





4,80 


2.1 


14,6 


33,00 




5 








212t 


9-22 


IOYR 4/4 


37.9 


31,20 


30, 


1.63 


0.141 


9.8 


. 


* 


14,90 


1.4 


22.0 


68.00 


4,2 


6 




PATAGONIA 


TYPIC 


2131 


22-34 


7, SYR 3/4 


65.9 


I, SO 


32, 


0,28 


0.029 


17.9 


. 





10,90 


0,3 


26.0 


42.00 


22,5 


4 








2832 


34-58 


7, SYR 5/4 


79.1 


20,30 


0, 


0.31 


0.031 


11.3 








1.00 


0.2 


21.3 


3.00 


76,6 


7 








2C2 


56-100 


7. SYR 4/3 


83.3 


5,70 


10, 


0.12 


0.010 


U.4 


- 


- 


2.90 


0.2 


18.5 


16.00 


13.1 







PALEARCtOS 








All 


0-8 


IOYR 4/3 


91. 


3,60 


3.40 


0.28 


0.028 


m 


2,78 


0,18 


0.29 


0.30 


3,27 


.9 


0,32 


7 




NORTHERN 


TYPIC 


A12 


8-32 


IOYR 3/3 


91, 


3,20 


3.00 


0,07 


0.023 


. 


2,43 


0,20 


0,33 


0.23 


3,32 


.9 


0,39 










82c 


32-55 


SYR 4/3 


54, 


4,00 


41,40 


0,42 


0.057 


- 


12,86 


3,93 


1.16 


0.94 


19.80 


.9 


0,43 










Al 


0-6 


IOYR 4/2 


49, 


26.7 


23,4 


1.07 


0.106 


2,2 





<- 


0,30 


0.40 


26,7 


.0 








FATACOHIA 


XE10U.IC 


212 ( 


6-22 


IOYR 3/2 


39, 


13,7 


46,9 


1,60 


0.156 


2,4 


. 


. 


0,30 


1,70 


26,4 


.0 


V 










3C 


22-60 


7.SYR4/2 


79, 


11*2 


12,8 


1.13 


0,013 


6,3 


- 


- 


0,30 


0.40 


18,3 


.0 


- 






HAPURCtD* 




NORTHERN 


MROU.IC 


Al 


0-25 


7.5YR 6/2 


81,4 


12,8 


5,80 


0,63 


0,069 


~ 


7,99 


1,27 


0,10 


0.41 


11,8 


0.85 


0,18 










82C 


25-50 


SYR 4/4 


35,4 


41,2 


23,4 


0,28 


0,041 


- 


13,40 


3,70 


0,23 


0,39 


21,2 


1.08 


0,27 










Al 


0*12 


IOYR 6/2 


45,8 


51,1 


21, 


1.08 


0.120 


0, 








0,70 


2,90 


23,1 


_ 












121t 


12-30 


7, SYR 4/4 


25,4 


45,0 


28, 


0.85 


0,067 


0, 


. 


. 


1,40 


3.40 


34.2 


4.00 









CENTRAL 


TYPIC 


122c 


30-45 


7, SYR 5/4 


16.7 


42,7 


24, 


0,43 


0,067 


U, 


. 


. 


2,20 


3,60 


38.9 


6,00 













13ca 


45*75 


7, SYR 5/4 


23,3 


30,2 


26, 


0,17 


, 


32, 


. 


. 


3,20 


2.60 


31.9 


10.00 


1,0 










Cca 


75-100 


7, SYR 6/4 


54. Q 


27,2 


21, 


0.08 


- 


10, 


- 


- 


6,30 


2.30 


35,3 


10,00 


2,4 










A 


0*11 


IOYR 3/3 


83.1 


7.2 


9. 


0.92 


0,069 


m 


6.4 


1.60 


0,3 


0.8 


15.3 


2,0 








PATACOMIA 


TYPIC 


212t 


11-33 


7, SYR V2 


54.5 


13.2 


30, 


1.69 


0,143 


2,7 






0,3 


0.4 


33. 1 






7 








3Cc 


33-100 


IOYR 8/2 




























1 











































Paleorthids 

These are associated with other Orthids and 
with some Entisols. They have developed in 
aeolian sediments and clayey material on ter- 
races, plains and tables. 

Camborthids 

They occur in association with other Aridisols 
and with Entisols. They have developed in aeo- 
lian/alluvial materials on table-shaped plains 
which are gently undulating, in extended de- 
pressions, and on gentle slopes. 

Salorthids 

These are associated with Argids and with 
some Entisols. They occur in depressions (salt 
deposits) on short gentle slopes and in poorly 
defined drainage ways consisting of depressions 
forming a line. The conductivity of these soils is 
45 mmhos/cm. 

Gvpsiorthids 

They are associated with other Orthids (Sa- 
lorthids). The parent materials are fluvial-lacus- 
trian sediments with high contents of calcium 
carbonate. They are located in depressional ar- 



eas, near salt deposits (Salazar and Godagnone, 
unpublished, 1984). 

Genetic Considerations 

The variety and distribution of the identified 
taxa as well as the peculiar geological and pedol- 
ogical characteristics existing in Argentina al- 
low some possible conclusions about the genesis 
of these Aridisols. Graphic 4 presents, in a ge- 
netic model, these conclusions, which mainly 
refer to the origin and distribution of the Subor- 
ders (Argids and Orthids). 

The starting point is to assume that the whole 
area with aridic SMR - and also most of the 
country - is covered by modern sediments (Pleis- 
tocene?, Holocene, and Recent) of aeolian origin. 
These sediments are friable, sandy and sandy 
loam textured material (P3 in the model) over- 
laying Pliopleistocene (P2) clayey materials. 
The thickness, distribution, nature, and grain 
size of these Holocene (P3) sediments are a re- 
sult of the drifting by winds, blowing from the S- 
SW-W direction, which originated in the South 



SCOPPA AND DI GlACOMO: ARIDISOLS OF ARGENTINA 



173 



GENETIC MODEL OF ARGENTINE ARIDISOLS 



ORDER 



REGIONS 



SUBORDERS 



UJ 



HOLOCENE 




PLIO. 
PLEISTOCENE 



PRE. 
QUATERNARY 



LANDSCAPE 



INTENSITY 



DIRECTION 



SOIL 

TEMPERATURE 
REGIMES 



VEGETATION 



LATITUDE 



R I 



NORTHERN 



ORTHIDS + 
ARGIDS 



CENTRAL 



ORTH1DS 



PAT AGON ICA 



ARGIOS 



Cloy + CQ^Co (Fluvioi ond 



Mountains 
and Plains 



nil 



w 



ME - TH 



Shrub and gra. 
mintoussfippt 
woodland 



22L.S. 



Plains 



Middlt 



W - SW 



TH - HY 



Shrub 

and gramlnious 
sttpp (1 2000m) 



Tablelands 



SW - S 



Graphic 4 



Pacific anticyclone. These winds are extremely 
intensive in the Patagonic region, causing defla- 
tion and drifting of materials which later on are 
deposited in the central and northeastern re- 
gions that become accumulation areas as winds 
loose intensity. Thus, the Holocene and Recent 
sediments, which are very thin or practically 
nonexistent in the Patagonia, become very thick 
in the plains of the central region and is medium 
in the northern region according to the charac- 
teristic landscape of tables and valleys (Graphic 
4). 

The geological, physiographic, and pedologi- 
cal features indicate that the Holocene sedi- 
ments - when thick enough - are the parent 
material of the Aridisols as a whole. On the con- 
trary, these Holocene sediments are the parent 
material of surface horizons when interrupted 



by other underlaying mate- 
rials such as the Pliopleisto- 
cene clays (P3) which would 
constitute the argillic hori- 
zon of the Aridisols. The clay 
and organic carbon contents 
of the Bt horizons in the Ar- 
gids and its Great Groups 
show this lithologic disconti- 
nuity. Therefore, the origin 
of the argillic horizon of the 
Argids is due to environ- 
mental conditions which are 
different from the present 
ones and which were present 
during the formation of the 
P2 material. The present pe- 
doclimatic conditions would 
have not influenced this ar- 
gillic horizon but would have 
acted almost exclusively on 
the A horizon developed 
from the Holocene sedi- 
ments (P3). 

The presence of Argids on 
Orthids depends fundamen- 
tally on the thickness of P3 
material, which is a conse- 
quence of the action, direc- 
tion, and intensity of the 
winds (main sedimentologi- 
cal agent) - more or less con- 
stant since the end of the 
Andean orogeny - passing 
through different land- 
scapes, climates and vegeta- 
tion. 

The differences at the Great Group level are 
due, in the case of the Argids, to the local charac- 
teristics of the P2 material (Pliopleistocene) 
which are a consequence of its own genesis. This 
is reflected in the variable content of calcium 
carbonate, gypsum, and sodium, grain size com- 
position, and grade of cementation, the presence 
and/or quantity of which results in conditiona 
for the development of Paleargids, Natrargids, 
and Haplargids. 

The taxonomic difference for Orthids can be 
explained by the aforementioned characteristics 
plus the presence of soluble salts in the Holocene 
sediments (P3), which are directly responsible 
for the presence or absence of Calciorthids, Gyp- 
siorthids, Camborthids, Paleorthids, Durorthids 
and Salorthids. The very abundant Recent sedi- 
ments resulting from the deposition of new ma- 



ME- TH 



Shrub and 
herbaceous sttpp* 



174 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



terials, as well as the reworking of those of the 
Holocene, have generated different types of Enti- 
sols (Torriorthents, Torrifluvents, Fluvaquents 
and Torripsaments) which are no doubt the most 
common in the region with aridic SMR in Argen- 
tina. 

References 

Van Wambeke, A. and Scoppa, C.O. Las taxas climticas de 
los suelos Argentines. Determinacion de las definiciones 
del Soil T axonomy mediante el modelo matem atico de 
Newhall y su resoluci on por computacion Fortran IV. 
RIA s3, v. XIII, N 1, pp. 7-39. 1976. 



Cabrera, A. Regiones Fitogeogr aficas Argentinas. Fasc. 1, 

Tomo II Enciclopedia Argentina de Agricultura y Jar- 

dineria. 2da. Edicibn. Ed. ACME SACI. Buenos Aires. 

1976. 
INTA. Atlas de Suelos de la Rep ublica Argentina (in 

press). 
Vargas Gil, R. and Culot, P. Los Suelos de la Puna. Actas 

IX Reunion Argentina de la Ciencia del Suelo, Tomo 

III, pp. 1.065-1.075. Parara. 1980. 
Salazar Lea Plaza, J.C. and Godagnone, R. Suelos de la 

Provincia de Rio Negro (in press). 



Classification of Aridic Soils, Past and Present: Proposal of a 

Diagnostic Desert Epipedon 

A. Souirji 1 



Abstract 

A diagnostic desert horizon, namely the torric epipedon, is proposed as 
an alternative to the use of the aridic moisture regime in the definition of 
high level taxa of aridic soils. 

The strong variation in moisture and air content of the torric epipedon 
induces physical dispersion, which play a major role in current desert soil 
formation. 

Eolian activity also plays an important role in desert soil formation, and 
the occurrence of eolian sand is proposed as a diagnostic criterion. 

The low organic matter content of desert soils is due to low litter produc- 
tion, its dispersion by wind and surface wash, and low incorporation, rather 
than to high decomposition. 



Introduction 

Early authors on aridic soils considered that 
there is little chemical and biological weather- 
ing in deserts because of low precipitation and 
scarce vegetative cover. Hence, only coarse-tex- 
tured, undifferentiated soils are to be found in 
these environments (Hilgard, 1906). These 
ideas stemmed directly from the concept of zonal 
soils at a time when little data were available on 
soils of the arid regions. 

With the development of irrigation in arid 
regions, more knowledge was accumulated 
about aridic soils, and their diversity became 
evident. "All manner of soils are found in arid 
areas and that to refer to 'desert soils' as if they 
constituted a homogeneous group is both erro- 
neous and misleading" (Jackson, 1957). 

All major soil classification systems had high 
level taxa of aridic soils. Brown desert soils in 
the USSR (Tiurin, 1965) and Light colored soils 
of arid regions in the USA (Thorp and Smith, 
1949) are two examples. Most of these classifi- 
cations distinguished aridic taxa at the highest 
level merely by their geographical location in 
arid areas. 

Some of the main characteristics of aridic 
soils, such as the presence of a vesicular crust, 
low organic matter content with dominance of 
fulvic acids, and surface accumulation of carbon- 
ates, were known to the Russian pedologists 
since the early decades of this century (Lobova, 
1960 ). 

However, the wealth of information gathered 
by the Russian pedologists did not lead to an 
effective classification of aridic soils because 



genetic factors of soil formation, soil processes, 
and geography were explicitly used in the defini- 
tion of the taxa instead of the soil properties 
themselves. 

The Australian classification before 1960 
(Stephens, 1954), the Israeli soil classification 
before 1979 (Dan and Koyoumdjiski, 1963), the 
French classification (CPCS, 1967), and the 
American classifications before the publication 
of Soil Taxonomy (USDA, 1975) all were influ- 
enced strongly by the Russian soil classifications 
and had the same limitations. 

The publication of Soil Taxonomy marked the 
start of a new era in soil classification. The 
main innovations that stimulated broad inter- 
national acceptance were: 

- The use of soil properties for classification. 

- The use of precisely defined diagnostic crite- 

ria and horizons to define the taxa. 

- The use of a rational nomenclature. 

- The organization as a key. 

However, the use of soil climate to define high 
level taxa has been criticized, and features asso- 
ciated with moisture regimes were proposed as a 
preferable alternative (Sombroek,1985 ). 

The latest update of the FAO-UNESCO soil 
map of the world (FAO,1988) supressed all ref- 
erence to soil climate in the definition of taxa. 
In lieu of that, a yermic phase was defined based 
on soil properties. 

In this paper the use of the aridic moisture 
regime for classification purposes will be dis- 
cussed first. Then desert soil forming processes 
and the associated soil properties will be re- 
viewed, leading to the definition of a diagnostic 
epipedon. 



Oman. 



Scientist, F.A.O., P.O. Box 5287, Sultanate of 



175 



176 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



LATERAL SUPPLY 



The Aridic Moisture 
Regime 

Definition of the Aridic 
Moisture Regime 

In the aridic (torric) mois- 
ture regime, the moisture 
control section in most years 
is: 

a.Dry in all parts more 

than half the time (cu- 
mulative) that the soil 

temperature at a depth 

of 50 cm is above 5C; 

and 
b. Never moist in some or 

all parts for as long as 90 

consecutive days when 

the soil temperature at a 

depth of 50 cm is above 

8C(USDA-SCS, 1975). 
"Soils that have an aridic 
or torric moisture regime are 
normally in arid climate." 
This statement extracted 
from Soil Taxonomy (USDA- 
SCS, 1975) emphasizes the 
close link which exists in 
most cases between the aridic 
soil moisture regime and the 
atmospheric climate. The 
question that arises then is 
whether or not the aridic 
moisture regime is a soil 
property. The analysis of the factors influencing 
soil moisture regimes in general may provide an 
answer to this question. 

Factors Influencing the Soil Moisture 
Regimes 

Figure 1. attempts to show the factors influ- 
encing the soil moisture regime as a system, by 
displaying the factors as a series of "valves". 

Valve 1 represents the atmospheric climate 
which controls entirely the precipitation and 
strongly the evaporation and transpiration. 

Valve 2 represents the geomorphological fac- 
tors controlling runon (water runoff from adja- 
cent soils). 

Valve 3 represents the soil-atmosphere inter- 
face, i.e., the soil surface. It controls infiltration 
and evaporation. Slope and infiltration rate are 
the most relevant soil characteristics. 

Valve 4 represents the soil characteristics 
that influence the transfer of water through it. 



ATMOSPHERIC 

CLIMATE 



TRANSPIRATION 




RUNOFF 



LATERAL DRAINAGE 



SUPPLY 
FROM GROUNOWATER 



DEEP PERCOLATION 



Figure 1: Factors influencing the soil moisture regime. 



Saturated and unsaturated hydraulic conduc- 
tivity are the most important in this regard. 

Valve 5 represents the hydrological and geo- 
morphological factors controlling moisture sup- 
ply from groundwater. 

Valve 6 represents the vegetation factor 
which strongly controls transpiration. 

Valve 7 represents the storage of water which 
is controlled by soil properties, e.g., water hold- 
ing capacity. 

It appears then that the moisture intake and 
storage by the soil is controlled by soil proper- 
ties, whereas the moisture supply is essentially 
an independant variable. The equilibrium of 
this system and its pattern of change with time 
constitute the soil moisture regime. 

It also appears from Figure 1 that, if valve 1 
(atmospheric climate) is closed, the soil, what- 
ever its properties and in the absence of ground- 
water supply, will remain dry. 

The aridity of desert soils is directly caused by 
the aridity of the atmospheric climate without 



SOUIRJI: CLASSIFICATION OP ARIDIC SOILS, PAST AND PRESENT: PROPOSAL OP A DIAGNOSTIC DESERT EPIPEDON 



177 



the involvement of soil properties or pedogenic 
processes. However, in some areas bordering 
deserts and having a semi-arid climate, some 
strongly sloping or crusted soils may have an 
aridic moisture regime. Since these occurences 
are very limited in extent, it can be stated that 
the aridic moisture regime is a property of the 
climate rather than of the soiL 

The Determination of Soil Moisture 
Regimes 

Although moisture regimes can be measured 
in situ, the procedure is costly, time consuming, 
and unsuitable for soil surveys, since at least 
one decade of measurement would be necessary 
to obtain statistically significant results. 

Therefore, mathematical models were devel- 
oped to compute moisture regimes from atmos- 
pheric climatic data. The model of Newhall 
(1980) is widely used. The only soil property 
taken into account in this model is the water 
holding capacity. The soil is considered as an 
inert reservoir; therefore the variations of the 
moisture storage are entirely determined by the 
atmospheric climate. 

It is likely that, in the future, better models 
will be elaborated, but they always will depend 
on climatic data, which are scarce in most arid 
areas. 

The Use of the Moisture Regime for Soil 
Classification 

Soils are the objects to be classified; therefore 
only soil properties are to be used as diagnostic 
criteria. 

Because of the important bearing of the mois- 
ture regime on soil genesis, moisture regime can 
be used to decide which soils belong together. 
The next step is to find out associated morphom- 
etric criteria to define the grouping. More re- 
fined classes of moisture regimes could be used 
at the family level to enhance the agronomic 
significance of the lower taxa (Sombroek, 1985). 
Indeed, the current classes of soil moisture re- 
gime are too broadly defined to allow useful ag- 
ronomic interpretations within the same cli- 
matic zone. 

Desert Soil Forming Processes and 
Associated Properties 

Strong moisture deficit, extreme tempera- 
tures, and high wind activity are the main fac- 
tors controlling soil formation in deserts. These 
factors, alone or in combination, are reflected in 



desert soils' composition, morphology, and be- 
havior. 

Organic Matter In Desert Soils 

In a virgin soil, the organic carbon content 
reflects the equilibrium between (mostly) vege- 
tal biomass production and decomposition. 

Vegetal biomass production depends on the 
availability of moisture and heat for a duration 
long enough for the vegetation to accomplish its 
biological cycle. 

In some deserts such favourable conditions 
occur in most years, whereas in others they oc- 
cur rarely and only locally. The dynamics of 
organic matter would therefore be quite differ- 
ent in deserts with a regular growing season 
and those without. 

Deserts With a Regular Growing Season 

These deserts are, rather, semi-deserts and 
are encountered in areas transitional to semi- 
arid. Some of the North- American deserts, such 
as those of Wyoming or Idaho, belong to this 
category. Precipitation is rather high, often 
more than 200 mm, and rainfall occurs in spring 
and early summer when soil temperature is op- 
timal. As a result, a dense vegetation of shrubs, 
forbs, grasses, and mosses develops. 

A sod can form and litter accumulates in the 
topsoil, protected from wind deflation by the 
dense bush cover. The soil fauna can also de- 
velop and contribute to the decomposition and 
incorporation of organic matter in the soil. 

In these semi-deserts, high amounts of or- 
ganic carbon, often more than 2-3 percent, occur 
in the A horizons. 

Deserts Without a Regular Growing 
Season 

These are the true deserts. They are either 
hot or cold but the amount of precipitation, gen- 
erally well below 200 mm, and the rainfall pat- 
tern are such that moisture supply to the vege- 
tation is both limited and erratic. 

In these conditions, a sparse woody vegeta- 
tion of shrubs tends to be dominant, to the detri- 
ment of grass, and a sod cannot form. Due to the 
scarcity and type of vegetative cover, the rate of 
litter production is very low and removal by 
wind and surface wash strongly reduce its in- 
situ accumulation. 

The initial decomposition of litter into smaller 
particles is normally done by insects, collem- 
bola, and other small invertebrates before fur- 
ther decomposition by microorganisms occurs. 



178 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



These primary decomposers are very scarce in 
true deserts. 

The final decomposition of organic matter is 
governed by microbial activity which itself de- 
pends on the available moisture, aeration, and 
temperature. Since the soils are dry most of the 
time and the temperature is often either very 
high or very low, desert environments are not, 
except for aeration, favorable to high decomposi- 
tion rates. Birch (1958) showed that, when a 
dry soil is moistened, a flush of decomposition 
occurs. Hence, the alternation of drying and 
moistening would favor rapid decomposition of 
organic matter. However, in true deserts rainy 
episodes are usually rare and brief, and there- 
fore the global rate of decomposition is rather 
low. 

Earthworms, which usually contribute sig- 
nificantly to the incorporation of organic matter 
into stable organomineral compounds, are to- 
tally absent from most desert soils. 

It seems, then, that the low organic matter 
content of true desert soils is primarily due to 
low litter production, dispersion, and low incor- 
poration, rather than to high decomposition. 

In true deserts the organic carbon content of 
the A horizons is usually well below 0.6 percent, 
and if the texture of the A horizon is loamy sand 
or coarser the organic carbon content is 0.2 per- 
cent or less (Souirji, 1987b; FAO, 1988). How- 
ever, in the driest deserts, such as those of the 
Arabian Peninsula or the Sahara, the organic 
carbon content of the A horizons is often less 
than 0.1 percent no matter what the texture is 
(Dutil, 1971; Souirji, 1990 in prep.). These val- 
ues are in good agreement with previously pub- 
lished data (Lobova, 1960; USDA, 1975; Kovda 
et al., 1979). 

Organic carbon content increases with mois- 
ture availability owing to poor drainage, irriga- 
tion, or both. It then can be higher than the 
above-mentioned values, but if so salinity in- 
creases beyond 4 ds/m (saturated paste extract) 
within 125 cm below the soil surface (Souirji, 
1987 b; FAO, 1988). 

Exceptions are encountered, however, in 
cases of soils having a good internal drainage 
flooded with low salinity waters. In virgin des- 
ert soils, this situation was only found to occur 
in some soils of flooded depressions (Souirji, 
1990 in prep.). 

It is noteworthy that organic matter contents 
can be higher in the subsoil than in the topsoil 
and that irregular distribution in the profile is 
common. 



The organic matter of typical desert soils con- 
tains more fulvic than humic acids and more 
waxes and resins than other soils (Lobova, 1960; 
Kovda et al., 1979). Iron, rather than calcium, 
seems to be the bonding agent between the or- 
gano-mineral compounds (Lobova, 1960). 

Physico-Chemical Behavior of the 
Air-Liquid-Solid Phase System 

In temperate areas and the humid tropics, the 
relative humidity of the air is usually high, 
around 70 percent or more. Lower relative 
humidity does occur but only for short dura- 
tions. 

Except in some coastal areas, air relative 
humidity is low in deserts. For example, in the 
arid part of Saudi Arabia, mean monthly rela- 
tive humidity is below 45 percent most of the 
year and below 25 percent for several consecu- 
tive months (Al-Zeidi et al., 1988). 

The pF of a soil in equilibrium with such rela- 
tive humidity (less than 48 percent) is above 6. 
Therefore all the pores having a diameter of 15 
microns or more are filled with air (Tessier, 
1984). 

A very high air content and a low water con- 
tent for extensive durations are very important 
characteristics of aridic soils. They are respon- 
sible for the occurence of physical dispersion and 
of a special type of chemical weathering. 

Physical Dispersion 

Structural crusts occur in soils of various cli- 
matic areas but are much more widespread in 
arid and semi-arid regions. "They are normally 
more compact, harder and more brittle than the 
soil just beneath them" (Evans and Buol, 1968). 

Structural crusts can form under the influ- 
ence of externally applied mechanical pressures 
(e.g., rain drops impact) or through the rear- 
rangement of the soil fabric following slaking 
upon wetting. 

It has been shown experimentally that the 
slaking of the soil upon wetting is due to either 
compression of air in the pores or to rapid swel- 
ling or both (Quirk, 1950; Tessier, 1984; Le Bis- 
sonais, 1989) . 
Compression of Air 

The above-mentioned authors and others 
have shown that, the lower the initial soil water 
content, the stronger its tendancy to slake upon 
wetting. By wetting the soil alternatively under 
atmospheric pressure and under vacuum, they 
have found that the compression of the air filling 



SOUIRJI: CLASSIFICATION OP ARIDIC SOILS, PAST AND PRESENT: PROPOSAL OP A DIAGNOSTIC DESERT EPIPEDON 



179 



the pores by the advancing moisture front 
causes the collapse of the aggregates. The in- 
tensity of slaking by air compression depends on 
the soil air content, which is determined by the 
climate and the pore-size distribution. 

In temperate climates, structural crusts occur 
mostly in silty soils, because these have larger 
pores that can dry out and become filled with air 
even during moderately dry climatic episodes, 
e.g., European summers. 

The surface soil in arid zones is usually ex- 
tremely dry (filled with air) for extensive periods 
and therefore presents ideal conditions for slak- 
ing by air compression. Hence, structural crusts 
occur in a much wider range of textures than in 
humid areas (Souirji, 1990 in prep.). 

Structural crusts of arid regions are generally 
vesicular (Jackson, 1957; Springer, 1958; Lo- 
bova, 1960; Miller, 1971; Kovda et al., 1979; 
Nettleton and Peterson, 1983; Souirji, 1987b, 
1990 in prep.). This peculiar feature is due to 
the fact that arid soils have a much higher air 
content and that the uppermost part of the crust 
dries out immediately after the rain stops, hence 
trapping the air which is bubbling up in the dis- 
persed soil. 

Swelling 

Clay minerals are known to have an internal 
deficit of positive electrical charges caused by 
isomorphous substitution of Si 4+ and A1 3+ ions by 
other cations of lower valence. The electrical 
deficit is compensated for by the adsorption of 
counter ions of an equal opposite charge on the 
surface of the mineral 

In presence of water, the adsorbed cations 
tend to diffuse away, but the electric attraction 
of the negatively charged mineral's surface 
forces them to remain in an adjacent layer of the 
soil solution. A diffuse double layer (D.D.L.) is 
formed which differs from the outer soil solution 
in that it contains a large excess of cations over 
anions. 

In case water is removed from the soil to the 
point that the D.D.L. shrinks below its potential 
thickness, we are in the presence of what Bolt 
(1976) has called a truncated diffuse double 
layer (T.D.D.L.). The greater the difference be- 
tween the potential (maximum) extent and the 
actual extent of the D.D.L., the stronger the 
tendency of the system to expand. 

Translated in terms of physical behavior, the 
system is similar to an osmometer, and it will 
develop, upon wetting, a very strong swelling 
pressure of the order of several tens of bar. This 



swelling pressure is higher if (1) the adsorbed 
counter ions are monovalent, (2) the soil solu- 
tion has a low electrolyte concentration, e.g., 
rainwater, and (3) higher specific area (e.g., 2/1) 
clay minerals are present (Bolt, 1976). 

In the arid regions, conditions are favorable 
for the occurence of T.D.D.L because the soils 
are usually very dry and sodium and potassium 
have a larger share of the base saturation than 
in humid areas. 

Consequences of Physical Dispersion 

Strong physical dispersion has the following 
consequences: 

- weak structure (often platy) in the topsoil 
and crusting 

- higher mobility of the clay fraction which 
can be leached (Souirji, 1990 in prep.). 

Indeed, strong dispersion of the clay fraction, 
if the topsoil is permeable, may cause the forma- 
tion of shallow argillic horizons. Also, since the 
peds are dispersed, clay coatings cannot form. 

Chemical Weathering 

In a soil that becomes very dry, the residual 
water has properties very different from an ordi- 
nary soil solution. 

The hydrogen atoms of the soil water mole- 
cules become about one million times more mo- 
bile, hence increasing their chemical activity, 
which may become equivalent to an H* concen- 
tration of about 0.5 N (Chaussidon and Pedro, 
1979). 

This strong variation of the pH of the soil so- 
lution, which also is accompanied by a variation 
in cations and anions concentration, may induce 
selective exchange with cations which are in iso- 
morphic substitution (e.g., adsorbed Mg ex- 
changing Fe). 

The elevation of the pH of the soil solution by 
two or more units upon wetting also may en- 
hance weathering of silica. The occurence of 
amorphous silica films on quartz particles of 
desert eolian sands is well documented (Le 
Ribault, 1977). If glass or opal (e.g., from roots 
decay) are present, one can speculate that, un- 
der an arid climate, marginal to ustic or xeric, 
weathering and leaching of silica may become 
strong enough to form duripans. 

Varnish on pebbles and stones of desert pave- 
ments also is probably formed by weathering 
caused by residual water from evaporation of 
moisture films deposited by the dew (Souirji, 
1990 in prep.). 



180 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



Aeolian Additions and Turbations 

The scarcity of vegetation and the occurence 
of strong thermic gradients cause the formation 
of violent and frequent winds that are charac- 
teristic of deserts. 

Because of the existence of large areas of bar- 
ren rocks undergoing physical weathering, des- 
ert winds carry large amounts of sand, which is 
redeposited as dunes or more commonly as 
small hummocks around vegetation or as thin 
sand Veils' on the soil surface. 

Playas and wadi beds can contribute substan- 
tial amounts of airborne clay and silt, which are 
redeposited in the neighboring areas or hun- 
dreds of kilometers away. Aeolian additions of 
clay, silt, sand, and salts to the soils are well 
documented (Lobova, 1960; Yaalon, 1973; Net- 
tleton and Peterson, 1983; Souirji, 1987a and b). 

Aeolian sands have a peculiar morphology 
(Cailleux et Tricart, 1959; Le Ribault, 1977). 
The individual quartz and feldspar particles are 
rounded (not necessarily spherical) and have a 
mat (dull) surface. Roundness is more frequent 
in particles larger than 0.250 mm. Souirji 
(1987a and b) proposed to use the occurrence of 
such aeolian sand as a diagnostic criterion for 
aridic taxa. 

Ventifacts, varying from perfect dreikanters 
to glossy abraded pebbles, commonly are found 
in desert pavements. The degree of wind shap- 
ing depends on the age of the pavement and its 
lithology. 

Recently deposited sandy alluvium often is 
reworked by wind and shows characteristic 'eo- 
turbations' in the form of cross stratifications 
(Souirji, 1987b). 

Salt Accumulation 

Because of limited leaching, the soils of arid 
regions often contain accumulations of carbon- 
ates, sulfates, chlorides, and nitrates. These 
accumulations can occur only if there is a source 
of the above-mentioned salts. The sources may 
be allochthonous (e.g., airborne dust) or autoch- 
thonous (e.g., parent materials or shallow water 
table). 

Differences in solubility product make the 
carbonates accumulate at shallow depth, 
whereas sulfates, chlorides, and nitrates accu- 
mulate at increasing depth. However, this is 
only valid for simultaneous deposition, and dif- 
ferent salts can be found in the same horizon if 
they were deposited under different moisture 
regimes. 



Many desert soils contain shallow lime accu- 
mulations separate from deeper relict calcic ho- 
rizons (Figure 2). These 'duplex* profiles are 
quite common (Lobova, 1960; Souirji, 1990 in 
prep.). 

Gypsum accumulations, when they occur, of- 
ten underlie calcic horizons. 

Acicular Clay Minerals 

The presence of acicular clay minerals, mostly 
palygorskite, in desert and semi-desert areas 
was reported by many authors (Millot et. aL, 
1969; Eswaran and Barzanji, 1974; Singer and 
Norrish, 1974; Aba-Husayn and Sayegh, 1977). 

Some authors attributed their origin to au- 
thigenic formation, whereas others favored in- 
heritance from the parent material or wind- 
borne additions. Conrad (1969) and Dutil 
(1971), who studied the soils of the Algerian 
Sahara, found a clear association between paly- 
gorskite occurence and sebkha (playa) or fluvio- 
lacustrine deposits. 

Zelazny and Calhoun (1977) identified pre- 
requisites for the occurence of palygorskite and 
sepiolite as a relatively closed soil system with 
minimal leaching with little or no influx of Al 
and low H 3 but high Si and Mg concentrations. 
These conditions are fulfilled in many playas 
and temporary lakes of the arid zone. 

Indeed, deserts, which generally have an en- 
doreic (not draining to the sea) hydrographic 
system or lack one, are extremely favorable to 
the formation of playas and lakes, especially 
when moister climatic periods follow drier ones. 
Such a climatic alternation is known to have 
occuredin most deserts; therefore, palygorskite- 
rich sediments could be quite common in these 
environments. 

These sediments can become palygorskite- 
rich soils and/or a source of palygorskite- (sepio- 
lite-) rich airborne dust to surrounding areas. 

However, the occurrence of weathering in 
surface horizons of desert soils, bringing in the 
soil solution soluble silica and magnesium, may 
justify the existence of pedogenic palygorskite. 

Whatever their origin, acicular clays when 
present in soils are a sign of very limited leach- 
ing, i.e., of aridity, unless they are protected in- 
side hard, impervious structures such as nod- 
ules or petro-calcic and petrogypsic horizons. 
Therefore, the presence of palygorskite has been 
proposed as diagnostic criterion to be used in 
conjunction with other criteria to define aridic 
taxa (Souirji, 1987a; FAO, 1988). 



Sounui: CLASSIFICATION OF ARIDIC SOILS, PAST AND PRESENT: PROPOSAL OF A DIAGNOSTIC DESERT EPIPEDON 



181 



Color of Desert Epipedons 

Desert soils generally have light-colored sur- 
face horizons. This is probably due to low or- 
ganic matter content and its type, the presence 
of iron oxides, and, often, a high lime content. 

A review by Souirji (1990, in prep.) has shown 
that desert A horizons generally have Munsell 
color value 3 or more when moist and 4.5 when 
dry, and a chroma of 2 or more when moist. 

Proposed Desert Epipedon 

General Considerations 

The shallow penetration of moisture in desert 
soils restricts current soil development to the 
upper horizons. Pedogenic development in 
deeper soil horizons of well drained desert soils 
is generally a relict of past moister climates. 
Epipedons are therefore the best markers of 
current soil fomation in deserts. 

A definition for a diagnostic desert epipedon is 
given in the next paragraph. This definition is 
designed the same way Soil Taxonomy (USDA, 
SCS, 1975) defines the mollic epipedon, in order 
to show similarities in the rationales. 

Definition of the Torric Epipedon 

Abstraction of properties common to 'mature' 
automorphic soils of the drier arid zones focuses 
immediate attention on the horizons at or near 
the surface rather than deeper ones. Virtually 
all these soils exhibit a relatively thin, light-col- 
ored, humus-poor surface horizon or horizons in 
which divalent cations are dominant on the ex- 
change complex and the grade of structure, of- 
ten platy parting to subangular blocky, is weak 
to moderate. This kind of horizon may be 
named the torric epipedon. 

From a genetic point of view, the properties of 
the torric epipedon are thought to develop under 
the influence of extreme dryness, extreme tem- 
peratures, and high wind activity. The forma- 
tion and accumulation of stable organo-mineral 
compounds are limited and a high percentage of 
the organic matter is fulvic acid, waxes, and 
resins. 

The soils having a torric epipedon have a 
widely varying mineralogical composition of the 
clay fraction but palygorskite, and less so sepio- 
lite, is frequently encountered. 

These soils are too dry for the cultivation of 
crops unless irrigated. Under cultivation, they 
have an adverse physical behavior, slake upon 
wetting, and are highly subject to wind erosion. 



Although the torric epipedon is a surface hori- 
zon that can be truncated by erosion, its many 
important accessory properties suggest its use 
as a diagnostic horizon at a high categorical 
level. 

The torric epipedon is defined in terms of its 
morphology rather than its genesis. It consists 
of mineral soil material. It is a surface A hori- 
zon or horizons unless it underlies a deposit of 
largely unaltered new material less than 50 cm. 
thick. It does not qualify as an anthropic, histic, 
mollic, plaggen, or umbric epipedons and has 
the following properties: 

1. The organic carbon content is less than 0.6 
percent if the texture is loamy very fine 
sand or finer and less than 0.2 percent if 
coarser (weighted average). If the soil has, 
somewhere within 125 cm below the sur- 
face, an electrical conductivity of the satu- 
rated paste extract of 4 ds/m or more, the 
conditions on organic carbon are waived. 

2. Both broken and crushed samples have 
Munsell color value 3 or more when moist 
and 4.5 or more when dry, a chroma of 2 or 
more when moist. 

3. Soil structure is weak or moderate and the 
consistence is soft when dry, although indi- 
vidual peds may be hard. 

4. After a rain or an irrigation, a structural 
crust forms at the surface of the aridic 
epipedon. It generally has vesicular pores 
and a sandy loam or finer texture. 

5. Base saturation is at least 75% by NH4 
OAc method. 

6. If the torric epipedon is calcareous, then it 
contains evidence of lime segregation in the 
form of coatings below gravels and coarse 
sand, pseudomycelia, or soft accumula- 
tions. 

7. If there is a source of sand in the landscape, 
the sand fraction in some subhorizon or in 
inblown material filling the cracks contains 
a noticeable proportion of rounded or 
subangular sand particles showing a mat 
(dull) surface. These particles make 10 
percent or more of the medium and coarser 
quartz sand fraction. 

8. If a surface accumulation of coarse frag- 
ments (pavement) is present, some of them 
are shaped by the wind or show the pres- 
ence of iron and manganese oxides accumu- 
lation (desert varnish) on their exposed 
surfaces or gypsum/calcium carbonate/so- 
dium chloride below. 



182 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



10. The lower limit of the torric epipedon is 
the upper limit of underlying B or C hori- 
zons. If the soil is cultivated and the Ap 
lies directly on a B or C horizon, the lower 
limit of the plow layer is also the lower limit 
of the torric epipedon. 

Description of a Typic Torric Epipedon 

A fully developed torric epipedon (Figure 2) 
has the following succession of features: 

- A pavement composed of one or two layers of 

pebbles or stones, often wind-shaped (1) 
and/or varnished. Most of the coarse frag- 
ments are embedded in the underlying ve- 
sicular crust, but some of them are 'floating' 
in loose sand. In old pavements the lower 
part of the coarse fragments is often weath- 
ered and has an orange tinge or accumula- 
tion of gypsum (2). 

- A sandy layer 0.5 to several cm thick (3). 
The upper part is often coarse sand and 
the grain size becomes gradually finer be- 
low. This sandy layer is sometimes discon- 
tinuous or absent. 

- A thin layer 1 to 3 mm thick, generally mas- 

sive, composed of clay, silt, and fine sand 
infiltrated from above (4). It is difficult to 
distinguish from the under-lying vesicular 
crust except above the cracks. 

- A vesicular crust 0.3 to a few cm thick, hav- 

ing the same textural composition as the 
layer above (5). The vesicles are generally 
1 to 3 mm in size. This crust is frequently 
thinly stratified and its transition with the 
underlying horizon is generally, but not 
always, abrupt. 

- Wedged-shaped cracks (6), often filled with 
inblown sand, divide the above-mentioned 
crust and the underlying horizon(s) into po- 
lygonal prisms. When the sand Veil' is 
removed carefully, the soil surface appears 
patterned into polygons by thin cracks (7). 

- A weak to moderate platy (8), often parting 

to fine subangular blocks, horizon. Its 
thickness is generally less than 10 cm but 
can sometimes be as thick as 25 cm. When 
it is calcareous, lime segregation occurs in 
the form of pseudo-mycellia, coatings below 
pebbles, or soft accumulations, which gen- 
erally increases in the lower part of the 
epipedon (9), which is often an Ak horizon. 

