Biodiversity

and Representativeness of

Research N atural A reason

N ational Wildlife Refuges

in M ontana

Designated Areas Within Benton Lak^ CharlesM . Russell,

LakeMason, Medidne Lakeland Red Rode Lakes

National Wildlife Refuges

FINAL REPORT
August, 1999

Submitted to the
U . S. Fish and Wildlife Service

P repared by:
Stephen V. Cooper and Bonnie L. Heidel

M M W 1' A N A

Natuml Heritaj

Biodiversity and

Representiveness of

Researcii N atural Areas on

N ational Wildlife Refuges in

M ontana

Designated Areas Within Benton Lake^ ChariesM. Russdl,

LakeMason, MedidneLalce^ and Red Rode Lakes

Nation^ WildlifeRefuges

August, 1999

©1999 M ontana N atural H eritage Program

state Libi^y Building. P.O. Box 201800. 1515 E^Sxth A\^ue. Hdoi^ MT .596201800. 406-444-3009

This document should be cited as follows:

Cooper, S. V. and B. L, Heidel. 1999. Biodiversity and representativeness of Research Natural Areas on National
Wildlife Refuges in Montana: designated areas within Benton Lake, Lake Mason, Medicine Lake, Red Rock Lakes and C.
M. Russell National Wildlife Refuges. Unpublished report to U.S. Fish and Wildlife Service. Montana Natural Heritage
Program, Helena. 63 pp. plus appendices.

I

There are fifteen Reasarch Netural Areas(RNAs) on National Wildlife

Abstract Refuges administered by the U.S. Fish and Wildlife Service in Montana Each

vjss inventDried for significant ecological and botanical attributes outstand-
ing plant association exampless rare plant association^ and M ontana plant
species of special concern. Tvjo more study sites with existing or prospective
special management designation were also considered in tine inventory \MDrk.
Biodiversity and representativeness information was prepared for each study
sit^ including a profile of all v^l-de/doped and uncommon native plant
association^ description of any rare plant species population^ and a summary
of biodiversity significance that incorporates this new data with originai
RNA designation records Related information was compiled to help put
results in contEKt for each sit^ including description of environment; land
LJ^ management notes^ and recognized non- biological values

As a result ten outstanding plant association examples, four rare plant
associations^ and four M ontoia plant ^ecies of special concern vj&s docu-
mented within twelve of the study sites M ost of the study sites are located in
the Great Plains complementing one another and generally representing
biodiversity features not otherwise undo" specif management designation in
Montana. The^ include riparian and dune system^ once- widespread grass-
land plant associations that have been drastically reduced d^where and rare
grassland plant associations thet have not been reported in M ontana before,
uncommon forest and woodland plant association^ and suites of successional
habitats associated with black-tailed prairie dog colonies Individually and
collectively, these RNAs help anchor the conservation of Great Plains natural
environments and their component plant associations and species

We recommend additional surveys that ©<tend beyond current RNA bound-
aries to identify areas that would fill gaps and achieve representation at scdes
more consistent with ecological processes and the historic nature of once-
widespread vegetation types The greatest potential for such areas is in the
Charles M . Russeli NWR and on surrounding public land^ which offer
unique opportunities for identification and consevation of representative
iarge-scale landscape systems

The expertise and interest of all U SFWS personnel vyith whom v\e worked is

AcknOWlGuOGmGntS gratefuHyazknowledged, along with the project support of the U.S. Fish and

Wildlife Service (USFWS| - U pper Missouri/ YeJIowjstone Rive" Ecosystem
Team and the U .S. Fish and Wildlife Service - Ecologies Services Office in
H dena We thank Steve M artin of Boiton Lake N ationd Wildlife Refuge
whose interest and support catalyzed this project J im Stutzman - U SFWS
M ontana Wildlife H abitat Office lent initial project support in an agreanait
b^ween the Montana Partners for Wildlife Program and Montana Natural
H eritage Program. R^uge coordination and access v\ere gratiousiy provided
by Mike Rabenberg (Medicine Lake National Wildlife Refuge), Ste^ Martin
(Benton LakeNat^onal Wildlife Refuge), MikeHedrick, Bill Berg, Bill
Haglan, Ever^t Russdl, and Matt DeRosier (Charles M. Russell National
Wildlife Refuge) and Daniel Gomez (Red Rock Lakes Nation^ Wildlife
Refuge); vwth able navigation and assistance provided by skippers Glen
Guenther and Jodyjonea The report was reviewed in draft form, and
comments and corrections were provided by Steve M artin, Bill H aglan, Tedd
Gutzk^ M ike Rabenberg, and Jim McCollum.

This work also ben^ited from the time and skills of M ontana N atural
Heritage(MTNH P) staff. Jim Vanderhorst provided botanical expertise in
fidd inventory at one site Scott Lee-Chadde digitized sampling locations
and contributed GIS map productSL Steve Chadde and Cedronjones
conducted the origir^l vrork in years prior to this project that s^ up the
databases with RNA information, subsequently used to plai this in\^entory
and provide a framework for compiling new inform^ion. The Biological
Conservation Database and its linked ^riesof datasefes represent the contri-
butions of many MTNHP staff, as v^l as the vjork of biologists statewide

This project was funded under two, separefce one->ear work order and
challenge cost-share ^reements betv\een tiie U . S. Fish and Wildlife Service
Program, theU. S. Fish and Wildlife Service- Ecological Services Office in
H den^ and the M ontana N atural H a"itage Program.

Ill

TABLE OF CONTENTS

Introduction i

Study Areas 4-

Methods 6

Results lo

Benton Lake National Wildlife Refuge 13

Mullan Trail Research Natural Area 13

CHARLES M. RUSSELL NATIONAL "WILDLIFE REFUGE 15

Fourth Ridge Research Natural Area 15

Hell Creek Potential Research Natural Area 17

Limber Pine Research Natural Area IS

Manning Corral Prairie Dog Town Research Natural Area 22

Missouri River Bottomlands Research Natural Area 24"

Prairie Dog Island Research Natural Area 28

Spring Creek Research Natural Area 30

Two Calf-Douglas-fir Research Natural Area 33

York Island Research Natural Area 3G

LAKE MASON NATIONAL WILDLIFE Refuge 39

Lalce Mason Research Natural Area 39

MEDICINE LAKE NATIONAL WILDLIFE REFUGE 4*1

Big Island Research Natural Area 411

Bruce's Island Research Natural Area 415

Homestead Research Natural Area 416

Medicine Lake Sandhills 4'7

Tepee Hills Research Natural Area 419

RED ROCK LAKES NATIONAL WILDLIFE REFUGE 53

Sheep Mountain Research Natural Area 53

DISCUSSION 57

CONCLUSIONS AND RECOMMENDATIONS 59

LITERATURE Cited ei

IV

FIGURES

Figure 1. Location ofU. S. Fish and Wildlife Service-administered Research Natural Areas in Montana 4*

Figure 2. Big Island Research Natural Area: Map of pi ant communities and associations 4i3

Figure 3. Tepee Hills Research Natural Area: Map of distribution of plant communities and associations 51

TABLES

Table 1. Target list of Montana plant species of special concern in the study area 8

Table 2. Synonyms among scientific names for dominant graminoids 9

Table 3. Matrix of plant communities / associations by Research Natural Area "'."v it h in Montana's National Wildlife

Refuges (arranged alphabetically within lifeform) 11

Table ^. Partial matrix of National Wildlife Refuge RNA criteria and sites in Montana 57

APPENDICES

Appendix A. Community survey form

Appendix B. Plant species of special concern survey form

Appendix C. Photographs of state-significant vegetation features

Appendix D. Vegetation constancy-cover sampling data - N Ot Available Online. See M T N H P for Details.

Appendix E. Element occurrence records for Montana plant species of special concern

Appendix F. Illustrations of Montana plant species of special concern

Appendix G. Vascular plants cited in this report, by common names, scientific names, and six-letter acronyms

INTRODUCTION

The purpose of this study is to develop a baseline of

ecological and botanical information on each Research
Natural Area (RI^TA) within the National "Wildlife
Refuges administered by the U.S. Fish and Wildlife
Service (USFWS) in Montana. The study results
provide a reference for refuge managers and
researchers, a standard for comparing throughout the
Refuge system in the Region, and a contribution to the
systematic evaluation of natural areas across the
Montana landscape as a whole. This report presents the
information on plant associations and rare plants
collected at all RNAs over the t'-'vo years of study,
replacing the previous Part 1 report that was submitted
as a drafl, and "■-"vhich described half of the RNAs.

Plant associations and species that are threatened,
endangered and sensitive are central "elements" of
biodiversity catalogued by the Montana Natural
Heritage Program statewide. The centralised database
and computer-assisted inventories focus on the state's
rarest animals and plants, as well as high-quality
examples of "natural" plant communities. As part of
the ongoing operations, we assess the "relative
endangerment of species and natural communities"
(Genter 1986)^ a daunting task in eastern Montana
with the relative paucity of information on
biodiversity features and their location. This was the
rationale in proposing an inventory of RNA biological
features among National Wildlife Refiiges in
Montana, emphasising community t3^es, and also
considering threatened, endangered and sensitive
plant species. It was designed to contribute to the
statewide framework for identifying and filling
representative natural areas targets in eastern
Montana, to identify the features protected by them,
and to increase the potential wildlife management
usefulness of existing RNAs for the USFWS while
also contributing to the understanding of ecological
and botanical resources.

From the early years of wildlife management and the
emphasis on regulating mortality and productivity for
individual species, the scope has broadened to
managing species' habitat, habitat processes, and the
fauna and flora at large.

The USFWS adopted an ecosystem approach to fish
and wildlife conservation in 1991-, defined as
^Protecting or restoring the function, structure, and
species composition of an ecosystem, recognising that
all components are interrelated" (Martin 1996).

Ecosystem management and sustainability hinge on
the maintenance of plant and animal species diversity
as well as natural processes, including disturbance (e.
g. fire, grasing), succession, and evolution. Biological
processes and biodiversity can be defined at a variety
of spatial and temporal scales, including genetic,
species, population, community, ecosystem, landscape
and regional (Noss 1983). Like the "ecosystem
management" term, "natural" has acquired numerous
potential meanings. A conceptual point of reference in
considering "natural conditions" is comparison to the
ecosystem's condition prior to European settlement,
though this is not readily reconstructed in grassland
landscapes, complicated by their dynamic nature at
several short- and long-term scales. Using a
compendium of historic information (Knowles and
Knowles 1993) and current information, preliminary
deductions and identification of geographic priorities
can be developed. On this basis, some of the National
Wildlife Refuges or areas within them offer the last or
best vestiges of natural conditions as reference areas
for ecosystem management.

Research Natural Areas are critical to ecosystem
management in the following ways:

Reference and Monitoring Sites :

The number of examples of natural ecosystems that
remain is finite and shrinking as landscapes are altered
and degraded (Noss 1987). It is judicious to manage
some ecosystems for their existing natural conditions
to reduce the risks associated with our limited
knowledge of ecosystem functions and to insure
ecosystem diversity, health, and sustainability.

Many natural resource management activities can be
conceived of as experiments; their outcome, including
changes in vegetation, animal populations, soils
quality, plant susceptibility to insect and disease
vectors, and changes in future productivity are, at
best, incompletely understood (Franldin 1992). As
such, reference points are needed to evaluate the
experiment's success. Regardless of the entity
monitored, small mammal demography, breeding bird
success, neotropical migrant birds, health of
endangered species populations, site productivity, or
impacts of road density on ungulate distribution,
reference points are essential. The reference or
benchmark function is one of the principal merits of
RNAs and similar areas for management and
environmental analysis. The availability ofRI'JAs as

sites for pure and applied scientific research is closely
linked to their importance as reference and monitoring
sites, for which research is nonmanipulative and
no ndest ru c t iv e.

Broader Research Applications :

RNAs provide more than a framework to answer
refiige or regional management questions. RNAs are
available to investigate the functioning of ecosystems
and the sustainability of both ecosystem processes and
community components. Theypresent an opportunity
for studying given ecological processes and the
natural range of ecosystem variability. Research
Natural Area systems are ideally pristine examples
that collectively represent the full range of ecosystem
types, and the accompanying range of biota, landform,
ecosystems, soils, climate, successional stages,
disturbance regimes and other ecological processes
(see Ryan et al. 1 99^ for the Rocky Mountain Region
types identified to date and Chadde et al. 1996 for the
Intermountain Region). In a similar tone, the Refuge
Manual states that "RNAS are intended to represent
the full array of North American ecosystems;
biological communities, habitats, and phenomena; and
geological and hydrological formation and conditions"
as part of a larger neh-vork for understanding
cumulative effects and large-scale changes.

Biodiversity Protection :

One of the stated goals of ecosystem management is
the protection of biodiversity. The RNA system
functions at the "fine filter" level in harboring
populations of rare or localized animals, plants, and
plant communities. The RNA system may also serve
as core areas of genetic diversity for common plant
and animal species and their habitats and as a safety
net for little loiown elements of biological diversity
(e.g. soil microflora and fauna, terrestrial and aquatic
invertebrates, etc.) and their contribution to
ecosystem processes. In this capacity they thus serve
as part of the "coarse filter" paradigm for protecting
biodiversity (Hunter 1991); all the more critical in

fragmented landscapes and patchworks of
management objectives.

Research Natural Areas are established consistent
with the Objectives Handbook of the National Wildlife
Refiige System (USFWS Refuge Manual S RM 10;
referred to as "Refuge Manual" in the rest of text).
Their establishment rests on the Handbook policy that
"The Service recognizes the importance of preserving
plant and animal communities in a natural state for
research purposes." They are categorised according to
one or more of the following biological or physical
features, consistent with their contribution to
ecosystem management:

A. Biological features

1. An ecological community significantly
illustrating characteristics of a physiographic
province or a biome.*

2. A biota of relative stability maintaining itself
under prevailing natural conditions, such as a
climax community.*

3. An ecological community significantly
illustrating the process of succession and
restoration to a climax condition following a
naturally caused disruptive change. A habitat
supporting a vanishing, rare, or restricted
species.*

^. A seasonal haven for concentrations of native
animals or a vantage point for observing
concentrated populations such as a
constricted migration route.

B. Physical features

1. Outstanding geological formations or
features significantly illustrating geological
processes.*

2. Significant fossil evidence.

3. Any site containing significant evidence
illustrating important scientific discoveries.

*(From USFWS Refiige Manual S HM 1 Of)

Many of the 15 RNAs were originally designated
based on their biological significance as providing
ecological communities characteristic of the
physiographic area. Others were cited as having
significance in providing relict habitat or habitat for
restricted species. This study was designed to evaluate
all of the 15 RNAs in Montana for their ecological and
botanical significance as they relate to five of the
criteria in the Refuge Manual (asterisked above).

Ecological communities are held a.s synonymouE with
plant communities, in the broadest sense, as the
logical units of inventory for the RNA system. Plant
communities and their variation are often interpreted
as an integrated expression ofbiotic and abiotic
influences (Pfister and Arno 1 9S0). This was the
thinking in setting the original conservation targets
each RNA, identified in terms of "Kuchler Types"
(Kuchler 1964i). They are coarsely defined potential
natural vegetation units used to characterize
prevailing vegetation across the country. Some but
not all Service-administered lands within these
vegetation units may represent the prevailing
vegetation features of the unit.

standard establishes the upper physiognomic

classification levels nationwide, the alliance and plant
association (floristic levels) have not been
standardised and are in progress. The latter are the
levels at which targets are set. Most of the detailed
classifications are from western Montana compared to
of eastern Montana (Pfister et al. 1977, Hansen and

Hoffman 1988, Hansen et al. 1995, DeVelice et al.
1996, Cooper et al. 1996). Nevertheless, a synthesis of
vegetation research results from eastern Montana and
adjoining states and provinces provides a sound
framework upon which to build and incorporate the
fijll breadth of Great Plains plant community
diversity.

This contrasts with a field-oriented approach that
focuses on plant associations. This "ground up"

approach was used in keeping with the plant
associations of Bourgeron and Engelking (1996). Plant
associations and alliances represent the existing, on-
site composition as recognized in the National
Vegetation Classification Standard (Federal
Geographic Data Committee 1997), rather than a
generalized mapping unit. While the new federal

There has not been an interagency synthesis of RI^-JA
information since the work by the Federal Committee
on Ecological Reserves (1977) at the national level. In
addition to all previously-mentioned objectives, this
project contributes new and standardized information
for incorporation into statewide, regional, and national
natural areas efforts and applications.

STUDY AREAS

Eight establ ished research natural areas (RNAs) were
inventoried in 1 997 and seven were inventoried in

199S, representing all designated RNAs administered
bj the U.S. Fish and Wildlife Service (USFWS) on the
national wildlife refuge (NWR) system in Montana
(Figure 1). Together they total 11,756 acres.

The fifteen research natural areas fall within five
National Wildlife Refuges (I'JWRe), including Benton
Lalce, Charles M. Russell, Lake Mason (administered
by Charles M. Russell), Medicine Lalce, and Red Rock
Laltes National Wildl ife Refuges. They are part of the
NWR System that includes more than 500 refuges
nationwide encompassing over 99 million acres of land
and water, supporting a diversity of flora and fauna,
and establ ished for many different purposes.

The five National Wildlife Refuges of this study are
among the largest NWP^ in the state, including most
of the NWRs east of the Continental Divide. They
were established to protect specific wildlife values,
briefly highlighted below. This summary provides a
basis for considering the contributions of the RNAs
within them to the overarching refuge goals.

Benton Lake NWR was established in 1929 as a
"refuge and breeding ground for birds." It is a
significant breeding ground and migration stopover
for ducks, geese and swans and is a recognized
shorebird site of the Western Hemisphere Shorebird
Reserve Network. It also harbors colonial nesting bird
Species of Special Concern including Franklin's gLjU,
double-crested cormorant, white-faced ibis, black-
crowned night-heron, black-necked stilt, common
tern, forster's tern and black tern, and upland
grassland birds declining elsewhere in their range.

Charles M. Russell I'JWRwas established as a national
game range in 1936, later converted to a national
wildlife refuge in 1976 in recognition of key game and
non-game species occupying its rugged terrain and
extensive habitat. They include: pronghorn antelope,
white-tailed and mule deer, reintroduced elk,
introduced Rocky Mountain bighorn, colonial nesting
birds, piping plover, raptors, mountain plovers, black-
tailed prairie dogs, upland grassland birds declining
elsewhere in their range, and reintroduced black-
footed ferret.

Lake Mason I'JWR was established in 194il and
provides habitat for breeding and migratory
waterfowl, shorebirds, passerines, raptors, and
antelope.

Medicine Lake NWR was established in 193^ through
the passage of the Migratory Bird Hunting Stamp Act,
as a "prodigious" waterfowl nesting area for Canada
geese, dabblers, and divers. It has been subsequently
recognized for its value for colonial nesting birds, as a
migration stopover, and as habitat for upland
grassland birds, including upland game, that are
declining elsewhere in their range.

Red Rock Lakes NWR was established in 1935
through the Migratory Bird Hunting Stamp Act as a
major trumpeter swan breeding and wintering area. It
has subsequently been recognized for its value for
threatened and sensitive raptors, reintroduced
peregrine falcons, waterfowl migration stopover, and
habitat for lacustrine Arctic gf^iying. Clarke's grebe,
black-crowned night-heron, colonial nestingbirds,
and a host of others.

Figure L Locations of US Fish and Wildlife Service-administered Research Natural Areas in Montana

bine's i^md AWA Jepee\ms RNA

METHODS

Two sets of information were compiled for each RNA
site before fieldwork. First, written information was

reviewed about the RNAs. This was in the RNA
estabhshment information as available from the U.S.
Fish and Wildlife Service. It had previously been
collected and entered by Montana Natural Heritage
Program in the Biological Conservation Database
(BCD) as representing recognised natural areas and
public lands (Site Basic Record Database, and
Managed Areas Database, respectively.)

Second, U.S.G.S. topographic maps ("■&) and available
aerial photos were assembled prior to or in
conjunction with fieldwork at each site, and RNA
boundaries were copied onto the maps. The
photographs were used for site stratification and
planning traverses across the major features of the
RNA. Mylar overlays were used to map out areas
having spectral signatures to consider for ground-
truthing^ and as base maps for future map production.
Often photos were not readily available, so that the
topographic maps were used to guide the site
traverses, focusing mainly on unique combinations of
slope, aspect and elevation throughout the site. In
addition, comments were routinely requested from
U.S. Fish and Wildlife Service biologists familiar with
the RNAs for information on biological features and
management, and for information and clarification
about access and boundaries. Additional botany and
ecology resources were compiled in technical
preparation (described separately in methods.)

Field investigations were restricted to established
RNA boundaries, with two additions. The Sandhills
area of the Medicine Lake NWR was included in
surveys because it shares some of the rare plant
species features as Big Island RNA, and has a special
designation as part of wilderness area. In addition, an
area west of Hell Creek State Park was identified by
Bill Haglan (Charles M. Russell ITWR) as possessing
features potentially worthy of considering for RNA
designation. We refer to the set of seventeen study
sites as including the Medicine Lalce Sandhills and
Hell Creek areas though they are not designated as
RNAs.

Ecological and botanical information collected in the
field was used to expand the RNA establ ishment
information, fully described in this report and
summarized in BCD. In addition, the individual rare
plant records have been entered in BCD, and

vegetation plot data is stored in vegetation databases
and draft classification documents.

Ecological Methods

Plant communities were identified and documented in
terms of their community composition, structure and
associated abiotic environmental parameters by
establishing representative 1/10 acre plots (37.2 fl
radius). Data were recorded on a standardised
Community Survey Form as used by Montana Natural
Heritage Program consistent "■-vith ECADS vegetation
ordination analysis (Ecosystem Characterization and
Description System, USFS 1996; see Appendix A).

At each RNA, plant associations were documented
that met one or more of the four following criteria;

1. Prevailing plant associations within designated
areas, i.e., the most extensive vegetation features
dominated by native plant species,

2. Plant associations that were the basis of original
designation, e.g., the Douglas fir forest at the
Two Calf-Douglas-fir RNA,

3. Well-developed plant associations that are
potentially rare statewide or rangewide, and

^. Well-developed plant associations in outstanding
ecological condition regardless or rarity of extent
at the site.

Vegetation sampling plots were placed within each
major natural vegetation type based on observed aerial
extent of the type. This approach provided
documentation for common vegetation types, but was
not intended for exhaustive sampling of localised or
atypical environments, large replications, or fiill
gradient representation. In some instances, a given
common community t3^e may span a range of
environments, in which case the attempt was made to
sample the modal expression of a community's
environmental range. Sampling sites were chosen
"subjectively, but without preconceived bias"
(Mueller-Dombois and EUenberg 1974) to meet the
criteria of homogeneous vegetation composition, least
disturbance, and representative setting. Plot points are
mapped on U.S.G.S. topographic maps with 300 feet
precision.

On the first Refuge visited. Medicine Lalte, excellent
quality aerial photography "was available at 8 inches /
mile that served as abase layer upon which vegetation

type maps could be traced. This was used
experimentally as a documentation tool in mapping,
The plant associations as units of inventory are
fundamentally different from species targets because
they are not often discrete in nature but intergrade
with one another, often with changes imposed by man
over space and time. Yet they are as real as for a
person to say "I live in Great Falls" without regard to
the dynamics or the definitions of Great Falls. In
addition, they have different geographic scales
attached to them in the region, over time, and within
each PINA. Some plant associations were prevailing
across the Great Plains landscape prior to European
settlement, some were restricted to large areas set
apart from the prevailing vegetation, or some were
restricted to small-sised patches of vegetation. The
presettlement extent of plant associations has been
changed in more recent times with such changes as
conversion to cropland and drainage or impoundment
constructions.

Plant associations were identified using major
classifications (Bourgeron and Engelking 1994,
Hansen et al. 1995, Schneider et al. 1997) and other
vegetation literature from Montana and the northern
Great Plains in general. These plant association
references have incorporated extensive literature
searches and are subject to refinements with new data.
Some among them also present a system and
compilation of "status ranlc" information that have
been developed statewide and rangewide, considering
past/present extent, as well as threats and uniqueness.
This assisted in identifying plant associations that
may be vulnerable or imperilled on a statewide and
rangewide basis.

Within the boundaries of each RNA, there are also
differences in extent attributed to each plant
association regardless of current landscape extent or
presettlement extent. Their landscape and
presettlement scales are noted as context in reviewing
collective significance and priorities among the plant
associations of all RNAs in the final discussion
chapter. Other site-specific information was used for
initial evaluations identifying plant associations of
outstanding quality and condition, including relative
species richness and exotic species component.

Botanical Methods

Montana plant species of special concern were
inventoried and documented on the Plant Species of
Special Concern Survey Form (Appendix B). Vouchers
specimens were collected if population numbers were
sufficiently large to avoid impact (Montana Native
Plant Society 1993) and documentary photographs
were talcen.

ground-tru thing and digitising the vegetation of

Tepee Hills and Big Island.

Tentative inventory targets were identified by
querying the Montana Biological Conservation
Database (BCD) for all records of state plant species of
special concern loiown within approximately a five
mile radius of the sites. This was based on the
Montana plant species of special concern list (Heidel
1997) and most current occurrence information in
BCD. The data search produced rare plant records in
the vicinity of the Research Natural Areas in the
Medicine Lake NWR. and fourteen rare species known
from the same counties as the Research Natural Areas
(Table 1).

Note: For purposes of this report, these species will be
collectively referred to as "rare" plant species. They
are sometimes referred to as "threatened, endangered
and sensitive" species, but there are no federally listed
plant species known in eastern Montana, nor species
recognised as sensitive or watch apart from those so of
the Bureau of Land Management and the U.S. Forest
Service.

Rare plant inventory was coupled with vegetation
sampling in visiting the RiW range of habitats. Those
habitats known to harbor rare species "were closely
inspected. Preliminary floristic information was
available at Medicine Lake NWR as represented by a
large mounted set of Refuge plant specimens. They
were examined for possible rare species and as habitat
indicators. In addition, a running list of the vascular
flora "was maintained over the course of the site visit,
and species that could not be positively identified in
the field were collected for later determination.
Specimens have been deposited at the University of
Montana (MONTU).

The main floristic references used included the Great
Plains Flora Association (1977, 1986), Dorn (1984),
and Hitchcock and Cronquist (1973). Nomenclature
used in this report is generally consistent with these
references, except for revisionary taxonomic
treatments, mainly for grasses (Kartess (1994i). This
means incotporating unfamiliar-sounding names for
some dominant and indicator species, particularly the
grasses in the Triticeae (Agropyron and Klymus in the
traditional sense). Thus, western wheatgrass, called
Agropyron srmthzi in Booth (1950) and Efymus srmthzi in
Dorn (198-4?), is referred to as Pascopyrum srmthii
(Tables.)

Table 1. Target list of Montana plant species of special concern in the study area

SCIENTIFIC NAME
COMMONNAKm

COUNTY

GLOBAL/

STATE

RANK'

NO OF
OCCURRENCES IN

COUNTY(IES) vs
INSTATE

Cyperus schweinitzn
Schweinitz' Flats edge

Sheridan

G5S1

1 /*

Ehdea hTigi':jagmata
Long Sheath Watervjeed

Phillips

G^G5Sl

e/*

Lobelm spicata
Pale-spiked Lobelia

Sheridan

G5SH

1/e

Mirabihs hirsuta
HairvFour o'clock

Sheridan

G5S1

1 ■ i

FJmcelm thermahs
Hot Spring' Phacelia

Garfield,
Phillips

G3G^ Si

e/ 3

FlagiGhothrys kptodadus

S 1 ende r-b ran ch ed P op corn-fl o w e r

Phillips

G^Sl

1 / s

Fsihcarphus hreinssimus
Dvj arf Wo o 1 ly-heads

Muss el Is hell,
Phillips

G5S1

^/7

Scirpus hetsrochaetus
Slender spikerush

Sheridan

G5S1

1 ■ 1

Sohdago sparsifiora
Few-flowered Golden rod

Garfield

GPSl

1/5

'Species and communities are evaluated and ranked by the Heritage Program on the basis of their global (rangewide)
status and their state (statewide) status according to a standardised procedure, using the following set of values and
accompanying definnitions.
Rank Definition

1 Critically imperiled because of extreme rarity (5 or fewer occurrences, or ^i^ry few remaining individuals) or
because of extinction-prone factors.

2 Imperiled because of rarity (6-20 occurrences), or because of other factors making it demonstrably vulnerable
to extinction.

3 Vulnerable because of rarity (21-100 occurrences) or found in a restricted range.

■4 Apparently secure, though it maybe quite rare in parts of its range, especially at the periphery.

5 Demonstrably secure, though it may be quite rare in parts of its range.

H Known only from historic records; possibly extirpated but concerted searches have not been conducted

8

Table 2. Synonyms among scientific names for dominant gTaniinoids

Coitiiiioii name

Bootli (195Q)» Great
Plaius FLoL'a Assoc.

(19S6)

Doru(l9S4)

Kartcsz (1994)

Western "'."v he at grass

Agropyron srmthii

'Elymus swi tJrn

Pasco pyrum srmthii

Bluebunch wheatgrass

Agropyron spicatum

Elymus spicatvs

Pseudoroegnena spicata

Sun sedge

Carex pensyha mca

Carex pensykja mca

Carex inops ssp. h^hophzla

Green needlegrass

Stipa vtndida

Siipa wndida

I'Jasella vindvla

tn this report, we have crosE-referenced each species
by both scientific name and common name the first
time the species is mentioned under each heading, and

by scientific name throughout the remainder of the
section. Appendix G is added as a synopsis of

common, scientific and six-character acronyms.
Common names are based on the list developed by the

U.S. Forest Service ofRegion 1, generally consistent
with major floras and the USFWS (Dittberner and
Olson 1983).

RESULTS

The fifteen Research Natural Areas and h-vo additional
study sites encompass over 60 plant associations,
including four that are potentially globally rare. The
plant associations provisionally identified as
significant representations of globally rare habitats
include:

• Douglas fir / littleseed ricegrass forest

{Pseudotsuga Tnenitissii / Chy^^psis rmcrantha
Forest) on Two Calf-Douglas-fir RNA of C. M.
Russell NWR

• Rocky Mountain juniper / Wyoming big
sagebrush Shrubland (JuTizpenis scopulorum/
ArtsTmsia tndetdata ssp. luyormngensu Shrubland
onFourthRidgeRNAofC. M.Russell NWR

• Porcupine needlegrass — thickspike wheatgrass
grassland (Shpa curhseta — Elymus lanceolatus
Herbaceous Vegetation) on Teepee Hills RNA of
Medicine Lake NWR

• Indian ricegrass / lemon scurf-pea barrens
[Chy^^psis hytnenoides / Psoraisa ianceolata Sparse
Vegetation) on Medicine Lake Sandhills
Wilderness area of Medicine Lake I'JWR

population of hotspring phacelia {Phacdia tkermalis)
was relocated on York Island. The records for the four
species that are tracked are presented in Appendix E,

and illustrations of them accompanied by descriptions
are presented in Appendix F.

This new information was added to prior information
in order to chai'acterise each RI'JA by its primary
biological attributes among the RNA criteria in the
Refuge Manual. These also include the composite
significance of landscape gradients, environmental
processes, and biological processes, whether they are
separate from or complementing single species and
plant association features.

Ten more plant associations are outstanding examples
of more common habitats. Each of the state- and
globally-significant plant communities are bold-faced
in the following table (Table 3. Matrix of plant
communities/ associations by Research Natural Area.)
This table represents all vegetation sampling
conducted in the course of the study for documenting
plant community biodiversity significance. Most state-
and globally-significant features are highlighted in
photographs presented in Appendix C, and all
vegetation sampling data is documented in constancy-
cover tables in Appendix D.

Two Montana plant species of special concern were
documented on the Big Island RNA, including plains

phlox (Phlox a.7id,icola.; G6 S2) and hairy four o'clock
(Khrahihs hirsvia G5 S3). Both species also occur in the
Medicine Lake Sandhills, along with two additional
rare species, Fendler cat's-eye (CTyptanthaJendlen ; G^
Si) and Schweinits' flatsedge (C)^^TT^ jirfee'^^^^fe^; G5
S2). Each of these is a "■-"v ide spread species but rare
from a state perspective. The Sandhills have the
highest number of rare plant species among the study
sites. We note that the Big Island and Medicine Lake
sandhills field evaluations provided the basis for
changing the status o^Mirahihs hzTsuta from a species
of special concern to "watch. In addition, the known

10

E

i

i

01

C
CD

ir

T3
<U

C
TO

©

a>
at

(U

-3J

l:
o

"id

Z

i3

w

X

[0

5^

0)

3

: X

[A

c:
B
c
o

5
<

I

2:

cc

ID
CO

a>
IT

><

CO

J5
o
o

3

E
E

m

Cl

CO

ro

2
3
ra

o
q:

C

o

:3
a;

JZ

O

X

X

X

&<

><

X

4

O

■^

■^

>^ E

If

CO OJ

<r m
5! ™

a

w

■a
o
o

n
o

I

.1)

I

.t

£

s

t

■S

n

c

1
2

!^ ^

fn ^ .(□ e

I

ra

.9-

ij}

g

I

1

li

II

i

t

,1

en

I

s

I

I

mti

•li 3

si

II a.

SfO

_- I'

1^
5 LO

^^"§
f s°

"St

m

I

2

pi

c 0. a c
Z ^ I il

™ 1-03 ■£.
■^ F " E
?^- ?

I

•9-

:: i3 ffl *

— f-i ^ I'i

11

The following pages present a summary of all
ecological and botanical data collected in the field, in
addition to observations and much background
information assembled for interpreting results.
Background information includes description of
environment, land use prior to and subsequent to
National Wildlife Refijge establishment, and
management comments as preliminary identification
of management concerns or questions associated with
ecological and botanical features.

An overall statement of biodiversity significance has
been drafted, building on previous RWA information.
Other non-biological values are also cited much as

they were addressed in the original RMA records.

This background information is all the more
important and difficult to compile in light of the
dynamic nature of the Great Plains vegetation, and
the absence of precise vegetation information for
reconstructing landscape conditions. The references
that are made to fire and grazing in the following
pages as historically widespread factors that shaped
the landscape are based on such works as Higgins
(1986) and Umbanhowar (199S) for fire, and on such
works as Hanson (19S4) and Peden et al. (1974) for
bison grazing. This is made in flill recognition that
there are different theories on how these apply to
current landscapes and management practices. One of
the common methods for investigating this is through
vegetation manipulation experiments with a control.
Grazing studies have often used comparative
vegetation sampling inside and outside exciosures, as
with a recent Rocky Mountain study of grazing affects
that included study sites on the Charles M. Russell
NWR(Stohlgren et al. 1999). The reader is referred to
such works in the management literature, and the
management notes that are included in the following
pages are rudimentary context for the vegetation data.

The study sites are sequenced alphabetically by refiige

name, and alphabetically by RNA name within
refiiges. Plant associations are described as they occur
in each RNA. They are sequenced by relative
extensiveness within the RMA, listing the most
widespread plant associations first. The classification
and characterization of major plant associations is
derived from vegetation plot sampling data. The plots
provide basic documentation of the existing
vegetation, and provided a basis for considering their
classification as well as their condition. The plot
information regarding species composition is arranged
in "synthesis" tables (Appendix D) in the same order
of presentation as in the text. "Constancy/ cover"
tables are also included to convey the variability
across a community type. Finally, we note less
extensive plant associations and provide qualitative
description.

12

Benton Lake National Wildlife Refiige

MuLLAN Trail Research Natural Area

ENVIRONlvlENT:

The Mullan Trail RNA is a 392 acre segment of
gently rolling terrace and fan landforms associated
"with Glacial Lake Great Falls. There are no surface
drainage features, and it lies within the closed-basin
topography surrounding Lalce Benton. The limited
relief in elevation ranges from 3628 to 3650 ft. Soils
have developed from alluvium and lacustrine deposits,
made up of fine-textured clays of the Pendroy Series
(h-YO map units represented) with slow to moderate
rates of runoff The semi-arid continental climate has
peak precipitation in May followed by June (Climate
data from Great Falls, Western Regional CI imate
Center).

"VEGETATION:

The vegetation is well-developed and relative
uniform, made up a single grassland plant association.
The overall visual impression is of homogeneity
across the "sea of grass." It is an island of intact
natural vegetation, a fragment of a formerly extensive
type, now surrounded by agricultural lands and
tamegrass.

