Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices.

■^<4 a

EVALUATING SUMMER
WATER DEFICIENCIES

ROBERT ZAHNER

Occasional

SOUTHERN FOREST EXPERIMENT STATION

Philip A. Briegleb, Director

Forest Service, U. S. Department of Agriculture

CONTENTS

Background and definitions. » Z

Estimating the water need o 3

Estimating actual evapo-transpiration , . . . 5

Calculating deficiencieSc 5

Applications . 9

Effects on forest growth . . . . . . ,13

Summary ,,o, o.,,, 1 5

Literature cited , . . .... 16

1

THE MID-SOUTH OFTEN
FEELS THE EFFECTS OF DROUGHT

High

mortality

Need for
replanting

Slow

growth

11

EVALUATING SUMMER WATER DEFICIENCIES

Robert Zahner

Southern Forest Experiment Station

Of all the raw materials used by a tree in its growth and mainte-
nance, that required in the largest amount is water. Drought, there-
fore, can be as serious to the forester as to the farmer.

It is important that the practicing forester understand what
drought is and how to evaluate its intensity. He already has indirect
measures of dry years, such as the number of days without rain or the
number of inches less than normal rainfall. But a more basic measure
is needed, a comparison between the water required and water available
and actually supplied to forest trees.

This paper describes a method for measuring water deficiencies.
Emphasis is placed on the upland pine-hardwood forests of the mid-
South, but with a little modification the method can be adapted to many
forest areas in the country.

Basically, the story is simple. Water for forest requirements
comes from moisture stored in the soil. When this moisture is not re-
plenished by rainfall, deficiencies occur. Correct analysis of soil-
moisture storage and depletion is necessary to evaluate the deficiency.

Readers who are interested in some of the applications of calcu-
lated water deficiencies, but who do not wish to read through the details
of theory and procedure, may turn directly to page 9.

BACKGROUND AND DEFINITIONS

Before water deficiency can be explained, other more basic defi-
nitions must be understood.

Evapo- transpiration. - -Large quantities of water are removed
from the soil by forest stands. The bulk of this water, however, is not
used by the trees in the manufacture of food or in the formation of wood.
It is transpired through the leaves into the atmosphere. This water loss,
combined with evaporation directly from the ground surface, is evapo-
transpiration. _!./

When e vapo-transpiration is occurring at its maximum rate, it is
supplying all the water that can be absorbed by the atmosphere under
existing weather conditions. Early workers called this maximum rate
"the evaporating power of the air" (7, 8, II, 12, 15, 28).^/ A better
term for maximum water loss is potential evapo- transpiration, which
has been defined as the amount of water that will be lost from a land
surface completely covered with vegetation if there is ample water in the
soil at all times for the use of the vegetation (14, 18, 19, 20). The
energy for potential evapo- transpiration is present in the atmosphere
whether or not vegetation and soil moisture are actually adequate for
maximum evapo-transpiration. The potential, therefore, is the atmos-
pheric demand for water.

Water need and supply, - -Rainfall alone is not a very meaningful
indicator of a wet or dry period. The water need of a forested area
must be considered along with the water supply. The need is the amount
of water that would be evapo-transpired if the energy of potential evapo-
transpiration were completely utilized. The supply is the amount
actually moving out of the soil into the air. The water thus moving may
come from current rainfall, or it may have been stored in the soil from
previous months.

In the mid-South the distribution of rainfall does not coincide
with water needs. Because the need during the winter months is low,

1/ Suggested for more detailed reading is H. W. Lull’s compilation of
related literature, Evapo-transpiration: Excerpts from Selected
References, U. S. For est Service South. Forest Expt. Sta. Occas.
Paper 131, 117 pp. 1953.

2/ Underscored numbers in parentheses refer to Literature Cited,

Po 1 6.

- 2 -

long periods without rain show no deficiencies and are of little conse-
quence. It is with the summer period that the forester is concerned.

At this time, actual evapo- transpiration can equal the high potential
only when the soil is very moist. As the moisture in the soil is re-
duced, the actual amount of water evapo- transpired falls far below
that which could potentially be passed into the atmosphere if ample
moisture were available.

Water deficiency. - -During any period when the forest soil can-
not supply the full amount of water which the energy of potential evapo-
transpiration could move into the atmosphere, there is a deficiency.

The magnitude of the deficiency may be so small that it has no serious
effect on forest growth and behavior. On the other hand, it can become
quite large, with correspondingly disastrous effects. To evaluate the
deficiency, both the potential evapo- transpiration (the water need) and
the actual evapo- transpiration (the water supply) must be known. The
numerical difference between the two can be termed the water deficiency.