- The lower horizons of old desert soils often 
contains a second, generally better devel- 
oped, calcic horizon (10). 




Figure 2. Morphology of a typical desert soiL 
(For legend, see text.) 



Identification of the Torric Epipedon 

The USDA (1975) defines recently deposited 
soil material (C horizon) as being thinly strati- 
fied. This definition is not adequate for sandy 
eolian materials because mass flow destroys the 
stratifications in very little time. Therefore, the 
absence of thin stratifications alone is not 
enough to indicate the presence of an A horizon. 

The mere presence of roots in a sandy surface 
horizon is also not sufficient to indicate enough 
soil development to form an A horizon, because 
roots can form anywhere moisture, nutrients, 
and air are available. 

The accumulation of organic matter in desert 
surface horizons being generally very limited or 
even less than in subsurface horizons, it cannot 
be used as criterion to define an A horizon. 

In deserts, as soon as a surface soil horizon 
becomes stable, dust deposition and physical 
dispersion induce the formation of a structural 
crust. This structural crust controls all subse- 
quent soil developement and gradually the 
other properties of the torric epipedon appear 
(Souiiji, 1990 in prep.). Actually the only type of 
A horizon that can form in desert environments 
is the torric epipedon, which can be considered 
the desert ochric epipedon. 

In the definition of the torric epipedon, refer- 
ence is made to "rounded or subangular sand 



SOUIRJI: CLASSIFICATION OP ARIDIC SOILS, PAST AND PRESENT: PROPOSAL OP A DIAGNOSTIC DESERT EPIPEDON 



183 



particles showing a mat (dull) surface make 

10 percent or more of the medium and coarser 
non-carbonate sand fraction." Since this deter- 
mination is not a routine procedure in soil sci- 
ence, it is useful to give some details. 

A composite sample of the A horizons is 
washed with water on a 250 microns mesh sieve 
to keep only the medium and coarser sand frac- 
tion. The sand is then put in a beaker and boiled 
with enough HC1 acid to eliminate carbonates 
and iron oxides. The composite sample weight 
has to be adjusted in order to obtain about 100 
sand grains. 

The clean sand grains are then mixed and a 
subsample of about 30 grains is examined with 
a strong hand lense (20X) or a binocular micro- 
scope. In Saudi Arabia at least 80 percent of the 
sand grains have a sub-angular or rounded and 
mat surface (Souirji, 1987, 1990 in prep.). If 
eolian sand is encountered in the landscape, the 
presence of mat sand can be assumed. 

Conclusions 

Soils having a torric epipedon generally be- 
long to the current order of aridisols (U.S.D.A., 
1975) and have subsurface diagnostic horizons. 
However, entisols of stable geomorphic surfaces, 
mostly torriorthents and torripsamments, do 
have a torric epipedon. The author suggests 
keeping these soils in the order of entisols and 
defining special great groups for them. Indeed, 
the presence of a torric epipedon has an impor- 
tant genetic and management significance and 
therefore should be reflected in soil classification 
and subsequently in soil maps. 

The use of the torric epipedon allows us to 
substitute field observation and simple labora- 
tory determinations for uneasy soil climate 
measurements. It has also the advantage of 
drawing attention to the properties of the topsoil 
which too often are neglected in the profile de- 
scriptions. Structural crusts are a good example 
of features that are often overlooked, although 
they reduce infiltration and can hamper or pre- 
vent seedling emergence. 

References 

Aba-Husayn, M. and Sayegh,A., 1977. Mineralogy of Al 
Hasa Desert Soils. Clay Minerals, 25: 138-147. 

Al-Zeidi et al, 1988. Guide for crop irrigation requirement 
in the Kingdom of Saudi Arabia. PAO - Ministry Of 
Agriculture And Water, Riyadh. 

Birch, H.F., 1958. The effect of soil drying on humus de- 
composition and nitrogen availability. Plant and Soil, 
19, 9-21. 



Bolt, G.H. andBruggenwert, M.G.M. (editors), 1976. Soil 

Chemistry A Basic Elements. Elsevier. 
Cailleux, A. et Tricart, J., 1959. Initiation a Tetude des 

sables et des galets, 3 tomes. Centre Docum. Univers. 

Paris. 
Chaussidon, J and Pedro, G. 1979. Role de 1'etat hydrique 

du systeme poreux sur 1'evolution du milieu. Realite de 

1'alteration dans les systemes a faible teneur en eau. 

Sciences Du Sol, numero double 2 et 3. 
Conrad, G., 1969. L'Evolution Continentale Post-Her- 

cynienne Du Sahara Algerien. Ph.D. thesis. Editions 

Du Centre National De La Recherche Scientifique, 

Paris, France. 
C.P.C.S., 1967. Classification Des Sols. Laboratoire de 

Geologie-Pedologie E.N.S.A. Grignon. Multi. France. 
Dan, J. and Koyumdjisky, H., 1963. The soils of Israel and 

their distribution. J. Soil. Sci., 14. 
Dutil, P. 1971. Contribution A L'Etude Des Sols Et Des 

Paleosols Du Sahara. Ph.D. thesis. Faculte Des Sci- 
ences de LTJniversite de Strasbourg, France. 
Eswaran, H. and Barzanji, A.F. 1974. Evidence for the 

Neoformation of Attapulgite in Some Soils of Iraq. 

Trans. 10th Int. Cong. Soil Sci., Moscow, 7. 
Evans,D,D and Buol,S,W, 1968. Micromorphological 

study of soil crusts. S.S.S-Am.Proc. Vol. 32 no 1. 
FAO., 1974-1981. Soil map of the world, vol. I-X, Unesco, 

Paris. 

FAO, UNESCO, ISRIC, 1988. Soil map of the world, Re- 
vised Legend. World Soil Resources Report 60. FAO, 

Rome. 

Hilgard, E.W., 1906. Soils. McMillan, New York. 
Jackson, E.A., 1957. Soil features in arid regions with 

particular reference to Australia. J. Austr. Inst. of 

Agric. Science. 196-209. 

Kovda, V.A., 1979. Soils processes in arid lands in Arid- 
Land Ecosystems, vol. 1. Goodall, D.W. and Perry, R.A. 

(editors). International biological Programme, 16. 

Cambridge University Press. 
Le Bissonais, Y, 1989. Analyse des processus de microfis- 

suration des agregats a Thumectation. Sciences Du Sol, 

Vol. 27 no 2. 
Le Ribault, L., 1977. L'exoscopie des quartz. Masson, 

Paris. 
Lobova, E.V., 1960. Soils of the Desert Zone of the 

U.S.S.R. Issued in translation from Russian by Israel 

Program for Scientific Translations, Jerusalem, 1967. 
Miller, D.E., 1971. Formation of Vesicular structure in 

soil. S.S.S. Am. Proc. Vol. 35, 1974. 635-637. 
Millot, G. et al., 1969. Neoformation de 1'attapulgite dans 

les sols a carapaces calcaires de la Basse Moulouya 

(Maroc Oriental). C.R. Acad. Sc. Paris, t. 268. 
Nettleton, W.D. and Peterson, F.F., 1983. Aridisols In: 

Wilding, L.P., Smeck, N.E. and Hall, G.F, (editors). 

Pedogenesis and soil taxonomy, II. The soil orders. 

Elsevier. 
Newhall, F., 1980. Calculation of soil moisture regime 

from climatic records. SMSS roneo. SCS, USDA, Wash- 
ington. 
Pye, K., 1987. Aeolian Dust And Dust Deposits. Academic 

Press. 
Quirk, J,P, 1950. The water stability of soil aggregates in 

relation to the water content at the time of wetting. 

C.S.I.R.O. Division of Soils, Report 12/50. 
Singer, A. and Norrish, K, 1974. Pedogenic Palygorskite 

Occurences in Australia. American Mineralogist, vol. 

59. 



184 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Sombroek, W.G., 1985. A quest for an alternative to the 
use of soil moisture regimes at high categorical levels 
in Soil Taxonomy. In Proc. of the 5th Intern. Class. 
Workshop, Sudan, 1982, Part I. Soil Survey Admini- 
stration, Sudan. 

Springer, M.E., 1958. Desert Pavement and vesicular 
layer of some soils of the Lahontan basin, Nevada. 
S.S.S. Am. Proc. 22, 63-66. 

Souirji, A., 1987a. Classification of Desert Soils. Land and 
Water Newsletter, 2, 1987, FAO, Rome. 

Souirji, A., 1987b. Classification of Desert Soils. Ministry 
of Agriculture and Water, Project PAO. UTFN/SAU/ 
015, Riyadh, Saudi Arabia. 

Souirji, A., 1990 in prep. Contribution To The Study Of 
Desert Soils, The Torric Epipedon. Ph.D. thesis. Fac- 
ulte Des Sciences Agronomiques de UEtat a Gembloux, 
Belgium. 

Stephens, C.G., 1954. The Classification Of Australian 
Soils. Proc. 5th Intern. Cong. Soil Sci. (Leopoldville), 4. 



Tessier,D, 1984. Etude experimentale de 1'organisation 
des materiaux argileux. Doctoral thesis. I.N.R.A., 
Paris. 

Thorp, J. and Smith, G.D., 1949. Higher categories of soil 
classification. Soil Science, 67: 117-126. 

Tiurin, V., 1965. The System of Soil Classification in the 
USSR, Main Stages in the Development of the Soil 
Classification System in the USSR. Pedologie, spec. 3. 

USDA. Soil Survey Staff, 1975. Soil Taxonomy. U.S. Gov- 
ernment Printing Office, Washington. 

Yaalon, D.H. and Ganor, E. 1973. The influence of dust on 
soils during the quaternary. Soil Sci., 116. 

Zelazny, L.W. and Calhoun, F.G., 1977. Palygorskite, 
Sepiolite, Talc, Pyrophyllite, and Zeolites. In "Minerals ' 
in Soil Environments" J.B. Dixon et al. editors, S.S.S. 
Am. 



Report on Site Specific Soil-Climate-Vegetation Relationships: Cold 

Vertisols-Aridisol ISCOM Tour (Montana - Idaho - Wyoming) 

G.J. Staidl and S.G. Leonard 1 
Abstract 

Rangeland vegetation community characteristics are highly varied due 
to the complex interaction between soil, ambient climate, and other envi- 
ronmental factors with specific vegetation requirements and community 
relationships. This study was conducted to investigate and develop soil- 
vegetation relationships for the rangeland soil sites of the Idaho, Montana, 
and Wyoming part of the Cool Vertisol-Aridisol ISCOM Tour. The results 
were incorporated in the tour guide for evaluation and use in the Soil Tax- 
onomy revision process. Documentation included soil pedon descriptions, 
soil characterization data, vegetation characteristics, climate data, and 
supporting literature. 

An on- site examination in the spring, 1989, emphasized the affect of soil 
physical and chemical properties in relation to climate on plants and com- 
munity characteristics. Each site reflected its own unique soil-climate- 
vegetation relationship. Geographic shifts in precipitation and air tem- 
perature patterns corresponded with vegetation shifts from grass domi- 
nance to shrub dominance. The soil physical and chemical properties (i.e., 
soil structure, bulk density, sodium absorption ratio, etc.), landscape, and 
microrelief have an effect on soil hydraulic conductivity and plant root size, 
abundance, and distribution within soil profile. The effects are evident in 
the variation in kind, amount, and proportion of vascular plants and associ- 
ated cryptogam communities within similar climate regimes. 

Soil taxonomic criteria provide a starting point for defining and extrapo- 
lating soil-climate-plant relationships. However, interactions with other 
environmental factors, more specific soil characteristics, and temporal in- 
fluences also are needed. 



Introduction 

The areas in northern and eastern Montana 
are generally described as a grama-needlegrass- 
wheatgrass (Boutelona-Stipa-Agropyron) poten- 
tial vegetation type in the central and eastern 
grasslands by (Kuchler, 1964). The remaining 
areas in western Montana, Idaho, and Wyoming 
lie within the sagebrush-steppe (Artemisia- 
Agropyrori) potential vegetation type of the 
western shrub and grasslands. The shift from a 
grassland potential vegetation to the sage- 
brush-steppe vegetation probably is tied more to 
timing and distribution of effective precipitation 
in relationship to growing season (Stoddart et 
aL, 1975) than to soils. 

The climographs (Kormondy, 1969) depicted 
in Figure la-e reflect the mean monthly precipi- 
tation and temperature patterns for Select Na- 
tional Oceanic and Atmospheric Administration 
(NOAA) Weather Stations that are representa- 
tive of the tour sites. The precipitation that falls 
when the air temperature is below 32 degrees F. 

'G.J. Staidl and S.G. Leonard are soil scientist, USDA, 
SCS, and range scientist, USDI, BLM, respectively, Na- 
tional Soil-Range Team, Univ. of Nevada Reno, 1000 Val- 
ley Road, Reno, NV 89512. 



generally is assumed to be at a time when the 
soil is frozen and in the form of snow. Where the 
area is windswept and the snow is blown off the 
soil surface, soil moisture recharge from precipi- 
tation occurring during this period is considered 
to be low. Depending upon the site location, this 
can occur over a period of three to 6 months. In 
areas where snowpack accumulations are com- 
mon, the reverse is also true. The importance of 
timing, amount, and distribution pattern of pre- 
cipitation when the soils are not frozen is re- 
flected in the efficiency of soil moisture recharge 
and kinds of vegetation communities. 

Within the grassland areas involved, the 
dominant plant species are relatively shallow- 
rooted. Much of the moisture comes during the 
growing season but does not result in deep soil 
percolation. The different range sites occurring 
within these areas therefore can be determined 
satisfactorily, for the most part, by the soil char- 
acteristics in the upper layers that affect mois- 
ture relationships (texture, structure, salinity, 
etc.) and by effective precipitation. 

Many of the grass species are constant be- 
tween range sites. The relatively gentle topog- 
raphy contributes to rather broad climatic and 



186 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



NOM STATIONS 

GLASGOW, MT 
MILES CITY, MT 
BWDGER MT 



1. 

ELEV MMT MAP 
(Ft) (F) (IN) 

2085 41.6 11.17 

237146.4 14.11 

3680 39 J5 112 



HQMSTATiONS ELB/ MAAT MAP 

FI) m OH) 

SAIMON.ID asoe? 3 

CHALUS.ID 427844.4 IB 

HOWE.SO 0S6242J i.4 

DSUJOH B MT 540S43.I 11.1 



7C 




10 



10 1.5 JLO 

MEAN MONTmy PREOPfTATtON 



2.3 




MEAM MONTHY PRECflTATION 



Figure 1. Climographs of representative NOAA Stations for each soil site stop along the tour. The vertical axis is 
temperature (F); the horizontal axis is precipitation (inches); numbers refer to months (l=January, 2=Febru- 
ary, etc.). Data are long-term monthly averages. 



vegetation gradients except where affected by 
microtopographic influences and mans manage- 
ment practices. The sites often are differenti- 
ated more by total potential annual production 
by air-dry-weight and the relative proportion of 
species within the climax plant community than 
by distinct species changes. 

Figure l.a. represents the moisture and tem- 
perature patterns for Glasgow, Miles City, and 
Bridger in eastern Montana. Note that the 
monthly precipitation exceeds .4 cm per month 
throughout the main growing season. This pat- 
tern, along with the timing and kind of storms, 
is most favorable to grass vegetation (Figure 2). 
Bridger Station (Warren site) has a dry-down 



period in July and August and, although grass 
dominates, shows a small increase in shrubs 
and a corresponding decrease in percent cover of 
grass. 

The sagebrush steppe is considerably more 
complex. Topography ranges from broad inter- 
mountain basins to rather narrow mountain 
valleys and mountain foothills. Elevation 
ranges within this type are great (699m - 
2432m) and sometimes locally steep. Soil par- 
ent material also is quite varied, including but 
not limited to lake sediments, sedimentary 
rocks, basalt, and granitics. The concept for 
most of the sagebrush steppe is that precipita- 
tion tends to be more in the form of winter snow 



STAIDL AND LEONARD: REPORT ON SITE SPECIFIC SOIL-CUMATE-VEGETATION RELATIONSHIPS 



187 



Flgufi to 


Flgur 14. R0UT8 U. 


WAA STATIONS ELEV MAAT IMP NOAA STATIONS ELEV MAAT MAP NOAA STATIONS ELEV MAAT MAP 


FI)(f] OK) (HHF) ON) (FT)(F) pN) 

BfGPiNET.WY 679834.6 8.01 LABARGE,Wf 660038.4 8.10 KEMMERER.WY 692739.0 10.19 


MERNA.WY 767534.2 15.2? FONTENElie, W* 6530 383 7.28 SAGE.WY 633237.8 10.19 




- f A 
















FONTE 


CLUE 0AM 
























7 
























^CA 


LABARGE 






y% %^Ti 




00 




BKIRhCY 






A/V 


STS^ \7 




Af, 












V 


UEAW. 








J; 


$ 






\K 








Ir A 








k \ 






i i 6 
















1 \ 














I ft I 1 




so 


> 


\\ 


gm 




V*^ 








M )/ 


' 




10 




)oi 






/ 


/j^S 




F 




/ )* // 






* 


/ / 
















/ 










/ 








/ 




40 






/ " /8 




40 






4C 




10 




7 








/ / 


/ ^ 






it 










1 


] / 


I 








f /w 


' / 
















t/ 


4 








'/' 


y 
















r 4 / 
/ 






$2 




/ ^ 


\ 




S2 






35 






/ 








/ 


uljL ^1 


\ 




30 


AS 






3fl 














/ 


/ ^ 


^ 








ft 










J 


11 








L 


r N 

% 


\ 






I 


11 


11 










11 








ii// 




^11 






J 










T 










\p 


A 


% 






1 


rt 








|| j 








to 


y 


3 






20 


! 






20 




}| j 








A 




V \ 




8 


2 








?^ 


^ 






A 




**** 


L ia 

^i 




2 I 




Jia 








2 ^Ji 








\\ 










1M 












4 

1 






10 


-- }/ 








10 










10 












j 


1^0 o 6 


.' 


1J> U 6 J 


\ \& 


MEAN MONTHLY PRECIPITATION MEAN MONTHLY PRECIPITATION MEAN MONTHLY PRECIPITATION 



and spring rains. A review of available climate 
data indicates this is not necessarily so at the 
eastern reaches of this vegetation type, but it is 
more arid than the grassland type. 

Western Montana and eastern Idaho (Figure 
l.b.) reflect the patterns similar to those shown 
in Figure l.a. but also show a drying trend. The 
Dillon area still reflects a dominance of grass 
cover (Figure 2), but the relative cover of shrubs 
is increasing. The monthly precipitation at this 
site remains at or above .4 cm per month 
throughout the majority of the growing season. 

In eastern Idaho the moisture pattern is 
somewhat dryer than at Dillon, MT, with the 
peak effective precipitation occurring during 
May and June and a dry-down occurring in July 
through October. The change in precipitation 
patterns has reduced the extended availability 
of surface soil moisture for grasses common to 
eastern Montana. This has allowed adaptable 
shrubs, with both lateral and taproot root sys- 
tems that use both ? jrface soil and deeper sub- 



soil moisture, to compete effectively with the 
grasses for available water during the growing 
season. The Leador ID area in Figure 2 reflects 
this trend with a near equal mix of shrubs and 
grass. 

Western Wyoming (Figures I.e., l.d., and I.e.) 
precipitation patterns, although dryer, are simi- 
lar, except the peak for moisture is generally in 
May, with a general drying starting in June. 
With surface soil moisture conditions less favor- 
able, the shrub communities (Figure 2) are more 
dominant, due to their competitive ability to use 
both surface and deeper soil moisture. 

Vegetation communities are often distinct 
within the sagebrush steppe, particularly be- 
tween the various sagebrush taxa. These vari- 
ous taxa have been used successfully as indica- 
tor species (in conjunction with productivity 
and/or associated species) for major soil proper- 
ties, as well as for range site differentiation. 
Figure 3 (Hironaka et aL, 1983) provides a con- 
ceptual relationship of sagebrush species and 



188 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



PERCENT COVER 


, , GRASS SH* 088 ~~~ FORBS 


40 - 




CRESTED Wl 


IEAT 


















/ 


VSEEOINGS 










\ 

\ 


/I 


s x 




30 - 




\ 


/ 


k 




f 


\ 


ii 

n 


\\ 




I 




\ 


/ 


\ 






^ 


n 


\ \ 

\\ 




b 








\ 






\ 


* t 


A 




jjjIO - 

QC 










"^>^ 




\ 


1 1 


^ 




IE 










/ / 


^**\:**<* 




/ 


it ,^^ 












/ 


/ 


>w ' 


\^^^ 


1 














/ 


/ 


^ 


*\ 


t 






10 - 


NN 


^ 




/ 


> 




\ 


i 








\ 




****%, ^ 


^ ^^ 






^ 


f 








X 


-'"" 




y 














^ 
LOC GIA 
I 


GO* * 
(T 1 


;cny WAf 

(T \ 


RN DH 
(T U 


ON looofl WWEY sewer fom 


WlE KEUUERER KEMMCAEFt 

nr iW' wf 

ttll J7MD1 fT&Vrt 


HIV, P*M) PWJ (H*n P* 7 * 1 ^' W| *"""' > ' ' *- " ' " SootfiwtitVVywiina 


WjtlwjlW>imftfr-j ^^ gjj:^^^^ W4tT ECENTRN.MI NW-WY NW-W W,C9aWL-Wr SYW SW-W 


Figure 2. Total percent cover of grass, forb, and shrub vegetation for each tour site. 


Data was taken from Table 1. 



subspecies with soil 
characteristics as 
observed in Idaho* 

West (1979) pro- 
vides some general 
relationships be- 
tween elevation, soil 
moisture, and vari- 
ous sagebrush spe- 
cies in the Great 
Basin and Colorado 
Plateau physiogra- 
phic regions (Figure 
4). Other authors, 
such as Zamora and 
Tueller (1973), Pas- 
sey et al. (1982), and 
Sasich and Nielson 
(1984), to name a 
few, have also dis- 
cussed soil, plant, 
and climate relation- 
ships of sagebrush 
species or communi- 
ties in different loca- 
tions. 

While there are 
consistencies in some relationships, there are 
also apparent differences expressed by the 
above authors. The soil, plant, and climate rela- 
tionships do vary slightly between geographic 
areas, and exceptions can be found within any 
geographic area. The geograpic differences can 
be accounted for, in part, by ecotypic variation 
within species or studies conducted at geo- 
graphic extremes in the adaptive range of a spe- 
cies. More importantly, there are many com- 
pensating factors between soil, climate, topogra- 
phy, and other environmental factors that pro- 
vide a physiologically similar environment for 
plant species (Leonard et al., 1988). Soil Taxon- 
omy criteria may not always correspond with 
specific physiological responses of range plants. 

Our preoccupation with soil, plant, and cli- 
mate relationships and management implica- 
tions associated with vascular plant communi- 
ties has virtually ignored or at best barely recog- 
nized the "other" plant community component of 
most arid and semiarid ecosystems - the crypto- 
gamic or nonvascular plant component. 

Harper and Marble (1988) point out the im- 
portance of the cryptogamic plant community 
best, as follows: 

Arid and semiarid rangelands commonly have 
blue-green algae, lichens, and mosses that often 



cover as much or more of the soil surface as vas- 
cular plants. Evidence suggests that most 
cryptogamic plants complement the effects of 
vascular plants relative to soil stability, water 
infiltration, and greater availability of nitrogen. 
Situations are documented in which crypto- 
gamic covers (mainly blue-green algae) have 
improved establishment and growth of vascular 
plant seedlings. 

The authors above also reference literature 
documenting certain potentially negative effects 
(i.e. allelopathy) associated with some crypto- 
gams. It appears that there are nonvascular 
plant species that have both desirable and unde- 
sirable values or characteristics, much like, 
some species with the vascular plant commu- 
nity. It also appears that there are definite soil, 
climate, and topographic relationships affecting 
these species. 

Methods and Materials 

Each site identified as rangeland in the 
Idaho, Montana, and Wyoming part of the 
Aridisol tour were visited by the Soil Scientist 
and Range Scientist members of the National 
Soil-Range Team. Documentation available in- 
cluded the soil pedon descriptions and Soil Con- 



STAIDL AND LEONARD: REPORT ON SITE SPECIFIC SOIL-CLIMATE-VEGETATION RELATIONSHIPS 



189 



DWARF 




DWAffF 
SAGEBRUSHES 




ARIOIC 



XERIC 



UOIC 



SOIL 



MOISTURE 



GRADIENT 



Figure 3. Conceptual relationship of sagebrush species 
and subspecies based on soil moisture, temperature 
and other soil characteristics. 



servation Service National Soil Survey Labora- 
tory characterization data. Supplemental sup- 
port documentation, where available, included 
soil survey area manuscripts, field sheets, iden- 
tification legends, geology reports, climate data, 
and vegetation data. The techniques and proce- 
dures used are the same as those outlined by the 
United States Department of Agriculture, Soil 
Conservation Service, in the National Soils 
Handbook, Soil Survey Manual, Soil Taxonomy 
and National Range Handbook. 

Soil Profile Documentation 

Each soil profile was examined in the field 
with an emphasis on the physical and chemical 
soil properties that would influence the present 
and potential plant community. The description 
of each pedon was compared with the soil pro- 
file. Particular attention was given to the size, 
abundance, distribution, and depth of plant 
roots. Any differences between the documenta- 
tion and the observable soil properties that 
would influence the kind, amount, and propor- 
tion of plans were noted. 



Vegetation Documentation 



Production and composition air-dry-weight 
(ADW) was estimated using ocular estimates 
by SCS area range conservationists familiar 
with the local plant communities, except in 
Wyoming. In Wyoming, 1988 production and 
composition data are presented from double 
sampling measurements made by a BLM 
range conservationist. Cover estimates were 
obtained by the National Soil-Range Team in 
cooperation with SCS range conservationists, 
using the Daubenmire technique (USDI, 
1985). Due to time constraints, only 20, 20 cm 
x 50 cm plots were estimated at each site, 
rather than the usual 40-50 plots. Only ma- 
jor plant species are listed, for the sake of 
brevity. 

Cover estimates and composition summa- 
ries are attached to the soil data pertinent to 
each individual site and are located in the 
field guide used during the tour. Table 1 in 
this paper is a composite of the above grass, 
forb, and shrub vegetation data by percent 
for each site. Figure 2 graphically displays 
the data obtained from Table 1, to allow the 
user to compare, at each site, relevant soil- 
vegetation-climate parameters. 

Except for the two sites near Kemmerer, 
WY, we attempted to quantify, as a cover 
component, the surface area affected by 
cryptogams. Cryptogamic surfaces were recog- 
nized by the roughened surface morphology or 
Type II surfaces (Eckert et al., 1986) associated 
with soil aggregates held together by mucilagi- 
nous secretions and/or the visual presence of li- 
chen, moss, or clubmoss on the soil surface. 
During the vegetation data collection, no at- 
tempt was made to differentiate cryptogam spe- 
cies. Cover estimates are general because there 
was a great variation in thickness and degree of 
cryptogam development. The intent here is 
merely to emphasize the species presence and 
the fact that we are grossly neglecting a major 
plant component in our ecological analyses of 
soil, climate, and vegetation relationships. 

At the Scobey site near Glasgow, MT, club- 
moss (Selaginella densa) is a primary ground 
cover component. Clubmoss is considered very 
detrimental to vascular plant production (Dolan 
and Taylor, 1972) and establishment. The re- 
mainder of the sites all have varying degrees of 
algal and lichen development in and upon the 
soil surface (cryptogamic crusts). These crusts 
appear to have a favorable effect on seedling 
establishment or at least no detrimental effects. 



190 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



A. GREAT BASIN 



High 



c 
o 



low 



A.!. 




Dry 



Soli Moisture 



Results and 
Discussion 

Soil- Vegetation and 
Relationships 

Montana - Scobev Soil 

The vegetation compo- 
nents associated with the 
Scobey Series (NSSL ID#: 
88PO849) produce an esti- 
mated 896 kg/ha (800 Ibs./ 
ac) with a cover dominated 
by grass, few shrubs, and 
intermediate amounts of 
forbs. Irrespective of the 
pedon description, the ma- 
jority (common or many) of 
plant roots are concen- 
trated within the upper 30 
cm of soil and are generally 
exped below 13 cm, where 
the argillic horizon is assumed to start. Below 
30 cm, the abundance of roots decreases dra- 
matically when the whole horizon is taken into 
account, except along ped surfaces. 

This soil lacks any apparent lithologic discon- 
tinuity in the upper soil profile that could re- 
strict hydraulic conductivity. The climate pat- 
tern (Figure l.a.) for Glasgow, MT, generally 
favors grass vegetation. A high amount of or- 
ganic carbon is common to 30 cm. Evidence 
suggests that the depth of frequent wetting is 
approximately 30-61 cm, as the concentrations 
of finely divided lime, the sodium absorption 
ratio, and the pH greater than 8.5 collectively 
increase within this zone. Occasional deep wet- 
ting is evident from the increased electrical con- 
ductivity, reflecting soluble salt movement to 
depths greater than 79 cm. The weak, very fine 
platy structure at the soil surface at present is 
assumed to have a minimal effect on soil mois- 
ture infiltration. Bulk density and soil structure 
also plays an important role in affecting roots at 
depths below 13 cm. 

Montana - Cambeth Soil 

The vegetation cover components on the Cam- 
beth Series (NSSL ID#: 88P0851) consist of a 
crested wheatgrass (Agropyron cristatum) seed- 
ing in a area that once was native range. The 
estimated production of crested wheatgrass is 
952 kg/ha (850 lbs./ac). This site is located in an 
area on the upper third of the sideslope, with 



8. COLORADO PLATEAU 
A.tv. 



Mtsic 




A.tw. 



'Att 



Soil Moisture 



Mesic 



Figure 4. Ordinations' of major sagebrush taxa against gradients of elevation 
and effective moisture. The symbnols represent the first letters of the ge- 
nus, species and subspecies names, therefore, A. = Artemisia; A,a.= A. Ar- 
buscula; A.b.= A. bigelovii; A-L= A. longiloba; A.n.= A. nova; A.s.= A. spines- 
cens; A.t.t.= A. tridentata tridentata; A.t.v.= A. tridentata vaseyana; and 
A.t.w.= A. tridentata wyomingensis. 



landscape position is slightly more conducive to 
runoff. Infiltration of precipitation may be re- 
tarded further by the presence of a moderately 
strong or strongly developed, 0.5-1.0 cm thick 
vesicular crust. 

Irrespective of the pedon description, plant 
roots range from many in the upper 14 cm to 
common to 24 cm and are reduced dramatically 
to few below 24 cm. This soil lacks any apparent 
lithologic discontinuity in the upper soil profile 
that could restrict hydraulic conductivity. The 
representative climate pattern (Figure l.a.) for 
Miles City, MT, generally favors grass vegeta- 
tion. 

High amounts of organic carbon is common to 
24 cm. Evidence suggests that the depth of fre- 
quent wetting is approximately 14-24 cm. The 
concentration of finely divided lime and the pH 
greater than 8.4 start to increase within this 
zone. Occasional deep wetting is evident from 
the high levels of finely divided lime, the sodium 
absorption ratio, and electrical conductivity re- 
flecting soluble salt movement at depths greater 
than 24 cm. 

Montana - Lonna Soil 

The vegetation cover components on the 
Lonna Series (NSSL ID#: 88P0852) consist of a 
crested wheatgrass (Agropyron cristatum) seed- 
ing in a area that once was native range. The 
estimated production of crested wheatgrass is 
15 kg/ha (1350 IbsVac). This site is located in an 



microrelief that is slightly convex. This type of area on the lower part of the sideslope com- 



STAIDL AND LEONARD: REPORT ON SITE SPECIFIC SOIL-CUMATE-VEGETATION RELATIONSHIPS 



191 



Table 1. Composite of \ 
( 

General Site Locations 


regetati< 
?oUecte< 

Total 


on Cover, Co 
3 by Soil at E 

Cr asses 


rn.positi.on an 
ach Site. 

Forbs 


id production 
Shrubs 


a Data 

Eat. 
_Prod,_ 

kg/ha 


Glasgow, MX 
(Hariaa) (721m) 
Site I Wheat Barley 
(Scob*y)(757m) 
Site 2 


% 
Cover 


% % 

Cover Comp 


% % 

Cover Conp 


% % 
Cover Covp 


29 


20 86 


9 8 


T 2 


896 


Milea City, MT 
(Cambeth) (Convex) (896m) 
Site 1 Created Wheat: 
(Loima) (Concave) (896n) 
Site 2 Created Wheat 


35 


32 


3 





952 
1512 


48 


46 


2 





Warren, HI 
(Storaitt) (1593n) 
Site 1 


35 


22 83 


5 A 


8 13 


392 


Dillan, MT 
(Crago) (1663m) 
Site 1 


38 


29 87 


4 10 


5 5 


336 


Laador, ID 
(Arbus) (2116m) 
Site 1 
(Not deaignated) (2116m) 
Site 2 


40 


18 45 


8 5 


14 50 


336 


47 


20 40 


5 8 


22 53 


672 


Big Piney, WY 
(Chedaey tax) (2554m) 
Site 1 
(Aahualog var) (2359m) 
Site 2 


54 


17 15 


17 14 


20 71 


739 


61 


11 16 


14 11 


36 73 


649 


Fontenelle Dam, WY 
(Treaano var) (2049m) 
Site 1 
(Traaano var) (2073m) 
Site 2 


35 


14 47 


5 10 


16 43 


554 


29 


5 19 


10 23 


14 58 


448 


Kammarer, WY 
(Luhon var) (2326m) 
Site 1 
(lyera var) (2198m) 
Site 2 


58 


18 37 


17 13 


23 50 


784 


77 


14 15 


31 6 


32 79 


739 



monly called the footslope and extends to the 
toeslope. The microrelief is slightly concave. 
This type of landscape position is slightly more 
conducive to receiving run-on water and addi- 
tional sediment from higher surrounding areas. 
The possibility of a snowpack on this soil also 
could be a consideration. Reduction of infiltra- 
tion characteristics may be only of a minor na- 
ture due to the weakly developed, 0.5-1.0 cm 
thick vesicular crust at the surface. 

The plant roots identified in the soil profile 
range from many in the upper 31 cm to common 
to 100 cm, and few below 100 cm. This soil lacks 
any apparent lithologic discontinuity in the up- 
per soil profile that could restrict hydrologic con- 
ductivity. The representative climate pattern 
(Figure l.a.) for Miles City, MT, generally favors 
grass vegetation. 

High amounts of organic carbon is common to 
31 cm. Evidence suggests that the depth of fre- 



quent wetting is approxi- 
mately 31-45 cm. The con- 
centration of finely divided 
lime, the sodium absorption 
ratio, and electrical conduc- 
tivity increase within this 
zone. Occasional deep wet- 
ting is evident from the high 
levels of the sodium absorp- 
tion ratio, electrical conduc- 
tivity, and pH at depths 
greater than 45 cm. 

The concave position on 
the landscape allows for in- 
creased opportunities to re- 
ceive run-on water and in- 
creased infiltration due to a 
lack of a strongly developed 
vesicular crust. This is re- 
flected by the differences in 
several soil characteristics 
such as the depth of wetting 
and vegetation production 
between this soil and the as- 
sociated Cambeth soil. 

Montana-Stormitt Soil 

The vegetation associated 
with the Stormitt Series 
(SSL ID#: 88P0853) pro- 
duces an estimated 392 kg/ 
ha (350 Ibs./ac) with a cover 
dominated by grass, few 
forbs, and slightly higher 
amounts of shrubs. The mi- 
crorelief for the area is 
smooth to slightly concave. Infiltration may be 
somewhat reduced by the presence of a less than 
0.5 cm thick, weakly developed vesicular crust. 
Review of the soil profile noted that the ma- 
jority of plant roots are concentrated in the up- 
per 33 cm and the roots are common to 48 cm 
and few below 48 cm. This soil has a lithologic 
discontinuity at approximately 48 cm that will 
affect hydraulic conductivity by restricting wa- 
ter movement. 

The representative climate patterns (Figure 
l.a.) for Bridger, MT, although seasonally 
warmer with a July-August dry-down, favor 
grass vegetation but at somewhat reduced pro- 
duction levels. Moderate amounts of organic 
carbon are common to 33 cm. Evidence suggests 
that the depth of frequent wetting is approxi- 
mately 33-48 cm. The concentration of finely 
divided lime and the pH increase within this 
zone. Occasional deep wetting may be inferred 



192 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



from the slight increase of electrical conductiv- 
ity, reflecting soluble salt movement to depths 
greater than 80 cm. The location and concentra- 
tion of plant roots reflect the above soil moisture 
relationships. 

Montana - Crago Soil 

The vegetation associated with the Crago 
Series (SSL ID#: 88P0854) produces an esti- 
mated 336 kg/ha (300 Ibs./ac) with a cover domi- 
nated by grass and including a few forbs and 
shrubs. The low production may be due in part 
to plant species composition, resulting in the fair 
condition class. 

Infiltration characteristics of this soil may be 
reduced by the presence of a 1-2 cm thick, mod- 
erately strong vesicular crust. Review of the soil 
profile noted that the majority of plant roots are 
concentrated in the upper 14 cm, and roots are 
common to 27 cm, common exped and along 
weak fracture plains to 83 cm, and none below 
83cm. 

This soil has a lithologic discontinuity at ap- 
proximately 27 cm that will have some affect on 
hydrologic conductivity by restricting water 
movement. The representative climate pattern 
(Figure l.b.) for Dillon, MT, is seasonally cooler 
and somewhat dryer than Glasglow and Miles 
City. The pattern continues to favor grass vege- 
tation but at reduced production levels. 

High amounts of organic carbon are common 
to 27 cm. Evidence suggests that the depth of 
frequent wetting is approximately 27-55 cm, as 
the concentration of finely divided lime and the 
pH increased substantially within this zone. 
Occasional deep wetting is evident from the in- 
creased electrical conductivity, reflecting soluble 
salt movement and the increase in the sodium 
absorption ratio at depths greater than 55 cm. 
The location and concentration of plant roots 
reflect the above soil moisture relationships 
quite well. 

Idaho-Arbus Soil 

The vegetation associated with the Arbus 
Series (NSSL ID#: 88P0848) produces an esti- 
mated 336 kg/ha (300 lbs./ac), with a cover 
dominated by approximately equal amounts of 
grass and shrubs with few forbs. The weak or 
very weakly developed, 0.5-1.5 cm thick, vesicu- 
lar crust at the soil surface is assumed to have a 
minimal effect on infiltration characteristics. 

Review of the soil profile noted that the ma- 
jority of plant roots are concentrated in the up- 
per 23 cm, and roots are common on rock frag- 
ment surfaces and along weak fracture plains to 



76 cm and generally no below 76 cm. This soil 
has a lithologic discontinuity at approximately 
23 cm that will have some affect on hydrologic 
conductivity by restricting water movement. 

The representative climate patterns (Figure 
l.b.) for Challis, Howe, and Salmon, ID are sea- 
sonally cooler and dryer than Dillon, Glasgow, 
and Miles City, MT. The seasonally dryer pat- 
tern, with precipitation peaking in June then 
decreasing throughout the remaining months, 
tends to shift the vegetation communities to- 
wards shrub domination. These low sagebrush 
(Artemisia arbusculd) communities are highly 
competitive for any available soil moisture 
throughout most of the growing season. Low 
sagebrush is highly adapted to conditions rang- 
ing from saturation as a result of a perched wa- 
ter early in the growing season, to extreme dry- 
ness later in the summer, or season-long 
droughtiness associated with wind swept ridges, 
etc. 

High amounts of organic carbon are common 
to 33 cm, with lesser amounts to 43 cm. Evi- 
dence of frequent wetting is approximately 23- 
33 cm. The concentration of finely divided lime 
and the pH increase within this zone. Occa- 
sional deep wetting is noted from increases in 
electrical conductivity and sodium absorption 
ratio at depths greater than 43 cm. The location 
and concentration of plant roots reflect the 
above relationships quite well. 

Idaho-unnamed Soil 

The vegetation associated with this unnamed 
soil (NSSL ID #: 88P0057) produces an esti- 
mated 672 kg/ha (600 lbs./ac), with a cover 
dominated by approximately equal amounts of 
grass, and shrubs, with few forbs. The soil 
shows ample evidence of numerous burrows 
from past animal activity, soil material scat- 
tered about the soil surface, and 5-10 cm in di- 
ameter krotovinas at depths ranging from about 
30-70 cm. The high nitrate concentration below 
42 cm also may be the end result of animal ac- 
tivity. Infiltration characteristics are assumed 
to be minimally affected because of a very 
weakly developed vesicular crust that ranges up 
to 1 cm. 