Pascopynim srmthii —I'Jasella vindula Herbaceous

Vegetation

CPASSMI-NASYIR^

western wheat grass — green needlegrass grassland

This is the one major plant association present in the
RI'JA. Its composition differs from place to place
■within the RNA but the two plots established at the
far ends of the area evidence a high degree of
similarity in

both composition and cover by the dominant species.
Western \"vheatgrass (PascopymrnsTtutkii)^ green
needlegrass [Utssetla vtrtdula), prairie junegr ass
[Kodena macrantha) and narrow-leaved sedge {Ca.rex
stsnophylla) are the dominant graminoids with this
component's total canopy cover ranging around 70%.
The forb component is low in species, with three that
are more common than all others: poverty-weed (Iiia
aj^llaTis), plains bahia (Pi^Tradeniopsis opposihfoha), and
scarlet globemallow (Sphaeralcsa coccinsa). It is
possible that foiis numbers and flowering levels are
suppressed by litter accumulation, but this association
as found on heavy soils is intrinsically limited in foii>
diversity. Patterns of variation from place to place

within the area may correspond with land use
differences or small-scale natural disturbance such as
burrowing animals.

The PASSMI-NASVIR plant association occurs in

Montana, North Dakota. South Dakota, Wyoming,
Saskatchewan, and southwestern Manitoba. It is
ranked G4 by TMC (Schneider et al. 1997). Nasella
vindula is both more palatable and more sensitive to
grazing than Pascopyni/m STmthzi and also has a
narrower ecological amplitude. In the Yellow Water
Triangle area Jorgensen (1979) notes the indicator
value o^Uasella vindvla for recognising sites with a
higher soil moisture status, such as swales, toeslopes
and moist terraces dominated by silver sage (ArtsTmsia
cana).

There is a need to refine U.S. and Canadian vegetation
classification as it involves this t3^e. Based on a study
of relict and near pristine sites, Coupland (1961)

identifies a porcupine grass — thickspike wheat grass
grassland (Stipa curiissta — Agropyron dasystachyuTn
Herbaceous Vegetation) as the major grassland type
on dark brown and brown soil zones of southern
Canada, essentially the prevaling mesic sites in
landscape. Coupland notes that south of the ■4?9'^
parallel the importance value of porcupine grass [Shpa
curhseta includes some or all of what has been treated
as Shpa spartea in Montana) as determined by cover
declines drastically and that Agropyron dasystach^m
(synonym; Klymus lancsolatus') exhibits a gradual
decrease as well. Shpa cwriiseta was noted as dominant
elsewhere in the RNA system at Tepee Hills. DeVelice
et al. (1995) documented the importance of a PASSMI
- NASVIR association across the northern tier of
Montana counties. They recognised PascopyruTnsTmthzi
and Kly?nus lancsolatus as ecological equivalents for site
identification and noted the difficulty of field
discrimination of these two species based on
vegetative or reproductive characters. |^Plots
NHMTECMT97SOD001, NHMTECMT97SCOOOq

OVERALL BIODIVERSITY SIGNIFICANCE;
The Mullan Trail RNA represents a good quality
occurrence of what may have been a common if not
the prevalent plant association of the Hi-Line
(Glaciated Plains Section) of Montana under
presettlement conditions. It is a mesic, productive
grassland type of the Great Plains biome, and it is
estimated that over 90 % of its original p re-settlement
extent has been plowed. While this is not globally

13

rare, few other occurrences are protected and
documented on public lands in Montana, and they are
smaller or lower quality/ condition. The absence of
surrounding natural vegetation does constrain options
for landscape-scale management if not also its value in
having landscape context. Nevertheless, it provides a
good rangeland reference and ecological baseline.

The RNA may contribute habitat to previously-
documented upland grassland bird Species of Special
Concern including Ferruginous Hawk, Burrowing
Owl, Loggerhead Shrike and Baird's Sparrow, but
does not contain the wetlands that provide primary
habitat for the waterfowl and colonial nesting birds
found elsewhere on the Refuge. Wildlife values were
not evaluated.

OTHER VALUES:

The RMA also preserves a segment of the Old Mull an

Trail, part of a 6^9 mile wagon road linking the

western-most navigable waters of the Missouri River
at Fort Benton with the eastern-most navigable
waters of the Columbia River at Walla Walla,
Washington.

laot:) USE:

Prior to and afler refijge establishment in 1929, the
area was grazed as a part of a large common grazing
allotment. A summer-fall season grazing permit
system was instituted in the 19^0s. After the refuge
was staffed and facilities developed in the early 1960s,
a new grazing management plan provided for a much-
reduced level of summer and fall grazing. Although it
is not possible to determine the exact grazing regime
applied to the RNA, grazing on the whole refuge
dropped from about 2,700 animal unit months
(AUMs) in 1960 to 1,631 AUMs in 1966. In 1976,
livestock grazing was terminated on the refiige, and
the area has been rested since that time. The existing
composition suggests that is was part of secondary
range or more likely a relatively recovered primary
range in good condition.

Since the time of RNA establishment, there has been
at least one experimental fertilizer application over
undefined segments of the area. Refiige records
indicate that it did not have the desired effect of
increasing productivity or stand structure, and was
discontinued. Records do not specify treatment area»
application concentrations, or include monitoring.

MAMAGEMENT COMMENTS:
Exotic species are uncommon at present. Although
both are present, populations of cheat grass (3ro7mis
teciorum) and intermediate wheatgrass (AgrGpyrvfi

inUrmediuTri) are at low levels within and outside the
RNA. The very aggressive yellow sweetclover
(Klehlotus officinalis) and crested wheatgrass (Agropyon
cnstabiTri) probably pose greater threats in the long
term. A narrow band of encroaching crested
wheatgrass is found along the road grade disturbance
zone along the west boundary of the area.

Historically, fire and bison grazing were two major
driving forces in this landscape, responsible for
renewing the vigor of the grasses, stimulating forb
numbers, and keeping shrub density low.
Re introduction of appropriately timed fire is a
management option to consider in containing nearby
weed populations and stimulating forbs; realizing that
it can help control or increase 3ro7nus tectomm a.r\.d
Melilotus officinalis depending on conditions.

14

Charles M. Russell National Wildlife Refiige

Fourth Ridge Research Natural Area

ENVIRONlvlENT:

The Fourth Ridge RNA spans 1,^80 acres
representing one among a repeating series of shale
ridges at the northeast end of Fort Peck Reservoir.
Outcrops of Bearpaw Shale are exposed at the surface.
Soils have developed from this parent material and
thus are heavy-textured "■-"vith clays predominating.
The shale outcrop landscape is predominantly gently
rolling with parts of the landscape prominently
erosion-sculpted with pitches and rolls that would be
registered only on a large-scale map. The semi-arid
continental climate has pealc precipitation in June
followed by July and May (mean annual precipitation
of 1 1 .6 inches; cl imate data from Fort Peck Power
Plant. Western Regional Cl imate Center, 1 956-1 997).

"VEGETATION:

The vegetation is made up of two extensive upland
plant associations that compose a mosaic of shrubland
and open woodland. The RNA does not include
ponderosa pine [Pinus ponderosa) vegetation types as
indicated in the establishment report, raising the
question of whether boundaries need to be reviewed.

ArtsTmsiiZ tndentata siip "ivyoTmngensis / PascopyruTn

srmtJm Shrubland

CARTTSW / PASSMI3

Wyoming big sagebrush / western wheatgrass

shrnjbland

This is the prevailing vegetation type on Fourth
Ridge RNA. Its occurrence is close to defining the
northeastern-most distribution of big sagebrush
(ArtsTnisia tndentata) as a species and as avegetation
type in North America, regardless of subspecies
(Shultz 1 98^). This shrubland occurs on benches and
gentle backslopes with fine-textured soils (silty clays
to silty clay loam) weathered from shale and claystone.
The amount of bare ground and litter is inversely
related and highly variable, perhaps depending on past
grazing history. The shrub layer is dominated by
Wyoming big sagebrush (ArtsTmsia tTidentnita ssp.
•iiTyoTmTigensis; the Great Plains subspecies) but total
canopy cover ranges between 10-20 %, seldom
exceeding 25 %, so that according to the National
Vegetation Classification Standard (1997) this
community is technically grassland with a shrub
component. The dominant and diagnostic grass is

western wheatgrass (Pascopyrumsrmthii) "with
subordinate graminoids like threadleaved sedge (Carex
Jihfoha), Sandberg's bluegrass [Poa sscanda)^ and
junegrass (Koehna moiTrantha) attaining only a fraction
of the iiQ % plus canopy cover of the dominant
graminoid. The highly palatable green needlegrass
{I'Jasdla vindula) is present in only trace amounts.
Forb diversity is low, not surpassing 15 per plot and
individual cover values seldom exceed trace amounts:
American vetch (Vicia aTnencand)^ white onion (Alhutn
textile), bastard toadflax (CoTrwiandra urtd>ellata)^
prickly pear (Opuntia polyacantha\ and yellow
sweetclover (Klehloius officinalis) have high constancy
in the community. Alehlohts officinalis is uncommon
and widely scattered in this type, and may be
increasing. ^Plots NHMTECFR97SC0001,
NHMTECFR97SC0003, NHMTECFR97SC00063

Junzpems scopudoruTn /Ariemzsia tndentata ssp.

wyormv^erisis Woodland

CJUNSCO / ARTTSW;]

Rocky Mountain juniper / Wyomingbig sagebrush

woodland

Rocky Mountain juniper / Wyoming big sagebrush
woodland (JuTupents scopuiorum / Artsrmsia tndentata
ssp. zvyormngensis Woodland) is an extensive t3^e
within this landscape, generally occurring on higher
positions with silty clay soils derived from one of the
subsidiary shale members of the Bearpaw Shale.
JUNSCO / ARTTSW is ranked globally imperiled
(G2; Schneider et al. 1997) and is cited to occur only in
Montana, Wyoming and Colorado. It is noted to
grade most frequently to the ARTTSW / PASSMI
community type, which occupies similar positions in
the landscape. Sometimes the difference in these
communities may reflect past disturi>ance, such as fire,
but the mosaic pattern at Fourth Ridge as it
corresponds with gentle dips may indicate edaphic
microhabitat differences. The Bearpaw Shale includes
mostly non-calcareous members but also has
calcareous and ben ton it ic shale beds.

Junipems scopulorwm is the only tree present, occurring
as short-statured and highly branched forms and in a
rather clumped distribution. Canopy height ranged
from 5-10 fl. At Fourth Ridge, as elsewhere along this
area of the Missouri River, its growth form is rounded
and generally without a central axis. It is not known if

15

the peculiar growth form is genetically- or
environment ally- induced. Its sporadic distribution
challenges accurate estimates of canopy cover, which
range from 15-30 %. placing these stands, according
to the parameters of the National Vegetation
Classification Standard, in both the woodland and
grassland categories. The relatively species diverse
shrub layer is dominated by Wyoming big sagebrush
(Artermsia indsntata ssp. ^^yowin^snsis), whose variable
canopy cover appears to be a function of competition
with neighboring trees. Graminoid and forb cover
vary depending on aridity, with western wheatgrass
{Pascopyru-m svutJni), prairie junegrass (Koelena
TnoiTrantha) and yarrow (Achillea Tmllefolium) in the
widespread, less arid conditions, and little bluestem
(Schz%acktyium scopanum), sun sedge (Carsx iTiops), and
few-flowered wild buclcwheat (Kriogonurn paitcifloru/m)
in driest places.

Piiccindha nuttlalhaTia Sparse Vegetation

CPUCNU'ri

Nuttall's alkaligrass barrens

This association is sparsely-vegetated with Nuttall's
alkaligrass (Piu:cinnelha nuttalhi) as dominant,
occurring as a broken stringer along an intermittent
drainage that feeds into Third Coulee. It constitutes
the vegetation band closest to the incised channel on a
floot^lain position with silty loam alluvial soils; salt
efflorescence was not observed but this community is
loiown to occur on salt-affected soils that have a
slightly wetter, temporarily inundated, moisture
regime. Within the TNC tracking system this
community type has been reported only from
Colorado as Gl? but Heidel and Cooper (1996) have
documented it from western plains of Montana near
the Rocky Mountain Front, noted it in field
reconnaissance, and cited it from the Canadian
literature (synonym: P'u^cinnellia airoidss, Dodd and
Coupland 1966).

The Fourth Ridge example of this type has low
diversity and is compositionally very similar to other
observed Montana occurrences with Piiccinelha

nuitalhana dominant at around ^0 % canopy cover;
inland saltgrass {DistiMts spicatd) and povertyweed
{Iva accdlans) are the only other forbs exhibiting more
than trace coverages. This community grades to
Tkshchihs spicata-dormn^l^d sites on drier posit ions.
Wyomingbig sagebrush [ArteTmsia tndentata s^.
-^yormngsnsis) in trace amounts was the only shrub
noted within the plot. Within the channelway, yellow
sweetclover (Mehlolus qfficiTiahs) was noted as forming
extensive, virtually unbroken s'-vathes in the same

position as PUCNUT and extending to the drier
Distichzhs stncta association positions as well.

Downstream from the PUCNUT sampling site a
comparable landscape position was occupied by what
has been described as western wheatgrass — inland
saltgrass grassland (Pascopyri£7n srmthzi — Distcklzs
spicata Herbaceous Vegetation; G^; WY, ND). This
type has not been formally described from MT, but
probably has been subsumed to date within the
Tkshchilis sincta or Pascopyrwm srmthit community
types of Hansen et al. (1996). |^Plot

i^JHMTECFRgrscooo^;]

Calamovilfa longifoha — Carex tnops

Herbaceous Vegetation

CCALLON - CARING^

prairie sandreed — sun sedge grassland

There are sites occurring as tiny woodland openings
at higher positions in the landscape that appear to be
developed on a more erosive shale member that
weathers to a fissile texture (functions as sandy soil
analogue) and may be acidic in its reaction. These
sites have a high percent of exposed soil (in excess of
80 %), a much reduced vegetation cover and the
composition in dominant vegetation is highly variable
across the landscape. They are in erodible settings,
which complicates interpretation. The sample plot
appears to be most similar in site and vegetation
parameters to the prairie sandreed — sun sedge
grassland {Calarriovilfa loTigifoha - Carsxtnops
Herbaceous Vegetation) that has been identified for
southeastern Montana (Hansen and Hoffman 198S).

The vegetative aspect is dominated by rhizomatous
graminoids, sun sedge {Carsxinops) and Cala-movilfa
longifoha (prairie sandreed) with plains reedgrass
(CalaTnagrostis montajienszs) and PiiscopyrwrnsTfutkii just
exceeding trace amounts. We hypothesise that an
acidic reaction of the substrate is reflected in the forb
component dominance by few-flowered buclcwheat
(Knogomim paucifloTUTn). Shrubs like prairie rose (Rosa
arkansana) and trees 1 ike Rocky Mountain juniper
(Juniperus scopvlorum) constitute less than 3 % canopy
cover and their population structure does not indicate
a change in their contribution. I^Plot
I^JHMTECFR97SC00053

OVERALL BIODIVERSITY SIGNIFICANCE;

Fourth Ridge RNA features a woodland community
dominated by Rocky Mountain juniper (Junipsrus
scopvlorum) in good condition It is part of one of the
most extensive Rocky Mountain juniper woodland
stands in the Great Plains portion of the state, and

16

near the northern 1 imits of its distribution. It also
represents JuTupents scopulorum as dominant in a low,

rounded growth form. It is not known if the peculiar
growth form is genetically- or environmentally-
induced, i.e, whether the plant association is
appropriately recognised as discrete from all others.
The rank, may be elevated accordingly.

The Wyoming big sagebrush/ western wheatgrass
shrubland {ArieTmsia tndentata ssp. zuyoTmngensis /
Pascopyrum srmihzi Shrubland) is also near its
northernmost extent, subject ofbiogeographic
interest, and in notably good condition. The NuttalTs
saltgrasE barrens {Pitccitmsllia tiuUalliaTia Sparse
Vegetation) signifies an under-documented vegetation
type of the northern plains. Wildlife values were not
evaluated. Overall values are enhanced by the
continuity with native vegetation on all upland
borders.

LAUD USE;

The area is grazed by livestock. The current condition
suggests that it is part of secondary range or a
grazing regime that maintains good ecological
condition. The area lies north of The Pines Recreation
Area- Signs of recreational use that were noted include
hunter and limited OHVuse.

MANAGEMENT COMMENTS:

There are few exotic species in this habitat, with the

exception of yellow s'-veetclover (A'ldilotus qfficimdis).

It is widespread but sparse throughout most of the

area, and particularly abundant along the ephemeral

watercourses.

There were no signs of fire in the landscape. Fire is
lethal to Juniperus scopidoruTn under most conditions, a
species that is highly-combustible whether it is dead
or al ive.

Hell Creek Potential Research
Natural Area

ENVIRONlvlENT:

This area includes rolling uplands at the head of Cold
Turkey Coulee but could certainly be expanded to
include some of the surrounding highly dissected

Missouri Breaks terrain. The climate is essentially
Continental (refer to the characterization of the
Missouri River Bottomland on the basis ofHaxby IS
SWandRoyS^NE Mobridge, Montana.) All of the
landscape is underlain by sedimentary formations. The
highest have sandstone caprock and the rest are
various shale and mudstone members that weather to

fine-textured soil. Thin, carbonate rich lenses occur
sporadically.

VEGETATION:

Hell Creek is an area notable for the fact that a

relatively recent wildfire has burned much of the
upland, rolling portion of the landscape and removed
the once-dominant Wyoming big sagebrush [Arteimsia
tndsntata ssp. "iayoTniTigensis). The terrain surrounding
the uplands portion is highly dissected and erosive,
dropping off into badlands, and supports primarily
Pondorsa pine-dominated t3^es, including ponderosa
pine / sun sedge (Pi nus ponderosa / Carex inops) and
badlands slopes with sparse shr^ab cover.

PascopyrwmsTmthii —I'Jasdla z/zndida
Herbaceous Vegetation

[:passmi-nasvir;]

western wheatgrass — green needlegrass grassland

Much of the landscape bordering the Missouri Breaks
is believed to have been occupied by Wyoming big
sagebrush / western wheatgrass — green needlegrass
shrubland (Artermsia tndentata ssp. •'jjyormngensis /
Pascopyrwm STmthzi — I'Jasella vindvla) that has been
burned. This had the result of killing all of the A.
tndentata, sometimes completely consuming the crown
and main stem to ground level, and leaving the
landscape dominated by grasses as a serai community.
The upland component of this landscape is
characterised as gently swelling benches to
moderately rolling lands with many different
exposures, all of which support this plant association,
making it a prevail ing type. Soils are derived from
fine-grained sedimentary strata (shale?) and are
primarily silty clay loams.

The length of time since fire is difficult to determine
but most of the landscape that once supported
ArtsTmsia tndentata ssp. zvyoTmngsnsis as a dominant, as
inferred from density of sagebrush skeletons, is only
very slowly returning to that status. No ArtsTmsia
tndentata seedings were found on the plot and only the
merest traces of fringed sage {Artermsiajilifoha) and
broom snakeweed [Guiierrsi^a sarothras) were noted.
The grass component strongly dominates this serai
phase; western wheatgrass {PascopymmsTnithii) is
relatively evenly distributed throughout the stand and
its cover (currently ■4?0-5O%) may still be increasing
following the bum (see Hansen and HofTman 1988 for
a comparison of grass production with and without A
tndentata). The appreciable cover of green
needlegrass {I-Jasella z^ndida) indicates the relative
mesic, productive nature of this site. Threadleaf sedge

17

(Carexfihfoha) and bluebunch wheatgrss
(Pseudoroegnsna spicata) are also important grasses in
the plot and across the stand. The forb component is
diverse but no one species is represented by more than
a trace. Of the native forbs, prairie smoke [Gewm
triflorum), shaggy fleabane (Kngeron purmhs), scarlet
globemallow {Sphneralcsa (XiCciTted) and dotted blazing-
star {hiains punctata) appeared to the most
consistently distributed across the landscape. I^Plot
WHMTECRN98SC0011;]

ArteTmsia tndentata ssp. "iayoTmngsnsis / Pseudoroegnena

spicata Shrubland

CARTTSW/PSESPi;]

Wyoming big sagebrush / bluebunch wheatgrass

shrubland

The representation of this association is highly
dependent on the extent of coarser-textured
substrates. Within the Hell Creek area this
association is found in small patches confined to the
uppermost portions, usually having vjestem or
southern exposures, of gentle slopes that are capped
"■.vith a sandstone member of the local mix of
sedimentary strata. Soil texture ranges from fine
sandy loam to fine sands.

Wyomingbig sagebrush [ArteTntsia tndentata ssp.
-MyormTigensis) dominates the shrub layer but its cover
is generally not sufficient (20% or less) to place these
stands as shrublands in the national classification.
Fringed sage {ArtsTmsaJhgida), 3Mcca {Tiu^ca glauca)
and fragrant sumac (Rhus aroviatica) are regularly
present with cover usually less than 1 or 2 percent.
Within the plot, threadleaf sedge {Carexfilijoha)
rather than bluebunch wheatgrass [Pseudoroegnena
spicata) is the dominant graminoid. but across the local
representation of this type dominance shifts among
three graminoids, also including needle-and-thread
(Stipa comata). This description of the type differs
from that of Hansen and Hoffman (1988) for
southeastern Montana in which Pseudoroegncna spicata
is uniquely dominant and may be an artifact of the
dissected terrain or it reaching the margins of its
distribution. The foii> component is shared with the
adjacent PASSMI - NASVIR community, with the
exception of silver-leaved scurf-pea (PsoraUa
argophylla) a species v/ell loiown to favor sandy
substrates. |^Plot NHMTECRN98SC0012^

Other VegetationT3^es: Little bluestem
(3chz%achynum scopanum) is a localized dominant on
coarse-textured knolls in the area.

The presence ofponderosa pine (Pitius ponderosa) in
surrounding lands is taken to represent the Pinus

ponderosa / Cartx inops Woodland. In addition, limber
pine {Pirms Jisjzlis) is known from ridgelines with
lenses of calcareos substrates in Hell Creek State Park
to the east, and could possibly occur in the potential
RNA area.

OTOIRALL BIODIVERSITY SIGNIFICANCE;
The Hell Creek uplands present a well -developed,
moderately extensive example of a productive,
widespread plant association of western wheatgrass —
green needlegrass grassland (Pascopyrwm srmthzi —
I'Jasdla vind-ula Herbaceous Vegetation) in excellent
condition. This area lies at the border between gentle
plains and Missouri Breaklands, and warrants
evaluation for its representation ofboth segments of
the Great Plains biome as well as the landscape
gradient. The habitat continuity with all of this
surrounding unglac iated terrain contributes to its
ecological value.

Small cage exclosures were noted, and it is possible
that this area is already being used in studies of fire
response or wildlife utilization. If not, itwouldbe
worthwhile to compile wildfire history information for
use of this otherwise well-suited area as a laboratory
for studying natural succession. Wildlife resources
were not evaluated.

LAI^JD USE:

The area has been part of a grazing allotment as

secondary range. It is not currently grazed. Though it

is isolated, it receives use by hunters, if not other

visitors.

MANAGEMENT COMIvlENTS:
The area is notably free of yellow sweetclover
(Mehlotus officiTialts). No exotic species management
problems were identified, though Japanese brome
(BroTnus japomcus) is present at low levels in all
communities across this landscape

Limber Pine Research Natural
Area

environment:

Limber Pine RNA encompasses 1,053 acres
representing a cross-section of Missouri River Breaks
habitat developed on residual soils weathered from
shales and non-calcareous sandstone mainly of the
Fox Hill Sandstone. It includes all of a large ravine
system incised to a maximum of about 300 feet, fed by
small springs, and emptying into the backwaters of
Fort Peck Reservoir. Extensive grasslands with
scattered outcrops span the upland benches and

18

exposed ravine slopes, Ehnjb-doniina.ted communities
are parts of the ravine, and small, scattered woodlands
are characteristic of nor th- facing ravine slopes and
segments of the narrow bottoms. The semi-arid
continental climate has peak precipitation in June
followed by July and May, and a mean annual
precpitation of 1 1 .6 inches (Climate data from Fort
Peck Power Plant. Western Regional Climate Center,
1966-1997).

"VEGETATION:

Represented on this site are at least three major
grassland plant associations and many other types of
small size or restricted ravine habitats. They readily
sort by topographic position, slope, and aspect; but the
highly dissected nature of the setting fosters a
compl icated vegetation pattern.

Shpa coTnata — Boutdoua gracilis — Carexfihfoha

Herbaceous Vegetation

CSTICOM - BOUGRA - CARFIL;]

needle-and-t bread — blue grama — thread-leaved sedge

grassland

This grassland association occupies the rolling

uplands and upland benches with well-drained soils
derived from sandstone; it also occurs on moderate to
steep slopes, usually those "with a southerly aspect.

This association is consistently dominated by needle-
and-thread (Stipa coTnata). Cover of the major co-
dominant species, threadleaved sedge (Carexjilifoha)
and blue grama (Bouteloua gracilis), is highly variable
and factors controlling this variation have not been
identified. Forbs constitute -^^ry little cover in this or
the following grassland types scarlet globemallow
(Sphaeralcea coccinsa), rush skeletonweed (Lygodesrma
juncea), and silver scurfpea (Psoralea argophylla) are the
foibs with greatest cover and constancy. Together
with the western wheatgrass — needle-and-thread
grassland (PascopyruTnsvuthzi —Shpa co-maia
Herbaceous Vegetation), they comprise the great
majority of the upland landscape. There was some
■western wheatgrass {F'ascopyru7ns7m^ai)'^v^s^r)\. in
nearly ^^^vy upland site. While there are
characteristically steep gradients from Pascopyrwm
j?rai/!Z2-dominated sites to those dominated by Shpa
<XiTnata in western Montana, these gradients are difRise
in eastern Montana. The break-point coverage
between these two associations is placed by Hansen
and Hoffman (1988) at the point where dominance (in
terms of canopy cover) shifts from one to the other
principal species. ^Plots NHMTECCR97SC0001,
I1HMTECCR97SC0OO2, NHMTECCR97BH0003;]

Pascopyrwm swithii — Shpa comata Herbaceous

Vegetation

CPASSMI-STICOM]

western wheatgrass — needle-and-thread grassland

This is the other major grassland association within
the RNA; it occurs on benches, concave topography to
swales and is associated with slightly finer-textured
soils (silt loams or finer, usually shale-derived) than is
STICOM-BOUGRA-CARFIL. It grades to the
STICOM- BOUGRA -CARFIL type of drier
exposures, coarser textured soils, and under intensive
grasing pressure. In more moist positions, such as
swales, it grades to the PASSMI - NASVIR
assoc iation.

CalaTnovilfa longifoha — Carex inops

Herbaceous Vegetation

[CALLON - CARING^

prairie sandreed — sun sedge grassland

I^Plot NHMTECCR97BH0002^ This community

type constitutes the most extensive vegetation on

exposed sandy slopes, though its representation on the

RNA has very reduced vegetative cover and much

more exposed substrate compared to literature
descrptions of the t3^e (Hansen and Hoffman 1988,
DeVelice et al. 1995). It borders on the RHUARO /

PSESPI and STICOM -BOUGRA - CARFIL
associations; oflen the ecotone beh.veen these types is

abrupt due to the rhizomatous nature of both

Calamovilfa longifolia (prairie sandreed) and Carex

inops (sun sedge), both t3^ically forming dense clones.

RNA examples of these sites are highly erosive and

this may constitute the difference between this type

and the adjoiningplant associations as well as explain

the differences between the Limber Pine RI'JA

expression of the type and those literature

descriptions of the t3^e. There are questions as to

whether an association should accommodate this much

variation in site parameters. Calamovilfa longifolia is

typically the site dominant, though shrub cover of

golden currant (Ribes aureum) and yucca (Yucca

glauca) may rival that of the graminoids.

Rhus aromahca / Psendoroegnena spicaia Shrubland

P^PIUARO/ PSESPQ
fragrant sumac / bluebunch wheatgrass shrubland

This community is associated with sandy, somewhat
unstable soils of the steep-slope ravines, particularly
southeast- through south west- facing exposures. It is

19

most often found as small stands extending from the
brow of the slope (where it grades to STICOM —
BOUGRA- CARFIL ofbenches) to mid-slope and is
occasionally ■weakly represented further downslope.
Total vegetation cover is low, seldom exceeding 25-^0
% and concomitantly the amount of base soil and rock
often exceeds 85 %. We speculate that these sites
differ from other associations strongly associated with
sandy sites (e.g. GALLON — CARFIL) by having more
exposed rock and gravel. Fragrant sumac (Rhus
aromahcd) shares dominance of the shrub layer with
yucca {Titcca glauca), and their relative proportions
shifting with no obvious environmental correlates.
The graminoid component is usually dominated by
low coverages (not exceeding 26 %) of bluebunch
wheatgrass (Pseitdoroegnsna spicatd) and considerably
lesser amounts of grasses associated with sandy soils
like Indian ricegrass [Ory^opsis h/menoides), little
bluestem [Schi%achfnurri scopariuTri), and prairie
sandreed (CalaTno'^lfa loTigifoha) Along with the
widespread rangeland forbs such as scarlet gaura
(Qiiiira coccima) and scarlet globemallow (Spkaeratcea
<^a^n£d), occur species that are restricted to sandy
sites like green milkweed [Asdepias 'Jindiflora), prairie
spiderwort (Tradescantia occidentahs) and nodding wild
buckwheat (Knogomim cernuuTn). |T*lot
imMTCR97SCOOCH'3

Jurapents scopxdorurfi/ Pseudoroegnena spicata

Woodland

CJUNSCO / PSESPI]

Rocky Mountain juniper / bluebunch wheatgrass

woodland

This woodland occurs in small patches on moderate to
steep northwest- to northeast-facing slopes from the
bottom of ravine slopes to midslope, with soils derived
from sandstone or interbeddings of sandstone and
shale/ muds tone. Some outcrops test positive for
calcium carijonate. Generally more than 70% of the
surface is exposed as soil and rock. These slopes are
moderately to highly erosive. Short-statured (less than
^-Q ft.) Rocky Mountain juniper [Jumpsnis scopulorum}
dominates the tree layer and generally its cover
exceeds 50 %, making it difficult to traverse stands.
Though representing some of the more mesic habitat
in the RNA, these are still stressful sites with
depauperate undergrowth. Bluebunch wheatgrass
{Pseudoroegnena spicata) and field milkvetch (Astragaliis
agreshs) are the only forbs occurring in greater than
trace amounts. This association is singular for the
occurrence of certain forbs, including false starry
Solomon's seal {STmlaciTia stellata), Missouri goldenrod
(Sohdago Tnissoune fists), and harebell {CaTnpanida
rotundifolia)

There is not a discrete pine woodland t3^e present in
the RNA, though pine trees are scattered across the
juniper woodland. Ponderosapine (Pinus ponderosa) is
widespread, but there are no areas where it is common
or dominant, as evaluated in studying aerial photos
and visiting areas of highest tree density on foot. All
probable locations were considered, such as north-
facing slopes that might have calcareous outcrops,
attempting to locate the limber pine {Pinus Jlexilis).
We found only P. ponderosa, though the search was
not exhaustive. Finding P Jlej:^l2s isplausible in light
of its presence in the Hell Creek State Park to the west
on Fort Peck Reservoir, and the Terry Badlands to
the southeast. Its presence here would signify an
intermediate location between other outlying stands;
however, failure to find it here does not diminish the
status of this RNA. ^Plot I'JHMTECCR97SC0005^

Chysothz?nmis riauseosus / Knogonu?n pa7/ci/lojiitn

Sparse Vegetation

CCHRNAU / ERIPAIJ]

common rabbitbrush / few-flowered wild buclcwheat

barrens

This small and localised community occurs on steep,
south-facing outcrop slopes at the bottom of the
ravine, representing a stressful and unique
environment. Soils exhibit salt efflorescence. Slopes
show signs of sheet and gullying erosion, with over
90% of the surface made up of exposed soil. A similar
vegetation assoc iation has been described by Branson
et al. (1970) in Valley County and by Vanderhorst et
al. (199S) for Carter County; both of their studies
indicated acid shales as the determinant of the unusual
and depauperate vegetation. The examples from the
literature occurred on gently rolling terrain whereas
this type was only represented on steep slopes on the
RNA Sites are species poor (<20 species) and total
vegetation canopy cover does not exceed 30% with
dominance shared by common rabbitbrush
(ChfysotkaTnniis nauseosus), few-flowered wild
buclc-"vheat (KriogonuTn paucifloruTri), and bluebunch
wheatgrass (PseudoroegneriG spicata).
CPlotNHMTECCR97BH0001^ I^Plot
NHMTECCR97BH00023

Jumpems hon^ontalzs / Pseudoroegnena spicata

Shrubland

CJUNHOR / PSESPI]

creeping juniper / bluebunch wheatgrass shrubland

This plant association has been reported in the Little

Missouri River badlands (Jensen et al. 1992) and
previously noted in the county in the course of

baseline botanical work (Heidel 1994), but it has not

20

previously been documented in Montana. This type
occurs in small patches on the RNA in relatively
broken topography in north-facing coulee settings.
Creeping juniper (JuTizpsrus hon^^ntalis) is documented
to occur with bluebunch wheatgrass [Psetidoroegnena
spicatd) and a number of other graminoids such as sun
sedge (Carex inops) and threadleaved sedge (Carex
Jilifoha). Juniperus kon^ontalis is oflen associated with
intrinsically erosive sites and such maybe the case
here. \T\o\i NHMTECCR97BH000^^

PiZscopyruTti snuthii —I'Jasella vindula Herbaceous

Vegetation

CPASSMI-NASYIR^

western wheat grass — green needlegrass grassland

This is a minor type, confined to swales and north-
facing slopes, usually on toe-slope positions. Both
dominants (also indicators of the type) are strongly
preferred forage by cattle and none of the sites had
green needlegrass (I'Jasella i^ndvla) cover values even
approaching those registered on some sites (within
the region) where grazing has been less intensive.
Needle-and-t bread (Shpa comata), threadleaf sedge
(Carexfikfoha) and blue grama [Boutdoiia gran^ilis) are
poorly represented on these sites

Note: With the incised drainages, considerable
topographic relief and some variety of parent
materials there are numerous habitats, including some
badlands topography, that was not adequately
surveyed for community types.

SPECIES:

Montana plant species of special concern were not
found. There are common species that might be
mistaken for rare species, including ^reen milkweed
{Asdepias "Jindijlora) and linear-leaf four o'-clock
{hhrabilis linearis^) There are a few rare species that
"were sought unsuccessfully on sandy habitat as found
on the south-facing slopes, including little Indian
breadroot {Psoralsa snneandrd) a species that blooms
early in the growing season, and nine-anther dalea
{T)alsa snneandra) which blooms late in the growing
season. Review of twinpod {Ph^ana spp.) specimens
collected on-site and in herbaria are pending. The
widespread species, common twinpod {Physana
didymocarpd), has been documented from Garfield
County (Booth and Wright 1966). The regional
endemic species, double twinpod {Physana hrassicoid^),
has recently been documented from Carter County
(Vanderhorst et al 1998).

OVERALL BIODIVERSITY SIGNIEICANCE;

This site is a significant representation of dissected
plains and the semi-open ravine systems on the
Missouri Breaks, with all the landforms and
vegetation t3^ical of the Eox Hills Sandstone. It offers
a larger array of xeric ravine habitat and associated
vegetation than the Spring Creek RNA. and
complements the combination of the Missouri River
Bottomlands RNA and the Two Calf —Douglas-fir
RNA as a Missouri River Breaks landscape on
Bearpaw Shale over 60 miles west.

Though none of the plant associations are rare or
unusual at this location in the state, there is a high
vegetation and plant species diversity as supported by
the broken topography, sheltered north-facing slopes,
array of substrates, and seasonal water. As such, it
represent a typical Missouri River Breaks gradient.

Wildlife values were not evaluated, though the RNA
with its diversity of habitats is presumed to
complement the overarching wildlife values of original
game preserve and national wildlife refuge
establ ishment. Overall values are enhanced by
continuity with native vegetation on all upland
borders.

LAI^JD USE:

This landscape has been grazed in the past and the
Rocky Mountain Juniper (Jumpems scopvioTitm) is
likely to have been cut for fencing Grazing practices
have contributed to the infestation of annual brome
grasses like BroTmis japonzcus (Japanese brome) and 3.
tectorwm (cheatgrass) which is quite apparent in the
western portion of the RNA on the uplands. The area
receives at least light hunting use.

