Development of a Montana land cover map from Landsat imagery
by Philip William Smith
A thesis submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE
in Soil Science
Montana State University
© Copyright by Philip William Smith (1981)
Abstract:
Landsat color composite transparencies were visually interpreted in an effort to produce an accurate
low cost land cover map of Montana. Seven categories of land cover were interpreted and the data
transferred directly to a mylar overlay registered to a 1:1,000,000 scale base map. Cover categories
included in the state map are range, forest, dryland crops, irrigated crops, alpine areas and rock
outcrops, water, and urban areas. The map was compiled at an estimated cost of $0.01/km^2 and
showed 90% agreement with existing county land use maps.
Upon completion, the land cover map was encoded along with the state soils map for conversion to
AREAS, South Dakota State University's computerized geographic information system. Output from
AREAS includes the hectares of land occupied by each cover type and each cover type's distribution on
Montana's soil associations. The results from AREAS show that range occupies 50%, forest 27%,
dryland crops 18%, irrigated crops 3%, alpine and rock outcrops 2%, water 0.5%, and urban less than
0.1% of Montana's total land area.
This study demonstrated manual interpretation of Landsat imagery to provide an inexpensive and
accurate alternative to conventional surveys for small scale land cover mapping. 
STATEMENT OF PERMISSION TO COPY
In presenting this thesis in partial fulfillment of the 
requirements for an advanced degree at Montana State University,
I agree that the Library shall make it freely available for 
inspection. I further agree that permission for extensive copying 
of this thesis for scholarly purposes may be granted by my major 
professor, or, in his absence, by the Director of Libraries. It 
is understood that any copying or publication of this thesis 
for financial gain shall not be allowed without my written 
permission.
Signature ^
Date
DEVELOPMENT OF A.MONTANA LAND COVER MAP 
FROM LANDSAT IMAGERY 
by
PHILIP WILLIAM SMITH.
A thesis submitted in partial fulfillment 
of the requirements for the degree
Of ' . . '
MASTER OF SCIENCE
in
. Soil Science
Approved:
Chairperson, Graduate CommBGtfee
Head, Majdr Dep^ytment
___________
Graduate Dean
MONTANA STATE. UNIVERSITY 
Bozeman, Montana
June, 1981
iii
ACKNOWLEDGMENTS
I wish to express my appreciation to the following individuals 
who helped make the completion of this thesis a reality:
Dr. Paul Kresge, whose patience perservered throughout this 
extended degree program of his first graduate student; Dr; Gerald 
Nielsen, for his input and help in finding enough work to keep me 
solvent; Dr. Earl Skogley, for his assistance in prescreening and 
ordering the imagery; and Dr. John Taylor. -
My sincerest gratitude is extended to my parents and wife,
Jan, whose tolerance, understanding and financial help made this 
whole endeavor possible.
TABLE OF CONTENTS
Page
VITA . . . i ............................................ ii
ACKNOWLEDGMENTS . . . . . . . . . . .  ...................  iii
LIST OF TABLES. . . . . ............... ...................  vi
LIST OF FIGURES . ................ ............... .. vii
ABSTRACT . . . .......... . . . . ! . ......................viii
INTRODUCTION..............i . . .......................  I
The Need for Land Cover Information
and Analysis.................................. I
LITERATURE REVIEW ........   3
Description of the Landsat Program .............. . 3
Monitoring Land Use from Space Platforms . . . . . .  5
Land Use Surveys from Manned
Spacecraft Photography .................. 5
Automated Analysis of Landsat Data .. .-........  6
Manual Analysis of Landsat Data .............. 7
Geographic Information Systems Using Remote
Sensing Data Input . ....................... . . 11
MATERIALS AND METHODS . . ............ .. . . . .........  15
Classification of Land C o v e r ...................... 15
Identification of Land Cover Types . . . ..........  16
Irrigated Crops . ........................   16
Dryland C r o p s .........................   17
Forest........    18
W a t e r ..........    18
U r b a n ........................................ 19
Alpine and Rock Outcrops...................... 20
R a n g e .........    20
Encoding of the State Land Coyer M a p .............. 21
Areal Tabulation of Land Cover Categories . ........ 22
Testing Land Coyer Map Accuracy...............   23
Source of Ground Truth D a t a .................. 23
Selection of Ground Truth Sites .............. 24
VTable of Contents (continued)
Page
RESULTS AND DISCUSSION . . .  . ...............' ..........  28
Land Cover Distribution in M o n t a n a ............. . 28
Comparison with Existing Montana Land
Use D a t a ...................................... 38
Land Cover Map Classification Accuracy ................  40
Cost of Land Cover Map Development . . . .............44
Comparison with Other Mapping Projects
and M e t h o d s .................................  45
CONCLUSIONS............■ . . . '...................... .. 48
LITERATURE CITED . . ....................................  50
APPENDICES
I INDEX OF LANDSAT IMAGERY USED TO PREPARE
THE MONTANA LAND COVER M A P ........'............  57
II SOURCES OF REMOTE SENSING IMAGERY ..... ..........  59
III MONTANA SOIL ASSOCIATION AND LAND COVER
COMBINATIONS . ..................................  65
IV MONTANA SOIL ASSOCIATIONS’ DESCRIPTIONS
AND PERCENT LAND COVER .............. . . . . . . .  74
vi
LIST OF TABLES
Table Page
1 Landsat MSS characteristics ...................... . 4
2 AREAS tabulations For land cover in Montana . . . . . .  30
3 Soil associations with significant range cover . . . .  31
4 Soil associations with significant forest
cover . ..............................   33
5 Soil associations with significant dryland
crop cover . . ................   34
6 Soil associations with significant irrigated
crop cover..........   36
7 Soil associations with significant alpine
or rock outcrop cover............................. ! 37
8 USDA, ESCS land use statistics for M o n t a n a .......... 38
9 Montana Situation Statement's areal tabulations
for water, range, forest, and irrigated crops ........ 40
10 Map accuracy by land cover type...............   41
11 Sample point distribution by land cover
categories...................................   43
12. Cost of producing 1:1,000,000 Montana land 
cover map from manual interpretation of
Landsat imagery......................  45
I-I Index of Landsat imagery used to prepare the
Montana land cover map . ............    58
III-I Montana Soil Association and Land Cover
Combinations.............................   66
IV-I Montana soil assocations1 descriptions and
percent land cover . i ...........    76
vii
LIST OF FIGURES
Figure Page
1. Area covered by county land use maps
used as ground t r u t h ...........    25
2. Photographic reduction of original 1:1,000,000
Montana land cover m a p .........     29
MAP SUPPLEMENT
I. Computer plotted Montana land cover map
viii
ABSTRACT
Landsat color composite.transparencies were visually inter­
preted in an effort to produce an accurate low cost land cover map 
of Montana. Seven categories of land cover were interpreted and 
the data transferred directly to a mylar overlay registered to a 
1:1,000,000 scale base map. Cover categories included in the 
state map are range, forest, dryland crops, irrigated crops, 
alpine areas and rock outcrops, water, and ur^an areas. The map 
was compiled at an estimated cost of $0.01/km and showed 90% 
agreement with existing county land use maps.
Upon completion, the land cover map was encoded along with 
the state soils map for conversion to AREAS, South Dakota State 
University's computerized geographic information system. Output 
from AREAS includes the hectares of land occupied by each cover 
type and each cover type's distribution on Montana's soil 
associations. The results from AREAS show that range occupies 
50%, forest 27%, dryland crops 18%, irrigated crops 3%, alpine 
and rock outcrops 2%, water 0.5%, and urban less than 0.1% of Montana's 
total land area.
This study demonstrated manual interpretation of Landsat 
imagery to provide an inexpensive and accurate alternative to 
conventional surveys for small scale land cover mapping.
■J
INTRODUCTION
The purpose of this research was to determine the capability 
and accuracy of manual interpretation of Landsat satellite imagery 
for inventorying land cover in Montana. Products of this research 
include a I:1,000,000 scale state land cover map for Montana and 
areal tabulations for each land cover category and its distribution 
on the state's soil associations.
The Need for Land Cover Information and Analysis 
Until now, no land cover/use map has been available for 
Montana. Montana's size and natural resource diversity create a 
need for a reliable and economical method of land resource inven­
tory. Montana is the fourth largest state in the nation with a 
total surface area of 235,421 square kilometers (USDI-BLM, 1969). 
Within this area, the complexity of Montana's geologic, climatic, 
vegetational and pedologic patterns combine to form a variety of 
natural ecosystems. Since the demands on these ecosystems are in­
creasing and often conflicting, there is a need for an efficient and 
cost-effective method of land resource inventory and analysis. The 
perspective offered by remote sensing from aerial and space plat­
forms provides for the rapid collection of current and detailed land 
resource data. Particularly valuable is the spatial context of the 
data, which provides insight into the distribution of resources,
2their areal extent, and proximal relationships (Cox, 1977). These 
assets make land resource inventories compiled from remote sensing 
data a viable alternative to more conventional surveys.
LITERATURE REVIEW
The recording of remotely sensed data from aerial platforms was 
underway in the late 1800's and early 1900's. In the following 
years, two world wars and several international crises stimulated 
rapid advances in remote sensing methods and technology.
The conservation-oriented programs of the New Deal helped to 
initiate large aerial mapping progams in the United States. Many of 
these programs are still conducted by the USDA and other agencies as 
part of their natural resource surveys.
The manned space program of the 1960' s brought space photog­
raphy into its own and demonstrated the potential for monitoring 
earth resources from space (Kroeck, 1976).
The Landsat satellite series, operational since 1972, adds 
another dimension to remote sensing from space. Providing repeti­
tive coverage of the earth's surface, Landsat makes possible the 
continuous monitoring of the earth's natural resources.
Description of the Landsat Program
The Landsat Program (formerly the Earth Resources Technology 
Program) was designed as a research and development tool to demon­
strate the usefulness of satellite remote sensing in the management 
of earth resources (Anonymous, 1977). Since the launching of 
Landsat-I (formerly ERTS-I) on July 23, 1972, two additional
4satellites, Landsats-2 and -3, have been put in orbit. Although 
Lahdsat-I exceeded its planned orbital life, at this writing only 
Landsats-2 and -3 are still producing data.