Source of water. - -In the upland pine-hardwood forests of the
mid-South, the total water available to tree roots usually is only that
held in storage by the soil particles themselves. Additional sources,
such as lateral underground seepage or shallow water tables, are
generally absent. The amount of stored water varies with soil texture
and structure, with thickness of different soil horizons, and with other
soil characteristics.

When rain has wet the soil for the entire depth of the root zone,
and after excess water has drained away, the quantity of available
water under a forest stand may vary on different soils from the equiva-
lent of only about 5 inches of rainfall to as much as 15 inches,

Evapo- transpiration occurs at its maximum rate when the soil
is at maximum storage capacity. At this level, water is held only
loosely by the soil, and is easily removed. As the soil is depleted of
moisture, the remaining water is held ever more tightly by individual
soil particles, and the rate of evapo- transpiration decreases pro-
portionately (1, 10, 17, 30, 31). As the soil approaches wilting point,
during long periods without rain, actual evapo- transpiration becomes
negligible. With each rainfall, the rate of evapo- transpiration picks
up immediately.

ESTIMATING THE WATER NEED

The evaporating potential of the atmosphere is difficult to
measure directly. Although solar radiation is the source of energy,

- 3 -

many factors--e. g„ , air temperature, humidity, barometric pressure,
and wind- = modify the rate of evapo- transpiration, Transeau’s (28)
precipitation- evaporation ratio was an early attempt to find a climatic
indicator of this energy. In measuring the evaporating potential, other
investigators have constructed indices involving complex moisture-
temperature relations and vapor-pressure deficits (7, 12),

Units of moisture-temperature indices and vapor-pressure
deficits, however, are difficult to apply, Thornthwaite (20) has de-
veloped an empirical method for calculating potential evapo- transpiration
from air- temperature data alone. His units of measure for potential
evapo- transpiration are readily converted into inches of water, the
standard measure of rainfall and other water-cycle variables, Thornth-
waite's method has found many practical applications (4, 6, 13, 21, 22,
£3, 24, £7, £9). “ “

From Thornthwaite ’ s (20) method, it is apparent that the climate
of the mid-South is sufficiently uniform to yield a nearly constant value
for normal potential evapo- transpiration over the whole of the forested
Gulf Coastal Plain area from Alabama to Texas, According to the
method, the average monthly potential evapo- transpiration for five wide-
ly separated locations (Crockett, Texas; Hammond, Louisiana; Crossett,
Arkansas; Tupelo, Mississippi; and Selma, Alabama) is 6. 7 0, 4 inches

for June, July, and August. Monthly variation within the summer
season is small.

There is evidence that Thornthwaite ' s method may underestimate
the summer water need in the mid-South. Forest stands in south
Arkansas have been found to remove from the soil more than a quarter -
inch of water per day through evapo-transpir ation (16, 31, 32), Un-
published data of summer soil-moisture depletion by upland forests in
Mississippi and in east Texas substantiate this figure. It would appear
that the water need during June, July, and August is about 8 inches per
month, and in the calculations outlined in this paper, this figure will
be used in preference to. the lower one (6, 7 + 0. 4 inches) obtained by
Thornthwaite’ s method.

Along with the hot summer months, the latter part of the warming
period in the spring and the early part of the cooling period in the fall
are important. Calculations by Thornthwaite ' s method from mid-South
weather data show that the w^ater need for May and September is about 5
inches for each month. That for October is half this amount, or about
2. 5 inches. These figures are in close agreement with measured rates,
and will be used in calculations of water deficiencies.

4

All values for water needs are estimates only, and are subject
to modification as experience and further investigation warrant.

ESTIMATING ACTUAL EVAPO- TRANSPIRATION

In contrast to potential evapo- transpiration, actual evapo- transpi-
ration is a highly variable factor. During the summer, when the water
need is normally greater than rainfall, actual evapo- transpiration is
equal to current rainfall plus changes in soil water storage. In June, for
example, if rainfall is 3 inches and soil water storage is reduced 3 more
inches, actual evapo- transpiration has supplied the atmosphere with 6
inches of water. Ignoring runoff, all rainfall of less than 8 inches for
June, July, and August, of less than 5 inches for May and September, and
of less than 2. 5 inches for October, is used in evapo- transpiration during
these months. Water entering the ground is only momentarily stored.
Trees soon remove it in transpiration. Even intercepted radnfall, al-
though never entering the soil, is evaporated directly from forest crowns
and litter and is utilized in actual evapo- transpiration.