MANAGEMENT COMIvlENTS:

There are almost no noxious weeds, except for Canada
thistle {CirsiuTn av^ense) at a springhead. Yellow
siuestdover (Mdilotus officinalis) is currently restricted
to shale slumps in lower ravine slopes, but has the
potential to occupy most of the terrain as judging by
results from other landscapes. The shores do not have
TaTnanx chinensu (tamarisk). Perhaps the most
abundant non-native species are the annual bromes,
mentioned previously.

Eire and bison grazing were driving factors with
which this landscape evolved. Reintroducing fire as a
management tool is an option on the rolling uplands
provided that it was planned to favor the natives over
the annual brome populations. With continued fence
maintenance, this RNA provides a good rangeland
reference and ecological baseline.

21

Manning Corral Prairie Dog
Town Research Natural Area

environment:

Miinning Corni! Prairie Dog Town Research Natural
Area encompasses a flat ridge, essentially a strip of

tableland at the edge ofbreakland topography. It is
typical of the Montana Glaciated Plains (Subsection d)
of the Northwestern Glaciated Plains Section (33lD).
Such areas have received Continental glaciation and
accompanying deposits of till and drift: over what is
essentially a planar to gently undulating surface of
soils developed from predominantly clay shales and
siltstone. There are numerous on-site exposures of
glacial drift to indicate this area has been glaciated. It
adjoins and is actually mis-mapped within the
Missouri River Breaks (Subsection f) of 33lD (Nesser
et al. 1 997); areas that are strongly and deeply
dissected terrain. West of the RNA the elevation
drops 600 feet to Rock Creek and east of the RNA are
the convoluted subdrainages of Seven Mile Creek).
There are bedrock outcrops to the immediate west
below the tableland to indicate that the overlying
glacial deposits are a thin veneer. The climate (nearest
station Haxby IS SW) verges on Continental with
cold» dry winters and the peak in precipitation comes
in May and June (36% of the year's total).

"VEGETATION:

The diversity of communities present corresponds in
part with the use patterns of the black-tailed prairie
dog (Cynomys Ijidozjzcianus) colony that died in a 1993
sylvatic plague episode five years earlier. The present
landscape is in a state of secondary succession. Three
areas were sampled that appeared to have
approximately the same environmental parameters but
that may represent different successional stages.

The RNA may have supported shrub-dominated

communities of Wyoming big agebrush / western
wheatgrass — green needlegrass (ArteTmsta tndentata
ssp. zvyomznge fists / PascopyruTnsTmihii) with or without
a major component of green needlegrass (L^Jasella
vtndjdiz). However, no traces of sagebrush skeletons
were found in the area occupied by the dog town.
Normally in these dry environments the woody
skeletons can persist scores of years if they are not
burned, even if only in dished-out rootcrowns. We did
not see evidence that they had decomposed or burned.
In glaciated terrain of north central Montana, such
sagebrush-dominated communities general decline
away from broken topography and with well-drained

soils. Information from the surroundings was

inadequate to examine cause and effect.

Bouteloua gracihs Herbaceous Vegetation

Cbougra;]

blue grama grassland

Across the formerly occupied prairie dog town is a
shortgrass prairie vegetation that covers most of the
gentle uplands of the designated area. This prevailing
vegetation is dominated by blue grama (Boiddoua
gracihs) but with discrete patchy islands where most of
the individual plants of midgr ass-height species are
concentrated around individual prairie dogburrows.
This community t3^e appears to be developed in the
identical setting as the two types described below. It is
distinguished from them by the severe reduction of
western wheatgrass (PascopynimsTmihii) cover, the
absence of Wyoming big sagebrush (ArtcTmsia
tndentaia ssp. -jjyomingensis), and the major increase of
tumblegrass (Sckedonnardus pamadatvs). It is further
distinguished by the presence of early succession
species like lemon scurf-pea (PsoraUa lanceohzta) and
conyza (or horseweed; Conyi^:a canadensis).

This association has very low vegetative cover, low
levels of litter accumulation, and much of the ground
surface made up of exposed gravels. We hypothesised
that these conditions were created when prairie dogs
occupied the site and the resulting cover removal
promoted wind deflation of the soil surface. Thus this
site has three times more exposed gravels than the
adjacent shrub-dominated site outside of the prairie
dog colony, as well as much higher cover of moss and
lichens.
CPlot NHMTECRN98SC00173

PascopyroTi srmthii - Boiiteloita gracilis — Carexfdifoha

Herbaceous Vegetation

CPASSMI -BOUGRA- CARFIL)

western wheatgrass —blue grama — threadleaf sedge

grassland

At the fringes of the formerly occupied prairie dog
town there is an abrupt transition between a
sagebrush-dominated community outside the colony
perimeter, and a grass-dominated community w ithin
the colony. A pair of adjoining plots were sampled for
direct comparison. Graminoids dominate the site; blue
grama (Boutdoua gracihs) and western wheatgrass
(Pascopyrum srmthii) are the most conspicuous. Trace
amounts of shrubs were noted, including Wyoming
big sagebrush (Arermsia tndentata ssp '•MyoTmngeTisis)^
but there are no shrub skeletons to indicate that this

22

lifeform previously dominated the site. There is major
overlap w ith BOUGRA in a grass composition that is
often associated with xeric or disturbed conditions,
including plains muhly (MuklerTbergza cuspidata),
Sandberg's bluegrass (Poa sscunda), and tumblegrass
{Schedonnadvs pa7iiadatus)y and the absence of green
needlegrass (IJasdla 'Jindida).

This communitymay represent a transition state
between Boutdoua gracilis grassland making up the
core of the colony and the colony perimeter. The
difference between this outer zone and the inner core
may reflect shorter occupancy and duration of
succession. Alternately, it may represent the greater
speed of recovery in this zone. The plot has the same
level of gravels exposed at the surface as the
preceding, but it does not have the "pedas tailing" of
wind erosion around the base of each piece of gravel.
It is also in a position for speedier recolonization with
the growth of rhizomes hj Pascopymm snuihiz from
directly adjoining areas. I^Plot
NHMTECRN98 S COOl 6^

Art^jnzsia trzdenlata ssp 'ivyormngerTsis / PascopyruTn

srmthii — J'lasella vtndjda Shrubland

i;arttsw / PASSMi - nasvir;]

Wyoming big sagebrush / western wheatgrass —
greeen needlegrass shrubland

This association or the 'very similar ARTTSW /
PASSMI, which lacks green needlegrass {I'Jasella
vtndjda), commonly exist as matrix or large patch
types associated with gently rolling bene hi and. This
particular example of the t3^e is a remaining fragment
outside of the discrete prairie dog colony, dominated
by Wyoming big sagebrush (ArtsTmsia tndsntata ssp
•'jjyormngsnsis) and western wheatgrass {Pascopynim
srmthii). It also has blue grama [Bouteloita gracilis)
present; the importance of the grass species in this
association are nearly the opposite of what was
recorded for the preceding association.

The site occurs at 2,980 fl: elevation at the head of a
draw in a slight swale that may receive additional
moisture through snow deposition, ostensibly causing
I-Jasella vindula to be present. The herbaceous layer is
dominated by Pasa)pyron STmtkii and has I'Jasella
vindvla as a subordinate, up to 10% cover. Canopy
cover of the gr az in g-sensitive I^Jasella 'Jindida is highly
variable in these communities depending on past
grazing intensities and it is used as an indicator
species at even reduced cover values. This is a
productive shrubland that has less than 50% bare
substrate (soil and gravel}, the ground cover
consisting mostly of litter, basal area, mosses and
lichens. Total shrub cover (25%) places this stand at

the break point between shrubland and herbaceous
vegetation. The subshrubs fringed sage [ArisTmsia
fngida) and broom snakeweed [Gutierre^a^a sarothrae)
are consistently present in bai'ely greater than trace
amounts. The usual complement of forbs including
scarlet globemallow (Sphasralcea coccinea), prairie aster
(AsUrfalcabi-s), and fleabane (Kngeron punulzs) are
present in trace amounts; only field millcvetch
(Astragalus agr^tis') exceeds trace amounts. I^Plot
I'JHMTECRN9SSC0015^

Other Vegetation Types: Side slopes were not
sampled, and the highly dissected terrain overlain by
recent fire contributed to a grassland mosaic. Areas of
localized dominance byplains muhly [MtihlefTbergza
ciispidatd) were noted, along with Ponderosa pine /
bluebunch wheatgrass (Pi mts ponderosa /
Pssitd.oroegnsna spicata) where pine survived the
recent burn.

OVERALL BIODIVERSITY SIGNIEICAMCE;
Like the Prairie Dog Island RMA, this site could be
used to track plant succession in the wake of prairie
dog use, and/or be considered for prairie dog
re introduction. It is the only one of the two sites that
has retained a dominance of native species, malting it
better suited in studying natural succession.

Apart from such natural succession, this RNA does
not represent biome features but has the potential. It
lies between rolling glaciated plains and south-facing
breakland topography. It is the only RNA with intact
plant associations on glacial deposits as opposed to
lacustrine or aeolian deposits, orunglaciated
landscapes.

Mountain plovers were previously documented in the
RNA, and the presence of burrowing owls was
mentioned in the original establishment record.
Wildlife values were not evaluated in this study.

LAI'JD USE:

The history of livestock use is evidenced in the site
name, a gathering point for southward cattle drives or
to disperse cows going north (Haglan pers. commun.)
The site is currently part of a large allotment.

MANAGEMENT COMIvIENTS:

A fire had burned at the south end within the recent

years. It appeared to have originated in the Rock
Creek valley, below, burning more of the surrounding
slopes than the uplands. There were no noxious weed
problems or exotic species invasions noted.

23

Missouri River Bottomlands
Research Natural Area

Note: Dillon Island and Grand Island were originally
recognized as separate research natural areas, but the
subsequent establishment of the Missouri River
Bottomlands RNA encompassed both islands and their
RNA boundaries.

ENVIRONlvlENT:

IvIisEouri River Bottomlands Research Natural Area
encompasses about 9 miles of free-flowing Missouri
River and associated valleybottom spanning 5,OS5
acres, including three large islands. It also represents
the downstream end of the Missouri River designated
Wild and Scenic by the National Park Service (1976),
i.e., the 9 miles at the downstream end of a 1^9-mile
segment. The valleybottom is over Ys mile wide in the
ai'ea, with many vestiges of intact bottomland
vegetation on islands and meandered slivers scattered
among homesteads and abandoned cropland,
encompassed within the rugged Missouri River valley
rising sharply at the valley edges over 600 fl. above
the River. The RNA boundaries follow legal
desc rations, zigzagging along midslope or at least
toeslope positions almost continuously on both sides
of the winding valley.

"VEGETATION:

The considerable relief, influence of water, and
ongoing successional processes accommodate a
complexity of vegetation. We have not tried in this
case to place the vegetation descriptions that follow by
their relative extent because they are nearly all
between 5-10% of the total landscape. Collectively, the
plant associations of black greasewood (Sarcohatus
'jermtculahis') are probably most extensive because they
are prevalent in lower valley slopes, as well as being a
major bottomland association. Stands of plains
Cottonwood (Popidus dsltoides) of various successional
stages are relatively well -rep re sen ted. Apart from the
River itself, about half of the RNA is made up of
bottomlands, and in each of the Bottomlands (named
for the settlers: Hess, Kendall, Knox, LeClair,
McNulty), over half of the bottomlands have been
plowed. We have not tried to characterize this major^
albeit more altered, segment of the landscape.

The studies ofRoberts and Sibbernsen (1979), which
focused on woodland and forest t3^es, and Mackie
(1965), which emphasized rangelands. were the first
major efforts at describing and classifying vegetation

for portions of the C. M. Russell National Wildlife
Refuge, and provide an important frame of reference.

Ariemzsa cana / Pascopyrum srmihii Shrubland

CARTCAN / PASSMi;]

silver sagebrush / western wheatgrass shrubland

Shrub stands dominated by silver sagebrush (ArtsTmsia
cand) are a recurrent bottomland landscape
component, occurring predominantly as small patches^
but ranging to large linear patches on river terraces as
well as the islands within the Missouri River. These
stands typically are developed on flat to -^^v^ gently
roll ing riverine and stream terraces, on medium-
textured (loam, silt loams, and silt) alluvial deposits.
Though this association can include stands with
wetland characteristics, as with temporarily flooded
hydrological regime and hydric soils, stands of the
RMA evidenced at most flooding and the vegetation
was not hydrophytic. Perched, or high water tables,
may influence the shrub rooting zone for a portion of
the year. Litter is the predominant ground cover,
usually in excess of 80 % cover, with small patches of
bare soil randomly distributed. This type grades to
western snowberry (Symphoncarpos occidentalis) on
moister sites, and Wood's rose (Rosa zvoodsii) or black
greasewood (Sarcohatvs verTmadatiis) dominated
community, sometimes directly to Wyoming big
sagebrush (Artemzsza tndsntata ssp. -^OTningensis) or
Sarcohatus 'i^erftucidatus-donim.^tad uplands that lack
dominance by rhizomatous grasses and have salt
affected soils. These stands probably received heavy
use by livestock prior to RNA establishment. Most are
in relatively good condition but there are portions of
these stands with a strong non-native graminoid
component.

These ArteTmsia cana stands are like the big sagebrush
stands (Artermsia indsntata) elsewhere on the ClvlR
Refiige in that the percent shrub cover ranges from
mid-20s to lower-30s, the somewhat arbitrary cutoff
between shrubland and shrubby herbaceous
vegetation. Shrub dominance is almost exclusively
contributed by 3-4i [^^^ feet tall Ariermsia cana.
Common rabbitbrush {ChrysotJut-mniis Tiaussasvs), black.
greasewood [Sarcohatus ZJeTymadatus)^ and western
snowberry

{Syjnphtxncarpos occidentalis) are nearly 100% constant,
but present in greater than trace amounts only at
ecotones to surrounding vegetation types. The
graminoid component is also consistent in
composition with western wheatgrass (Pascopyrum
srmthzi), green needlegrass (I'Jasella z/zndida), and Po/z
juncifoha (alkali bluegrass). Despite sample stands
being chosen for appearing among the least disturbed,
all stands have some measure of non-native grasses

24

including Kentucky bluegrass (Poa pratensts\ Japanese

brome (BroTnus japomcus), and on occasion,
intermediate wheatgrass [Agr^pyron iniermsdium).
Areas "with abundant "■-"veed populations have very little
ISlaseiia vindida, indicating it maybe susceptible to
grazing and competitive efTects. The foib component
is sparse and species poor; only the non-native
increasers yellow sweetclover (Afdilotus qffici7ialis\
"white sweetclover (Klehlotus alba), and goat's beard
(Tragopogon dubzus) occur in greater than trace
amounts. Yarrow (Achillea mzlkjbhum) is the only
native "■-vith greater than 50% constancy.

Atnplex gardnen Dwarf Shrub land

Catrgar;]

Gardner's saltsage dwarf shrub land

This plant association is found along the valley slopes
as small patches in a complex mosaic of other sparse
vegetation that is more widespread. Surrounding
vegetation includes stands dominated by black
greasewood {Sarcohatiis verrmadahis) and Wyoming
big sagebrush (ArteTmsia tndsTitata ssp zvyoTmngsnsis).
Our sample plot differed from these types only in
having but a trace o^ Sarcohatus verrmculahis but clearly
there is a continuum in substrate properties that is
reflected to some degree in the vegetation mosaic. It
should also be noted that in the course of
reconnaissance Gardner's saltsage (Atnplex gmrdncn)
■was found to occur with coverages greater and less
than 10% (the sparse cover cutoff) which would give
some of these stands a sparse cover designation.
Overall the cover is greater than 10% giving stands an
aspect closer to the type as described for southeastern
Montana (Vanderhorst et al. 199S). Within the RI'JA,
ARTGAR occurs both on slope aprons, where allcal i-
laden fine-textured slopewash accumulates, as well as
on slope shoulders and narrow crests and even mid-
slope positions of any aspect; virtually an3r^here
bentonite lenses or unusual shale substrates are
exposed.

The vegetation is close to being a monospecific layer
of the dwarf-shrub Atnplex gardnsn, its cover ranging
from 6 to 30 (4iO)%. Depauperate specimens of black
greasewood {Sarcohatus verrmcidahis) generally
constitute the only other shrub present. Bottlebrush
squirreltail (Sitanion hystnx)^ Indian ricegrass
[Oryi^x)pszs hymsTioid^) and thick-spike wheatgrass
{Symus lanceolatvs) are the graminoids most often
found here, but not greater than trace amounts. Tali
seablite [Suaeda moquinzi) and plains bahia
(Picrademopsts oppositifoha) are the only recurrent
forbs in this community.

Populus ddtoidss / Symphoncarpos occidentalis

Floot^lain Woodland

CPOPDEL / SYMOCC]

plains Cottonwood / western snowberry floodplain

woodland

Floodplain woodlands are areally extensive
communities as broken bands along the Missouri
River on the older portion of alluvial bars and outer
edges of the river's floodplain; see Hansen et al. (1995)
for a discussion of riverine geomorphology explaining
the genesis of these stands. Some of the stands
representing this community type are flooded
virtually ev^ry year or at least heavily influenced by
the seasonally fluctuating watertable; other stands are
more removed, do not experience yearly flooding and
their roots are less watertable influenced. During
reconnaissance, all degrees of anthroprogenic
modification were noted including plowing, seeding to
alien species, cutting, domestic stock grazing and
browsing in this type. Some stands of plains
cottonwoood {Populus ddtoides) had nothing more than
combinations of quackgrass (Agropyron rspens), smooth
brome [BroTnus insrmis), Kentucky bluegr ass {Poa
pratensis), leafy spurge (Kupkorbia ssidd), and American
licorice (Glycyrrhii^:a lepidotd) in the understory. Stands
were sampled that appeared least disturbed, but that is
not to say they were undisturbed.

Hansen et al. (1995) interpret this community as both
a mid-seral stage of floodplain development and a
browsing-induced disci imax (by whitetail deer?) of the
plains Cottonwood/ redosier dogwood forest {Popvlus
deltoidss / Cormts ser^c^us Forest). If that were the case,
then animal scouring of these stands is phenomenally
thorough because our inventory was able to find no
more than one stem of Cornus senseus in the RI'^A. The
common chokecherry (Pmnvs virgimaTia) and western
serviceberry [ATneliznchisT alnifolia) were also scarce.
In younger or more mesic representations of this type,
canopy cover of the Populus ddtoides may exceed the
60% crown cover limit for woodlands. The three
sampled stands of this type are relatively mature to
''old growth" with average diameters of 16-20 inches
in b.Yo stands and the third with 3^-^^ inch stems
remaining and number of downed and dying veterans.
Rotten hearb.vood precluded obtaining ages on these
stems. The mortality in the stand with trees of the
largest diameter has resulted in less than 30% tree
canopy cover and with no Popidiis deltoid.^
reproduction because of no fresh alluvium being
deposited. This stand will is liltely to become a
Symphoncarpos occ^denialis-doniin:ited shrub land.
Woods rose {Rosiz zvoodsit) is the second leading shrub

25

species followed distantly by species of gooseberry
[Ribes spp.) and willow (Sahx spp), which are seldom
present with greater than 5% cover. In the better
condition stands, western wheatgrass (PacopyniTn
STruthii) and green needlegrass (I'Tasdla z/zrzdida) are the
dominant graminoids and their cover approachs 30%.
Commonly, alien graminods such as smooth brome
{3romus inermis), quaclcgrass (Aropyron rspens),
intermediate w he atgr ass [A^ropyronirderTnedvwm), and
Kentucky bluegrass {Poa pratsnsts) dominate this
component. In the sample stands, the foii> compoent
is minor; the only time a significant forb presence was
noted was in the case of exotic species such as Canada
thistle [CiTszurfi avjenss), yellow sweet clover [Mehlohis
officinalis)^ and blaclc medic [Klsdicago lupuliTia).
Canada goldenrod {Sohdago cartade fists) was the only
native forb occurring in more than half the study
plots.

[;Plots NHMTECRN9SSC0018,
UHMTECRN98SC0026, NHMTECRN9SSOD030^

Sarcohatus verrmadabis / Ainplex gardnen

Intermittently Flooded Shrubland

CSARVER / ATRGAR;]

greasewood / Gardner's saltsage intermittently

flooded shrubland

This association, or one affiliated to it in name, has
previously been identified only from the southeastern
and Bighorn Basin regions of Montana and adjoining
lands in Wyoming. It has been characterized as a
small to large feature occurring on alkali-affected
alluvial flats, thus the title of intermittently flooded
shrubland. More recently this same type has been
noted in Carter County to occur en badland
formations (Vanderhorst et al. 1998). The two plots
representing this association on the RNA occur on
dissected, Beai' Paw Shale uplands, bentonite
inclusions, and slope aprons with rill, gully and sheet
erosion and with plant pedicelling. Ground cover is a
monotonous expanse of grayish-tan exposed clayey
substrate lacking soil development and with traces of
gravel. The vegetative physiognomy of these plots
does not technically qualify as shrubland but rather as
dwarf shrub lands verging on sparse vegetation (<10%
total canopy cover). It may be advisable to change the
modifier name of the type because even the
intermittent flooding is inferred and may not if fact be
a significant ecological driver. This type is extensive
and recurrent along the Missouri River valley, with so
much exposed valley slope. Part of this landscape
mosaic supports patches of Gardner's saltsage
(Atnplex gardnen) or rill scale (A suckleyi) alone,
without Sarcohatus verTtuadatus These species may
define two different associations in the provisional

statewide classification, but do not necessarily
represent two distinctly-different environments.

Dominance in the shrub canopy shifls between
Atnplex gardnen and Sarcohatus ZJerffucidahts but their
cover^ singly or combined, does not exceed 15-20%.
Wyoming big sagebrush {Ariermsia tndsTitata ssp,
zvyoTmngensis) is present in trace amounts. The
graminoid layer is depauperate to nonexistent and the
forb component usually follows suit, except for the
sporadic, unexplainedly high coverages of rillscale
[Atnplsx sucklcyi). Tall ^e^Vite {Suaeda Tnoquinii) is
consistently present in trace amounts. |^Plots
NHMTECRN9SSOD023, I1HMTECRN9SSC00293

Sarcohatus ZJerTmadatiis / Pascopyrum STmthii Shrubland

CSARVER / PASSMq

greasewood / western wheatgrass shrubland

This community is predominantly a large patch type
that develops on the heavy silt to clay loam soils of
alluvial fans, toeslope or slope apron positions and old
river terraces on nearly level terrain with at most 2-
3% slope. Many of these stands receive considerable
slopewash including entrained fines from the adjacent
erosive uplands. It differs markedly from the
previously described black greasewood (Sarcohatus
vsrtmculatvs)- type immediately above in that it is
sufficiently productive that Utter forms and
accumulates leaving little exposed soil. Most often
this type is noted to grade to the ArisTmsia cana /
PascopyTU/msTTnthzi (silver sagebrush / western
wheatgrass) association, that occurs on less salt-
affected positions. Together with the Ariermsia cana
shrubland, it is an integral component of the riparian
mosaic in this landscape, though none of the
representations inventoried on the RNA met the
requirements for a jurisdictional wetland.

The shrub canopy is dominated by Sarcobatvs

vervmculatus, usually in excess of 25% canopy cover.
Both silver sagebrush [Ariermsia cana) and rubber
rabbbitbrush (Ckrysotkamnus nauseosvs) are consistently
present, their cover seldom exceeding 10%. The three
plots represented sites inferred to be in good condition
by their high cover values (>60%) for the dominant
grass, western wheatgrass [PascopyrK/msvutkii). Other
high constancy graminoids are alkali bluegrass {Poa
juncifohd), green needlegrass (I^Jasella vindida) and the
annual weed Japanese brome (Bromvs japortzcus); the
combined cover of native graminoids apart from
Pascopyruvi STTntkii ui^ua[\y do^^ not exceed 10-16%.
The forb composition reflects past disturbance with a
variable assemblage of weedy, increaser species
present, but their coverages usually don't exceed trace

26

amounts. Stands and stand segments were noted

"where Poa secanda (Sandberg's bluegrass) "was

dominant in place of, or in addition to, Pascopyrwm

STmthii, a feature that is likely to be a disturbance

response.

[;Plots NHMTECRN9SSC0O21,

WHMTECRN9SSC002S, NHMTECRN9SSC0032^

Pinvs ponderosa / Carex inops var. W^o^fe^iZ Woodland

CPINPON / CARING]

Ponderosa pine / sun sedge woodland

This woodland t3^e of limited extent is found as small
patches within breakland or highly dissected and

slumpy topography on moderate to steep slopes
having cooler exposures. The highly erodible and
immature soils are derived from shales. Sheet, rills and
gully erosion, as well as plant pedicelling, are evident.
This type exists in a mosaic "■-vith two other woodland
communities; Ponderosa pine [Pinvs ponderosa)-
dominated patches having such erosive substrates that
the undergrowth is virtually absent (< 6% canopy
cover) and on cooler aspects small patches of the
Douglas-fir / small-seeded Indian ricegrass
(Psevdoisu^a merisziesti / Ory%opsis rmcrardhxi)
association are found.

Prior to this report, the type has onlybeen identified
and described by Hansen and Hoffman (1988) for
southeastern Montana and adjacent portions of the
Daltotas and by Hoffman and Alexander (1987) for
Wyoming. It is notable that this type, as described by
the above-cited authors, is outside the distribution
limits of Douglas fir {Pssudohuga men^iestiy Though
the stands on the RNA are dominated bj an open
canopy o^Pzrms ponderosa in the uppermost layer, as
well as the reproductive layers, these sites are not so
severe as to be beyond the limits o'i Psendotsuga
men?a.€sii establishment and growth. Coring of the
largest pine trees {Pinvs pondtTosa^\'^-\Q inches dbh,
38 fl. tall) reveals ages of at least 130 years; none of
these trees have fire scars. The canopy cover of Rocky
Mountain junper (JuTupervs scopulorum) ranges widely
but, usually is in the 10-20% range.

The undergrowth of this association is dominated by
graminoids. Shrubs and forbs are poorly represented.
Fragrant sumac (Rhus arojTiatica) and Wood's rose
(Rosa zvoodiii) are consistently present, occurring in
trace amounts. Sun sedge {Carex iTiops) dominates the
undergrowth with coverages generally not exceeding
30%. This contrasts with its representation in
southeastern Montana stands where it is nearly
sward-like, coverages mostly exceeding 80%.
Bluebunch wheatgrass {Psevdoroegnena spncata) and

western wheatgrass (PascopyniTnsmithz-i) are

consistently present in low coverages. Yarrow

{Achillea Tmllifohum) and American vetch [Vicia

amerzcand) appear to be the forbs most consistently

present.

CPlot MHMTECRN98SC0O313

Popvius deltotdes / Comus senceiis Temporarily Flooded
Forest

CPOPDEL / corser;]

plains Cottonwood / red-osier dogwood temporarily
flooded forest

Mote: This stand was sampled directly oii"tside of the

Refuge within James Kipp Campground area. It had
previously been sampled by the Montana Riparian
Association (Hansen et al. 1995); this data has been
weighted heavily in characterising dogwood as the
undergrowth dominant in "natural", undisturbed
Cottonwood stands. This stand was sampled to satisfy
our curiosity as to the composition and landscape
position o^ ^Popidus ^^^oz ^-dominated stand that
differed from all other such stands noted in the RNA-
In terms of landscape position, relationship to the
watertable and flooding regimes this stand appeared
no different than those of the P deltoides /
Syjnphoicarpos occidentahs association seen upriver from
this point. In extensive reconnaissance of the upriver
bottomland stands only once did we find a shoot of
red-osier dogwood (Cornvs senceus). Hansen et al.
(1995) interpret POPDEL/SYMOCC as a browsing-
induced serai expression of the POPDEL / CORSER
community. This may be a plausible explanation for
what was observed for Popidus deltoides stands on the
Missouri River Bottomlands. But it begs the question
as to why the James Kipp Campground stand could
escape browsing altogether for a period sufficient for
CoTfvus senceus to attain a height putting its foliage
beyond the reach ofbrowsers. This stand is also
somewhat anomalous in either lacking other shrub
species like western serviceberry (Avielanchzer almfoha,
common chokecherry (Prumts ZJirgtmaTia), gooseberry
species (Rzbes spp.) or having their coverages much
below ranges cited as typical for this community type
as in the case of Wood's rose (Rosa zvoodsii) and
western snowberry (Symphoncarpas occidentahs). For
both the plot and the small examined portion of the
total stand, the herbaceous cover was extremely
depauperate. Clasp ing-leaved dogbane [Apocynum
sihincum) was the only forb noted.

Other Vegetation Types: The shoreline vegetation
zones were not briefly described. They often included
off-shore emergent bands of Three-square (Scirpus
pungens), water sedge (Carex aquatilis) along the

27

shorelines, thickets or interrupted strips of coyote
"willow (Sahx e3::igiia) on the banks a.nd flats, and open
flats colonized by wild licorice [G-lycyrrhzi^^a lepidota)
3.nd Pascopyrufn svuthzi. There were occasional grassy
banks with scattered patches of prairie cordgrass
{Sparhna psctinata) and Canada wild-rye {^ymus
canadensts), but they were heavily invaded by
quackgrass {Agropyron repens) and smooth brome
(^ormis inermis)^ so it was not clearwhether this is a
discrete local vegetation feature.

OVERALL BIODIVERSITY SIGNIEICANCE:
The Missouri River Bottomlands RNA has
significance as representing a major Great Plains river
valley and its accompanying natural succession and
geological processes. It is a fitting example with three
major islands, an array of cottonwood stands at
different successional stages, bottomlands with the full
suite of shrubland communities^ and boundaries oJong
app. nine miles that take in many ofthe valley slope
communities.

It is contiguous with and is accessible overland via the
Two Calf-Douglas-fir RNA, discussed below. The
Douglas fir forest is not othet"-^ise represented in
Missouri River Bottomlands RNA, and the b-vo
together represent an outstanding example ofthe
Missouri Breaks gradient and complement one
another. Together they contrast with and complement
the drastically different vegetation ofthe Limber Pine
RNA, also in a Missouri Breaks setting over 60 miles
east, but with a predominantly sandstone, rather than
shale bedrock.

This RNA is known to harbor the following animals
species of special concern: bald eagle {Hahaeeius
l^yiXfcephalus), Ferrugineous hawk [Butso regahs),
Sturgeon chub [Hybopszs gdida), Sicklefin chub
{Hybopszs Tneeki), and a major Missouri River
population of pallid sturgeon (ScaphzrhyTickus albiis)
above Fort Peck. Most of these have territories that
extend beyond the limits of RNA boundaries. Wildlife
values were not evaluated.

laot:) USE:

Extensive segments ofthe Missouri River
bottomlands were plowed, representing over half of
the bottomlands running through the center ofthe
study area, and including all accessible, large flats.
These were subsequently seeded to non-native species.
Homesteaders also based their ranching operations in
the valley, and prior to settlement, woodhawkers cut
timber to supply passing ferries.

Most ofthe area has not been grazed since the late
1960s or the early 1970s (Haglan pers. commun.)

Boating, motorized travel on existing roads, and
hunting are regular recreational activities. The site
adjoins James Kipp Campground and Boat Ramp, and
is the lower 9 mile end ofthe 1^9 mile-long V/ild and
Scenic segment ofthe Missouri River.

MANAGEMENT COMlvlENTS:
The flooding regime is modified by upstream
impoundments, diminishing the magnitude of floods,
increasing the rates ofwater erosion, and perhaps
altering the character of ice jams. The highwater
conditions of 1997 were reflected in overtopping
cutbanks and new- or newly-expanded silt and gravel
deposits.

Habitats that are flooded have the continuous threat of
invasion by water-borne exotics like Russian

knapweed (Centaurea rsptns)^ one of the most invasive
exotic species at present in the bottomlands ofthe
RNA. V/ater is not its only dissemination vector, and
the bottomland plantings of non-native species are
vulnerable to its expanded invasion in general. One of
the most serious Icnapweed invasions is on Hess
Bottoms, located above the best-condition cottonwood
stands. Invasion by Taviansk chine nsis (tamarisk) is
another serious potential threat to riverside habitat
though it was not observed in the limited study area
visit.

In the uplands, yellow s'-veetclover (AleUlotus
qfficiTiahs) is widespread and appears to be particularly
abundant on shale uplands and some brealdand areas
that have burned.

Prairie Dog Island Research
Natural Area

ei'jvironment:

Prairie Dog Island is a 1 5 acre island in the upper end
ofthe Dry Arm» a major south-trending arm at the
eastern end of Fort Peck Reservoir. Its highest point
rises less than 60 feet above water level. This site is
somewhat atypical of Missouri River Brealcs
Subsection (f) ofthe Northwestern Glaciated Plains
Section (33lD, Nesser et al. 1997) in that it is not
dissected but gently rolling terrain and the
predominant soil texture is a fine sandy loam,
indicating a weathering from mainly sandstone. The
climatic regime is essentially Continental with hot
summers and frigid winters; 35% of average annual

28

precpitation (1 1.5 in.) occurs in May and June (data
averaged overyears 1966 to 1998, from Fort Peck

Power Plant, Western Region Climate Center).

VEGETATION:

The great majority of the island was at one time a

black-tailed prairie dog [Cynomys ludovzcianvs) colony,
that has been extirpated as the result of sylvatic
plague. During the colonys existence or following its
demise, 3ro7mis tectorum established in dense s'-"vards
over the area occupied, or once occupied, by the
colony. Only two plots were established to document
the island's vegetation composition, one representing
this community within historic prairie dog colony, and
the other representing the probable undisturbed
native vegetation.

Brornvs feir^or^?^ Disturbance Vegetation

[EROTEC^

cheatgrass disturbance vegetation

Cheatgrass [Brormis tectorum) is the dominant species
in the area formerly occupied by prairie dog colony.
Cheatgrass cover is variable, but generally is in excess
of 60% and ranges as high as 90%. A number of
species that were once prominent on the site, inferred
from composition of adjacent landscape, are still
present, including silver sagebrush (Ariermsia cana),
fringed sage (ArteTmsiaJrigida), blue grama (Bouteloiia
gracilis), needle-and-thread (Stipa comata), Indian
ricegrass [Oiy^opsis hymsnoidss) and scarlet
globemallow (Sphaeralceacoccinsa).}rioYje,-^^T, the site
is dominated by weedy, increaser species that also
include common dandelion (Taraxacum officinale),
tumblemustard (SzsymbnuTn altzsstjnum), goat's beard
(Tragopogon did>ius), Russian thistle (Salsola kali), and
rough pennyroyal (FLedeoTna kispidum). I^Plot
NHMTECRN98 S COOO^]

Stipa coTnata — Boutdoua gracilis — Carexjilifoha

Herbaceous Vegetation

CSTICOM-BOUGRA-CARFIL^

needle-and-thread — blue grama — threadleaf sedge

grassland

This is the prevailing grassland where sandy deposits
are extensive, as at Prairie Dog Island. The
composition of the sampled site, representing the

fraction of vegetation not impacted by the "dog town",
matches the modal conditions described for this plant
association with Stipa coTmnta dominant and Bouteloua
gracilis and Carexfilifolia being subsidiary graminoids
(Hansen and Hoffman 1988, Jensen et al. 1992). Of

only four forbs present, scarlet globemallow

(Spkasralcea cocciTied) is the dominant. The presence of
silver sagebrush (ArtsTmsia cana) probably reflects the
high water table. This site has also received some past
disturbance, as evidenced by several alien species
being present. [[Plot NHMTECRM98SC0005^.

The unvegetated sandy shore ends abruptly in

cutbank with no inter'vening vegetation gradient
between it and the grassland.

OVERALL BIODIVERSITY SIGNIFICANCE;
This site has limited merit as an RNA due to the
explosion of weed populations, particularly &[>7???ty
tectorum (cheatgrass). This condition may reflect the
combined history of prairie dog use and surrounding
land use. Even if this site were in pristine condition, it
would have limited vegetation significance because it
represents a single major community t3^e; one that is
best represented as part of a large, contiguous
landscape.

The presence of colonial nesting birds was evident and
warrants enumeration. Cutbanlts on the leeward
(southeastern) shore were favored as perches and
shoreline nest sites. The original establishment
record said that this site also provides habitat for
burrowing owl; it is unclear whether this meant that
breeding of this species had been documented or
suspected. Existing and potential wildlife significance
may warrant further consideration.

OTHER VALUES:

Like the Manning Corral RNA, this site could be used
to track plant succession in the wake of prairie dog
use, or be considered for prairie dog re Intro duct ion.

LAI^JD USE:

Grazing by large ungulates, if it occurs at all, would
be confined to winter when the frozen-over reservoir
provides the only access, but it was once part of
primary range for livestock grazing. Until recently, it
was also subject to foraging from prairie dogs in the
main colony.