Each satellite was launched into.a near polar, sun-synchronous 
orbit at an average altitude of 910 kilometers. The sensor design 
for all three Landsats is similar except for modifications made on 
Lapdsat-3. Landsat's main sensor system is the multispectral scanner 
(MSS). The MSS collects radiometric data in four discrete spectral 
bands within the visible and near infrared spectrum (Anonymous,
1977) (Table I). Upon collection, the data are either transmitted 
directly to one of the ground receiving stations at Fairbanks,
Alaska; Goldstone, California; and Greenbelt, Maryland or stored on 
tape until the satellite comes in view of a ground station.
Table I. Landsat MSS characteristics.
Band Wavelength (urn) Spectral Region
4 0.5 - 0.6 green
5 0.6 - 0.7 red
6 0.7 - 0.8 near IR
7 0.8 - 1.1 near IR
5The spacecraft motion produces a continuous strip of data. 
During image processing at the Ground Data Handling System Facility, 
this strip of imagery is formed into separate images. These images, 
called scenes, incorporate a ground coverage area of 185 x 185 
kilometers. The lateral ground resolution of the MSS is approxi­
mately 80 meters. Each picture element (pixel) covers a ground area 
of about 0.4 hectares, but the minimum area resolved is Usually 
closer to 2 to 4 hectares (Seevers and Peterson, 1978).
' Landsat data are available in either computer compatible tapes
(CCTs) containing the digital data of all spectral bands or in a
1
variety of photographic products.(Gonzoles and Sos, 1974).
Monitoring Land Use from Space Platforms 
Starting with photographs from early manned flights, numerous 
land use studies have.been performed using imagery taken from space 
platforms.
Land Use Surveys from Manned 
Spacecraft Photography
Manual interpretation of Gemini and Apollo manned spacecraft 
photography was employed to produce generalized land use maps at 
scales of 1:250,000 and I:1,000,000 for portions of the southwestern 
United States. . The broad view provided by the photos was found to
6Skylab photography was interpreted visually in an attempt to 
map land use in the Fairfax, Virginia area. It was established that 
the Skylab photography filled the gap between the coarser resolution 
Landsat MSS data and high altitude U-2 plane photography. The 
Skylab photography enabled the interpreters to map urban structural
detail from orbital altitudes. In comparison, urbanized areas could
'
be delineated on Landsat MSS images but it was difficult to deter­
mine the specific activities (commercial, industrial, etc.) present 
(Lins, 1976).
Automated Analysis of Landsat Data
The format of Landsat products makes possible either manual or 
automated (computer) interpretation of the MSS' data. Automated 
techniques, involving the analysis of computer compatible tapes have 
advantages over visual interpretations resulting from the rapid 
quantitative identification of land resource categories and the 
immediate generation of summary statistics (Welch, et al., 1979).
Automated analyses of Landsat data have been employed in several 
studies involving general land use (Dqrnbach and McKain, 1973;
Welch, et al., 1979) and agricultural land use (Richardson, et al.,
be important in the study of regional relationships of both physical
and cultural features (Thrower, 1970).
71977; Hanuschak, et al., 1979; Bauer, 1978). In the automated 
analysis of Landsat data, Draeger, et al. (1974) emphasized the 
importance of human-computer interaction, concluding that the most 
efficient and cost-effective method for producing "automated" 
surveys involves the use of manual interpretation at several stages 
of the process, particularly in the preliminary stratifications.
Manual Analysis of Landsat Data
The analysis of Landsat data by manual interpretation tech­
niques requires less specialized training and less expensive equip­
ment than automated procedures (Heller and Johnson, 1979). In addi­
tion, visual interpretation of Landsat imagery permits the rapid 
construction of regional and local map products (Welch, et al., 
1979).
Using a variety of techniques but emphasizing visual interpre­
tation methods, small scale land use maps of the northeastern por­
tion of the Plain of China and the Nun River Basin were produced 
from Landsat imagery.. To help identify crop types, aids including 
crop calendars, reflectance spectra and the use of images recorded 
on dates maximizing reflectance differences between crops were 
applied (Welch, et al:, 1979).
Manual interpretation methods were also used to generate a land 
use map for the state of Iowa from Landsat imagery. The map was
I ■
8commercial/industrial., urban open, transportation networks, extrac­
tive land, agricultural land, forest, water, and reservoir flood. 
pools. Summer images were used since it was determined that they 
provided the most complete land use information (Anderson, 1977).
Landsat color composite prints (a single photographic product 
produced by transmitting blue, green and red light through images of 
bands 4, 5 and 7 respectively) were interpreted visually and the 
data mapped directly onto an acetate overlay to produce a land use 
map of Boone County, Missouri. The interpreters used ancillary data 
including geologic, hydrologic, topographic, soils, road and pre­
vious land use maps in addition to familiarizing themselves with 
crop calendars, weather data and existing aerial photography. Using 
this methodology, maps with accuracies between 70 and 90 percent 
were easily and quickly generated (Elifrits, et al., 1978).
The impending development of Wyoming's fuel reserves requires 
that a quick and efficient method of land resource inventory be 
employed to help manage the state's forthcoming growth; The use of . 
photo interpretive techniques for analyzing Landsat imagery were 
investigated in a preliminary study of Wyoming's Powder River Basin. 
Landsat data were able to provide much of the needed physiographic
drawn at a scale of 1:250,000 and then printed at 1:500,000 and.
includes nine categories of land use: Urban residential, urban
9and land use information more quickly and efficiently than conven­
tional inventory methods. Although Landsat imagery did not furnish 
the detail needed for some land use applications, it was found to be 
capable of providing a great deal of broad scale land use data on a 
continuing basis (Breckenridge, et al., 1973).
Cropland inventories comparing 1:125,000 scale color infrared 
(CIR) high altitude photography and Landsat 1:1,000,000 scale (band 
5) black and white imagery were conducted in Kern County, California. 
Inventories completed using imagery acquired on a single date re­
vealed no significant differences in accuracy between the two imagery 
sources. In terms of absolute accuracy though, the optimum results 
were^achieved by a multidate analysis of the Landsat imagery. This 
is significant in that even though the scale had been decreased and 
the resolution degraded compared to the 1:125,000 photography, the 
analysis of Landsat data achieved comparable results and in this 
study slightly higher absolute accuracy (98% vs. 97% accuracy) than 
the CIR photos. The availability of multidate Landsat scenes and 
the lower cost of Landsat imagery for an equivalent ground area give 
Landsat additional advantages over high altitude CIR photography.
In any case, surveys completed from either high altitude CIR photos 
or Landsat imagery can be accomplished at 3% to 5% of the cost and 
in 95% less time than conventional cropland surveys (Jensen, et al., 
1975).
10
The repetitive coverage of Landsat enables the interpreter to 
monitor changes in land resources over specific time periods. For 
California's San Joaquin Valley, photo interpretation of Landsat 
imagery was used to record land use changes during a one year period. 
The imagery proved adequate in monitoring and mapping macrolevel 
land use changes. However, it was determined that to achieve the 
best mapping accuracy the interpreters should be familiar with the 
areas1s landscape features and their geographic distribution (Estes, 
et al., 1974).
To estimate irrigated acreage in a section of southwestern 
Idaho, Landsat false color composites were manually interpreted and 
the data recorded on a grid corresponding to a commonly used map 
base. Each map grid cell represented a ground equivalent area of 
3.6 x 3.6 kilometers and the imagery was selected to match the time 
when the bulk of the irrigated crops were at maturity. Imagery 
covering a three year time span was chosen to determine the extent 
of irrigated cropland expansion during that period. The data was 
mapped at 1:1,000,000 and found to have a sampling error of 6-10%, 
an error rate quite acceptable for this type of survey (Heller and 
Johnson, 1979).
A similar inventory of irrigated lands was completed for the 
Klamath River Basin of Oregon. For the inventory it was. necessary 
to use imagery from both mid-summer and late-summer to discriminate
between irrigated and dry farmland. In July all irrigated fields 
appear red on the imagery.as do a few dryland areas. However* by 
September all dryland crops have matured and appear tan on the 
imagery while most of the irrigated lands still have a red appear­
ance. By comparing the images from the two dates the irrigated and 
dryland fields could be definitely separated (Draeger, et al., 
1976).
Place (1974) used manual interpretation of Landsat imagery in 
an attempt to map land use change in the Phoenix, Arizona area. 
Changes detected included conversion of crop and rangeland to resi­
dential development, desert to new cropland, and new reservoir
fill-up. Land use change detection from interpretation of high
.
altitude U-2 plane photos was more detailed than the Landsat inter­
pretation but the total area and pattern of change were similar for 
both interpretations. An attempt to enhance change detection by 
overlaying images of different dates using different color filters 
proved ineffective. The primary differences shown by enhancement 
proved to be different stages of vegetative growth and not land use 
changes.
Geographic Information Systems Using 
Remote Sensing Data Input
The formulation of geographic information systems (GIS) has 
been a recent development in the management of land resource data.
Such systems commonly contain data on soils, geology, vegetation, 
transportation, political units and other land related factors 
(Ford, 1978). Shelton and Estes (1979) discussed the need for the 
integration of geographic information systems with remotely sensed 
data. Hill-Rowley and Enslin (1979) noted that the incorporation of 
remotely sensed land cover information into regional computer grid 
systems should be a defined objective of regional planning agencies.
New York State's LUNR (Land Use and Natural Resource Inventory) 
was one of the first systems developed that incorporated remotely 
sensed data on a large scale. Initiated in 1966 by. the Center for 
Aerial Photographic Studies at Cornell University, LUNR used visual 
interpretation of conventional black and white aerial photographs to 
compile a detailed inventory of land use and natural resources in 
New York. This information is stored on computer discs and products 
of the system include overlay maps of areal, point, or linear data; 
printouts listing information about each cell; and computer maps 
indicating the location of quantitative analyses performed on the 
data (Ford, 1978; Hardy and Shelton, 1970; Shelton and Tilman,
1978).