During long periods without rain, actual evapo- transpiration
cannot keep pace with the water need, A forest can meet large water
demands for a short time with the reserves stored in the soil, but soon
stored water makes up only a part of the water need, and this part be-
comes ever smaller as the soil dries.

Although there is some variation with soil type, for practical
consideration it can be assumed that the rate of moisture depletion by
upland pine-hardwood forests is directly proportional to the moisture
content of the surface six feet of soil- -the effective root zone, F or
example, when a soil which is capable of holding 10 inches of water in
the root zone is depleted of all but five inches, the rate of actual evapo-
transpiration should be about half the maximum rate.

CALCULATING DEFICIENCIES

A detailed account of what is happening to the balance between
supply and demand for water in any particular location can be kept as
a bookkeeping procedure. As Thornthwaite and Mather (27, p. 350)
state, *’The moisture in the soil may be regarded as a bank account.
Precipitation adds to the account; evapotranspiration withdraws from
it. *’ With irrigation, the farmer can keep his books balanced. The
forester, although without irrigation, would like to keep an accurate
account of the deficits.

- 5 -

Sample computation^ --Summer water deficiencies can be system =
atically computed by employing the basic concepts discussed aboveo
The following procedure has been adapted from Thornthwaite et al.

(26, 27)s, for special use by foresters in the mid-South. The raw data
necessary are rainfall records and the approximate soil moisture
storage capacity of the site under consideration.

To illustrate the method, a sample computation is presented on
p. 7, using hypothetical rainfall and soil storage data. In the step-by-
step explanation, the reader may find it helpful to follow the procedure
by means of the bar diagram. Water needs for the mid-South are given
each month. Actual evapo-transpir ation is assumed equal to these water
needs during those months in which rainfall is equal to or greater than
the need- -as in May, August, and October, in this example. When
rainfall is less than the need, as in June, July, and September, then
actual evapo- transpiration is equal to the sum of rainfall plus the
change in soil moisture storage. Usually soil storage cannot complete-
ly make up the difference, and the part not made up is the deficiency.

The actual amount of water supplied by the soil during a par-
ticular month depends on the storage capacity of the soil and on how
much depletion has occurred during previous months. These values are
taken from table 1, which has been constructed to simplify computations.
This table is based on the premise that the amount of water supplied
from soil storage becomes proportionally less as the soil dries; it shows
amounts of water actually supplied by soil storage for any amount of
water need in excess of rainfall. For comparisons between specific
sites of known storage, the table gives water supplied by soils of four
different storage capacities - = 6, 8, 10, and 12 inches. In this example,
as for most practical purposes, an "average” storage capacity of 1 0
inches of water is assumed.

The application of table 1 requires some explanation. In the
example, 7. 5 inches of water are required during June for maximum
evapo-transpiration. Reading down the 10-inch maximum storage
column in table 1, it is seen that the soil can supply 5, 3 inches. In
July, 5. 5 inches are lequired. Now computations must account for the
accumulated two-month difference between rainfall and the water need.
This is 13. 0 inches, made up of 7. 5 for June and 5. 5 for July. Table 1
shows that the soil can supply 7, 3 inches toward a 13. 0 need. But 5. 3
inches of this has previously been supplied during June, which leaves
only 2, 0 inches to be used in July. This accumulated effect may con-
tinue, with the soil becoming ever drier, supplying less and less water
for actual evapo-transpiration, as long a.s rainfall is below the water
need.

- 6 -

MAY JUNE JULY AUG. SEPT. OCT.

CALCULATION OF WATER DEFICIENCY ON A
HYPOTHETICAL FORESTED SITE WITH A MAXIMUM
SOIL STORAGE OF TEN INCHES.

MAY INCHES

RAINFALL 8.5

NEEDED 5.0

DIFFERENCE (Excess water) 3.5

REMAINING IN SOIL 10.0

JUNE

RAINFALL 0.5

NEEDED 8.0

DIFFERENCE -7.5

SUPPLIED FROM SOIL STORAGE 5.3

DEFICIENCY 2.2

REMAINING IN SOIL 4.7

JULY

RAINFALL 2.5

NEEDED 8.0

DIFFERENCE -5.5

SUPPLIED FROM SOIL STORAGE 2.0

DEFICIENCY 3.5

REMAINING IN SOIL 2.7

AUGUST

RAINFALL 8.5

NEEDED 8.0

DIFFERENCE (Soil storage recharge) 0.5

REMAINING IN SOIL 3.2

□

□

Rainfall used in evapo-transpiration.

Soil storage used in evapo-transpiration.
Rainfall used in soil storage recharge.
Excess rainfall.

Water deficiency.