MANAGEMENT COMlvlENTS:

Weeds and increaser species are prevalent on this site.
Native graminoids and forbs are present, but there is
no indication that they can outcompete the weed
population to reestablish their dominance. Though
there are various theories on cheatgrass invasion
(Young and Allen), the projections are similar. On
comparable sites of the Columbia Basin and Great
Basin, despite a 4iO year hiatus in disturbance,
cheatgrass (Br^Tnus tectorum) has continued - and even

29

increased - its site dominance (Mack. 1981 ). There is
low potential for the return of natural vegetation on
Prairie Dog Island. A large sward of Canada thistle
(CiTsium arvense) adj oins the southeast end where bird
use seemed heaviest. The plains cottonwood (Popidus
deltozdes) and tamarisk (Ta?rn2T^k chineTTsis) that had
established along the shoreline "were innundated and
killed in the high water conditions of 1997-98.

Spring Creek Research Natural
Area

ENVIRONlvlENT:

Spring Creek RNA is a 160 acre tract that
encompasses an incised ravine system, adjoining
tableland, and valley slopes along the Dry Arm of
Fort Peck Reservoir. It is part of the Missouri River
Breaks (Subsection f) of 33lD {Nesser et al. 1997).
The 190 fl. ofvertical relief dissects both sandstone
and underlying siltstone bedrock. It is situated north
of Spring Creek and Spring Creek Bay, with an
ephemeral streamcourse that empties westward into
the Reservoir. Upland soils, upper ravine slopes, and
ravine bottom soils are fine sandy loams developed
from the underlying bedrock. Lower ravine slopes,
benches and valley slopes are silt loams (shale
derived). The climate is essentially Continental (refer
to the characterization of Fort Peck monitoring data,
presented for Prairie Dog Island RNA.)

"VEGETATION:

We 11 -developed expressions of tableland and ravine
slope plant communities are found, as well as a
stringer of woody draw vegetation that has a
prevalent green ash (Fra3::inus pensyliiamca) component
in the canopy mixture. The original designation of
this RNA to preserve an unusual stand of aspen
(Popid-us treTmdoides) may have been based on the
interpretation that this species, being the tallest in the
ravine, is the stand dominant. The complement of
plant associations does not have one that is clearly
more extensive than any other. Perhaps the most
extensive, though variable, cover t3^e is comprised of
the Wyoming big sagebrush shrubland {ArteTmsia
inderdata ssp. zvyormngensis Shr^abland) on benches at
the mouth of the ravine and valley slopes above the
reservoir.

Artermsia tndentata ssp. •'JuyormngsTtsts / Pascopyru-in

sTmtJm Shrubland

CARTTSW / FASSMI]

Wyoming big sagebrush / western wheatgrass

shrubland

This association is a major vegetation type both in the
study area and across the Northern Great Plains from
Colorado north to Saskatchewan, on both glaciated
and non-glaciated surfaces. The majority of the type
occurs in Wyoming and Montana. It generally
occurs, as on this RNA, over large areas, except at the
perphery of its range, where it may occur in small
patch fragments. There are several recognised names
and variations for the Wyoming big sagebrush/
western wheatgrass shrubland (including Artevusia
tndentata / PascopyruTn srmthii, ArtsTnisia tndentata ssp.
tndentata / Pascopyrum srmthii, ArtsTmsia tndentata /
Efymus lanceolabis, A. tndentata / PascopymTti srmthii —
(Efymus lanceolatus) (^Schneider et al. 1997,
Vanderhorst et al. 1998^.) Some of these differ only in
that the subspecies of big sagebrush was not stipulated
or known at the time the investigations were
conducted. Only the Artemisia tndsTitata ssp. tndentata
association (ARTTST) represents a distinct habitat,
one more associated with swales and drainages and
foundto the west oftheARTTSW/PASSMI type.
The ARTTSW / PASSI type is generally associated
with low relief uplands, benches, plateaus, or rolling
terrain but within most of the RNA these flatter
surfaces are sandstone capped and favor needle-and-
thread [Shpa ^%^??M:t^^dominated grasslands, whereas
the ArteTmsia tndentata communities are found
downslope on gentle inclines with heavier-textured
soils (silty clay loams). This association, including the
RNA representation, has considerable exposed
surfaces (soils mainly), generally in excess of 50%.

The shrub canopy o^ Artemisia tndentata ssp.
luyormngensis is variable in cover, but generally in the
vicinity of 20-30%, straddling the shrubland to herb-
dominated structural break according to NVCS.
Winter fat (Krascheramtikovia lanata) and Fringed sage
(Artermsiajrigidd) are present in trace amounts,
though areas of disturbance support more Artermsia
Jhgida. Western wheatgrass (Pacopyru/m STmthzi) and
threadleaved sedge {Carexjilifbhd) are the dominant
graminoids. their combined cover usually not
exceeding 50%. Green needlegrass {I'lasdla z/zndida) is
present, mostly under the protective canopy of -^.
tndentata, raising some speculation, at least for the
footslope positions, that ARTTSW / PASSMI -
I'Jasdla vindvla maybe the potential community type.
Certainly in the past, livestock grasing pressure might
have been extreme and led to significantly reduced
coverage for the highly palatable I^Jasella wndida. The
forb component is both sparse and species poor. In the
plot, two of the four species, includingplains prickly-
pear (Opuntia polyacantha) and brittle prickly-pear
(Opuniiajragilis) are associated with overgrasing,

30

though their low cover here would not necessarily
indicate overgrazing at present or in the recent past.
Scarlet globe mallow (Sphasralcea coccinsa) is a forb
found here and across all manner of range sites. I^Plot

WHMTECRN9S s cooos;]

Stipa co-maia - Boutdoua gracihs - Carexfikfoha

Herbaceous Vegetation

CSTICOM - BOUGRA - CARFIL;]

need] e-and-t bread - blue grama — threadleaved sedge

grassland

This is one of the most extensive of Great Plains
grassland community types, occurring from the
Midwest to the Rocky Mountain Front of Montana
and north well into Saskatche'-van and Alberta. In
eastern Montana and North Dalcota it occurs, as at
this site, on soils with a higher percentage of sand
than is represented in soils of adjacent communities; it
occurs on both glaciated and unglaciated landscapes.
In eastern Montana landscapes these sites are
frequently on ridge systems where sandstone strata
are exposed. In the shale- and siltstone-dominated
plains of eastern Montana it is often manifested as a
small patch type on projecting ridge crowns and
hillocks.

It has also been the subject of some vegetation
classification uncertainty because there have also been
two other plant associations named with needle-and-
thread in separate combination with the two other
species. There is no existing unequivocal key to
vegetation types that can distinguish among these
types, so the most inclusive name was chosen, one that
appears to fit published descriptions of the type (Allen
et al. 1999). This t3^e is potentially extensive on the
sandstone-underlain benchlands that cap the local
landscape. It is in particularly good condition as it
occurss within the RNA whereas this type has
undergone a gr as ing-induced conversion to a fringed
sage (ArtsTmsiaJhgida) and blue grama {3outdoua
gracihs)-dounn.^led disci imax on directly adj oining
tracts separated by fence. Litter and lichens dominate
the ground cover within the RNA type and ai'e a
decidedly reduced outside the fenceline, where
exposed soil is the dominant surface category.

Within the RWA's expression of this type, Artewisia
JTigida is the only shrub-like plant, present in trace
amounts. Nee dle-and-th read {Shpa coTnata) strongly
dominates the herbaceous component. Throughout
this grassland, threadleaved sedge (Carexfilifoha) has
cover values ranging from 60 to 70 %, though blue
grama {^Bouidona gracilis), is still an important
component. Western wheatgrass (PascopyruTn smithii)

is consistently present in trace amounts in contrast to
its greater cover on adjacent heavier-textured soils.
There are no forbs with greater than trace amounts of

cover; those noted to have with high constancy in this
t3^e and present throughout the stand are rush
skeletonweed (Lygod^rma juncsa) and scarlet
globemallow (Sphsusralcsa coccinea). I^Plot
I^JHMTECRM98SC0006^

Fraxinus psnnsyhanica / Prum/s ZJzrgtmana

Temporarily Flooded Forest

CFRAPEN / PRUVIR^

green ash / chokecherry temporarily flooded forest

Typically this t3^e occurs along riparian corridors,
springs and ponds and other floot^lain positions, but
in this landscape it is associated with v-shaped ravines
known colloquially as "woody draws". It is very
similar to the green ash-American elm/ western
snowberry forest [Fraxinus psnnsylvanica — Ulmus
amencana / Symphoncarpos oa^dentalis Forest)
identified for North and South Dakota, though in
Montana only slightly more than 10% of the stands g^
this t3^e have UlTmis a??^T7£i?;?tz present. Most of this
stand is confined to the ravine bottom and toeslope
positions. The forest floor is nearly completely
covered with litter, the limited ungulate trails being
the only areas where soil that is sandstone-derived,
fine sandy loam is exposed. This example of the type
is of good to moderate quality due to the dominance of
the alien Poa pratensis (Kentucky bluegrass) and the
somewhat low diversity of the forb component,
though noxious weeds are not present

The upper canopy is rather open, appropriately
classified as woodland cover (^0-60%) and height ('25
fl.), w ith Fra3^nus pennsylvaTuca generally the
dominant tree species, as well as being represented in
all layers of the multi-storied canopy. Conks (bracket
fiingi) were present on almost all larger Fra3:imis
stems, though elevated mortality rates were not
evident. In some portions of the stand quaking aspen
[Popidus tre-rmdoides) is codominant with the Fraxinus.
Both Popndits irerfudotdes and Rocky Mountain juniper
(Juniperus scop/uloruTn) also occur in multiple size
classes throughout the stand. The shrub component is
relatively diverse with at least seven species
consistently represented. Common juniper (Jumpervs
commums) and western snowberry {Symphoncarpos
occidentalis) are the dominants, but if the browsing
pressure on chokecherry (Prunvs 'Jirgimana) and
western serviceberry (Amelanchisr alnifoha) were
relaxed, it is possible that these species might increase
in cover. Past disturbance is probably the reason that
Kentucky bluegrass [Poa prats nsis) is the dominant
grass. Other important grasses are Canada wildrye

31

(Klymus canadensis) and bearded wheatgrass (Klymus
trachycaulis). Starry Solomon-plume (Svulacin^ stdlata)
and horsemint (Klonardafistidosa) are uniformlywell-
distributed, the only forbs among the eight total
occurring in greater than trace amounts. Absent are
forbs such as purple meadowrue (Tkahctmm
dasycarpum), northern beds t raw (Gahufn apanrie) and
Sprengel's sedge (Carsx sprengdii) that quite often
inhabit these sites; their absence couldbe attributed to
poorly-developed soils or to habitat conditions. I^Plot
WHMTECRN98 S COOO?^

Juntpsms konrxftitahs / ScIn^ack/nuTnscopanu7nY)'7^^Ti

Shrub land

CJUNHOR / SCHSCO3

creep ingjunper / little bluestem dwarf shrub land

This association, both at large and ■within the RNA, is
characterised as a topoedaphic climax, found on
moderate to steep, potentially highly erosive slopes of
fine sands to sandy loams, with north- to east-facing
exposure. In some instances it does occur on flatter
slopes, but still the substrate is prone to erosion. For
the most part, patch size is dependent upon local
vertical relief of appropriate substrate, which is limited
in the RWA and thus the type is exemplified by small
patches occurring on cooler exposures of steeply
incised ravines. There are several other associations
having creeping juniper (Jumperus hori'^iontalis)
dominant but they differ somewhat with regard to the
graminoid component. All occur on coarse-textured,
erosive soils, but some, such as Jumperus hon^^ntalis /
Carex inops, are confined to steep cool exposures
(Hansen and Hoffman 19SS).

Jumperus kon^^nialzs generally has greater than 60 %
cover at these sites and is the primary substrate
binding agent; other shrubs occur in trace amounts.
Within the plot, the dominant graminoid is threadleaf
sedge (Carexfilifoha) but the indicator graminoid is
actually bluebunch wheatgrass (PssudorosgTiena
spicata). Across the local landscape there was a
fluctuation as to which of these two species is
dominant. The grass for which the t3^e is named,
little bluestem (Schz^^^achynujnscopa.num), is at best,
sporadically distributed in this RNA, though overall
its distribution is largely coextensive with that of
Pseudoroegnena spicata in considering sites of this
nature and in this region (thus its attribution as an
indicator species as well). Other grasses present and
typical of sandy sites include plains reedgrass
(Calaviagrostis Tnontanensts') and prairie sandreed
(CalaTnoi'ilfa loTigifohd). The forb component is
diverse, ranging up to 20 species on a plot, but only
one or two species, usually standing milkvetch

(Astragalus adsurgsns) or purple prairie clover

(PetalostsTnon purpureurri), are present in greater than

trace amounts.

CPlot NHMTECRN98SC00093

Rhus aromaiica / Pseudoroegnena spicata Shrubland

CRHUARO / PSESPQ
fragrant sumac / bluebunch wheatgrass shrubland

The sample plot is representative of steep, erosive and
high solar intensity slopes^ mostly of upper slope and
slope shoulder positions. This association occurs
predominantly as small patches. This severe and
heterogeneous environment has a concomitantly
sparse and variable vegetation composition with
dominance in the shrub component alternating, in no
readily explained manner, between soapweed yucca
(Tu^ca glauca) and fragrant sumac (Rhus aromaticd).
The soils are fine sands to sandy loams in texture and
rills, gulleys and faceted slopes are testimony to their
erosive nature. There are at least six closely related
plant associations that occur as small patch t3^es on
sites with abiotic parameters comparable to those of
RHUARO / threadleaf sedge (Carexjihjbha),
RHUARO / little bluestem (ScJwzachryiumscopanurd),
RHUARO / plains muhly [Khihlenhergia cuspidata),
Tucca glauca / Calamovtlfa loTigifoha, and Tucca glauca
/ Pseudoroegnena spicata (Hansen and Hoffman 19S8,
Jensen et al. 1992, Schneider et al. 1997, DeVelice et
al. 1995). There are no vegetation keys that permit
one to unequivocally identify /differentiate these
communities, but the site descriptions and vegetation
parameters most closely match the RHUARO /
PSESPI association described by DeVelice et al.
(1995) for northcentral Montana and RHOARO/
PSESPI (Shallow Depth Ecological Type) by Jensen
et al. (1992) for western North Dalcota.

Rhus aromatica, Tucca glaiica, Jurdperus kon'^ntalis and
ArteTmsiaJhgida are ubiquitous shrubs in this type, but
only the first three listed exhibit even 5% canopy
cover within this landscape; all other shrubs are
present in trace amounts. These low shrub coverage
values are not in accordwith the modal description o?
the type on a regional basis. For the sample plot, and
most of association as developed on the RI'JA, there is
not a dominant graminoid; rather there exists an
assemblage of graminoids typical of coarse-textured,
well-drained sites. In approximate order of
importance, in decreasing cover, these include: plains
muhly (KluMenhergia cuspidata), 1 ittle bluestem
[ScJu^ackryium scopanum), bluebunch wheatgrass
[Pseudoroegnena spicata)^ prairie sandreed (Calamoz/ilfa
longifolia), indian ricegrass (Qry^opsis kymcnotdes),

32

needle-and-thread (Shpa comata), sand dropseed

(Sporobolus ctyptaTidnis) and red threeawn (Anstida
iongiseta). Other short graminoids, not necessarily
associated with coarse-textured soils, can also
dominate these sites. The forb component is low in
cover and extremely heterogeneous, with high
diversity (30 plus species / plot-sised area) in some
areas and scarcely one third that in others. Almost
ubiquitous within the t3^e are hairy golden-aster
(Usterotheca '^llosa)^ American vetch (Ficia aTnencana)^
scarlet globemallow (Sphaeralcsa cocinea), scarlet gaura
(Qaura cocciTisa), silver-leaf scurf-pea {PsoraUa
argopkylla) and woolly groundsel (Senecio canus).

Other Vegetation Types: Where the woody draw
opens up and widens some distance above the
reservoir, ]ih.Q ArtsTmsia canii / Pascopyrurn STmihii
(silversage / western wheatgrass) association is found
contained within what becomes a broad, shallow
drainage. This community is in relatively good range
condition as indicated by the high cover o^ Pascopyrum
srmtkii and low cover of Kentucky bluegrass (Poa
pratensis). Comparatively large specimens o^ Ariermsia
indsntata ssp. luyormngensu contribute to the high
shrub cover.

OVERALL BIODIVERSITY SIGNIEICANCE:
The boundaries were set to encompass the ravine, a
particularly well -developed landform with its full
complement of associated vegetation. The ravine itself
is an unusually mesic woodland in its composition for
the Northern Great Plains biome, though limited in
development and extent as is the case for most
Missouri Breaks ravines due to their narrowly- incised
settings of limited length. This gives it all the more
contrast and development in north-south ravine slope
vegetation as a consequence. A segment of the
surrounding upland grassland features are
serendipitously included that are representative of a
prevailing Northwestern Unglaciated Plains
landscapes in excellent condition. Such grassland
habitat extends onto adjoining lands to provide
landscape continuity, though adjoining lands are not
in as good a condition. As such, Spring Creek RNA
potentially affords a rangeland reference area for land
managers and ecology researchers, and a striking
fenceline contrast in range condition between
adjoining pastures. Wildlife values were not evaluated.

LAUD USE;

The site has been protected from livestock grazing to
keep the area in a natural state, presumably since
establishment in 1991. The current excellent
condition and paucity of exotic species suggests that it

had been managed in good-excellent range condition
prior to establishment.

IvIANAGEMENT COMlvlENTS:
Mo immediate management issues or concerns were
identified; there is a nearly complete absence of
noxious weeds and exotic species invasions with only
limited yellow sweetclover (Kfelilotus offtcnwhs) along
the western bottoms. Vehicle access is limited, and
weeds were found along the corridor.

Historically, fire and bison grazing were two major

driving forces in this landscape, responsible for
renewing the vigor of the grasses, stimulating forb
numbers, and keeping shrub density low.
Re introduction of appropriately timed fire is a
management option to consider in stimulating grass
upland grass vigor and foii> flowering, within the
wildlife management framework.

The grassland had relatively low forb numbers and
litter accumulation. Apolicy excluding wildfire
suppression under discrete terms, if not a rotating
prescribed bum treatment, may warrant fiirther

consideration. The consideration of any treatment
must factor in yellow sweetclover life history and
responses.

Two Calf-Douglas-fir Research
Natural Area

environment:

Two Calf-Douglas-fir Research Natural Area is a 160
acre block of Missouri Breaks that encompasses an
almost 500 feet vertical gradient of moderately to
highly dissected terrain developed in highly erodible
shales. It encompasses the upper slopes of Knox
Ridge and extends northward down to the Missouri
River; thus the mostly steep slopes have contrasting
north- and south-facing exposures. Barely reaching
the southern boundary, in the vicinity of an ephemeral
stream feeding Two Calf Creelc, a wildfire has burned
the predominantly pine forest, leaving scattered
blackened snags. At the northern base of the ridge, the
ridge slope tapers into highly convoluted slump block
terrain with small-scale relief that is not revealed by
^0 feet contour intervals. The climate is essentially
Continental (refer to the characterization of Mobridge
monitoring data, presented for Missouri River
Bottomlands RNA.)

33

"VEGETATION:

Bisected as it is by a east-west oriented ridge, the
RNA's predominant vegetation cover includes
moderately to densely forested north-facing slopes

and open woodlands and sparsely vegetated south-
facing slopes.

Pseiidotsuga men^issii / Ory^opsis Tmcrantha Eorest

[:PSE1vIEN / ORYMIC3

Douglas fir / little-seed ricegrass forest

This forest t3^e is of very limited rangewide
distribution, found only in the Missouri River Breaks
of Montana. It was originally characterized by
Roberts and Sibbernsen (1979) as Douglas fir / plains
muhly forest {Pseudotsuga me/f^ersii / AhJderThergza
cuspidate Eorest) as a result of mis identifying the
dominant grass in vegetative condition. It occupies
moderate to steep slopes with northwest to northeast
aspects. This is a major type within the RNA, where
it is best-developed on steep north-facing slopes that
are very undulating in both the horizontal and
vertical. It appears to be developed on the same
erodible shale substrate that supports other forested
types as "well as long-leaved sagewort/indian ricegrass
barrens on south exposures. The ground surface has a
patchy cover of mosses and lichens, the combined
cover of which generally exceed 50%; about '^% is
contributed by litter and the remaining 10% is bare
soil which shows sheet and rill erosion in places.
There were no fire scarred trees or buried charcoal
which tends to support the contention of Roberts et al.
(1979) that these sites experience low fire frequencies.
This type usually grades to ponderosa pine woodland
on drier/ warmer slopes or on flats and toeslopes
below.

The overstory approaches canopy closure with
Pseudotsuga menio^esii (Douglas fir) strongly dominant
and PiTius pondsrosa (Ponderosa pine) scattered. The
understory has numerous stems o^ Pssvdotsuga
Tnen^^esii^ and Rocky Mountain juniper (Jumperus
scopuloTum) in a distinctly shrubby form. The canopy
is too dense for Pim^s ponderosa reproduction, clearly
xi^ZikAn^Pssxidoisuga 7nen%iesti the climax dominant and
apparently serai dominant as well.

Eurther evaluation of old-growth characters may be
warranted. Tree ring studies were conducted among
Douglas fir at a study site referred to by the nearby
"James Kipp Recreation Area" out of the National
Laboratory of Tree Ring Research; they documented
the oldest age among sampled Douglas fir trees to be
■4?91 years (L. Smith pers. commun. to J. McCollum,
1982).

The rhizomatous western snowberry [Symphoncarpos
occidentahs) is present in patches as the dominant
shrub; squaw currant (Rihss cerium} is consistently
present, as is the intensively browsed common
chokecherry (Prunus virgimaTia). The dominant herb,
little-seed ricegrass (Qty^zopszs Tmcranthd) is highly
variable in cover; the sample plot represents the high
end (4iO%) of this species' cover values. Sun sedge
[Carex inops) and bluebunch wheatgrass
(PssiidoTsognena spicata) are consistently present w ith
coverages generally not exceeding 5%. Oregon
woodsia (IVood^ia oregana) is a fern occurring in more
than trace amounts; forb cover is lower than this.
Yellow sweetclover (Mehlotus officinalis) is also present
in trace amounts despite the shaded environment,
confirming its aggressive nature and broad ecological
amplitude. ^Plot NHMTECRN98SC00133

PiTius ponderosa / C(?r^j:z/io/is Woodland

CPINPON/CARINO;]
Ponderosa pine / sun sedge woodland

On the steep, south slopes a complex of open pine
woodland vegetation encircled the sparse vegetation
associations of the long-leaved sagewort / indian
ricegrass barrens. The open pine stands represent a
wooded shale barrens complex in which there were
frequent clumps of sun sedge {Carsx mops'), but
undergrowth vegetation was sporadic, sparse over
most of the area, and variable in composition. Plains
reedgrass (Calamagrostis montanensis) and bluebunch
wheatgrass (Pseudoroegnena spicata) were also locally
abundant. Multiple plots would be needed to make
generalisations and characterize the highly variable
structure. This plant association has been documented
as a distinct woodland community on isolated shale
outcrops such as the War Horse Area of Critical
Environmental Concern (Lesica 1987), and on
secondary tributaries of the Missouri River such as
Woodhawk Creek (Heidel 1996) where it is generally
better-developed under less harsh conditions. |^No
Plot^

The virtually ubiquitous ponderosa pine of the plains
has been employed for reconstructing climate history
as indicated by tree ring patterns. Ponderosa pine
elsewhere on the Refuge have been cored to document
Great Plains drought history (Meko 1982. 1992).

Artermsta lon^foha / Ory^zopsis h/menoides Spai'se

Vegetation

CARTLON / ORYHYlvT]

long-leaved sagewort / indian ricegrass barrens

34

This community was found on steep, south-facing
slopes eroded from acid shales that heat up under
direct exposure to the sun. Both of the community co-
dominants, long--leaved sagewort (Arisrmsia longifoha)
and few -flowered wild buckwheat {Knogonum
paiiciflourn) are highly associated with soils derived
from acid shales (in Montana the Bearpaw, Colorado
and Clagget Shales) and bentonite. These shales are
intrinsically highly erosive, with sheet, rill and gi-jlly
erosion evident on site. Plant establishment is further
hindered by low pH values (< 5) and low values for
moisture available to vegetation. Thus, these sites are
-^Gvy stressful for vegetation and support a suite of
uniquely adapted species, which individually, or in the
aggregate, seldom exceed 10% canopy cover (the value
defining the break between sparse/not sparse in the
NVCS). We have placed the inventoried stand into
this association using the key of DeVelice et al. (1995);
the congruence between our plot and their description
of this type, regarding both environment and
vegetation, is close. Several species present are
generally associated with sandy soils such as soapweed
yucca (Tiicca glaiica), indian ricegrass (Chy^^pszs
hymenoides), and prairie sandreed (Cala-movilfa
longifoha). They are found on these clay shales because
the weathering process produces a substrate of
predominantly sand-sized platy shards rather than the
clay-sised particles that are the ultimate product of
shale decomposition. Forbs t3^ical of disturbed sites
occur in trace amounts.

Other vegetation: In the northern portion of the RNA,
where the ridge slope tapers into highly convoluted
slump block terrain, there is a fragmented and
repeating pattern in plant communities within a short
distance, with most of the communities occupying
only a few square meters. Some of the communities
noted but not formally sampled were Pinus pond.srosa /
Carex ntops and a Pinus pojiderosi2-doni\n:ited type that
had virtually no undergrowth due to the highly
erosive nature of the substrate. These types were not
sampled because the surface was so rolling and
convoluted that there was no portion extensive
enough to accommodate a plot sample.

Other Vegetation T3^es: Small patches of the
following types were noted; ChrysothaTnniis naussosvs —
KnogonuTn paucijlorum (a variation of ARTLON-
ERIPAU), PASSMI; SARVER-ATRGAR and
ARTTSW/ PASSMI.

OVERALL BIODIVERSITY SIGNIFICANCE:
This is significant as the only RI^-JA that contains the
uncommon to rare PSElvlEN/ORYMIC Forest. It

encompasses an interesting contrast of vegetation for
a small area; juxtaposed with the north slope
PSEMEN/ORYMlCc.t. is ARTLON/ ORYHYM
occurring on the steep south slopes. As such, this

RNA captures an unusual slice of the Great Plains
bio me.

Two Calf-Douglas-fir RNA is contiguous with and
provides access to the Missouri River Bottoms RNA
below. While the latter does not provide additional
PSEMEN/ORYMIC habitat, the two together
represent an outstanding example of the Missouri
Brealcs gradient and complement one another.

The relatively dense canopy of the PSEMEN /
ORYMIC and other forested types of northerly slopes

constitute important thermal and hiding cover for
large native ungulates. The palatable shrubs of these
slopes, including serviceberry (ATnslanchzer almfoha.)
and chokecherry [Pruniis virgimana) have been
reduced to stubs only inches high presumably due to
intensive wildlife browsing.

The skewed orientations of otherwise straight tree

trunlts ("drunk forest") raised questions about the
history of slumping. The massive slope wasting
phenomena currently under intact vegetation signifies
an interesting subject for research into "natural"
stability /instability of this landscape.

LAI-JD USE:

The general area has been grazed in the past but local
conditions are unfavorable ^ot such use. It may have
been subject to selective removal of suitable trunlcs of
Juniperus scopulorwm cut for fencing, and trunks of
Psevdotsuga men^iesii cut for fuel and building material
from the perimeter of the stand where access and
removal were practical.

MANAGEMENT COMlvIENTS:

This 160 acre patch is too small to encompass the
disturbance regimes (wildfire, wildlife browsing and
grazing) affecting the Missouri Brealcs landscape. If a
wildfire were to burn the northern slope, it would
probably crown-out and be stand replacing for most of
the forested landscape, setting back the forested
landscape to an early serai stage. Tree seedling
establishment in such an environment is very
sporadic, and it talces many years for an establ ished
forest to produce a mature stand in this dry
enviroment. Given the rarity of the type and relative
lack of its protection in Montana, it would be prudent
to either add more area of this association to the
current RNA or find additional examples of high
qualityPSElvlEN/ORYMIC that could be placed in

35

RNA status. Mote: Fire suppression was identified in
the original establishment record as needed to
maintain the vegetation.

There is currently not a weed threat to the area, but
the introduced yellow sweetclover (Jslehlobis qfficiTtahs)
is aggressive and can be seen invading environments
as disparate as PSEMEN / ORYMIC andARTLON /
ORYHYM. "While it has the greatest potential for
expansion on the shrub and grassland sites within the
area, it could also proliferate with fire or other major
changes to forested community structure.

The Knox Ridge road is a maintained BLM road
running through the area that is a potential corridor
for new invasions of exotic species. Any road-grading
work on such a steep-sided, narrow ridge also presents

the possibility of destabilising the slopes that drop off
on either side.

York Island Research Natural
Area

ENVIRONlvlENT:

York Island is a 120 acre island in the eastern end of
Fort Peck Reser'voir in a highly exposed setting at the
juncture of the main reservoir and the Dry Ar'm. Its
highest point rises less than SO feet above water level.
The undulating to sharply incised surface is t3^ical of
Missouri River Breaks Subsection (f ) of the
Northwestern Glaciated Plains Section {33lD, Nesser
et al. 1997) where dissected river breaks have formed
in shale, sandstone and siltstone. Most of the island's
communities are developed on soils weathered from
fine-textured sedimentary parent materials, including
montmorillinitic clay, i.e. bentonite, awater deposited
volcanic ash. The climatic regime is Continental with
hot summers and frigid winters; 35% of average
annual precipitation (total 1 1,6 in.) occurs in May and
June (data averaged over years 1 956 to 1 998, from
Fort Peck Power Plant, Western Region Climate
Center).

"VEGETATION:

This site has two major community t3^es, both
dominated by Wyoming big sagebrush (ArteTtusia
tndentata ssp. zvyoTmngsnsis). While the cover of this
species is quite variable, it averages in the mid-20%
range, i.e., close to the 25% threshold for
distinguishing between shrubland and herbaceous
vegetation (Federal Geographic Data Committee —
Vegetation Subcommittee 1997.} We have described

only the ArtsTmsia tndentata shrub assoc iations (shrub
canopy cover > 26%) because they appeared to be
more prevalent and their intergradation with the

herbaceous associations is structurally and
compositionally overlapping without a discretely
different herbaceous type at another extreme.

Ariemisiiz tndentata ^^ zvyoTmngensis / PascopyTU/tn

snuthzi Shrubland

CARTTSW / PASSMQ

Wyoming big sagebrush / western wheatgrass

shrubland

This is the most extensive of the island's vegetation
types, occurring on fine-textured silt and clay loams
derived from shale and siltstone, and found on upland
benches and gently to moderately inclined slopes of all
aspects. Its varies between the more densely
vegetated undulating uplands and more sparsely
vegetated south-facing slopes, inversely related to the
amounts of exposed soil and gravels (less than 20% -
over 80%, respectively). Traces of scattered rounded
rock are testimony to past glaciation but veneers of till
were not found as part of the soil profile.

Upland sites have a notably well-developed
microbiotic crusty including crustose lichens and algae,
that constitute as much as 80% cover. This is
circumstantial evidence that the site is a refiige from

grazing ungulates, and previously had light use when
it was contiguous with the mainland.

As noted above, Wyoming big sagebrush (ArtsTmsia
tndentata ssp ivyormngensis) dominates the shrub
component, coverages varying from approximately 15
to 30% but giving a distinct impression of a shrubland.
Other shrubs/ subshrubs present, generally in less
than trace amounts, are fringed sage (Arte-rmsia
fngida), broom snakeweed (Gutisrrs^ia sarotkr^y and
fragrant sumac (Rhus tnlohata). The graminoid
component is dominated by western wheatgrass
(PascopymmsTmikii) with higher coverages (to 60-70%)
occurring on rolling uplands. Other graminoids
commonly present but in amounts not exceeding 5%,
include: narrowleaved sedge (Carex stenophylla)^ sun
sedge [Carex inops), needle-and-thread (Siipa coTnatd),
blue grama (Bouteloua gracilis), and prairie junegrass
(Koelena Tnacranika). Green needlegrass (Masella
vindida) is sporadically present; higher coverages of
I'Jasdla 'i^indida, generally on lower north-facing
slopes or toeslope postions, indicate a transition to
more mesic and relatively scarce habitats (and the
ARTTSW / PASSMI-NASVIR association). The
forb component constitutes little cover, the most
constant species being scarlet globemallow

36

(Sphaeralcea coccinsa), northern fairy-candelabra

(A?idrosace ssptsntnoTialis), and Nuttall's pussy-toes
(AnUnnana parvifiora)

Yellow s'-veetclover (Melztotus officinalis) is uncommon
in the sampled stand but ■widespread; other
expressions of this association are densely carpeted
■with this introduced species and it has high potential
for expansion. Other introduced species such as goat's
heard (Tragopogon duhius) and Japanese brome (BroTmis
japomcus') generally have insignificant populations.
CPlot WHMTECRN9SSC0001;]

Artermsia tridentizta s^ wyoTmngensis / Pseudoroegnena

spicata Shrub land

CARTTSW/PSESPi;]

Wyoming big sagebrush / bluebunch wheatgrass

shrub! and

This association is typical of the mid- to upper-slope
positions of steeper slopes of all aspects, associated
with glacial drift soils, both coarser-textured (mostly
sandy loams) and having greater amounts of gravel
than the ARTTSW/PASSMIp.a. It grades to
ARTTSW / PASSMI both at downslope positions
and at slope shoulders. General ly both the amount of
bare soil and the exposed gravel/rock comprise
upwards of 70% of the substrate; the erodible surface
probably accounts for the lack of a microbiotic soil
crust. Wyoming big sagebrush {ArteTmsia tndeniata
ssp. WyormTtgcTtsis) ranges in cover from 10% to
upwards of 35 % but generally exceeds 20%. Fringed
sage {Artermsiajrigida), broom snakeweed [Guiisrrs^ia
sarothras), soapweed yucca {Yucca glaitca) and aromatic
sumac (Rkus tnlobata) are the shrubs consistently
present in trace amounts. Bluebunch wheatgrass
(PssvA-orosgnsna spicata) is the dominant graminoid,
though its cover does not much exceed 30%.
Graminoids consistently present with low covers and
associated with coarser-textured substrates, or well-
drained xeric sites include, plains muhly (Muhlet^ergia
cuspidata\ prairie sandreed (CalaTnovilfa longifoha), and
sand dropseed (Sporoholvs CTyptandrus'). Little bluestem
(Schz^achynum scopaTium) is often present. The foii>
component is very weakly represented; no one species
can be said to be abundant but scarlet globemallow
(Sphasralcea coccinea), brittle prickly-pear [Opunha
Jragihs)^ Hood's phlox (Phlox hood^ii) and American
vetch (Vicza amencana) are consistently present. I^Plot
NHMTECRN98SCOOOq

Chrysothamnvs naitseosvs / 'Knogonu-m paitafiorum

Sparse Vegetation

CCHRWAU / ERIPALT]

common rabbitbrush / few-flowered wild buckwheat

barrens

This localized and small patch association is found on
the island's uppermost outcrops and has been
previously described in the Limber Pine RNA, in
Valley County (Branson et al. 1970) and in Carter
County (Vanderhorst et al. 199S). Soils of this site
possess no horizons and are weathered from a very
dark grey, possibly acidic, shale and bentonite.
Though the ultimate result of weathering is clay-sized
particles, much of the substrate has just been broken
down to sand-sise particles and thus has better
drainage than would be expected of a soil high in clay;
it is also highly erosive, rills and gullies abound. This
association mostly occupies upper hill slopes or crests
and has depauperate vegetation (< 10% canopy cover).
There is a suite of species adapted to these sites
including long-leaved sagewort (Artermsia longifoha),
common rabbitbrush (Chtysothamnvs nauseosus), few-
flowered wild buclc-vheat (Knogonum pauciflorum K
hrevicaids ^m\ hrevicaide in southeast Montana), indian
ricegrass (Chyi^psis ky?n£7ioid^) and western
wheatgrass (Pascopymm snuikii) that regularly appear
in various mixes and quantities. On York Island.
Knogofmrn paucifloruTn has greater cover than
associated species, and Chysotkamnus Ttauseosus cover is
less than 3%. The 19 species of the sample plot is an
unusually high number; normally species richness
does not exceed 5-10 for this type. I^Plot
NHMTECRM9SSC0003^

Other Vegetation; Other patchy or restricted plant

associations were noted on York Island. The

southern east em point had the best development of the

Rocky Mountain juniper / indian ricegrass woodland

(Juniperus scopulorum /Or^opsis micrantha
Woodland; JUNSCO/ORYMIC) on the steep, north-
facing slope of a small hill. Small patches of western
snowberry shrubland (Symphoricaipos occidental is
Shrubland) are confined to swales and drainage
courses. Sandy ridgetops at the north end have prairie
sandreed — sun sedge (Calamagrostis longifolia —
Carex inops), and the drainage courses graded in
places into the western wheatgrass — green
needlegrass grassland (Pascopyrum smithii - Nasella
viridula Herbaceous Vegetation).