Many geographic information systems are now using information 
derived from the interpretation of Landsat satellite imagery as a 
source of land resource data. In Minnesota, Landsat imagery has 
been applied to updating land use data for the Minnesota Land
12
13
Management Information System (Sizer, 1974). It was established 
that high quality Landsat images could yield more detailed land use 
data than previously existed for Minnesota (Brown, et al., 1973).
The United States Geological Survey has tested the capability 
of Landsat as a source of environmental data for CARETS (Central 
Atlantic Regional Ecological Test Site), an.experimental regional 
environmental information system. Like LUNR, CARETS is a geographic 
information system with overlay capabilities for computer map pro­
duction. This system has been used in test studies to detect land 
use changes along the mid-Atlantic coast of the United States with 
favorable results (Alexander, 1973 and 1974).
CRIES (Comprehensive Resource Inventory and Evaluation System), 
originated in 1975 by the United States Department of Agriculture 
and Michigan State University for estimating agricultural growth 
potential in developing countries, also employs Landsat data as a 
source of land use information. In the Dominican Republic manual 
interpretation of Landsat imagery proved to be accurate, practical, 
and more discriminating than computer classifications for inputs 
into CRIES. The use of manual techniques for interpreting Landsat 
imagery is also advantageous in that! the technology required for 
such interpretations is easily transferred to workers in developing 
areas (Shelton and Tilman, 1978).
14
Input of remotely sensed data into a GIS allows combination 
with other spatial resource data to aid in planning and zoning 
activities. Such an analysis was undertaken near the Black Hills of 
South Dakota where land use data from visually interpreted high 
altitude photography were composited with soils information to 
assist in land use decisions (Cox, 1977).
Landsat data were added to a computer management system for a 
power plant siting study on the Delmarva (Delaware, Maryland and 
Virginia) Peninsula. The result was the production of computer 
drawn composite maps showing areas having acceptable land resource 
parameters for either nuclear or fossil fuel power plants (Halpern, 
et al., 1975).
There are several advantages to computer processing of digi­
tized data contained within a geographic information system. Digi­
tal data are registered to geographic coordinates and stored in 
easily accessible form for overlay and monitoring capabilities. In 
some systems the planimeter task can be done by computer rather than 
by manual methods and spatial display of the data can be provided at 
the desired map scale by computer plotter. This eliminates the need 
for cartographic or photo lab renditions of the data including 
enlargements and reductions, making a variety of hardcopy products 
readily available (Cox, 1977).
MATERIALS AND METHODS
Manual interpretation of Landsat false color composite trans­
parencies was used to produce a land cover map for Montana at a 
scale of 1:1,000,000.. The thirty-five images needed for statewide 
coverage were selected based on percent cloud cover and acquisition 
during the growing season. The images were purchased from the 
Aerial Photography Field Office of the USDA-ASCS in Salt Lake City, 
Utah. A list of the images used is given in Appendix I. It was 
felt that cloud free imagery obtained during the growing season 
would be the most useful in identifying land cover. Therefore, only 
images from the months of May through September with cloud cover of 
10 percent or less were chosen. The images used in this study range 
in date from 1972 to 1976.
To produce the state land cover map each image was registered 
to rivers, lakes and other identifiable landmarks displayed on a 
translucent mylar base map at the 1:1,000,000 Albers Equal-Area 
Projection. The interpretations were done with the aid of a light 
table and transferred directly to a matte-finish mylar overlay 
registered to the base map.
Classification of Land Cover
The term land cover is preferred over land use since "use" 
carries the connotation of a specific activity being present. Such
16
activity may or may not be detectable on remote sensing imagery. 
Burley (1961) defined land cover as "the setting in which action 
takes place, i.e., the vegetation and artificial constructions 
covering the land surface." Seven categories of land cover were 
delineated in the production of the state land cover map: Irrigated
crops, including irrigated pasture and subirrigated lands; dryland 
crops; forest; range; water; alpine and rock outcrops; and urban 
areas having a 1970 census population exceeding 20,000.
Identification of Land Cover Types
A number of factors were considered in identifying land cover 
categories. These included a landscape feature's color, texture, 
size, shape and its association with other features. The following 
sections describe in detail some of the factors considered in the 
interpretation of land cover types.
Irrigated Crops
Irrigated crops are often located in valley bottoms and along 
natural drainages in Montana. Since they have a high reflectance in 
the near infrared, irrigated crops are identified during the growing 
season oh the false color composite imagery by their bright red 
appearance. To avoid confusion with non-irrigated vegetation, 
imagery should correspond to dates after which noh-irrigated crops
17
have begun to ripen. On imagery acquired earlier, non-irrigated 
vegetation may resemble irrigated crops due to the continued 
presence of adequate soil moisture.
Some irrigated land is readily identified. Such is the case 
with irrigated crops under center pivot irrigation systems which can 
be recognized by their distinctive circular shape. However, some 
land mapped as irrigated crops is associated with riparian and sub­
irrigated lands. Since it is difficult to distinguish between 
irrigated crops and riparian and subirrigated lands, no effort was 
made to separate them in areas where they were iinterspersed.
Dryland Crops
As explained in the previous section it is best to use imagery 
from the latter part of the growing season to avoid confusing dryland 
with irrigated crops. As dryland crops mature, their appearance on 
the color composite imagery changes from progressively lighter 
shades of fed, to pink and then to a whitish hue upon crop ripening. 
In addition to their color, the linear boundaries of many dryland 
crop fields aid in their identification. Strip-cropped areas can be 
readily recognized by their contrasting bands of cropped and fal­
lowed land. Areas under fallow at the time of image acquisition 
were included in the dryland crop classification. On the imagery, 
fallow areas appear from dark to light grayish blue-green depending
18
on the tillage techniques employed and the soil surface conditions. 
Fallowed areas containing stubble appear light brown to gray and may 
be difficult to distinguish from range cover, particularly where 
dryland crops, and dryland pasture are mixed.
Forest
Areas classified as forest consist of all identifiable coni­
ferous woodland areas. Forest areas, like other concentrations of 
vigorous vegetation, reflect highly in the near infrared giving them 
a red appearance on the Landsat imagery. In contrast to the bright 
red of irrigated crops, forest areas are represented by a darker red 
on the imagery. Other factors which may aid in the identification 
of coniferous forests include their granular appearance on the 
imagery and their irregular boundaries which often follow variations 
in aspect and topography.
Water
Reservoirs and lakes were classified as water on the land cover 
map. Clear water absorbs a large percentage of the incident electro­
magnetic radiation causing it to appear black on the imagery. With 
increasing turbidity the appearance of water changes from black to 
light blue. In mountainous areas, water bodies may be hard to iden­
tify because of the masking effects of shadows from the rough ter­
19
rain. Generally, however, water bodies are easily identified and 
mapped from Landsat imagery.
Urban
Urban areas mapped include only those cities whose populations 
were greater than 20,000 at the 1970 census. Although visible on 
the imagery, smaller urban areas were not mapped due to their com­
paratively insignificant areas and the problems associated with 
their representation on a map of such small scale.
The appearance of urban areas on Landsat imagery varies from 
red and flesh^colored tones to white and/or mottled blue patterns. 
Residential areas appear either flesh-colored or shades of red due 
to the integration of white tones produced by highly reflective 
objects such as roofs, sidewalks and streets, and red tones from the 
high near infrared reflectance of trees, lawns and other vigorous 
vegetation. Residential areas, especially low density residential 
areas, may be confused with dryland and irrigated crops in the 
spring. To avoid confusion it may be beneficial to use imagery 
taken later in the summer when residential vegetation begins to 
suffer from lack of moisture, mowing, and other stresses (Anderson, 
1977).
Due to the lack of vegetation and presence of highly reflective 
objects, main business and industrial areas are represented by white
20
to blueish tones on the imagery. Radiating out of the hub of the 
central business district is often a network of highways, blueish- 
gray in appearance, which may also serve as an aid in identifying 
urban centers.
Alpine and Rock Outcrops
The classification, alpine and rock outcrops, includes all land 
areas above the treeline and locations having sparse vegetation 
where rock outcrops predominate. The appearance of alpine areas 
differs in response to snow cover and vegetation condition. Snow is 
highly reflective, hence, snow-covered areas are white on the 
imagery. Snow-free alpine areas also may appear white where little 
vegetation is present and the area is dominated by light-colored 
rocks. Alpine areas having sufficient vegetation vary in appearance 
from pink to red depending on the conditions and quantity of the 
vegetation present.
The appearance on the imagery of areas dominated by large out­
croppings of rock is dependent on the color of the rock. Light 
colored rock outcrops are highly reflective and are represented by 
light tones while less reflective rocks appear dark on the imagery.
Range
The range classification includes treeless areas below treeline 
dominated by grasses and/or shrubs, dryland pasture, and badland
21
areas. Range varies considerably in its spectral response depending 
on soil moisture conditions, vegetation condition, and other vege­
tation and soil influenced reflection properties. Range appears as 
shades of gray or gray-brown on the color composite imagery. To 
avoid confusion with irrigated or dryland crops, range is best 
depicted on imagery obtained in the latter part of summer. At this 
time the moisture stressed range vegetation is easily distinguished 
from the red color of healthy irrigated crops and the uniform pink 
to white response of ripening dryland crops.