WATER NEEDS, SUPPLIES, AND DEFICIENCIES
FOR THE SAMPLE COMPUTATIONS AT THE
RIGHT.

SEPTEMBER

RAINFALL 4.0

NEEDED 5.0

DIFFERENCE -1.0

SUPPLIED FROM SOIL STORAGE 0.2

DEFICIENCY 0.8

REMAINING IN SOIL 3.0

OCTOBER

RAINFALL 3.5

NEEDED 2.5

DIFFERENCE (Soil storage recharge) 1.0

REMAINING IN SOIL 4.0

TOTAL SUMMER DEFICIENCY 6.5

- 7 -

Table 1. - -Relationship between water required and the water actually supplied by storage
Water required is the accumulated difference between rainfall and need

Accumulated

difference

between
rainfall and
need (inches)

Total maximum storage

Accumulated

difference

between
rainfall and
need (inches)

Total maximum storage

6

inches

8

inches

10

inches

12

inches

6

inches

8

inches

10

inches

12

inches

Inches of water s

upplied by storage

Inches of water supplied by storage

0. 2

0. 20

0. 20

0. 20

0. 20

8. 0

4. 45

5. 10

5. 54

5. 87

. 4

. 39

.40

.40

. 40

8. 2

4. 50

5. 17

5. 63

5. 97

. 6

. 58

. 59

. 59

. 59

8.4

4. 55

5. 24

5. 72

6. 07

. 8

. 76

. 78

. 78

. 78

8. 6

4. 60

5. 31

5. 81

6. 17

8. 8

4. 65

5. 38

5. 89

6. 27

1. 0

. 94

. 96

. 96

. 96

1. 2

1. 11

1. 14

1. 14

1. 14

9. 0

4. 70

5. 45

5. 97

6. 37

1.4

1. 27

1. 31

1. 32

1. 32

9. 2

4. 74

5. 51

6. 05

6.46

1. 6

1. 43

1.48

1.49

1. 50

9.4

4. 78

5. 57

6. 13

6. 55

1. 8

1. 58

1. 64

1.66

1. 68

9. 6

4. 82

5. 63

6. 21

6. 64

9. 8

4. 86

5. 69

6. 29

6. 73

2. 0

1. 73

1. 80

1. 83

1. 85

2. 2

1. 87

1. 96

1. 99

2. 02

10. 0

4. 90

5. 75

6. 36

6. 82

2. 4

2. 01

2. 11

2. 15

2. 19

10. 2

4. 94

5. 81

6.43

6.91

2. 6

2. 14

2. 26

2. 31

2. 35

10. 4

4. 98

5. 86

6. 50

6.99

2. 8

2. 27

2. 40

2.46

2. 51

10. 6

5. 01

5. 91

6. 57

7. 07

10. 8

5, 04

5. 96

6. 64

7. 15

3. 0

2. 39

2. 54

2. 61

2. 67

3. 2

2. 51

2. 68

2. 76

2. 83

11. 0

5. 07

6. 01

6. 71

7. 23

3. 4

2. 63

2. 81

2. 91

2. 98

11. 2

5. 10

6. 06

6. 78

7. 31

3. 6

2. 74

2. 94

3. 05

3. 13

11.4

5. 13

6. 11

6. 84

7. 39

3. 8

2. 85

3. 07

3. 19

3. 28

11. 6

5. 16

6. 16

6. 90

7.47

11. 8

5. 19

6. 21

6. 96

7. 55

4. 0

2. 96

3. 19

3. 33

3. 43

4. 2

3. 06

3. 31

3.46

3. 57

12. 0

5. 22

6. 25

7. 02

7. 62

4. 4

3. 16

3.43

3.59

3. 71

12. 5

5. 30

6. 35

7. 17

7. 80

4. 6

3. 25

3. 54

3. 72

3. 85

13. 0

5, 35

6.45

7. 32

7. 97

4. 8

3. 34

3. 65

3. 85

3. 99

13. 5

5.40

6. 55

7.46

8. 14

14. 0

5.45

6. 64

7. 57

8. 29

5. 0

3.43

3. 76

3. 97

4. 13

14. 5

5. 50

6. 72

7. 70

8.44

5. 2

3. 52

3. 87

4. 09

4. 26

5.4

3. 60

3. 97

4. 21

4. 39

15. 0

5. 55

6. 79

7. 82

8. 59

5. 6

3. 68

4. 07

4. 33

4. 52

16. 0

5. 61

6. 94

8. 02

8. 86

5. 8

3. 76

4. 17

4.44

4. 64

17. 0

5. 66

7. 07

8. 22

9. 11

18. 0

5. 71

7. 17

8. 39

9.35

6. 0

3. 83

4. 27

4. 55

4. 76

19. 0

5. 76

7. 27

8. 54

9.55

6. 2

3. 90

4. 36

4. 66

4. 88

6.4

3. 97

4.45

4. 77

5. 00

20. 0

5. 81

7. 37

8. 69

9.75

6. 6

4. 04

4. 54

4. 87

5. 11

22. 0

5. 88

7. 49

8. 91

10. 09

6. 8

4. 11

4. 63

4. 97

5. 22

24. 0

5. 93

7. 59

9. 11

10. 39

26. 0

5.96

7. 69

9. 29

10. 63

7. 0

4. 17

4. 71

5. 07

5. 33

28. 0

5. 98

7. 79

9. 39

10. 83

7. 2

4. 23

4. 79

5. 17

5. 44

7. 4

4. 29

4. 87

5. 27

5. 55

7. 6

4. 35

4. 95

5. 36

5. 66

7. 8

4.40

5. 03

5.45

5. 77

-8-

August shows no deficit, since rainfall is greater than the need,
September again requires water from soil storage. The accumulated
difference between need and rainfall is now 13. 5 inches, made up of
13,0 inches from June and July, minus the 0, 5-inch recharge in August,