Most of the perimeter was ringed by sparsely-
vegetated shoreline flats of shale fragments colonized
by yellow sweetclover (Mshlotus officinalis), pummeled
by the waves and pounded into rack 1 ines that set off
backwater wetlands in what were previously bays.
These backwater wetlands are colonized and variably
dominated by adventive and exotic species like
common sunflower {Hdianthvs annuus), tumbleweed
(Ajnaranthus alhiis), [Powell's amaranth (Amaranthus
pozvdlu), red orache [AtripUx rosea), slimleaf goosefoot

37

(Ckenopodzu-m leptopkyllum), and Russian thistle (Sabola
kali). The shorel ine is interrupted by scattered, high
cutbanlcs.

OVERALL BIODIVERSITY SIGNIEICANCE:
York Island is a small but representative example of

the Missouri River Breaks segment of the Great
Plains biome. As such, it affords a rangeland reference
area for land managers and ecology researchers. The
island is small and vertical relief is limited, and though
there are two or three major substrate types, overall
biological diversity of the uplands is limited.

The survival of Hotsprings Phacelia (Phaceha
therTnahs) on York Island is confirmed, restricted to
relatively sparse, successional vegetation zones
created by the Fort Peck Reservior; including scoured
beaches and the drawdown zone in "■.vetland
baclc-vaters. Places where Montana rare plant species
of special concern are confined to zones of man-made
disturbance raise questions whether the disturbance
mimics natural habitat conditions or the species is
adventive by nature. One other collection of this
species has been made on the Refuge^ in Douglas-fir
habitat near the former Slippery Anne Guard Station
(E0#001), suggesting that the species occupies natural
habitat elsewhere and the York Island disturbance
may mimic natural habitat conditions. The Refuge is
the only place "where this species occurs on public land
in Montana, and even though York Island does not
have biodiversity significance as a representative site
for this species' conservation, it points to the
possibility of finding such sites elsewhere on the
Refuge.

OTHER VALUES:

York Island is also a Fort Peck Reservior landmark
and shelter for boaters. Archeological artifacts may be
present. Wildlife values were not evaluated.

LAUD USE;

This landscape has been grazed in the past and the
Jumpents scopidoruTn woodland was probably cut for
fencing/ firewood. Juniperus scopuLorum. stem density
probably approaches p re-disturbance conditions. No
signs of grazing were evident. It is otherwise idle
except for occasional visits by passing boaters.

MANAGEMENT COMMENTS:
Exotic species present included Canada thistle
(Cirsium arvense) at scattered shoreline and backwater
locations, and minor upland populations of cheatgrass
{3romus tsctoTU/m) and Japanese brome {Bromus
japomcus). Yellow sweetclover [M^diloius officinalis)
dominates beach vegetation where it may interfere

with shoreline bird use, and forms dense populations
in limited upland areas. It is currently the most
abundant non-native species and may have the
potential to occupy virtually all habitats on the island,
with orwithout disturbance, as evidenced by its
mainland patterns of distribution.

38

Lake Mason National Wildlife Refiige

Lake Mason Research Natural
Area

ENVIRONlvlENT:

Lalte Mason RNA is in a broad, open natural basin,
comprised of two parcels totaling 1,420 acres, lying on
either side of Lake Mason at the lalte perimeter. The
lake outlet is South Willow Creelt, and it has a
spillgate to artificially maintain lake levels. Lacustrine
deposits and alluvium derived from shale and
sandstone are the primary parent materials in the
basin, though residuum derived soils occur in the
"western half of Section 22. All soils appeared to have a
heavy texture, with silty clays and silty clay loams
predominant. The semi-arid continental climate has
peak precpitation in June followed by May, and a
mean annual precipitation of 12.-4? inches (Climate data
from Roundup, Western Regional Climate Center,
191^1997), This RNA occurs within the Montana
Sedimentary Plains Subsection of the Powder River
Basin Section where annual precipitation ranges from
11 to lii inches annually, about a third ofwhich is
snow.

"VEGETATION:

The three main vegetation types are distributed
primarily according to soil moisture regimes, which
vary with distance from Lalte Mason and South
Willow Creelt with the exception of the uplands in
Section 22. All but the standing water (herbaceous
emergent) plant communities were sampled.

Pascopyrum STmthzi Herbaceous Vegetation

CPASSMi;]

western wheatgrass wet meadow

The extensive alluvial flats are dominated by western
wheatgrass (Pascopymfnsrmthzi). There are at least six
plant associations named across the Northern Great
Plains that have Pascopyrum srmthii as the first-named
indicator species; only the t3^e named here is defined
by the virtual monospecific dominance ofP srmtkii
and is rated G3G5 by TNC. This type is t3^ically
strongly associated with sub irrigated alluvial flats and
most of this plant association in the RNA meets
definitions of jurisdictional wetland (Hansen et al.
1996). Consistent with this characterization, both
sample plots displayed gleyed and mottled soils. It is
dominated by a sward of P. snuthzi ranging in cover
from 50 to in excess of 80 percent, making up a
relatively homogeneous expanse on the flats (and

beyond) around the perimeter of the lake. At least in
the sampled locations, weedy or increaser with
disturbance species, e.g. Japanese brome (BroTnus
japotacus), povertyweed (Iva accdlans), wild lettuce
Q^actuca canadensis)^ flixweed tansymustard
(P>€scwrai7\ia sophid), and common dandel ion
{Taraxacwm officinale) dominate the heiijaceous layer
to the near exclusion of native species except grasses.
This may reflect a history of heavy grazing by
livestock. [[Plots NHMTECLM97SCOO01,
NHMTECLM970006^

Pascopyru/m STTnthzi —I'Jasdla vindida
Herbaceous Vegetation

[[passmi-nasvir;]

western wheatgrass — green needlegrass grassland

This association is the prevailing type on non-wetland
alluvial flats and on gentle upland slopes; it
constituted a major plant association prior to
agricultural development. It is ranked as G^ and
occurs in North Dakota. South Dakota, Wyoming, and
Saskatchewan. Stands on the flats grade to the
western wheatgrass wet meadow {Pascopyrum STmtkii
association) and those of the uplands grade to western
wheatgras- needle-and-thread grassland {PascopyruTn
srmthzi / Stipa comata) of drier sites. In the vicinity of
South Willow Creek this t3^e appeared to be in good
to excellent condition with I'Jassella ZJindula (green
needlegrass) canopy cover ranging from 10 to ^0
percent, exceeded only by that of Pascopyrufn STfutkii.
The uplands in northwest corner of Section 22 also
support fair to good quality occurrences of this type.
Other graminoids represented with more than 6
percent cover include needle-and-thread (Stipa coTruzta)
and prairie junegrass (Kodena macrantha) Exotic
graminoids, including Japanese brome (3ro7mis
japonzcus), crested wheatgrass (Agropyron cnstatum),
and Kentucky bluegrass (Poa prate nsis), are much less
prevalent than on the association described above. In
the curtailment of grazing, populations o^^Brumvs
japonicus and Agropyron cnstatum oflen decline, but Poa
pratensis has shown a propensity on similar mesic sites
to increase without disturbance. Povertyweed (Iva
a3::iUans) and scarlet globemallow (Sphaeralcea coccined)
are the only forbs occurring with more than trace
coverages; other high constancy heiijs include prickly
pear (Opuntia polyacanihd), American vetch (Vicia

39

izmencaTta) and Holboell's rockcress (Arahis holhodhi).
Fringed sage (Arte-misiajrigidd) is the only shrub-like
plant consistently present; however, it does not exceed
3 percent in canopy cover. |^Plots
WHMTECLM97SC0003, NHMTECLM9 7 30000^3

Sar(x>hatus vermictdaius / Pascopyrwmsmithii Shrubland

CSARVER/PASSMi;]

black greasewood / western ■wheatgrass shrubland

This is a common type on the RNA, especially
extensive on the eastern side of the lake where it
occurs upslope by a matter of a few tenths of a foot or
more from the PASSMI alluvial flat t3^e. These sites
are more salt-affected than those "with Pascopynim
STmthii alone. This t3^e as found in a lacustrine setting
typically develops under conditions in which salts
from alkaline lakes are deposited by wind and water
on the leeward shores. The deposition process and
probably other habitat conditions are altered by the
artificially maintained water levels on Lake Mason.
Black greasewood (SarcohGhis ii^rmiculatus) occupies
slightly raised mounds, perhaps as a result of an
ensuing erosion processes. Its canopy cover is low,
varying between 5 and 20%. but the visual aspect is
that of shrubland due to the stature of the Sarcohahis
verrruadahis relative to that of the associated
undergrowth. This association in the RNA has been
as affected by grazing as those of the PASSMI type
judging by the dominance of increaser species,
foremost among which are Japanese brome (Btovi;us
japomais) and povert3'weed [Iva a3::illans). We noted a
micro-patterning within this type as the dense patches
o^ BroTmis japonicus seemed to have little Pascopynnn
STtuthn and conversely where tillering of P STmtJm was
especially dense there was little B.japomcits. This is
not a high-quality occurrence of this type due to the
altered environment and abundance of weeds. I^Plot
WHMTECLM97SC0002^

Atnplex gardnen / PascopyruTn srmthii E>warf Shrubland

CATRGAR / PASSMq

Gardner's saltsage / western wheatgrass dwarf

shrubland

This association occurs as small stands that are
fractions of an acre, across salt-affected
alluvial /lacustrine flats on the west side of Lake
Mason. These sites undoubtedly have standing water
during spring runoff and are slow to dry given the
clay soils. At least one site had mottled and
incipiently gleyed soil, indicating an oxygen depleted
condition developing during extensive inundation.
Canopy cover of Gardner's saltsage (AtnpUx gardnen)
is typically not greater than 20 % and usually
exceeded by that of a suite of graminoids whose cover

contributions are quite variable, including western
wheatgrass {Pascopyrum STmthii) as usual dominant,
meadow barley (Hordewmbrackyantkenijn), foxtail
barley (H.ordeu7njid)ahim)^ bottlebrush squirreltail
(Sitamon h^fstnx), and Nuttall's alkaligrass (Pucanelha
nuUalhana). Graminoid cover is higher for the Lake
Mason examples than has been seen elsewhere for this
association and probably reflects the relatively
favorable soil moisture of these sites. The herbaceous
component is dominated by weedy species or ones that
increase with disturbance, including povertyweed [I'ija
axillaris), clasping pepperweed (hepidium perforatuTn),
wild lettuce (l^actitca canadensis), and common
dandelion (Taraxaaim officinale). The suite of
weeds/ increaser species indicates past disturbance and
decreases the baseline value of these sites. j^Plot
NHMTECLM97SC00063

Emergent Wetlands: We did not sample the semi-
permanent emergent wetlands that are part of the
RNA. They are dominated by hardstem bulrush
(Scirpus acutus) with abundant sago pondweed

[Potarriogeton pectinatiis), chara {Chara spp.), and water
buttercup (Ranunculvs spp.).

OVERALL BIODIVERSITY SIGNIFICANCE;

The Lake Mason RNA represents a fair example of the

western wheatgrass wet meadow {PascopyruTnsrmtkiijy
and a once common plant association, Pascopyrum
STmthii-l'Jasella iiindula with isolated areas in good
condition. There are also weed infested occurrences of
a less common type (PASSMI alluvial bottom) as well
as several associations (SARVER / PASSMI,
ARTGAR / AGRSMI) of more restricted occurrence
also plagued by weeds. We are not prepared to
address the affects of elevated water table levels to this
low-lying RNA. In comparison with Mullan Trail
RNA and its glaciolacustrine setting, it protects more
plant associations and more of the hydrological
gradient, though the overall ecological condition is
not as high. The RNA is contiguous in places with
surrounding grasslands connecting to the larger
landscape of the surrounding basin slopes.

LAI-JD USE:

Lake Mason was intensively grazed in the past.

Livestock grazing ceased in 1980.

MANAGEMENT COMMENTS:
This RNA maybe suited to study of community
succession and habitat values with and without
restoration practices. The elevated water levels may
affect restoration potential and efforts to simulate the
historical ecological drivers of grazing and fire.

40

Medicine Lake National Wildlife Refuge

Big Island Research Natural
Area

ENVIRONMENT: Big Island RNA is the second
largest of h-vo major islands in Medicine Lalte at 961
acres. It has a loioll on the south end that rises 35 ft.
above the lake, and two distinct wetland swales north
of the knoll, but most of the island is less than 10 feet
above lake level and gently rolling. A large bay at the
south end is set offby two long isthmuses that are at
or below the lake water level and covered mainly by
robust grasses. The island is mapped as Blanchard fine
sand, 4i-20% slope (Richardson and Hanson 1977)
though most of the island soils appear to be loamy
sands; possibly with glacial till on the knoll at the
south end. The controlled lalte level affects the island
shore, regulated at the Lake Creek outlet, with a dam
and spillgate to artificially maintain lalce levels. The
semi-arid continental climate has peak precipitation in
June followed by July and May, and a mean annual
precpitation of 13.25 inches (Climate data from
Medicine Lake, Western Regional Climate Center.
1911-1997).

Note: This area and b.vo others in Medicine Lake are
part of the 11,366 acres designated as Medicine Lake
Wilderness Area.

VEGETATION:

The array of plant communities forms a grassland-
shrubland mosaic. It can be explained by both small-
scale relief as evidenced in height above the lalce level,
which need vary only a fraction of a foot in order to
influence vegetation, and by soil texture. See Figure 2
for an occurrence map of Big Island vegetation types.

Syrriphoncarpos occidentalis Shrubland
CSYMOCq

western snowberry shrubland

Western snowberry shrubland constitutes the most
extensive vegetation type on the island and is among
the most common shrub communities in the Northern
Great Plains. Across the island it occurs on gently
undulating topography and s'.vales. some ofwhich may
be seasonally inundated, intermittently flooded, or
sub irrigated via subsurface connection to Medicine
Lake. It has been characterized in regional
classifications as a "temporarily flooded" system, but

this is the case on Big Island only when ground thaw
is delayed. As noted by Hansen et al. (1995) and
exemplified on Big Island, it spans an environmental

range from mesic upland slopes to wetlands (hydric
soils and wetland hydrology). Almost none of the
sites in the RNA would qualify as jurisdictional
wetlands because the dominant, western snowberry
(Symphoncarpos ocadentahs), and all associated species
including western wheatgrass (PascopyruTnsTmthzi),
smooth brome (BroTnus inerrms), and Kentucky
bluegrass (Poa prats nsis) are rated FACU (Facultative
Upland, i.e. only occurring in wetlands less than 55%
of the time and conferring no wetland status) by the
U.S. Fish and Wildlife Service.

This t3^e has nearly continuous cover of
Symphoncarpos occidentalis, a shrub that produces
sucker shoots emanating from stout, spreading
rhizomes. Given the density o^ Symphoncarpos
occid^ntahs, it is perhaps not surprising that the only
other native species found with regularity are also
rhisomatous (mentioned above). Weedy species such
as flixweed tansymustard (Descuraima sophia), pinnate
tansymustard (Descu/rairfza pinnata)^ and leafy spurge
(Kupkorhia esida) occur in patches, perhaps areas
formerly disturbed. The Euphorbia esvia is widespread
on the island in these relatively moist communities
and appears to be expanding, forming dense clones.
Several species of spurge fleabeetles have been
introduced on the island to provide leaiy spurge
control.

Ordinarily SyTnphoncarpos occidentalis shrubland occurs
in small stands rather than as a prevalent vegetation

feature. It is recommended that recent aerial
photographs of the island be compared with historic
photos if it is possible to determine from them
whether shrubland has been present since early years
ofRefuge establishment. A literature review and
dialogue with other refuges of species' management
responses and wildlife benefits or deterrents might
also be helpful in applying current vegetation
information to wildlife management and noxious weed
management matters.

Shpa comata - Boutdoiia gracilis — Cartxjilifolia

Herbaceous Vegetation

CSTICOM -BOUGRA-CARFIL)

needle-and-thread — blue grama — threadleaved sedge

grassland

41

This grassland association is found on soils ranging
from sandy loam to fine sand. It occurs on Big Island
across higher ground and warmer exposures. It is

dominated by needle-and-thread (Stipa comala),
usually having greater than ^0% canopy cover. Blue
grama (Bouteloiia gracilis) and threadleaved sedge
(Carexjilifohd) are consistently present, often as co-
dominants. Their cover can exceed that o^ Stipa coTnata
and varies greatly across the landscape v/ ith no
obvious correlation to site factors, "'.vhether due to
distuiijance patterns, imperceptible environmental
disturbances, or chance. This type usually grades to
grasslands dominated by western wheatgrass
(Pascopyrum simthzi) including PASSMI - CARFIL or
PASSMI -BOUGRA While western wheatgrass is
consistently present in this prevailing association, it
has low cover values. Forijs are a minor component;
only pricklyp ear [Qpurdia polyacantha) consistently
occurrs in more than trace amounts, favored by the
sandy substrates or else the land use history. Fringed
sage (ArieTmsiafngida) is consistently present, but
seldom exceeds trace amounts. This community type
generally has only minor populations of weedy or
exotic species; for example, crested "■-"vheatgrass
(AgropyroR cnstatum) occurs in scattered blocks.
STICOM-BOUGRA-CARFIL is aprevailing
grassland association of the Northern Glaciated Plains
where medium- to coarse-textured soils are found, as
previously described for Spring Creek RNA. I^Plot
MTWHECRA97SC0007^

Hansen and Hoffman (1988) recognize this type by the
dominance oi PascopymTnsjmthzi over Stzpa coTnata and
generally this occurs only on lower positions in this
landscape or those having planar surfaces and/ or

having finer-textured soils. Where PascopyruTti srtuthii
and Stipa coTnata are co-dominant, or nearly so, we
have assigned these sites to PASSMI -BOUGRA -
CARFIL due to the appreciable cover o? Pascopyrum
sTmthii indicating the higher moisture status of these
sites (technically, several examples of this community
with Ci7?"^j:spp. dominant do not "key out" to any
type). Mote that Schneider et al. (1997) in the
provisional Great Plains vegetation classification of
The Nature Conservancy, recognize a Pascopynctn
svuthzi —Shpa comata community type but cite no
parameters for its recognition. The whole complex of
community t3^es involving PuscopyniTn STmthzi, Shpa
comata, Carexjilijbiia, Carex stejiopkylla, Boutdoita
gracilis, and I^Jasdliz vindida needs extensive work to
separate intrinsically different environments from
disturbance induced states and a worlcable key for
discriminating types one from another. This, too, is a
broadly distributed Northern Great Plains plant
association (CO, MT, ND, SD, SK, WY).

Pascopyrwm svuthii — 3outeloua gracilis — Cartxjilifolia

Herbaceous Vegetation

CPASSMI-BOUGRA- CARFIL]

western wheatgrass —blue grama — threadleaved

sedge grassland

This is an uncommon grassland asociation <:>t\ the
island because fine-textured soils are limited. Western
wheatgrass (Pascopymjn sjTuthzi) is typically the
dominant graminoid in this grassland association, but
narrowleaved sedge (Carex stsnophylla; synonym: C
eleochans) was the dominant graminoid (50% c.c.) in
the one island plot. Only two shrub-like plants,
fringed sage (ArteTntsiafngida) and broom snakeweed
(Quhsrrei^ia sarrothrae), regularly occur but with low
cover values. Rush skeletonweed {hygodesTmajuTicsa),
American vetch (Ficia aTnencana)^ pricldypear (Opuntia
polyacantha) and scarlet globemallow (Sphaeralcsa
coa:iT!£a) are the forbs with high constancy but seldom
do their coverages exceed 1 or 2% under natural
conditions. The occasional bunch of crested
wheatgrass (Agropyron cnstatitTn) and patch of fl ixweed
tansymustard (Descuraima sophia) indicate past
disturbance. ^Plot NHMTECRA97SC000S]

42

Fi&Jiirc 2-

0,05 0.05 .1 0.1 5 02 C2S
Scale in MjIeb

'.■ IS

Mg^a^ine Lake

Vegetation of Big Island RNA

Medicine Lake NWR

Shrub Cammualitl^S.

'•'■■i[ ] Symphoricarpos occid^ntalEs

^ La :| 5ymphcn"lcai"pos occidentalis (BrDmus inermis)

lb I Symphoricarpos occid^ntalis (Euphorbia esula)
2.1 Symphoricarpos occidentalis (Agropyran smithii)

■^ j S^phcricarpos ocddent?li5 (Prunus vlc^iniana)

.^ I Symphoricarpos occidentalis (Distichiis strictsi.
Euphorbia esula]

4 I Symphoricarpos occidentalis (Spartina pectinate)

'? i^l Frunus virjiniana

fe I Prunus virginiarta (Symphoricarpus occidentails)

. S I Sarcobatus vermiculatus {Prunus vJrginiana.
A^ropyron cristatum)

: 7. I Sarcobauis vermiculatus {Agropyron cristatum.
Hordpgm jubatiam^ vartaus 6XO-tiC£)

Grassland Communities

8 Agropyron smithiii-Stipa Comata

s^ Agropyron smithii-Sdpa comaiia

(CaEamoviifa longifolia-Agropyron smitliii)

9 Agropyron smPthii-Stlp-a comata-Bouteloua gi^cilis

10 Distichlis scricca

1 1 I Juncus balticus-Care* praegracllis

iia. Juncus balticus-Carex praegracilis (Poa prac^nsps)

12 Spartina pectinata (Phragmites communis)

Disturbed Connmunities

13 j Bromus inermis

14 I Bromijs inermis- Agr-Dpyi"&n f rlstacumi

15 I Descurainia sophia {Pelacan damage)
mPH Urikown Plant Communities

Bg^ch
Water
Ephemeral water

At the pvmniap scale nnl all commuiuEy or -dUTmoani^ E>pes i:atL be
elTeclively majjped. ■which. ii«£teilatC5. miillLpJc labels ajid licsiflnatiMja
nf LBciusiona. Buckets Lrn;:]&£e^onniiLmil> L>pcs or sp^Jes ll:iin wngitutc
in tsncpy aiiva it^tir^^A nf a jKjlygoti; uf poLyE^qja Idbclc^d li lajving L>»0
commLimly t>pei the TaraL JjsLed is Ulc more ext^SSivc intl separated Jrnm
Ihc iCKEL^r b^ s.arj1c>n. V-awward ^ils-ihcs {''} ^■rdii^L^b [-) dCpur^Lc ihf
ron^iLlunri Species by ^lucti a pJarJ coisijiJUJiityoc association i^ r^^

Vegdaiiim miipiiniTs fromS"'l Tnite (I :f92fl3 serial pliotography.
Field verifitUJitnc.l^S'/. AJtras RquaJ ArcaPrajcctiAjn DalLimHAD^?.

June J.'^, iy93f

Muniar^ Maiunil J[enla»rt Pmj^jin, 1^'5 Ri3S( Sivl:h Ave, flclcna, MT.Wfj^il

Other Vegetation T3^es: In the lost portions of the

landscape where the water table is shallow.
subirrigated conditions are found. In such settings
"with soils that are salt-afTected. inland saltgrass
(DistzMzs spicatd) forms extensive and nearly pure
communities or occurs in various mixes with Nuttall's
allcaligrass (Piu:cimlha nuttalhana), prairie cordgrass
(Spartifm pechnata) or scattered blaclc greasewood
(Sarcohatus ZJerTmcidatits). Another wetland association
of Baltic rush — clustered field sedge (JuntTus halticus —
Carex praegracilis) occurs in alkaline habitats as
stringers or small patches that are a few tenths of an
acre. Kentucky bluegr ass (Foa prate nsis) often
establishes and outcompetes the natives in this habitat.
Wet areas that are not so salt-affected have Sparhna
psctinatm dominant, with orwithout an abundance of
common reed (Phragrmtss commuTus).

On the island's west shore there is a mix of woody
species and communities that may be more a result of
past, or ongoing, disturbance than any intrinsic site
differences. Associated with near-shoreline locations
and ostensibly subirrigation are several stands of
common chokecherry (Prunus virginiana). some
serving as nesting sites and perches for black-crowned
night herons. They have a margin and sometimes a
low-shrub layer of western snowberry
(Symphoricarpos occidental is) and herijs that are
mainly non-native grasses. Also present is a band of
black greasewood (Sarcobatus vermiculatus), largely
dominated by exotics such as Agropyron cristatum or
opportunists such as Hordeum jubatum and H.
brachyantherum. This vrgetation type is usually
present only on the shores of alkali lalces, perhaps
reflecting the original nature of Medicine Lake.

On the sandiest substrates associated with needle-and-

thread (Stipa comata)-dominated communities were

fragments of associations codominated by prairie

sandreed including Calamovilfa longifolia — Carex

stenophylla and Calamovilfa longifol ia — Pascopyrum

smithii plant associations. These were too small to

map other than noting as inclusions. The abruptness

of the transition between the Stipa- and these small

Cal am ovil fa-dominated communities was noted with

no discernible soil or other enviromental differences to

account for the pattern.

Some of the most heavily-used wildlife habitat has no
associated native vegetation. The places of
concentrated nesting by the American white pelican
colony shifts over time, and the history of use has
created areas with extensive bare ground and patches
of annuals/biennials, oflen dominated by flixweeds
[Descuraima sophia and D pinnata.)

Two Montana plant species of special concern were
documented on Big Island. Site information is
presented on the Element Occurrence Records in
Appendix E, and annotated illustrations are in
Appendix F.

Hairy four o'clock (Alznzbtlts hirsuta) is represented by
widely scattered plants in -^^ry low numbers at
different ends of the island, among plant associations
dominated by both prairie sandreed (CaLzmozJzlfiz
longifoka) and needle-and-thread [Stipa comata). It has
the lack of habitat specificity on Big Island and in the
Sandhills that is characteristic of adventive species.
While it is only known from 3 different counties in
Montana, there have been reports that it is adventive
and more common than records indicate. This study
lent support to the case for moving it offto the watch
list.

Plains phlox (Phlox artdicola) occurs on the north-
facing slopes of the IcnoU on Big Island. It is likely to
be scattered across most of this slope, but the species
was at the ^^vj end of flowering at the time of visit, so
it could not reliably be located to determine
population numbers and extent. It, too, is present on
the Medicine Lake Sandhills. In general, it is
restricted to sandy soils and was previously known in
Montana only from southeastern counties. The
numbers of records in recent years provides the basis
for changing its state rank from Si to S2 (potentially
imperiled); this rank is subject to review with
additional fieldwork in early summer.

OVERALL BIODIVERSITY SIGNIFICANCE:

Big Island RNA supports a spectrum of community
types, from slivers of typical prairie wetlands with
Sparhna pectinata and Dishchlis stncta to dry prairie
and to tall shrub copses. This community diversity
affords a wildlife habitat diversity, even if the plant
assoc iations are not rare or in noteworthy condition.
The complex mosaic of communities may not
representative of the distribution of these communities
in the local landscape due to the regulated
lake water level.

Wildlife values were not evaluated, though it has
previously been noted that Big Island harbors nesting
subpopu Iations of piping plovers, federally listed as
threatened, apopulation of black-crowned night
herons and a large colony of American white pelicans,
both ofwhich are state species of special concern. It is
productive forwaterfowl and sharp-tailed grouse
nesting. Big Island directly contributes to the core
waterfowl production mandate, as well as providing
habitat for colonial nesting birds. It is one of two

44

major islands in the lake, and among the fev/ large
islands in the Prairie Pothole landscape of
northeastern Montana, particularly important in
reducing mammalian predation.

LAUD USE HISTORY;

The island was not previously contiguous "■-"vith the
mainland, but livestock were brought out prior to
Refuge establishment. The vegetation-altering affects
of grazing history compared to raised "'.vater levels and
colonial bird use were not evaluated. Whatever the
cause(s), there are areas that are covered by nothing
but the exotic grasses 3romus inerrms (smooth brome)
and quackgrass (Klymus repsns) to the extent that
native communities are not identifiable. Usually
■where these grasses have invaded native communities,
there are vestiges of the native communities.

MANAGEMENT COMMENTS:

Threats are posed to this whole landscape by exotics

and noxious weeds. Leafy spurge [Kuphorhia esvia) is
by far the most aggressive and tenacious of noxious
"weeds, and is well established. It has a strong presence
in the southern end of the island. Several species of
spurge fleabeetles are established on the island
(Rabenberg pers. commun.) Their potential for
curbing seed production is particularly important
because the seeds of leafy spurge are readily
disseminated by water and wildlife vectors as whitetail
deer and mourning dove.

The far north and south ends have much Canada
thistle (CiTsiuTn a7\jsnse). Smooth brome (BroTnus
insTmis), quackgrass (Elyrmis repens), and crested
wheatgrass (Agropyron i7nstatu/m) are also present
throughout the island. 3ro7ms inertms appears to be
aggressively displ^c in^ PascopyruTn snuthiz from
Sympkoncarpos occidentahs- and PascopyruTnsvutkii-
dominated communities. Other common exotic
grasses include cheatgrass (Bromus tedoji/m), Japanese
brome (Bromvsjapomcus), fowl bluegrass (Poa palustns)
and Kentucky bluegrass (Poa pratensis).

Exceptionally high litter accumulation levels were
noted over most of the Island in both grassland and
shrubland habitat. As mentioned previously, it is
possible that the extensive shrub cover of
Sympkoncarpos occidentahs is an artifact of the land
being left idle. This same shrubland type is present in
trace amounts in Tepee Hills RNA, and though the
two sites have different settings and substrates, they
have some comparative value. The management
options for addressing these situations depends on
desired vegetation structure for wildlife and the
framework for integrating noxious weed management.

Bruce's Island Research Natural
Area

environment:

Unlike Big Island, which is mostly low-lying terrain,
Bruce's Island is a ridgeline that had once been
connected to the mainland, made up of a hump and
toeslopes together totaling 367 acres. It has little of
the micro topography patterns as found on Big Island.
The high shores on the north are eroding in places as
cutbanks, and the gentle toe slopes on the south are
temporarily inundated. Soils are mapped as Dooley
fine sandy loams on the high north end, Dimmick silty
clay in a low-lying middle band, and Williams loam,
undulating at the south end (Richardson and Hanson
1977). The controlled lalce level affects the island
shore, regulated at the Lake Creek outlet, with a dam
and spillgate to artificially maintain lake levels. The
semi-arid continental climate has pealc precipitation in
June followed by July and May, and a mean annual
precipitation of 13.25 inches (Climate data from
Medicine Lake. Western Regional Climate Center,
1911-1997).

Note: This area and two others in Medicine Lake are
part of the 1 1,366 acres designated as Medicine Lake
Wilderness Area.

VEGETATION;

Approximately one half ofBruce's Island on the
higher elevations of the north has been plowed. In this
area, as well as unplowed uplands, Agropyron cnstatu-m
(crested wheatgrass) is the prevailing cover t3^e. As a
whole, the uplands have been sufficiently altered so
that they no longer support intact native vegetation,
instead reduced to small, irregular patches of native
species among the exotics. The potential prevailing
matrix community t3^e of the uplands is probably
western wheatgrass —blue grama — threadleaved
sedge grassland (Pascopyrum STmthzi — Bouteloita gracilis
— Carexjilifoha Herbaceous Vegetation) or western
wheatgrass — needle-and-thread grassland (PascopyruTn
STns-ihn. —Stipa j%f???i?:if? Herbaceous Vegetation), Sample
plots were not established in the course of field
reconnaissance of this RNA because of the lack of
intact vegetation.

The lower lying terrain on the island's southern
portion, particularly along the shorelines, has well-
developed palustrine emergent vegetation. Inland
saltgrass (Lkshchlts spncatd) is among the most
extensive wetland vegetation types, occurring
predominantly as a broad ecotone between wetter
sites dominated by bulrush (Scirpus spp.) or alkali
cordgrass (Spartina gracilis) and uplands. In some

45

locations salt efTlorescence was noted in the Dishchlis
spicata flats, indicating it exists along the capillary
fringe of wetland sites, ^kali bulrush (Scirpus
Tnanhmiis), hardstem bulrush (Scirpus aaitus'), and
sharp bulrush (S puTigens) dominate the communities
at the shoreline fringes in positions that are nearly
continuously flooded. Prairie cordgrass (Spartina
gracilis) was noted in shoreline patches. The Baltic
rush —clustered field sedge meadow (Juncus haltiais-
Carex pras gracilis Emergent Vegetation) is found on
wetland sites that appear to be only temporarily or
intermittently flooded. Where disturbance occurs in
this vegetation, American licorice {Glycyrrki^^a lepidota)
can be an important component. Canada thistle
[QirsiuTn arvsnse) is scattered in with the Glycyrrhi%a
Ispidota and has patches of abundance in the
Sympkoncarpos occidsfdalis Shrubland where it forms a
discontinuous and narrow fringe between true
wetland sites and upland grasslands. There are also
gentle mudflats and a small bay on the eastern shore.

OVERALL BIODIVERSITY SIGNIFICANCE:
There are no intact upland plant assoc iations
represented on Bruce's Island. In the southern portion
of the island there are typical Northern Glaciated
Plains Section wetlands types represented- These
wetland types may be more appropriately sought as
RI'JA targets among natural wetland basins, and it is
expected that they are represented elsewhere on the
refuge system in Montana.

Wildlife values were not evaluated. It has previously
been documented that Bruce's Island harbors nesting
subpopu Iations of pipingplovers, federally listed as
threatened. It is said to be productive for waterfowl
and upland game bird nesting, as well as harboring
significant numbers of Baird's sparrows and
grasshopper sparrows. Site biodiversity significance
may hinge on the contribution of Big Island avifauna
to the Medicine Lake landscape as a whole. The
artificially maintained lalce levels ensure the isolation
of Bruce's Island as an island, directly contributing to
the core waterfowl protection mandate. It is one of
two major islands in the lake, and among the few large
islands in the Prairie Pothole landscape of
northeastern Montana,

LAND USE:

Bruce's Island is covered by a tamegrass planting or
"goback" of crested wheatgrass (Agropyron cnstatum)
over the higher northern half of the island, and has
been grazed in the past. It has been an island since
lake levels were raised.

MANAGEMENT COMMENTS:

The site has limited potential to serve as a natural area

ecology/botany baseline despite its wildlife values

Homestead Research Natural
Area

environment:

The Homestead RNA is a 39 acre tract on a gently-
rolling glacial till deposit above the mouth of Lake
Creek on Big Muddy Creek. Soils are a mixture of
Bowdoin and Lohler clay loams in the northwest
corner, with Manning coarse sandy loam along the
eastern margin (Richardson and Hanson 1977). The
semi-arid continental climate has pealt precipitation in
June followed by July and May, and a mean annual
precipitation of 13.25 inches (Climate data from
Medicine Lake, Western Regional Climate Center,
1911-1997).

VEGETATION:

The grassland is unbroken but heavily invaded or

seeded into smooth brome {Broviiis insrmis) and
quackgrass (Efymvs repens). The closest semblance to a
natural plant association is composed of small, weedy
patches of western wheatgrass - blue grama grassland
(PascopyruTnsTmthzi —Boutdoua gracihs Herbaceous
Vegetation). Sample plots were not established in the
course of field reconnaissance of this RNA because of
the lack of intact vegetation.

OVERALL BIODIVERSITY SIGNIFICANCE;
The RNA was originally designated in recognition of
its value as a lek for breeding sharptail grouse, as well
its reported vegetation features. Wildlife values were
not evaluated.

LAI'JD USE:

There are abandoned quarry sites present. The name
for the area comes from the nearest town called
'Homestead." The site is bordered by roads on two
sides, with deep ditches to drain water from the
roadbed, lowering the water table. Cottonwood trees
have become established in the ditch on the western
margin.

MANAGEMENT COMMENTS:

The site has limited potential to serve as a natural area

ecology/botany baseline despite its wildlife values.