Encoding of the State land Cover Map 
A computerized geographic information system developed by Gary 
L. Ford is available.at Montana State University (Ford, 1978). Upon 
completion, the state land cover map was digitized and added to this 
system. This increased the number of maps presently stored in the 
system to seventeen. These maps include:
1. Soils
2. 1941-70 Average Annual Precipitation
3. 50 Year Peak Precipitation
4. Number of Strong Chinooks per 100.Years
5. Potential Evapotranspiration
6. Average Annual Effects of Erosive Rains
7. Average Length of Frost-Free Season
8. Average Date of First Freeze
9. Average Date of Last Freeze
10. Consumptive Use of Water
11; 1968-72 Average Annual Snowfall
12. Geology
13. Percent of Annual Precipitation Falling from April I 
thru July 31
22
14. Percent of Annual Precipitation Falling from May I 
Thru July
15. Climax Vegetation
16. Elevation
17. Land Cover
Products of the system include both plotter-drawn maps and 
alpha-numeric line-printer maps. The system is capable of out- 
putting data from a single map or performing a joint analysis of 
several maps (i.e., soils, precipitation, length of growing season, 
etc.) to produce a composite map locating areas containing a par­
ticular set of environmental parameters.
In order to be encoded into the system, a map must, first be 
prepared at a 1:1,000,000 scale with Albers Equal-Area projection. A 
plotter-drawn mylar grid at the same scale and projection is used to 
encode the map data. The grid is based on latitude and longitude 
and has an average cell size covering 21.3 square kilometers. The 
grid is overlain on the map and the data are coded for each cell by. 
the dominant category present. If no category is dominant then the 
class at the cell's center is encoded. The data is then recorded 
and keypunched and the resultant computer drawn map is checked for 
errors (Ford, 1978).
Areal Tabulation of Land Cover Categories
The encoded state land cover and general soils (USDA-SCS, 1978) 
maps were sent to the Remote.Sensing Institute at South Dakota State
23
University for conversion to AREAS (Area Resource Analysis Systems) 
for areal tabulations. AREAS is a computerized geocoded cellular 
information system. The output received from AREAS included a 
breakdown of the hectares of each soil mapping unit under a specific 
land cover (Appendix III). These data were then analyzed for soil- 
land cover relationships throughout Montana (Appendix IV).
Testing Land Cover Map Accuracy 
Source of Ground Truth Data
Due to the prohibitive expense and amount of time required for 
the collection of ground truth data by field observations, existing 
land use maps for eighteen Montana counties were used to check the 
land cover map accuracy. The maps were prepared by the Montana 
Statewide Cooperative Land Use Mapping Program, coordinated by the 
Department of Community Affairs, Helena, Montana, and the 
Yellowstone-Tongue Area Wide Planning Organization, Broadus, Montana. 
The Department of Community Affair's maps include: Broadwater,
Carbon, Cascade, Deerlodge, Hill, Lewis and Clark, Mineral, Missoula, 
Park, Pondera, Silver Bqw and Teton Counties. Counties embodied by 
the Yellowstone-Tongue Area Wide Planning Organization maps are: 
Carter, Custer, Fallon, Powder River, and the northern portion of
.
24
Rosebud County. The shaded areas of Figure I show where these 
counties are located in association with the remaining portion of 
the state.
Selection of Ground Truth Sites
Ground truth sites were chosen by a stratified random sampling 
of sections located within the described eighteen county area. To 
select each section, rapdom numbers corresponding to tier and range 
designations were computer generated to specify township and random 
numbers between I and 36 were generated to determine section. To 
locate each section oh the land cover map, a map of identical scale 
and projection displaying township locations was overlaid on the 
land cover map. Each specified section was then located with the 
aid of 16 lines per inch graph paper. Each cell on the graph paper 
closely approximates a section at a scale of 1:1,000,000 (I cell 
represents 0.99 square miles). The dominant land use within each 
section was then used to classify each sample site. The resulting 
classification was then checked for agreement with the classifi­
cation within the same section on the county land use maps.
NJ
Ln
Figure I. Area covered by county land use maps used as ground truth (shaded regions)
26
The minimum number of sample sites needed to check each land 
cover category was determined by the following formula:
2 2
N = .
L
Where N is the minimum number of samples needed, 1.96 is the t-table
2
value for a confidence interval of .95, S is the estimated variance 
for a binomial distribution and L corresponds to the acceptable 
degree of error.
Assuming a map accuracy of at least 80% with a .95 confidence 
interval yields:
H (1.96)2(0.80)(0.20) _ 615 _
o.io2
To insure an adequate sample, a running total was kept for each 
category until 70 checks were performed on each classification. This 
figure is in excess of the minimum sample size of 50 recommended by 
Hay (1979) for sampling land use map classification accuracy.
Exceptions to a sampling number of 70 are the water and urban 
classifications. Due to the difficulty in obtaining seventy random 
samples on the small relative area occupied by water in Montana, 
only 25 sample checks on the water classification were performed. It 
was felt that this number would be adequate, particularly since
27
surface waters are easily identified.on the satellite imagery. In 
addition, the selection of ground truth sites was not limited to the 
eighteen county area previously described. Instead, USGS topographic 
maps were used as ground truth so the sampling area for water could 
be expanded to include almost all of Montana.
The insignificant area occupied by urban areas in Montana made 
sampling of these areas impractical. Therefore, no checks were 
performed on the urban cover classification.
Upon completion of the sampling scheme, the overall classifi­
cation accuracy for all samples was calculated. In addition, the 
accuracy for each classification category was determined.
RESULTS AND DISCUSSION
Land Cover Distribution in Montana
Figure 2 is a photo-reduced copy of the 1:1,000,000 Montana 
land cover map produced from manual interpretation of Landsat 
imagery. A 1:1,000,000 computer plotted version of the land cover 
map is presented as a map supplement to this thesis. The computer 
drawn map does not possess the detail of the original hand-drafted 
map but it still demonstrates general land cover distribution in 
Montana.
The results of the AREAS program's tabulations of the areal 
extent of the seven land cover categories (Table 2). show range to 
be the dominant cover type in Montana. Range occupies approxi­
mately 50% (18,890,000 hectares) of Montana’s total land area and 
is found predominately in the eastern two-thirds of the state and 
in the intermontane valleys of the west. The largest extents of 
uninterupted rangeland occur in east central Montana where soils 
have developed on soft sandstones, siltstones and claystones. In 
this area, topography is locally rugged due to differential erosion 
between beds of softer and harder materials (Veseth and Montagne, 
1980). Many of the soil associations having extensive range cover 
are located on these rough, dissected bedrock plains (Table 3). 
Significant acreages of rangeland also, occur on the glaciated till
s
KEY 'O'
□  IRRIGATED CROPS
□  DRYLAND CROPS
■  FOREST
□  RANGE
H ALPINE a  ROCK OUTCROPS
■  URBAN (» 2 0 ,0 00  pop )
■  WATER
N)
<D
Photographic reduction of original 1:1,000,000 Montana land 
cover map.
Figure 2.
Table 2. AREAS tabulations for land cover in Montana
Land cover 
type Hectares
(thousands)
Acres
Percent 
of total
Range 18,890 46,642 50
Forest 10,029 24,763 27
Dryland . 6,614 16,330 18
Irrigated ' 1,199 2,960 . 3
Alpine and 
Rock Outcrop 734 1,812 2
Water 204 505 0.5
Urban 23 57 < 0.1
Total Area 37,693 93,069 100
31
Table 3. Soil associations (USDA-SCS, 1978) with significant
range cover.
Mapping
Unit
Symbol
Association 
Description ,
Hectares 
of Range 
Cover
(thousands)
%  of Map­
ping Unit 
in Range
Location of Major 
Range Cover 
Areas
Th Entisols-Aridisols: 
Strongly sloping soils 
on dissected sedimen­
tary bedrock plains 
and hills
3,305 86 Extensive areas 
in east central 
Montana
Vh Entisols-Aridisols: 
Predominantly clayey, 
strongly sloping soils 
on dissected shale 
plains
1,896 89 Areas throughout 
eastern Montana
Sg Mollisols-Aridisols: 
Nearly level to hilly 
soils on glacial till 
plains
1,173 36 Northern east 
central Montana
Nh Entisols-Inceptisols- 
Mollisols: Strongly
sloping soils on sedi­
mentary bedrock plains 
and hills
1,117 66 Southern east 
central Montana 
and scattered 
areas in the. 
northeast
Tp Aridisols-Entisols: 
Nearly level to moder­
ately steep soils on
1,113 78 Scattered areas 
throughout south­
eastern Montana
plains
32
plains of north central Montana. Elsewhere in the state range is 
often interspersed with other cover types.
Forest occurs on almost 27% of Montana's total land area.
Forest cover is primarily limited to the western mountains and the 
highly dissected plains and hills of eastern Montana (Table 4). 
Dominant cover species vary from Pinus ponderosa Dougl. and Juniperus 
scopulorum Sarg. (ponderosa pine and Rocky Mountain Juniper) in 
eastern Montana to Thuja plicata Bonn, and Layrix occidentalis 
Nutt, (western red cedar and western larch) in the northwest (Ross 
arid Hunter, 1976). Crown covers differed from 15% for areas in 
eastern Montana to about 100% in the western mountains.
Dryland crops occupy over 6.6 million hectares in the state
’ ' -I
comprising about 18% of Montana's total land area. Observation of 
the dryland crop distribution (Figure 2) reveals two extensive 
areas in Montana; the state's northeast corner and the Triangle 
area located in north central Montana. These areas correspond to 
the extent of the last continental ice sheet advance with the 
dominant dryland cropped areas being located on glacial till plain 
soils (Table 5). One soil association mapping unit, Sg,
(USDA-SCS, 1978) contains over 31% of the state's total dryland 
area. This unit is extensively cropped in the Triangle area. The 
Judith Basin contains significant dryland crop cover on soils of
33
Table 4. Soil associations (USDA-SCS, 1978) with significant forest
cover.
Mapping
Unit
Symbol
Association
Description
Hectares % of Map- 
of Forest ping Unit 
Cover in Forest 
(thousands)
Location of Major 
Forest Cover 
Areas
Me Inceptisols-Alfisols: 
Moderately sloping to 
very steep soils on 
mountains
2,810 92 Mountains of 
western Montana
Mf Inceptisols: Moder­
ately sloping to very 
steep soils on mountains
2,044 81 Mountains of 
northwestern 
Montana
Mo Mollisols-Inceptisols- 
Alfisols: Gently
sloping to very steep 
soils on mountains
1,050 77 Mountains east 
of the divide
Ma Inceptisols (Andie): 
Steep and very steep 
soils on mountains 
mantled by volcanic ash
827 94 Northwestern
Montana
Lg Inceptisols-Alfisols: 788
Undulating to rolling 
soils in valleys and on 
foothill glacial moraines
91 Mountain valleys 
of northwestern. 