plus 1, 0 inch in September. Table 1 shows that the soil can supply 7. 5
inches of this total. However, 7. 3 inches of this have previously been
used during June and July, leaving only 0. 2 inch for use in September.

The winter months have very low water needs. Rainfall in
excess of the need goes into soil storage recharge until the full 10-inch
capacity is reached. Further excess rainfall then is lost through runoff
and internal drainage.

Effect of runoff, --In the discussion above, it has been assumed
that no runoff occurs when precipitation is below the potential evapo-
transpiration. In the example just given, it was assumed that all of the
rainfall of 0, 5 inch in June and 2, 5 inches in July entered the ground or
was intercepted by foliage and litter. The total 3, 0 inches, therefore,
was assumed available for actual evapo- transpiration during June and
July,

Depending on the intensity of showers, some of the rainfall
during the growing season is actually lost as runoff. This amount is
usually small, however, and in most summer showers on dry forested
uplands of the mid-South can be ignored for practical purposes. In
any case, water lost through runoff is not available for actual evapo-
tra.nspiration. The assumption that there is no runoff during the grow-
ing season therefore results in a conservative estimate of the actual
water deficiency.

APPLICATIONS

The evaluation of water deficiencies has real meaning when
applied under specific conditions. Two variables affect deficiencies:
soil and weather conditions. The latter changes with time and place;
i. e, , from year to year and from, one geographic location to another.
When time and geographic regions are held constant, it is possible to
evaluate water deficiencies of various sites within a region. When a
site within a region is held constant, comparisons of deficiencies are
possible among years, or even among months. Thus, it is possible
to evaluate water deficiencies for many combinations or circumstances
within the mid»South area. Following are some examples of how cal-
culated deficiencies can be applied to specific cases.

- 9 -

Comparison among sites. - -As emphasized earlier, the amount
of stored water available for forest consumption varies greatly from
site to site. The greater the amount of stored water, the longer the
forest can supply the atmospheric
demand for water during the sum-
mer period, and the smaller the
deficiency. Figure 1 illustrates
this principle for two sites on the
Crossett Experimental Forest in
south Arkansas. Normal rainfall
data were used in computing the
deficiencies, so that the average
effects of the two sites might be
compared.

The site storing 12 inches
of available water in the root zone
suffers a deficiency of nearly 7
inches during a normal summer.
In contrast, the deficiency of an
adjoining site storing only 6 inches
of water is 40 percent greater--
10 inches of water. On the latter
site, a deficiency normally begins
in May, whereas on the site stor-
ing more water, soil storage sup-
plies the May requirements.

6 inches soil storage 12 inches soil storage

dRainfall used in evapo-transpiration.
^^Soil storage used in evapo-transpiration.
[^Rainfall used in soil storage recharge.
U Water deficiency.

During a normal summer,
all stored water is not used on the
site storing 12 inches of water,
and wilting point is not reached.

On the site storing only 6 inches
of available water, however, wilt-
ing point is closely approached.

Figure 1.- -Water needs and supplies
for a normal summer at Crossett on
two sites, one storing 6 inches and
one storing 12 inches of available
water in the root zone.