Review of the soil profile noted that the ma- 
jority of plant roots are concentrated in the up- 
per 61 cm., are common to 100 cm, and few be- 
low 100 cm. This soil has a lithologic discontinu- 
ity at approximately 100 cm that may have 
some affect on hydrologic conductivity, assum- 
ing soil moisture penetrates to this depth. 



STAIDL AND LEONARD: REPORT ON SITE SPECIFIC SOIL-CLIMATE-VEGETATION RELATIONSHIPS 



193 



The representative climate patterns (Figure 
l.b.) for Challis, Howe, and Salmon, ID, are sea- 
sonally cooler and dryer than Dillon, Glasgow, 
and Miles City, MT. The seasonally dryer pat- 
tern, with precipitation peaking in June and 
then decreasing throughout the remaining 
months, tends to shift the vegetation communi- 
ties towards shrub domination. Big sagebrush 
(Artemisia tridentata) communities are highly 
competitive in using both shallow and deep soil 
moisture throughout most of the growing sea- 
son. 

High amounts of organic carbon are common 
to 61 cm. Evidence of frequent or occasional 
wetting may by masked due to animal activity. 
However, wetting max be inferred from the elec- 
trical conductivity and sodium absorption levels 
increasing at 42 cm and reaching the highest 
levels at 61-100 cm. Finely divided lime is fairly 
evenly distributed to 100 cm and increases sub- 
stantially along with pH below depths of 100 
cm. This also could reflect previous early Holo- 
cene or late Wisconsin Pluvial climates. 

Wyoming - Chedsey Soil 

The vegetation associated with the Chedsey 
Taxadjunct Series (NSSL ID #: 88P0856) pro- 
duces an estimated 739 kg/ha (660 Ibs./ac), with 
a cover dominated by near equal amounts of 
grass, forbs, and shrubs, with shrubs comprised 
mainly of low sagebrush (Artemisia arbusculd) 
being slightly higher in proportion. 

The thin vesicular crust on the soil surface is 
assumed to have some impact on infiltration 
characteristics. Review of the soil profile noted 
that the majority of plant roots are concentrated 
in the upper 23 cm, common to 45 cm, and few 
below. This soil lacks any apparent lithologic 
discontinuity in the upper soil profile that could 
restrict hydraulic conductivity. 

The climatic pattern (Figure I.e.) for Merna, 
Wyoming, generally represents the site. It is 
seasonally cooler and dryer than the Montana 
sites, but is slightly more moist than the Idaho 
sites during the growing season. The expected 
moisture distribution pattern is conducive to 
leaching and deeper soil water availability 
through the short growing season. The higher 
production and near equal amounts of grass, 
forbs, and shrubs reflect this climate pattern. 

High amounts of organic carbon are common 
to 23 cm, with slightly lesser amounts to 45 cm. 
Sufficient seasonal moisture is available to 
leach carbonates and reduce the base saturation 
above 10 cm. Evidence suggests that the depth 
of frequent wetting is approximately 45-74 cm, 



as the concentration of pedogenic finely divided 
lime increases substantially. Occasional deep 
wetting is evident from the increased electrical 
conductivity and the pH at depths greater than 
74 cm. Plant root penetration also ia reflected in 
the changes to bulk density at 23 to 45 cm and 
below 45 cm. The high bulk density also affects 
other soil properties that will affect plants, such 
as total pore space and available water capacity. 

Wyoming - Ashuelog Variant Soil 

The vegetation components associated with 
the Ashuelog Variant (NSSL ID#: 88P0855) 
produce an estimated 649 kg/ha (580 lbs./ac), 
with a cover dominated by shrubs and including 
smaller amounts of grass and forbs. The pres- 
ence of a vesicular crust was not evaluated at 
this site. Review of the soil profile noted that 
the majority of plant roots were concentrated in 
the upper 33 cm, with few roots from 33 to 66 
cm. Further root and soil moisture penetration 
is restricted at depths below 66 cm by a petrocal- 
cic horizon. 

The climatic pattern (Figure I.e.) for Big 
Piney, Wyoming, generally represents the site. 
It is cooler and dryer than the Merna, Wyoming, 
Idaho, and Montana sites. The seasonally dryer 
pattern with precipitation peaking in April and 
May, with a dry-down during the remainder of 
the growing season, tends to shift the vegetation 
community toward shrub domination. The 
black sagebrush-early sagebrush (Artemisia 
nova - Artemisia longiloba) communities are 
highly competitive for any available soil mois- 
ture throughout mpst of the growing season. 

High amounts of organic carbon are common 
to 33 cm with lesser amounts to 66 cm. The 
depth of frequent and occasional wetting is diffi- 
cult to determine in this profile. The zone from 
33-66 cm may be considered as a degrading pet- 
rocalcic horizon or as a newly forming calcic ho- 
rizon, as this is the major zone of finely divided 
lime accumulation. Either case could restrict 
roots. The presence of few roots within this 
layer also may indicate a lack of frequent deeper 
moisture penetration restricting major plant 
use of the soil to the upper 33 cm. 

Wyoming - Tresano Soil 

The vegetation components associated with 
the Tresano Variant (NSSL ID#: 88P0858) pro- 
duces an estimated 554 kg/ha (495 lbs./ac), with 
a cover dominated by shrubs, and including 
slightly lower amounts of grass and few forbs. 
Infiltration characteristics of this soil may be 
reduced by the presence of a 1-3 cm thick vesicu- 



194 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



lar crust. Review of the soil profile noted that 
the majority of plant roots were concentrated in 
the upper 54 cm with few or no roots occurring 
below. 

The climate pattern (Figure l.d.) for Fon- 
tenelle Dam and LaBarge, Wyoming, generally 
represents the site. It is slightly warmer than 
the Big Piney and Merna, Wyomging, sites and 
is cooler and dryer than the Idaho and Montana 
sites. The seasonally dryer pattern, with pre- 
cipitation peaking in April and May with a dry- 
down during the remainder of the growing sea- 
son, tends to shift the vegetation communities 
toward shrub dominance. The Wyoming big 
sagebrush (Artemisia tridentata ssp. Wyomin- 
gensis) communities are highly competitive in 
using both shallow and deep soil moisture 
throughout most of the growing season. 

Moderate amounts of organic carbon are com- 
mon to 25 cm, with lesser amounts to 54 cm. 
Evidence suggests that the depth of frequent 
wetting is approximately 25-54 cm, as the con- 
centration of pedogenic finely divided lime and 
the pH increase substantially. Occasional deep 
wetting is noted from the increase in electrical 
conductivity reflecting soluble salt movement at 
depths greater than 79 cm. The sodium absorp- 
tion ratio, which starts to increase at 25 cm, also 
reaches its highest levels below a depth of 79 
cm. 

Wyoming - Tresano Variant Soil 

The vegetation components associated with 
the Tresano Variant (NSSL ID#: 88P0861) pro- 
duces an estimated 448 kg/ha (400 Ibs./ac), with 
a cover dominated by shrubs, and including few 
grasses and intermediate amounts of forbs. In- 
filtration characteristics of this soil may be re- 
duced by the presence of a 1-3 cm thick vesicular 
crust. Review of the soil profile noted that the 
majority of plant roots are concentrated in the 
upper 26 cm, with the total amount dropping off 
with depth, as they are generally confined to 
exped surfaces. 

The climate pattern (Figure l.d.) for Fon- 
tenelle Dam and LaBarge, Wyoming, generally 
represents the site. It is slightly warmer than 
the Big Piney and Merna, Wyoming, sites and is 
cooler and dryer than the Idaho and Montana 
sites. The seasonally dryer pattern, with pre- 
cipitation peaking in April and May with a dry- 
down during the remainder of the growing sea- 
son, tends to shift the vegetation communities 
toward shrub dominance. The Wyoming big 
sagebrush (Artemisia tridentata spp. Wyomin- 
gensis) communities are highly competitive in 



using both shallow and deep soil moisture 
throughout most of the growing season. 

Moderate amounts of organic carbon are com- 
mon to 18 cm, with lesser amounts to 26 cm. 
Evidence suggests that the depth of frequent 
wetting is approximately 18-26 cm, as the con- 
centration of pedogenic finely divided lime and 
the pH increase substantially. Occasional deep 
wetting is evident from the increased electrical 
conductivity reflecting soluble salt movement at 
depths greater than 80 cm. The sodium absorp- 
tion ratio which starts to increase at 26 cm, also 
reaches its highest levels below a depth of 80 
cm. The plant rooting characteristics are re- 
flected in the above relationships. 

Wyoming - lyers Variant Soil 

The vegetation components associated with 
the lyers Variant (NSSL ID#: 88P0860) produce 
an estimated 739 kg/ha (660 Ibs./ac), with a 
cover dominated by a near equal amounts of 
shrubs and forbs and lower amounts of grass. 
This site is located in an area on the lower part 
of the sideslope commonly called the footslope, 
extending to the toeslope. The microrelief is 
slightly concave. This type of landscape position 
is more conducive to receiving run-on water in 
addition to the possibility of a snowpack accu- 
mulation. Infiltration characteristics of this soil 
may be reduced by the presence of a thin vesicu- 
lar crust and clay textures. Reduced infiltration 
may be offset by the natural tendency of this 
high clay soil to form deep wide cracks when 
partially dry, allowing the soil profile to remois- 
ten from below. 

Review of the soil profile noted that the ma- 
jority of plant roots are concentrated in the up- 
per 25-30 cm. Smaller numbers of roots that 
generally occur along ped faces and cracks ex- 
tend to deeper depths. 

The climate patterns (Figure I.e.) for 
Kemmerer and Sage, Wyoming, generally repre- 
sent the site. They are seasonally the same as 
the other Wyoming sites and are cooler and 
dryer than the Idaho and Montana sites. The 
peak precipitation months are April and May, 
with a dry-down period during the remainder of 
the growing season. This pattern is more ori- 
ented to shrub dominance, as the early sage- 
brush (Artemisia longilobd) communities are 
highly competitive for available soil moisture 
throughout the growing season. 

High amounts of organic carbon are common 
to 13 cm, with moderate amounts to 90 cm. 
Evidence suggests that the depth of wetting is 
approximately 45 to 90 cm, as the electrical con- 



STAIDL AND LEONARD: REPORT ON SITE SPECIFIC SOIL-CUMATE-VEGETATION RELATIONSHIPS 



195 



ductivity increases, reflecting soluble salt accu- 
mulation. The shrink-swell characteristic of 
this soil is an important feature to consider, as it 
affects kinds of plants and their natural rooting 
habitat. 

Wyoming- Luhon Soil 

The vegetation components associated with 
the Luhon Variant (NSSL ID#: 88P0859) pro- 
duce an estimated 784 kg/ha (700 Ibs./ac), with 
a cover dominated by shrubs and including 
slightly lower proportions of grass and forbs. 
Review of the soil profile noted that the majority 
of plant roots are concentrated in the upper 58 
cm, with few roots along ped surfaces extending 
to deeper depths. 

The climate data (Figure I.e.) for Kemmerer 
and Sage, Wyoming, are the nearest data for the 
site. However, because of the elevational differ- 
ence, the climatic pattern from the Merna site 
(Figure I.e.) may be more representative. They 
are seasonally the same as the other Wyoming 
sites and are cooler and dryer than the Idaho 
and Montana sites. The peak precipitation 
months are April and May, with a dry- down 
period during the remainder of the growing sea- 
son. 

This pattern is more oriented to a shrub domi- 
nance, as the early sagebrush (Artemisia longi- 
loba) communities are highly competitive for 
available soil moisture throughout the growing 
season. 

High amounts of organic carbon are common 
to 18 cm, with moderately high amounts to 107 
cm. Evidence suggests that the depth of fre- 
quent wetting is approximately 18 to 58 cm, as 
the pH and the concentration of finely divided 
lime increase substantially. Occasional deep 
wetting is evident from the decreased amount of 
roots occurring on ped surfaces at depths 
greater than 58 cm. 

Summary and Conclusions 

The natural plant community best adapted to 
a particular rangeland soil is dependent on the 
climate-soil interactions and the physiological 
adaptation of individual species to extracting 
moisture and nutrients at available times and 
locations within the soil profile. Climate-soil 
interactions of course, are, affected by geo- 
graphic location and topographic position. The 
plants (nonvascular as well as vascular) also 
affect the climate-soil interactions through 
moisture interception, shading, evapotranspira- 
tion characteristics, and many other factors. 



There are also biological relationships (symbio- 
sis, allelopathy, etc.) and temporal relationships 
(surface disturbance, climate fluctuations, use 
and management, etc.) that affect the commu- 
nity characteristics. Major relationships affect- 
ing potential plant communities are reflected by 
the physical and chemical properties of each 
soil. Variations in present productivity, compo- 
sition, etc. often are reflected by temporal fac- 
tors and biological interactions. 

Individual and combined results of these com- 
plex interactions could be observed throughout 
the tour. General relations of ambient tempera- 
ture and precipitation distribution with soil 
moisture availability were evident in the transi- 
tion from grass to shrub dominance. Physiologi- 
cal plant adaptations were evident from the 
various sagebrush species and subspecies re- 
sponse observed in Idaho and Wyoming. Vari- 
ations in productivity on any soil can be attrib- 
uted in part to biological interactions (clubmoss 
effects on the Scobey soil) or temporal relation- 
ships (drought and range condition on the Crago 
soil). It is, therefore, unrealistic for Soil Taxon- 
omy to address all combinations of these com- 
plex interactions. 

The perception is that Soil Taxonomy should 
address natural plant communities to a suffi- 
cient extent to make general interpretations of 
soil-climate-plant relationships and guide the 
user toward more specific information needs. 
Soil moisture and temperature regimes are a 
concern in natural plant communities and par- 
ticularly for Aridisols. These concerns are per- 
haps best expressed as questions. 

Adjacent soils mapped in complex or associa- 
tion may have other classifications (i.e., molli- 
sols in association with aridisols within the 
same map unit). Is soil moisture regime a func- 
tion of climate, soil properties, or the combined 
effect on plant productivity? 

If soil moisture regime (and temperature re- 
gime) relates to productivity, how much and 
under what conditions? Blaidsell (1958) reports 
a continuum of beginning growth temperatures 
for range plants starting near 0C, well below 
the 5C standard in Soil Taxonomy. This and 
other physiological adaptions (rooting charac- 
teristics) suggest a major portion of plant 
growth may be associated with shallower or 
sometimes deeper soil moisture availability 
rather than with what typically is considered as 
the soil moisture control section. Is additional 
soil moisture control section information and 
criteria for natural plant communities needed? 



196 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Many range plants, shrubs in particular, 
seem able to function physiologically at soil 
moisture tensions of 6.0 to 7.0 MPa (Caldwell, 
1985). Is the soil really "dry* to these plants, or 
is it merely dry according to our agronomic per- 
ceptions of a 1.5 MPa tension wilting point? 

Acknowledgements 

Special thanks to the Idaho, Montana, and 
Wyoming Soil Scientists and Range Conserva- 
tionists of the Bureau of Land Management and 
Soil Conservation Service who provided assis- 
tance and information for this report. 

References 

Blaisdell, J.P . 1958. Seasonal development of native 
plants on the upper Snake River plains and their rela- 
tionship to certain climatic factors. USDA Tech. Bull. 
1190:68. 

Caldwell, M. 1985. Cold desert. In: Physiological ecology 
of North American plant communites. New York, Lon- 
don, Chapman and Hall. 

Dolan, J.J. and J.E. Taylor. 1972. Residual effects of 
range on dense clubmoss and associated vegetation. 
Range Management. 25:32-37. 

Harper, K.T. and J.R. Marble. 1988. A role for nonvascu- 
lar plants in management of arid and semiarid 
rangelands. In: Tueller, P.T. (ed.) Vegetation science 
applications for rangeland analysis and management. 
Kluwer Academic Publishers. Dordrecht. Pp. 135-169. 

Hironaka, M., M.A. Fosberg, and A.H. Win ward. 1983. 
Sagebrush-grass habitat types of southern Idaho. For- 
est, Wildlife and Range Experiment Sta. Bull. 35. Uni- 
versity of Idaho. Moscow, ID. 44 p. 



Kormondy, E.J. 1969. Concepts of ecology. Jn:McElroy, 
W.D. and Swanson, C.P. (Ed.) Concepts of modern biol- 
ogy series. Prentice-Hall, Inc. Publishers. 209 p. 
Kuchler,A.W. 1964. Potential natural vegetation of the 
conterminous United States. Special Publication No. 
36. American Geographical Society. New York. 

Leonard, S.G.,R.L. Miles and P.T. Tueller. 1988. Vegeta- 
tion-soil relationships on arid and semiarid 
rangelands. In: Tueller, P.T. (ed.) Vegetation science 
applications for rangeland analysis and management. 
Kluwer Academic Publishers. Dordrecht. Pp. 225-252. 

Passey, H.B., V.K. Hugie, E.W. Williams and D.E. Ball. 
1982. Relationships between soil, plant community and 
climate on rangelands of the intermountain west. 
Tech. Bull. No. 1669. USDA, SCS. Washington, B.C. 

Sasich, J. and G.A. Nielsen. 1984. Indicators of soil cli- 
mate used by Montana soil classifiers: a survey. Mon- 
tana Ag. Exp. St. Special Report 4. Montana State 
University, Bozeman. 45 p. 

Stoddart, L.A., A.D. Smith and T.W. Box. 1975. Range 
Management. McGraw-Hill, Inc. New York. 532 p. 

USDA. 1975. Soil Survey Manual. U.S. Dept. Agric. 
Handb. 18, 503 pp., Illus. (Supplements Replacing pp. 
173-188, Issued May 1962) 

USDA. 1975. Soil Taxonomy: a basic system of soil classi- 
fication for making and interpretating soil surveys. 
Soil Conservation Service, U.S. Dept. of Agric. Hand- 
book 436, 754 pp., Illus. 

USDI. 1985. Rangeland monitoring trend studies. Tech- 
nical Ref. 4400-4. USDI, Bureau of Land Management. 
Denver Service Center. Denver, CO. 130 p. 

West, N.E. 1979. Basic syn ecological relationships of 
sagebrush-dominated lands in the Great Basin and the 
Colorado Plateau. In: Anon. The sagebrush ecosystem: 
A symposium. Utah State University College of Natu- 
ral Resources. Logan, Utah. 

Zamora, B. and P.T. Tueller. 1973. Artemisia arbuscula, 
A. longiloba, and A. nova habitat types in Northern 
Nevada. Great Basin Naturalist. 46:3349-352. 



Soil Management Research on the Clay Soils 
of Southwestern Ontario - A Review 

J. A. Stone* 
Abstract 

Intensive row crop agriculture on the clay and clay loam soils of south- 
western Ontario has resulted in surface soil structural deterioration and ac- 
companying decreases in productivity and increases in nonpoint source 
surface water and groundwater contamination. Maintaining and/or im- 
proving soil quality depends upon understanding the extent and causes of 
structural deterioration and upon introducing a structure-improving com- 
ponent into row cropping systems. This paper presents a review of com- 
pleted and ongoing research conducted on the clay soils of southwestern 
Ontario, addressing: 1) methods of measuring changes in structure; 2) the 
effect of long-term row crop production, tillage, and traffic on structural de- 
terioration, and 3) methods for improving structure. 



Introduction 

Soil degradation, the depletion of productive 
capability, is a major concern across Canada 
(Anonymous, 1984). This paper focuses on the 
clay and clay loam soils in southwestern On- 
tario, generally defined as south and west of 
metropolitan Toronto (south of 43 30' North 
latitude, Fig. 1.), which are major areas of corn 
(Zea mays L.) and soybean (Glycine max) pro- 
duction. On these soils, degradation is evi- 
denced primarily by a gradual deterioration in 
the ability of the surface soil to retain an ar- 
rangement of solid and void space favorable for 
crop production. This is referred to as surface 
soil structural deterioration and is contributing 
to increased erosion, runoff, and compaction. 

Surface soil structural deterioration is be- 
lieved to be a result of moldboard plow based, 
intensive row crop production (Ketcheson, 1980; 
McKeague et al., 1987). The soil structure of the 
surface 30 cm appears to have deteriorated to an 
equilibrium level and is considered compact 
(McKeague et al., 1987). A contributing factor is 
the misuse of modern farm machinery which 
enables a farmer to till excessively and to till 
when the soil is too wet. In addition, the period 
when the soil is dry enough to till satisfactorily 
is reduced if the soil structure is poor, and a 
cycle is likely to develop in which the soil is in- 
creasingly damaged. 

Soil structural deterioration not only lowers 
productivity but also contributes to contamina- 
tion of surface water and groundwater, as well 
(Miller et al., 1988; Stone and Logan, 1989). 
Surface runoff of sediment-sorbed nutrients in- 



* Agriculture Canada Research Branch, Research Sta- 
tion, Harrow, Ontario NOR 1GO, Canada. 



creases algal growth in lakes and streams. The 
application of increased amounts of agricultural 
chemicals, particularly nitrogen and pesticides, 
in response to losses in productivity, is resulting 
in the contamination of groundwater. The 
International Joint Commission addressed 
these concerns in the Revised Great Lakes Wa- 
ter Quality Agreement (Anonymous, 1987) and, 
as a result, major Federal and Provincial pro- 
grams have been introduced which address soil 
degradation problems in southwestern Ontario. 
Maintaining and/or improving soil and water 
quality depends upon understanding the extent 
and causes of structural deterioration and upon 
introducing a soil structure-improving compo- 
nent into row cropping systems. The objective of 
this paper is to present a review of completed 
and ongoing research conducted on the clay soils 
of southwestern Ontario, addressing 1) methods 
of measuring changes in soil structure, 2) the 
effect of long-term row crop production, tillage, 
and traffic on soil structural deterioration, and 
3) methods for improving soil structure. 

Soils and Climate 

The majority of fine textured soils in south- 
western Ontario were formed on till or lacus- 
trine sediments. The major clay and clay loam 
soil series is Brooks ton (Evans and Cameron, 
1983). It extends over 8000 km 2 , primarily in 
extreme southwestern Ontario, and is one of the 
most productive soils in Ontario when tile 
drained. 

The Brookston series soils generally are re- 
garded as being poorly drained, fine textured, 
relatively stone free, and located on level to gen- 
tly undulating topography. Family particle size 
is primarily fine clayey but ranges from 21 to 



197 



198 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



4230' 

Lake St. Clair 



Fig. 1. Major areas of clay and clay loam soils in southwestern Ontario. The smooth curves delineate the heat units 
available for corn production. 




65% (Anonymous, 1978). Texture is generally 
uniform to a depth of 1 to 1.5 m. Organic carbon 
in the surface horizon (Ap) averages 3.4%, ex- 
cept for extreme southwestern Ontario, where it 
averages 2.8%. Clay mineralogy consists pre- 
dominately of illite, with subordinate amounts 
of chlorite, vermiculite, and hydroxy-interlayerd 
vermiculite (Evans and Cameron, 1983). Other 
major series of lacustrine origin, present in 
lesser amounts, are the very fine textured 
Haldimand and Lincoln soil series. 

The climate of southwestern Ontario is influ- 
enced by the Great Lakes. The growing season 
ranges from approximately 135 to 175 frost free 
days (Wicklund and Richards, 1961). The heat 
units available for corn production range from 
approximately 2300 to 3600. Precipitation 
ranges from 800 to 1000 mm per year and is 
distributed fairly evenly throughout the year. 

Measuring Changes in Clay Soil 
Structure 

Quantifying soil structural deterioration re- 
quires a technique which is sensitive to changes 
but is relatively easy to apply. Changes in soil 



structure usually are assessed by performing a 
battery of tests, because no one technique is 
completely satisfactory. 

Techniques that have worked best include soil 
core porosity and bulk density, and aggregate 
stability. Techniques that have not worked 
well, because of insufficient sensitivity and/or 
time required, include dry aggregate size distri- 
bution, aggregate shear strength, turbidity and 
instability index (Stone et al., 1985), and pen- 
etrometer resistance (Heslop et al., 1986). 

This paper discusses core sampling, for bulk 
density and porosity determinations, and aggre- 
gate stability. 

Bulk Density and Porosity 

Core sampling with a typical hand held, ham- 
mer driven, double cylinder core sampler (Blake 
and Hartge, 1986) for the determination of dry 
bulk density, total porosity, and air-filled poros- 
ity has been used for many years. However, 
year-to-year fluctuations in bulk density and 
porosity, determined from soil cores, have been 
observed which are independent of soil and crop- 
ping treatments (Stone et al., 1985; Bolton et 
al., 1982). 



STONE: SOIL MANAGEMENT RESEARCH ON THE CLAY SOILS OP SOUTHWESTERN ONTARIO - A REVIEW 



199 



Adjustment of long-term core data for volume 
changes resulting from variations in soil water 
content at the time of sampling were found to 
reduce, but did not eliminate, the year-to-year 
fluctuation (Stone and Wires, 1989). Stone 
(1989) compared hammer driven and hydraulic 
core sampling techniques and found that the 
core sampling technique significantly affected 
the overall mean values of bulk density, total 
porosity, and air-filled porosity. However, the 
sampling technique generally did not affect the 
ability of the core parameters to detect differ- 
ences in soil structure at field soil water con- 
tents ranging from 12.6 - 23.8%. 

These results (Stone, 1989; Stone and Wires, 
1989) indicate that seasonal fluctuations in bulk 
density and porosity of Brookston clay loam oc- 
cur independently of variations in water content 
at the time of sampling or in the sampling tech- 
nique. Stone et al. (1985) and Bolton et al. 
(1981) found that seasonal fluctuations in bulk 
density and porosity related best to early spring 
rainfall. Bolton et al. (1981) concluded that 
heavy May-June rainfall contributed to a "slak- 
ing" effect which made the soil more susceptible 
to wheel traffic induced compaction. Horn 
(1988) explains that aggregate bulk density in- 
creases during cycles of wetting and drying, but 
varies with the extent of drying. Therefore, fluc- 
tuations in the water content of moist soil may 
promote rearrangement of particles and result 
in increases in bulk density. 

Aggregate Stability 

The most successful method for detecting 
changes in soil structure has been the determi- 
nation of wet aggregate stability (WAS) on 0.5 
to 2.0 mm air dry aggregates wetted by immer- 
sion (Kemper and Rosenau, 1986). Changes in 
WAS due to cropping and tillage have been 
shown (J. A. Stone, 1989, unpublished data; 
Stone and Heslop, 1987). However, in ongoing 
studies, WAS has been found to vary within the 
growing season, presumably due to soil mois- 
ture antecedent to sampling and the number of 
seasonal wetting and drying cycles (Stone, 
1988). Research currently is underway to inves- 
tigate the effect of water content on wet aggre- 
gate stability and dispersible clay (J. Caron, 
1989, personal communication) and the rela- 
tionship between aggregate stability and micro- 
bial biomass (J. A. Stone, 1989, personal com- 
munication). 



Soil Structure and Row Crop 
Production 

Long-term Corn Production 

Long-term plots established on Brookston 
clay loam have provided information about the 
extent and effects of soil structural deterioration 
resulting from row crop production systems. 
McKeague et al. (1987) reported changes in soil 
structure resulting from long-term continuous 
and rotation corn in comparison with a never 
cultivated soil. The long-term com plots (con- 
tinuous and rotation) compared to the never cul- 
tivated plots had a massive as opposed to strong 
ped structure, lower macroporosity, isolated as 
opposed to interconnected pores, and greater 
bulk densities. They reported that the surface 
horizons of the rotation corn plots differed from 
those of the continuous corn plots in having 
more abundant biopores. However, below a 
depth of 30 cm, comparable horizons from all 
plots were similar in ped structure, bulk den- 
sity, and water characteristic curves. There was 
no distinct plow pan or restriction in root distri- 
bution (Stone et al., 1987) resulting from long- 
term corn production. 

Stone et al. (1985) summarized data for the 
period 1957-1982 on bulk density and porosity 
of Brookston clay loam soil from a long-term 
corn fertility experiment. They showed that the 
bulk density tended to slowly increase and po- 
rosity to slowly decrease with time in the 5 to 15 
cm layer for both rotation and continuous corn. 
The rotation plots had lower bulk densities and 
higher porosities than the continuous corn plots. 
However, the fertility level had no effect on bulk 
density or porosity of the continuous corn plots. 

The overall similar rates of deterioration of 
the continuous and rotation corn plots led Stone 
et al. (1985) to conclude that continuous corn by 
itself appeared to play a minor role in the dete- 
rioration of soil structure. By 1957, 4 years af- 
ter the plots were established, dry bulk densities 
of all the plots were considerably higher than 
those of the never cultivated plot reported by 
McKeague et al. (1987). Compaction of the sur- 
face horizon apparently occurred prior to the 
monitoring of bulk density which began in 1957. 

Bryant et al. (1987) analyzed tile flow data 
from long-term plots established on Brookston 
clay loam to determine the effect of cropping 
system on tile drain discharge. They found that 



200 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



continuous corn and bluegrass generally con- 
tributed to a greater tile discharge volume than 
rotation corn, because of a longer flow period. 
For the bluegrass plots, this may be attributed 
to a more favorable soil structure and, therefore, 
more water storage in the surface profile. For 
the continuous corn plots, this may be a result of 
the compact soil structure and, therefore, 
ponding and longer wet periods. 

Tillage and Traffic 

Developing improved management practices 
to stabilize or improve soil structure requires 
information about the contribution of tillage and 
traffic to soil structural deterioration. Studies 
conducted on Brookston clay loam have pro- 
vided conflicting information about the effect of 
tillage on soil structural deterioration. 

Tillage Method 

Stone and Heslop (1987) measured organic 
carbon, wet aggregate stability, and bulk den- 
sity in a study comparing ridge, fall moldboard 
plow, and blade cultivator tillage in a corn - soy- 
bean - corn rotation. After 3 years, mid - season 
organic carbon ranked by tillage treatment was 
blade cultivator > ridge > moldboard plow, but 
the blade was not significantly different from 
ridge tillage. The values of wet aggregate sta- 
bility were ranked as ridge > blade cultivator > 
moldboard plow. 

There were no significant differences in bulk 
density between treatments. Increases in or- 
ganic carbon and wet aggregate stability due to 
treatments were evident below the depth of till- 
age (20 to 30 cm). This suggests that improve- 
ments in soil structure resulting from these re- 
duced tillage systems may be a result of root 
exudates and the byproducts of residue decom- 
position. 

In another 3-year study conducted on Brook- 
ston clay loam (Stone et al., 1989), mid-season 
measurements of organic carbon, aggregate 
shear strength, and wet aggregate stability did 
not provide any consistent or meaningful differ- 
ences between ridge, fall moldboard plow, or 
zero tillage (J. A. Stone, 1989, unpublished 
data). Corn - corn - soybean and soybean - soy- 
bean - corn rotations were compared at adjacent 
sites in this study. 

Average organic carbon levels were lower at 
the site of this study than the site utilized by 
Stone and Heslop (1987). However, soil texture 
and wet aggregate stability were similar at both 
sites. Tile drain spacing was 12.2 m and 7.6 m 
for the sites utilized by Stone et al. (1989) and 



Stone and Heslop (1987), respectively. The 
slopes of the two sites were essentially equal 
(<0.5%), but surface drainage was somewhat 
restricted in the study of Stone et al. 

The contrary results of these two studies sug- 
gest that improvements in soil structure result- 
ing from reduced tillage on clay and clay loam 
soils may be enhanced by better drainage. They 
also suggest that, in the short-term, expected 
reductions in erosion and runoff resulting from 
reduced tillage are more likely induced by the 
physical effects of surface residue rather than 
from improvements in soil structure. 

Wheel Traffic 

Stone (1987) reported the results of a 3-year 
experiment conducted to determine the cumula- 
tive contribution of surface soil compaction to 
structural deterioration. 

Fall vehicle compaction prior to moldboard 
plow tillage did not significantly contribute to 
surface soil structural deterioration. Fall com- 
paction was presumably alleviated by fall plow- 
ing, freezing and thawing, spring secondary till- 
age, and wetting and drying cycles. Spring 
compaction generally contributed to structural 
deterioration; however, there were no cumula- 
tive detrimental effects of fall or spring compac- 
tion on soil structure. 

It was concluded that the contribution of ve- 
hicle compaction, at soil moisture contents suit- 
able for tillage, to the compact surface structure 
of the intensively cropped clay soils in south- 
western Ontario, does not appear to be long- 
term and can be minimized by controlling spring 
wheel traffic. Similarly, Bolton and Aylesworth 
(1959) reported that the effects of excess tillage 
on pore space were small and that there was 
little or no association between porosity values 
and crop yield on this soil. 

Although spring wheel traffic results in sea- 
sonal compaction and tillage is a major factor in 
the rate of drying in the spring, wheel traffic and 
tillage do not appear to be contributing to addi- 
tional structural deterioration on clay soils in 
southwestern Ontario. This information is the 
basis for additional studies on the role of 
rhizosphere effects of grass and legume forages 
in improving soil structure. 

Improving Soil Structure 

The long-term productivity of the clay soils in 
southwestern Ontario appears to depend upon 
introducing a structure-improving component 
into the row cropping system. Grass and legume 
forages generally are accepted as the most effec- 



STONE: SOIL MANAGEMENT RESEARCH ON THE CLAY SOILS OP SOUTHWESTERN ONTARIO - A REVIEW 



201 



tive means of improving soil structure. Recent 
studies have shown that forages may improve 
soil structure within one cropping season (Stone 
and Buttery, 1989; Stone, 1988; Angers and 
Mehuys, 1988). 

Forages 

Stone and Buttery (1989) reported that reed 
canary grass improved soil structure compared 
with other grass and legume forages, due to in- 
creased root mass. The frequency of VA mycor- 
rhiza hyphae was not associated with improve- 
ments in structure, although it varied between 
forages. Drury et al. (1989) measured microbial 
biomass C and N under corn, soybeans, red clo- 
ver, alfalfa, reed canary grass, orchard grass, 
and bare soil at monthly intervals during the 
third growing season on Brookston clay loam 
soil. They found that reed canary grass resulted 
in the greatest biomass C content and corn in 
the least. Varying the extent of the wetting and 
drying cycles in the bare soil did not affect bio- 
mass C or biomass N levels. Reed canary grass 
also showed the greatest improvement in simul- 
taneous measurements of aggregate stability. 

Microbial Factor 

Fysun and Oaks (1989) have established that 
there is a microbial factor associated with leg- 
umes which dramatically increases the growth 
of corn under greenhouse conditions. Ongoing 
research (A. Oaks, 1989, personal communica- 
tion) is being directed toward isolation and iden- 
tification of the beneficial microorganisms. 

Preliminary field studies carried out on 
Brookston clay loam indicate that inoculation of 
soils that have been under corn for 30 years 
leads to significant growth responses. Fysun 
and Oaks (1987) reported that planting corn in 
soil that has been under alfalfa leads to larger 
soil sheaths (soil adhering to roots) on seedling 
roots than when the corn seedlings have been 
grown in soil that has been under corn. The size 
of the sheath may relate to the detrimental ef- 
fect of continuous corn on soil structure, and the 
simple measurement of this parameter may pro- 
vide a useful assay of the degree of soil struc- 
tural deterioration. 

Intercropping 

Probably the most economical method of in- 
corporating the beneficial effects of forages into 
an intensive row crop production system is by 
interseeding the row crop with a forage, thereby 
providing surface cover, improvements in soil 
structure, and biologically fixed nitrogen. Con- 
siderable intercropping research was conducted 



in the 1950s, with limited success, due primarily 
to the lack of satisfactory weed control. Re- 
cently, researchers have successfully intercrop- 
ped corn with a forage using chemical weed con- 
trol (Scott et al., 1987; R. W. Sheard, 1988, per- 
sonal communication). 

Ongoing experiments on Brookston clay loam 
are designed to evaluate corn intercropped with 
legume and grass forages under conventional 
and conservation tillage, relative to improve- 
ments in soil structure while maintaining ac- 
ceptable yields (J. A. Stone, 1989, personal com- 
munication). Related work is being con- 
ducted for a range of soil textures at the Univer- 
sity of Guelph (B. D. Kay, 1989, personal com- 
munication). 

Summary 

Intensive row cropping of the clay and clay 
loam soils of southwestern Ontario has resulted 
in a gradual deterioration of the surface soil 
structure which is contributing to increased 
runoff, erosion, and compaction. Research is 
addressing: 1) methods of measuring changes in 
soil structure; 2) the effect of long-term row crop 
production, tillage, and traffic on soil structural 
deterioration, and 3) methods for improving soil 
structure. 

Changes in structure have been assessed sat- 
isfactorily by determining bulk density, poros- 
ity, and aggregate stability. However, seasonal 
fluctuations exist which have not been fully ex- 
plained. Ongoing research is investigating the 
role of seasonal wetting and drying cycles. 

Evaluation of long-term corn plots has shown 
that the soil structure has deteriorated to an 
equilibrium level under continuous row crop- 
ping. Structural deterioration is restricted to 
the surface 30 cm and has not resulted in a dis- 
tinct plow pan or restriction in root distribution. 
However, it results in a greater volume of dis- 
charge through tiles because of a longer flow 
period. Wheel traffic and tillage do not appear to 
be contributing to additional structural deterio- 
ration. 

Long-term soil productivity appears to de- 
pend upon introducing the structure-improving 
attributes of forages into row cropping systems. 
Research has shown that forages differ in the 
ability to improve soil structure and indicates 
that there is a relationship between the micro- 
bial population associated with different forage 
species and the ability to improve soil structure. 
Ongoing research is being directed toward isola- 



202 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



tion and identification of beneficial microorgan- 
isms. Corn - forage intercropping studies are 
underway to determine the feasibility of incor- 
porating the beneficial effects of forages into an 
intensive row crop production system- 
Literature Cited 

Angers, D. A. and G. R. Mehuys. 1988. Short-term effects 
of cropping on macro- aggregation of a marine clay soil. 
Can. J. Soil Sci. 68:723-732. 

Anonymous. 1978. The Canadian System of Soil Classifi- 
cation. Canada Soil Survey Committee. Pub. 1646, 
pg.115-118. Agriculture Canada Research Branch, 
Central Exp. Farm, Ottawa, Ont. K1A OC6. 

Anonymous. 1984. Soil at risk: Canada's eroding future. 
Published by the Standing Committee on Agriculture, 
Fisheries, and Forestry, Committees and Private Leg- 
islation Branch, The Senate of Canada, Ottawa, On- 
tario K1AOA4. 

Anonymous. 1987. Revised Great Lakes Water Quality 
Agreement of 1978. International Joint Commission, 
Great Lakes Regional Office, 100 Ouelette Ave., Wind- 
sor, Ont. N9A 6T3. 

Blake, G. R. and K. H. Hartge. 1986. Bulk Density. In A. 
Klute (ed.) Methods of soil analysis. Part I. 2nd ed. 
American Society of Agronomy, Madison, WI. 

Bolton, E. F. and J. W. Aylesworth. 1959. Effect of tillage 
traffic on certain physical properties and crop yield on 
Brookston clay soil. Can. J. Soil Sci. 39:98-102. 

Bolton, E. F., V. A. Dirks, and M. M. McDonnell. 1981. 
Effect of fall and spring plowing at three depths on soil 
bulk density, porosity and moisture in Brookston clay. 
Can. Agric. Eng. 23:71-76. 

Bolton, E. F., V. A. Dirks and M. M. McDonnell. 1982. The 
effect of drainage, rotation and fertilizer on corn yield, 
plant height, leaf nutrient composition and physical 
properties of Brookston clay soil in southwestern On- 
tario. Can. J. Soil Sci. 62:297-309. 

Bryant, G. J., R. W. Irwin, and J. A. Stone. 1987. Tile 
drain discharge under different crops. Can. J. Agric. 
Eng. 29:117-122. 

Drury, C. F., J. A. Stone, and W.I. Findlay. 1989. Seasonal 
changes in microbial biomass under grass and legume 
forages on a clay loam soil. (In press, copies available 
from C. F. Drury, Agriculture Canada Research Sta- 
tion, Harrow, Ont. NOR 1GO). 