Montana
Th Entisols-Aridisols: . 494 13 Southeastern
Strongly sloping to steep Montana
soils on dissected sedir 
mentary bedrock plains 
and hills
34
Table 5. Soil associations (USDA-SCS, 1978) with significant dryland 
crop cover.
Mapping
Unit Association 
Symbol Description
Hectares 
of Dry­
land 
Cover
(thousands)
% of Map- Location of Major 
ping Unit Dryland Cover 
in Dryland Areas
Crops
Sg Mollisols-Aridisols: 2,077
Nearly level to hilly 
soils on glacial till 
plains
Og Mollisols-Entisols: 958
Undulating to strongly 
rolling soils on 
glacial till plains
Gb Mollisols: Nearly 419
level to steep soils 
on fans, benches, and 
terraces
Np Mollisols-Entisols: 417
Nearly level to strongly 
sloping soils on sedi­
mentary bedrock plains
Nh Entisols-Inceptisols- 402
Mollisols: Strongly
; sloping to steep soils
on sedimentary bedrock 
plains and terraces
64 Triangle area
66 Northeastern
Montana
32 Judith Basin
42 Wibaux area
25 Northeast
Montana and 
Judith Basin
35
alluvial fans, benches and terraces, while other areas of exten­
sive dryland crops occur on sedimentary bedrock plains in eastern 
Montana.
Together range, forest and dryland crops make up approxi­
mately 95% of Montana's land cover. The remaining 5% is occupied 
by irrigated crops (3%); Alpine and rock outcrops (2%); water 
(0.5%); and Urban (<0.1%).
Irrigated crops are present in the intermontane valleys of 
western Montana and along major river drainages in the east. The 
western intermontane valleys have intermittently acted as deposi- 
tional basins since at least late Eocene. Adjacent to the basins' 
modern floodplains, pediment surfaces extend up the mountain 
fronts. These pediments are often covered by alluvial fans, 
benches or terraces (Fields and Petkewich, 1971). It is on these 
geomorphic features; floodplains, fans, benches and terraces; that 
much of Montana's irrigated crops occur. Outside of the inter­
montane valleys, large irrigated cropland areas are located along 
major river drainages such as the Milk and Yellowstone Rivers 
(Table 6).
Alpine and rock outcrops are limited to the higher elevations 
and ridge crests of the western mountains. Table 7 presents the 
major soil mapping units having significant areas of alpine cover.
36
Table 6. Soil associations (USDA-SCS, 1978) with significant
irrigated crop cover.
Mapping
Unit Association 
Symbol Description
Hectares 
of Irri­
gated 
Cover
(thousands)
%  of Map- Location of Major 
ping Unit Irrigated Cover 
in Irri- Areas 
gated Crops
Ap Entisols-Aridisols- 
Mollisols: Nearly
level to strongly 
sloping soils on low 
terraces, fans, and 
floodplains
256 15 Milk, Yellowstone 
Big Horn, and 
Clark's Fork 
Rivers
Av Entisols-Inceptisols- 
Mollisols: Nearly level 
to strongly sloping 
soils on low terraces 
fans, and floodplains
194 43 Beaverhead, 
Bitterroot, 
Deerlodge, 
Gallatin, Flat- 
head arid Town­
send Valleys
Ct Mollisols nearly level 
to moderately sloping 
outwash terraces, fans, 
and benches
98 55 Bitterroot,
Flathead
Valleys
Kb Aridisols-Mollisols: 
Nearly level to moder­
ately steep fans, 
terraces and benches
88 15 Bitterroot,
Flathead
Valleys
Dt Mollisols-Alfisols: 
Nearly level to strongly 
sloping soils on lacu­
strine terraces
52 - 34 Flathead Valley
Tb Aridisols and Mollisols: 
Nearly level to steep 
sloping soils on fans, 
benches, and terraces
52 18 Yellowstone 
River between 
Billings and 
Hysham
Pb Mollisols and Aridisols: .48 
Nearly level to steep 
fans, terraces and benches
26 Fairfield Bench
B Mollisols: Nearly level 
soils in cold wet basins
48 56 Big Hole and 
Centenial Valley
37
Table 7. Soil associations (USDA-SCS, 1978) with significant 
alpine or rock outcrop cover.
Mapping
Unit
Symbol
Association
Description
Hectares % of Map- 
of Alpine ping Unit 
and Rock in Alpine 
Outcrop Rock Out- 
Cover crops
(thousands)
Location of Major 
Alpine and Rock 
Outcrop Cover 
Acreage
R Mountain peaks and 433
gently sloping to very 
steep alpine grasslands
56 High mountains 
of western 
Montana
Me Inceptisols-Alfisols: 
Moderately sloping 
to very steep soils 
on mountains
146 5 High mountains of 
western Montana
Mf Inceptisols: 
Moderately sloping 
to very steep soils 
on mountains
100 4 Northwestern
Montana
Surface water locations are scattered throughout the. state 
while the recorded urban areas correspond to those cities whose 
populations exceeded 20,000 in the 1970 census. These cities are: 
Billings, Bozeman, Butte, Great Falls, Helena and Missoula.
38
Comparison with Existing Montana land Use Data 
Although not directly comparable, the AREAS tabulations of 
land cover are similar to previously existing information on 
Montana land use. The USDA, ESCS (Frey, 1974) report on major 
land use in the United States presents similar acreages (Table 8) 
for land under crops, range and forest uses.
Table 8. USDA, ESCS land use statistics for Montana (Frey, 1974).
Land
Use Hectares
(thousands)
Acres
Percent 
of total
Cropland 6,489 16,021 17.19
Grassland, 
Pasture, & 
Range 20,633 49,465 53.09
Forest .8,059 19,899 21.36
Special uses 1,876 4,633 4.97
Other uses 1,279 3,158 3.39
Total 38,336 93,176 100.00
39
The lower percentage of forestland listed in the USDA, ESCS 
report compared to the AREAS figure is partially due to inclusion 
of some forest cover areas in the special uses category. The 
special uses classification includes land used for recreational 
activities which embraces a significant amount of forest cover 
lands, reducing, the forest use area total in the USDA» ESCS 
figures.
The larger USDA, ESCS figure for land in range may be the 
result of the inclusion of grazeable woodland areas in this cate­
gory. This would also contribute to the lower areal figure for 
forestland compared to the AREAS figures.
The Montana Situation Statement's (Anonymous, 1979a) tabula­
tions (Table 9) for lands under irrigation and forest compare 
favorably with the AREAS computations. However, the report's 
figures for surface waters is noticeably larger than that obtained 
by AREAS. The lower AREAS figure for surface waters is likely due
to the method used to encode the original hand-drafted map into
2the computer mapping system. The large cell size (21.3 km ) used 
to encode the mapped land cover data precluded all but the larger 
water bodies from being encoded. Another method for measuring 
surface water areas (i.e., planimeter or grid dots) may produce 
more compatible results.
40
Table 9. Montana Situation Statement's (Anonymous, 1979a) areal
tabulations for water, range, forest and irrigated crops.
Land Percent of
Use Hectares Acres Montana's
(thousands) total area
Range 21,870 54,000 57
Forest 9,392 23,189 25
Irrigated
Crops 1,053 2,600 3
Water 398 982 I
Land Cover Map Classification Accuracy 
The percentage agreement between the state land cover map and 
the county land use maps used as ground truth was determined and 
is shown in Table 10. The overall agreement for all land cover 
categories mapped was 90%. This figure, assuming the county land 
use data are accurate, is well above the 85% minimum accuracy 
level recommended by Anderson (1976) for land cover surveys using 
remote sensing data.
It should be emphasized that interpretation errors detected 
by the ground truth procedure demonstrate only a lack of agreement
41
Table 10. Map accuracy by land cover type.
Cover
Type
Number
of
Samples
Total
Correct
Samples
Percent
Samples
Correct
Irrigated Crops 70 61 87
Dryland Crops 70 62 89
Forest 70 65 93
Range 70 ‘ 66 94
Alpine and 
Rock Outcrops 70 58 83
Water 25 25 100
Total 375 337 90
between the land cover map and the county land use maps. It is 
not known whether an error's source is within the state land cover 
map or the county land use maps. Therefore, the accuracy figures 
in this text show only percent agreement between the maps.
Only one individual category, alpine and rock outcrops, fell 
below the 85% mark. Table 11 shows a tendency to confuse forest 
with alpine areas, resulting from an inability to always success­
fully delineate the treeline. This problem was particularly acute 
in the Absaroka and Beartooth Ranges southeast of Bozeman, one of
42
the major ground truth areas for the alpine classification. The 
Landsat image used to interpret this area was dated September 30 
when snow was covering the higher elevations. This created a 
masking effect which made it difficult to discern the treeline 
causing forest to be mapped incorrectly as alpine cover.
A tendency to confuse range with irrigated crops is also 
demonstrated in Table 11. Reexamination of the satellite imagery 
revealed that the majority of these errors were the result of 
mistaking moist range sites for irrigated crops, especially when 
the range sites were interspersed with irrigated lands.
Many of the land cover map errors resulting in misidentifica- 
tion of range as dryland crops originated as location errors. 
Locational errors occurred wjien dryland crop fields were not 
positioned at the same coordinates on the land cover map as on the 
maps used as ground truth. These locational misalignments pre­
vented the verification of some dryland areas, resulting in class­
ification errors. Also contributing to the classification errors 
between range and dryland crops was the inclusion of small range 
areas in the dryland mapping unit. When these range inclusions, 
mapped as dryland crops, were checked against the ground truth 
they were declared as errors.
43
Table 11. Sample point distribution by land cover categories.