As has been mentioned, recharge of depleted soil water begins
only after the demand by potential evapo-transpiration becomes less than
is supplied by rainfall. For the Crossett area, recharge by rainfall does
not normally begin before October. Assuming no runoff, all rain falling
on the site storing 12 inches of water goes into the ground as storage until
late December, when, in a normal year, recharge has completely replaced
the 8. 2 inches of depleted storage. This normal recharge occurs by early
December on the site which stores only 6 inches of available water, as

10 -

only 5.4 inches of stored water is used. After recharge of the soil, rain-
fall in excess of potential evapo-transpiration is surplus water and is
lost to both sites by runoff or by internal drainage.

Comparison among years. - -Figure 1 shows that water deficien-
cies of about 7 inches can occur on a good site in the mid-South during a
normal growing season. The forester is quite well satisfied with a
’’normal” year. The dry year, or drought year, is of concern.

Figure 2 shows the trends cf precipitation and evapo-transpiration
for the 6-year period of 1950 through 1955 for an average site on the
Crossett Experimental Forest. The summers of 1950 and 1951 were
considered wet, those of 1952, 1953, and 1954 were dry, and that of 1955
intermediate. Winter surpluses did not vary greatly from year to year,
and averaged about 20 inches after soil storage recharge. The total ex-
cess of nearly 120 inches for the 6-year period certainly does not reveal
the severity of the three drought years.

mjjaso mjjaso mjjaso mjjaso mjjaso mjjaso

1950 1951 1952 1953 1954 1955

Figure 2. --Water needs and supplies for 1950-55 on the Crossett Ex-
perimental Forest. Computations were made for a site storing 10 inches
of available water in the root zone.

Analyzed season by season, figure 2 reveals outstanding differ-
ences in water deficiencies for the Crossett area during the 6-year
period. Deficiencies varied from 0. 7 inch in 1950 to 16. 7 inches in 1952.
The result is a simple quantitative comparison of failures to meet the

11

water need from one growing season to another. Wilting point of the soil
on this average site was closely approached in all years except 1950 and
1951. Sixty-five percent of available soil moisture had been removed
from this site by the end of June in 1952, and about 55 percent had been
lost by this date in 1953 and 1954. Ninety percent of available water was
used during all three years by the end of August, with a dry September
still ahead in all cases.

Dry periods within a season canbe easily delineated andanalyzed
in detail. For example, of the 11.8-inch deficiency for Crossett in 1953,
4. 8 inches (40 percent) occurred during September and October, while
only 3.7 inches (30 percent) occurred during June and July. In 1952, dry
conditions in May and June resulted in most of the stored water being
used early in the season, a circumstance which greatly increased de-
ficiencies from July through October. Other interesting examples are
noted in June of 1953 and 1954. Even though monthly rainfall was only
about one -half inch for each of these instances, actual monthly deficien-
cies were not high- -in the neighborhood of only 2 inches each. In these
periods stored soil water supplied most of the water required by poten-
tial evapo- transpiration, thus partly compensating for the lack of rain.

Comparison among locations. -- The diagrams in figure 3 have
been constructed from weather records of four locations throughout the
mid-South. The normal trends of precipitation and evapo- transpiration
are compared with those for the year 1954. An average soil storage of

HRoinfall used in evapo-tronspiration.

j^Soil storage used in evapo-transpiration.

MJJASO MJJASO MJJASO MJJASO MJJASO MJJASO MJJASO MJJASO
Normal 1954 Normal 1954 Normal 1954 Normal 1954

Crockett, Texas Hammond, Louisiana Tupelo, Mississippi Selmo, Alabama

Figure 3. - -Water needs and supplies for normal conditions and for the
1954 season, at four locations. Computations were made for a site
storing 10 inches of available water in the root zone.

12 -

10 inches of water is assumed for all four locationSo It is felt that this
storage capacity is representative of most forested areas of upland
Coastal Plain soils in the mid-South, For specific sites, summer de-
ficiencies will be more or less severe than shown in the figure, depend-
ing on the actual amount of water stored in the root zone,

Crockett, Texas, normally has a rather large water deficiency,
9o 5 inches, whereas Hammond, Louisiana, has relatively little, 2, 5
inches. More than 80 percent of available soil water is used during the
normal summer at Crockett, while only half this amount is used at
Hammond, Forests at Tupelo, Mississippi, and Selma, Alabama,
normally use about 75 percent of the stored water, and suffer a de-
ficiency of between 6 and 7 inches of water.

During the summer of 1954, all stations except Hammond show
severe water deficiencies, with Selma having the largest absolute de-
ficiency, Although Hammond rainfall was less than one-half inch for
August 1954, soil storage supplied most of the 8-inch water need, and
the deficiency was only 3, 1 inches for the month. At Selma, low rain-
fall early in the season used up most stored water, with the result that
continuing dry weather accumulated a very large water deficiency in
1954.