46

Medicine Lake Sandhills

(Part of Medicine Lake "Wilderness
Area)

ENVIRONlvIENT:

The Medicine Lake Sandhills are of recent Holocene
geological development, formed when prevailing
"winds from the northwest scoured sediments from
dried Pleistocene lake beds. The lacustrine beds
contained all particle size classes. Silt and clay
fractions "were carried far downwind but the sand-
sized particles were deposited immediately to the
southeast, and partially reworked by the winds to
form sand dunes. Much of the landscape is rolling but
portions have typical choppy dune features, with
blowouts and associated stages of dune stabilization.
Not all of the springs and seeps were developed for
livestock use, and they contribute significantly to
species and habitat diversity. The semi-arid
continental climate has peak precipitation in June
followed by July and May, and a mean annual
precq:>itation of 13.25 inches (Climate data from
Medicine Lake, Western Regional Climate Center,
1911-1997).

Note: This area and two others in Medicine Lake are
part of the 11,366 acres designated as Medicine Lake
Wilderness Area. This report is a very preliminary
descrption of the refiige portion (about 2,300 acres of
the Wilderness Area) of the entire sandhills area that
in the aggregate covers about 21 square miles, the
largest dune complex in Montana.

"VEGETATION:

The sandplains and sand dunes harbor community
types that are possibly unique in Montana,
underdocumented, and pending classification review in
the northern Great Plains states and provinces. The
landsc:^e is primarily composed of graminoid-
dom in ated vegetation, especailly in blow-out areas,
though in swales and bottomlands shrub-dominated
vegetation types are common.

CalaTnozfilfa longifolia — Stipa coTnaia Herbaceous

Vegetation

[GALLON - STICOlvTI

Prairie sandreed (-) needle-and-thread grassland

Stabilized sites constitute the vast majority of the
landscape, particularly the more planar areas while the
blow-out patches are very scattered. The more
stabilised states have high canopy coverages of needle-
and-thread [Stipa comata) usually in excess of 50%.
There are far lesser amounts of the next most
prevalent grass, prairie sandreed (Cala-mD'-nlfa

loTigifoha), with lemon scurf-pea [Psoralea laTicsolata) as
the most common species among aweakly represented
forb component (total cover not exceeding 10%). It
was tentatively place in the Calamoi'ilfa longifoha —
Sapa comata Herijaceous Vegetation type (prairie
sandreed — needle-and-thread grassland). This plant
assoc iation may in turn be an early and long-
persisting serai stage of to Stipa comata —3outeloiia
gracihs — Care3:jilifoha Herbaceous Vegetation, but no
examples of the latterwere found in either the choppy
or the gently roiling terrain. Additional field sampling
may be required to adequately describe the plant
associations, their relation to successional processes,
and the site variables. |^Plots NHMTECRA97SC0004,
NHMTECRA97SC0005. NHMTECF^97SC0006^

SyTnphortcarpos occidentalis Shrub I and

CSYMOCC]
western snowberry shrubland

Shrubland dominated by western snowberry
(Syjnphoncarpos ocadentahs) is a recurrent community
type within the Sandhills, usually occurring in swales
but extending upslope in the area, though with
reduced stem density. In the physiognomic portion of
the classification it is referred to as a temporarily
flooded type which could hardly be the case for the
Sandhills sites; there have to be some unappreciated
circumstances that favor the establishment o?
Sympkoncarpos occidentalis and other shrubs on such
seemingly droughty sites. The undergrowth is
dominated by western wheatgrass (PascopyTH/m STmthzi)
and --j^ry few herbaceous spec ies (usually fewer than
five).

Elaeagnus coTmnutata /Stipa co-maia Shrubland

CELECOM / STICOM]

silverberry / needle-and-thread shrubland

This is community type has not been previously
named or described. A rare silverberry /western
wheatgrass shrubland (Elaeagnus coTmrtutata /
PascopyTumsTrathzi Shrubland; G2) has been named and
cited as occurring in MT, SK, ND, and IvlB. The
combination of ElaeagTuts cofmnutata and Stipa comata
is unique in that Elaeagnvs commutata is usually
associated with moist sites in the landscape and Stipa
comata with drier, sandy soils. Quite possibly the
deeper-rooting K coTmnutata is tapping a watertable
unavailable to the herbaceous component. This is
supported by the fact that common chokecherry
(Pmnus ZJzrgimaTta) and western snowberry
[Symphoncarpos occidentalis) occur as community
dominants adjacent to the Klaeagnus coTnmutata-
dominated community, ostensibly on the same site.

47

Both of these other species require moisture in levels
above that supplied through precipitation alone or by
compensating environments where evaporative losses
are mitigated.

The Pmmis •i^rgimaniZ-doTn.m.^y.^d community has an
undergrowth with Shpa coTnata dominant but with
Poscopyrum srmihzi prominent whereas the
Sympkoncarpos occideniahs-domin^lad communities
have an undergrowth with the abundances of these
undergrowth species switched [Shpa comata relegated
to merely present in most cases). It would be difficult
to envision a scenario wherein the Pascopynim srmihii
was grazed out of the ELECOM / STICOM p.a. and
not the adjacent P virginiana- and S occiderdahs-
dominated communiities. In keeping with the sandy
substrate, by far the dominant forb in ELYCOM /
STICOM was slimleaf scurfpea {Psoralea lanceolata);
other forbs were present in only trace amounts.
Wildlife browsing has been intensive on the Elaeagmis
with shrubs attaining only 3-3.5 fl. height in 1 1 to 12
years. |^Plot NHMTECRA57SC0003]

In general, Klaeagnus co-rmmdata is rarely regarded as a
shrubland dominant in the south of the i'S'^ Parallel
except possibly as a localized feature on limestone,
including montane settings, or on well-drained
Northern Great Plains grasslands in idle conditions
and with ample moisture.

Prunus virgtmana Shrubland

CPRUVIR]

common chokecherry shrubland

Common chokecherry (Prunus z/zrgiTuan^) dominates

'very small stands within the Sandhills; the stems had
been exceedingly hedged by wildlife browsing and the
leaves fed on by insects. The undergrowth has both
needle-and-thread (Shpa comata) and western
wheatgrass {PascopyruTnsTnithzi) as dominants, their
roles apparently shifting by site. This t3^e is found
primarily on the undulating flats but also occurs on
slopes of arrested dunes, sites seemingly too dry for a
species normally associated with mesic sites. |^No
Plot^

Oryfi^pszs hymenoidss / Psoralea lanceolata Sparse

Vegetation

CORYHYM / PSOLAiri

Indian ricegrass / lemon scurf-pea barrens

A recent blowout early in the process of stabilization
was sampled that represents an indian ricegrass /
lemon scurf-pea barrens (Oryzopsts hymenoides /

Psoralea lanceolata Sparse Vegetation). The
unconsolidated sand substrate constitutes over 90 % of
the cover at the surface, and the vegetation canopy
cover is less than b %. This is a fundamentally
different vegetation than the Centennial sandhills, but
these two sites have one rare plant species in common
{Cryptantha feTidUn) as well as analogous successional
processes (Lesica and Cooper 1999).

SPECIES:

Eour Montana pant secies of special concern have
been documented in the Medicine Lalte Wilderness
Area; two in the course of this study. Species site
information is presented on the Element Occurrence
Records in Appendix E, and annotated illustrations
are in Appendix E. Detailed information is lacking to
compare their numbers in the Refuge to elsewhere on
the Sandhills for providing concise statements of
botanical significance. Nonetheless, for its habitat
uniqueness and accrued botanical information, it
represents the highest known concentration of rare
plants in the Sandhills and in the county.

Eendler's cat's-eye [CryptanthafindUri) occurs in the
Sandhills on discrete zones of unstable sand, oflien the
leeward rim of active dune blowouts. It was found in
two of the most active blowouts on the Refuge,
perhaps a small segment of a much larger population
complex alluded to by Lesica in estimating total plant
numbers in excess of 10.000 across the entire
Medicine Lalte Sandhills.

Schweinitz' flatsedge (Cyperus sckzveimtzzi) also occurs
on unstable sand, often in the hollow at the head of an
active blowout. It was found at a single site on the
Refijge, presumable part of a much larger population
complex alluded to by Lesica in estimating total plant
numbers as "many thousands" across the entire
Medicine Lalte Sandhills.

Hairy four o'clock (JS/hrabihs Jursuta) is widely
scattered in low numbers across a range of sandy
habitats. It shows no discernible habitat specificity in
relation to composition or structure. It is present in
both the Medicine Lake Sandhills and Big Island. It
has since been documented outside of the Refuge in
disturbed settings including roadsides and CRP. It
exhibits the distribution pattern of an adventive
species, thus providing the basis for removing it from
the list of Montana plant species of special concern to
the watch list. Yet it is only Icnown from three
counties and seven collection records so that field data
will continue to be compiled on it for further
evaluation.

48

Plains phlox (Phlox andicola) is locally common on
gently rolling sandhill slopes under a sparse canopy of
Klaeagniis commutata (silverberry) I^Plot
HHMTECRA97SCO003;] where only the vestiges of
flo'werE remained at the time of visit. It was
occasional in the best condition grassland habitat of
Big Island, as found on a steep, north-facing slope. It
may be under-documented in the sandhills area
because it is inconspicuous except during its early
flowering, but there is not enough information to
confirm or refiite this idea at present.

OVERALL BIODIVERSITY SIGNIEICANCE:
The Medicine Lake Sandhills are part of the largest
sandhills in Montana, followed by the Centennial
Sandhills in southwestern Montana which overlaps
with Red Rock Lalces I'JWR. These landscapes, and
their dune system in particular, constitute highly
significant landscapes, harboring unique plant
communties and rare species.

This characterisation is at best a preliminary highlight
of the sandhills vegetation and rare plant species.
Wildlife values were not evaluated.

OTHER VALUES:

Archeological and cultural values are oflen associated

with sandhills.

LAUD USE;

The area has a long history of grazing by livestock.
Grazing leases were recently cancelled on the area.
Upland segments near roads have been planted into
crested wheat grass (Agropyron cnstabimy

MANAGEMENT COMMENTS:
V/idely-scattered patches of leafy spurge (Kupkorbza.
esidd) have been identified by Refuge personnel
(Rabenberg pers. commun.} There are several spurge
flea-beetle release sites in the Sandhills.

Prior to the settlement, this landscape was maintained

in various successional states by periodic disturbance,
mostly via buffalo, pocket gophers, fire and wind. The
processes operative here are probably analogous to
those within the Centennial Sandhills. Studies have
demonstrated for the Centennial ecosystem, that
without periodic disturbance and with the healing of
existing blow-out and deposition surfaces, the result is
reduced community and species diversity, particularly
of rare species associated with early serai states
(Lesica and Cooper 1999). Cattle have served as
surrogate disturbance agents in the Centennial system
and in the Medicine Lake Sandhills in some measure.

Weed problems could be exacerbated w ithout close
control of management and weed population
responses.

Portions of the sandhills native communities are
overwhelmed by dense populations of increaser

species, for example, sagewort (Artevusia caTnpestns)^
brittle prickly-pear (Opuntiajragilis)^ flixweed
tansymustard (Descuraima sophza.)^ jZTi^pinnate
tansymustard (D. pinnatd) that may have resulted from
past grazing practices. Leafy spurge (Kuphorhia es-ula)
is present in w idely-scattered patches and spurge flea-
beetles have been released in an effort to control it
(Rabenberg pers. commun.). Canada thistle (CirsiuTn
arvense) and other exotic species like smooth brome
(BroTniis iTtermzs), Japanese brome (3. japomcus)^ and
crested wheatgrass (Agropyon cnstahim) are present in
low numbers.

Re introduction of appropriately timed fire may
possibly be a management option to consider in
containing neaiijy weeds, reducing litter
accumulation, setting back woody species, and
stimulating forbs.

The undeveloped areas of natural spring and seep
features and associated riparian habitat that were
noted are in relatively good ecological condition and
are among the segments of the landscape warranting
closer investigation. Water developments below them
reduced grazing pressure in the hills above.

Tepee Hills Research Natural
Area

environment:

Tepee Hills is developed on a till/outwash plain that
has been down-cut by an abandoned meander channel
of the Missouri River. It straddles the crest of the
slopes above Medicine Lake, with mainly south-facing
slopes mapped as Zahill clay loam» steep (Richardson
and Hanson 1977) that have overlying gravelly,
cobbly water-worked deposits of Elaxville gravel
(Witkind 1 959). The center of the RNA is dissected by
a coulee. The RNA comprises only 50 acres but with
the variation in relief aspect and soil depth, a range of
environments are represented. The semi-arid
continental climate has peak precipitation in June
followed by July and May, and a mean annual
precipitation of 13.95 inches (Climate data from
Medicine Lake, Western Regional Climate Center,
1911-1997).

49

"VEGETATION:

The variation in aspect, soil depth as it affects water

holding capacity, and slope runofTproduce a number

of distinct environments in a relatively confined area
(see Figure 3. Tepee Hills RNA plant communities
and associations).

Stipa co-maia — Boutdoua gracilis — Carexjilifoha

Herbaceous Vegetation

[:STICOM - BOUGRA - CARFIL;]

needle-and-thread — blue grama — threadleaved sedge

grassland

This is one of the most extensive and broadly
distributed of upland plant associations within the
Northern Great Plains, occurring in Manitoba,
Saslcatchewan, Nebraska, Wyoming. North and South
Dalcota and in Montana is a major matrix t3^e from
the base of the Rocky Mountain Front eastward. In
the Medicine Lalce landscape it is restricted to the
most xeric exposures, moderate to steep, south- to
southwest- facing, mostly convex slopes having
shallow soils. ??Compositionally the RNA examples
of this type are not close to the modal description
"wherein Shpa comata is the dominant graminoid and
^outsloua gracilis has 100% constancy (but cover
values not exceeding 30 %). Quite possibly past
grazing, favoring short-statured rhizomatous species,
has influenced the composition of this site. I^Plot
NHMTECRA97SCOOO 1 J

Shpa cartiseta — Klyjnus laTtceolatus Herbaceous

Vegetation

CS TI CUR - EL YL AN]

porcupine needle-grass (-) thick-spike wheatgrass

grassland

This association is found only on moderate to steep,
north-facing slopes with well-developed soils, as
restricted to the coulee. It has been described from
similar settings in northern Valley and Phillips
Counties (DeVelice et al. 1995) and in northwestern
North Dalcota. However, w ithin the Canadian prairies
or prairie parldands it occurs on planar and roll ing
surfaces as an extensive^ prevailing type in mesic
settings. In the Tepee Hills representation of this
type, porcupine needle-grass {Stipa curtiseta) is mono-
dominant (in excess of 50 % canopy cover) and other
graminoids, including thickspike wheatgrass {Symus
lancsolahis) which are said to be dominant or co-
dominant in Canadian settings, comprise little more
than trace amounts. This suggests the need to further
evaluate if not reclassify this plant association. There
is a noteworthy and relatively luxuriant diversity of

native forbs, reflecting the mesic environment,
totaling 36 species in a single plot. They include

outlying Rocky Mountain plant species such as small-
flowered penstemon (Penstemon procenis) that are of
biogeographic interest. I^Plot
I'JHMTECRA97SC00023

Pascopyrumsmithzi — Boutdoua gracilis — Carexjilifblia
Herbaceous Vegetation
CPASSMI -BOUGRA- CARFIL;]
western wheatgrass —blue grama — threadleaved
sedge grassland
This is a broadly distributed t3^e Northern Great
Plains plant association, which we have distinguished
from ELYLAN-STICOM because it appeared to be
present lower in the landscape, on the flats and
toeslope positions, than was the ELYLAN - STICOM
community t3^e. This distinction may be somewhat
artificial but their respective distributions appeared
distinct at the time of sampling. This type generally
occupies heavier soils and more poorly drained sites
than does ELYLAN - STICOM. Western wheatgrass
(PascopynijnsTnithii) is the dominant graminoid with
cover usually in excess of ^0 %; the cover of blue
grama (Boutdoua gracilis) and threadleaved sedge
(Carsxjilifolia) is usually less than that o^ Pascopyrum
srmthii. Which species has greater cover seems to vary
randomly across the landscape. Within this RNA,
crested wheatgrass (Agropyron cnstatum) has
established significant coverages in this community
type, mostly by volunteer seeding from adjacent
agricultural lands. |^No Plot^

Pascopymm srmthii Herbaceous Vegetation

CPASSMI3

western wheatgrass grassland

This association represents, along with westerm

smpwberry shrubland. the most mesic sites within the
uplands of the refijge. It is a widely distributed across
the Northern Great Plains from Montana to Nebraska
and south to Colorado, It occupies, as a narrow band,
the heaviest alluvium soils of toeslopes and ephemeral
drainages; oflen this t3^e is assumed to be
sub irrigated and occasionally it can qualify as a
jurisdictional wetland (no examples of this on RNA),
In its native state, this type is characterized by
virtually

50

FigLre 3. ^

Vegetation of Tepee Hills RNA

Medicine Lake NWR

^t.;-;j Synphori carpus occidentalis

Gi^iland Commjnltl^
^ AEr9p>ron smrthii-Sripa Com&ci-Bouteloua jracilis {Agrop^ron crtetatuni)
IS Agr&pyiNin iinidtll (Agropyron crlstatum. Broihu* Irenuls, Caragana arhoresf eni)

0.1

ff.i

17 i Stipa coma^^Go lAcbUti I'mc ills -Care k fillfolhi

IS I StipacurcUsita-El^mjA Unce&latus

3^ Stipa. connata-Elymus lanceolatus jAgropyron crJataUim}
K) Agrofi^roii ^mltfiii-Scipa viridola

I riediciieUkeNWH

I Private Land

Water

Stole bMile

Atbers Bqiul AimPrEijBtldQQ

DfitLjniHAD27

At Ok givnj map a^ab Qijl all«nmjumLly«dijmiaflJxctj'pcs-<^QiLlff rfl^'vdyi^^^
uceaiUtJitscEaiikpJcJabcLs-BDiJ d^^fIU1iaDBatlIl?LuaLC^ Br^kn? cncloffCvn^pi'iO'L^pv?
ar4tfiLifi&tluto:in!!tJtLit>^En izjjk^ ca>v£^ up ti20^<yri&i:ilyg;^Q;ofpab'£cci2]Bl>clKiQstLsvli]e
Lw c^rmni^ilv Cyf4(l]i«r[$;]ij^-if^lheTfu:itdexL^^ivA ^d^^p^rncdfram-^ lEBvrtiyB colon.
Forvard Blaskcs i^Oofi^BsbBl-jKparHHttic Mostinicm:qKdBby¥ili]A4planic<fl>TTTiBiiiTiyoT
aisa:uitioD is lujnzd.

VFBrt3t]mniapumta-fc[mS":a imip(i:7S'2y;j ncrialpioiciEriyjhy, field v?ri-flGKl June 1M7

June 15, i99S

1515£35JSSTthAvc
ITGBFTia.MrS5620

monoEpecific dominant PascopyruTn STmthii (coverages
generally in excess of 60 %) and low forb diversity.
Within the RNA, almost all of these habitats have
been either seeded to, or invaded by, Agropyron
cnstatuTTi (crested wheatgrass) and BroTnvs inerrms
(smooth brome) though Pascopymm srmthzi has often
maintained dominance or co-dominance. |^Mo Plot^

Symphorictsrpos oa^dsntalis Shrubland

CSYMOCC]

western snowberry shrubland

This is a common Northern Great Plains type of
subirrigated settings or those receiving overland flow,
draws and swales, positions that on the RNA are
merely mesic uplands and not wetlands or riparian
habitat. The RNA examples of this type are small
inclusions in more extensive types and are in
relatively good condition with the density of western
snowberry [Symphoncarpos oa^dentahs) sufficient to
exclude most other species except for the rhisomatous
grasses like western wheatgrass [Pascopyrum smithit),
smooth brome (BroTnus inermis, and Kentucky
bluegrass (Poa pratsnsis). |^No Plot^

OVERALL BIODIVERSITY SIGNIEICANCE:
Tepee Hills has the most intact vegetation overall
among the Medicine Lake Refuge's established RNAs,
a slice of Great Plains landscape. Perhaps the most
significant ecological feature captured by this RNA is
the Stzpa curtiseta — Efymus lanceolatus plant association,
a relatively high quality example ofwhat is considered
as an important vegetation type of the Canadian
glaciated plains. There are no other protected
examples of this community documented within the
state.

OTHER VALUES:

Tepee Hills has archeological values, featuring a
historic Native American encampment, recognized on
the National Registry ofHistoric Places.

LAND USE:

The area was grazed prior to RNA establishment. It
has light non-motorized recreational use. It is
bordered by a crested wheat grass planting on the west
that may extend w ithin RNA boundaries.

MANAGEMENT COMMENTS:
Tepee Hills is a relatively small area, surrounded
mainly by agricultural lands and man-made features
that can be expected to provide a continuous threat
through weed introductions and simple fragmentation
of populations and habitat. It is situated between a

CRP planting ofcrested wheatgrass (Agropyron
cnstahijri) to the north and aweedy roadside right-of-
way to the south. A large area of smooth brome
(Brovus inerrms) has become established on the western
end of the ridgetop. Planted windbreak species within
the RNA include Siberian pea-shrub (CaragaTia
arhor^cens) and green ash (Fra-xiniis pennsyivamca). It
was burned in the spring of 1 994i (Rabenberg pers.
commun.)

54

Red Rock Lakes National
Wildlife Refuge

Sheep Mountain Research Natural
Area

environment:

Sheep Mountain RNA rq^resents an 86 acre segment
of a unique environment both ■within the USFWS
refiige system in Montana and the state at large. The
Centennial Range is Montana's only large mountain
range whose main axis is oriented east-west. As such,
it is in position to intercept cells of moist air that
originate in the Gulf of Mexico and drift north"'-"vard in
mid to late summer. These cells are the source of
afternoon thunders how ers that can be quite intense
and can cause mountain meadows to remain green
long into the growing season. Annual precipitation at
Lakeview (6,700 fl.)» in the Centennial Valley at the
^ery base of the mountains, is ^.1 inches, which is
quite high for a valley location (compare to Wisdom,
MT 1^6,100 ft elevation^ which receives 11. S in.
annually). Near the crest of the range annual
precipitation probably exceeds 50 inches. About 27%
of annual precipitation falls in May and June, which is
typical for western Montana's mountainous areas. Soil,
snow, winds, and snow slides also shape its
uniqueness, as recognised in the original
establishment record. The Sheep Mountain RNA,
ranging in elevation from 7,600 to 8,'K)0 ft., is but a
partial representation of a 3,000 vertical feet long
mountain gradient developed wholly on the calcareous
(predominantly Madison limestone) north flanlc;
quartzite is also reported to be present here according
to the original establishment record. The limestone-
derived soils are generally thin and have a low water
holding capacity. An avalanche chute is located along
theRNAs north edge.

VEGETATION:

The vegetation features of Sheep Mountain RNA are
consistent with Society of American Foresters (SAF)
cover type (c.t.) targets originally identified for the
site, including the Engelmann sprue e-subal pine fir c.t..

Interior Douglas Fir c.t., and limber pine c.t. They are
in noteworthy old-growth form. In addition, grassland

communities and the avalanche chute successional

features are present.

Four tree species are the climax dominants in the
forest series on Sheep Mountain RNA: Engelmann
spruce (Pzcea engelmanmi), subalpine fir (Ahies
lastocarpa), Douglas fir (Fseudotsuga Tnen^nssii) and
limber pine (Pinvs Jles^hs). In addition, a grassland
ridgeline opening is near the lower end toward the
east, and an avalanche chute is near the upper end
toward the west. The predominantly north-facing
slopes of the RNA support plant associations of forest
series even at the lowest elevations because of the high
precipitation. This is in contrast to other portions of
southwestern Montana, where at the elevations
represented on this RNA. grasslands are prevalent and
any forest series present would be only the Pseudotsuga
meni^iesii or Pinvs Jls3::ihs series. High elevation sites
that have thin soils, are on wind-exposed or ridgeline
positions, or have warmer exposures, regularly
support the Pseudotsuga 7nen^a,esii and Pinusjlexilis
series. Pseudotsuga men^iesii, Pirms jiexihs and even P.
engdmatimi tend to be favored over Ahies iasiocarpa and
lodgepole pine {Pirtus cotitoria) by calcareous
substrates. In fact, Pitius contorta was rarely seen in
the course of our RNA transect, which appeared to
traverse only limestone. Thus, these three species.
Pseudotsuga merisziesn, Picea engelmanmi, andPinus
Jlexdis, tend to have greater cover on these mesic
slopes than would be predicted from precipitation and
temperature alone. Where thin soils combinewith
exposed positions and warmer exposures, non-forested
environments are produced and usually dominated by
bunchgrasses like bluebunch wheatgrass
[PseudoToegnena spicata) and/or Idaho fescue [Festitca
idahoensts\

Ahies iasiocarpa / Thxihctrum occidentals Forest

CABILAS / THAOCC^

subalpine fir / western meadowrue forest

The vast majority of this RI'JA is considered to be in
various serai stages of this potential or climax plant
association. The subalpine fir is used to name the
association even though Douglas-fir is strongly
dominant in stands representing this type. This
naming convention is used because the national
vegetation classification, at least in the western United
States, has been based, up to now, on plant
associations named in the context of potential natural
vegetation or habitat types (Pfister and Amo 19S0).
The existing vegetation or serai associations that
occurwithin habitat t3^es (potential natural vegetion

55

based associations) are yet to be documented. The area
capable of supporting the subalpine fir / western
meadowrue forest (Abies lastocarpa / Thahctrwm
occidentaie Forest; ABILAS/THAOCC) ranges from
the steep, north-facing slopes at the lo\"vest elevations
of theRNAto the upper slopes (7,900 ft. plus) where
it extends to warmer slopes as increasing elevation
■with colder temperatures and increased precipitation
compensate for aspect with increased solar insolation
load. The ground surface is continuously carpeted
"■.vith litter, having virtually no stone or gravel
exposed. This plant community could be "typed" in
two ways, using Pfister et al. 1977 (which is specific to
Montana) and Steele et al. 1983 (which is specific to
eastern Idaho and western Wyoming); the
undergrowth and associated mesic environment better
correspond to the ABILAS / THA.OCC cl imax
association described in Steele et al. (19S3) as a minor
type in northwestern Wyoming.

Most of the stands are rather open (verging on
woodland at less than 70% canopy cover), not
exceeding 65 to 75 ft. in height, single-aged to two-
aged and dominated by Psendotsuga in the upper
canopy. At least two old-growth stands were
encountered, where Pssud.otsuga exceeding 20 inches
and 200 years were common. Though there are
occasional mature Abies specimens in the upper
canopy, P eTigelmtznmi is the more common
representative of mature to late serai tree species. At
the start of reconnaissance from the slope bottom, it
was especially notable that virtually all smnller Abies
projecting above the snowline had been browsed,
ostensibly by moose. Given that most of these Abies
stems were decidedly shorter than they should have
been given the thickness of their stems, it is inferred
that this snowline browsing has continued for years
and is the primary reason these stands will never
become Abies dominated. Serai Ptnusjlej::ilis is perhaps
the most abundant canopy tree after Psevdotsuga and
Ptcea.

Undergrowth cover, which varies inversely w ith the
degree of canopy shading, ranges from just barely
more than trace amounts to 50% plus and is
dominated by forbs; those with the greatest cover and
constancy include showy aster (Aster cons picuus),
western meadowrue (Tkahctrwm occidentale), mountain
sweet-cicely (Qsmorhi%a chzlensis\ heart-leaved arnica
(Armca cordifblia), northern valerian (Falenana dioica)
and slender cinquefoil (Potenhlla gracilis). Shrub cover
barely exceeds trace amounts; various Ribes species
(currant or gooseberry) and mountain snowberry
(Symphoncarpos oreophilus) are regularly present. The
grass component is also depauperate with nodding

bluegrass (Poa refle.xa) and pinegrass (Calaviagr^sUs
mbescens) usually the only species present and always
with low cover, usually not exceeding 5%. |^Plots
NHMTECRM9SSC0033, WHMTECRN9SSC003q

Abies lasiocarpa / Juniperus communis Woodland

CABILAS / JUNCOM]

subalpine fir / common juniper woodland

This is a vary common plant association, identified

from the drier mountain ranges of eastern Oregon and
Washington, eastward into Montana and Wyoming
and south as far as New Mexico and Arizona (see
explanation under ABILAS / THAOCC association as
to why these stands dominated by Douglas-fir
[Pse-udotsiiga m^n^zesiz) are named for subalpine fir
[Abies lasiocarpa) Common juniper (JuTuperus com.-mums)
is a relatively stress-tolerant shrub. Within the
context of this relatively mesic, generally north-facing
ftank of the Centennial Range it represents habitats
experiencing greater moisture stress than are reflected
by the presence of other forested associations
commonly encountered that instead have subalpine fir
(Abies lasiocarpa) in association with either western
meadowrue (Thalictmfn occidentale), pinegrass
[Calamtzgrostis n^esceTis) or with shiny-leaf spireaea
(Spiraea betuhfoha). On the RNA, ABILAS /
JUNCOM was encountered on warmer exposures,
those with a westerly component, and above 7,800 ft.,
though it is capable of occurring at much lower
elevations. It generally grades to ABILAS / mountain
gooseberry (Ribes rnontigenuTn), which is present on the
RNA as narrow patches where snowpacks are deeper
than on surrounding terrain.

Being a woodland, tree canopy cover is generally
below 60% and tree form approaches "stunted" with
heights barely exceeding -I^O feet at more than 350
years of age. Engelmann spruce (Picea eTigelmnfinii)
and Abies lasiocarpa are present mostly in the
sapling/seedling layer, though scattered mature and
old-growth Picea engelmanTui are typically present as
well. This association is at the dry extreme o^ Abies
lasiocarpa distribution and this species probably will
never achieve canopy dominance. The canopy
dominant over most of the stand is Pseud-otsuga,
though limber pine [Pinusjlej::ilis) is a major
component in patches. Common juniper (JumpeTits
coTtimunis) dominates the undergrowth, its cover
generally exceeding 10%. The graminoid element is
especially depauperate with only traces of nodding
bluegrass (Poa reflexa) and Ross sedge (Carex rossiil).
Showy aster (Aster conspicuus) is the forb with highest
cover in the plot and it and lanceleaved stonecrop

56

(Sedufn lanceolahim) were noted as the prevalent forbs
throughout the drier "woodland environments. I^Plot
NHMTECRN98 S C0037^

Piniisjlss^hs / Pseiidoroegnsna spicGta

Woodland[;PINKLE / PSESPQ

limber pine / bluebunch wheat grass woodland

This plant association was found on a very rocky,
thin-soil, limestone ridge with a northwest- and "■-vest-
facing aspect and stretched up and downslope
approximately 120 vertical feet from the 8,160 fl.
contour. The ground surface is more than S5%
exposed gravel and rock with bare soil constituting
another 5-10%. There is no soil profile development
and of the upper 10-20 cm. more than 50% is rock
(gravel size or larger); this site verges on being a scree
slope.

In this old growth stand of stunted limber pine {Pirms
Jlexiiis) (?^^Douglas fir (Pssudotsuga 7nen^a,esii\
maximum height of 300 plus yr. old trees 22-24 ft.) all
the veteran trees have very battered crowns and boles
emblazoned with numerous lightning scars, often
having more bare bole and scar tissue than
functioning bark. Tree canopy cover ranges between
30 and 60%, composed of only the above named
species; there are scattered seedling and samplings but
the mid-sized age classes are missing. Shrubs like
mountain snowberry (Sympkoncarpos orsophilvs) and
common juniper (JuTuperus coTmnums') occur in only
trace amounts. The herbaceous layer is very sparse
(total cover <10 - 12%) and dominance shifts across
the stand, some portions (or patches) being dominated
by bluebunch wheatgrass (PssitdorosgTisna spicata)
others by grayish cymopteris (CyTnopUrus glauc^is),
lanceleaved stonecrop [Ssdumlancsolatus), weedy
milkvetch {Astragalus Tmser) and even many-flowered
phlox (Phlox Tnultiflora). Overall Pseureognena spicata
appears to consistently have the highest coverage,
though this is generally less than 5%.

This is among the oldest PINKLE / PSESPI stands
that have been inventoried in south"'-Yestern Montana;
it is so old and lightning struck that none of the trees
cored yielded a core that was countable beyond several
hundred years, extrapolation yields ages in excess of
500 years. Stand structure is rather typical of xeric-
site, old-growth with very scattered reproduction and
few intermediate-aged trees. This stand represents
the moisture stressed extreme of a type that is known
as one of the driest of the woodland vegetation types
in Montana with exception of Jumpems spp.-
dom in ated woodlands. I^Plot
WHMTECRN9S S C00363

Pseudoroegnsna spicata — Poa secunda Herbaceous
Vegetation

CPSESPI-POASEC]

bluebunch wheatgrass — Sandberg's bluegrass

grassland

This association is found as small patches on the very
driest of spur-ridges that project to the north from the
main east-west trending ridgeline of the Centennials.
The combination of thin, rocky, limestone-derived
soils, the western exposures of the spur-ridges (having
the highest solar insolation values in a landscape with
primarily northern exposures) and the prevailing
southwesterly winds which scour snow from the
windward slopes (west) and crests causes these ridges
to be the most moisture-limited of any features in this
landscape. Wind deflation causes more than 80%
exposed limestone gravels; the depressed interstices
are occupied by soil. Litter and basal area together
comprise less than 5% of the surface area.

These ridges are so dry as to be incapbale of
supporting much biomass, the total cover approaches
the 10 % cutoff of sparsely vegetated communities.

Though we have classed the plot as belonging to the
bluebunch wheatgrass — Sandberg's bluegrass
association {Pseudoroegnena spicata —Poa secunda
Herbaceous Vegetation), its position and composition,
both in alpha diversity and in the number of cushion
plant spec ies prominent, place it closer to the P spicata
/ "Cushion Plant" community described by Cooper et
al. (1995) for southwestern Montana (not yet
recognised in TNC's Western Region Classification).
The shrub component is almost nonexistent; Woods
rose {Rosa zvoodsii) and green rabbitbrush
(ChtysoihammiTns viscidiflorus) are so thinly scattered
and depauperate as to be obscured by the herbaceous
layer. Only two graminoids were present in the plot;
the dominant P spicata and a trace of Idaho fescue
(Fstuca idakoensts) which is virtually ubiquitous in
moutain grasslands of southwestern Montana. Low-
growing, cushion-like plants or those more typically
found in exposed subalpine to alpine environments
(e.g. grayish cymopterus ^X/^Tnopterus glaucus^, lance-
leaved stonecrop {Sedum laTiceolatum), Rocky
Mountain douglasia {Douglasia montane), Cut-leaved
daisy (Engeron compositus), Parry's townsendia
(To'!vnsendia panyi) and sheep cinquefoil (Potenhlla
ovina) are conspicuously reperesented, though
individually not exceeding 5% cover.

OVERALL BIODIVERSITY SIGNIEICANCE;
The Sheep Mountain RNA as currently defined is a
small, intact sample of predominantly old-growth
forest in the Rocky Mountains biome. It is in the

57

middle of a much larger and more diverse ecosystem,
the whole north face of the Centennial Range, where
the escarpment begins in shrub land/ grassland at
approximately 6,700 ft. and sweeping upwai'd
uninterrupted to the highest alpine sites at 9,600 feet.
The RNA is encompassed by the Red Rocks Lalces
Wilderness, affording additional protection to it and a
much larger area. However, neither the RNA nor the
surrounding wilderness area that is centered in the
valleybottom capture the scale of the processes that
operate in this landscape or more than a fraction of the
habitat diversity existing in it.

Avalanches constitute one of the more salient of these
processes; very steep terrain at the highest elevations
causes avalanche chutes that fringe parts or all of two
RNA borders. These chutes represent a feature of
geological process as well as natural succession, and
only two plots were taken in the array of wet-to-dry
habitats in these avalanche features.

gradient ostensibly spans several jurisdictions (Red
Rock Lalces National Wildl ife Refuge, BLM
Centennial Mountains Wilderness Study Area, USFE
Targhee National Forest. Agricultural Research
Service's Sheep Experiment Station).

The RNA directly adjoins the only known extant

occurrence ofWhipple's Beardtongue (Penstsmon
zvhzppleanus) in Montana, a species that was
documented incidental to the baseline sensitive species
work in the Centennial Valley for the Bureau of Land
Management (Culver 1993). The avalanche chutes
and talus slopes are also potential habitat for dwarf
goldenweed {JrLaplopGppus nanus), known only in
Montana from a historical collection on the slopes
south of Red Rock Lakes. There was not adequate
time for completing a systematic survey of these two
species across the RNA.