Interpreted
Data
GROUND
Irrigated Dryland 
Crops Crops
TRUTH DATA 
Forest Range
Alpine Sc. 
Rock Out­
crops Water
Irrigated
Crops
61 I I 7
Dryland
Crops
62 - 8 - -
Forest - - 65 5 - -
Range I I 2 66 - -
Alpine & Rock 
Outcrops
- - 12 - 58 -
Water - - - - - 25
Only minor problems were encountered in distinguishing 
dryland from irrigated crops. Initially there was concern over 
the ability to discriminate between dryland and irrigated crops on 
satellite imagery from a single date. However, only once during 
the ground truth process were irrigated and dryland crops confused. 
This occurred when a small area of dryland crops, incorporated
44
into an area containing irrigated crops, was mapped as irrigated 
land. Subsequently, the ground truth procedure exposed this 
inclusion and it was counted as an interpretation error.
Similar to the dryland-irrigated error described above, the 
inclusion of small clearings (range cover) in the forest mapping 
unit resulted in conflicts with the ground truth data. However, 
this lumping of forest areas was necessary to avoid overcrowding 
the small scale land cover map with excessive detail.
Range was the most accurately interpreted land cover type. 
Errors in the range mapping unit were evenly distributed between 
the major cover types with no significant bias. The high accuracy 
levels of range and forest cover categories are in part a reflec­
tion of the relatively large land area they occupy.
Cost of land Cover Map Development
Development of the 1:1,000,000 Montana land cover map from
manual interpretation of Landsat imagery required 169 man-hours of
2
labor and cost approximately $0.01/km . Included in this figure 
are expenditures for imagery acquisition, interpretation, mapping, 
materials, and ground truth expenses (Table 12).
45
Table 12. Cost of producing I:1,000,000 Montana land cover map 
from manual interpretation of Landsat imagery.
Imagery 35 color 
composites
$ 852
Interpretation and mapping* 150. hours 1,050
Materials 20
Maps used for ground truth 4
Ground truthing* 19 hours 133
Total $2,059
*Based on a pay rate of $7;00/hour.
Comparison with Other Mapping Projects and Methods 
The cost incurred in the production of the Montana land cover 
map compares favorably with similar mapping projects. A pre­
liminary 1:250,000 scale land use map of Iowa was drafted for
9
about $0.03/km involving 90 man-hours of Landsat image inter-
2pretation (Anderson, 1977). Higher costs, $0.10/km , were ex­
perienced in the construction of a 1:250,000 Landsat derived land 
use map for Boone County, Missouri. This increase in expenses 
reflects the entire Landsat image cost being charged to the area 
covered only by Boone County. Since 5 to 10 counties are usually
46
represented on a single image, the price per unit area for mapping 
only Boone County was inflated (Elfrits, et al., 1978).
Direct transfer of data manually interpreted from Landsat 
imagery was used to produce land cover maps at 1:250,000 and - 
1:150,000 for New York state. The imagery was enhanced by a diazo 
process and the maps produced at a cost of $0.39/km (Hardy, et 
al., 1973). Similar diazo enhanced imagery obtained from Cornell 
University proved to have little additional value over standard 
color composite imagery for delineating land cover in Montana. Use 
of the diazo processed imagery was therefore not deemed worth the 
added expense for visually interpreting land cover in Montana.
The expense of producing the state land cover map from Landsat 
imagery is a small fraction of the construction cost for a similar 
map using conventional means. Conventional aerial photo coverage 
is available for all of Montana at various dates, scales and film 
types. The National Cartographic Information Center (NCIC) in 
Heston, Virginia provides an Aerial Photography Summary Record 
System (APSRS) listing the types of air photo coverage for Montana 
that may be purchased.
Statewide land cover mapping would require the interpretation 
of an estimated 4,500; 1:40,000 scale, air photos at a minimum 
cost of $9,000 for the photos alone. The use of smaller scale
47
photos would reduce the photography and interpretation expenses 
but the cost would still exceed that of a survey employing Landsat 
imagery as the data source. These findings support the conclu­
sions demonstrated by Mertz (Watkins, 1978) that vegetational land 
cover maps can be obtained from Landsat data for about 10% of the 
cost of conventional survey methods. The reduced cost makes 
Landsat imagery an attractive alternative to conventional survey 
methods for mapping land cover at small scales over large areas.
CONCLUSIONS
Manual interpretation of Landsat imagery proved an accurate
and economical method of mapping land cover statewide in Montana.
2
Its low cost ($0.01/kilometer ) and accuracy (90%) make Landsat a 
viable alternative to conventional aerial photo and ground surveys. 
Landsat imagery is particularly suited for small scale regional 
mapping projects such as state and county inventories.
Repetitive coverage and low analysis costs make manual inter­
pretation of Landsat imagery practical for updating existing land 
cover information. Additionally, statewide land cover data inter­
preted from Landsat imagery was easily entered into the Montana 
Computer Graphic System for comparison with soil and other envi­
ronmental information.
Data timeliness and image quality are two problems associated 
with Landsat imagery. The satellites' passover dates for a speci­
fic. area are predetermined, thus prohibiting the researchers from 
defining the dates of acquisition. . Further limitations involve 
image quality and cloudiness. Poor image quality hampers inter­
pretation and cloudiness may render an image entirely useless. 
Fortunately, the semi-arid climate of Montana provides much cloud- 
free weather, and cloudiness was not a major problem in this 
Study. However, careful imagery prescreening for date, image
49
quality and cloudiness is important for any land cover mapping 
involving Landsat imagery. For Montana, imagery acquired during 
late July through mid-September proved the most useful for inter­
pretation of the land cover categories mapped in this project.
The Landsat project's future should be of major concern to 
researchers interested in using Landsat imagery for impending pro­
jects. At present, the Landsat series is an experimental project 
only. Plans to give Landsat operational status by turning part or 
complete control of the project to commercial interests have not 
been worked out (Anonymousj 1979b). Landsat1s usefulness in 
natural resource inventories makes it imperative that these 
problems be resolved soon to insure continuity of Landsat data.
LITERATURE CITED
LITERATURE CITED
Alexander, Robert H. 1974. An experimental regional information 
system using ERTS data, p . 505-522. In S. C. Freden, E . P. 
Mercanti and D. B. Friedman (ed.) Proc. Third ERTS-I Symp., 
Washington, DC. 10-14 Dec. 1973. NASA SP-356, Scientific and 
Technical Information Office, Washington, DC.
Alexander, Robert H. 1973. Land use classification and change 
analysis using ERTS-I imagery in CARETS. p. 923-930. In S.
C. Freden, E. P. Mercanti and M. H. Becker (ed.) Proc. Symp. 
on Significant Results from ERTS-I, New Carrollton, MD. 5-9 
Mar. 1973. NASA SP-327, Scientific and Technical Information 
Office, Washington, DC.
Anderson, James R., Ernest E. Hardy and John T. Roach. 1976. A 
land use and land cover classification system for use with 
remote sensor data. USGS Circ. 964,
Anderson, Raymond R. 1977. Production of a land use map in Iowa 
through manual interpretation of Landsat imagery, p. 827-836. 
In Proc. Ilth Int. Symp. on Remote Sensing of Environ., Ann 
Arbor, MI. 25-29 Apr. 1977. Environ. Res. Inst, of Mich., Ann 
Arbor, MI.
Anonymous. 1979a. Montana situation statement. USDA State Commit 
tee for Rural Development.
Anonymous. 1979b. U. S. Senate ponders operational Landsat.
Plain Brown Wrapper. Ames Res. Center, Moffet Field, CA.
2(1):1-4.
Anonymous. 1977. Data analysis for resource management: Supple­
mentary documentation on digital image acquisition, appli­
cations, analysis, and statistics, ESL-TM 748. ESL Inc., 
Sunnyvale, CA.
Bauer, Marvin E., Marilyn M. Hixon, Barbara J. Davis and Jeanne 
B. Etheridge. 1978. Area estimation of crops by digital 
analysis of Landsat data, Photogram. Eng. Remote Sensing, 
44(8):1033-1043.
52
Breckenridge, Ray M., Ronald W. Marrs and Donald J . Murphy. 1973 
Remote sensing applied to land use studies in Wyoming, p. 
981-989. In S. C. Freden, E . P. Mercanti and M. H. Becker 
(ed.) Proc. Symp. on Significant Results from ERTS-I, New 
Carrbllton, MD. 5-9 Mar. 1973. NASA SP-327, Scientific and 
Technical Information Office, Washington, DC.
Brown, Dwight, James Gamble, Steven Prestin, Dale Tripper, Merle 
Meyer, Joseph Ulliman and Ralph Eller. 1973. ERTS-I appli­
cations to Minnesota land use mapping, p. 991-996. In S. C. 
Freden, E. P. Mercanti and M. H. Becker (ed.) Proc. Symp. on 
Significant Results from ERTS-I, New Carrollton, MD. 5-9 Mar. 
1973. NASA SP-327, Scientific and Technical Information 
Office, Washington, DC.
Burley, Terrence M. 1961. Land use or land utilization? Prof. 
Geogr. 13(6):18-20.
Cox, T. L. 1977. Integration of land use data and soil survey 
data. Photogramm. Eng. Remote Sensing. 43(9):1127-1133.
Dornbach, J. E. and G. E. McKain. 1974. The utility of ERTS-I
data for applications in land use classification, p. 439-455. 
In S. C. Freden, E . P. Mercanti and D. B. Friedman (ed.) Proc. 
Third ERTS-I Symp., Washington, DC. 10-14 Dec. 1973. NASA 
SP-356, Scientific and Technical Information Office, 
Washington, DC.
Draeger, W. C., J. D. Nichols, A. S. Benson, D. G. Larrabee, W. M. 
Senkus and C. M. Hay. 1974. Regional agricultural surveys 
using ERTS-I data. p. 117-125. In S. C. Freden, E. P. 