Evans, L. J. and Cameron, B. H. 1983. The Brookston se- 
ries in southwestern Ontario: characteristics, classifi- 
cation and problems in defining a soil series. Can. J. 
Soil Sci. 63:339-352. 

Fysun, A. and A. Oaks. 1989. Growth promotion of maize 
by legume soils. Plant and Soil (In press, copies avail- 
able from A. Fysun, Dept. of Biology, McMaster Univ., 
Hamilton, Ont. L8S 4K1). 

Fysun, A. and A. Oaks. 1987. Physical factors involved in 
the formation of soil sheaths on corn seedling roots. 
Can. J. Soil Sci. 67:591-600. 

Heslop, L. C., J. A. Stone, and B. Compton. 1986. Portable 
loading frame for field penetrometer measurements. 
ASAE paper no. 86-1039. ASAE, 2950 Niles Rd., St. 
Joseph, MI 49085. 

Horn, R. 1988. Strength of structured soils due to loading 
- a review of macro- and microscale processes; Euro- 
pean aspects./n W. E. Larson, G. R. Blake, R. R. All- 



maras, and S. C. Gupta (ed.) Mechanics and Related 
Processes in Structured Agricultural Soils. Kluwer 
Academic Publishers, Dordrecht. ISBN 0-7923-0342-3, 
Irwin, R. W., G. J. Bryant, M. R. Toombs, and J. A. Stone. 

1987. Evaluation of a drainage coefficient for a poorly 
drained soil. Trans. ASAE 30:1343-1346. 

Kemper, W. D. and R. C. Rosenau. 1986. Aggregate stabil- 
ity and size distribution. In A.- Klute (ed.) Methods of 
soil analysis. 2nd ed. Part I. American Society of 
Agronomy, Madison, WI. 

Ketcheson, J.W. 1980. Long-range effects of intensive cul- 
tivation and monoculture on the quality of southern 
Ontario soils. Can. J. Soil Sci. 60:403-410. 

McKeague, J. A., C. A. Fox, J. A. Stone, and R Protz. 1987. 
Effects of cropping system on structure of Brookston 
clay loam in long-term experimental plots at Woodslee, 
Ontario. Can. J. Soil Sci. 67:571-584. 

Miller, M. H., Groenevelt, P. H. and Stonehouse, D. P. 

1988. Stewardship of soil and water in the food produc- 
tion system. Notes on Agriculture. 22:5-12. Published 
by the Ontario Agricultural College, Univ. of Guelph, 
Guelph, Ont. NIG 2W1. 

Scott, T. W., Mt. Pleasant, J., Burt, R. F., and Otis, D. J. 
1987. Contributions of ground cover, dry matter, and 
nitrogen from intercrops and cover crops in a corn pol- 
yculture system. Agron. J. 79:792-798 

Stone, J. A. 1987. Compaction and the surface structure of 
a poorly drained soil. Trans. ASAE 30:1370-1373. 

Stone, J. A. 1988. Field crop and forage effects on seasonal 
values of structural properties of a poorly drained clay 
loam soil. In W. E. Larson, G. R. Blake, R. R. Allmaras, 
and S. C. Gupta (ed.) Mechanics and Related Processes 
in Structured Agricultural Soils. Kluwer Academic 
Publishers, Dordrecht. ISBN 0-7923-0342-3. 

Stone, J. A. 1989. Comparison of hammer and hydraulic 
core samplers for soil bulk density and porosity deter- 
minations. ASAE/CSAE Paper no. 89-2191. ASAE, St. 
Joseph, MI. 

Stone, J. A., N. H. E. Allen, and C. D. Grant. 1985. Corn 
fertility treatments and the surface structure of a 
poorly drained soil. Soil Sci. Soc. Am. J. 49:1001-1004. 

Stone, J. A. and B. R. Buttery. 1989. Nine forages and the 
aggregation of a clay loam soil. Can J. Soil Sci. 69:165- 
169. 

Stone, J. A. and L. C. Heslop. 1987. Blade cultivator, 
ridge, and moldboard plow tillage comparison on a 
poorly drained soil. Trans. ASAE 30:61-64. 

Stone, J. A. and L. L. Logan (ed.). 1989. Agricultural 
Chemicals and Water Quality in Ontario. Proc. Work- 
shop, Kitchener, Ont., Nov. 17-18, 1988. ISBN 0- 
921447-05-1. (Copies available from J. A. Stone, Agri- 
culture Canada Res. Branch, Res. Sta., Harrow, Ont. 
NOR 1GO). 

Stone, J. A., J. A. McKeague, and R. Protz. 1986. Corn root 
distribution in relation to long term rotations on a 
poorly drained clay loam soil. Can. J. Plant Sci. 67:231- 
234. 

Stone, J. A., T. J. Vyn, H. D. Martin, and P. H. Groenevelt. 

1989. Ridge-tillage and early-season soil moisture and 
temperature on a poorly drained soil. Can. J. Soil Sci. 
69:181-186. 

Stone, J. A. and K. C. Wires. 1989. Moisture content and 
soil core volume on Brookston clay loam. Can. J. Soil 
Sci. (In press, copies available from J. A. Stone, Agric. 
Canada Res. Branch, Res, Sta., Harrow, Ont. NOR 
1GO). ). 



STONE: SOIL MANAGEMENT RESEARCH ON THE CLAY SOILS OF SOUTHWESTERN ONTARIO - A REVIEW 



203 



Wicklund, R.E., and Richards, N.R. 1961. The soil survey 
of Oxford County. Report No. 28 of the Ontario Soil 
Survey. Ontario Institute of Pedology, Guelph Agric. 
Centre., Box 1030, Guelph, Ont. N1H 6N1. 

Answers to Questions 

1- Larry Wilding: What are the problems in using the 
Guelph permeameter to characterize soil degrada- 
tion? 

First, I'd like to say that although the Guelph per- 
meameter has not worked well as a measure of soil struc- 
tural degradation for me on clay loam soil, I have not 
worked extensively with the technique. The major prob- 
lems have been 1) boring a hole that does not intersect 
with a crack in the soil and 2) the amount of time required 
to obtain a measurement (because of the low hydraulic 
conductivity of these soils). I think that 1) can be overcome 
by conducting measurements when cracks are not present 
(in the spring or fall) and 2) by using a bank of permeame- 
ters. However, seasonal labour may not be available in the 
spring and the fall and using a bank of permeameters is 
expensive. 



2. Larry Wilding: Have you explored with the applica- 
tion of micromorphology as a analytical tool to docu- 
ment structural degradation and changes in shape, 
size, and distribution of macroporosity? 
Yes, macro- and microstructure were described and 
water desorption characteristics were measured for 
Brookston clay loam soil under different cropping sys- 
tems. This information is presented in the following publi- 
cation: 

McKeague, J. A., C. A. Fox, J. A. Stone, and R. 
Protz. 1987. Effects of cropping system on structure of 
Brookston clay loam in long-term experimental plots at 
Woodslee, Ontario. Can. J. Soil Sci. 67:571-584. 



204 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Soil Classification Related Properties of Salt-Affected Soils 

L Szabolcs 1 

Abstract 

Salt-affected soils which occur under different environmental conditions 
and have diverse morphological, chemical, physical, physico-chemical, and 
biological properties should be included in one big group, according to their 
common feature, the dominating influence of electrolytes on their forma- 
tion and properties. 

In the various soil classification systems, the names and places of salt- 
affected soils on the taxonomic levels are diverse, even though their mor- 
phological diagnostic features and their chemical, physical, and physico- 
chemical properties are very similar. 

The following properties are significant for salt-affected soils in most 
classification systems: 

I/ Morphology of the soil profile (presence or absence of diagnostic hori- 
zons). 

2/ Significant physical properties (mainly for the Solonetz group). 

3/ Chemical and physico-chemical properties: 

/a/ Salt content, salt composition, and salt distribution in the profile 
and, in some cases, also in the ground water. 

fbi Exchangeable sodium percentage and sodium adsorption ratio. 
/c/ pH conditions and the existence of sodium carbonate. 

Most of the listed properties are included in the main international or 
national classification systems but on different levels with different limit 
values and sometimes with different interpretations. 



Definition and Characteristics of 
Salt-Affected Soils 

Salt-affected soils occur in more than a 
hundred counties, from above the polar circle to 
the equator. They occur under different envi- 
ronmental conditions and have diverse morpho- 
logical, chemical, physical, physico-chemical, 
and biological properties. 

Although the definition of salt-affected soils, 
and the taxonomic level at which they are classi- 
fied, differ in the different countries and classifi- 
cation systems, all types, kinds, or varieties of 
these soils mainly are characterized by the pres- 
ence and/or influence of water 
soluble salts in certain horizons 
or layers of the soil profile. 

While the decisive influence 
of electrolytes is common in all 
salt-affected soils, their chemi- 
cal composition and concentra- 
tion vary, depending on local 
conditions. Consequently, dif- 
ferent types of salt-affected soils 
develop with diverse properties. 

Table 1, based on the above 

1 Research Institute for Soil Science 
and Agricultural Chemistry of the 
Hungarian Academy of Sciences, 
P.O.B. 35. H-1525 Budapest, Hun- 
gary. 



considerations, demonstrates a simple but prac- 
tical grouping system for salt-affected soils. It 
classifies them in compliance with the types of 
electrolytes causing salinity and/or alkalinity. 

The taxonomic units of Table 2 have been se- 
lected according to this paper's definition of salt- 
affected soils and their grouping in Table 1. 

In the so-called genetic soil classification sys- 
tem which is common in the USSR (Kovda- 
Rozanov, 1988) and many other countries, So- 
lonetz and Solonchak and sometimes Solod soils 
are high taxonomic units subdivided into types, 
subtypes, and varieties. 



Table 1. Grouping of salt-affected soils 


Electrolytes(s) causing 
salinity and/or 
alkalinity 


Type of sal t- 
aflected soil 


Environment 


Main adverse effect 
on production 


Method for 
reclamation 


Sodium chloride and 


Saline soils 


Arid and 


High osmotic pressure 


Removal of 


sulphate (in extreme 




semi-arid 




excess salt 


cases - nitrate) 








(leaching) 


Sodium ions capable of 


Alkali soils 


Semi-arid 


Alkali pH 


Lowering or 


alkaline hydrolysis 




Semi-humid 


Effect on water 


neutralizing 






Humid 


physical soil 


the high pH 








properties 


by chemical 










amendments 


Magnesium ions 


Magnesium 


Semi-arid 


Toxic effect, high 


Chemical 




soils 


Semi-humid 


osmotic pressure 


amendments 










Leaching 


Calcium ions 


Gypsiferous 


Semi-arid 


Acidic pH, toxic 


Alkaline 


(mainly CaSO 4 ) 




Arid 


effect 


amendments 


Ferric and aluminum 
ions (mainly sulphates) 


Acid sulphate 
soils 


Sea shores, 
lagoons with 


Strongly acidic pH, 
toxic effect 


Liming 


heavy, sulphate 






containing 










sediments 







SZABOLCS: SOIL CLASSIFICATION RELATED PROPERTIES OF SALT-AFFECTED SOILS 



205 



Most of the other soil classification systems 
place the different salt-affected soils similarly to 
one or another of the three classification sys- 
tems described above, (1. SZABOLCS, 1989; 2. 
Key to Soil Taxonomy, 1985; 3. Kovda and 
Rozanov, 1988), or in a manner which combines 
these systems' approaches. 

The soil classification system in Canada is 
based on soil genetical principles but uses the 
terms and taxonomic structures of both the clas- 
sification in the FAO/UNESCO World Soil Map 
and the US Soil Taxonomy (The Canadian Sys- 
tem of Soil Classification, 1978). 

The recent Australian soil classification prin- 
cipally is based, even more than the Canadian 
system, on traditional genetic terms and ele- 
ments, but in this system a nearly complete cor- 
relation with the US Soil Taxonomy and with 
the Legend of the FAO/UNESCO World Map 
has been elaborated (Soils, an Australian view- 
point, 1983). 

Discrepancies among classification systems' 
heirarchical placement of various salt-affected 
soils often make it difficult to find an approxi- 
mate equivalent in one system of a type or vari- 
ety defined in another system. However, the de- 
finitive role of electrolytes in the characteriza- 
tion of all salt-affected soils can facilitate the 
conversion of terms for such soils among the dif- 
ferent classification systems. 

Main Properties of Salt-Affected 
Soils Underlying Classification 

Morphological and Physical Soil 
Properties Related to the Classification of 
Salt-affected Soils 

In the three above-mentioned soil classifica- 
tion systems, and in almost all common soil clas- 
sification systems, the main diagnostic morpho- 
logical and physical feature for the characteriza- 
tion of Solonetz soils, Solodic soils, and the ma- 
jority of alkali soils is the existence of a natric 
horizon that is used in the US Soil Taxonomy 
and the FAO/UNESCO Soil Map Legend. In the 
genetic and related soil classification systems, 
the criterion is the presence of nearly identical 
horizon, namely the horizon B. 

The natric horizon is an argillic horizon which 
has prisms or, more commonly, columns in the 
subsurface, as well as blocky structures and 
tongues of an eluvial horizon. Other character- 
istics of natric horizons are mentioned in the 
next section of this paper. 



Table 2. Salt-affected soil in the 
hierarchy of the US 


Soil Taxonomy 


Order 


Suborder 


Great Group 


Alfieols 


Aqualfe 


Natraqualfe 




Boralfe 


Natriborals 




Udalfe 


Natrudalfe 




Ustalfe 


Natrustalfs 




Xeralfs 


Natrixeralfe 


Aridisols 


Argids 


Natragids 




Orthids 


Salorthids 






Gypsiorthids 


Entisols 


Aquents 


Sulfaquents 


Inceptisols 


Aquepts 


Sulfaquepts 


Mollisols 


AquollB 


Natraquolls 




Borolls 


Natriborolls 




Ustolls 


Natrustolls 




Xerolls 


Calcixerolls 






Natrixerolls 



The depth of the natric horizon along the pro- 
file is not referred to in the US Soil Taxonomy or 
in the FAO/UNESCO Soil Map Legend. How- 
ever, in the genetic soil classification system, 
the position of this horizon along the profile de- 
termines the subdivision of Solonetz soils: 

- If the natric horizon is 0.8 cm - shallow So- 
lonetz 

- between 9-15 cm - medium Solonetz 

- more than 16 cm- deep Solonetz 

As a diagnostic feature, the natric horizon or 
B horizon is the most universal property among 
the main soil classification systems. 

The physical and morphological properties of 
salt-affected soils are closely interrelated with 
their physico-chemical diagnostic properties, 
which jointly serve as crieteria in the soils' clas- 
sification. 

Chemical and Physico-chemical Soil 
Properties Related to the Classification of 
Salt-Affected Soils 

ESP, SAR values and related physico-chemical 
properties 

The sodium adsorption ratio (SAR) and the 
exchangeable sodium percentage (ESP) play sig- 
nificant roles. According to the US Soil Taxon- 
omy, if SAR is >13 (or 15 percent or more satura- 
tion with exchangeable sodium) in some sub- 
horizon within 40 cm of the upper boundary; or 
if there is more exchangeable magnesium plus 
sodium than calcium plus exchange acidity (at 
pH 8.2) in some subhorizon within 40 cm of the 
upper boundary if the SAR is ^13 (or ESP ^15) 
in some horizon within 2 m of the surface. 

The 2 m depth in this definition seems too 
deep, because the diagnostic natric horizon 
must be much higher in the profile if it deter- 
mines the soil type. 



206 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



The diagnostics of Solonetz soils in the FAO/ 
UNESCO Soil Map Legend are very similar to 
those in the US Soil Taxonomy. In the genetic 
soil classification systems, the diagnostic char- 
acteristics besides the morphological features 
(see in (A)) are based mainly on the ESP values 
in the B horizon. The limits are very close to 
those in the US Soil Taxonomy. 

It should be noted that the ESP values must 
be always considered together with the morpho- 
logical features of the B horizon or illuvial hori- 
zon, because in some places typical morphology 
of natric horizon develops with comparatively 
low ESP and/or SAR values (e.g., ESP <7-10, in 
the Ukraine), while in other places no such 
natric horizon develops even in cases of high 
ESP value (ESP >20-25 in Sudan). The causes 
of such discrepancies can be different. Local dif- 
ferences in the definitions of appropriate diag- 
nostic values of ESP in the sodic horizon need 
further study. 

Nevertheless, the similarity of diagnostic ESP 
values in the different soil classification systems 
makes possible the correlation of sodic soils or 
alkali soils or solonetz soils or natric soils in the 
different soil classification systems. 

Salt content of the soil profile 

The salt content of soils (often called salinity) 
plays a significant role in all soil classification 
systems as diagnostic criteria for saline soils, 
solonchak soils, etc. 

According to the US Soil Taxonomy the salic 
horizon is the main diagnostic feature for saline 
soils, which in this system belong mainly to the 
Aridisols. The diagnostics in FAO/UNESCO 
Soil Map Legend are similar for Solonchak soils. 
In the genetic soil classification system, too, 
analogously to the salic horizon, a saline horizon 
is the key diagnostic feature for Solonchak soils. 

In the US Soil Taxonomy, the salic horizon is 
a horizon 15 cm or more thick that contains a 
secondary enrichment of salts more soluble in 
cold water than gypsum. It contains at least 2% 
salt. 

In the FAO/UNESCO Soil Map Legend as 
well as in the genetic soil classification system, 
the limit values for salt content are somewhat 
lower; they do not restrict the salt accumulation 
to secondary enrichment. The genetic soil clas- 
sification system also includes the gypsum con- 
tent in the total salinity. 



In spite of the fact that the convertibility of 
terms for the different saline soils is possible 
based on the main diagnostic features of the salt 
content of the salic horizon, certain discrepan- 
cies exist among the diagnostics of different soil 
classification systems in respect to salinity. The 
more important ones are as follows: 

- While in the US and many other countries 

salinity commonly is determined by meas- 
uring the electric conductivity of soil paste 
or saturation extract (EC values), in the 
USSR, and many other countries where the 
genetic soil classification system is used, 
the chemical analysis of 1:5 or 1:1 aqueous 
extract is usual. Such differences make it 
difficult to convert the analytical values of 
classification units of different systems. 
Because of the different chemical composi- 
tion of salts, the chemical analysis of aque- 
ous extracts may produce salinity values 
which differ from those calculated from the 
EC values. 

- The measurement of electric conductivity 
does not reveal the chemical composition of 
salts in the soil which is disclosed by the 
chemical analysis of an aqueous extract. 
EC determination does not show either 
whether the soluble salts, mainly sodium 
salts, are neutral, like sodium chloride, or 
are sodium salts capable of alkaline hy- 
drolysis, like sodium carbonate or sodium 
bicarbonate. 

The US Soil Taxonomy does not distinquish 
between saline soils dominated by neutral so- 
dium salts and those dominated by sodium salts 
capable of alkaline hydrolysis. In the genetic 
soil classification system, soda Solonchak soils 
are separated from chloride and sulfate Solon- 
chak soils. This approach should be considered 
because pedological properties, agronomical 
value, and the possibilities of reclamation of sa- 
line soils and reclamation strategies strictly 
depend on whether the soils have nearly neutral 
pH or strongly alkaline pH values. 

The limit values for salinity should be differ- 
ent, depending on the chemistry of salt composi- 
tion of the soils. Evidently if the salt carbonate 
dominates the salt composition, much lower 
concentrations result in adverse soil properties 
than if sodium chloride or sodium sulfate is 
dominant. Another condition exists when the 
presence of Mg salts dominate. This conditions 
is the easiest to reclaim. 



SZABOLCS: SOIL CLASSIFICATION RELATED PROPERTIES OP SALT-AFFECTED SOILS 



207 



A Few More Questions to be 
Answered 

In spite of the fairly good correlation of the 
properties of salt-affected soils among the differ- 
ent classification systems, some problems need 
study, discussion, and agreement. From those 
only a few important ones are mentioned below. 

(1) If we accept the definition of salt-affected 
soils described in the first part of this pa- 
per, gypsic soils, calcic soils, sulfatic and 
sulfidic, including acid sulfate, soils should 
be included in the system. In this case, 
according to the US Soil Taxonomy, the soil 
classification related properties of salt-af- 
fected soils can include calcic horizons and 
k horizons; partly gypsic, petrogypsic, and 
sometimes petrocalcic horizons; and even 
sulfuric horizons. (Also see Tables 1 and 2). 

(2) In the practice of irrigation, the RSC (Re- 
sidual Sodium Carbonate content in meq/L) 
can be a diagnostic feature for irrigation 
water quality. If the highly alkaline pH 
value of the soil and/or the free sodium car- 



bonate plus biocarbonate content of the dif- 
ferent chemical types of soil salinity, a more 
advanced classification could be elaborated 
which meets the practical requirements of 
irrigation and soil reclamation. 
(3) The possibility of the occurrence of a salic 
and/or natric horizon in Vertisols and even 
Histosols should be studied. 

Literature Cited 

The Canadian System of Soil Classification. 1978. Can- 
ada Soil Survey Committee, Research Branch, Canada 
Department of Agriculture. Publication 1646. 

FAO/UNESCO Soil Map of the World. 1988. Revised 
Legend. World Soil Resource Report 60, FAO, Rome. 

Keys to Soil Taxonomy. 1985. Technical Monograph No. 
6, Agronomy Department, Cornell University, Ithaca, 
New York. 

Kovda, V.A. and E.G. Rozanov. 1988. Pochvovedenie (Soil 
Science) I-II. Vyshaya Shkola, Moscow. 

Soils, an Australian viewpoint. 1983. CSIRO/Academic 
Press, Melbourne. 

Szabolcs, I. 1989. Salt-affected soils. CRC Press Inc., Boca 
Raton, Florida. 



208 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Clayey Soils of Northern Canada and the Cordillera* 

C. Tarnocai 1 , G.E Mills 2 , H. Veldhuis 3 , EL Luttmerding 4 , and A. Green 5 

Abstract 

Clayey soils which developed on glaciolacustrine, marine, and glacial till 
materials are found dispersed throughout northern Canada and the Cordil- 
lera. Permafrost occurs in all soils in the arctic, in many soils in the subarc- 
tic, and in some, usually poorly drained, soils in the boreal. Most of the soils 
containing permafrost have strongly cryoturbated horizons. In the boreal 
and the southern part of the Cordillera, well and imperfectly drained clayey 
soils usually have either a brownish B horizon (Bm) or a B horizon with sig- 
nificant clay accumulation (Bt). In the extreme southern part of the Cordil- 
< lera, the clayey soils often have a moderately thick, organic-rich surface 

horizon (Ah). When located in poorly drained, depressional areas, they have 
gleyed horizons* 

Arctic and subarctic clayey soils show no evidence of slickensides; crack- 
ing, if present, is due to thermal or frost cracking, rather than desiccation. 
Slickensides have been observed in some of the boreal (Manitoba) and Cor- 
dilleran (southern interior of British Columbia) clayey soils. Some of these 
boreal and Cordilleran soils also show evidence of cracking, but without sig- 
nificant grumic or churning characteristics. 

According to present criteria, clayey soils in northern Canada and the 
Cordillera can not be classified in any of the Vertisol suborders of the 
American Soil Taxonomy. Vertic properties, if present, are not strong 
enough to warrant the soils being classified in the proposed Cryert suborder. 



Introduction 

The study area covers northern Canada, in- 
cluding the Arctic, Subarctic, and North Boreal 
Ecoclimatic Provinces, as well as the Cordilleran 
Ecoclimatic Provinces (Figure 1). This large 
area lies north and west of the Grassland and 
South Boreal Ecoclimatic Provinces (Figure 1), 
the subject of the presentation by Acton et al. 
(these proceedings). 

Soils having clayey textures occupy small, but 
significant, areas in this large region. The most 
common materials oh which these clayey soils 
have developed are glaciolacustrine and marine 
deposits, clayey glacial till, and residual materi- 
als derived mainly from shales. 

The objective of this study is to describe the 
distribution and properties of clayey soils and 



"Land Resource Research Centre Contribution No.: 90- 
33. 

l Land Resource Research Centre, Research Branch, 
Agriculture Canada, K.W. Neatby Bldg., Ottawa, Canada, 
K1A OC6. 

2 Manitoba Soil Survey, Room 362, Ellis Bldg., Univer- 
sity of Manitoba, Winnipeg, Canada, R3T 2N2. 

3 Canada Soil Survey, Agriculture Canada, Room 362, 
Ellis Bldg., University of Manitoba, Winnipeg, Canada, 
R3T 2N2. 

4 British Columbia Ministry of the Environment, Policy 
and Planning Branch, 777 Broughton Street, Victoria, 
Canada, V8V 1X5. 

5 Canada Soil Survey, Agriculture Canada, 6660 N.W. 
Marine Drive, Vancouver, Canada, V6T 1X2. 



their development as influenced by the range of 
climatic conditions and associated vegetation 
across the study area. The effects of cryogenic 
processes on the morphology of northern soils 
affected by permafrost is contrasted to the effect 
of shrink-swell processes on the morphology of 
southern clayey soils not affected by permafrost. 
The criteria used to classify permafrost-affected 
soils are compared with those used in the classi- 
fication of Vertisols. 

Physical Environment 

Physiographic Regions 

Physiographically and geologically, Canada is 
composed of two large and distinctly different 
areas: a core of old, massive, Precambrian crys- 
talline rocks forming the Canadian Shield, and 
a surrounding younger, mainly sedimentary 
rock area forming the Borderlands (Figure 2). 
Further subdivisions of these two major physi- 
ographic units in northern and western Canada 
are shown in Figure 2, with the description de- 
rived mainly from the work of Bostock (1970) 
and Prest (1970). Clayey soils occurring 
throughout these areas have developed under a 
diverse range of climate and vegetation condi- 
tions (Figure 3). 

A major part of Canada has been subjected to 
repeated glaciation during the Pleistocene ep- 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OP NORTHERN CANADA AND THE CORDILLERA 209 



ECOCLIMATIC PROVINCES 

LEGEND 
ECOCUMATIC PROVINCES 

1 ARCTIC 

2 SUBARCTIC 

NOftTHtOREAL 

35 SOUTH BOREAL 

4 COOL TEMPERATE 

5 MODERATE TEMPERATE 
IA AMD ORASSLAND 

IB TRANSmOMAL GflASfiLAMD 

7 SUBAACTK COHOUERAN 

I NORTH MTEMOft COflMlERAN 

B SOUTH HfTEfflM COdOtUHUN 

H PAORC COmUJLEflAN 



KFWPX 

&>^^vr$vSi 




Southern Umtt of the Study Area 



Figure 1. Ecoclimatic provinces of Canada (Ecoregion Working Group 1989). 



och. As a result, the Canadian Shield is charac- 
terized by a thin cover of stony, sandy till. Fine- 
textured (silty clay and clay) glaciolacustrine 
deposits and gravelly or sandy beach materials 
have been deposited in the basins of former gla- 
cial lakes. These clayey glaciolacustrine depos- 
its are common in north-central Manitoba and 
parts of northwestern Ontario, but sporadic or 
non-existent in the rest of the Canadian Shield. 

The Hudson Region is a low-lying, peatland- 
dominated area on the south and west coasts of 
Hudson Bay. A large part of this region was 
inundated by the sea during the early Holocene 
epoch and clayey deposits are now found in the 
northern Ontario portion of this area. 

The Interior Plains Region is composed of a 
series of plains, hills, and plateaus which rise 
gradually from the arctic coast southward. The 
area is covered mainly by clayey or loamy till. 
During the retreat of the glacial ice, a series of 
glacial lakes occupied most of the upper and 



central Mackenzie River Valley and the present 
Great Slave and Great Bear Lakes. Clayey and 
sandy deposits are common in the basins of 
these former glacial lakes. 

The Arctic Lowlands Region consists of rela- 
tively gently rolling terrain. The loamy till 
cover is generally thin and bedrock outcrops are 
common. Clayey deposits usually are found only 
in areas along the coast which were inundated 
after the retreat of the glacial ice. 

The Innuitian Region has a varied topogra- 
phy and is more rugged than the surrounding 
physiographic regions. The terrain is composed 
largely of bedrock associated with a veneer of 
sandy and loamy material. Along the coast the 
deposits are generally clays of marine origin. 

The Arctic Coastal Plains Region is a strip 50 
to 100 km wide along the shores of the Arctic 
Ocean from Meighen Island to Alaska. Some 
clayey marine deposits are found in the present- 
day Mackenzie River delta area and on Herschel 



210 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Island. The eastern part of 
the Yukon coast is covered 
by clayey till; the central 
and western parts are char- 
acterized by clayey colluvial 
deposits. 

The Cordilleran Region is 
divided longitudinally into 
three belts, the Eastern, In- 
terior, and Western Sys- 
tems, each with its own 
characteristic geology and 
physiography. Glacial 
lakes developed in valleys 
temporarily blocked by gla- 
cial ice. The deposits left 
behind in the larger basins 
often are composed of 
clayey materials. The larg- 
est of these occur in the cen- 
tral interior of British Co- 
lumbia, the Whitehorse 
area of the southern Yukon, 
and the Old Crow area of 
the northern Yukon. 
Clayey materials of marine 
origin occur along the 
southern coastal areas of Vancouver Island and 
in the lower Fraser Valley of British Columbia. 

Ecoclimatic Provinces 

The ecoclimatic provinces of Canada were 
mapped and described by the Ecoregion Work- 
ing Group (1989) and their distribution is shown 
in Figure 1. According to this work, the Arctic 
Ecoclimatic Province encompasses all those ar- 
eas lying north of the arctic tree line, which 
reaches its most northerly limit in the Macken- 
zie River delta (approximately Lat. 69) and its 
most southerly limit along the shores of Hudson 
Bay (approximately Lat. 57). The Arctic 
Ecoclimatic Province occupies approximately 
40% of the area of Canada and includes the 
northern continental area and all of the arctic is- 
lands. 

The Subarctic Ecoclimatic Province encom- 
passes the area south of the arctic tree line that 
has open coniferous forest vegetation. The 
North Boreal Ecoclimatic Province encompasses 
those areas south of the Subarctic Ecoclimatic 
Province dominated by closed canopy coniferous 
forest vegetation. The Cordilleran Ecoclimatic 
Provinces, which include the Subarctic Cordille- 
ran, North Interior Cordilleran, South Interior 
Cordilleran and Pacific Cordilleran Provinces, 
encompass the mountains and plateaus of west- 



Figure 2. Physiographic regions of Canada (Bostock 1970). 




em and northwestern Canada. Climate-vegeta- 
tion-soil relations throughout the Cordillera are 
governed by vertical as well as latitudinal zona- 
tion. 

Atmospheric Climate 

Climatic data for some selected stations in the 
study area are presented in Figure 4. The cli- 
mate of the Arctic Ecoclimatic Province is char- 
acterized by long, cold winters and short, cool 
summers with low annual precipitation. Ap- 
proximately 40 to 60% of the total precipitation 
occurs as snow. The winter is characterized by 
little daylight, the summer by long periods of 
daylight. 

The climates of the Subarctic and North Bo- 
real Ecoclimatic Provinces are characterized by 
long, cold winters and short, cool summers with 
moderate precipitation. The summers become 
warmer as one proceeds southward. 

The climates of the Cordilleran Ecoclimatic 
Provinces vary greatly from south to north and 
with elevation. In the north the climate is simi- 
lar to that described for the Subarctic Ecocli- 
matic Province. In the south, however, at low 
elevations the climate is temperate with warm, 
dry summers and cool winters; at higher eleva- 
tions the climate becomes increasingly severe 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OP NORTHERN CANADA AND THE CORDILLERA 211 




Figure 3. Distribution of clayey soils in the study area and the location of 
climatic stations. 



and Is similar to that of mountainous areas to 
the north. Along the Pacific Coast the climate is 
oceanic with mild winters, cool summers and 
high precipitation. 

In Figure 4 both precipitation and Thornth- 
waite potential evapotranspiration are plotted 
for ten climatic stations. These graphs provide a 
general picture of the atmospheric moisture bal- 
ance of the study area. In the Pacific Cordille- 
ran (Victoria and Abbotsford stations) and 
South Interior Cordilleran (Armstrong) Ecocli- 
matic Provinces, a moisture deficiency occurs 
because of low precipitation during the summer. 
In the North Interior Cordilleran (Dawson 
Creek, Whitehorse, and Mayo), North Boreal 
(Wabowden, Fort Simpson, Kapuskasing, and 
Moosonee), Subarctic (Norman Wells), Subarc- 
tic Cordilleran (Dawson City) and Arctic 
(Tuktoyaktuk and Resolute) Ecoclimatic Prov- 
inces, most precipitation occurs during the sum- 
mer months. The increased summer precipita- 
tion at the stations given for these areas, how- 
ever, is not sufficient to compensate for 
evapotranspiration, and thus moisture deficien- 
cies also occur. In the Ontario portion of the 
North Boreal Ecoclimatic Province, however, 
precipitation is significantly higher than in 
other portions of this ecoclimatic province, with 



the result that little mois- 
ture deficit occurs. 

Throughout northern 
Canada and the Cordil- 
lera, soils with perma- 
frost never completely 
dry out at depth, because 
of the gradual release of 
moisture to the soil as the 
seasonal frost melts. Soil 
moisture deficiency is 
therefore minimal in soils 
containing permafrost, 
even in areas that have 
an atmospheric moisture 
deficiency. 

Soil Climate 

The study area falls 
into four broad soil cli- 
matic regions: the Arctic, 
Subarctic, Cryoboreal, 
and Boreal (Clayton et 
al., 1977; Tarnocai, 
1978), a modified version 
of which is shown in Fig- 
ure 5. The Arctic Soil Cli- 
mate is characterized by 
extremely cold soil temperatures. The MAST 
(mean annual soil temperature at 50 cm depth) 
is <-7C and the MSST (mean summer soil tem- 
perature at 50 cm depth) is between 0C and 
<3C. The moisture regime of these soils ranges 
from aquic to humid. 

The Subarctic Soil Climate is characterized 
by very cold soil temperatures. The MAST 
ranges from -7* C to <2C, the MSST from 3C to 
<6C. The moisture regime of these soils ranges 
from aquic to humid. 

The Cryoboreal Soil Climate is characterized 
by cold to moderately cold soil temperatures. 
The MAST ranges from 2C to <5.5C, the 
MSST from 6C to <12C. The moisture regimes 
of these soils are humid to subhumid in north- 
ern British Columbia and Alberta, humid to 
aquic in northern Saskatchewan and Manitoba, 
and subaquic to perhumid in northern Ontario. 
The Boreal Soil Climate occurs only in the 
southern interior of British Columbia and is 
characterized by moderately cold soil tempera- 
tures. The MAST ranges from 5.5 C to <8C, 
the MSST from 12C to <18C. The moisture 
regime of these soils ranges from subhumid to 
semiarid. 



212 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



PACIFIC CORDILLERAN 
Victoria, Brltiafi Columbia 

4839'N 12326'W 19m 



PACIFIC CORDILLERAN 
Abbotaford, Brttiaft Columbia 

49-2'N 122-22'W 56 m 




Mean annual temperature; 9.5C. Total annual precipitation: 872.9 mm 



SOUTH INTERIOR CORDILLERAN 
Armatrong, Brltlah Columbia 

5026'N 119*12'W 375m 



c 


MM 




50 


160 


- 


40 


140 




30 

20 


120 
100 


/ \ 

x X V 






' * * 


10 


80 


X % 

/ X 





60 


f \ * 


-10 


40 


' .Nf * / 4^"^ 4 +A ^ 


-20 


20 


. * V^*-^ \ 


-30 





* * '^ ' ' ' ' ' ' \ J- 



MAMJJASOND 
Month* 

Mean annual tamperature: 7.1 C C. Total annual predoHaiion: 456.8 mm 



NORTH INTERIOR CORDILLERAN 
WhHahoraa, Yukon Tarritory 

60'43'N 135'4'W 703m 



c 


MM 




30 


140 


- 


20 


120 


X* ^ ^ 






/ v 






f * V 


10 


100 


' \ 






/ ' 





80 


X \ a 






1 * 


-10 


60 


. * ! x ^ 






f * * 


-20 


40 




30 
-40 


20 






J F M A M J J A S N D 



Months 
Maan annual temperature: -1 .2XX Total armud precipitation: 261. 2 mm 



*C MM 






240 







220 


- 


f 


200 






180 






50 160 






40 140 






30 120 


y 




20 100 


xV^, / 






X * \ T" 




10 80 


. ' '*v_ X 





60 


* '' \ / N 







X \ J* 




10 40 


\S 


X 

> 




X*-* 


\ 


20 20 




^% 


30 


/" ,,,,,,,, 


**x 


JFMAMJ JAS 


N D 


Months 




Mean annual temperature: 9.5C. Total annual precipitation: 1 513.0 mm 


NORTH INTERIOR CORDILLERAN 




Dawaofi Craafc, Brltlah Columbia 




55'44'N 120T1VW 655m 




C MM 






50 160 






40 140 


- 




30 120 


*-'\ 






. % 






' X 




20 100 






10 80 


J * * * \ 




60 
10 40 


, X /-+--+ % x 

L. 1 / A. * 
' / \\ 

+ . * / / v 


* 


-20 20 


~^* 


^4- 


jn A 


*, A. +* \ I 1 1 1 1 


1 \ ^ 


JFMAMJ JAS 


N 


Months 




Mean annual temperature: 0.9C. Total annual precipitation: 503.7mm 


NORTH INTERIOR CORDILLERAN 




Mayo, Yukon Territory 




'C MM 63 " 37 ' N 135852 ' W S 04 




40 140 


. 






-f-'^\ 




30 120 


/ \ 




20 100 


/ 




10 80 


t * \ 




60 


: ^ v 





-10 40 


I /^^V I 






f ' /* Xy V 




20 20 
-30 








JFMAMJ JASOND 

Months 
Mean annual tmperatur:-4.0*C. Total annual pradpftatfon: 306.3 mm 



Figure 4. Mean monthly temperature and precipitation and Thornthwaite potential evapotranspiration for climatic stations 
in areas of major clay concentrations (Atmospheric Environment Service 1982, a and b). These climatic stations and the 
associated ecoclimatic provinces are as follows: Victoria, B.C. and Abbotsford, B.C. - Pacific Cordilleran; Armstrong, B.C. - 
South Interior Cordilleran; Dawson Creek, B.C. and Whitehorse, Y.T. - North Interior Cordilleran; Wabowden, Man. and 
Fort Simpson, N.W.T. - North Boreal; Norman Wells, N.W.T. - Subarctic; Tuktoyaktuk, N.W.T. and Resolute, N.W.T. - Arctic, 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OP NORTHERN CANADA AND THE CORDILLERA 213 



NORTH BOREAL 
Wabowden, Manitoba 

54'55'N 96*38'W 233m 



NORTH BOREAL 
Fort Simpson. N.W.T. 



C MM 


_-, ., ., Lcutrnu 




50 160 


-I Precipitation 50 160 






X Potarrtial Evapotransptratton 


X 

s \ 


40 140 


j*~ 40 140 


[ x' \ 


30 120 


/ \ 30 120 


' x 


20 100 


* V 
/ % \ 20 100 


/ 






/ .\ 


10 80 


/ , \ . 10 80 


X * 






9 \ 


60 


* f \ * W 


1 \ 

I ^4_4. ^ * 


-10 40 


i^~ \, -10 40 


' J^"^ "^4^ 




* S* V4 ~* h-^i_ 


.^^ ^^. 


-20 20 


-f 14^" ! K * -20 20 
<i * \ 


- -K^^^ V 1 "^ v 4 """" 1 ^ 


-30 




ft^J,J?l , , , , \i, ^ ^ 


JFMAMJJASOND JFMAMJJASO N 6 


Months Months 



Mean annual temperature: -2.2C. Total annual precipitation: 464.2 mm 



Mean annual temperature: -4.CTC. Total annual precipitation: 350.5 mm 



C MM 


NORTH BOREAL 
Kapu*kalng, Ontario 

49-25'N 8228'W 226m 


*C MM 


NORTH BOREAL 
Moosonee, Ontario 

51*16'N 80"39'W 10m 


40 140 
30 120 


x 
/ \ 

/ \ 


40 140 
30 120 


A\. 


20 100 


^ 

9 / \ 


20 100 


*/+ ^ 


10 80 


9 ^ 4 "^" r "" f "^ fNs +--f 


10 80 


. "/V^N-V^.^. 