It was suggested in the original establishment record
that the unique, exposed climatic conditions found
within this RNA affords an excellent opportunity for
studies in forest ecology and plant physiology. The
same records ascribed wildlife values to this RNA, but
they were not evaluated in this study.

LAND USE:

The site is essentially pristine; no timber cutting or
evidence of domestic stock use was found It receives
limited recreational use from hunters and hikers.

MANAGEMENT COMMENTS:
This site would ideally be core of a much larger RNA
spanning the fiill range of environments and processes
of the Centennial Mountains. It lies within Red Rock
Lalces Wilderness Area, which spans the lower
segment of the 3,000 ft Sheep Mountain slope. The
RNA could be expanded lengthwise within wilderness
area boundaries, but any expansion of the vertical

58

DISCUSSION

The 1 5 RNAs and two additional study sites
encompass a highly significant array of natural
landscapes for Montana and the upper Missouri
River watershed. Over 50 different plant
associations are present, representing about 10% of
all the recognised terrestrial plant association t3^es
in Montana. About a third of these examples are
truly noteworthy in their quality and condition to be
considered good or outstanding representatives of
the biodiversity embodied in Montana's natural
vegetation (represented by shaded cells in Table 3,

pgii)

These noteworthy plant associations are interpreted
as representing significant biome features, the first

of the RNA criteria in the Refuge Manual. RNAs
may also represent stability in ecological
communities, succession in ecological communities,
habitat for threatened, endangered or sensitive
species, and geological processes. Twelve RNA sites
met one or more of these RNA establishment criteria
in our partial assessment, as highlighted in Table 4i
(below). We refer to this as a partial assessment
because it did not address wildlife features.

Most of the five RNAs that do not contain
exemplary ecological or botanical features were
originally nominated based on wildlife values, and
this study simply provides background habitat

description.

Table 4. Partial matrix of National Wildlife Refuge RNA criteria and sites in Montana

SITE

BIOME

CLIMAX ■ SUCCESSION

TES

PLANT

SPECIES

GEO.

PROCESS

Mullan Trail RNA

Yes

Fourth Ridge RNA

Yes

Hell Creek area

Yes Yes

Limber Pine RNA

Yes

Manning Corral Prairie Dog
Town RNA

Yes

Missouri River Bottomlands
RNA

Yes

Yes

Yes

Spring Creek RNA

Yes

Two Calf - Dou^las-fjr RNA

Yes

Yes

York Island

Yes

Sheep Mountain RNA

Yes

Yes

Yes

Yes

Medicine Lake sandhills area

Yes

Yes

Yes

Yes

Tepee Hills RNA

Yes

57

Among the significant examples of plant
associations, as determined by their outstanding

quality and condition, several are considered
potentially rare or vulnerable across their entire
distribution. The Douglas fir / littleseed ricegrass

forest (Pseudotsuga menziesii/Oryzopsis micrantha
p. a.) is a well-developed plant association found only
in central Montana that is rare on account of its
geographical restriction, even if it is not under
widespread threat. Three other plant associations
may possibly be globally rare and are in varying
stages of status evaluation in cooperation '.vith
adjoining states and provinces. They include: Rocky
Mountain juniper/ Wyoming big sagebrush
woodland (Juniperus scoulorum / Artemisia
tridentata ssp. wyomingensis Woodland), Indian
ricegrass / lemon scurf-pea sparse vegetation
(Oryzopsis hymenoides / Psoralea lanceolatap.a) as
found in sand dunes, and the porcupine needlegrass -
thickspilce wheatgrass grassland (Stipa curtiseta /
Elymus lanceolatus Herbaceous Vegetation). We
believe that most or all of the other high global
ranks (Gl-GS) for plant associations on Table 2 are
artifacts of gaps in research or literature review.

of Stipa comata — Bouteloua gracilis — Carex filifolia,
but only Spring Creek had more than 10 acres in
good condition and surrounded by more-or-less
intact upland landscape approaching good
representation of the grassland system and
processes. Even the two RNAs established to
represent prairie dogs towns were studies in
contrast: an exotic species had talten over in Prairie
Dog Island RNA whereas native species associated
with early serai conditions prevail across the prairie
dog town site of Manning Corral Prairie Dog Town
RNA

Other recurrent patterns appear in collectively
considering these 12 sites. They include some of the
few protected public lands in eastern Montana with
intact mesic, productive plant associations. Such
inventory features include the once-widespread
Pascopyrum smithii-Nasella viridula p.a. ofMullan
Trail RNA as mentioned above. The other
associations of high biomass are highly localized
features like the Stipa curtiseta — Elymus lanceolatus
p.a. of Tepee Hills RNA, restricted to north-facing
slopes.

There was relatively little overlap between plant
association features at different sites. This may
reflect the system of selecting sites or the inherent
diversity within the NWR system. Even in cases of
overlap, the ''redundant" plant association features
diflered in their ecological context. For example, two
significant stands of Artemisia tridentata ssp.
wyomingensis / Pascopyrum smithii were identified:
on York Island, and on Fourth Ridge just to the
west. Yet they differed in that the York Island
shrubland is pervasive across the uplands and
represents a typical landscape, while the Fourth
Ridge shrubland is part of a juniper woodland
mosaic in an extreme example in an unusually harsh
setting. Two sites of Pascopyrum smithii -Nasella
viridula p.a were documented at Mullan Trail RI'JA
and in the Hell Creek area, the former in a
glaciolacustrine setting and the latter in an
unglaciated setting where the community is a post-
fire serai stage. Many sites had vestiges or patches

The largest RNA, the Missouri River Bottomlands

RNA, is in a class by itself, encompassing riverine
processes and succession, and containing relatively
large Missouri River islands, relatively large stands
of plains cottonwood, and erodible valley slopes. The
presence of intact landscape processes, as well as the
plant association components, are enhanced by
representation of active geological processes, which
enhance system sustainability. Geological processes
are also captured in the sandhills segment of the
Medicine Lake Wilderness Area, the largest
sandhills in Montana with its aeolian processes and
succession. The Medicine Lake sandhills also have
the highest numbers of Montana plant species of
special concern among study sites because of the
uncommon sand dune habitat. The third notable
RNA example of geological process are the
avalanche chutes of the Sheep Mountain RNA,
although the RNA includes only small portions of
two chutes.

58

CONCLUSIONS AND
RECOMMENDATIONS

In addition to background and habitat information for

each site, this report provides a baseline for assessing
the diversity of ecological features and processes
represented in Montana's U.S. Fish and Wildlife
Service RNAs. Together "■."vith information on U.S.
Forest Service RNAs and BLM "Areas of Critical
Environmental Concern" (ACECs), this assessment
can be used to help systematically identify protected
or unprotected habitats and landscapes in Montana
and the region.

While boundai'y review per se was not the focus of
this project, information we collected suggests some
possible changes that would better fulfill
establishment or representation objectives for h-vo of
the RNAs studied. These comments focus on a
landscape perspective, including gradients and
processes, which are important to the long-term
viability of communities and species within the sites.
Some of the RNAs already encompass broad gradients.
The Spring Creek RNA encompasses a well-developed
ravine system with its full complement of habitats.
The Limber Pine RNA encompasses a typical
Missouri Breaks cross-section with a complementary
suite of plant associations. The Missouri River
Bottomland in combination with the Two Calf-
Douglas-fir RI'JA similarly encompasses a cross-
section of Missouri Breaks landscq:>e, though the
difference between the vegetation on the former with
its sandstone and siltstone bedrock is a striking
contrast with the vegetation of the latter on Bearpaw
Shale and bentonite.

The value of the Missouri River Bottomlands RNA
(representing the valley slope gradient) is enhanced by
the adjacent Two Calf-Douglas-fir RNA. However,
the boundary may be inadequate to effectively
represent the latter forest type and accompanying
upland processes, and boundary review for the latter is
recommended-

The Sheep Mountain RNA area is not large enough to
represent viable stands and landscape processes, but is
surrounded by designated wilderness on the Refijge.
The Refuge extends to midslope positions in the

Centennial Range so any recognition of intact
landscape gradients would involve collaboration with
other agenc ies. Sheep Mountain is also in a
geographic class by itself among Fish & Wildlife
Service RNAs as a Rocky Mountain site rather than a
Great Plains site, with intact old-growth plant
associations that are otherwise incompletely
represented in the Forest Service RNA system in
Montana. We recommend that the Service consider
expanding the RNA lengthwise on Refuge lands and
explore elevational expansion of the RNA to
encompass the unbroken ecological gradient that
extends into higher elevations onto BLM and USES
lands.

Though a "gap analysis" and exploration of alternative
or additional sites was beyond the scope of this
project, some observations emerged from our studies.
Most important is that despite the array of plant
associations within this USFWS RNA system, it does
not include large areas of once-extensive plant
associations that covered the Great Plains. However,
some RNAs we studied occur within larger areas
where these important systems are represented in
good condition. The Charles M. Russell NWR offers
outstanding and unique opportunities to identify and
sustain large, intact plains landscape features not
found elsewhere on publ ic lands in Montana. Further
field assessment is recommended beyond the RMA
boundaries to document the locations and condition of
key communities and landscape complexes to provide
information that can assist w ith management and
conservation of key ecological features and areas on
the Refiige.

On as smaller scale, the Manning Corral Prairie Dog

Town site could include representative south-facing
breaklands habitat in addition to prairie dog town
succession. We also noted that few. well-developed
plant associations or wetland settings with intact
hydrological gradients were found, and these
represent a gap in types represented within existing
RNAs.

In conclusion, we recommend a "next phase" of effort
focussed on identifying areas that would fill gaps and

59

achieve representation at scales more consistent with
ecological processes and the historic nature of once-
widespread types. Much of this effort should be
focussed on the Charles M. Russell NWR and
surrounding public lands, where there maybe
outstanding representation of large scale landscape
systems and conservation opportunities potentially
unique in Montana and the region. Future work
should also include assessment ofwildlife
representation and values, emphasizing rare, declining
and keystone species.

60

LITERATURE CITED

Allen, L., S. Cooper, D. Faber-Langendoen and G.
Jones. 1999. P. Comer (ed.) Selected shrubland
and grassland communities of the northern
Great Plains. A report to the Nebraska National
Forest by The Nature Conservancy, Boulder,
CO. 12^ pp.

Booth, W.E, 1950. Flora of Montana, part 1,

Conifers and Monocots. Research Foundation at
Montana State College, Boseman. 232 pp.

Bourgeron, P. S. and L. D. Engelking (eds.). 1991?.
A preliminary vegetation classification of the
Western Unitied States. Unpublished report
prepared by the Western Heritage Task Force
for The Nature Conservancy, Boulder, CO. (not
paginated)

Branson. F. A., R F. Miller and I. S. McQueen.

1970. Plant communities and associated soil and
water factors on shale-derived soils in
northeastern Montana. Ecology 5l:391-'K)7.

Cooper, S. V., P. Lesica, R. L. DeVelice and J. T.

McGarvey. 1995. Classification of southwestern

Montana plant communities "with emphasis on
those of Dillon Resource Area, Bureau ofLand
Management. Montana Natural Heritage

Program. Helena, MT. 152 pp.

Culver, D. 1993. Sensitive plant inventory in the
Centennial Valley, Beaverhead County.
Unpublished report to the Bureau ofLand
Management. Montana Natural Heritage
Progrm, Helena. ^2 pp. + app.

DeVelice, R L., S. V. Cooper, J. T. McGarvey, J.
Lichthardt, and P. S. Bourgeron, 1995. Plant
communities of northeastern Montana: A first
approximation. Montana Natural Heritage
Program, Helena.. 113 pp.

Dittberner. P.L. and M. R Olson. 19S3. The Plant
Information Neh-vord DataBase: Colorado,
Motana, North Dalcota and Wyoming. U. S.
Fish and Wildlife Service FWS/OBS-S3/36. 7S6
pp.

Dorn, R D. 19S4. Vascular Plants of Montana.
Mountain Press Publishing. Cheyenne, V/Y.

Federal Committee on Ecological Reserves. 1 977. A
Directory of Research Natural Areas on Federal
Lands of the United States of America. Published
through the U.S. Forest Ser'vice, Washington,
D.C.

Federal Geographic Data Committee — Vegetation
Subcommittee. 1997. National Vegetation
Classification Standard, June 1997. U S.
Geological Survey, Reston, VA. 60 pp.

Franklin, J. F. 1992. Scientific basis for new

perspectives in forests and streams. In: Naiman,
R J. ed. Watershed Management; Balancing
sustainability and environmental change. New
York. Springer-Verl ag. pp. 25-72.

Genter, D. 1986. The Montana Natural Heritage
Program. Pp. 3'h-35, In- Proceedings of the 1986
Montana Natural Areas Conference. 1 59 pp.

Great Plains Flora Association. 1977. Atlas of the
Flora of the Great Plains. University of Iowa
Press, Ames. 600 pp.

Great Plains Flora Association. 1986. Flora of the
Great Plains. University Press of Kansas,
Lawrence, KS. 1^02 pp.

Hansen, P. L. and G. R Hoffman. 1987. The
vegetation of the Grand River/ Cedar River,
Sioux, and Ashland Districts of the Custer
National Forest: A habitat t3^e classification.
USDA Forest Service, Rocky Mountain Forest
and Range Experiment Station. General
Technical Report RM-157. Rocky Mountain
Forest and Range Experiment Station, Fort
Collins, CO. 68 pp.

Hansen. P. L., R D. Pfister. K. Boggs. B. J. Cook, J.
Joy, and D. K. Hinckley. 1995. Classification
and management of Montana's riparian and
wetland sites. Miscellaneous Publication No. 5-4?,
Montana Forest and Conservation Experiment
Station, School of Forestry, University of
Montana, Missoula, MT. 6^6 pp.

61

Hanson, J. R. 19S4i. Bison ecology in the northern
plains and a reconstruction of bison patterns for
the North Daltota region. Plains Anthropologist

29(10^):93-113.

Hanson, J. R. 198-11. Bison ecology in the nooorthern
plains and a reconstruction of bison patterns for
the North Dalcota region. Plains Anthropologist
29: 95-115.

Heidel, B. L. 1996. Woodhawk. botanical survey,
Fergus County. Unpublished report to the

Bureau of Land Management. Montana Natural
Heritage Program, Helena. ^5 pp + app.

Heidel, B. L. 1997. Montana plant species of special
concern. Unpublished list. Montana Naturla
Heritage Program, Helena. 5^ pp.

Higgins. K. F. 1986. Interpretation and compendium
of historical fire accounts in the Northern Great
Plains. U.S. Fish & Wildlife Service, Resource
Publ. 161. 39 pp.

Hitchcock, C. L. and A. Cronquist. 1973. Flora of the
Paciic Northwest. University ofWashington
Press, Seattle, WA 730 pp,

Jensen, M. E. and P. S. Bourgeron. 1993.
Ecosystem management; Princples and
applications. U. S. Department of Agriculture,
Forest Service, Northern Region. Missoula, MT.
597 pp.

Kartesz, J. A. 199^. A synonymised checklist of the
vascular flora of the United States, Canada, and
Greenland. 2 volumes. Biota of North America
Program, Timber Press. Portland, OR 806 pp.

Knowles, C. J. and P. R Knowles. 1995. A

bibliography of literature and papers pertaining
to presettlement wildlife and habitat of Montana
and adj acent areas. A report to USDA Forest
Service — Region 1, Missoula.

Kuchler, A. W. 196^. Potential natural vegetation of

the conterminous United States — manual to
accompany the map. American Geogrq:>hical

Society. New York, NY.

Lesica, P. 1987. The vegetation of acid shale pine
forests, Petroleum and Fergus Counties,
Montana. The Nature Conservancy Field Office,
Helena, MT. Unpaginated.

Lesica, P. and S. V. Cooper. 1999. Succession and
disturbance in sandhills vegetation: constructing
models for managing biological diversity.
Conservation Biology 15(2): 293-302.

Lesica, P. andJ. S. Shelly. 1991. Sensitive,

threatened and endangered vascular plants of
Montana. Montana Natural Heritage Program
Occ. Publ. No. 1. 88 pp.

Mack, R. N. 19S1. Invasion ofSroTnus tecforumh.
into western North America; An ecological
chronicle. Agro-ecosystems 7: 757-775.

Maclde, R J. 1965. Range ecology and

relationships of mule deer, elk. and cattle in the
Missouri breaks. Montana. Unpublished
dissertation, Montana State University,
Bozeman. 229 pp.

Martin, S. 1996. The U.S. Fish and Wildlife Service
ecosystem approach for the Upper
Missouri/Yellowstone River and Columbia River
Basin Ecosystems. I ntermoun tain Journal of
Sciences 2(1):45.

Meko. D. M. 19S2. Drought history in the western
Great Plains from tree rings, pp; 521-526. In:
International Symposium on Hydro meteorology.

Meko, D. M. 1992. Dendroclimatic evidence from
the Great Plains of the United States, pp. 312-
550. In: Climate Since AD. 1500., RS. Bradley
and P.D. Jones, eds. Routledge, London.

Mueller-Dombois, D. and H Ellenberg. 197^. Aims
and methods of vegetation ecology. New Yorlt.
John Wiley and Sons. 5^7 pp.

62

Noss, R. 1983. A regional landscape approach to
maintain diversity. Bioscience. 33: 700-706.

Noss, R. 1987. Protecting natural areas in

fragmented landscapes. Natural Areas Journal.

7: 2-13.

Peden, D. G., G. M. Van Dyne, R W. Rice and R M.
Hansen. 1 97^. The trophic ecology ofSzson bzson
L. on shortgrass plains. J. Applied Ecology
ll{2):'*89-^97.

Pfister, R D., B. L.Kovalchilc, S. F. Arno. andR C.
Presby. 1977. Forest habitat types ofMontana.
USDA Forest Service, Intermountain Forest and
Range Experiment Station. General Technical
Report INT-3^. Intermountain Forest and
Range Experiment Station. Ogden. UT. 17^ pp.

Richardson. R. E. and L. T. Hanson. 1977. Soil
Survey of Sheridan County, Montana. Soil
Conservation Service (NRCS)» Boseman, MT. 60
pp. + maps

Roberts, D. W. and J. I. Sibbernsen. 1979. Forest
and woodland habitat types of north central
Montana. Vol. 2: The Missouri River Brealcs.
Unpublished report to the Bureau of Land
Management, Billings. Intermountain Forest and
Range Experiment Station, Missoula, MT. 2^ pp.

Ryan, M. G., L. A. Joyce, T. Andrews, and K. Jones.
199^. Research Natural Areas in Colorado,
Nebraska. North Dakota, South Dalcota, and
parts of Wyoming. USDA Forest Service. Rocky
Mountain Forest and Range Experiment Station.
General Technical Report RM-251. , Rocky
Mountain Forest and Range Experiment Station,
Fort Collins. CO. 57 pp.

Salwasser, H. 1992. From new perspectives to

ecosystem management: response to Frissel et al.

and Lawrence and Murphy. Conservation
Biology. 6; ^69-^72.

Schneider, R. E.. D. Faber-Langendoen, R. C

Crawford, and A. S.Weakley. 1997. The Status
ofBiodiversity in the Great Plains: Great Plains
Vegetation Classification. Supplemental
Document 1. In W. R. Ostlie, R. E. Schneider, J.
M. Aldrich, T. M. Faust, R L. B. McKim, and S.
J. Chaplin, The Status ofBiodiversity in the
Great Plains. The Nature Conservancy,
Arlington, VA. 75 pp + X

Stohlgren, T. J., L. D. Schell. and B. VHeuvel. 1999.
How grazing and soil quality affect native and
exotic plant diversity in Rocky Mountain
grasslands. Ecol. Appl. 9(l):45-6^.

Umbanhowar, C 1996. Recent fire history of the
northern Great Plains. Amer. Midi. Nat.
135(l):il6-121.

Vanderhorst, J. , S. Y. Cooper, and B. L. Heidel.

199S. Botanical and vegetation survey of Carter

County, Montana. Unpublished report to Bureau
of Land Management. Montana Natural
Heritage Program, Helena. 116 pp. + app.

Witkind, L J. 1969. Quaternary geology of the
Smoke Creek-Medicine Lake-Grenora Area,
Montana and North Dakota. U.S. Geolo. Survey
Bull. 1073. pp. 1-SO.

Young, J. A. and F. L. Allen. 1997. Cheatgrass and
range science: 1930-1950. J. Range Manage.
60:630-636.

63

Appendix A
Community Survey Form

COMMUNITY SURVEY FORM (MTNHP) GENERAL PLOT DATA:

A. IDENTIFICATION AMD LOCATION:

MANUAL: UNITS: ft, m

PLOT NO. :MON. DAY: YEAR: EOCODE:_

EXAMINER(S);

POTENTIAL. NAT. COMM.: C.T:

BAILEY CLASS.: SECTION SUBSECTION LTA

POLYGONNO.: POLYGON NAME:

SITE NAME: ^ STATE: COUNTY:

USGS QUAD NAME: ^ QUAD CODE:

EXTENT C.T./P.AW/IN LANDSCAPE: Matrix, Lg. Patch, Sm. Patch (circle) COMMUNITY. SIZE (acres):

GPS REF. NO.: Field UTM X_ _TnE Field UTM V___ mN

Corrected UTM X mE Corrected Field UTM V mN UTM Zone

Public Land Survey:T, NorS ; R, Eor W ; Sec, _; 1/4S ; 4/4 ; 4/4/4 ;, ^!^fiiA ;

LATITUDE: (degO; _ _(min.):_ _._ _(sec.): LONGITUDE: (deg.);_ _(min.);_^ _■_ Jsec);

OWNERSHIP (circle): Private {Name: ). U. S- ForesI Service, BLM, Tribal, Bur. of Rec , State MT, ^__„

PLOT TYPES: PLOT SIZE: RADIUS/LN; WIDTH SURVEY:

PHOTOGRAPHY: {type, azimuth, etc.)

DIRECTIONS (to plot): _.

ENVIRONMENTAL FEATURES:

DL: SOILRPT:

SOIL UNIT: SOIL TAXON:

SOIL TEXTURE ;(i:ircic one) clayi sandy day: Silly clay; cloy Joflmi dliy clay loam; sandy day Joam; loam; sill Eoam; ?ili; sandy loam; loamy sand; sand; x= unable (0 assess

PARENT MATERIAL(S): LANDFORM:

PLOT POSITION: SLOPE SHAPE: (vert,)^ ; (hor.) ASPECT(c): SLOPE (%):

ELEVATION: (ft. ot M) EROSION POT.; EROS. TYPE(S):

HORIZON ANGLE: N ; E ; S __; W IFSLP: ^^__^^_ IFVAL:

SPECIAL FEATURE(S):

GROUND COVER (by cover cla^): son.-!- GRAVEL+ ^ROCK + LITTER. + WOOD + MOSS + BASAL VEG -^ OTHER =100^

(baresoil = <2mmfracliC)n;gfav^ = 2nnnn to <lOcm,rocK|1m:, cobbles, b[HiWeis] = ^ i an; olher = water, iHjhea spadfy

DISTURBANCE HISTORY (include estimation of weed populations here; type, mtensity, frequency, season):

RIPARL4NAVETLAND FEATURES:

COWARDIN CLASS.; SYST. Palmlnne, Utmirine, RivErinc (drdo) SUBSYST, CLASS.

SUBCLASS. DOMINANCE TYPE

HGM CLASS.: _^

VALLEY FLOOR GRADIENT: FLOODPLAIN WIDTH: (m, ft,) BEDMATEIUAL:

CHANNEL WIDTH: CHANNEL DEPTH: CHANNEL ENTRENCH.;

SURFACE (STANDING) WATER DEPTH: (cm or in. observed): MEAN MAXIMUM

DIST. FROM WATER: AVE. ANN. HIGH WATER: ^__^_ (observed or esumate. circle)

PONDING EVIDENCE: (a Acrldl pholo, Sbaibdrd vc^. C njcks w/ Wo tftlmiiflle omr, D aedimenl depoartiorj L idcJq Wand v^oliclion. Rliaib wrack lines, SwaleiVBiilslairas)

DURATION oflNUNDATION: (days, ihis year)

INUNDATION PERIOD/HYDRO. REGIME: ,-Mc-..jPann-iicnUyFlG(nled; Siwurai«li Semipennanentiv Flooded, StflEonillyFloodcJ' Temporarily Md ; InienniitHilly Fid

SEDIMENT DEPOSITION: ^COVER f/-), DEFFH (cm or in.)

BANK STABILITY: Rills, Gully Cutting, Headcuts, Slumps, Undercut Bank:

CAPILLARY FRINGE: ^DEHTHTO CAP. FRINGE; Tl-IICKNESS CAP. FRINGE; DEPTH to SATURATION ff«c*ii«>

ORGANIC HORIZON THICKNESS (cm or in): MEAN MIN. MAX.; Oa Oe ^Oi

OCULAR PLANT SPECIES DATA:

PLOT NUMBER; . NO. SPECIES: PNC:.

MINIMUM COVER VALUE:

TREES: T6TAL CV.
TALL CV.
LOW CV,

m£AH HT-
MED. CV.
GRND-CV,

FOftSS: ^<:^TAL CV.
MED, CV.
GRND. CV.

MEAN KT,
LOW CV.

Tree Height | Canopy Cover by Dia, Class
SPECIES IDENT/ >18" <1B" O" <5" 1''

T 1
T 2

T 3
T 4

T 7

T e

;

/

/

_/_

/

/

/

/

/

/

/

/

/

/

/

/

/

/

i

/

/

/

!

/

/

/

!

_/_

1

/

/

i

/

SPECIES IDENTIFICATION
F 1

SHRUBS: TOTAL CV.

TALL CV.
LOW CV.

MEAN HT-
MED CV.
GRND.CV.

1
2
3
4
5

S

S

S

S

S

3

S

S 8

S 9

SIO

311

312

7

/
/
/

7"

/

/

/

/

"/"
7"
7'

GRAMINOIDS: TOT. CV.
MED. CV.
GRND. CV.

MEAN HT.
LOW CV.

F

F

F

F

F

F

F 8

F 9

FIO

Fll

F12

F13

F14

F15

F16

F17

F18

F19

F20

F21

F22

F23

F24

F25

F26

F27

F28

HT. CCC

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ :

/ [ ]

/ [ ]

/ [ ]

/ [ 1

/ [ ]

/ [ ]

/ [ ]

/ [ )

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

/ [ ]

G 1

/

F29
F30
F31
F32
F33
F34
F35
F36
FERNS

tot;

LOW
F 1
F 2
F 3
F 4
F 5 _

G 2

/

/

G 3

G 4

/

G 5

/

G 6

/

/
/

G 7

G 8

G 9

/

AND ALLIED FORMS (S.(
\L CV. MEAN HT.
CV. GRND CV.

J EQUISETUM, 1

GIO

/

MED. C

Gil

/
/

G12

G13

/
/
/

G14

G15

G16

/

BRYOIDS £ LICHENS: TOTAL CV.

Broids :
/
/
/
/
/
/
/
/

TOTAL
L 1
L 2
L 3

L 4
L 5
L 6
L 7
L 8

CV. Lichens:

B 1
B 2
B 3
B 4
B 5
B 6
B 7
B 3

]
]
]
]
]

■

/
/
/
/
/
/
/
/

/ [ ]

/ [ ]

/ [ ]

I

r

[

[.

[ ]

[ ]

[ ]

[ ]
[ ]
[ ]
[ ]

[ ]

[ ]

[ ]

" Tree canopy cover for nrnture (> 5 in. dbh) and Beed lings /saplings {< 5 in. dbh.J is Che minimum breetidown for tree stratum, lor any species;

' Canopy Couer CUssea (Percent Values) ; 0] T - >0, <.\: P • 51, <%: ! = >5,<1S; 1 - zlS, <25; 3 - ?25, <35; 4 = i3S, <15; 5 - 545, <55; 6 = £%^, <65j

7 = £S5, <1^i e - i^S, <&5; 9 - ^95, <&5; F - j95

'Fiifit three letters uf genus ami apeciesp' wjrite tomplete tpeciss naiiie if confusion poeslbie within lifeforrn; uan S to innJicBte collected toxou

Appendix B
Plant Species of Special Concern Survey Form

Plant Species of Special Concern Survey Form

MONTANA NATURAL HERITAGE PROGRAM
P.O. Box 201800, 1515 E. Sixth Avenue, Helena, MT 59620-1800

SCIENTinC NAME;
OBSERVER(S):

WORK LOCATION / ADDRESS:

SicatiOnrmttachseOBy pfpertment'7.¥-or35

COUNTY:

TOWNSHIP:

RANGE:

ADDITIONAL T/R/S, SECTIONS OR V* SEC:

DATE OF SURVEY:

USGS QUAD: .

SEC.(s): __

NATL. FOREST DISTRICT/BLM DISTRICT RA/OTHER: ,

DIRECTIONS TO SITE (Refer to towns, roads, trails, other geographic features): .

'A SEC.:.

TOTAL NUMBER OF INDIVIDUALS (estimated or exact population count; sum clumps or stems if vegetative}:

NUMBER OF SUBPOPULATIONS AND SUBTOTALS (if applicable):.

SIZE OF AREA COVERED BY POPULATION (acres):

PHENOLOGY (% flowering, fruiting, vegetative):

EVIDENCE OF DISEASE, PREDATION, INJURY;

EVIDENCE OF SEED DISPERSAL, ESTABLISHMENT: ,

POPULATION TREND/OTHER COMMENTS:

paHitat: iB^s™be.thejli3fcgtiis^H^atores»o£itht;^^^^^^^ _ fj

^edes'Jiabitatin the^etting):P_^ '*^ " ' " ''^'^''^'

ELEVATION (mean or range): ' ASPECT: QN QNE QE QSE QS QSW QW QNW

% SLOPE: SLOPE SHAPE Q Concave D Convex O Straight Q Other

TOPOGRAPHY: Q Crest □ Upper Q Mid Q Lower Q Bottom QOther .

MOISTURE: D Dry D M^'S^ D Saturated □inundated □ Seasonal seepage □ Other ^

UGHT EXPOSURE: □ Open □ Shaded □ Partial shade □ Other _

PARENT MATERLVL: ' _

SURFACE COVER (TOTAL %): MOSS/LICHEN BASAL VEG. BARE GROUND

SOIL TEXTURE/SERIES: . ^

CANOPY COVER: TREE (%) SHRUB {%) FORB (%) GRAMINOID(%).

PLANT COMMUNITY; (dominant species at present, age and structure notes): ^ . —

CLIMAX VEGETATION (if not above): .

ADDITIONAL ASSOCIATED PLANTS (include most common, conspicuous, and characteristic spp.):_

EVIDENCE OF DISTURBANCE: .

PHOTOGRAPH TAKEN ? (if so, mdicate photographer and depository) :
SPECIMEN TAKEN? (if so, list collector, collection #, and repository): _

IDENTIFICATION (name of person making determination, and/or flora used):.

ECODATA PLOT NUMBER (attach photocopied data sheets}: .
OTHER DOCUMENTATION OR REFERENCES:

Appendix C
Photographs of State-significant Vegetation Features

Glaciated example of the once-widespread western wheatgrass — green needlegrass prairie [Pascopyrum smithii — Nasetta viridula Herbaceous Vegeta-
tion) at Mullan Trail RNA- Benton Lake NWR

Good example and unusual mosaic pattern formed by Rocky Mountain juniper woodland {Juniperus scopulorum/ Artemisia tridentata ssp. wyomingensis

Woodland] a! Fourth Ridge RNA -Charles M. Russell NWR

Unglaciated example of the once-widespread western wheatgrass - green needlegrass prairie {Pascopyrum smithii - Nasetia viridula Herbaceous
Vegetation) at Heii Creels potentiai RNA - Charies M. Russell NWR

Good representation of Missouri Breaks gradient at Limber Pine RNA - Chartes M. Russell NWR, foreground includes prairie sandreed - sun sedge prairie
[Caiamoviiia longifolia - Carex inops Herbaceous Vegetation] and in the background Juniperus scopulorum and Pinus ponderosa dominate northern
exposures and ravines

Natural succession with blue grama prairie [Bouteloua gracilis Herbaceous Vegetation) as dominant across former prairie dog town at Mannhg Corral
Prairie Dog Town RNA - Charles M. Russell NWR; light-colored grass present In extensive patches Is the native tumblegrass {Schedonnardus panicutatus)

Good example of intact island communitieB on Two Ca[f Island tn Missouri River Bottomlands RNA — Charles W. Russefl NWR; Including silver sagebrush
shrubland {Artemisia cana/ Pascopyrum smithii Herbaceous Vegetation), plains cottonwood woodland [Populus deltoides/ Symphoricarpos occidentalis
woodlands) and willow {Salix spp.) thickets

Intact stand ol plains cottonwood/ western snowberry woodland {Populus deltoides/ Symphoricarpos occidentatis Woodland) at Hess Bottoms of Missouri
River Bottomlands RNA - Charles Nfl. Russell NWR; mature plains cottonwood mostly 16-20 inches in diameter

Representa!Jve black greasewood / western wheatgrass shrubland [Sarcobatus vermicutatus I Pascopymm smithii Shrubland) at Missouri River Bottom-
lands RNA - Ciiarles M. Russell NWR; black greasewood is also a major component of the vegetation on steep valley slopes to the right

Head of ravine woodland with aspen bordering green ash/ chokecherry woodiand {Fraxinus pennsylvanica I Prunus virginiana Woodland] at Spring Creek
RNA - Charies M. RusseH NWR; sandstone slopes to the right support a bunchgrass mosaic \n which skunl^brush sumac {Rhus aromatica) and yucca
{Yucca glauca) are consistent components.

Fenceline contrast with excellent condition prairie (left) of needle-and-thread — blue grama [Slipa comata - Bouteloua gracilis Herbaceous Vegetation] at
Spring Creek RNA — Charles M. Russell NWR; poor range condition landscape (right] is dominated by 6. graciiis and fringed sage (Artemisia fftgida).

Good example of the widespread Wyoming big sagebrush/ western wheatgrass shrubland {Artemisia tridentata ssp. wyomingensis I Pascopyrum smithii
Shrubland) at York Island RNA— Charles M. Russell NWR; In the background just below horizon are seen patches of yellow sweetclover

Good example of the uncommon Douglas fir/ litlle-seed ricegrass forest [Pseudoisuga menziesii/ Oryzopsts micrantha Forest), at Two Calf - Douglas fir
RNA - Charles M. Russell NWR; undergrowth is dominated by O. micrantha and bright green lichens

Good example of the indJan ricegrass - slimleaf scurpea barrens {Oryzopsis hymenoides - Psoralea ianceolata Sparse Vegetation) serai community
surrounding an active dune blowout area at Medicine Lal^e Sandshllis - Medicine Lake NWR

Good example of uncommon porcupine grass — thick-spike wheatgrass prairie [Stipa curtiseta - Elymus lanceolatus Herbaceous Vegetation] at Tepee Hjlls

l=^NA - Medicine Lal<e NWR; a diverse community type found on nortfi-facing slopes

Lower end of Sheep Mountain RNA — Red Rock Lakes NWR^ with an open ridge of bluebunch wheatgrass - Sandberg's bluegrass prairie [Pseudoreogneria
sptcata - Poa secunda Herbaceous Vegetation); these wind-impacted ridgeiine baids constitute tlie driest environments within the RNA

Old-growth limber pine/ bluebunch wheatgrass Woodland {PInus flexilis / Pseudoroegneria spicata Woodland) at Sheep Mountain RNA - Red Rock Lakes
NWR; numerous dead limbs and soars radiating from the crown downward testify to multiple lightnhg strikes on these veteran trees

Old-growth subalpine fir / common juniper woodland {Abies lasiocarpa / Thaiictrum occidenfate Woodland) with Douglas fir {Pseudoisuga menziesif)
contributing the great majority of tree cover, at Sheep Mountain RNA— Red Rocks Lakes NWR; heavy browsmg by moose prevents A. lasciocarpa from
growing out of the seedling/sapling size class

Appendix D
Vegetation Constancy-cover sampling data

Not AvailableOnline- SeeMTNHP for Details

Appendix E

Element Occurrence Records for
Montana Plant Speqes of Special Concern

USFWS RNA RECORDS - MONTANA PLANT SPECIES OF SPECIAL CONCERN

Scientific Name: CYPERUS SCHWEINITZII
Common Name: SCHWEIWITZ' FLATSEDGE

Global rank: G5 Forest Service status:
State rank: SI Federal Status:

Element occurrence code: PMCYP06360. 001
Element occurrence type :

Survey site name: MEDICINE LAKE SANDHILLS

EO rank:
EO rank comments:

County: SHERIDAN

USGS quadrangle: SUNNYHILL SCHOOL

CAPENEYS LAKE

Township: Range: Section: TRS comments:
031N 057E 24 13, 14, 20, 23, 26
031N 058E 7 13; 19

Precision: S

Survey date: - Elevation: 2100 -

First observation: 194 3 Slope/aspect :

Last observation: 1997-07-02 Size (acres) :

Location:

MEDICINE LAKE SANDHILLS, SOUTHEAST OF MEDICINE LAKE, CA. 25 AIR MILES
NORTH OF CULBERTSON,

Element occurrence data :

RHIZOMATOUS, MANY THOUSANDS OF STEMS IN SANDHILLS AREA.