Mercanti and D. B. Friedman (ed:) Proc. Third ERTS-I Symp., 
Washington, DC. 10-14 Dec. 1973. NASA SP-356, Scientific and 
Technical Information Office, Washington, DC.
Elifrits, C. Dale, Terry W. Barney, David J. Barr and C. J.
Johannsen. 1978. Mapping land cover from satellite images: A
basic low cost approach. NASA Contractor Report 2952.
Estes, J. E., R. R. Thaman, and Leslie W. Senger. 1974. Appli­
cation of ERTS-I satellite imagery for land use mapping and 
resource inventories in the central coastal region of 
California, p. 457-490. In S. C; Freden, E . P. Mercanti and 
D. B. Friedman (ed.). Proc. Third ERTS-I Symp., Washington, 
DC. 10-14 Dec. 1973. . NASA SP-356, Scientific and Technical. 
Information Office, Washington, DC.
53
Fields, Robert W. and Richard M. Petkewich. 1971. Tertiary
stratigraphy and geologic history of the upper Jefferson, Ruby, 
lower Beaverhead and lower Big Hole river valleys, p. 71-77.
In Robert W. Fields (ed.) Geology of western Montana. Univ. 
Montana, Missoula.
Ford, G. L. 1978. A computer graphic system for interactive 
display of land qualities and potential. Ph.D. thesis,
-Montana State University, Bozeman, MT.
Frey, T. H. 1974. Major uses of land in the United States: 1974. 
USDA, ESCS Agricultural Economics Rep. No. 440.
Gonzoles, L. and J. Y. Sos. 1974. ERTS operations and data pro­
cessing. J. Environ. Sci. Mar./Apr.:8-14.
Halpern, Jack A., Lawrence D. Alexander, and Dennis M. O'Regan.
1975. The application of remote sensing data to geographic- 
based information systems, p. 351-358. In Proc. IOth Int.
Symp. Remote Sensing Environ., Ann Arbor, MI. 6-10 Oct. 1975. 
Environ. Res. Inst, of Mich., Ann Arbor, MI.
Hanuschak, G. R., M. Craig Sigman, M. Ozga, R. Luebbe, P. Cook,
D. Kleweno and C. Miller. 1979. Obtaining timely crop area 
estimates using ground-gathered and Landsat data. USDA, ESCS 
Tech. Bull. 1609.
Hardy, E. E. and R. L. Shelton. 1970. Inventorying New York's 
land use and natural resources. Food and Life Sci. Qtrly.
3(4):4-7.
Hardy, E. E., J. E. Skaley, and E. S. Phillips. 1974. Evaluation 
of ERTS-I imagery for land use/resource inventory information, 
p. 491-495. In S. C. Freden, E. P. Mercanti and D. B. Friedman 
(ed.) Proc. Third ERTS-I Symp., Washington, DC. 10-14 Dec. 
1973. NASA SP-356, Scientific and Technical Information 
Office, Washington, DC.
Hay, Alan. M. 1979. Sampling designs to test land use map
accuracy. Photogramm. Eng. Remote Sensing. 45(4):529-533.
Heller, Robert C. and Kim A. Johnson. 1979. Estimating irrigated 
land acreage from Landsat imagery. Photogramm. Eng. Remote 
Sensing. 45(10):1379-1386.
54
Hill-Rowley, Richard and William R. EnsIin. 1979. An evaluation 
of Michigan land cover/use inventories derived from remote 
sensing: Characteristics and costs, p . 1251-1259. In Proc.
13th Int. Symp. on Remote Sensing of Environ., Ann Arbor, MI. 
23-27 Apr. 1979. Environ. Res. Inst, of Mich., Ann Arbor, MI.
Jensen, J. R., L. R. Tinney, J. E. Estes. 1975. An analysis of 
the accuracy and cost-effectiveness of a cropland inventory 
utilizing remote sensing techniques. p. 1149-1158. In Proc. 
IOth Int. Symp. on Remote Sensing of Environ., Ann Arbor, MI. 
6-10 Oct. 1975. Environ. Res. Inst, of Mich., Ann Arbor , MI.
Kroeck, Dick. 1976. Everyone's space handbook. Pilot Rock Inc., 
Areata, CA.
Lins, Harvey F. 1976. Land use mapping from Skylab S-190B photo­
graphy. Photogramm. Eng. Remote Sensing. 42(3):301-307.
Place, John L. 1974. Change in land use in the Phoenix
(1:250,000) quadangle, Arizona between 1970 and 1973: ERTS as
an aid in a nationwide program for mapping general land use. 
p . 393-423. In S. C. Freden, E . P. Mercanti and D. B. Friedman 
(ed.) Proc. Third ERTS-I Symp., Washington, DC. 10-14 Dec. 
1973. NASA SP-356, Scientific and Technical Information Office, 
Washington, DC.
Richardson, A. J., C. L. Wiegand, R. J. Torline, M. R. Gautreaux,
1977. Landsat agricultural land use survey. PhotOgramm. Ehg. 
Remote Sensing. 43(2)=207-216.
Ross, R. L. and H. E. Hunter, 1976. Climax vegetation of Montana 
based on soils and climate. USDA-SCS. Bozeman.
Seevers, P. M. and Rex M. Peterson. 1978. Application of remote 
sensing in agricultural analysis, p. 259-269. In B. F. 
Richason (ed.) Introduction to remote sensing of the environ­
ment. Kendall/Hunt Co., Dubuque, IA.
Shelton, Ronald L. and John E. Estes. 1979. Integration of
remote sensing and geographic information systems. p. 463-483. 
In Proc. 13th Int. Symp. on Remote Sensing of Environ., Ann 
Arbor, MI. 23-27 Apr. 1979. Environ. Res. Inst; of Mich.,
Ann Arbor, MI.
55
Shelton, Ronald L. and Stephen E. Tillman. 1978. Remote
sensing, geographic information systems, and national land 
planning: Some Central American and Caribbean experiences,
p. 587-585. Jn Proc. 12th Int. Symp. on Remote Sensing of 
Environ., Ann Arbor, MI. 20-26 Apr. 1978. Environ. Res. Inst, 
of Mich., Ann Arbor, MI.
Sizer, J. E. 1974. ERTS-I role in land management and planning 
in Minnesota, p. 341-350. S. C. Freden, E . P. Mercanti, and 
D. B. Friedman (ed.) Proc. Third ERTS-I Symp., Washington, DC. 
10-14 Dec. 1973. NASA SP-356, Scientific and Technical Infor­
mation Office, Washington, DC.
Thrower, Norman J. W. 1970. Land use in the southwestern United 
States from Gemini and Apollo imagery (map supplement No. 12). 
Annals Assoc. Amer. Geographers 60(1).
USDI-BLM. 1969. Public land statistics.
USDA-SCS. 1978. General soil map Montana. Montana State Univer­
sity, Ext. Misc. Pub. No. 16.
Veseth, Roger and Clifford Montagne. 1980. Geologic parent
materials of Montana soils. Montana Agric. Exp. Stn. and 
USDA-SCS Bull. 721.
Welch, R., H. C. Lao, and C. W. Pannell. 1979. Mapping China's
new agricultural lands. Photogramm. Eng. Remote Sensing.
45(9):1211-1228.
Watkins, Thayer. 1978. The economics of remote sensing.
Photogramm. Eng. Remote Sensing. 44(9):1167-1172.
APPENDICES
APPENDIX I
INDEX OF LANDSAT IMAGERY USED TO PREPARE THE MONTANA
LAND COVER MAP
58
Table I-I. Index of Landsat imagery used to prepare the Montana 
land cover map.
Index Identification
No. Area Date No.
I Libby, MT 10/5/72 1074-18082
2 Whitefish, MT 7/14/74 1721-17544
3 Thompson Falls, MT 9/11/73 1415-18022
4 Glacier National Park, MT 7/13/74 1720-17490
5 Flathead Lake, Missoula, MT 7/18/73 1360-17572
6 Bitterroot Valley, MT 7/13/74 1720-17495
7 Cutbank - Shelby, MT 6/11/73 1323-17514
8 Choteau, MT 8/27/72 1035-17513
9 Bittte, MT . 7/17/73 1359-17521
10 Clark Canyon Reservoir, MT 9/22/74 1791-17415
11 Hill County, MT 9/13/72 1052-17452
12 Great Falls, MT 9/13/72 1052-17454
13 Bozeman, MT 9/13/72 1052-17461
14 . West Yellowstone, MT 9/21/74 1790-17361
15 Havre, MT 7/15/73 1357-17395
16 Lewistown, MT 7/15/73 1357-17401
17 Big Timber, MT 7/15/73 1357-17404
18 Gardiner, MT 9/30/72 1069-17405
19 Malta, MT 9/6/73 1410-17332
20 Petroleum County, MT 9/6/73 1410-17334
21 Billings,MT 9/6/73 1410-17341
22 Cooke City, MT 8/1/73 1374-17351
23 Glasgow, MT 9/5/72 1409-17273
24 Ft. Peck, MT 9/5/73 1409-17280
25 Hardin, MT 9/5/73 1409-17282
26 Lodgegrass, MT 9/5/73 1409-17285
27 Scobey, MT 5/3/76 5380-16402
28 Circle, MT 8/17/73 1390-17223
29 Miles City, MT 8/17/73 1390-17230
30 Decker, MT 5/19/73 1300-17243
31 Plentywood,MT 5/2/76 5379-16344
32 Sidney, MT 5/2/76 5379-16351
33 Ekalaka, MT 5/2/76 5379-16353
34 Broadus, MT 6/18/74 1695-17110
35 Thompson Falls, MT - 
Spokane, WA 10/5/72 1074-18085
APPENDIX II
SOURCES OF REMOTE SENSING IMAGERY
SOURCES OF REMOTE SENSING IMAGERY
The purpose of this appendix is to serve as a user guide to 
selecting and ordering Landsat and other types of remote sensing 
imagery.