60 
-10 40 
-20 20 
-30 


: ,^-y '\ \ 

x \ 

( 1 1 1 1 I 1 1 1 1 1 I 


60 
-10 40 
-20 20 
-30 


- ; ^] \ ^ 
- 

, , , , ; , , , , , , \ , . 


J FMAMJ J A S O N D 


JFMAMJJASOND 


Months 


Months 



Mean annual temperature: 0.5 C. Total annual precipitation: 858.2 mm 



Mean annual temperature : -1 .1 *C. Total annual precipitation: 727.7 mm 



SUBARCTIC CORDILLERAN 
Dawfton City, Yukon Territory 

64-03'N 139^6'W 320m 



SUBARCTIC 
Norman Wells, N.W.T. 

65*1 T N 1 26*48' W 73m 



C MM 


C MM 




50 160 


50 160 


x-'^ 


40 140 


/ X ^ 40 140 


' \ 


30 120 


/ \ 30 120 
/ X 


: ' \ 


20 100 


j \ 20 100 


m \ 




/ \ * 


1 * % 


10 80 


f N 10 80 

; \ * 


J I* 


60 


1 1 60 


x \ 

i y^""^ x 




1 _L * 




-10 40 


1 ,^^-H 4v | -10 40 


/ or \\ 


.20 20 


* ' S NiJL-4- 
. , 1 Y I m20 2 


- i^-f^^-f^-f V"^^-r- 




i i T"""i 1 < i t i i JL i ,-0 o 


a *""""! """"a? . i . V * *-- 


-30 


J > M A M J J A S N D 'J' F M K M J J A S O N D 


Months Months 



Mean annual temperature: -5.1 C. Total annual precipitation: 306.1 mm 



Mean annual temperature: -.4*C. Total annual precipitation: 328.4 mm 



Figure 4. 



214 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



SOIL CLIMATES OF NORTHERN CANADA 

,00 C 300t**t 



so*. MOOTune M-CMMC CLASSES 



AOLNC REGIME 



UfcSATURATED 
REGIME 






Soil Climate 


Soil TmpraturCI43ea * 


Symbol 


Pagans and 
Sotxogtorv 


MAS1 

c 


MSST 
^C 


Numb*rof 
days>8*C 


Dgr<)ay 
>5"C 


1 


Afcrte 


<i-7 


<2 


Oto<10 





1.1 


H>9h Arctic 




<0 








1.2 


jyWAftac 




Oto<2 








14 


lowAfCfte 




OtO<2 


Oto<10 





2 
1 


Subafenc 


7 to<) 


2 tot 4 


10to <80 


010<55 


3.1 


CoMCryoboraai 




4to<8 


80 to < 180 


S&UOOO 


3.2 


Mediately Cold 

CryobOrMl 




8to<"!2 


160U><190 


160UX1000 


4 


BGTM) 


5S10<8 


i2u><ia 


1 90 to < 220 


1000 to< 1730 


it 


CoWOowal 






>190 


I000lo< 1600 


U 


Mod*iittB*yCotd 

Boraaf 






<KO 


I600UX1720 



DKription of Gasse* 



Sol not dry in any part ai 
oay in rnoet years 

Soitdryinaorrwpartowi 



EXRLAMAT)ON OF MAP SYMBOL 

Soil Ornate SubrQon 

^ 1 . Wi 



Figure 5. Soil climates of the study area (Tarnocai; 1978, Clayton et aL, 1977). 



Permafrost 

Permafrost is a thermal condition in soil or 
rock in which temperatures below 0C persist 
over at least two consecutive winters and the 
intervening summer (Harris et aL, 1988; Brown 
and Kupsch, 1974). Approximately one-half of 
Canada lies in the permafrost region (Figure 6), 
which is further divided into two zones - discon- 
tinuous in the south and continuous in the north 
(Brown 1967). 

In the southern part of the discontinuous 
zone, permafrost occurs as scattered islands a 
few square metres to several hectares in size. 
The permafrost is confined to peatlands, wet 
clayey soils, and north-facing slopes. In the 
northern part of this zone, the permafrost be- 
comes increasingly widespread and occurs in a 
great variety of terrain types. Finally, in the 
continuous zone, permafrost exists in all terrain 
types under all moisture conditions. The only 
exception is in newly deposited sediments where 
the climate has not yet imposed its influence on 
the soil thermal regime. 



Soils of the Study Area 

All areas of the Arctic Ecoclimatic Province 
are covered by Cryosolic soils (Bentley ed., 1978; 
Tarnocai, 1978). Most of these Cryosols (Agri- 
culture Canada Expert Committee on Soil Sur- 
vey, 1987), classified as Pergelic Inceptisols (Soil 
Survey Staff, 1975) in the American soil classifi- 
cation system (see Table I for the correlation 
between the two soil classifications), are 
strongly cryoturbated and are associated with 
permafrost, patterned ground, and tundra vege- 
tation. 

Cryosolic soils dominate the Subarctic Ecocli- 
matic Province, although Brunisolic soils (Incep- 
tisols, Soil Survey Staff, 1975) are also present 
in the southern part of this province. Patterned 
ground features are common, especially poly- 
gons, earth hummocks, and circles. 

The Boreal Ecoclimatic Province is dominated 
by Brunisolic and Luvisolic soils (Inceptisols, 
Boralfs and Udalfs, Soil Survey Staff, 1975). 
Organic soils (Histosols, Soil Survey Staff, 1975) 
and some Cryosols are found on peat materials. 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OP NORTHERN CANADA AND THE CORDILLERA 215 



The Cordilleran Ecocli- 
matic Provinces are domi- 
nated by Luvisols, Podzols 
(Spodosols, Soil Survey 
Staff, 1975) and Brunisols, 
with smaller amounts of 
Chernozemic (Mollisols, 
Soil Survey Staff, 1975) and 
Cryosolic soils. 

Properties of Clayey 
Soils 

Some Properties of 
Clayey Soils in the 
Yukon and Northwest 
Territories 

Most of the Northwest 
Territories lies within the 
Arctic and Subarctic Ecocli- 
matic Provinces, although a 
small portion lies within the 
North Boreal Ecoclimatic 
Province. The Yukon has a 
more complex ecoclimate, 
dominated by the North 
Interior Cordilleran Ecoclimatic Province. 
Also present in significant areas are the 
Subarctic Cordilleran and Subarctic Ecocli- 
matic Provinces. Only the extreme northern 
portion of the Yukon lies in the Arctic Ecocli- 
matic Province (Figure 1). Most of the infor- 
mation presented in this section refers to the 
clay soils occurring in the North Boreal, 
Subarctic, and Arctic Ecoclimatic Provinces. 

Morphology 

The clayey soils of the Arctic and Subarctic 
Ecoclimatic Provinces are classified as Cry- 
osols and are associated with permafrost. The 
permafrost table usually occurs within 80 cm 
of the surface. Earth hummocks, a type of 
patterned ground, provide the microrelief (Fig- 
ure 7), ranging in height from 40 to 60 cm and in 
diameter from 80 to 160 cm (Tarnocai and 
Zoltai, 1978). In the High Arctic Ecoclimatic 
Subprovince clayey soils are very often severely 
cracked and heaved (Figure 8). 

The internal morphology of these soils is 
dominated by cryogenic features such as dis- 
placed soil horizons, mixed soil materials, and 
organic smears** resulting from the movement 

""Organic smears result from movement of soil materi- 
als by cryoturbation. When organic materials are moved 
they leave a trail of thin, darker organic material smeared 
on the lighter mineral material. 



gyyj Continuous Pertnelroat Zone 

Southern Limit of Continuous 

Permafrost Zone 

DiscortlmuOuf Permafrost Zone 
Widespread Permafrost 
Southern Fringe of Permafrost Region 
Southern Limit ol Permafrost 

Patches of Permafrost Observed in Pest 
Bogs South ol Permafrost Limit 

Permafrost Areas at High Altitude in 
n South ol Ptrmelrost Limit 




Figure 6. Permafrost regions of Canada (Harris et aL, 1988; Brown, 1967). 



Table 1. Correlation between the Canadian and American soil 
classification systems/ 



Canadian Soil Classification 1 



American Soil Classification 3 



Brunisol 

Orthic Eutric Brunisol 

Orthic Dystric Brunisol 
Chrnozem 

Orthic Dark Gray Chrnozem 

Eluviated Dark Brown Chernozem 
Cryosol 

Orthic Turbic Cryosol 

Brunisolic Turbic Cryosol 

Orthic Static Cryosol 
Gleysol 

Hurnic Luvic Gleysol 
Luvisol 

Orthic Gray Luvisol 

Solonetzic Gray Luvisol 
Organic 
Podzol 
Grumic Families of various orders 



Inceptisol 

Cryochrept, Eutrochrept 

Cryochrept, Dystrochrept 
Boroll 

Boralfk Boroll, Argic Boroll 

Typic Boroll 
Pergelic subgroups 

Pergelic Ruptic Cryochrept 

Pergelic Ruptic Cryochrept 

Pergelic subgroups 
Aqu-suborders 

Aquoll, Humaquept 
Boralf&Udalf 

Cryoboralf 

Cryoboralf 
Histosol 
Spodosol 
Vertisols 



*Only those soils included in this paper are correlated. 
Agriculture Canada Expert Committee on Soil Survey, 1987. 
2 Soil Survey Staff, 1975. 



of soil materials by cryoturbation. An example 
of a cryoturbated clayey soil associated with 
earth hummocks is shown in Figure 9. 

The granular structure of the surface mineral 
horizons of these clayey Cryosolic soils is the 
result of repeated freezing and thawing cycles 
and the associated movement of soil materials. 
The granular horizons grade into massive, 
structureless subsurface horizons (Table 2) 
overlying a perennially frozen horizon in which 
ice lenses, vein ice, and pure ice layers com- 
monly occur. Organic materials occur in all 
subsurface horizons, including the frozen hori- 



216 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 






Figure 7. Earth hummocks associated with clayey soils in the Canadian 
arctic. Earth hummocks viewed (A) from the air and (B) from the ground. 
Cross sections of earth hummocks (C) with a frost crack and (D) with an 
icerich perennially-frozen soil horizon (z). 



zon. This organic material 
has been moved downward by 
cryoturbation and forms dis- 
tinct bodies, layers, and 
smears. 

In the southern part of the 
Yukon (North Boreal Ecocli- 
matic Province), the well to 
imperfectly drained clayey 
soils are Brunisols. These 
soils have B horizons with 
subangular blocky structure. 
The generally level or, in 
some cases, slightly hum- 
mocky microrelief is masked 
by a layer of forest humus (L, 
F, and H horizons). In poorly 
drained areas, the clayey soils 
generally have a peaty sur- 
face layer. Permafrost is com- 
mon in these poorly drained 
soils, as a result of the insu- 
lating properties of this peat 
layer. 

Physical Properties 

The clay contents of the B 
horizons of some clayey soils 
are presented in Table 3 and 
range from 24 to 60%. Very 
slight clay accumulation 
sometimes occurs in the B 
horizons of clayey Brunisolic 
soils, but no such accumula- 
tion occurs in any of the hori- 
zons of the clayey Cryosolic 
soils. 

The clay mineralogy for 
Soil 3 (Table 2) is given by 
Pettapiece et al. (1978) and is 
dominated by smectite, ver- 
miculite, and mica. Traces of 
chlorite and kaolinite are also 
present. 

Soil Moisture and Ice 
Content 

The near surface soil mate- 
rials of clayey Cryosolic soils 
have volumetric moisture 

contents of approximately 

30%. The moisture contents begin to increase layers it can be much higher, reaching 100% in 

rapidly close to the permafrost table. The in- pure ice layers (Figure 10). 

creases are primarily the result of water being Active Layer 

released from the ice-rich subsoil. The volumet- The active layer is that layer of the soil, lying 

ric moisture (ice) content of the frozen soil mate- above the permafrost, that is subject to annual 

rials is generally 30 to 50%, although in ice-rich thawing and refreezing. Its thickness is con- 





Figure 8. Severely frost-cracked and heaved lacustrine clayey soils in the 
Lake Hazen area, northern Ellesmere Island (Lat. 81 49*N, Long. 71 18*W). 
Photograph C shows a frost crack which is about 10 cm wide. 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OP NORTHERN CANADA AND THE CORDILLERA 217 



Site:Y67 




Figure 9. Cross section of an earth hummock devel- 
oped on clayey materials, showing the cryotur- 
bated soil horizons (Soil 5, Table 3). Location: 68 
46'N 134 03'W (N.W.T.). 



trolled by the soil texture and moisture, the sur- 
face organic cover, the vegetation cover, and the 
latitude. The relationships between the active 
layer depth and these factors are shown in Fig- 
ure 11. The data presented in this figure indi- 
cate that the thickness of the active layer in well 
to imperfectly drained clayey soils ranges from 
27 to 38 cm in the High Arctic Ecoclimatic Sub- 
province, from 35 to 56 cm in the MidArctic sub- 
province, from 34 to 62 cm in the Low Arctic 
subprovince, and from 70 to 98 cm in the 
Subarctic Ecoclimatic Province (Tarnocai, 
1978). The thickness of the active layer in 
poorly drained clayey soils and soils with 
peaty surface layers is 10 to 20 cm less 
than that of well to imperfectly drained 
clayey soils in the Arctic Ecoclimatic Prov- 
ince and 30 to 50 cm less in the Subarctic 
Ecoclimatic Province (Tarnocai, 1978). 

Frost Cracking 

Frost cracking is the fracturing of soil 
material by thermal contraction in 
subfreezing temperatures (Washburn, 
1979) (Figure 12). Ice is the critical mate- 
rial in frozen soils. Pure ice has a coeffi- 
cient of linear contraction of about 45xlQ- 6 
C-1 at -40 C. The factors affecting crack- 
ing are the salt content, the highly variable 
distribution of ice in the frozen soil, and the 
insulating effect of the snow. 

Frost cracks can develop in both sandy 
and clayey soil materials as well as in or- 
ganic materials. Fine textured, moisture- 
rich soils are probably the soils which are 
most susceptible to frost cracking. In per- 



20 



I 40 



60- 



80 



100L- 



Waler or Ice Content % on a Volume Basis 
20 40 60 80 



HI 



1 - 1 - 1 - \ - T 



100 



* 



permafrost table 



moisture content 
T Ice content 



Figure 10. Moisture and ice content of a clayey 
hummocky soil from the Canadian North* The 
permafrost table is located at the 60 cm depth. 
Location: 68* 46'N 134* 03'W (N.W.T.). 



mafrost soils, the initial frost cracks start at the 
surface and can extend to a depth of 4 m 
(Mackay 1972, 1974). 

Cracks are primarily imprinted in the perma- 
frost layer, rather than in the active layer. 
Thus, once the frost cracks are initiated, they 
recur in the same place year after year. The 



100 


- 








90 


- 








80 








x <- 


70 


- 






/ X 


s Of Active Layer 
(cm) 

8 6 8 8 


: K 


-ih 


-|~r-i'' 




| 20 










K 10 





















Figui 
in 
19' 
ba 
br 


High Arctic 


Mid-Arctic 


North South 
Low Arctic 


Subarctic 


Increasing Vegetation Cover And Increasing Thickness Of Surface Organic Layer * 

'e 11. Thickness of the active layer in various clayey soils 
the Arctic and Subarctic Ecoclimatic Provinces (Tarnocai 
78), The range of active layer thicknesses is indicated by a 
r and the mean active layer thicknesses are indicated by a 
oken line (- - -). 



218 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Table 2. Morphological description of some northern clayey soils. 


Horizon Depth Colour Texture" Structure* 


Special Features 


(cm) (moist) 




1. Orthic Eutric Brunisol 60 50'N, 135 20W 




Ah 3-0 10YR2/1 




Bal 0-13 10YR4/3 SiCL sbk 




Ba2 13-28 10YR5/4 SiCL sbk 




Ba3 28-39 10YR 5/3 SiC sbk 




Ck 39+ 7.5Y6/2 SiC pr 




2. Orthic Eutric Brunisol 60 40'N, 119 04W 




L, H 8-0 10YR3/2 




Bm 0-14 7.5YR5/4 C sbk 




BC 14-55 10YR6/3 SiC sbk 




C 55-100 10YR7/1 SiC sbk 




3. Orthic Turbic Cryosol 63 37'N, 123 39W 




OH 24-9 7.5YR 3/2.5 




O2 9-4 10YR 2/1.5 




Om 4-0 SYR 2.5/1.5 




Bmy 0-22 5YR 3.5/1 SiC gr 




BCy 22-47 10YR3/1 SiC mss 


strongly cryoturbated 


Ahyz 47-59 10YR3/1 SiC mss 


strongly cryoturbated 


Cyz 59+ 10YR4/1 SiC mss 


strongly cryoturbated 


4. Orthic Turbic Cryosol 68 46'N, 134 03W 




L, H 1-0 10YR3/2 




Bmy 0-20 10YR 5/4 C gr 


strongly cryoturbated 


BCgy 20-60 10YR 5/2 C gr 


strongly cryoturbated 


Cyzl 60-88 5Y5/1 CL mss 


strongly cryoturbated 


Ahyz 88-100 10YR 3/1 CL 


strongly cryoturbated 


5. Brunisolic Turbic Cryosol 68 23'N, 133 44*W 




Om 10-0 5YR 2.5/1 




Bm 0-20 10YR4/3 C gr 




BCyl 20-70 10YR 4/3 C mss 


strongly cryoturbated 


BCy2 70-100 10YR 4/2 C mss 


strongly cryoturbated 


Ohy 100-110 SYR 3/2 


strongly cryoturbated 


Cg 110-130 10YR4/1 C mss 




Cz 130+ 10YR4/1 C mss 




6. Orthic Turbic Cryosol 69 35'N, 139 05 W 




Ahky 0-10 10YR 2/2 SiC sbk 




Bmky 0-17 5Y5/2 SiC gr 


strongly cryoturbated 


Ckyl 0-28 10YR3/2 SiC sbk 


strongly cryoturbated 


Cky2 0-40 10YR 3/2 SiC pr 


strongly cryoturbated 


Ckz - 5Y5/2 SiC 




"Texture: C-clay; CL-clay loam; SiC-silty clay; SiCL-silty clay loam 


^Structure: sbk-subangular blocky; gr-granular; mss-massive; 


pr-prismatic 



crack can occur as a single crack more or less lin- 
ear in form, but most commonly, the cracks oc- 
cur in a polygonal pattern (Figure 12C). The 
cracks are usually filled by snow and melt water 
or, in dry high arctic areas, by sand. As a result, 
frost cracks in permafrost soils are commonly 
occupied by either ice wedges (Figure 12D) or 
sand wedges. In the arctic regions, however, ice 
wedges are very common and sand wedges are 
relatively rare. 

Cryoturbation 

Most northern Canadian soils are subject to 
cryoturbation, a process which affects the entire 
soil system. As a result of cryoturbation, the soil 
surface is unstable and, internally, soil materi- 
als are mixed together and soil horizons dis- 
rupted and displaced. These properties are as- 
sociated with Turbic Cryosols and with those 
Brunisols and Gleysols which occur in the north- 
ern part of the North Boreal Ecoclimatic Prov- 
ince or above the timberline in some mountain- 
ous areas. 



ods 



Chemical Properties 

The sola of clayey Cryosolic soils are 
acid to mildly alkaline, ranging in pH 
between 3.8 and 7.9 (Table 3). The sola 
of clayey Brunisolic soils are slightly acid 
to neutral (Soils 1 and 2, Table 3) with 
pH values somewhat lower than those of 
the parent material as a result of leach- 
ing. 

The organic carbon content of the 
clayey Brunisolic soils decreases with 
depth. The clayey Cryosolic soils, on the 
other hand, have high levels of organic 
carbon in all mineral horizons as a result 
of cryoturbation. In the Cryosolic soils, 
an organic-rich mineral horizon com- 
monly occurs in the lower part of the pe- 
don, in the vicinity of the permafrost 
table. 

The cation exchange capacity ranges 
between 10 and 20 meq/lOOg, with cal- 
cium and magnesium being the domi- 
nant exchangeable cations (Table 3). 
Although the electrical conductivity is 
generally low (0.1 to 0.3 mmhos/cm), in 
some cases, the parent material has a 
higher conductivity (Cz horizon of Soil 6, 
Table 3). In the Arctic Ecoclimatic Prov- 
ince, clayey soils very often develop a salt 
crust on the soil surface during the 
warmer summer days, especially in peri- 
with no precipitation when moisture is 



drawn to the surface where it evaporates. 
Present Use 

Almost all of the clayey soils occurring in the 
Yukon and Northwest Territories are under 
native forest or tundra vegetation. The main 
problems associated with the use of these soils 
result from the high amounts of ice in the frozen 
subsoil. Removal of the vegetation and organic 
surface horizons causes the soil temperature to 
rise and the ice in the subsoil to melt. This re- 
sults in pronounced subsidence and an uneven 
topography. 

When these soils occur on sloping topography, 
they are subject to solifluction and erosion if the 
thermal balance is disturbed. Clayey soils with 
level topography are very often subject to ther- 
mokarst. This thermokarst process results from 
the melting of ice-rich material and leads to 
subsequent thaw settlement. All activities in- 
volving these high ice content clayey soils 
should be carried out in such a way that melting 
of the ice is minimized. 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OF NORTHERN CANADA AND THE CORDILLERA 219 




Figure 12. Frost cracks at the permafrost ex- 
perimental site on Richards Island in the 
Mackenzie River Delta area (Lat. 69 29'N, 
Long. 134 35'W). These frost cracks, which 

are seven years old and about 15 to 30 cm wide, have developed in drained lake 
sediments (A and B). In these two photographs the vertical stakes with the 
horizontal rods are used to monitor the widening of the frost crack. Ice wedge 
polygons with wide (1 to 3 m) frost cracks (trenches), some filled with water 
(C). The ice wedges are beneath the trench. An ice wedge developed in a poly- 
gon trench (D). Photographs C and D were taken in the Mackenzie River 
Delta area. 



Some Properties of Clayey Soils in the 
Northern Portions of Ontario, Manitoba, 
and Saskatchewan 

Most of this area lies within the North Boreal 
Ecoclimatic Province. Only the extreme north- 
ern part of the area lies within the Subarctic 
Ecoclimatic Province (Figure 1). The informa- 
tion presented in this section refers only to the 
North Boreal Ecoclimatic Province, since this is 
where most of the clayey soils occur. 

Morphology 

Gray Luvisols have developed on clayey mate- 
rials throughout the North Boreal Ecoclimatic 
Province (Figure 2), under the influence of forest 
vegetation. Their main characteristics are con- 
tinuous light-coloured eluvial horizons and illu- 
vial textural B horizons. These developed 
through leaching of the soluble decomposition 
products of forest litter and consequent down- 
ward translocation and immobilization of clays 
with other associated colloidal materials. 

Increasingly severe climate towards the 
northern edge of the North Boreal Ecoclimatic 



Province (Figure 2) results 
in the persistence of perma- 
frost below the 1 m control 
section in some Luvisols 
and the widespread occur- 
rence of permafrost within 
1 m in Static Cryosols and 2 
m in Turbic Cryosols. The 
Cryosols are usually associ- 
ated with varying degrees 
of cryoturbated horizons 
and patterned ground fea- 
tures on the soil surface. 

Other clayey soils in 
these northern landscapes 
include poorly drained 
Gleysolic soils with thin 
surface veneers of peat. 
Very poorly drained lower 
slopes and depressions are 
characterized by Organic 
soils in which peat accumu- 
lation ranges from 40 cm to 
in excess of 1.6 m. These 
Organic soils are underlain 
by clay sediments. 

The most southerly oc- 
currence of permafrost is in 
organic materials; on pro- 
gressing northward, perma- 
frost occurs next in poorly 
drained clay soils and eventually is encountered 
in well drained clays near the northern bound- 
ary of the North Boreal Ecoclimatic Province. 
The occurrence of permafrost affects soil mor- 
phology in terms of cryoturbation, soil moisture 
regime, ice content and active layer depth. 

Variation in degree of cryoturbation in well to 
imperfectly drained clay soils is summarized in 
Table 4. Soil 1 has developed continuous undis- 
turbed horizons. Soil 2 is characterized by low 
microhummocky relief with continuous horizons 
(Figure 13a). The hummocky surface of this soil 
is a relict feature from a climatic period during 
which cryoturbation and earth hummock forma- 
tion occurred. Soil 3 occurs in the Subarctic 
Ecoclimatic Province and is affected by perma- 
frost and more pronounced cryoturbation (Fig- 
ure 13b). This Cryosolic soil is weakly devel- 
oped with irregular, broken and convoluted hori- 
zons and strongly mounded microrelief. 

Physical and Chemical Properties 

Clayey parent materials in northern Ontario, 
Manitoba, and Saskatchewan are generally 
moderately to strongly calcareous, although a 



few areas of neutral, noncal- 
careous clay occur. The min- 
eralogy of these materials is 
mixed; high shrink-swell clays 
such as smectite originate from 
the shale bedrock of the Interior 
Plains Region to the south and 
west. Dilution of smectite with 
illitic clay derived from Precam- 
brian crystalline rock occurs in 
many of the clayey sediments in 
the Canadian Shield. Physical 
and chemical properties of three 
clay soils from northern Mani- 
toba are presented in Table 5. 

All soils have been leached to 
varying degrees (removal of sol- 
uble salts and CaCO 3 and re- 
duced pH in the sola). The well 
developed Luvisols (Soils 1 and 
2, Table 5) have deeper sola and 
distinct horizonation resulting 
from translocation of clay. The 
clayey Cryosol (Soil 3, Table 5) 
is leached to a shallow depth, 
with weakly developed horizons 
and no pronounced accumula- 
tion of translocated clay. Devel- 
opment of this soil is impeded by 
the permafrost table and the un- 
stable soil surface resulting 
from cryoturbation. 

Soil Moisture and Ice Content 

Clayey Luvisolic soils occur under moisture 
regimes in which annual precipitation and po- 
tential evapotranspiration result in only short 
periods of soil moisture deficit (Figure 4, 
Wabowden). The moisture regime of clayey 
Cryosolic soils is influenced by the presence of 
permafrost and the slow release of moisture 
from the ice-rich subsoil as it thaws during the 
growing season. As a result, Cryosols may be- 
come dry to a very shallow depth for only short 
periods. 

Active Layer 

The depth of the active layer in well and im- 
perfectly drained clayey Luvisols containing 
permafrost varies from 0.7 to 2 m in northern 
portions of the North Boreal Ecoclimatic Prov- 
ince. Although these soils do not contain perma- 
frost in the control section in all years, they are 
the coldest Luvisols. In the Subarctic Ecocli- 
matic Province, the active layer in well and 
imperfectly drained clayey soils decreases to 



Table 3. 


Some selected physical and chemical characteristics of some northern 
clayey soils. 


Horizon 


Depth Total 
(cm) sand 


Silt 


Clay 


pH 


Cond. C 
(m mhos/cm) 


CEC 


Exchangeable Cations 
(meq) meq/100 g 
Ca Ms K Na 


Ah 


1. Orthic Eutric Brunisol 6050'N 


, 135 20^ 














f\ii 
Bml 


0-13 10.0 


56.7 


33.2 


5.4 


0.1 


2.3 





13.9 


12.5 


0.8 


0,1 


Bm2 


13-28 2.7 


63.6 


33.7 


6.2 


0.1 


1.1 





12.1 


10.2 


0.7 


0.2 


Bm3 


28-39 1.2 


58.0 


40.7 


6.4 


0.1 


1.0 





14.0 


12.7 


0.5 


0.3 


Ck 


39+ 4.9 


49.8 


45.3 





0.1 


2.1 





31.3 


12.5 


0.1 


0.5 


2. Orthic Eutric Brunisol 60 40'N, 11904W 


L,H 


8-0 








6.3 


0.4 


52.3 


88 


71.9 


7.6 


0.6 


0.2 


Bm 


0-14 24.6 


34.8 


40.6 


6.9 


0.3 


1.9 


24 


20.7 


2.6 


0.4 


JTr* 


BC 


14-55 4.1 


45.9 


50.0 


7.7 


0.2 


1.3 


19 














C 


55-100 6.2 


65.4 


28.3 


7.8 


0.2 





10 














3. Orthic Turbic Cryosol 6337*N, 


123*39^ 














on 


24-9 








3.7 





41.9 





20.9 


9.0 


1.4 





02 


9-4 








5.9 





37.2 





88.6 


25.5 


0.7 





Om 


4-0 








5.9 





33.4 





96.8 


26.6 


0.2 





Bmy 


0-22 9.0 


49.7 


41.3 


6.7 


0.10 


2.3 





17.8 


7.6 


0.2 





BCy 


22-47 7.3 


52.7 


39.9 


6.7 


0.10 


8.4 





42.8 


12.3 


0.2 





Ahyz 


47-59 9.5 


56.0 


43.5 


6.7 


0.17 


10.4 





46.6 


11.5 


0.3 





Cyz 


59-*- 8.4 


54.5 


37.2 


6.8 


0.15 


8.5 





43.6 


11.3 


0.5 





4. Orthic Turbic Cryosol 68 46'N, 134 OSW 


L,H 


1-0 








4.5 





44.2 





12.7 


8.9 


4.1 


0.3 


Bmy 


0-20 18.6 


36.2 


45.2 


3.8 


0.10 


2.5 


22 


0.4 


1.9 


0.3 


Tr* 


BCgy 


20-60 19.3 


36.4 


44.3 


4.0 


0.05 


1.5 


20 


1.3 


2.1 


0.2 


0.1 


Cyzl 


60-88 27.0 


37.7 


35.2 


4.4 


0.28 


4.7 


22 


3.5 


3.8 


0.6 


0.1 


Ahyz 


88-100 23.4 


39.9 


36.8 


4.4 


0.32 


12.9 


33 


6.2 


5.3 


0.5 


0.1 


Cyz2 


100-125 22.2 


40.0 


37.8 


4.3 


0.50 





21 


7.5 


3.3 


0.7 


0.4 


Om 


5. Brunisolic Turbic Cryosol 
10 


68 23*N, 133 44W 


Bm 


0-20 3.0 


49.0 


48.0 


3.8 





2.0 


27 


0.1 


0.2 


0.3 


0.1 


BCyl 


20-70 2.0 


45.0 


53.0 


3.7 





1.2 


25 


0.2 


0.3 


0.5 


0.1 


BCy2 


70-100 3.0 


47.0 


50.0 


3.7 





2.3 


27 


0.6 


0.5 


0.6 


0.1 


Ohy 


100-110 








4.1 





17.4 


78 


2.4 


1.3 


0.5 


0.2 


Cg 


110-130 2.0 

1Ort , 


48.0 


50.0 


3.7 





2.7 


25 


6.8 


3.8 


0.7 


1.2 


6. Orthic Turbic Cryosol 69 35'N, 


139 OSW 














Ahky 


0-10 5.4 


43.3 


51.3 


7.4 


0.50 


4.5 

















Bmky 


0-17 2.9 


37.2 


59.9 


7.6 


0.30 


2.2 

















Ckyl 


0-28 4.0 


49.2 


46.8 


7.6 


0.20 


2.6 

















Cky2 


0-40 3.1 


48.8 


48.2 


7.5 


0.20 


2.6 

















Ckz 


3.8 


51.2 


45.0 


7.5 


0.20 


2.9 


_ 














*Tr - trace 

























about 0.5 to 1.2 m, depending on organic surface 
material thickness and vegetative cover. Some 
part of the control section remains frozen 
throughout the year. 

Cryoturbation 

Cryoturbation is evident in Luvisols in which 
permafrost occurs only periodically within the 
control section and also in Luvisols no longer 
affected by permafrost. Subdued earth hum- 
mocks occur on the surface of these cold Luvi- 
sols, but horizon development is continuous and 
not disrupted or convoluted (Figure 13a). Such 
patterned ground is a relict feature from cli- 
matic periods when cryoturbation was more ac- 
tive. Clayey Cryosols characterized by perma- 
frost within the control section have shallow 
active layers and a strongly mounded mi- 
crohummocky surface pattern. Cryoturbation is 
sufficiently active to disturb and disrupt soil 
horizons, cause unstable soil surfaces, and per- 
mit only relatively weak horizon development 
(Figure 13b). 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OP NORTHERN CANADA AND THE CORDILLERA 221 



Vertic Properties 


Table 4. Morphological description of some northern clayey soils. 


tent and mineralogy that have the po- 


Horizon Depth Colour Texture" Structure* Special Features 
(cm) (moist) 


tential to develop vertic properties. 


1. Solonetzic Gray Luvisol 55 32'N, 98 OSW 


Vertic features have been observed in 


L,F, H 6-0 5YR 2.5/2 
Ae 0-7 10YR4/2 C gr 


well and imperfectly drained Luvisolic 


AB 7-10 10YR5/3 C sbk vertical cracking 


clay soils, but the frequency and extent 


Btnj 10-17 10YR 4/2.5 HC sbk vertical cracking 
Bt 17-30 10YR4/2 HC gr 


of their occurrence has not been the 


BC 30-36 10YR5/2 HC gr 




Ckgj 36-100 10YR 4.5/3 HC mss 


subject of any quantitative study. 


2. Solonetzic Gray Luvisol 55 55'N, 97 42*W 


Cracks that are open to the surface sel- 


L,H 2-0 10YR2/1 


dom extend below the upper solum of 


Ae 0-6 10YR5/2 SiC gr 
AB 6-16 10YR4/3 C pi vertical cracking 


Luvisolic soils in the North Boreal 


Btnj 16-34 10YR4/2 SiC abk vertical cracking 

P>*. <*A Kf\ inVP A/ 1 ^. Qip aMr 


Ecoclimatic Province. 


DC UTk-OV/ lUJJvTt/O Olw SDK 

Ckl 50-98 10YR4/4 C sbk 


Slickensides have been observed in 


Ck2 98-138 10YR4/4 HC gr 


southern Luvisols in the study area, but 


3. Orthic Turbic Cryosol 57 50'N, 99 36' 
L,F 6-0 10YR5/4 


appear to be uncommon in more north- 
erly clayey soils, particularly those af- 


Bmyl 0-5 10YR5/2 SiCL gr strongly cryoturbated 
Bmy2 5-12 2.5Y4/2 SiCL sbk strongly cryoturbated 
Cgjyl 12-30 2.5Y 3.5/2 SiCL mss strongly cryoturbated 


fected by permafrost and cryoturbation 


Cgjy2 30-60 2.5Y 3.5/2 SiCL mss strongly cryoturbated 
Cgjy3 60-90 2.5Y 3.5/2 SiC mss strongly cryoturbated 


(Mills et al., these proceedings). This 


Cz 90-98 2.5Y 3.5/2 SiCL mss 


can be attributed to shorter periods 


4. Gleyed Gray Luvisol 48 32'N, 81 04' 
Ap 0-13 10YR2/1 SiCL sbk 


when the soil becomes sufficiently dry 


AB 13-18 10YR5/2 SiCL sbk 


to develop desiccation cracks and con- 


Btgj 18-40 10YR4/3 C sbk 
Ckg 40+ 10YR6/3 SiCL sbk 


centration of turbation effects in the 


Texture: C-clay; HC-heavy clay; SiC-silty clay; SiCL-silty clay loam 


active layer above the permafrost table. 


+Structure: pl-platy; abk-angular blocky; sbk-subangular blocky; gr-granular; mss- 
massive 


Present Use 






The vast majority of northern 


Table 5. Some selected physical and chemical characteristics of some northern 


clayey soils remain in their na- 


Manitoba and Ontario clayey soils. 


tive state. Better drained clayey 


Horizon Depth Total Silt Clay pH Cond. C CEC Exchangeable Cations 


soils in the North Boreal Ecocli- 


(cm) sand (%) (%) (mmhos/cm) (%) (meq) meg/100 g- 
Ca Mg K Na 


matic Province support commer- 


1. Solonetzic Gray Luvisol 55 32'N, 98 03W 


cial stands of coniferous and 


L, F,H 6-0 _____ 5.1 47.6 103 40.0 14.4 1.6 0.1 
Ae 0-7 5 39 56 5.5 5.7 46 22.5 10.7 0.8 0.1 


mixed- wood forests. The extent 


AB 7-10 3 31 66 5.7 2.0 33 17.1 9.2 0.8 0.1 


of commercial forests, however, 


Btnj 10-17 1 13 86 6.1 1.4 34 18.3 10.3 0.9 0.2 
Bt 17-30 1 6 93 6.5 0.7 34 20.7 10.5 0.9 0.2 


decreases rapidly with increas- 


BC 30-36 1 5 94 7.4 1.1 35 25.8 9.9 0.8 0.2 
Ckgj 36-100 1 19 80 7.6 11.1 27 


ing latitude. Forest and other 


2. Solonetzic Gray Luvisol 55 55'N, 97 42W 


vegetation also have great value 


L, H 2-0 5.9 32.4 87 50.5 8.3 1.7 0.4 


as wildlife habitat. Hunting and 


Ae 0-6 3 39 58 5.5 4.0 38 16.5 4.4 0.6 0.1 
AB 6-16 1 38 61 5.6 2.5 31 15.8 4.3 0.6 0.1 


trapping are of both commercial 


Btnj 16-34 1 46 53 5.9 1.0 35 18.9 7.2 1.0 0.2 
Bt 34-50 54 46 7.0 0.5 34 19.5 6.5 1.1 0.2 


and social value to many native 


Ckl 50-98 40 60 7.6 8.3 31 18.8 6.1 0.8 0.2 


people. 


Ck2 98-138 31 69 7.7 7.7 31 30.8 5.9 1.0 0.2 


IT Jr 

Small areas of clayey soils 


3. Orthic Turbic Cryosol 57 50'N, 99 36W 
L,F 6-0 ______ 3.5 36.6 67 10.2 6.2 1.8 0.6 


have been cleared and broken for 


Bmyl 0-5 6 64 30 4.2 6.2 29 3.7 4.8 0.5 0.3 
Bmy2 5-12 7 61 32 4.8 3.8 20 8.6 5.0 0.5 0.2 


agriculture. Cereal crops, for- 


Cgjyl 12-30 8 61 31 5.7 1.6 15 10.5 5.4 0.4 0.2 


age, and potatoes have been 


Cgjy2 30-60 7 59 34 5.8 16 12.4 5.5 0.4 0.2 
Cgjy3 60-90 6 53 41 6.5 17 13.5 5.6 0.5 0.2 


grown with varying success due 


Cz 90-98 75 38 6.6 19 13.5 5.5 0.5 0.2 


to climatic conditions. Potential 


4. Gleyed Gray Luvisol 48 32'N, 81 04' 
Ap 0-13 8 44 48 7.0 3.0 


for vegetable production has 


AB 13-18 7 50 43 7.1 0.9 ________ 


been proven by the operation of 


Btgj 18-40 2 27 72 7.7 0.7 _ _ _ _ _ 
Ckg 40+ 1 44 55 7.9 ______ 


some market gardens developed 




. _ 1 A. 



by native communities and individuals for pri- Where permafrost exists, 
vate use. Townsite development and associated problems with roadbed subsidence and shifting 
infrastructure such as road construction have of building foundations can be expected if appro- 
occurred on northern clayey soils, as well. If no priate design measures are not taken, 
permafrost is present, problems encountered 
are similar to those experienced in the south. 



222 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Some Properties of Clayey Soils in British 
Columbia and Northern Alberta 

Almost all of British Columbia lies within the 
Cordilleran Ecoclimatic Provinces, which in- 
clude the Pacific Cordilleran (referred to as the 
"coast"), the South Interior Cordilleran (referred 
to as the "southern interior"), and the North In- 
terior Cordilleran (referred to as the "northern 
interior") Ecoclimatic Provinces (Figure 1). The 
extreme northeastern corner of British Colum- 
bia, together with northern Alberta, lies within 
the North Boreal Ecoclimatic Province. For 
more information on the North Boreal Ecocli- 
matic Province, see the preceeding section. 

Morphology 

The upper solum of well drained coastal 
clayey soils (Soil 1, Table 6) generally has 
subangular blocky structure, while that of the 
poorly drained coastal clayey soils has a granu- 
lar structure. The well drained clayey soils in 
the southern interior have granular to subangu- 
lar blocky surface soil structures. The clayey 
Luvisolic soils in the northern interior (Soil 6, 
Table 6), however, have platy surface soil struc- 
tures. The B horizons of all of these clayey soils 
have prismatic structures which break to suban- 
gular or angular blocky. The structures of all of 
these B horizons are very pronounced and well 
developed. 