General site description;

OPEN SAND HILLS; MOST CONSISTENTLY FOUND IN BLOWOUTS; WITH ORYZOPSIS
HYMENOIDES, PSORALEA LANCEOLATA, STIPA COIyiATA, SPOROBOLUS CRYPTANDRUS,
CRYPTANTHA FENDLERI .

Land owner /manager :

MEDICINE LAKE WILDERNESS

PRIVATELY OWNED LAND (INDIVIDUAL OR CORPORATE)

STATE LAND - UNDESIGNATED

Comment s :

Information source: LESICA, PETER, REPORT TO THE NATURE CONSERVANCY. UNDATED.
Specimens: ROTCHKISS, N. (6869). 1943, MONT.

USFW5 RNA RECORDS - MONTANA PLANT SPECIES OF SPECIAL CONCERN

Scientific Name: PHLOX ANDICOLA
Conunon Name: PLAINS PHLOX

Global rank; G4 Forest Service status:
State rank: S2 Federal Status:

Element occurrence code: PDPLMODOBO. 006
Element occurrence type :

Survey site name: MEDICINE LAKE SANDHILLS

EO rank:
EO rank comments:

County: SHERIDAN

USGS quadrangle; CAPENEYS LAKE

Township: Range: Section: TRS comments:
031N 057E 19 SW4SW4

Precision: S

Survey date: Elevation: 2000 -

First observation: 1997-07-02 Slope/aspect: 2-20%/NORTH

Last observation: 1997-07-03 Size (acres):

Location:

CA. 15 MILES SOOTH AND EAST OF MEDICINE LAKE (TOWN)-

Element occurrence data:

LOCALLY COMMON IN ROLLING SANDPLAIN PLOT; OCCASIONAL IN ISLAND KNOLL
PLOT. IN FRUIT IN JULY.

General site description:

VARIOUS WELL-DRAINED, EXPOSED SETTINGS.

Land owner/manager:

MEDICINE LAKE WILDERNESS

MEDICINE LAKE NATIONAL WILDLIFE REFUGE

Coinments :

PRESUMED TO CORRESPOND WITH THE MATERIAL COLLECTED AND IDENTIFIED AS
PHLOX ACULEATA SOUTH OF THE REFUGE IN SANDHILLS, DOCUMENTED IN ECODATA
PLOTS.

Information source: COOPER, S. V, AND B, L. HEIDEL. 1999. BIODIVERSITY

INVENTORY AND REPRESENTATIVENESS ASSESSMENT OF
RESEARCH NATURAL AREAS ON NATIONAL WILDLIFE
REFUGES IN MONTANA- UNPUBLISHED REPORT TO U.S.
FISH AND WILDLIFE SERVICE. MONTANA NATURAL
HERITAGE PROGRAM, HELENA.

Specimens :

USFWS RNA RECORDS - MONTANA PLANT SPECIES OF SPECIAL CONCERN

Scientific Name: PHACELIA THERMALIS
Common Name: HOT SPRING PBACELIA

Global rank: G3G4 Forest Service status :
State rank: 31 Federal Status:

Element occurrence code: PDHYD0C4L0 . 002
Element occurrence type:

Survey site name;

EO rank:

EO rank comments:

County: GARFIELD
USGS quadrangle :
Township : Range :
025K 041E

Precision:
Survey date;
observation:
observation;

First
Last
Location: YORK ISLAND, CHARLES H,

YORK ISLAND

BC

PROTECTED EO OF LIMITED SIZE AND UNNATURAL

HABITAT.

YORK ISLAND
Section: TRS comments:
08 SW4; SW4NW4
S

1998-07-17
1978
1998-07-17

Elevation: 2250
Slope/aspect ;
Size (acres) ; 1
RUSSELL NWR.

Element occurrence data:

CA. 50 PLANTS IN 3 SUBPOPULATIONS; THEY ARE OUTLIERS OF 1-2
INDIVIDUALS EXCEPT FOR THE SUBPOPULATION IW THE EASTERNMOST BACKWATER.
IN FRUIT AWD LATE FLOWER 17 JULY 1998.

General site description:

BEACHES OF REWORKED SHALE, AKD EPHEMEEALLY POWDED BACKWATERS SET OFF
FROM FORT PECK RESERVOIR BY WRACK LINE OF SHALE FORMED BY WAVE ACTION.
HIGHEST SPECIES NUMBERS ARE AT THE EDGE OF AN OPEN, EVAPORATED
BACKWATER POOL SURROUNDED BY A ROBUST WEEDY COMMUNITY OF CHENOPODIUM,
HELIANTHUS ANNUUS, AND LEPIDIUH SATIVUM.

Land cwner/manager: CHARLES M. RUSSELL NATIONAL WILDLIFE REFUGE
YORK ISLAND RESEARCH NATURAL AREA

Comments:

ALMOST ALL SUITABLE HABITAT WAS SURVEYED IN 1998. THE RESERVOIR
REACHED MAXIMUM POOL CAPACITY IN 1997. IN THE SPRING OF 1998 WATER
LEVELS WERE LOW BUT HAVE RISEN 5 FEET WITH JUNE RAINS, THESE CHANGES
AFFECT SPECIES BIOLOGY [DISPERSAL, ESTABLISHMENT) AS WELL AS HABITAT
AVAILABILITY. OBSERVED BY B. HEIDEL, S. COOPER, AND C. GUENTHER IN
1998.

Information source:

COOPER, S. V. AND B. L. HEIDEL. 1999. BIODIVERSITY
INVENTORY AND REPRESENTATIVENESS ASSESSMENT OF
RESEARCH NATURAL AREAS ON NATIONAL WILDLIFE
REFUGES IN MONTANA. UNPUBLISHED REPORT TO U.S.
FISH AND WILDLIFE SERVICE. MONTANA NATURAL
HERITAGE PROGRAM

Specimens: LACKSCHEWITZ, K. H. (8248), 1978. SPECIMEN #819^33. MONTU.

USFWS RNA RECORDS - MONTANA PLANT SPECIES OF SPECIAL CONCERN

Scientific Name: CRYPTANTHA FENDLERI
Common Name: FENDLER CAT'S-EYE

Global rank: G4 Forest Service status;
State rank: SI Federal Status:

Element occurrence code: PDBOROAOXO . 005
Element occurrence type:

Survey site name: MEDICINE LAKE SANDHILLS

EO rank:
EO rank comments:

County: SHERIDAN

USGS quadrangle: CAPENEYS LAKE

Township: Range: Section: TRS comments:
031N 057E 20 SW4SW4

Precision: S

Survey date: Elevation: 2065 -

First observation: 1997-07-02 Slope/aspect: 20%

Last observation: 1997-07-02 Size (acres) :

Location:

MEDICINE LAKE SANDHILLS. CA. 2 AIR MILES NORTHWEST OF BEVERLY SCHOOL.

Element occurrence data:

COMMON IN SMALL AREAS OF TWO OF THE MOST ACTIVE BLOWOUTS, A HIGHLY
LOCALIZED HABITAT.

General site description:

UNSTABLE RIM OF OPEN SAND BLOWOUT HABITAT, WITH PSORALEA LANCEOLATA
AND ORYZOPSIS HYMENOIDES.

Land owner /manager :

MEDICINE LAKE WILDERNESS

MEDICINE LAKE NATIONAL WILDLIFE REFUGE

PRIVATELY OWNED LAND (INDIVIDUAL OR CORPORATE)

Comments:

THERE IS POTENTIAL HABITAT BETWEEN ELEMENT OCCURENCES 005 AND 003, AND
THEY ARE LIKELY TO BE PART OF A COMPLEX POPULATION,

Information source: COOPER, S. V. AND B. L. HEIDEL. 1999. BIODIVERSITY

INVENTORY AND REPRESENTATIVENESS ASSESSMENT OF
RESEARCH NATURAL AREAS ON NATIONAL WILDLIFE
REFUGES IN MONTANA. UNPUBLISHED REPORT TO U.S.
FISH AND WILDLIFE SERVICE. MONTANA NATURAL
HERITAGE PROGRAM, HELENA.

Specimens;

Appendix F

Illustrations of

Montana Plant Species

OF Special Concern

JJIuslration by Jeanne It Janish, From
^VascuJar Plants of the Pacific Northwiat'

CRYPTANTHA FENDLERI
FENDLER CAT'S-EYE

Fcndler Cat's-e>e is an annual with simple or branched stems that are 5-20 cm high. The alternate, narrow, strap shaped
leaves may be up to 5 cm long; those at the base are usiially brown by the time the plant is fruidng. Foliage is sparsely
covered with spreading hairs. Tin>; white flowers are bome on coiled stalks Ihat unwind and elongate as flowering
progresses from the base upward. The corolla is ca. 1 mm high and has a small, united poilion below and 5 spreading
petals above. The calyx is covered with stiff, straight hairs and becomes 4-6 mm long in fruil. Within each fiuiti ng cal jx are
4smooth, shiny, narrowly lancc-shaped nutlets that are ca. l,5mmiongandl/3 as wide; 1 or more of these may be
missing. Flowering in May-early July.

Annual species of CRYPTANTHA are distinguished by characters of the seeds. C. FENDLERI is distinguished by having
4 smooth, shiny nutlets that are lance-shaped and 1/3 as wide as they are long. A hand lens or microscope are needed for
positive identiflcation.

Reprinted wiih pt^mission ^om the
New Britton and Brown niuslrated
FlOTa of the NoitheasTein United
States and Adjacent Caoada,
Vol. i, page 253, Copyright 1952,
The New York Btrtaiiieal Garden.

CYPERUS SCHWEINITZII
SCHWEIMTZ' FLATSEDGE

Schweifiitz' Flatsedge is a grass-like perciinia} witli stems that are 10-40 cm high, arising Ax)ni short, irregularh' swollen
rhizomes. The leaves are 1-4 mm wide and located mostly near Jhebaseoftheslem. The inflorescence is subtended b>' 3-6
long Jeaf-like bracts, some of which are wider than the leaves. The innoresence is made up of ascending dusters of flatten
spikelets that are 5-25 mm long and bome on stalks that are veiy short to long. Tlie flowers are crowded opposite each other
and consist only of a small, pointed scale, tliat is ca. 3-4 mm long and subtends 3 stamens and an ovary. The seed is triangular
in cross-section. Fruit mature in late June-July.

This is our only peremiial CYPERUS and is the only one occurring in upland habitat

llluslration by Jeanne R. Janish,

From 'Vascular PlanK of the Pacific Northwest'

PHACELIA THERM ALIS
HOT SPRING PHACELIA

Hot Spring Phacelia is an annual that is branched from the base, with prostrate or ascending stems. The alternate leaves
have broadly lance-sliaped blades that are 1-9 cm long widi toothed and deeply lobed margins and well-developed
petioles- Foliage isglandular-hairy. The short-stalked flowers are bome in crowded, narrow, l-sided, curved spikes that
are up to 10 cm long. The spikes unwind as they mature and originate in the leaf axils. The lavendar to whitish flowers
each have a 5 lobed tubular corolla that is 3-4 mm long and 5 narrowly lancc-shapcd, hairy sepals that are as long as the
corolla in flower but twice as long in ihiit. The stamens are included in the corolla tube. The fruit is a capsule with 2-A
seeds covered by a honeycomb pattern. Flowering in June.

PHACELIA rVESIANA differs from P. THERMALIS in that it has strap shaped sepals and is not as densely glandular-
hairy, P. LUTEAhas yellow flowers and only shallowly lobed leaves.

Illusiraiion by Debbie McHiel

PHLOX ANDICOLA
PLAINS PHLOX

Plains Phlox is a perennial with loosely tufted stems that are 4-10 cm high arising from creeping rhizomes. The 5-8 pairs
ofopposite, linear leaves have prominent midveins and whitish bases and are 10-25 mm long, ca. 1 nun wide and come to
a sharp point. Foliage is glabrous to sparsely hairy. Stems are while. 1-5 white flowers are home at the stem tips. Each
flower has 5 petals and a tubular corolla. The calyx is also tubular, with 5 lobes, tangled long hairs, and 6- 1 1 mm length.
Flowering in May-eaiiy June.

Distinguished from PHLOX HOODll by leaf length over 10 mm long, and from R ALYSSlFOLlAby leaf width less than 2
mm wide. Flowers aje needed tor positive identification, and hybridization between these species is reported elsewhere in
the range. \

Appendix G

Vascular Plants Cited in This Report,
BY Common Names, Scientific Names, and Six-letter Acronyms

Appendix G,

Trees

Ccxle

ABILAS

FRAPEN

JUNSCO

PICENG

PINFLE

Scientific Name

Abies lasiocarpa
Fraxinus pennsyivanica
Junlperus scopuiorum
Picea engelmannii
Pinus flexilis

Common Name

Subalpine Fir

Green Ash

Rocky Mountain Juniper

Engelmann Spruce

Limber Pine

Code Scientific Name

PINPON Pinus ponderosa

POPDEL Popuius deitoides

POPTRE Popuius tremuloides

PSEMEN Pseudotsuga menziesH

Common Name

Ponderosa Pine
Great Plains Cottonwood
Quaking Aspen
Douglas-fir

Shrubs

Code Scientific Name

/\j-A EALM Ameianchier ainifoiia

ARTCAN Artemisia carta ssp. cana

ARTFRI Artemisia frigida

ARTTSW Artemisia tridentata ssp.

wyomingensis

ATRCON Atripiex confertifoiia

ATRGAR Atripiex gardneri

BERREP Berberis repens

CERLAN Ceratoides ianata

CHRNAU Chrysoihamnus nauseosus

CHRViS Chrysothamnus viscidifiorus

CLELIG Ciematis iigusticifoiia

CLEOCC Ciematis coiumbiana

CORSTO Cornus stoionifera

ELACOM Eiaeagnus commutata

GUTSAR Gutierrezia sarothrae

JUNCOM Juniperus communis

JUNHOR Juniperus horizontaiis

LEPPUN Leptodactyion pungens

LONUTA Lonicera utahensis

PRUVIR Prunus virginiana

Common Name

Western Serviceberry
Silver Sagebrush
Fringed Sagewort
Wyoming Big Sagebrush

Shadscale
Gardner's Sallsage
Creeping Oregongrape
Winterfat

Common Rabbitbrush
Green Rabbitbrush
Western Virgins-bower
Columbia Clematis
Red-osier Dogwood
Silverberry
Broom Snakeweed
Common Juniper
Creeping Juniper
Prickly Phlox
Utah Honeysuckle
Common Chokecherry

Code Scientific Name

RHUARO Rhus aromatica

RIBAUR Ribes aureum

RIBCER Ribes cereum

RIBLAC Ribes iacustre

RIBMON Ribes montigenum

RIBSET Ribes setosum

RIBVIS Riijes viscosissimum

ROSARK Rosa arkansana

ROSWOO Rosa woodsii

RUBPAR Rubus parvifiorus

SALEXI Saiix exigua

SALLUT Saiix iutea

SARVER Sarcobatus vermicuiatus

SHECAN Shepherdia canadensis

SPIBET Spiraea betuiifoiia

SYMALB Symphoricarpos aibus

SYMOCC Symphoricarpos occidentaiis

SYMORE Symphoricarpos oreophiius

TOXRYD Toxicodendron rydbergii

YUCGLA Yucca giauca

Common Name

Fragrant Sumac
Golden Currant
Squaw Currant
Swamp Currant
Mountain Gooseberry
Missouri Gooseberry
Sticky Currant
Arkansas Rose
Woods Rose
Thimbleberry
Sandbar Willow
Watson Willow
Black Greasewood
Canada Buffaloberry
Shiny-leaf Spiraea
Common Snowberry
Western Snowberry
Mountain Snowberry
Poison Ivy
Yucca

IRAMINOIDS

CcxJe Scientific Name

AGRCRI Agropyron cristatum

AGRINT Agropyron intermedium

AGROPY Agropyron spp.

AGRREP Agropyron repBns

AGRSMI Agropyron smithii

(= Pascopyrum smithii)
AGRSPI Agropyron spicatum

(= Pseudoroegneria spicata)
AGRTRA Agropyron trachycaulum

(= Agropyron caninum)
ANDHAL Andropogon haliii
ARILON Aristida iongiseta
ARIPUR Aristida purpurea

(= Aristida tongiseta)
BOUGRA Bouteioua gracilis

BROINE Bromus inermis

B RO JA P Bromus japonicus

BROTEC Bromus tectorum

GALLON Calamovilfa iongifoiia

CALMON Calamagrostis montanensis

CALRUB Calamagrostis rubescens

CAREX Carexspp.

CARFIL Carex filifolia

CARHEL Carex heliophila

(= Carex pensytvanica)

CARING Carex inops

(= Carex pensytvanica)

CARPEN Carex pensyivanica

GARROS Carex rostrata

(= Carex utriculata)
GARSTE Carex stenophylla
GYPSGH Cyperus schweinitzii

DISSTR Distichlis stricta

(= Distichlis spicata)
ELYCAN Elymus canadensis

Common Name

Code

Scientific Name

Grested Wheatgrass

ELYTRA

Elymus trachycaulus

Intermediate Wheatgrass

(= Agropyron caninum)

Wheatgrass
Ouackgrass
Western Wheatgrass

FESIDA

Festuca idahoensis

FESOCT

Festuca octo flora

HORJUB

Hordeum jubatum

KOEMAC

Koeierla macrantha

Bluebunch Wheatgrass

(= Koleria cristata)

MUHGUS

Muhienbergia cuspidata

Bearded Wheatgrass

ORYASP

Oryzopsis asperifolia

ORYHYM

Oryzopsis hymenoides

Sand Bluestem

ORYMIG

Oryzopsis micrantha

Red Threeawn

PASSMI

Pascopyrum smithii

Red Threeawn

POACOM

Poa compressa

Blue Grama

POAGUS

Poa cusickii

Smooth Brome

POAINT

(= Poa fendleriana)
Poa interior

Japanese Brome

(= Poa glauca)

Cheatgrass

POAJUN

Poa junci folia

Prairie Sand reed

(= Poa second a)

Plains Reedgrass

POAPRA

Poa pratensis

Pinegrass
Sedge

POAREF

Poa reflexa

POASAN

Poa sandbergii

(= Poa secunda)

Thread-leaved Sedge

POASEG

Poa secunda

Sun Sedge

PSESPI

Pseudoroegneria spicata

Long-stolon Sedge

PUGNUT

Puccinetlia nuttalliana

SGHPAN

Schedonnardus paniculatus

Long-stolon Sedge

SGHSGO

Schizachyrium scoparium

Beaked Sedge

(= Andropogon scoparius)

SITHYS

Sitanion hystrix

Narrow- leaved Sedge

SPOGRY

Sporobolus cryptandrus

Schweinitz's Flalsedge

STICOM

Stipa comata

Alkali Saltgrass

STIGUR

Stipa curtiseta

STIVIR

Stipa viriduta

Ganada Wildrye

(= Nasella viriduta)

Common Name

Bearded Wheatgrass

Idaho Fescue
Six-weeks Fescue
Foxtail Barley
Prairie Junegrass

Plains Muhly
Roughleaf Ricegrass
Indian Ricegrass
Little-seed Ricegrass
Western Wheatgrass
Ganada Bluegrass
Gusick's Bluegrass

Inland Bluegrass

Alkali Bluegrass

Kentucky Bluegrass
Nodding Bluegrass
Sandberg's Bluegrass

Sandberg's Bluegrass
Bluebunch Wheatgrass
Nutt all's Alkaligrass
Tumblegrass
Little Bluestem

Bottlebrush Squirreltail
Sand Dropseed
Needle-and-thread
Porcupine Needlegrass
Green Needlegrass

FORBS

Code

Scientific Name

ACHMIL

Adiiiiea millefofium

AGOGLA

Agoseris glauca

ALLCER

Allium cernuum

ALLIUM

Allium spp.

ALLTEX

Allium textile

ALYDES

Alyssum desertorum

ANDSEP

Andfosace septentrionalis

ANEMON

Anemone spp.

ANEPAT

Anemone patens

ANTCOR

Antennaria corymbosa

ANTENN

Antennaria spp.

ANTMIC

Antennaria microphylla

ANTPAR

Antennaria parvi folia

APOCYN

Apocynum spp.

APOSIB

Apocynum siblrlcum

AQUILE

Aqullegia spp.

ARABIS

Arabis spp.

A RAG LA

Arabis glabra

ARAHIR

Arabis hirsuta

ARAHOL

Arabis hotboellil

ARNCOR

Arnica cordi folia

ARTCAM

Artemisia campestrls

ARTDRA

Artemisia dracunculus

ARTLON

Artemisia iongi folia

ARTLUD

Artemisia iudoviciana

ASCSPE

Asciepias speciosa

ASCVER

Ascieplas verticliiata

ASCVIR

Asciepias viridi flora

ASTADS

Astragalus adsurgens

ASTAGR

Astragalus agrestis

ASTBIS

Astragalus bisulcatus

ASTCER

Astragalus coram leu s

ASTCON

Aster conspicuus

ASTDRU

Astragalus drummondil

ASTER

Aster spp.

Common Name

Code

Common Yarrow

ASTFAL

Pale Agoseris

ASTFLE

Nodding Onion

ASTGIL

Onion

ASTKEN

Textile Onion

ASTLAE

Desert Alyssum

ASTLOT

Northern Fairy-

ASTMIO

candelabra

ASTMIS

Anemone

ASTPAN

Pasqueflower

ASTPEC

Meadow Pussy-toes

ASTPER

Everlasting Pussy-toes

ASTPUR

Rosy Pussy-toes

ASTRAG

Nuttall's Pussy-toes

ATRHOR

Dogbane

ATRIPL

Clasping-leaved Dogbane

ATRROS

Columbine

ATRSUC

Rockcress

Towermustard

BAHOPP

Hairy Rockcress

BUPAME

Holboell's Rockcress

CALELE

Heart-leaf Arnica

CALELE

Pacific Sagewort

Tarragon

CALNUT

Long-leaved Sagewort

CAM ROT

Prairie Sagewort

CAMSAT

Showy Milkweed

CASLIN

Whorled Milkweed

CENREP

Green Milkweed

CERARV

Standing Milk-vetch

CHADOU

Field Milk-vetch

CHEALB

Two-Groove Milk-vetch

CHEBOT

Painted Milk-vetch

CHEGIG

Showy Aster

CHELEP

Drummond's Milk-vetch

CHENOP

Aster

Scientific Name
Aster falcatus
Astragalus flexuosus
Astragalus gliviflorus
Astragalus kentrophyta
Aster laevls
Astagalus lotlflorus
Astragalus missourlensis
Astragalus miser
Aster pansus
Astragalus pectlnatus
Aster pereiegans
Astagalus purshll
Astragalus spp.
Atrlplex hortensis
Atrlplex spp.
Atrlplex rosea

Atrlplex suckleyl
(= A triplex dioica)
Bahia opposltifolia
Bupleurum amerlcanum
Calochortus elegans
Caiopiaca eiegans
(=Xar}thona elegans)
Calochortus nuttallll

Campanula rotundl folia
Cameiina satlva
Castiiieja tinariifoiia
Centaurea repens
Cerastium arvense
Chaenactis douglasll
Chenopodium album
Chenopodlum botys
Chenopodium gigantospermum
Chenopodlum leptophyllum
Chenopodium spp.

Common Name

White-prairie Aster
Wiry Milk-vetch
Plains Orophaca
Thistle Milk-vetch
Smooth Aster
Lotus Milk-vetch
Missouri Milk-vetch
Weedy Milk-vetch
Tufted White Prairie Aster
Tine-leaved Milk-vetch
Elegant Aster
Pursh's Milk-velch
Milk-vetch
Garden Orache
Saltbush
Red Orache
Rillscale

Plains Bahia

American Thorough-wax

Northwest Mariposa

Sego-lilly
Harebell

Gold-of-pleasure
Narrow- leaved Paintbrush
Russian Knapweed
Field Chickweed
Hoary Chaenactis
Lambsquarter
Jerusalem-oak Goosefoot
Maple-leaved Goosefoot
Slimleaf Goosefoot
Goosefoot

FORBS

Ccxle

Scientific Name

CHEPRA

Chenopodium pratericola

CHRVIL

Chrysopsis viiiosa

CIRARV

Cifsium arvense

CIRUND

Cirsium undulatum

CLESER

Cleome serrulata

COLLIN

Coliomia linearis

COMUMB

Comandra umbeifata

CONARV

Convoivuius arvensis

CONCAN

Conyza canadensis

CONORI

Conringia orientalis

CORMIS

Coryphaniha missouriensis

CREACU

Crepis acuminata

CREPIS

Crepis spp.

CRYCEL

Cryptantha ceiosioides

CYMBIP

Cymopterus bipinnatus

CYMGLA

Cymopterus giaucus

CYMHEN

Cymopterus hendersonii

DALCAN

Daiea Candida

DALPUR

Daiea purpurea

DESCUR

Descurainia spp.

DESPIN

Descurainia pinnata

DESRIC

Descurainia richardsonii

DESSOP

Descurainia sophia

DOUMON

Dougiasia montana

DRAOLI

Draba oilgosperma

ECHANG

Echinacea angustifoiia

EPIANG

Epiiobium angustifoiium

ERICAE

Erigeron caespitosus

ERICER

Eriogonum cernuum

ERICOM

Erigeron compositus

ERIFLA

Eriogonum fiavum

ERIPAU

Eriogonum paucifiorum

ERIPUM

Erigeron pumiius

Common Name

Slimleaf Goosefoot
Hairy Golden-aster
Canada Thistle
Wavy-leaved Thistle

Rocky Mountain Bee
Plant

Narrow-leaf Coliomia
Bastard To ad -flax
Field Morning-glory
Horseweed
Mustard Hare's Ear
Nipple Coryphantha
Tapertip Hawks beard
Hawks beard
Northern Cryptantha
Hayden's Cymopterus
Grayish Cymopterus
Henderson's Cymopterus
White Prairie-clover
Purple Prairie-clover
Tansymustard
Pinnate Tansymustard
Richardson's
Tansymustard
Flixweed Tansymustard

Rocky Mountain
Dougiasia
Few-seeded Draba
Pale Purple Coneflower
Fire weed
Tufted Fleabane
Nodding Wild Buckwheat
Cut-leaved Daisy
Yellow Buckwheat
Few-flowered Wild
Buckwheat
Shaggy Fleabane

Code

Scientific Name

Common Name

ERIUMB

Eriogonum umbeiiatum

Sulfur Buckwheat

ERYASP

Erysimum asperum

Plains Wallflower

ERYREP

Erysimum repandum

Spreading Wallflower

EUPROB

Euphorbia robusta

Rocky Mountain Spurge

EUPSER

Euphorbia serpyliifoiia

Thyme-leaf Spurge

FILARV

Filago arvensis

Field Filago

FRASPE

Frasera speciosa

Giant Frasera

FRAVES

Fragaria vesca

Woods Strawberry

GAIARI

Gaiiiardia aristata

Blanket-flower

GALBOR

Galium boreale

Northern Bedstraw

GALIUM

Galium spp.

Bed straw

GAUCOC

Gaura cocci nea

Scarlet Gaura

GAYDIF

Gayophytum diffusum

Spreading Groundsmoke

GERVIS

Geranium viscosissimum

Sticky Geranium

GEUTRI

Geum triflorum

Prairie Smoke

GILCON

Giiia congesta

(= Ipomopsis congesta)

Ballhead Gilia

GLYLEP

Glycyrrhiza lepidota

American Licorice

GOOOBL

Goodyera oblongifolia

Western Rattlesnake-
plantain

GRISQU

Grindelia squarrosa

Curlycup Gumweed

HACDEF

Hackeiia defiexa

Nodding Stickseed

HACFLO

Hackeiia fioribunda

Showy Stickseed

HAPACA

Hapiopappus acaulis

Cushion Goldenweed

HAPSPI

Hapiopappus spinulosus

Spiny Goldenweed

HEDDRU

Hedeoma drummondii

Drummond False
Pennyroyal

HEDHIS

Hedeoma hispida

Rough Pennyroyal

HELANN

Heiianthus annuus

Common Sunflower

HELIAN

Helianthus spp.

Sunflower

HELPET

Heiianthus petiolaris

Prairie Sunflower

HETVIL

Heterotheca viiiosa
(= Chrysopsis viiiosa)

Hairy Golden-aster

HEUPAR

Heuchera parvi folia

Small-leaved Alumroot

HEURIC

Heuchera richardsonii

Richardson's Alumroot

HYMFIL

Hymenopappus fHifolius

Columbia Cut-leaf

IRIMIS

iris missouriensis

Rocky Mountain Iris

FORBS

Code

Scientific Name

Common Name

IVAAXI

Iva axillaris

Poverty-weed

LACCAN

Lactuca canadensis

Wild Lettuce

LACSER

Lactuca serrioia

Prickly Lettuce

LACTUC

Lactuca spp.

Lettuce

LEPDEN

Lepidlum densifiorum

Prairie Pepperweed

LEPPER

Lepidium perfoiiatum

Clasping Pepperweed

LEPSAT

Lepidium sativum

Garden Cress

LEPSAT

Leptogium satumium

LESLUD
LESQUE
LIAPUN

Lesquereiia iudoviciana
Lesquereiia spp.
Liatris punctata

Silvery Bladderpod
Bladderpod
Dotted Blazing-star

LINLEW
LINPER

Linum iewisii
(=Unum perenne)
Linum perenne

Wild Blue Flax
Blue Flax

LINRIG

Linum rigidum

Yellow Flax

LITARV

Lithospermum arvense

Corn Gromwell

LITINC

Lithospermum incisum

Yellow Gromwell

LUPARG
LUPLEP
LYGJUN
MACCAN

Lupinus argenteus
Lupinus iepidus
Lygodesmia juncea
Machaeranthera canescens

Silvery Lupine
Prairie Lupine
Rush-like Skeletonweed
Hoary Aster

MACGRI

Machaeranihera grindeiioides

Nuttall's Goldenweed

MEDLUP

Medicago iupuiina

Black Medic

MEDSAT

Medicago saiiva

Alfalfa

MELALB

Mefaneiia aibertana

MELALB

Meiiiotus aiba

White Sweet-clover

MELOFF

Meiiiotus officinalis

Yellow Sweet-clover

MIRABI

Mirabilis spp.

Four-o'clock

MONFIS

Monarda fistuiosa

Horsemint

MUSDIV

Musi neon divaricatum

Leafy Musineon

OENCES
DENSER

OPUFRA

Oenothera cespitosa
Oenothera serrulata
(= OerKithera rhombipetaia)
Opuntia fragiiis

Desert Evening-primrose

Shrubby Evening-
primrose
Brittle Prickly -pear

OPUPOL

Opuntia poiyacantha

Plains Prickly-pear

Code

Scientific Name

Common Name

OROFAS

Orobanche fasclculata

Clustered Broomrape

OROLUD

Orobanche Iudoviciana

Suksdorf's Broomrape

ORTLUT

Orthocarpus luteus

Yellow Owl-clover

OSMCHI

Osmorhiza chilensis

Mountain Sweet-cicely

OXYCAM
OXYLAM

Oxytropis campestris
Oxytropis lambertii

Slender Crazyweed
Purple Locoweed

OXYTRO
PARPEN
PARSES

Oxytropis spp.
Parietaria pensylvanica
Paronychia sessili flora

Crazyweed
Pennsylvania Pellitory
Stemless Whitlow-wort

PENPRO
PENSTE

Penstemon procerus
Penstemon spp.

Small-flowered

Penstemon

Penstemon

PETCAN
PHACEL

Petaiostemon candidum
(= Dalea Candida)
Phacelia spp.

White Prairie-clover
Phacelia

PHAFRA

Phacelia frankllnii

Franklin's Phacelia

PHALIN

Phacelia linearis

Threadleaf Phacelia

PHLALY
PHLAND

Phlox alyssi folia
Phlox andicola

Alyssum-leaved Phlox
Moss Phlox

PHLHOO

Phlox hoodii

Hood's Phlox

PHLMUL

Phlox multi flora

Many-flowered Phlox

PHLOX

Phlox spp.

Phlox

PHYSAR
PICOPP

Physaria spp.
Picradeniopsis oppositifoiia

Twin pod
Plains Bahia

PLAARI

Plantago aristata

L arge- bract ed Plantain

PLAELO
PLALAN

Plantago elongata
Plantago ianceolata

Slender Plantain
Buckhorn Plantain

PLAPAT
POLALB

POLALB

Plantago patagonica
Pofychidium aibociiiatum
(= Leptochidium albociHatum)
Polygala alba

Indian-wheat
White Milkwort

POLAVI

Polygonum avicuiare

Prostrate Knotweed

POLERE

Polygonum erectum

Erect Knotweed

POLTRA
POLYGO

Poianisia trachysperma
Polygonum spp.

Clammy-weed
Knotweed

POTARG

Potentitta arguta

Tall Cinquefoil

FORBS

Code

Scientific Name

Common Name

POTENT
POTPEN
PSOARG

Potentilla spp.
Potentiiia pensylvanica
Psorafea argophylia

Cinquefoil
Prairie Cinquefoil
Silver-leaved Scurf-pea

PSOESC

Psoraiea esculenta

Indian Bread-root

PSOLAN

Psoraiea lanceolata

Lemon Scurf-pea

PYRSEC

Pyroia secunda

One-sided Wintergreen

RATCOL

Ratibida columnifera

Prairie Coneflower

RUMVEN

Rumex venosus

Wild Begonia

SALKAL

Salsola kali (= Safsola ibehca)

Russian Thistle

SEDLAN

Sedum lanceolatum

Lance-leaved Stonecrop

SENCAN

Senecio canus

Woolly Groundsel

SENINT

Senecio integerrimus

Western Groundsel

SENPLA

Senecio plattensis

Prairie Groundsel

SENSTR
SILSCO

Senecio streptanthif alius
Silene scoulerl

Rocky Mountain
Butterweed
Scouler's Silene

SISALT

Sisymbrium altissimum

Tumblemustard

SMIHER

Smilax herbacea

Carrion-flower

SMIRAC

Smilacina racemosa

False Spikenard

SMISTE

Smilacina steilata

Starry Solomon-plume

SOLCAN

Soildago canadensis

Canada Goldenrod

SOLIDA

Solldaqo spp.

Goldenrod

Ferns and Fern Allies

EOULAE Equisetum iaevigatum
SELDEN Selagineiia densa
WOOORE Woodsia oregana

Smooth Scouring Rush
Compact Selagineiia
Oregon Woodsia

Code

Scientific Name

SOLMIS

Soiidago missoun'ensis

SOLMOL

Sofidago moiiis

SOLMUL

Soiidago muitlradiata

SOLNAN

Soiidago nana

SOLNEM

Soiidago nemoraiis

SONARV

Son^us arvensis

SPHCOC

Sphaeraicea cocci nea

STERUN

Stephanomeria runcinata

SUAMOQ

Suaeda moquinii

TAROFF

Taraxacum officinaie

THAOCC

Thaiictrum occidentalo

THAVEN

Thaiictrum venuiosum

THERHO

Thermopsis rhombifoiia

TOW PAR

Townsendia parryi

TRADUB

Tragopogon dubius

TRAGOP

Tragopogon spp.

TRAMIS

Tragopogon misceiius

TRAOCC

Tradescantia occidentaiis

VICAME

Vicia americana

VIOLA

Vioia spp.

VIONUT

Vioia nuttaiiii

Common Name

Missouri Goldenrod
Velvety Goldenrod
Northern Goldenrod
Low Goldenrod
Gray Goldenrod
Field Milk-thistle
Red Globe-mallow
Runcinate-leaved
Skeltonweed
Tall Seablite
Common Dandelion
Western Meadowrue
Veiny Meadowrue
Round-leaved
Thermopsis
Parry's Townsendia

Goat's Beard

Salsify

Hybrid Salsify

Prairie Spiderwort

American Vetch

Violet

Yellow Prairie Violet