In terms of departure from normal, the 1954 deficiency at
Crockett was not so severe as that at Selma or Tupelo, Although Tupelo
had a smaller absolute deficiency than did Crockett, it suffered a
greater departure from normal, Hammond showed only minor departure
from noimal deficiency in 1954,

EFFECTS ON FOREST GROWTH

This pamper does not offer experimental evidence of the effect of
summer water deficiencies on tree growth. Instead, it is merely in-
tended to present a useful and new concept of how summer water de-
fiency can best be expressed. Certain generalizations about the relation-
ship between tree growth and water deficiency appear in order, however.

The rate of for est tree growth is generally known to be related
to the amount of available water in the soil (2, 3, 5, 9)o There is in-
creasing evidence that whenever soil moisture is deficient, the rate of
forest growth is affected. New plantations are probably affected differ-
ently than stands of pulpwood or poles. These in turn may be affected
differently than m.ature timber.

13 -

It is not yet known exactly at what level of water deficiency forest
growth is affected. As long as actual evapo-transpiration keeps pace
with the potential, the rate of growth should be at the maximum that the
other factors (e. g, , temperature, nutrient supply, photoperiod) will allow.
When actual evapo-transpiration falls below the potential, then the water
factor theoretically may affect growth. Even during a normal year in the
Crossett, Crockett, Tupelo, or Selma areas, for example, maximum
growth would not be attained during the June through September period-^
half of the growing season. During specific years, forest growth in these
areas actually is proportionately higher or lower than normal, depending
on the magnitude of water deficiency.

Comparisons of water deficiencies by sites, years, and geo-
graphic locations offer possible explanations for corresponding differ-
ences in forest behavior. Thus while the dry summer of 1954 was felt by
forests in other parts of the mid-South, growth in the Hammond area may
have suffered very little that year. Soil storage on the average site pro-
vided all but a very little of the water required during the one dry month.
This situation extends to the normal year. Normal deficiencies are one-
fourth to one- third less in the Hammond area than in the Crockett, Tupelo,
or Selma areas. Moreover, calculations show that a soil at Hammond
storing only 6 inches of available water normally suffers a smaller water
deficiency than a soil at Crockett which stores 12 inches of water. Forest
growth on poor soil in the Hammond area might therefore be as much as
that on good soil in the Crockett area. Although it is not known whether
water deficiency is largely reflected in reduced height growth, reduced
diameter growth, or both, site index differences can probably be traced
to differential water deficiencies.

Year-by-year comparisons of deficiencies on the average site in
one location should yield corresponding forest growth comparisons. At
Crossett, for instance, water relations were adequate for excellent
growth throughout the summer of 1950. During the summers of 1952,

1953, and 1954, however, growth was adversely affected by the pro-
portional amounts of the respective water deficiencies. In 1951, growth
may not have been affected before August, for there was no deficiency in
June and July.

Conservation of soil moisture through stand management- -control
of basal area, eradication of underbrush, removal of cull trees--mini-
mizes the ill effects of water deficiency. Further investigations may
discover practical ways of overcoming part of the deficiency felt nearly
every summer by forests of the mid-South.

14 -

SUMMARY

Summer water deficiencies of upland forest land can be evaluated
only when the water need is compared to the water supply. The atmos-
pheric potential for evapo- transpiration is a, measure of the need. Current
rainfall plus stored soil water make up the supply. This paper describes
a method for calculating water deficiencies by taking into account these
factors of need and supply.

Normally, summer rainfall in many parts of the mid-South
supplies only about half the water required by the high potential evapo-
transpiration. Part of this difference is made up by water stored in the
soil and removed by forest vegetation. That part not supplied is the de-
ficiency. The more available water a soil inherently stores in the root
zone, the less the deficiency during a dry summer period.

As long as the soil is moist, it supplies a large portion of the
difference between rainfall and the water requirements of the atmosphere.
As the soil dries, however, it supplies less and less water for evapo-
transpiration, and the deficiency becomes greater and greater. During
summer droughts, it is common for forest land to build up deficiencies
of 15 to 20 inches of water in the mid-South.

Because the water need is considered relatively uniform over
large areas in the mid-South, water deficiencies vary primarily with
current rainfall and soil storage capacities. Using a minimum of data,
it is possible to evaluate and compare deficiencies from site to site,
from year to year, and from geographic location to location within the
mid-South region.

15 -

LITERATURE CITED

(1) Ashcroft, G, , and Taylor, S» A,

I953„ Soil moisture tension as a measure of water removal
rate from soil and its relation to weather factors.

Soil Sci. Soc, Amer. Proc. 17: 171-174.