Cracks approximately 1 to 2 cm wide and ex- 
tending to a depth of 1 m or more are common in 
the clayey soils of the southern interior (Soils 3, 
4 and 5, Table 6). During very dry summers, the 
coastal clayey soils (Soils I and 2, Table 6) also 
develop similar cracks. These cracks, however, 
are less visible because of the friable surface 
materials. Cracking occurs less frequently in 
the northern interior clayey soils (Soil 6, Table 
6) than in the soils of the southern interior. This 
lack of cracking is probably due to the organic 
surface horizons (forest litter) and the usually 
higher moisture levels. Slickensides have been 
noted in the interior clayey soils, but none have 
been reported on the coast. 

Physical Properties 

The texture of these soils is silty clay to clay, 
with the highest clay content usually occurring 
in the B horizon (Table 7). The upper horizons 
of these soils are usually coarser textured be- 
cause of eluviation. The low permeability re- 
sults in periodically perched water tables, espe- 
cially during the winter months in the coastal 
clayey soils. Perched water tables also occur in 
the northern interior clayey soils because of sea- 



20 20 40 60 . 80 100 120 140 160 180 200 cm 





20 
40 
60 
80 

oo- 

20 



T 



T" 



T" 



~T 



T" 



nr 




20 


20 
40 

60 

80 

100 



20 40 60 80 100 120 140 160 160 200 cm 
I | | | [- I \ I | 1 




Figure 13. a. Cross section through subdued mi- 
crohummocks and continuous soil horizons. 
Soil 2, Table 4. Solonetzic Gray Luvisol, 63P22. 
Location: Manitoba 55 55'N, 97o 42W Jb. Cross 
section of an earth hummock with strongly 
mounded microrelief and cryoturbated hori- 
zons. Soil 3, Table 4. Orthic Turbic Cryosol, 
64G1. Location: Manitoba 57 50'N, 99 36'W. 



sonal frost in the subsoil. This frost remains 
well into June, especially in sites having a dense 
moss cover. 

The clay mineralogy of the coastal clayey soils 
is dominated by vermiculite and chlorite, in both 
the coarse and fine clay fractions. Smaller 
quantities of smectite and mica also occur. The 
interior clayey soils are dominated by smectite 
and vermiculite. 

Chemical Properties 

Both the coastal and northern interior clayey 
soils have strongly to very strongly acid surface 
mineral horizons and slightly acid to mildly al- 
kaline parent materials (Table 7). The southern 
interior clayey soils, on the other hand, are 
slightly acid to neutral, with mildly to moder- 
ately alkaline parent materials. 



TAJRNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OF NORTHERN CANADA AND THE CORDILLERA 223 



Horizon 



Depth 

(cm) 



0-6 

6-18 

18-25 

25-31 

31-50 

50-75 

75-96 

96-127 

127+ 



0-15 

15-25 

25-32 

32-48 

48-70 

70-92 

92-120 

120+ 



0-21 

21-30 

30-60 

60-93 

93-115 

115-135 



0-14 

14-20 

20-35 

35-49 

49-82 



The organic carbon content of 
the surface mineral horizons of 
well to imperfectly drained clayey 
soils is about 2%, with coastal 
clayey soils having slightly higher 
contents than those in the inte- 
rior. The poorly drained clayey 
soils (Soil 2) generally have higher 
carbon contents than do the im- 
perfectly and well drained soils 
(Table 7). 

The cation exchange capacity 
varies from 11 to 45 meq/lOOg. 
These values are largely depend- 
ent on the soil horizon, clay miner- 
alogy, and organic matter con- 
tent. 

Present Use 

The clayey soils in British Co- 
lumbia and northern Alberta are 
used for agriculture and forestry. 
Along the coast, most of these 
soils are used for pasture and hay 
production and for forestry, al- 
though in the lower Fraser River 
Valley they also are used for grow- 
ing cereal crops. In the southern 
interior almost all clayey soils are 
used for agriculture and are 
among the best agricultural soils 
for growing hay, cereal grains 
and, where the climate is suitable, 
tree fruits. In the northern inte- 
rior, these soils are used for forage 
production, pasture, and forestry. 

The main limiting factors for 
use are the shallow rooting depth, 
the slow warming in spring, the 
maintenance of the surface or- 
ganic matter content in the northern interior, 
and the poor trafficability because of the periodi- 
cally perched water tables. There are also prob- 
lems in maintaining the soil structure and limit- 
ing compaction. 

Classification of Tin-bated Soils 

Both cryoturbation and shrink-swell proc- 
esses in soil result in physical disturbance af- 
fecting the entire soil system. Soils character- 
ized by such turbation processes occur under 
environments extending from the arctic to the 
tropics. 

In the Canadian soil classification, soils asso- 
ciated with permafrost are classified as Cryosols 



Table 6. Morphological description of some British Columbia clayey soils. 



Colour 
(moist) 



Texture" Structure* Special Features 



L 

F,H 

Bmcc 

Bml 

Bm2 

Bm3 

Bm4 

BC 

Cl 

C2 

IIC 

Ap 

Aeg 

AB 

Btgl 

Btg2 

BC 

Cgl 

Cg2 

Apl 

AB 

Bt 

BCcal 

BCca2 

Ck 

Ap 
Ae 
Bt 
BC 
Ck 

Ap 

Ac 

Btl 

Bt2 

BCca 

Ck 

L,F,H 

Ae 

AB 

Btl 

Bt2 

BC 

C 



1. Orthic Dystric Brunisol 49 17TC, 124 54W 

8-3 

3-0 

L gr 

L gr 

SiC gr 

SiC gr 



SYR 5/6 

7.5YR 4/3 

10YR4/3 

2.5Y4/4 

2.5Y4/4 

2.5Y5/2 

2.5Y5/2 

2.5Y4/4 

2.5Y4/2 



iron-manganese concretions 



SiL 

L 

L 

L 

SL 



sbk 
Pi 



2. Humic Luvic Gleysol 49 06*N, 122 37*W 



10YR2/2 

2.5Y5/2 

2.5Y4/2 

2.5Y4/2 

5Y4/3 

5Y4/1 

5Y4/1 

5Y4/1 



SiC 

SiL 

SiCL 

SiC 

SiC 

SiC 



sbk 



pr 
pr 
mss 
mss 



3. Orthic Dark Grey Chernozem 5025'N, 119 IIW 



10YR4/2 
10YR5/2 
10YR 4/3 
10YR 5/4 
2.5Y4/2 
2.5Y4/2 



C 

HC 
HC 
HC 
HC 
HC 



pr 
abk 
abk 
mss 



4. Eluviated Dark Brown Chernozem 49 49'N, 119 37W 



10YR5/2 

10YR5/2 

10YR5/2 

2.5Y5/2 

5Y7/2 



SiC 
SiC 
SiL 
SiC 
HC 



abk 

Pi 
pr 
pr 
Pi 



5. Orthic Gray Luvisol 



0-9 

9-24 

24-39 

39-64 

64-90 

108-125 



10YR3/2 
10YR5/2 
10YR4/2 
10YR5/2 
10YR6/2 
10YK4/2 



6. Orthic Gray Luvisol 



5-0 

0-5 

5-13 

13-25 

25-51 

51-62 

62-110 



10YR5/2 
10YR5/2 
10YR 3/3 
10YR 3/3 
10YR 3/3 
10YR 3/3 
10YR 3/3 



5026'N, 119"irW 
SiL sbk 

SiCL abk 

C pr 

C pr 

CL abk 

CL pi 

54 23'N, 124 16 W 



C 

HC 
HC 
HC 
HC 
HC 



Pi 

abk 

col 

pr 
pr 



"Texture: C-clay; HC-heavy clay; SiC-silty clay; SiCL-silty clay loam; SL-sandy loam; CL-clay 
loam; L-loam 

+Structure: pl-platy; abk-angular blocky; sbk-subangular blocky; gr-granular; mss-massive; pr- 
prismatic; col-columnar; sg-single grain 



(Agriculture Canada Expert Committee on Soil 
Survey, 1987) while in the American soil taxon- 
omy such soils are recognized only at the sub- 
group level in the various soil orders. All tex- 
tural classes of these permafrost soils commonly 
are affected by cryoturbation processes result- 
ing in churning of soil material and development 
of various kinds of patterned ground. Additional 
cryogenic forces are responsible for the forma- 
tion of cracking in these soils. The morphology 
and properties of clayey soils associated with 
permafrost are provided in the foregoing de- 
scription of the Arctic, Subarctic, and North 
Boreal Ecoclimatic Provinces. 

In the American soil taxonomy, strongly tur- 
bated soils south of the permafrost region are 



224 



SIXTH INTERNATIONA! S oil C OSSIFICATION WORKSHOP 



classified as Vertisols (Soil 
Survey Staff, 1975). These 
soils have developed in clayey 
materials with high shrink- 
swell potential, resulting in 
both the movement of soil 
materials and cracking. The 
volume change required for 
vertic soil properties to de- 
velop is greatest under envi- 
ronments which permit great- 
est oscillation hetween wet 
and dry soil moisture content 
(Wilding and Tessier, 1988). 

Soils affected by shrink- 
swell processes are recognized 
in Soil Taxonomy (Soil Survey 
Staff, 1975) at the order level 
as Vertisols. In contrast, simi- 
lar degrees of physical distur- 
bance resulting from cryo- 
genic forces in cold soils are 
recognized at the subgroup 
level in various soil orders. 
Similarity in degree of physi- 
cal disturbance and soil crack- 
ing suggests that the logic of 
this ordering of soil properties 
should be reconsidered in any 
future revision of Soil Taxon- 
omy. Establishment of an or- 
der of soils (equivalent to the 
Canadian Cryosolic Order) to 
recognize the morphology re- 
sulting from cryogenic forces 
associated with permafrost 
would correspond to recogni- 
tion of shrink-swell soils oc- 
curring in southern regions in 
the Vertisol Order. Dr. Guy Smith (1986, p. 61) 
suggested that establishment of such an order 
should be considered by a small international 
committee. 

Summary 

Representative clayey soils from the Arctic, 
Subarctic, North Boreal, and Cordilleran Ecocli- 
matic Provinces were evaluated in order to de- 
termine their properties and classification. 

All clayey soils in the Arctic and Subarctic 
Ecoclimatic Provinces are associated with per- 
mafrost and earth hummocks and are classified 
as Cryosols. Their morphology is dominated by 
cryoturbated features and the occurrence of per- 



Table 7. Physical and chemical characteristics of some British. Columbia clayey 
soils. 


Horizon 


Depth 
(cm) 


Total 
sand 


Silt 


Clay pH 


Cond. C 
(m mhos/cm) 


CEC 


Exchangeable Cations 
(meq) meq/100 g 
Ca Mg K Na 


L 


1. Orthic Dystric Brunisol 49 17'N, 124 54*W 
8-3 6.0 


39.4 


96 


78.1 


7.7 


2.0 


0.3 


F,H 


3-0 








5.0 





40.3 


101 


59.0 


16.3 


1.4 


0.3 


Bmcc 


0-6 


31.1 


41.3 


27.6 4.7 





3.3 


27 


1.8 


0.4 


0.1 


0.1 


Bml 


6-18 


24.4 


47.8 


27.7 4.7 





2.4 


26 


1.7 


0.4 


0.1 


0.1 


Bm2 


18-25 


15.7 


49.3 


35.0 4.8 





2.3 


27 


4.6 


0.6 


0.1 


0.1 


Bm3 


25-31 


10.5 


51.9 


37.6 4.9 





1.5 


25 


7.8 


0.9 


0.1 


0.1 


Bm4 


31-50 


11.7 


48.2 


40.0 5.2 





0.9 


33 


17.6 


1.8 


0.1 


0.1 


BC 


50-75 


40.5 


38.8 


20.7 5.8 





0.3 


34 


22.2 


1.7 


0.1 


0.1 


Cl 


75-96 


46.6 


35.6 


17.8 6.4 





0.3 


32 


22.4 


1.5 


0.1 


0.1 


C2 


96-127 


37.0 


44.4 


18.6 6.1 





0.3 


29 


19.2 


1.2 


0.1 


0.1 


IIC 


127+ 


65.5 


32.4 


2.0 6.9 





0.1 


4 


2.0 


0.1 


0.0 


0.0 


2. Humic Luvic Gleysol 


4906'N f 122 37*W 


Ap 


0-15 


2.0 


64.0 


34.0 5.6 





6.7 


36 


5.6 


2.6 


0.3 


0.5 


Aeg 


15-25 


2.0 


70.0 


28.0 5.6 





0.7 


18 


4.3 


4.6 


0.1 


0.3 


AB 


25-32 


5.0 


56.0 


38.0 5.9 





0.5 


26 


7.3 


10.1 


0.1 


0.6 


Btgl 


32-48 


1.0 


50.0 


49.0 6.7 





0.4 


35 


10.5 


18.4 


0.2 


2.2 


Btg2 


48-70 


1.0 


49.0 


50.0 7.8 





0.2 


35 


10.1 


16.4 


0.2 


4.3 


BC 


70-92 


1.0 


49.0 


50.0 7.9 








32 


9.9 


13.7 


0.3 


4.1 


Cgl 


92-120 


1.0 


49.0 


50.0 7.8 








31 














Cg2 


120+ 








8.0 








25 


8.1 


14.3 


0.4 


4.1 


3. Orthic Dark Gray Chernozem 50 25*N, 119 1VW 


Ap 


0-21 


8.2 





58.0 6.0 





3.0 


43 


7.7 


11.5 


1.8 





AB 


21-30 


5.0 





72.2 6.4 





2.2 


50 


16.1 


16.4 


1.9 


0.3 


Bt 


30-60 


5.0 





74.5 6.7 





2.0 

















BCcal 


60-93 


0.0 




90.0 7.1 





0.7 

















BCca2 


93-115 


1.0 




88.5 7.3 





0.5 

















Ck 


115-135 


1.0 




94.2 7.5 


_ 


0.2 














_. 


4. Eluviated Dark Brown Chernozem 49 49'N 


119 37'W 


Ap 


0-14 


16.8 


47.7 


35.5 6.8 





3.8 


32 


19.6 


6.2 


2.9 


0.1 


Ae 


14-20 


19.1 


49.3 


31.5 6.7 





1.6 


26 


14.9 


6.3 


1.8 


0.1 


Bt 


20-35 


7.6 


43.0 


49.6 6.7 





0.8 


33 


20.3 


10.0 


1.3 


0.1 


BC 


35-49 


1.4 


66.2 


32.4 7.0 





0.7 


29 


18.5 


11.9 


0.8 


0.1 


Ck 


49-82 


1.6 


38.4 


60.0 7.7 








31 


18.9 


11.6 


0.6 


0.3 


5. Orthic Gray Luvisol 50 26'N, 119 irW 


Ap 


0-9 


10.2 


31.7 


58.0 7.4 


0.13 


4.8 

















Ae 


9-24 


5.2 


25.7 


69.0 7.4 


0.13 


2.7 

















Btl 


24-39 


2.0 


17.0 


81.0 7.6 


0.08 


1.3 

















Bt2 


39-64 


0.0 


14.5 


85.5 8.2 


0.23 


0.9 

















BCca 


64-90 


0.0 


16.0 


84.0 8.2 


2.60 


0.5 

















CBca 


90-108 


0.0 


10.0 


90.0 8.1 


4.70 


0.5 

















CK 


108-125 


3.0 


55.5 


41.5 8.1 


4.50 


0.5 








_ 


_ 


- 


6. Orthic Gray Luvisol 54 23*N, 124 6W 


L,P,H 


5-0 








5.4 





24.6 


104 


39.6 


27.8 


4.3 


0.3 


Ae 


0-5 








4.9 





2.6 


36 


8.3 


11.1 


0.8 


0.1 


AB 


5-13 


7.0 


27.0 


66.0 4.8 





1.4 


27 


5.7 


10.3 


0.5 


0.1 


Btl 


13-25 


1.0 


16.0 


83.0 4.5 





1.0 


37 


8.4 


16.7 


1.1 


0.2 


Bt2 


25-51 


1.0 


14.0 


85.0 4.5 





1.1 


43 


9.4 


23.0 


0.7 


0.7 


BC 


51-62 








4.7 








37 


8.2 


27.2 


0.5 


0.9 


C 


62-110 





31.0 


69.0 5.6 








31 


7.5 


21.4 


0.3 


1.3 



ennially frozen soil horizons. If these soils occur 
on sloping topography, they are subject to solif- 
luction and erosion when the thermal balance is 
disturbed. Cracking is common in these soils, 
but these cracks are the result of thermal con- 
traction, not desiccation. Clayey soils occurring 
on level topography are subject to thermokarst. 
Well to imperfectly drained clayey soils occur- 
ring in the North Boreal Ecoclimatic Province 
have either a brownish Bm horizon without sig- 
nificant clay accumulation or a Bt horizon with 
clay accumulation. Those soils having Bm hori- 
zons are classified as Brunisols, those with Bt 
horizons, as Luvisols. Poorly drained clayey 
soils in this region commonly have a peaty sur- 
face horizon and strongly gleyed and mottled 



TARNOCAI, MILLS, VELDHUIS, LUTTMERDING, AND GREEN: CLAYEY SOILS OP NORTHERN CANADA AND THE CORDILLERA 225 



mineral horizons. These soils are classified as 
Gleysols or, if permafrost is present, as Cryosols. 

In the South Interior Cordilleran Ecoclimatic 
Province, those clayey soils having a dark L-F-H 
or organic-rich Ah (or Ap) horizon are classified 
as Luvisols or Chernozems. When characterized 
by pronounced mottles or matrix colours of low 
chroma, they are classified as Gleysols. 

Arctic and subarctic clayey soils show no evi- 
dence of slickensides and desiccation cracking. 
These soils, with the exception of the surface 
horizons, are moist to wet throughout the sum- 
mer because the cool climate results in lower 
evapotranspiration than is found in the warmer 
southern soils. The melting of ice in both the 
seasonally frozen layer and the near-surface 
permafrost continuously releases moisture dur- 
ing the summer months, keeping these soils 
moist or wet. Slickensides have been observed 
in some of the clayey Luvisols in the boreal 
(Manitoba and Saskatchewan) and Cordillera 
(southern interior of British Columbia). These 
soils also show evidence of cracking due to desic- 
cation, although no grumic or churning charac- 
teristics have been observed. 

According to the present criteria, some of the 
clayey soils may fit some of the Vertisol subor- 
ders of the American soil taxonomy. At the pres- 
ent time, however, there is not sufficient data to 
verify this. Vertic properties, if present, do not 
appear to be strong enough to warrant the soils 
being classified under the proposed Cryert sub- 
order. 

References 

Acton, D., R. Smith and G. Coen. These proceedings. 
Clayey soils of the grasslands and southern boreal. 

Agriculture Canada Expert Committee on Soil Survey. 
1987. The Canadian system of soil classification. Re- 
search Branch, Agriculture Canada, Publication No. 
164, second edition, 164 p. 

Atmospheric Environment Service. 1982 a. Canadian cli- 
mate normals, temperature, and precipitation: the 
North - Yukon Territory and Northwest Territories. 
Environment Canada, Atmospheric Environment Serv- 
ice, UDC:551.582(712), 55 p. 

Atmospheric Environment Service. 1982b. Canadian cli- 
mate normals, temperature, and precipitation: Prairie 
Provinces. Environment Canada, Atmospheric Envi- 
ronment Service, UDC:55 1.582(712), 429 p. 

Bentley, C.F. (editor). 1978. Photographs and descrip- 
tions of some Canadian soils. Volume 4, Eleventh Con- 
gress of the International Society of Soil Science, Ed- 
monton, Canada, 98 p. 

Bostock, H.S. 1970. Physiographic subdivisions of Can- 
ada, in Geology and Economic Minerals of Canada. 
Department of Energy, Mines and Resources Canada, 
Geological Survey of Canada, Economic Geology Report 
No. 1, p. 9-30. 



Brown, R.J.E. 1967. Permafrost in Canada. Geological 
Survey of Canada, Department of Energy, Mines and 
Resources, Map 1246A, First Ed. 

Brown, R.J.E. and W.O.Kupsch. 1974. Permafrost termi- 
nology. National Research Council of Canada, NRCC 
14274, Technical Memorandum No. Ill, 62 p. 

Clayton, J.S., W.A. Ehrlich, D.B. Cann, J.H. Day and I.E. 
Marshall. 1977. Soils of Canada. Research Branch, 
Canada Department of Agriculture, 2 vols. and accom- 
panying maps. 

Ecoregion Working Group. 1989. Ecoclimatic regions of 
Canada, first approximation. Ecoregion Working 
Group of the Canada Committee on Ecological Land 
Classification. Ecological Land Classification Series 
23, Sustainable Development Branch, Canadian Wild- 
life Service, Conservation and Protection, Environment 
Canada, Ottawa, Canada, 119 p. and map. 

Harris, S.A., H.M. French, J.A. Heginbottom, G.H. 
Johnston, B. Ladanyi, D.C. Sego and R.O. van Everdin- 
gen. 1988. National Research Council of Canada Tech- 
nical Memorandum No. 142, NRC Report No. 27952, 
156 p. 

Mackay, J.R. 1972. Some observations on ice-wedges, 
Garry Island, NWT. in Kerfoot, D.E., ed. Mackenzie 
Delta area monograph, 22nd International Geological 
Congress, Montreal, Brock University, St. Catharines, 
Ontario, 174 p. 

Mackay, J.R. 1974. Ice- wedge cracks, Garry Island, 
Northwest Territories. Canadian Journal of Earth Sci- 
ence, 11:13361383. 

Mills, G.F., R.G. Eilers and H. Veldhuis. These proceed- 
ings. Thermal regime and morphology of clayey soils in 
Manitoba, Canada. 

Pettapiece, W.W., C. Tarnocai, S.C. Zoltai and E.T. 
Oswald. 1978. Guidebook for a tour of soil, permafrost 
and vegetation relationships in the Yukon and North- 
west Territories of northwestern Canada, Tour 18. 
llth Congress of the International Society of Soil Sci- 
ence, Edmonton, Canada, 165 p. 

Prest, V.K. 1970. Quaternary geology, in Geology and 
Economic Minerals of Canada. Department of Energy, 
Mines and Resources Canada, Geological Survey of 
Canada, Economic Geology Report No. 1, p. 675-764. 

Smith, G.D. 1986. The Guy Smith interviews: Rationale 
for concepts in soil taxonomy. T.R. Forbes (editor). 
SMSS Technical Monograph No. 11, Washington, D.C., 
259 p. 

Soil Survey Staff. 1975. Soil taxonomy. Soil Conservation 
Service, U.S. Department of Agriculture, Agriculture 
Handbook No. 436, Washington, D.C., 754 p. 

Tarnocai, C. 1978. Distribution of soils in northern Can- 
ada and parameters affecting their utilization, llth 
International Congress of Soil Science Transactions, 
Volume 3, Symposia Papers, p. 281-304. 

Tarnocai, C. and S.C. Zoltai. 1978. Earth hummocks of 
the Canadian arctic and subarctic. Arctic and Alpine 
Research 10:581-594. 

Washburn, A.L. 1979. Geocryology. Edward Arnold, 
London. 406 p. 

Wilding, L.P. and D. Tessier. 1988. Genesis of Vertisols: 
shrink-swell phenomena. In: Vertisols: Their distribu- 
tion, properties, classification and management. 

L.P. Wilding and R. Puentes (editors), Texas A&M Univer- 
sity System and Soil Management Support Services, 
College Station, Texas, p. 5581. 



226 



SIXTH INTERNATIONA! S oil C LASSIFICATION WORKSHOP 



Discussion 

Don Gross - What part of the soil has the thermal prop- 
erties to expand and contract to form the thermal contrac- 
tion cracks? Does the mineral fraction have such thermal 
contraction properties? 

C. Tarnocai - Frost cracking is the result of thermal 
contraction of the soil in subfreezing temperatures. When 
the frost crack develops, the soil is in a frozen state. That 
is, the active layer (the layer which thaws and refreezes 
annually) is frozen to the permafrost. The initial frost 
crack starts at the surface of the soil and can extend to a 
depth of 4 m. The cracks are primarily imprinted in the 
permafrost rather than in the active layer and, thus, they 
recur at the same place year after year. Thermal contrac- 
tions (frost cracks) develop in both mineral and organic 
soils. 

J,L. Sehgal - The arctic and subarctic soils of northern 
Canada have been grouped as CRYOSOLS. We have such 
soils in the high Himalayas which qualify as Cryochrepts. 
How would you classify these soils in the U.S. System of 
Soil Taxonomy? 

C. Tarnocai - I am not experienced in classifying per- 
mafrost soils using the U.S. Soil Taxonomy. We tried to 
use this soil classification in the early 1970s, but had great 
difficulty. The northern soils classified as Cryosols in the 
Canadian System of Soil Classification fit into the Pergelic 
subgroups (e.g. Pergelic Cryochrept and Pergelic Ruptic 
Cryochrept) of the U.S. Soil Taxonomy. 

Warren Lynn - You mentioned that frozen soils cracked 
at weak points. What are the weak points and can they be 
predicted? 



C. Tarnocai - The weak point could be the change in ice 
content or textural differences. There is no definite way, 
as far as I know, to predict these weak points. If a frost 
crack develops, however, the soil will crack at the same 
place in the future. 

Jim Richardson - Please cover the mechanism for frost 
cracking? 

C. Tarnocai - Frost cracking is the fracturing of soil 
material by thermal contraction in subfreezing tempera- 
tures. All mineral soil materials contract a certain 
amount at low temperatures but; since ice is the critical 
material in frozen soils, its contraction plays an important 
role. Pure ice has a coefficient of linear contraction of 
about 45xlO- 6 C-1 at -40C. Soil material, however, is 
non -homogeneous and thus a complex system. Other fac- 
tors affecting cracking are the salt content, the highly 
variable distribution of ice in the frozen soil, and the insu- 
lating effect of the snow. Frost cracks can develop in both 
coarse and fine textured soil materials as well as in or- 
ganic materials. Fine textured, moisture-rich soils are 
probably the soils which are most susceptible to frost 
cracking. 

A. Qsman -What is the origin of the sodium salts in the 
soils of northern Canada and how are they formed? 

C. Tarnocai - The salts in the northern soils originate 
from the parent materials. Soils which developed in resid- 
ual or till materials derived from shale and marine depos- 
its are commonly associated with salts. A whitish salt 
crust can develop on the surface of these soils as a result of 
evaporation of moisture from the soil surface during 
warmer summer days. 



Vertisols of France 
Daniel Tessier 1 , Ary Bruand 2 , and Yves-Mary Cabidoche 3 

Abstract 

This paper deals with a general presentation of Vertisols and related soils 
of France. Vertisols and related soils are mainly developed on clayey geo- 
logical substrata. Clay minerals are either 2:1 interstratified illite-smectite 
clays or smectite-rich materials. Shrinkage curves on undisturbed samples 
exhibit a normal shrinkage in a wide range of water content. The shrinkage 
amplitude is highly variable and appeared to be dependent on both clay 
content and mineralogy. Water retention curves differed considerable from 
one soil to another. A statistical study showed the respective contribution 
of clay content, clay mineralogy (CEC and EGME surface areas), and soil 
fabric to soil physical properties. 



Introduction 

The geographic situation and climatic condi- 
tions encountered in France are not generally 
favorable to the development of Vertisols. How- 
ever, some soils found on a large range of clay- 
rich parent materials can be classified following 
Soil Taxonomy in the order of Vertisols 
(Duchaufour, 1982). Moreover, the behavior of 
many other related soils is strongly influenced 
by the presence of vertic B horizons, although 
these soils are not classified as Vertisols. This is 
particularly the case for Pelosols such as defined 
in the German classification (Bonneau et al., 
1965 and 1967; Begon and Jamagne, 1973). 
Because these vertic B horizons play an essen- 
tial role in soil behavior, numerous research 
works have been aimed at establishing a rela- 
tionship between intrinsic characteristics such 
as clay mineralogy, texture, and behaviors. 

This paper first presents the geographic situ- 
ation, the climatic conditions, and the distribu- 
tion of Vertisols and related soils in relation to 
the nature of parent materials. It then exam- 
ines the main properties of these soils, especially 
their physical properties. 

Geography and Climate 

France extends from 42 to 52 north latitude 
and from 5 and 8 west and east longitude, re- 
spectively. The characteristics of the climate 
prevailing over Vertisols and related soils in 
France are presented in Fig. 1. Except for 
mountain areas, rainfall ranges from 600 to 900 



Station de Science du Sol, INRA, Route de S* Cyr, 
78026 Versailles, France. 

2 SESCPF, INRA, Centre d'Orleans, Ardon, 45160 Ol- 
ivet, France. 

3 CRAAG, INRA, Centre De Guadeloupe, Petit Bourg, 
BP 1232 97184 Pointe a Pitre, Guadeloupe, Frenc Indies. 






500 - 800 mm 

l^m] 600 - AGO mm 

aoo . 1200 

>> 180 mm 



Figure 1. General climatic data: (a) isotherms in 
January; (b) isotherms in July; and (c) mean an- 
nual precipitation. 



mm per year. In the north-west part of the 
country, rainfall is distributed regularly. The 
oceanic temperate climate is characterized by 
mild temperatures, and 7C and 18 C are the 
mean temperatures in January and July, re- 
spectively, with 180 days of rainfall in the Brest 
area. Further eastward the climate changes 
from oceanic to continental. Winters are dry 
and cold, whereas summers are wet and 
warmer. In the southern part of France the 
rainfall distributed irregularly. The dry season 
is well expressed, with about 60 rainfall days 
per year, while mean temperatures reach 7C 
and 23 C for the colder and warmer months, 
respectively, i.e., January and July. The gen- 
eral elevation of Vertisols and related soils 
ranges from sea level to 500 m above sea level. 



228 



SIXTH iNTERNATIONAl Soil CLASSIFICATION WORKSHOP 



Spatial Distribution 

Vertisols and related soils (i.e., those with 
vertic B horizons) occur in France on areas with 
intensive agricultural activities. They cover 
about 5 percent of cultivated soils and are lo- 
cated mainly in the Paris and Aquitanian Basin 
(Fig. 2). Vertisols were described in the south- 
ern part of the Beauce Plain near Orleans by 
Fedoroff and Fies (1968), Horemans (1984), and 
Arrouays (1987). In the south of France, Bonfils 
(1988) found Vertisols close to the Mediterra- 
nean Sea. Vertisols in Guadeloupe the French 
West Indies have developed on the coastal plain 
under climatic conditions very different from 
those in metropolitan France ( Jaillard and Cabi- 
doche, 1984). 

Parent Materials 

Vertisols occurring on basalts are present 
near Agde in the south of France under Mediter- 
ranean climate (Bonfils, 1988). Other Vertisols 
have developed on clayey sedimentary deposits 
which are usually interstratified illite-smectite 
or smectite-rich materials of both secondary and 
tertiary ages (Bornand et al., 1975 and 1984; 
Jamagne et al., 1970; Fedoroff and Fies, 1968). 
The more frequent situation corresponds to the 
presence of clay minerals similar to those pres- 
ent at the time of the sediment formation. 

Clay minerals are sometimes transformed 
during pedogensis, as described by Robert et al., 
(1973). For instance, in the Paris Basin the 
Albian and Cenomanian glauconite-rich materi- 
als are weathered into iron-rich smectites, and 
Vertisols have developed in these smectites (Is- 
ambert, 1984). Soils related to Vertisols appear 
on numerous sedimentary clayey outcrops. 
These soils are characterized by vertic struc- 
tured B horizons. Parent materials are some- 
times clayey marls on Trias and Lias sediments 
(Nguyen Kha, 1973; Bonneau et aL, 1967), on 
glauconitic Cenomanian formation (Isambert, 
1984), or on clayey Quaternary sediments 
(Salin, 1983). According to the type of sedimen- 




sroas with Varitaofs and r^at<J oM 

Figure 2. Schematic map of Vertisols and related soils 
with both A and B clayey horizons. 



TABLE 1: Main characteristics of 9 vertic B horizons and related soils 


Soil 


Organic 
Depth 
cm 


Clay 


Silt 
2-50}im 


Sand 
50|im-2mm 


pHin 
water 


CaCO, 


CEC.j 
m.e.g 


Carbon 


ST FRANCOIS 
FREVILLE 
BETHONVILLIERS 
VTLLERS-STONC. 
THIFON-ALO 
CHEZAL-BENOIT 
LABOUZOULE 
LAVIAUDE 
THIANGES 


60-90 
24-44 
25-60 
25-45 
85-105 
60-95 
20-50 
25-40 
25-70 


85.4 
82.6 
52.8 
66.2 
72.6 
92.9 
56.0 
58.2 
63.5 


9.4 
14.2 
37.3 
30.6 
17.7 
4.5 
42.2 
41.4 
26.5 


5.2 
3.2 
9.9 
3.2 
9.7 
2.6 
1.8 
0.4 
10.0 


8.0 
8.2 
7.1 
7.5 
4.8 
4.7 
8.1 
8.6 
7.4 


0.9 
0.5 
0.6 



0.9 
9.2 



0.47 
0.43 
0.39 
0.35 
0.33 
0.26 
0.20 
0.18 
0.17 


0.23 
0.67 
0.42 
0.55 
0.23 
0.19 
0.71 
0.69 
0.16 



tation and the importance of diagenesis, parent 
clay minerals are either 2:1 interstratified illite- 
smectite clays or smectitic-rich materials. 

Particle Size, Mineralogy, and 
Chemistry 

Table 1 summarizes the main characteristics 
of 9 B horizons belonging to Vertisols or related 
soils. The < 2 jam fraction is higher than 50 per- 
cent. These horizons have developed upon non- 
calcareous materials or are the final stage of a 
complete decarbonation process. Carbon con- 
tent is always low (less than 1 percent) and cat- 
ion exchange capacities are saturated with Ca, 
Mg, K, and Na cations, with Ca being dominant. 
Fig. 3 shows the main types of mineralogy 
encountered. Clays have been classified accord- 
ing to cation exchange 
capacities and accessible 
surface to ethylene glycol 
mononethyl (EGME). 
Both characteristics are 
generally very well corre- 
lated. X-ray diagrams 
and transmission elec- 
tron micropscopy show 
that the proportion of 



TESSIER, BRUAND, AND CABIDOCHE: VERTISOLS OF FRANCE 



229 



000 



900 



^ 400 

i 
o 



300 



ST FRANCOIS 
- LORRIS ALo 



BETHONVN.LIERS 



AUV, ( B ) 



\ SMECTITES 



214 m 

HARAS OU PIN 



VIL.ST. 
& LtCH 



H VK5W1ERE8 

\ 



AUV.Ap 
LA JUSTICE 



Z1B 

INTERSTRATIFIED 

LA VIAUOC 



BCHEZAl 8EWOIT 
\ 
\ 



ILLITES 



0,2 



0,4 



o,* 



0,4 



CEC ( mEq Q-i) 

Figure 3. Cation exchange capacities and EGME 
surface areas as a function of clay mineralogy. 



opened 2:1 layers ranges from about 30% to 
nearly 75%. Tessier et al. (1989) established 
that the higher the CEC the lower the clay par- 
ticle size, consequences of shrink-swell phenom- 
ena, air entry point and structural stability 
were deduced. Thus, EGME and CEC appeared 
to be the best criteria to characterize soil clay 
mineralogy. 

It was also noticed that soils developed on the 
same clay-rich parental material is mineralogi- 
cally homogeneous and thus typical of the par- 
ent material. 

Water Retention and Shrink-Swell 
Phenomena 

These soil characteristics are generally con- 
sidered in laboratory studies on centimetric 
clods (5 to 10 cm 3 ). To study them, large undis- 
turbed samples are collected, preferably at the 
end of winter or at the beginning of spring (i.e., 
full and homogeneous wetting). Drying is 
avoided before measurements. Clods are ob- 
tained by hand breaking and equilibrated with 
a range of water potentials from -1 KPa to -2 
MPa. Water contents and bulk densities are 
measured (Tessier and Berrier, 1979; Bruand 
and Prost, 1987). Similar measurements also 
can be made on clods in equilibrium with differ- 
ent relative humidities. 



0.60 



0.80-- 



0.70 



0.60-- 



0.50- r 




0.10 



0.20 0.30 



0.40 



0,50 



water content cm 



" 1 



Figure 4. Shrinkage curves of two vertic B horizons 
with 1.1 g.cm-3 (-.-.-) and 1.5 g. cm-3 ( ) bulk densi- 
ties at field capacity, repectively. 



0.4 



Q> 



0.2 




2 
log(P) 



3 
(hPa) 



Figure 5. Water retention curves of three vertisols. 



On the studied B horizons, drying curves ex- 
hibit a normal shrinkage in a wide water con- 
tent range (Fig. 4). Normal shrinkage mainly 
corresponds to a complete water saturation. 
Nevertheless, Cabidoche and Voltz (1987) 
showed the presence of an incomplete water 
saturation due to biopores. On the other hand, 
water retention curves can be different to a 
large extent (Fig. 5). For instance, at a given 
water potential, water content can be twice as 
much in one vertisol as in another. 

A device also was used to measure soil verti- 
cal movement (Cabidoche and Voltz, 1988). The 
results were interpreted in terms of horizontal 
and vertical cracking and water dynamics in 
biopores and clay matrix pores. 



230 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Relationship Between Measurable 
Intrinsic Properties and Behaviors 

Vertisols and related soils of France have 
quite varied clay contents, clay mineralogies, 
water retention, and shrinkage curves. At- 
tempts have been made to establish a relation- 
ship between measurable intrinsic properties 
and behaviors. 

A study was carried out on the < 2(im clay 
fraction extracted from 8 vertisols and related 
soils (Tessier et al., 1990). Sample preparation 
was standardized: drying at 48% RH, slow 
re wetting up to a given water potential. A corre- 
lation between clay hydration properties and 
macroscopical clay swelling on one hand, EGME 
surface area and CEC on the other hand, was 
found. 

These results were compared to water behav- 
iors of clods sampled at field capacity (Bruand et 
al., 1988). The statistical relationship has 
shown the respective contribution of clay con- 
tent, mineralogy, and fabric to soil properties in 
a range of water potentials from -32 KPa to -1.6 
MPa (Bruand et al., 1988). Thus, it was estab- 
lished that taking into account soil CEC allows 
better prediction of water content than clay con- 
tent does. At -32 KPa, the latter accounted for 
58% of water content variance, in contrast to 
79% for the former (Table 2). Similar results 
were obtained at -1.6 MPa. 

Furthermore, the natural soil fabric also can 
be taken into account by measuring the pore 
volume due to the clay phase in the samples 
near field capacity. A simple way to take this 
parameter into account was to consider the bulk 
volume of the natural clods sampled at filed 
capacity. Pore volume is correlated positively 
with water content at different water potentials. 
This correlation allows 93 and 87% of the water 
content variance to be explained at -32 KPa and 
-1.6 MPa, respectively (Bruand et al., 1988). 
Similar results were obtained with shrinkage. 
The shrinkage amplitude appeared to be de- 
pendent on both clay content and mineralogy. 
The higher the bulk volume (or the lower the 
bulk density), the higher the shrinkage. 

It was concluded that the prediction of soil 
physical properties is considerably improved by 
taking CEC into account. A further significant 
improvement was obtained when soil fabric, 
numerically expressed by the bulk volume, at 
near field capacity, was considered. 



TABLE 2: Percentage of explained variance accounted for 

by clay content, C.E.C. and bulk volume on water content 

at different water potentials. 



Water 
potential 


Clay 
fraction C.E.C. 


Bulk 
volume 
at field capacity 


-32 KPa 
- 1.6 MPa 
- 100 MPa 


58 79 
62 83 
42 91 


87 
93 
73 



Water Regime and Soil Structure 

Shrinkage is a very important aspect of Verti- 
sols and related soils. Examination of volume 
changes, as a function of water loss on centimet- 
ric or larger clods, revealed a normal shrinkage 
in a large range of water potentials. This means 
that each volume of water lost is equivalent to 
the observed volume change. Consequently, a 
network of cracks is not normally observed on 
centimetric clods, and only pores resulting from 
clay particle arrangement become closer (Wild- 
ing and Tessier, 1988). Thus, water is located in 
very fine pores (< l|im in size - Tessier, 1984), 
hydraulic conductivity of the system is low, and 
cohesion can become very high. 