(2) Boggess, W, Rc

1953. Diameter growth of shortleaf pine and white oak during
a dry season. Univ„ 111. Agr. Expt, Sta. Forestry
Note 37, 7 pp.

(3) Byram, G. M. , and Doolittle, W. T.

1950. A year of growth for a shortleaf pine. Ecol. 31: 27-35.

(4) Fletcher, H. C. , and Rich, L. R.

1955. Classifying southwestern watersheds on the basis of
water yields. Jour. Forestry 53: 196-202.

(5) Clock, W. S.

1955. Tree growth, 11. Growth rings and climate, Bot. Rev.
21: 73-188.

(6) Hare, F. K.

1950. Climatic and zonal divisions of the boreal forest

formation in eastern Canada. Geog. Rev. 40: 615-635.

(7) Huffaker, C. B.

1942, Vegetational correlations with vapor pressure deficit
and relative humidity. Amer. Midland Nat. 28:
486-500.

(8)

(9)

(10)

(11)

(12)

(13)

Kittredge, J.

1938. The magnitude and regional distribution of water loss

influenced by vegetation. Jour. Fore stry 3 6: 775- 778.
Kozlowski, T. T.

1955. Tree growth, action and interaction of soil and other
factors. Jour. Forestry 53: 508-512.

Lassen, L. , Lull, H. W. , and Frank, B.

1952. Some plant- soil- water relations in watershed manage-
ment. U. S, Dept. Agr. Cir. 910, 64 pp,

Livingston, B. E.

191 lo A study of the relation between summer evaporation in-
tensity and centers of plant distribution in the United
States. Plant World 14: 205-222.

1916, A single index to represent both moisture and temperature
conditions as related to plants. Physiol. Res. 1:
421-440.

Mather, J. R,

1954. The determination of soil moisture from climatic data.

Bui. Amer. Met. Soc. 35: 63-68.

16 -

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

Mather, J. R.

1954. The measarement of potential evapotranspiration. Pub.

in Climatology 7(1). Johns Hopkins Univ. Lab. of
Climatology.

McDoagall, E.

1925. The moisture belts of North America. Ecol. 6:325-332.

Moyle, R. C. , and Zahner, R.

1954. Soil moisture as affected by stand conditions. U. S.

Forest Serv. South. Forest Expt. Sta. Occas. Paper

137, 14 pp.

Rowe, P. B. , and Colman, E. A.

1951. Disposition of rainfall in two mountain areas of

California. U. S. Dept. Agr. Tech. Bui. 1048,

84 pp.

Thornthwaite, Co W,

I94I0 Atlas of climatic types in the United States, 1900-1939o
U, So Depto Agr. Misc, Pub. 421, 58 pp.

1946.

The moisture-factor in climate. Trans. Amer. Geophy
Union 27: 41-48, illus.

(— >

v£)

4^

00

An approach toward a rational classification of climate.
Geog. Rev. 38: 55-94.

1952.

Grassland climates. Proc. Sixth Int. Grassland Cong. ,
pp. 667-675. State College, Pa.

1955.

Climatology in arid zone research. Paper presented
at the AAAS International Arid Lands Meeting,
Albuquerque, N. M.

and Hare, F, K.

1955,

Climatic classification in forestry, Unasylva 9: 50-59.
and Mather, J. R.

1954.

Climate in relation to crops. Met. Monog. 2: 1-10.
and Mather, J, R.

1955.

Climatology and irrigation scheduling. Weekly Weather
and Crop Bui. , Nat. Summary of June 27,
and Mather, J, R.

1955.

The water balance. Pub. in Climatology 8(1), 104 pp. ,
illus. Drexel Inst, of Technology, Lab. of
Climatology, Centerton, N. J.
and Mather, J. R.

1955. The water budget and its use in irrigation. Water, Agr.
Yearbook 1955, 751 pp, , illus.

17 -

(28)

(29)

(30)

(31)

(32)

Transeau, E. N,

1905. Forest centers of eastern North America. Amer. Nat.
39: 875-889.

Van Bavel, C. H. M.

1953. A drought criterion and its application in evaluating
drought incidence and hazard. Agron. Jour. 45:
167-172.

West, E, S, , and Perkman, O,

1953. Effect of soil moisture on transpiration. Austral.

Jour. Agr, Res. 4: 326-333.

Zahner, R.

1954. Soil moisture depletion by pine and hardwood stands

during a dry season. Forest Science 1(4): 258-264.

1954. Plantation exhausts soil water rapidly. U. S. Forest
Serv, South. Forest Expt. Sta. South, Forestry
Notes 95.

- 18 -