Consequently, in vertisols developed under 
very contrasted climate such as Vertisols in 
French Guadeloupe (Jaillard and Cabidoche, 
1984), water distribution in the profile remains 
heterogeneous even after a long wet season. In 
France, this heterogeneity in water distribution 
occurs in summer but disappears during winter. 

Conclusion 

Even if the extension of vertisols in France is 
limited, numerous related soils with B vertic 
horizons are present. Except for studies carried 
out on Vertisols of Guadeloupe, research works 
on this subject were devoted to understanding 
vertic B horizons (Tessier et al., 1989-2). They 
also incorporate the main clayey soils developed 
in France. The soil structure developed under 
natural conditions during soil formation and 
weathering appeared as a determining factor to 
explain and predict soil behaviors. When pos- 
sible, soil behavior has been studied in the labo- 
ratory on undisturbed and undried samples. 
There is no contradiction with classical studies 
using standardized data (tests). 

Nevertheless, it is stressed that sample 
preparation can modify the system and only 
express a potential behavior, which does not 



TESSIER, BRUAND, AND CABIDOCHE: VERTISOLS OP FRANCE 



231 



necessarily correspond to field behavior. Study- 
ing undisturbed sample changes by measuring 
water content, shrink-swell phenomena, and 
presence of air in soils is also a way to develop 
numerically express parameters which can be 
introduced in predicting models. 

Literature Cited 

Arrouays D. 1987. Carte des sols 1/50 000. Notice de la 
feuille de Bellegarde du Loiret. INRA-SESCPF, INRA 
publ. Versailles, 152 p. 

Begon J.C. and Jamagne M. 1973. Les notions de pseu- 
dogley et de pelosol dans la signification des sols de la 
R.F.A. Science du sol 4:223-239. 

BonfilsP. 1988. Carte pedologique de France y 100 000. 
Feuille de Lodeve. INRA-SESCPR, INRA publ. Ver- 
sailles. 

Bonneau M., Duchaufour Ph., Le Talon F. and N'guyen 
Kha. 1967. Note sur quelques sols developpes sur 
substratum argileus. PedologieXVII-l:106-108. 

Bonneau M., Duchaufour Ph., Millot G. and Paquet H. 
1965. Note sur certains sols vertisoliques en climat 
tempere. Bull. Soc. Geo. Als. Lorr. 17-4:325-334. 

Bornand M., Dejou J. and Servant J. 1975. Les Terres 
Noires de Limagne: Leurs differents facies et leur 
place dans la classification francaise des sols. C.R. 
Acad. SC. Paris, 281-D: 1689-1692. 

Bruand A., Tessier D., and Baize D. 1988. Contribution a 
1'etude des proprietes de retention d'eau des sols argi- 
leux: importance de la phase argileuse. C.R. Acad. Ac. 
Paris. 307-11:1937-1941. 

Bruand A. and Prost R. 1987. Effect of water content on 
the fabric of a soil material: an experimental approach. 
J. Soil Sci., 38:461-472. 

Cabidoche Y.M. and Voltz M. 1987. Variations isotropes 
de volume en sols argileux heterogenes. Controle ex- 
perimental dans le cas d'un vertisol calcique de Guade- 
loupe, p. 143-159. In Tranfert dans les milieux poreux 
deformables, INRA, Department de Science du sol, 
Versailles publ., p. 143-159. 

Duchaufour Ph. 1982. Pedology. Pedogenesis and classi- 
fication. George Allen and Uniwin Publ., London, 448 

P- 

Fedoroff N. and Fies J.C. 1968. Les vertisols du sud-est 
de la Beauce. Bulletin de TAssociation francaise pour 
1'Etudedu sol, 1:19-32. 

Horemans P. 1984. Notice explicative de la carte pedolo- 
gique de France a 1/250 000. Feuille de Paris. 
SESCPF, INRA publ., Versailles. 202 p. 



Isambert M. 1984. Carte pedologique de France 1/100 
000. Feuille de Chateaudun. Notice explicative. 
INRA-SESCPF, INRA, publ., Versilles, 259 p. 

Jaillard B. and Cabidoche Y.M. 1984. Etude de la dy- 
namique de 1'eau dans un sol argileux gonflant: dy- 
namique hydrique. Science du sol 3:187-198. 

Jamagne M., Bliet L. and Remy J.C. 1970. Contribution a 
1'etude pedologique et agronomique des sols argileux du 
Bassin Parisien. La haute Brie. Annales Agronomiques 
21-2:119-157. 

Nguyen Kha. 1973. Recherches sur revolution des sols a 
texture argileuse en conditions temperees et tropicales. 
Thesis Univ. Nancy. 166 p. 

Robert M., Hardy M., and Elsass F. 1989. Crystallochem- 
istry and organization of soil clays derived from major 
desimentary rocks in France. Abstract Int. Clay Con- 
ference, Strasbourg, France, 1989. 

Robert M., Isambert M., and Tessier D. 1973. Etude et 
premiees interpretations de revolution des glauconites 
dans les sols. C.R. Acad. Sc. Paris 277-D:1129-1139. 

Salin R. 1983. Etude pedologique dans les Marais de Ro- 
chefort au 1/25 000. SESCPR-INRA, INRA publ., Ver- 
sailles 200 p. 

Tessier D. 1984. Etude experimentale de 1'orgamsation 
des materiaux argileux. Thesis Univ. Paris VII INRA, 
Versailles publ. 361 p. 

Tessier D. 1987. Identification of clays. Data from inves- 
tigations with strongly hydrated systems. Proc. 20th 
Colloquium Int. Potash Institute, 1987, Int. Potash Int. 
publ., Bern, Switzerland 45-63. 

Tessier D., Beaumont A., and Pedro G. 1990. Influence of 
rewetting rate on clay microstructure. Proc. Int. Micro- 
morphology Meeting, San Antonio, Texas 1988, L. 
Douglas publ., 115-121. 

Tessier D. and Berrier J. 1979. Utilisation de la micro- 
scopie electronique a balayage dans Tetude des sols. 
Observation de sols humides a differents pF. Science 
du Sol 1:67-82. 

Tessier D., Bruand A., and Beaumont A. 1989. Relation- 
ship between clay minerzlogy and soil behavior in Paris 
Basin clayey soils. Abstract Int. Clay Conf. Strasbourg 
1989. 

Wilding L.P. and Tessier D. 1988. Vertisols: Their distri- 
bution, properties, classification, and management. 
Article in the book: Genesis fo vertisols: shrink-swell 
phenomena, 55-81, L.P. Wilding and R. Puentes ed., 
Texas A&M Univ. Publ., College Station, Texas. 



234 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Surface 




Surface 







US 



45-50 % 



40-45 % 



1 2 3 4 5 

Distance (m) 

Figure 3: Minimum volumetric seasonal soil moisture 
content (%) (after Spotts, 1974). 



I 

4 



12345 

Distance (m) 

Figure 4: Maximum volumetric seasonal soil moisture 
content (%) (after Spotts, 1974). 



bility of cracked, dry microlows is initially 
greater than that of microhighs; most water is 
transmitted down macrovoids to junctures with 
deep subsoil clay. Following recharge, the per- 
meability of microlows becomes extremely slow. 
When wet, the permeability of Vertisol micro- 
highs is reported to be 3-4 times greater than 
microlows. They found no consistent pattern for 
water retention relationships (i.e. field capacity, 
wilting point, or available water) between gilgai 
elements. 

Spotts (1974) also observed that strain (per- 
cent dimensional change - either expansion or 
contraction) was in the zone of maximum 
change in soil water content (Fig. 6). The most 
active site was 45 cm below the surface of the 
microhigh. In the depression, the zone from 65- 
80 cm was the most active layer, but it was not 
as active as in the microhighs. 

While these zones may express the greatest 
active movement, they may not correspond to 
zones which have the greatest shear failure, 
commonly observed between 1-2 m below the 



Table 2. General trends in selected physical 
properties between microlow and microhigh 
gilgai elements. 


Property Microlow 


Microhigh 


Cracking Expression 
(Width & Depth) Higher >* 
Water Movement 
External Runon < 
Internal Higher > 
<** 
Moisture Content (%) Higher > 
Clay Content (%) Higher > 
Specific Surface Area (mVg) Higher > 
COLE (cm/cm) Higher > 
Cohesive Strength Lower > 
Plasticity Index Higher > 
Shrinkage (%) Higher > 


Lower 

Runoff 
Lower 

Lower 
Lower 
Lower 
Lower 
Higher 
Lower 
Lower 


*> Commonly reported trends. 
**< Reverse trend has been reported. 



surface (Yaalon and Kalmar, 1978). Wilding 
and Tessier (1988) and Thompson and Beck- 
mann (1982) postulated that the microlows 
would have the greatest change in moisture 
content between wet/dry states and greatest 
shear failure because of topographic position, 
cracking patterns and vegetative communities 
that extract water to greater depths and lower 
matric potentials. However, this may not be the 
case, based on Spotts (1974) work. Additional 
studies of this nature are needed to clarify soil 
moisture/strain relationships as a function of 
gilgai elements. 

Yule and Ritchie (1980) sampled microhigh 
and microlow gilgai elements of Houston Black 
(Udic Pellusterts) and Burleson (Udic Pel- 
lusterts) to compare soil shrinkage relation- 
ships. They observed higher 15-bar water re- 
tention and total shrinkage in microlows. Clay 
contents in microlows were also higher than in 
microhighs. The coefficient of linear extensibil- 
ity (COLE) gave a good estimate of total vertical 
shrinkage but the expected 1:1 relationship was 
not found. They further observed that vertical 
shrinkage could be estimated from CEC almost 
as well as from COLE indices. 

Expansion and contraction of Vertisols is re- 
lated to the amount of water added or removed 
during seasonal cycles of desiccation/rewetting 
cycles (Wilding and Tessier, 1988). Measured 
elevational changes have been as much as 8 cm 
when dry soil is wet to field capacity (Aitchison 
and Holmes, 1953). Relative changes between 
gilgai elements are not well-known, but it is 
postulated by Thompson and Beckmann (1982) 
that, at different seasonal periods, the soil of one 
part of the gilgai complex may be relatively ex- 



WILDING, WILLIAMS, MILLER, COOK, AND ESWARAN: CLOSE INTERVAL SPATIAL VARIABILITY OP VERTISOLS 



235 





Distance (m) 

Figure 5: Maximum change in volumetric water con- 
tent (%) from driest to wettest state (after Spotts, 
1974). 



Distance (m) 

Figure 6: Percentage soil movement (strain) in micro- 
high and microlow (after Spotts, 1974). 



panded, while the adjacent part is contracted, 
providing for maximal or minimal elevation dif- 
ferences. 

For example, immediately following a wet 
period, the microhighs would begin to dry out 
even though water was ponded in adjacent mi- 
crolows. However, when the water of the mi- 
crolows has discharged, vegetation in these posi- 
tions would flourish. Then, the microlows would 
dry out more rapidly than microhighs. With 
cracking, microlows would become desiccated to 
depths of 1 m or more while microhighs would 
still be moist at 20 cm below the surface (Th- 
ompson and Beckmann, 1982). Vegetation in 
microlows varies also with depth and duration 
of ponded water conditions; in deeper microlows, 
there is often no vegetation (Tucker et al., 1989). 
This is a fruitful field for further study that has 
been little explored. 

Clay Contents and Clay Mineralogy 

Thompson and Beckmann (1982) found no 
consistent pattern in clay content or clay depth 
functions between gilgai elements. Yule and 
Ritchie (1980) found that Vertisols in Texas 
have only slightly higher clay contents in mi- 
crolows than in microhighs. These differences 
are likely in response to higher carbonate con- 
tents in microhighs. Likewise, clay mineralogy 
for Texas Vertisols was dominantly smectite 
with trace to small quantities of kaolinite, mi- 
cas, and quartz for both microlows and micro- 
highs. Calcite occurred in quantities reflecting 
total carbonate content in these soils. The layer 
clay minerals of these systems are indigenous to 
parent calcareous sediments (Dixon, 1982). 



Black Earths (mostly Pellusterts) of Australia 
have similar clay mineral suites (Thompson and 
Beckmann, 1982; Stace et al., 1968). 

Cracking Patterns 

Only sparse information is available on crack- 
ing patterns between microhigh and microlow 
gilgai elements. Yaalon and Kalmar (1978) re- 
corded seasonal cracking and crack infilling 
trends in Vertisols of Israel, but these did not 
have gilgai. Thompson and Beckmann (1982) 
observed that a network of cracks develop 
mainly in the depressions during prolonged 
drying periods. They form rough circular pat- 
terns delineating mounds. Visible cracks on the 
mounds were less frequent and were apparently 
much finer and shallower than in depressions. 
In a comparison of paired gilgai elements of 
Black earths, Stirk (1954) found that shrinkage 
cracks developed at lower moisture tensions in 
depressions than in mounds. Verification of 
cracking patterns, crack closure hysteresis, and 
cracking depths as a function of seasonal soil 
moisture within gilgai elements is an acute area 
of research need. 

Chemical Properties 

Table 3 provides some generalized relation- 
ships between selected chemical properties of 
the microhigh and microlow gilgai elements. 
While general trends are evident for many prop- 
erties, reversals are apparent for some of the 
more temporal conditions influenced by sea- 
sonal sampling period and site specificity (i.e. 
soluble salts, ESP, etc.). 



requires a multi-discipline approach in- 
cluding: expertise of a field soil scientist 
to locate a representative sampling site; 
local soil conservation staff to obtain per- 
mission for access to property and to as- 
certain past cultural history; a plant tax- 
onomist to make a detailed vegetative 
survey of gilgai elements; an engineer to 
conduct a detailed topographic survey of 
the specific sites; soil scientists to de- 
scribe and sample the selected site; and 
laboratory analysts to characterize the 
physical, chemical, and mineralogical 
properties. 

Selection of a Representative 
Sampling Site 

Perhaps one of the most critical steps 
in a microvariability study is making 
sure the site sampled is a good represen- 
tation of the field conditions for a given 
soil. This phase of the work was con- 
ducted by co-author, Mr. Wesley Miller, 
Area Soil Scientist, USDA-SCS, Victoria, 
Texas, who was party chief for the soil 
survey of Victoria County. 

To initiate site selection, six alterna- 
tive areas from numerous mapping units 
of the Lake Charles clay within Victoria 
County were considered. Several condi- 
tions were imposed on the selected areas, 
namely, they had to be: 1) within large mapping 
units of the Lake Charles clay; 2) uncultivated; 
and 3) have gilgai cycles with periodicities be- 
tween 5-15 m. Periodicity of gilgai and prospec- 
tive sites were screened from the office by view- 
ing older aerial photosheets of the survey area 
(1:14,840 and 1:7,920 scales). Ground truth was 
collected from near the center of the mapping 
units with linear point transects at 0.6 m inter- 



Pedon Chemical Variability 


as 


- 








(Microlow - 1m') 


H 













c 
















020 


- 














1 
















2 
















P> 

'o 

c 

u 10 

<* 


: 








1 


,*. 








i 

'$ 


".* 
fe 




1[U 




J 

5 


- 






*"*t%* 

1 **'! 




HI 


| 


W& 

i 




1 1,1 


m 







. ' " 




'f, : ,4 




rf 


1 






$$j : f=^ tV^NJ 


El 


No 


H< 


-3 ct so 4 EC SA'R 




Saturation 


Exlrac! J 


I 15 


- 








o 












" Y 






j 


JRorig* Among 
Al., Bw I C Horiiont 


"o 




^p 


1 








e 




- 


1 


1 




1 










N v 

1 


s! 


%J&t 


9 
O 




*j[ 


^ 
















1 


<> o 




vF". 




w< 




^L?.:^ 


rr 


1 


sVv 1 


$ 


Eitroct CEC 
No 


ESP 


PH Co00 3 


Figure 8: Observed chemical 


variability within a sampling 


unit of 1 m 2 for Houston Black (Udic Pellustert) microlow 


gilgai element (after Wilding, 1985). 





/V""^ 




/ \ 




t I 


25 


/^ I 




10YR 3/1 J5 \ 




1 \ 


50- 


'" 




/ o | 


75^. 


/ \\ 


100- 


' ' \l 

>'/ * 

-^ 2.5Y 5/2 




caic. 

" -. - 



6/2N 



Distance (m) 
Figure 9: Elevation and soil transect of Lake Charles clay (lypic Pelludert). 



vals such that at least three consecutive micro- 
highs and microlows were crossed. Field 
transects were oriented with calcareous micro- 
highs in the middle of the transect, and the 
transect run in a straight line or at a slight 
angle to an adjacent microhigh. 

At each of the transect points, a 4 cm core was 
taken with an hydraulic power probe to at least 
1 m depth (a few to 1.75 m) and a soil description 
written. Table 4 illustrates 
gilgai periodicity, diapir perio- 
dicity, length of transect with 
surface colors of specified 
thickness, and pedon classifi- 
cations for four of the six 
transects run. Two transects 
were not included because 
they inadvertently were ori- 
ented lengthwise along micro- 
highs, which inflated the per- 
centage of Entic versus Typic 
Pellusterts observed- 

In summary, the mean pe- 
riodicity of the gilgai cycles 



Soil Surface 



IOYR 2/0 

10YR 2/1 
IOYR 3/1 



WILDING, WILLIAMS, MILLER, COOK, AND ESWARAN: CLOSE INTERVAL SPATIAL VARIABILITY OP VERTISOLS 



239 



5 


6.30 


6.10 


1 


3.15 


3.05 


2.44 


0.61 + 








2 







1.83 


1.22 + 




9.70 


6.10 


3 


4.85 


3.05 


2.44 


0.61 + 








4 







1.83 


1.22 + 






3.66 


5 




1.83 












6 







0.00 


1.83 + 


6 


6.10 


5.80 


1 


3.05 


2.90 


1.37 


1.52 + 



Table 4. Gilgai periodicity, diapir periodicity, length of transect with given surface thickness/color 
criteria and pedon classification for Lake Charles clay, Victoria County, Texas. 



Transect 

# 



Peridicity (m) 



gilgai 



diapir* 



Pedon 

No.** 



4.25 



4.85 



6.10 



6.00 



4.00 



4.26 



3.35 
3.35 



4.26 
4.26 



7.00 



5.80 



3.04 
4.58 



3.04 



4.88 



2.12 



3.05 



3.00 



2.00 



2.13 



1.68 



1.98 



Total 



58.52 



24 



Typic Pelluderts**** 
Entic Pelluderte 



*Diapir defined as zone with thinnest and lightest color portion of microhigh 

**Pedon defined as 1/2 distance between diapirs 

***Pedon classification based on 1/2 diapir periodicity as concept of pedon 

****Pedon classification based on total length of transects with given surface colors of specified thickness 



was 4.9 m 
and ranged 
from 3.1-7.0 
m. Local 
relief differ- 
ential be- 
tween adja- 
cent micro- 
highs and 
microlows 
ranged from 
15-35 cm 
and cen- 
tered on 20 
cm. By pe- 
don classifi- 
cation, 71% 
of the 24 
pedons ex- 
a m i n e d 
were Typic 
Pelluderts 
(> 30 cm of 
10YR 3/1 or 
darker sur- 
face hori- 
zons over 
half of the 
pedon) while 29% were Entic Pelluderts (colors 
not meeting Typic and most being 10 YR 4/1 or 
lighter). Considering the proportion of the 58.5 
m total transect distance, 65% of the linear dis- 
tance consisted of Typic Pelluderts and 35% 
Entic Pelluderts. Figure 9 illustrates a crossec- 
tional profile of one of these transects very simi- 
lar to the site selected later for detailed study. 

Several insights about the short-range spatial 
variability gained from these detailed mapping 
unit transects were as follows: 

1. Calcareous diapirs (chimneys) were more 
pronounced and occupied a larger percent- 
age of the smaller diameter microhighs 
than the larger diameter microhighs (20- 
35%). 

2. The pedon is different depending on 
whether it is defined as one-half of the gil- 
gai periodicity or one-half the distance be- 
tween the thinnest and lightest-colored 
surface horizons (diapir periodicity). This 
definition difference will influence the pe- 
don concept and classification. A higher 
percentage of Entic Pelluderts will occur if 
the pedon is based on one-half the gilgai pe- 
riodicity rather than one-half the diapir pe- 
riodicity. 



Pedon Length with Surface Colors 
Pedon Length 10YR 3/1 10YR 3/1 

gilgai diapir** or darker (<30 cm thick) & 

1^30 cm thick) 10YR 4/1 or lighter 



Pedon Classification*** 
Typic Entic 

Pelluderts Pelluderts 



2.13 



2.43 2.13 



3.50 
2.90 



1.52 



2.29 



1.52 



2.44 



1.98 
1.52 
2.13 
2.13 
1.22 
1.98 
1.83 
1.22 
2.90 
1.52 
1.52 
1.68 
1.83 
0.61 
0.30 
1.53 
1.37 



0.91 
0.61 
0.00 
0.00 
0.91 
1.52 
1.68 
1.68 
0.00 
0.00 
0.00 
0.61 
0.46 
0.91 
1.22 
0.91 
1.07 



37.94 



20.57 



17 



65% 



35% 



71% 



29% 




Distance (m) 

Microhigh - 76% of area 

f|HTrncri and Sampled Padont 



Mlcrolow - 24% of area 



Figure 10: Microtopography, trench and sampled 
pedons in Lake Charles clay (Typic Pelludert). 



240 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



3. Calcareous diapirs 
are not always in 
the center of the 
microhigh but may 
be on the shoulder 
or intermediate 
position between a 
microhigh and mi- 
crolow. Some- 
times, they are a 
small, positively 
elevated area in 
the middle of the 
microhigh. They 
are most easily 
identified by 2-5% 
calcium carbonate 
nodules on the sur- 
face and associated 
gray or grayish- 
brown clay at the 
surface or within 2- 
6 cm of the surface. 

4. The transition 
area (10YR 4/1 col- 
ors) between micro- 
high chimneys and 
microlows occupies 
more area than 
previously consid- 
ered. 

5. Microhighs are 
more variable in 
morphological 
properties than 
microlows. Better 
classification and 
morphological data 
could be obtained if 
a 10 m 2 area were 
gridded at 0.5-1 m 
intervals with ele- 
vation control to yield 



Table 5. Vegetative composition of microhighs and microlows for Lake Charles clay at 
ICOMAQ site in Victoria County, Texas. 


Species 


Composition (%) 


Common Name 


Scientific Name 


Microhigh 


Microlow 






Mean 


Range 


Mean 


Range 


Grasses (perennial): 












Indiangrass 


Sorghastrum nutans 














Brownseed paspalurn 


Paspalum plicatulum 


1.5 


0.7-2.8 


.4 


2.4-8.6 


Unknown paspalum (#10) 


Paspalum spp. 








0.1 


0.0-0.5 


Texas wintergrass 


Stipa leucotricha 


4.6 


2.8-8.1 


3.7 


1.0-3.7 


Little blueetem 


Andropogan saccharoides 














Broom s edge blueetem 


Andropogan virginicus 


3.6 


2.8-4.2 


1.3 


0.0-3.7 


Bushy bluestem 


Andropogan glomeratus 








0.1 


0.0-0.4 


Scribner's dicanthelium 


Panicum scribnerianum 


0.3 


0.0-1.1 








Dropseed species (smutgrass) 
Blue-eyed grass species 


Sporobolus spp. 
Sisyrinchium spp. 


3.7 
2.2 


2.7-4.2 
1.1-2.4 


1.3 
0.1 


0.0-20.0 
0.0-0.4 


Little barley 


Hordeum pusillum (Nutt.) 








0.7 


0.5-1.6 


Fescue 


Festuca spp. 








0.1 


0.0-0.4 


Phalaris species 


Phalaris spp. 








1.0 


0.5-2.4 


Common bermudagrass 


Cynodon spp. 








0.6 


0.0-1.9 


Unknown grass #8 




- 















15.9 




13.4 




Forbe (perennial): 












Yellow nuteedge 


Cyperrus esculentus 


8.5 


3.8-13.6 


0.9 


0.0-2.4 


Eleocharis species 


Eleocharis spp. 








8.5 


6.3-13.4 


Yellow woodsorrel 


Oxalis dill e nil 


2.3 


1.0-6.4 


3.5 


2.8-4.3 


Upright prairie coneflower 


Rudbeckia spp. 


0.9 


0.0-1.6 








Dotted gay feather 


Liatris punctata 














Button snakeroot 


Liatris pycnostachya 














Thistle species 


Cirsium spp. 














Maximilion sunflower 


Helianthus maximiliani 














Houstoni a species 


Houston! a spp. 














Indigo species 


Baptisia spp. 


1.7 


0.0-5.9 








Yellow neptunia 


Neptunia lute a 


0.6 


0.7-1.1 


0.9 


0.0-1.9 


Butterfly pea 


Centrosema virginianum 








0.1 


0.0-0.4 


Catclaw sensitivebriar 


Mimosa strigillosa 


1.9 


0.3-4.5 








Plantain species 


Plantago spp. 


0.9 


0.0-2.2 


0.3 


0.0-0.5 


False dandelion 


Pyrrhopappus spp. 


2.5 


2.5-3.4 


0.3 


0.0-1.0 


Frogfruit 


Phyla incisa 








1.8 


0.0-3.1 


False garlic 


Nothoecordum spp. 


0.3 


0,0-1.1 


0.4 


0.0-0.8 


Goldenrod 


Solidago spp. 


4.0 


1.7-8.0 


4.7 


5.3-7.3 


Green antelopehorn 


Asclepias virdiflora 





0.0-0.3 


0.4 


0.0-1.3 


Western ragweed 


Ambrosia psilostachya 


0.6 


0.0-1.41 


2.4 


2.0-2.9 


Euphorbia spp. 


Euphorbia spp. 


4.2 


3.4-4.3 


8.1 


0.0-13.4 


Eryngo species 


Eryngium spp. 








1.4 


0.5-2.8 


Unknown forb #4 




0.3 


0.0-0.7 


0.7 


0.0-1.6 


Unknown forb #6 




2.8 


0.0-5.6 


2.8 


0.0-5.3 


Unknown forb #7 




JLL 




L2. 


0.0-3.2 






32.6 




38.5 




Forbfi (annual): 












Annual broom weed 


Xanthccephalum 












s phaeroce phalli m or 


0.3 


0.0-0.7 


1.3 


1.4-2.4 




Xanthocephalum 












texanum 










Unknown annual forb #9 
















Unknown annual forb #10 
















Miscellaneous forbs 




IL5 


0.0-1.1 


<QL2 


1.0-2.8 


0.8 




1.6 








Bare Ground 




38.6 


20.0-50.0 


14.4 


5.0-25.0 


Litter 




12JL 


5.0-25.0 


22J. 


20.0-50.0 






100% 




100% 





a 3-D topographic 

surface net of topography and horizonation. 
After field transect studies were conducted, a 
site was selected for detailed study which re- 
flected the mean conditions of transects, for 
which access could be gained and for which a pit 
could be opened and remain open for about one 
week. Transect #7 (Table 4) represents the area 
chosen for the field study and is close to where 
the trench was placed in the detailed study area. 
It is also close to the mapping unit type location 
of Lake Charles clay in the Victoria County Soil 
Survey. 



Detailed Study Area 

After selecting the representative study area, 
three transects were run across multiple micro- 
highs and microlows to determine trench loca- 
tion and orientation for best light exposure. To 
locate calcareous diapirs in the center of micro- 
highs, the trench face was "V" shaped with the 
apex in about the center of the pit wall. At that 
point, the pit was about 10 off the straight line. 

Before the trench was excavated, the location 
of microhighs and microlows were sketched to 
scale and the location of the trench placed on 
this sketch (Fig. 10). The pit extended through 



WILDING, WILLIAMS, MILLER, COOK, AND ESWARAN: CLOSE INTERVAL SPATIAL VARIABILITY OF VERTISOLS 



241 




Figure 11: Distinct soil zones or polyhedrons are 
delineated by white string. 



two microhighs and two microlows. The study 
area was approximately 10 m 2 , but varied, de- 
pending on size and shape of gilgai. Study area 
boundaries were outlined for detailed vegetative 
and topographic surveys. 

Vegetative Survey 

The species composition, bare ground, and lit- 
ter were recorded for three microhighs and three 
microlows. Compositional percentages (means 
and ranges) for the plant survey are given in 
Table 5. These data were collected at the site of 
the ICOMAQ study, within a few kilometers of 
the detailed study area. The range condition of 
the study area was severely overgrazed. The 
percentage bare ground is higher and plant lit- 
ter lower on microhighs than microlows, reflect- 
ing areal percentage of apparently active cal- 
careous diapirs that rise to the surface of micro- 
highs. 

Species preference between microhigh and 
microlow positions was not strong but noted dif- 
ferences were as follows: Brown seed paspalum 
(microlow), broomsedge bluestem (microhigh), 
rattail smutgrass (microhigh), dotted blue-eyed 
grass (microhigh), yellow nutsedge (microhigh), 
Eleocharis spp. (microlow), false dandelion 
(microhigh), western ragweed (microlow), and 
Euphorbia marginhea (microlow). The lack of a 
more striking difference in species com- 
position between gilgai elements may be 
due to the fact that this site was over- 
grazed. It also may reflect the seasonal- 
ity of the vegetative survey with dy- 
namic changes in vegetative composition 
occurring with changing moisture con- 
tents. 



Topographic Survey 

Within the 10 m 2 study area, a topographic 
survey is under construction on a grid at 1 m 
intervals; recordings will be taken at closer in- 
tervals as needed. This work is still in progress. 
Future studies will include surface cracking pat- 
terns as a function of soil moisture conditions 
changing with time that will be drawn to scale 
with reference to microtopographic highs and 
lows of the study area. 

Trench Description and Sampling Scheme 

A pit was excavated along the transect se- 
lected to cross two microhighs and two mi- 
crolows. The major pit wall was slightly "V" 
shaped with the trench face 1.2 m wide, 2 m 
deep and 10 m long. A "T" leg about 1/3 the 
length of the pit face was constructed in a 
wedge-shaped design to allow better light expo- 
sure of the pit face during midday. The "T" ex- 
tended from the bottom of the pit and sloped 
upward to the surface over a length of about 5 m 
for easy access. 

Following excavation, the entire pit face was 
picked to expose horizon configuration, natural 
structural peds, and slickenside orientation. 
After careful examination of all observable fea- 
tures, nails were inserted into the face to outline 
all areas having common features such as color, 
structural arrangement, slickenside patterns, 
carbonate and ferromanganese segregations, 
and other distribution patterns. White string 
was used to connect the nails and enclose com- 
mon distinct soil zones or polyhedrons (Fig. 11). 

Seven pedons (sampling units) were selected 
for detailed sampling, which included two mi- 
crolows, two microhighs, and three intermediate 
positions. The pedons were sampled for field 
moisture content, shortly after pit excavation, 
at depths of 10, 60, 110 and 190 cm. This was to 
establish initial moisture conditions relative to 
marked differences observed in moist soil consis- 
tencies between gilgai elements (Table 6). The 
Bk horizons of microhighs had much firmer soil 
consistency than microlows, and it was hypothe- 



Table 6. Spatial variability in soil moisture content (weight 
percentage) of Lake Charles clay 


Lateral site location 


Depth 
(cm) 


Pedon #1 Pedon #2 Pedon #3 Pedon #4 Pedon #5 Pedon #6 Pedon #7 
(MH)* (Int.) (ML) (Int.) (MH) (Int.) (ML) 


10 
60 
110 
190 


26.8 
23.9 
18.5 
23.2 




27.1 
22.4 
18.3 
20.0 


28.7 
25.5 
20.9 
19.8 


25.6 
24.9 
23.3 
23.7 


23.6 
23.0 
22.6 
23.0 


28.6 
24.6 
21.5 
22.8 


32.0 
23.8 
25.9 
21.9 


* Gilgai 


position - 


MH 
Int. 
ML 


- microhigh 
- intermediate between MH and ML 
- microlow 



242 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



sized that this may be due to different mois- 
ture states. While microlows did tend to 
have slightly higher moisture contents than 
microhighs or intermediate positions, differ- 
ences were not great nor consistent. At 190 
cm, microhighs had slightly higher moisture 
retention than microlows. Differences in 
moisture content observed do not seem to be 
adequate to explain the distinctly greater co- 
hesive strength and firmness of microhigh 
subsurface structural units vs. microlow 
units. 

Detailed profile descriptions were written 
of the seven pedons. Bulk samples and bulk 
density clod samples were taken of each ho- 
rizon in all pedons. Bulk samples of distinct soil 
zones (layers) that were not collected in the 
seven pedons sampled were collected as satellite 
samples. All samples were shipped to the Na- 
tional Soil Survey Laboratory, Lincoln, Ne- 
braska, for laboratory characterization analy- 
ses, except for field soil moisture contents, which 
were determined in the Texas Soil Characteriza- 
tion Laboratory, Texas A&M University, Soil 
and Crop Sciences Department, College Station. 
Laboratory data are not yet available except for 
initial moisture status. 

Selected morphological characteristics along 
the pit face were plotted on graph paper in the 
field at a scale of 1:4. The properties were plot- 
ted from a string line placed on the level about 
25 cm above the microhighs for elevation control 
(Fig. 11). From this crossectional profile, the 
percentage of the pit face comprised of micro- 
high, intermediate and microlow gilgai ele- 
ments, with corresponding horizon percentages 



Table 7. Percentage of pit face comprised of microhigh, 
intermediate and microlow gilgai elements with corresponding 

horizon percentages of given colors. 


Gilgai 
element 


Gilgai Horizonation 


Soil colors 
(Munsell Notations) 


element (%) 

(%) 


Kind 


Microlow 
Intermediate 
Microhigh 
Total Trench 


29 72 
8 
20 
52 9 
49 
42 
19 4 
18 
78 
38 
8 
22 
32 


(A, A2, Bw) 
(Bk) 
(Bk) 
(A) 
(Bk) 
(Bk) 
(A) 
(Bk) 
(Bk) 
(A, A2, Bw, Bk) 
(A, Bk) 
(Bk) 
(Bk) 


10YR 2/1, 3/1 
10YR 4/1 to 6/1 
5Yand2.5Y6/2,7/2 
10YR 2/1, 3/1, 4/1 
10YR 4/1, 5/1 or 6/1 
5Yand2.5Y6/2, 7/2 
10YR4/1 
10YR 5/1 
5Yand2.5Y6/2,7/2 
10YE 2/1, 3/1 
10YR4/1 
10YR 5/1, 6/1 
2.5Y and 5Y 6/2, 7/2 



of given colors, was planimetered or determined 
from grid areas (Table 7). These data will be 
discussed under the next section. The crossec- 
tional profile data also were input into a Geo- 
graphical Information System (GIS, ARCINFO) 
at South Technical Center, Fort Worth, Texas, 
so morphological and analytical databases may 
be overlayed. Further, this allowed presenta- 
tion of the crossectional profile at variable 
scales. GIS reductions the crossectional profile 
and corresponding areas with similar zip pat- 
terns are presented in Figures 12 and 13. 

Summary of Field Observations of 
Sampling Site 

The site selected was an excellent example of 
the extreme variability in many Vertisol proper- 
ties corresponding to microhighs and microlows. 
We hope this study will illustrate the complexity 
of soil morphology, chemical, physical, and min- 
eralogical properties, water movement, root 



ssr 



CASE STUDY 

fV OP LAKE CHAQL.ES CLAY 




fta n it m r 



VISTtBCI It CK 



Figure 12: A reduction of the field-constructed crossectional profile using a Geographical Information System 
(ARCINFO). 



WILDING, WILLIAMS, MILLER, COOK, AND ESWARAN: CLOSE INTERVAL SPATIAL VARIABILITY OP VERTISOLS 



243 



development, and crop/vegetative adaptivity 
and response to microlow and microhigh gilgai 
topographic elements. Following are a few spe- 
cific notes and suggestions relative to this case 
study: 

Morphological Properties Associated With 
Gilgai Elements 

Striking morphological differences were ob- 
served in three gilgai positions, namely, micro- 
high, microlow, and intermediate positions. The 
two microhighs of pedons #1 and #5 had were 
narrow (30-70 cm wide) diapirs "chimneys" of 
grayish, calcareous clays extending from the 
lower Bk horizon to the surface. They did not 
appear to be due to crack infillings, but rather, 
these "tepee-shaped" structures appear to have 
been pushed or squeezed up cat steep angles 
along slickenside planes that border microlows. 
The chimneys are not always located in the cen- 
ter of the microhighs but may be offset onto the 
intermediate position. 

The lows are generally black clays with little 
or no carbonates either as nodules, soft segrega- 
tions, or disseminated carbonates. In lower Bw 
and Bk horizons, a few carbonate nodules occur 
randomly distributed throughout these horizons 
but lack soft carbonate coatings or encasements 
that are common in Bk horizons of microhigh 
and intermediate positions. It appears that 
these carbonates are undergoing dissolution 
and leaching because they are rough- surfaced, 
entirely free of soft segregation coatings, and 
exhibit clean sand grains protruding from car- 
bonate nodules. 

Between the microlows and microhighs was 
an extensive zone of calcareous, gray clay, 
darker in color than the microhigh, but lighter 



than the microlow. It contained common car- 
bonates as soft segregations, hard concretions 
encased in soft calcareous rinds, and dissemi- 
nated forms. These calcareous materials were 
commonly banded at angles corresponding to 
bordering slickenside planes and oriented to- 
wards the microhigh. Further, they were virtu- 
ally continuous with lower Bk horizons. 

In the microlow of pedon #5, the Bw and Bk 
horizons had circular crossectional, pedotubule - 
structures about 3-5 cm in diameter. They were 
oriented vertically and are quite possible krotov- 
inas from crayfish activity. 

Quantification of Gilgai Elements 

Quantitative determinations of the crossec- 
tional area of the trench face comprised of differ- 
ent gilgai elements and corresponding horizona- 
tion are given in Table 7. Over half (52%) of the 
trench was comprised of the intermediate gilgai 
position, followed by the microlow (29%) and the 
microhigh (19%). Black or very dark gray A and 
Bw horizons comprised nearly three-fourths of 
the microlow, with 28% comprised of lighter 
gray Bk horizons. 

In contrast, intermediate and microlow posi- 
tions were dominated by gray and grayish- 
brown Bk horizons (91% and 96%, respectively), 
with the remainder comprised of thin black, 
very dark gray, or dark gray A horizons. Of the 
total trench, 38% was comprised of black or very 
dark gray clays, 8% by dark gray clays, 22% by 
gray clays, and 32% by light gray and light, 
brownish-gray calcareous clays. 

Slickenside Orientation and Spacing 

Large slickenside planes tend to outline the 
microlow and microhigh gilgai elements; they 



Biciemiuii.m r im emails CUT 






189 1)0 9 8 JO Sfl )t 




Figure 13: Crossectional profile illustrating zones within trench with common soil colors. Graph drawn using 
a Geographical Information System (ARCINFO). 



244 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



form a "bowl structure" or "cone of revolution' 1 in 
the microlow where the angle of dip (plane of the 
pit face) at the apex or base was only about 15. 
Along the upward- tending diapirs of the micro- 
highs, the angle of dip was steep (60-75). In the 
trench exposure, there were 6 or 8 large slicken- 
side planes which were carefully hand-picked 
back into the face of the pit (Fig. 11). These 
planes were 2-5 cm apart as they approached 
the diapirs of the microhigh but 4-10 cm apart in 
the intermediate and microlow pedons. The 
major slickenside faces were polished and 
grooved with a network of ridges and valleys. 
The ridges had a local relief differential of 1-6 
cm higher than the troughs. The troughs were 
4-25 cm wide. Roots tended to follow slickenside 
faces but some extend through the faces. 

Starting at a microhigh of pedon #1, the slick- 
enside planes extended downward at a dip angle 
of 60-70 from the horizontal. The slickenside 
planes ran in a diagonal concave curved arc 
across the intermediate pedon #2, to 180 cm 
depth, where they flattened out to 15-20 in the 
microlow of pedon #3. The same set of major 
slickensides extended in a concave curved arc 
upward across the intermediate pedon # 4, to 63 
cm below the surface of the microhigh of pedon 
#5, where the angle of inclination was 60-75. 
Similar slickenside planes were observed be- 
tween intermediate pedon #6 and microlow pe- 
don #7. 

Perpendicular to the pit face, the angle of 
strike for major slickensides ranged from 25-60 
but was commonly 50-55. No pattern for strike 
angle was noted relative to angles of dip or gil- 
gai positions. 

Carbonate Forms and Patterns 

As previously mentioned, common to many, 
soft and hard segregations of carbonates are 
randomly distributed in lower Bk horizons of all 
pedons sampled and ubiquitous to all Bk hori- 
zons of intermediate and microhigh positions. 

In pedon #1, a zone of soft segregations ex- 
tended to within 53 cm of the surface and fol- 
lowed a downward concave arc along slickenside 
planes towards the microlow of pedon #3. In 
pedon #5, soft segregations were translated 
along slickenside planes to near the surface on 
the extreme right side of this pedon and follow 
the edge of a large slickenside plane between 
pedons #5 and #6. 

The microlows of pedons #3 and #7 are either 
leached of carbonates or have a few hard carbon- 
ate nodules but no soft segregations of carbon- 
ates. Along the outer concave "bowl" edges of 



the microlows, hard carbonate concretions occur 
in a 20-30 cm wide band in the shape of a con- 
cave upward arc. This is believed to have been 
translated upward from the lower Bk between 
two bordering slickenside planes. Adjacent to 
this zone of hard concretions is a 10-15 cm wide 
band of both hard concretions and soft segrega- 
tions of carbonates. 

Distribution of Fe-Mn Nodules 

Few to common, fine, distinct ferromanga- 
nese rounded nodules and segregations occur 
throughout the Bk horizons of all gilgai ele- 
ments. In the microlow, they also occur in lower 
Bw and Bk horizons. The nodules are most pro- 
nounced in the strongly calcareous lower Bk 
horizons. It is not clear whether these segrega- 
tions represent contemporaneous alternating 
redox states or may be relict from wetter paleo- 
environments. 

Based on current measurements of hydrology 
and redox in a Lake Charles site for the ICO- 
MAQ Study (Griffin et al., 1989), it does not 
seem reasonable that the Fe-Mn segregations 
are contemporaneous because, over the past 
year, neither the microhigh nor microlow has 
been saturated. Further, the redox states would 
need to be very low to reduce oxidized phases of 
these compounds in the presence of free carbon- 
ates. An alternative explanation is that most of 
the Fe-Mn segregations are the result of more 
strongly reducing conditions following post- 
deposition of the deltaic-fluvial Beaumont for- 
mation sediments. Such conditions would have 
been present during the period of marshy envi- 
ronment up through the dewatering stage and 
ripening of these sediments. 

Future Aspects 

Close-interval microvariability in Vertisols 
that corresponds to gilgai elements, oscillatory 
horizonation, and contorted or interrupted 
zones has received inappropriate attention. To 
better elucidate soil genesis, classification, man- 
agement, use, and distribution of Vertisols, the 
following areas justify increased research em- 
phasis: sampling schemes, hydrological behav- 
ior, cracking patterns, stress/strain/shear fail- 
ure, root distribution patterns, soil chemistry/ 
fertility properties, and plant distribution/re- 
sponse relationships. These databases should 
be assembled over a broad geographical area 
considering macro/micro-climate, parent mate- 
rial, vegetation, soil management, and land use 
variables. Such knowledge will serve as the 



WILDING, WILLIAMS, MILLER, COOK, AND ESWARAN: CLOSE INTERVAL SPATIAL VARIABILITY OF VERTISOLS 



245 



foundation for reevaluating the adequacy of the 
pedon concept for Vertisols, sampling schemes, 
engineering qualities, pollution hazards, cul- 
tural practices, and technology transfer under 
mechanized and subsistence agriculture. 

Sampling Schemes 

The sampling design used in this study is an 
approach to better represent and quantify the 
spatial distribution of properties. It suggests 
that positions intermediate between microhighs 
and microlows comprise a significant landscape 
component. In future studies, the trench should 
be cut perpendicular to the major face to allow 
better visualization and sampling of spatial 
variability in the third dimension, thus fostering 
a 3-D concept of spatial variability. 

In field observations of core samples, the 
sampling scheme as a 1 m grid, with appropri- 
ate satellite observations at shorter intervals, 
would allow the most times and cost-effective 
method to quantify morphological variability. 
Such a grid should be accompanied by elevation 
control so computer-generated topographic sur- 
face nets can be constructed for horizons and 
features of interest. The grid should be of suffi- 
cient size to include a minimum of three micro- 
highs and three microlows. 

Proposed Revision in Pedon Concept 

The concept of a pedon as a sampling unit 
should be reappraised and revised to encompass 
the variability within an area no greater than 1 
m 2 . In Vertisols, such an area will include sig- 
nificant variability in some cases and relatively 
little in others. Such a sampling scheme will 
focus attention on variable topographic and 
horizonation elements within the landscape sys- 
tem and not bias the data towards microlows. 
This concept would also bring sampling design 
into conformance with other soils with less vari- 
able attributes. Current state of soil behavior 
knowledge demonstrates the importance of veri- 
fying spatial variability in the physical, chemi- 
cal, and biological properties of these soils over 
short lateral distances. The revised concept 
would further enhance such a microvariability 
knowledge base. 

If such a pedon revision were accepted, then 
the mapping and classification of Vertisols 
would not be so dependent on the 7 m cycle of 
gilgai elements. The classification would be 
based on the distribution of attributes found 
within the mapping unit and or their pattern of 
occurrence. In many cases, a complex of two or 



three soil conditions would be recognized, based 
upon verified microvariability and the impor- 
tance of these conditions to use and manage- 
ment. In many Vertisols, insufficient variability 
may occur within the polypedon area to justify a 
complex; in this case, the soil would be mapped 
as a consociation. 

Such a revision also would be consistent with 
the definition of the pedon as a sampling unit 
and the polypedon as the classification unit with 
landform expression. The pedon would be stan- 
dardized so it was of fixed dimensions and not 
variable in response to diapir or gilgai periodici- 
ties. The latter generates ambiguity in pedon 
definition and application for soil classification. 

Hydrological Behavior 

Water and chemical transfer through Verti- 
sols has been studied extensively (Ritchie et al., 
1972; Kissel et al., 1973; Bouma, 1988) but rela- 
tively little information is available to quantify 
soil moisture relations and transfer between gil- 
gai topographic elements (Spotts, 1974). For 
example, what is the saturated and unsaturated 
hydraulic conductivity between these mi- 
crosites, what influence do slickenside planes 
have on these hydraulic parameters, and what 
is the relative flux of water and chemical con- 
stituents between microlows and microhighs? 

Monitoring saturated and near-saturated soil 
moisture conditions between microlows and 
microhighs has been initiated through an ICO- 
MAQ Study in Texas (Griffen et al., 1989). Pie- 
zometers, tensiometers, and boreholes have 
been replicated in a series of microhigh and mi- 
crolow sites of the Lake Charles clay (Typic Pel- 
luderts). 

Seasonal influences on runofi/recharge/dis- 
charge characteristics and evapotranspiration 
that may be influenced by cracking, rooting, and 
macrostructural properties are also of vital 
interest to plant distribution patterns and re- 
sponses to hydrological behavior for irrigation 
and drainage (Bouma, 1988). Likewise, such 
dynamics influence nutrient availability, leach- 
ing losses, mineralization, reduction, and immo- 
bilization of plant nutrients (Kissel et al., 1973; 
Wilding and Rehage, 1985). Such knowledge 
has vast agronomic and engineering implica- 
tions but to date is not available. 

Soil Fertility Relationships 

Plant response due to spatial variability in 
Vertisols is not well partitioned to physical, 
chemical, and biological effects or interactions. 



246 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



In Texas, few accounts are available showing 
that crop response is variable from microhigh to 
microlow gilgai positions, though native vegeta- 
tion frequently corresponds to these microsite 
environmental habitats. It is apparent from 
Australian research (Russell et al., 1967; Stace 
et al., 1968; Thompson and Beckmann, 1982) 
that crop responses are well correlated with dif- 
ferences in gilgai soil properties. It would ap- 
pear that the interactive effect between soil 
moisture patterns and soil chemistry/fertility 
patterns should be further appraised. 

In Texas, on Vertisols cropped to cotton, at- 
tempts were made to relate gilgai patterns to 
the incidence of cotton root rot (Phymatot- 
richum omnivorum), but this research did not 
verify such a correlation. In Vertisols of the Gulf 
Coast Prairie region of Texas, Fe chlorosis pat- 
terns appear to be related to gilgai spatial vari- 
ability, but this hypothesis is untested. 

Soil fertility relationships to gilgai offer an- 
other fruitful area for research in the U.S, espe- 
cially considering the ability to apply differen- 
tial rates of water, chemicals, herbicides, and 
pesticides with computer-controlled, real-time 
sensing capabilities. 

Stress/Strain/Shear Failure Relationships 

Until stress/strain/shear failure interactions 
are better verified as related to changes in soil 
moisture patterns of gilgai elements, a more 
comprehensive knowledge of Vertisol genesis, 
management, and use implications is precluded. 
For example, we do not have a verified database 
on extent of elevation changes, expression of 
slickenside failure zones, soil strength attrib- 
utes, or cracking patterns between gilgai mi- 
crosites. There are few studies of the 3-D shear 
failure structural analysis in unsaturated soil 
zones (Knight, 1980). These soil mechanical 
properties are just as germane to agronomic 
practices and crop management as to construc- 
tion engineers. A better verification of these 
properties would enhance our knowledge of irri- 
gation frequency, cracking dynamics, pattern 
and spacing of gilgai, soil tillage attributes, and 
nutrient use efficiency. Future studies of this 
nature should be initiated only after close coor- 
dination, liaison, and collaboration with civil 
and agricultural engineers with expertise in soil 
mechanics. 



References 

Ahmad, N., 1983. Vertisols. In L.P. Wilding, N.E. Smeck 
and G.F. Hall (eds.). Pedogenesis and Soil Taxonomy II. 
The Soil Orders. Developments in Soil Science IIB. 
Elsevier Publ. Co., Amsterdam, pp. 91-123. 

Aitchison, G.D. and J.W. Holmes, 1953. Aspects of swel- 
ling in the soil profile. Aust. J. Appl. Sci., 4:244-259. 

Beckmann, G.G., G.D. Hubble and C.H. Thompson, 1970. 
Gilgai forms, distribution and soil relationships in 
north-eastern Australia. In Proceedings of Symposium 
on Soils and Earth Structures in Arid Climates, 
Sydney, pp. 116-121. 

Bouma, J. and J. Loveday, 1988. Characterizing soil water 
regimes in swelling clay soils. In Vertisols: Their Dis- 
tribution, Properties, Classification and Management, 
Texas A&M University Press, College Station, Texas. 

Bouma, J. and P.A.C. Raats, 1984. Proceedings of the ISSS 
Symposium on Water and Solute Movement in Heavy 
Clay Soils. ILRI Publication No. 37, Wageningen, The 
Netherlands. 

Dixon, J.B., 1982. Mineralogy of Vertisols. In Vertisols 
and Rice Soils of the Tropics, Symposia Papers II, 12th 
ICSS, New Delhi, pp. 48-59. 

Dudal, R. and H. Eswaran, 1988. Distribution, properties 
and classification of Vertisols. ]n L.P. Wilding and R. 
Puentes (eds.), Vertisols: Properties, Classification 
and Management. Texas A&M University Press, pp. 1. 

Dudal, R., 1965. Dark clay soils of tropical and subtropical 
regions. FAO Agricultural Development Paper No. 83, 
Rome. 

FAO, 1983. Proceedings of the 5th Meting of the Eastern 
African Subcommittee for Soil Correlation and Land 
Evaluation. Wad Medani, Sudan. World Soil Resource 
Report No. 56, Rome. 

Hallsworth, E.G., 1968. The gilgai phenomenon. In H.C.T. 
Stace, G.D. Hubble, R. Brewer, K.H. Northcote, J.R. 
Sleeman, M.J. Mulcahy and E.G. Hallsworth (eds.). A 
Handbook of Australian Soils. Rellim Tech. PubL, 
Glenside, South Australia. 

Hallsworth, E.G., G.K. Robertson and F.R. Gibbons, 1955. 
Studies in pedogenesis in New South Wales, 7. The gil- 
gai soils. J. Soil Sci., 6:1-32. 

Harris, S.A., 1959. The classification of gilgaied soils: 
some evidences from northern Iraq. J. Soil Sci., 10:27- 
33. 

Harris, S.A., 1958. The gilgaied and bad structured soils of 
Central Iraq. J. Soil Sci., 9:169-185. 

Hubble, G.D., R.F. Isbell and K.H. Northcote, 1983. Fea- 
tures of Australian soils. In Soils: an Australian View- 
point. Division of Soils, CSIRO, Melbourne, pp. 17-47. 

IBSRAM (International Board for Soil Research and Man- 
agement Inc.), 1987. Management of Vertisols under 
semi-arid conditions. Proceedings of the First regional 
Seminar on Management of Vertisols under Semi-Arid 
Conditions, Nairobi, Kenya, 1-6 December 1986. 

INWOSS, 1988. Transactions of the International Work- 
shop on Swell-Shrink Soils, Nagpur, India. 

ISSS, 1982. Vertisols and Rice Soils of the Tropics. Trans- 
actions, Symposium Papers II, 12th International Con- 
gress of Soil Science, New Delhi. 

Kissel, D.A., J.T. Ritchie and E. Burnett, 1973. Chloride 
movement in undisturbed swelling clay soil. Soil Sci, 
Soc.Am. Proc. 37:21-4. 



WILDING, WILLIAMS, MILLER, COOK, AND ESWARAN: CLOSE INTERVAL SPATIAL VARIABILITY OP VERTISOLS 



247 



Knight, M.J. 1980. Structural analysis and mechanical 
origins of gilgai at Boorook, Victoria, Australia. Geod- 
erma 23:245-283. 

McGarity, J.W., E.H. Hoult and H.B. So, 1984. The prop- 
erties and utilization of cracking clay soils. Reviews in 
Rural Science, No. 5, University of New England, Armi- 
dale, Australia. 

Newman, A.L., 1986. Vertisols in Texas: some comments. 
USDA-SCS, Temple, Texas (Unpublished mimeo- 
graphed draft.), pp. 206. 

Paton, T.R., 1974. Origin and terminology for gilgai in 
Australia. Geoderma, 11:221-242. 

Prescott, J.A., 1931. The soils of Australia in relation to 
climate, CSIRO Bull. 52. 

Probert, M.E., I.F. Fergus, B.J. Bridge, D. McGarry, C.H. 
Thompson and J.S. Russell, 1987. The Properties and 
Management of Vertisols, CSIRO, CAB International, 
Wallingford, U.K 

Ritchie, J.T., D.E. Kissel and E. Burnett. 1972. Water 
movement in undisturbed swelling clay soil. Soil Sci. 
Soc. Am. Proc. 36:874-879. 

Russell, J.S., A.W. Moore and J.E. Coaldrake, 1967. Rela- 
tionships between subtropical semiarid forest of Acacia 
Harpophylla (Brigalow), microrelief, and chemical 
properties of associated gilgai soil. Aust. J. Bot., 
15:481-98. 

Soil Survey Staff, 1975. Soil Taxonomy. Soil Conservation 
Service, USDA, Handbook 436, 754 pp. 

Soil Survey Administration, 1985. Taxonomy and man- 
agement of Vertisols and Aridisols. Proceedings of the 
5th International Soil Classification Workshop, Sudan, 
2-11 November, 1982. 

Spotts, J.W., 1974. The role of water in gilgai formation. 
Ph.D. Dissertation, Texas A&M University, College 
Station, Texas. 

Stace, H.C.T., G.D. Hubble, R. Brewer, K.H. Northcote, 
J.R. Sleeman, M.J. Mulcahy, and E.G. Hallsworth, 
1968. A Handbook of Australian Soils. Rellim Tech. 
Publ., Glenside, South Australia. 



Stirk, G.B., 1954. Some aspects of soil shrinkage and the 
effect of cracking upon water entry into the soil. Aust. 
J. Agric. Res., 5:279-290. 

Thompson, C.H. and G.G. Beckmann, 1982. Gilgai in 
Australian black earths and some of its effects on 
plants. Trop. Agric., 59:149-156. 

Thompson, C.H. and G.G. Beckmann, 1959. Soils and land 
use in the Toowoomba Area, Darling Downs, Queen- 
sland. CSIRO, Aust. Div. Soils, Soils and Land Use 
Series No. 28. 

Tucker, R.J., R.C. McDonald and M.D. Godwin, 1989. 
Soils of the Magoa River, Emerald irrigation area, 
Queensland. Land Resources Bull., Queensland Dept. 
of Primary Industries, Brisbane, Queensland, Austra- 
lia. (In Preparation) 

Wilding, L.P., 1985. Genesis of Vertisols. Proceedings of 
Fifth International Soil Classification Workshop, Su- 
dan, 2-11 November 1982. 

Wilding, L.P. and R. Puentes, (eds.), 1988. Vertisols: Their 
Distribution, Properties, Classification and Manage- 
ment., Texas A&M University Press. 

Wilding, L.P. and J.A. Rehage. 1985. Pedogenesis of soils 
with aquic moisture regimes. Wetland Soils: Charac- 
terization, Classification, and Utilization. Proceedings 
of International Workshop. IRRI, Los Banos, Philip- 
pines, pp. 139-157. 

Wilding, L.P. and D. Tessier, 1988. Genesis of Vertisols: 
shrink-swell phenomena.In L.P. Wilding and R. Puen- 
tes (eds.), Vertisols: Their Distribution, Properties, 
Classification and Management, Texas A&M Univer- 
sity Press. 

Yaalon, D.H. and D. Kalmar, 1978. Dynamics of cracking 
and swelling clay soils: displacement of skeleton 
grains, optimum depth of slickensides, and rate of in- 
tra-pedonic turbation. Earth Surface Processes, 3:31- 
42. 

Yule, D.B. and J.T. Ritchie, 1980. Soil shrinkage relation- 
ships of Texas Vertisols, I: small cores. Soil Sci. Soc. 
Am. J. 44:1285-1291. 



248 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



Geomorpfaology of Cold Deserts 
Robert C. Palmquist 1 

Abstract 

The cold desert is one in which at least one month has a mean tempera- 
ture below O'C so that frost action is possible. Most cold deserts occur in the 
mid latitudes as a result of continentality and/or orographic affects. The 
typical cold desert is of the mountain-and-basin type, wherein uplands of 
resistant rocks rise above a basin of less resistent rocks. These deserts are 
often tectonically unstable. The altitudinal differences give rise to through- 
flowing streams heading in the glaciers, snow fields, or forests of the more 
moist mountains. 

The typical landforms of the cold desert are the pediment, stream ter- 
race, alluvial fan, lake basin, and badlands. The repeated climatic and tec- 
tonic changes of the cold desert regions lead to sequences of relict land- 
forms rising above small areas of active landscape development. The most 
common relict landforms are the pediment and terrace, each with a stable 
hillslope rising above it. Most of these relict surfaces increase in age with 
increasing height in multiples of 100,000 years, reflecting control by the cli- 
matic cycles of the Late Cenozoic. 



Introduction 

A desert is any naturally occuring area of 
sparse vegetation; as such there are lithological 
deserts on bare rock outcrops or gravelly sub- 
strates and climatic deserts in arid or very cold 
regions. This paper deals with deserts which 
are both arid and cold, although it avoids the 
extremely cold deserts of the polar regions. Cold 
deserts have at least one month with a mean air 
temperature of 0C which is indicative of severe 
winters and frozen ground (Meigs, 1953). They 
are located in the middle latitudes and in the 
higher elevations of the low latitudes. In the 
mid latitudes, the aridity results from a combi- 
nation of continentality, such as the deserts of 
Mongolia and the southern USSR, and from oro- 
graphic effects, i.e. rain shadow deserts, as occur 
in North America. Bordering the deserts are 
extensive areas of semiarid climates which have 
similar landforms and soils (Figure 1). 

Another important distinction between types 
of deserts is that of tectonism, i.e., the presence 
or absence of recent or ongoing geologic defor- 
mation. Most of the low-latitude deserts occur 
in areas of extreme stability of long duration. 
These shield or platform deserts are characteris- 
tic of Africa, Arabia, Australia, and India and 
are characterized by gently sloping erosional 
surfaces extending from granitic hills or table 
lands developed on either horizontal strata or 
weathering crusts of calcrete or silcrete. 

In contrast, the mid-latitude deserts gener- 
ally occur in regions of more active deformation 



with mountains rising above basins. These 
mountain-and-basin deserts are typified by the 
Basin and Ranges of Utah and Nevada and the 
intermontane basins of Wyoming and Colorado. 
Here, the elevational contrasts lead to climatic 
contrasts with more moist uplands contributing 
water from snow melt, or glaciers, or increased 
precipitation to streams which drain into or 
across the lowland deserts. 

Another important distinction between the 
two types of deserts is that the basin-and-range 
desert has uplands composed of resistant rocks 
and basins composed of less resistant rocks. The 
combination of active or recent deformation and 
lithologic contrasts leads to a characteristic set 
of landforms. 

This short paper attempts to make four 
points. First, the climate of all deserts has re- 
peatedly changed with the climatic cycles of the 
Late Cenozoic. Second, the weathering proc- 
esses of warm and cold deserts are similar ex- 
cept for the addition of frost action in cold des- 
erts. Third, the variation in degrees of tectonic 
and climatic stability leads to the juxtaposition 
of relict and active landforms and to sequential 
landform development, which brings us to the 
fourth characteristic: most aridosols are likely 
to have developed under multiple climatic regi- 
mens. I will be unabashedly provincial in this 
paper and draw upon experiences and examples 
from the cold deserts and semiarid regions of the 
northern United States. This is an area with 
which I am familiar and through which you will 
be traveling on this trip. 



Northwest College, Powell, WY, 82435 



PALMQUIST: GEOMORPHOLOGY op COLD DESERTS 



249 




Figure 1. Location of deserts and semi- arid regions in 
North America as defined by vegetation. Cold des- 
erts and semi-arid regions occur north of the 30F 
isocline and warm deserts and semi-arid regions 
occur to the south of the 40"F isocline. A transition 
zone lies between. (Complied from several 
sources.) 



cent mountains. The consequent variations in 
the load/discharge ratios of the mountain 
streams lead to the development of relict pedi- 
ments, terraces, and alluvial fans. 

Although the nature of the climatic change 
will vary with the particular setting of each des- 
ert, the timing of the changes is well described 
by the isotopic marine record. In this record can 
be read both the variations in mean sea surface 
temperatures and the volume of continental ice. 
The record indicates that the global climate has 
oscillated between colder and warmer (Figure 2) 
with approximately a 100,000-year frequency 
for the last 600,000 years and with a shorter, 
less regular frequency between 600,000 years 
and 2.5 million years, when the first evidence of 
extensive continental glaciations appears. Dur- 
ing each cycle, the climate of an area varied 
greatly, as exemplified by the Fairbridge Cycle 
for mid latitudes (Figure 3a). The soils and 
landforms of the deserts reflect these climatic 
oscillations. The common perception that gla- 
cial climates were cooler and wetter than inter- 
glacial climates is an incorrect generalization. 
In the mid latitudes during maximal glaciation, 
the climate was colder but with about the same 
precipitation as at present (Figure 3b). 

The cold deserts contain evidence of perigla- 
cial features such as ice-wedge casts, ground- 
wedges, and Mima-like mounds which are inter- 
preted to have developed under arid to semi- 



Climatic Variations 

The Late Cenozoic is char- 
acterized by repeated oscilla- 
tions between colder and 
warmer and/or wetter and 
drier climates. The nature of 
these fluctuations is depend- 
ent upon the location of the 
desert. For instance, the 
warm deserts were subjected 
to an alternation of arid and 
savanna climates, which 
leads to the distinctive in- 
selberg landscape developed 
by alternating chemical 
weathering and regolith 
stripping (Thomas, 1974). In 
contrast, the cold deserts 
have undergone an alterna- 
tion of colder and warmer cli- 
mates, which leads to the ex- 
pansion and contraction of 
alpine glaciers in the adja- 



6- 



FREQUENCY 



(NO./, 



STABILITY EPISODES 
TERRACE SEQUENCES 



'SOky) 



DEEP-FRESH 

DEEP-BRACKISH 

SHALLOW- BRACKISH 

SHALLOW-SALINE 

DRY- PLAYA 

2.0 

PDB i'. 
0.5- 





** N \ , \ \ i \'\,'\, > h /\ \ 

L_ r^' \ \ F ^ : ' / V : J( ^l 1 .^\ ^.\ Moving average 




1.0 A 
AGE(X10 6 YRS) 



Figure 2. Geologic evidence of long term cyclic behavior. Bottom: The isotopic 
record of marine core V28-239; even numbered stages represent intergla- 
cials and odd numbered stage represent glacials (after Shackle ton and 
Opdyke, 1976). Middle: Variations in salinity of Searles Lake, California 
based upon the sedimentology of a deep core (after Smith, 1984). Top: Fre- 
quency distribution of terrace ages along rivers in the Rocky Mountain 
West as estimated by the procedures in Palmquist (1983). 



250 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



arid, windswept environments 10 to 13 C 
colder than today (about 5C). (Mears, 1987). 
Ice wedge casts require a moist climate, such as 
that of northern Alaska, and ground wedges a 
drier climate, such as that of Anarctica. 

Weathering Processes 

The cold deserts are characterized by both 
physical and chemical weathering. The desert 
floors are littered with fragments of shattered 
rocks and sand from the granular disintegration 
of rocks. The shattering is the result of both 
insolation weathering and frost shattering, 
whereas the granular disintegration results 
from hydration, hydrolysis, and salt weathering 
perhaps accentuated by freezing. Chemical 
weathering processes such as oxidation, hy- 
drolysis, solution, and leaching must exist or 
have existed in arid regions to explain such 
uniquely arid features as cavernously weath- 
ered boulders, duricrusts (silcretes and calcre- 
tes), and aridosols. Given the climatic vari- 
ations of the Late Cenozoic, any weathering 
process could have operated at some time during 
the history of any desert landform or soil older 
than a few thousand years. 

Landforms in Space and Time 

The major landforms of the cold desert are the 
pediment, alluvial fan, stream terrace, and plu- 
vial lake basin. These piedmont features border 
hillslopes developed on either the resistent rocks 
of the mountain flanks or the badlands of the 
basinal uplands and reflect differences in lithol- 
ogy and/or tectonic stability and climatic vari- 
ations. Their present activity influences the 
distribution of aridosols, inceptisols, and enti- 
sols. 

Pediments 

The major erosional feature of the piedmont is 
the pediment (Figure 4) which is best defined as 
a bedrock erosional surface of low relief extend- 
ing basinward from a backing hillslope from 
which it is separated by an abrupt change of 
gradient (freely modified from Marbutt, 1977). 
Most pediments in cold deserts are mantled 
with debris from the adjacent uplands and are 
termed mantled pediments. An exception is the 
relatively small badlands pediment, which is 
free of a mantle and is properly termed a pedi- 
ment. Mantled pediments are generally concave 
up in longitudinal profile and have a steeper 
gradient than stone-free pediments. In trans- 



MD -LATITUDE GLOBAL 

CLIMATE TEMPERATURE TERMINOLOGY 



o 
o 

o 
o 
o 



WET -WARM 

4 

DRY -COOL > .. 
WET-MILDJ alt 

DRY -COLD 
MOIST-MILD 

VERY DRY- 
VERY COLD ,# 

WET -COL 




INTERGLACIAL 



GLACIAL 



COLD DESERTS 




J L 



30" N 



60" N 



Precip > evap 
(humid) 



Precip < evap 
(arid) 



90 N 



Figure 3. Top: Fairbridge Cycle of climatic change 
within a single interglacial-glacial cycle as defined 
by one odd and even numbered isotopic stage (Fig. 
2). Cycle based upon a mid-latitude, maritime cli- 
mate and illustrates the complex variation in mois- 
ture and temperature (simplified from Fairbridge, 
1972). Bottom: Effective variation of precipitation 
in the northern hemisphere between an interglacial 
(today) and glacial illustrating the latidutinal 
changes of the cold deserts during a glacial cycle. 
Soild line represents present conditions; broken 
line indicates glacial conditions. The vertical lines 
mark the latidutinal range of present day cold des- 
erts. Diagram does not consider the affects of rain 
shadow in the latidutinal distribution of desertic 
conditions. (Modified from Flohn, 1953.) 



verse section, pediments undulate with low in- 
terfluves separating drainage lines. The sedi- 
ment on the interfluves may be less well sorted 
than that along the drainages. 

All hypotheses to explain the genesis of pedi- 
ments demand that they form in an environ- 
ment with a stable baselevel. This limitation 
restricts the formation of large pediments to 
areas of tectonic and climatic stability. Given 
the climatic fluctuations of the Late Cenozoic, it 
is not surprising that most pediments in cold 
deserts are relict and are presently being dis- 
sected. In my experience, the youngest relict 
pediments have ages which place them in one of 
the last three interglacials, that is, in isotopic 
stage 5, 7, or 9 (Figure 2), with the youngest 
mountain front pediments generally falling into 



PALMQUIST: GEOMORPHOLOGY OF COLD DESERTS 



251 



stage 7 or 9 and the youngest badlands or basin 
pediments falling into stages 5 or 7. 

Badlands pediments which develop rapidly 
through slope retreat in the soft clays may be 
either active or relict. In any extensive bad- 
lands area, most of the larger pediments are rel- 
ict and stand above the modern drainage. Only 
the smaller pediments that border and are 
graded to the modern drainage are actively 
forming. The active pediments have developed 
since the end of the Pleistocene about 11,000 
years ago, and most are related to the end of the 
Altithermal about 3,000 years ago. The inactive 
pediments are older than stage 5 and are bor- 
dered by inactive back slopes. 

Alluvial Fan 

The alluvial fan is found in areas of active 
stream incision which may result from, active 
tectonism, recent climatic change, or master 
stream incision. Alluvial fans are constructed 
from deposition either from streams or debris 
flows or a combination of the two. Generally, 
the more arid the region, the more dominant the 
deposition from debris flows. Debris flows result 
from the supersaturation of the stream by de- 
bris eroded from the valley floors of the uplands. 
On the more permeable fan, the interstitial wa- 
ter is lost and the lobate flow comes to rest. 
Generally the flow deposit is coarse-grained and 
poorly sorted and is confined to the channel or 
its immediate environs. Alluvial deposits are 
more heterogeneous and consist of finer- grained 
overbank deposits cut by linear sandy or grav- 
elly channel deposits. 

Most alluvial fans are concurrently being 
built by deposition while at the same time they 
are being destroyed by erosion (Figure 5). The 
active portion of the fan is characterized by ex- 
posed silts, sands, and gravel deposited from 
debris flows or streams. This portion of the fan 
is generally at a slightly higher elevation than 
the stable or actively degrading portions. The 
stable portions are mantled by a one-stone thick 
layer of gravel, desert pavement, which is lag 
from erosion by wind or overland flow of runoff. 
The longer that the surface has been stable, the 
better developed the pavement and its dark 
coating of manganese dioxide, desert varnish. 
The portions of the fan being degraded contain 
numerous gullies which head upon the fan sur- 
face. 

This complex activity occurs within one de- 
positional cycle. Given the repeated climatic 
changes of the Late Cenozoic and the repeated 



MANTLED PEDIMENT 




Figure 4. Block diagram illustrating the features of a 
mantled pediment. The thick alluvium at its base 
either may be the product of deposition within a 
closed basin or be shallower in depth and the 
product of deposition of stable stream in an open 
basin. (Modified from Packard, 1974, as presented 
in Selby 1982.) 



alternations of depositional and erosional cycles 
engendered by it, any fan surface is likely to 
have a wide variety of Aridisols and Entisols. 

Stream Terraces 

Stream terraces are flood plains abandoned by 
the rapid incision of the channel. As such, they 
stand above the modern flood plain as a bench 
and testify to the increase in erosional power of 
the channel at some time in the past. Terraces 
are common in the mountain-and-basin deserts 
because an increase in erosional power will re- 
sult from (1) increased gradient - perhaps as the 
result of tectonic tilting or incision of the master 
stream, (2) increased discharge - probably the 
result of a climatic change, or (3) decreased load 
- most likely the product of a climatic change. 
The Late Cenozoic variations in glacial activity 
caused the mountain-bred streams to have re- 
peated variations in their load/discharge ratios, 
which lead to alternating episodes of stability 
and incision. 

Most available ages for geomorphic features 
suggest that glacial-age stability or deposition 
alternated with interglacial-age incision, as ex- 
emplified by the terrace sequences along the 
rivers in the Rocky Mountain West (Figure 2). 
The result of such alternations is a chronose- 
quence grading from Typic Calciorthids on the 
lowest terrace to Ustollic Haplargids on the 
higher terraces but with gradations in soil de- 
velopment not reflected in the terminology 
changes (Reheis, 1987a, 1987b). Any hillslope, 



252 



SIXTH INTERNATIONA! Soil CLASSIFICATION WORKSHOP 



pediment, or alluvial fan graded to 
a terrace is likewise relict and with 
an equivalently developed soil. 

Desert Lake Basins 

Within many of the closed basins 
of the mountain-and-basin deserts 
of Utah and Nevada, fresh water 
lakes developed during the cooler 
glacial stages. Some of these lakes, 
such as glacial Lake Bonneville at 
Salt Lake City, were very extensive 
and deep and developed around 
their shorelines deltas, beaches, 
wave-cut cliffs and benches. These 
lakes varied from fresh water to 
highly saline, depending upon the 
relationship between the elevation 
of the water surface and that of the 
outlet. This relationship is best 
determined from the composition of 
the sediments (Figure 2). Within 
each cycle, as the water level fell, a 
series of shorelines were exposed, 
on which have developed a complex 
chronosequence. 

Desert Hillslopes 

Bordering the piedmont features 
are hillslopes which range from the 
debris mantled or bare rock slopes 
of the mountain front to the bad- 
lands slopes of the basins. The 
slopes vary greatly in their vegeta- 
tion cover and stability. Many of 
the mountain front slopes are ei- 
ther exposed limestones, sand- 
stones, or granites, with scattered 
shrubs growing from joints, or de- 
bris-mantled with scattered shrubs 
and grasses. Most of these slopes 
are stable and lead down to relict 
pediments at their bases. 

Within the basins, hillslopes flank relict pedi- 
ments, alluvial fans, and terraces or form the 
intricately dissected mosaic of the badlands. 
Those slopes flanking relict piedmont features 
are usually gentler, more subdued, and more 
vegetated than the slopes of the badlands and 
have vegetated drainages at their base. It is 
difficult to escape the conclusion that they are 
likewise relict and that the only active slopes 
are those of the badlands. 

Badlands occur in three situations. The most 
common is at the head of actively incising tribu- 
taries leading to a through flowing master 



Present channel 
Older channel 
Youngest fan surface 
Older fan surface 
Oldest fan surface 
Debris flow levees 
Earth flow 

* Fault scarp 
-*- Minor wash 







Figure 5. Variation in surface ages and processes on the Shadow 
Rock Fan, California illustrating the localized nature of deposi- 
tion. The older surfaces will have increasingly well developed 
desert pavement, desert varnish, and aridosols (after Hooke, 
1967). 



stream. Generally these tributaries have 
breached terraces and alluvial fans to dissect 
portions of the otherwise stable slopes rising 
above these features. The second situation is at 
the head waters of major basinal streams where 
steep slopes exist because of cap rocks, and the 
third is where a stream of any size is undercut- 
ting the valley wall at the outside of a meander. 
The distribution of badlands indicates that 
the dissection of a hillslope is controlled by the 
activity of its base. A hillslope rising above a 
terrace, relict pediment, or relict alluvial fan is 
stable, whereas a slope rising above an incising 
base is actively being degraded. In my experi- 



PALMQUIST: GEOMORPHOLOGY OF COLD DESERTS 



253 



ence, few degrading hillslopes have stable bases 
upon which colluvial slopes or small alluvial 
fans are being constructed. The implication of 
this observation is that the present arid envi- 
ronment is incapable of slope dissection unless 
aided by slope oversteepening through either 
basal erosion or height exaggeration resulting 
from a cap rock. 

Spatial and Temporal Variability 

As has been pointed out in the previous para- 
graphs, the climate and/or tectonism of the cold 
deserts has undergone repeated changes 
throughout the Late Cenozoic. The result is 
that these deserts are a complex mosaic of relict 
and active surfaces. The relict surfaces, such as 
pediment and stable backslope, rise as a series 
of benches above the modern drainage with ages 
that generally increase with height in multiples 
of 100,000 years. Between the sets of stable 
surfaces are actively degrading badlands, per- 
haps containing landslides, along the margins of 
and in the headwaters of incising streams. All of 
the incising streams are tributary to a through- 
flowing, mountain-bred, master stream. 

The result of this spatial juxtaposition of var- 
ible-age surfaces is a mosaic of soils which vary 
in a complex manner. The variation in soils re- 
sults from differences in parent materials, topo- 
graphic position, microclimate and vegetation, 
and age. In many respects within the cold des- 
ert, age is the most important of all of the soil 
forming factors. Age influences not only the 
degree of development of the soils but also the 
number of climatic cycles through which the soil 
has developed. Alternation of climatic cycles 
leads to such features as carbonate-engulfed, 
argillic horizons and frost shattered, petrocalcic 
horizons, corrosion surfaces within carbonate 
nodules, and the stripped and redeposited epipe- 
don of the Paleargid. All of these characteristics 
require a fluctuation between either colder and 
warmer climates and/or wetter and drier cli- 
mates. 

One might properly ask the questions 
Which characteristics of the cold desert aridosol 
are the product of the environment in which 



they are presently found? What landforms of 
the cold deserts were formed under the present 
environment? Did the arid characteristics of the 
cold desert aridosol form under interglacial-age 
aridity or glacial-age aridity? 

The cold desert and its soils are a complex 
ensemble with distinctive characteristics 
formed by its unique history of climatic and tec- 
tonic variations. Its similarities to the warm 
deserts are obvious, but their different geologic, 
climatic, and hence pedogenic histories make 
them different entities. 

References Cited 

Denny, C.S. 1967. Fans and Pediments. Am. J. Sci. 
265:81-105. 

Fairbridge, R.W. 1982. Climatology of a glacial cycle. 
Quat. Res. 2:283-302. 

Flohn, H. 1953. Studien uber die atmosphariische 
Zirkulation in der letzten Eiszeit. Erdkunde 7:266-275. 

Hooke, R. leB. 1967. Processes on arid-region alluvial 
fans. J. Geol. 5:438-60. 

Mabbutt, J.A. 977. Desert Landforms. MIT Press, Cambr- 
idge. 

Mears, B., Jr. 1987. Late Pleistocene periglacial wedge 
sites in Wyoming: an illustrated compendium. Menoir 
No. 3. The Geological Survey of Wyoming, Laramie. 

Meigs, P. 1953. World distribution of arid and semi-arid 
homoclimates. Rev. of Res. on Arid Zone Hydrology. 
Arid Zone Programme 1. UNESCO, Paris. 203-210. 

Palmquist, R.C. 1983. Terrace chronologies in the Bighorn 
Basin, Wyoming. 34th Ann. Field Conf., Wy. Geol. As- 
soc. Guidebook. 217-231. 

Reheis, M.C. 1987a. Soils in granitic alluvium in humid 
and semiarid climates along Rock Creek, Carbon 
County, Montana. US Geol. Sur. Bull 1590. 

Reheis, M.C. 1987b. Climatic implications of alternating 
clay and carbonate formation in semiarid soils of south- 
central Montana. Quat. Res. 27:270-282. 

Shackleton, N. J. and N. D. Opdyke, 1973. Oxygen isotope 
and paleomagnetic stratigraphy of equatorial Pacific 
core V28-238: oxygen isotope temperatures and ice vol- 
umes on a 10 5 year and 10 6 year scale. Quat. Res. 3:39- 
55. 

Smith, G.I. 1984. Paleohydrologic regimes in the south- 
western Great Basin 0-3.2 my ago, compared with 
other long records of "global" climate. Quat. Res. 22:1- 
17. 

Selby, M.J. 1982. Hillslope materials and processes. Ox- 
ford University Press. 

Thomas, M.F. 1974. Tropical Geomorphology. John Wiley 
and Sons, New York.