APPENDIX II
Federal Sources of Remote Sensing Imagery 
EROS Data Center
A d m i n i s t e r e d by the Geological Survey of the Department of the
Interior, the EROS Data Center (EDC) contains the largest single
federal collection of. satellite, and aircraft imagery. In addition
to a large collection of aerial mapping photography; data from the
Landsat, Skylab, Apollo and Gemini space programs are available
from the EDC. To order or inquire about available imagery or
assistance the prospective user is instructed to contact the EDC by
phone, letter or personal visit:
User Services Unit
EROS Data Center
Sioux Falls, South Dakota 57198
Phone: 605/594-6511
By submitting to EROS the geographic location of the areas of 
interest, a user may institute a computer search listing available 
imagery for that area. The user should also specify the dates and
61
seasons, acceptable cloud cover, and image format desired. 
Geographic computer inquiry forms are available.upon request from 
the EDC.
There are a variety of data types available from EROS.
Landsat images are available in color or black and white trans­
parencies ranging in size from 16 millimeters to 40 x 40 inches and 
scale from 1:3,369,000 to 1:250,000. The color composite, produced 
by combining three of Landsat's MSS bands and exposing them through 
color filters to produce a color image is one popular product. Also 
available are computer compatible tapes (CCTs) containing the 
digital information for each Landsat scene. The CCTs are available 
in nine track tapes having either 800 or 1600 bites/inch.
Imagery from the Skylab, Apollo and Gemini space programs and 
NASA research aircraft is available in color or black and white 
prints and transparencies from 16 millimeters to 40 inches on a 
side. A limited number of CCTs obtained from sensors used in the 
manned space programs may also be purchased. Conventional aerial 
mapping photography may be procured in either black and white 
prints or transparencies.
Prices for standard products obtained from EROS are included 
with the order forms. Prices are generally comparable with costs 
of obtaining the same product from other federal sources. Prices
62
range from around $3.00 for a 9 x 9  inch black and white print to 
$50.00 for a 40 x 40 inch color print and $200.00 for a CCT of all 
four Landsat bands. Custom processing to special scales, formats, 
and rapid delivery orders are available but pricing is increased 
accordingly. If a Landsat scene desired by a user previously had a 
color composite produced then a color composite may be purchased at 
the standard product price. However, if a color composite has not 
been produced then there is a substantial service charge added to 
the product cost.
All orders to EROS must be accompanied by check, money order, 
purchase order or an authorized account identification. A standing 
account at EROS may be.opened with a minimum deposit of $100.
USDA-ASCS
The ASCS division of the USDA has Landsat and Skylab imagery 
available in addition to USDA photography for about 80% of the U.S. 
excluding Alaska. Standard product prices are comparable to those 
at the EROS Data.Center. Air photo coverage is updated approxi­
mately every seven years with the majority of the coverage at 
1:20,000 (8 1/2 inch lens) and new photography at 1:40,000 (6 inch 
lens). However, it is possible to request photo products at other 
specified scales.
63
To obtain NASA aerial photography, Landsat and Skylab imagery
the user is instructed to contact:
Aerial Photography Field Office 
ASCS-USDA
2222 West 2300 South 
P. 0. Box 30010 
Salt Lake City, Utah 84125 
Phone: 801/524-5856
\
Orders for USDA aerial photography can be initiated by contacting 
the local ASCS office or by contacting the Aerial Photography Field 
Office in Salt Lake City.
To obtain a listing of aerial photography available for a 
specific state the user should contact the National Cartographic 
Information Center (NCIC) and request an Aerial Photography Summary
Y
Record System booklet for the state of interest. Each booklet is
updated twice a year and may be obtained from:
National Cartographic Information Center 
507 National Center 
Reston, Virginia 22092 
Phone: 703/860-6045
■■ ■
Other Sources of Remote Sensing Imagery
Integrated Satellite Information Services Ltd. (ISIS) is a 
private Canadian enterprise providing quick processing and delivery 
of Landsat imagery. ISIS has available almost total coverage of 
the United States and Canada. The firm promotes its ability to 
produce imagery within 2 hours of receiving the data and provide 
delivery within 24 to 48 hours of data reception.
64
A daily microfiche record (ISISFICHE) of all the Landsat 
images received in Canada may be obtained by subscription. Stan­
dard image products from ISIS include either 8 x 10 ipch or 70 x 70 
millimeter prints or transparencies (Kroeck, 1976):
Integrated Satellite Information Services Ltd.
P. 0. Box 1630 
Prince Albert, Saskatchewan 
Canada S6V 5T2 
Phone: 306/764-3602
764-4259
Additional sources of aerial photography include state 
agencies such as the Department of Highways and the Department of 
Natural Resources. Often privately owned local aerial photography 
firms also have their own photo depositories. If they have already 
flown the area of interest the user can usually arrange to purchase 
copies of the desired.photos.
APPENDIX III
MONTANA SOIL ASSOCIATION AND LAND COVER COMBINATIONS
66
Table III-I. Montana Soil Association (USDA-SCS, 1978) and Land 
Cover Combinations /
Soil Land. Use Hectares . Soil Land Use Hectares
B • W 2083 . JF R 141671
B R 31251 JF I . 2083
B I 47918 JF F 33334
B F ' 4147 JF D 125.00
R R 4167. KF W 6250
. R M 433346 KF R 81252
R I 2083 KF F 22917
R F 337510 KF D 14595
X W 10417 KT R 45835
X R 472930 KT . I 18751
X I 2083 KT D 8334
X F 183339
X D 16667 LT R 2083
LT. F . 29168
AW W 2083
AW R . . 18751 MA ' W 4167
AW I 2083 MA R 4167
AW D 33334 MA M 43751
MA . F . 827107
ET R 4167
ET F . . 58335 . OH R 183339
OH D 185422
FG R 8334 • .
FG . I 4167 OT R 25001
OT D 237507
HF R 2083
SH R . 100003
IG R . 87502 SH D 2083
IG ■ F . 25001
. ST R . 22917
JB R 100003 ST I 2083
JB I .. 29167 ST D . 120837
JB F 10417
JB D ■ 8334
67
Table III-I. . (continued)
Soil Land Use Hectares Soil Land Use Hectares
API R 179172 AV3 R 25001
API I 75002 AV3 I 20834
API D 27084 AV3 F ' 29167
AP2 R 479180 AV4 U 2083
AP2 I 100003 AV4 R 64585
AP2 F .18751 AV4 I 35418
AP2 D 27084 AV4 F 4167
AV4 D 6250
AP3 R .258341
AP3 I 2083 AV5 R 27084
AP3 F 2083 AV5 I 8334
AP3 D 27084 AV5 D 6250
AP4 R 137504 CTl R 31251
AP4 I 47918 CTl M 2083
AP4 D 62502 CTl I . 72919
CTl F 37501
AP5 W 2083 CTl D 2083
AP5 R 27918
AP5 I 4167 CT2 R 6250
AP5 D 79169 CT2 I 25001
AP6 R 29167 DTl ' U ' 4167
AP6 I 6250 DTl R 2083
AP6 F 2083 DTl I 12500
DTl F 79169
AP7 R 41668 DTl D 2683
AP7 I . 20834
AP7 D 41668 DT2 R 10417
DT2 I 39585
AVI R 52085 DT2 F 2083
AVI I 79169
Avl F 6250 GBl R 20834
Avl D 6250 GBl I 20834
GBl D 60418
AV2 R 8334
AV2 I 52085
AV2 F 22917
68
Table III-I. (continued)
Soil Land Use Hectares Soil Land Use Hectares
GB2 R 227090 GF5 R 175005
GB2 I 29168 GF5. F 2083
GB2 F 20834 GF5 b 2083
GB2 D 27084
GGl R 16667
GB3 R 29168 GGl I 2083
GB3 I 4167 GGl F 12500
GB3 F 12500
GG2. W 2083
GB4 R 97919 GG2 R 112503
GB4 I 12500 GG2 D 4167
GB4 F 4167
GB4 D 45835 HFl R 218756
HFl I 18751
GB5 R 306259 HFl F 79169
GB5 I 50001 HFl D 6250
GB5 F 4167
GB5 D 277091 HF2 . R - 183339
HF2 I 2083
GB6 . W 2083 HF2 F 60418
GB6 R 52085 HF2 D 10417
GB6 I 4166
GB6 . D 8334 HF3 R 193756
HF3 I 2083
GFl R 54168 HF3 F 41668
GFl F 16667 HF 3 D 2083
GFl D 4167
HF4 R 181255
GF2 R 70835 HF4 F 2083
GF2 I 2083 HF4 D 2083
GF2 b 4167
H42 R 4167
GF3 R 91669
GF3 F 10417 IBl R . 41668
IBl I 8334
GF4 R 208339 IBl F 2083
GF 4 D 6250 IBl D 2083
69
Table III-I. (continued)
Soil Land Use Hectares Soil Land Use Hectares
IB2 R . 129170
IB2 I 43751
IB2 F . 25001
IB3 R 77086
IB3 D 4167
IFl R 60418
IFl F 52085
IFl D 6250
IF2 R 216673
IF2 F 50001
IF3 R 485430
IF3 I 4167
IF3 F 125004
IF3 D 6250
IF4 R 58335
IF4 I 2083
IF4 F 64585
KBl W 10417
KBl U 4167
KBl R 233340
KBl I 50001
KBl F ' 6250
KBl D 58335
KB2 R 118753
KB2 I 37501
KB2 D 20834
KB3 R 35418
KB4 R 297925
KB4 F 6250
KB4 D 18750
LGl . W 8334
LGl R 43751
LGl I 10417
LGl F 662519
LG2 R 10417
LG2 F 62503
LG3 R 2083
LG3 F 62502
MEl W 10417
MEl R 18751
MEl M 56252
MEl F 997945
ME2 R 2083
ME2 M 2083
ME2 I 2083
ME2 F 179172
ME3 R 29168
ME3 M 12500
ME3 F 618768
ME4 R 37501
ME4 N 2083
ME4 M 43751
ME4 F 862525
ME5 M 31251
ME5 F 152088
MFl W 8334
MFl R 31251
MFl N 2083
MFl M 4167
MFl M 4167
MFl F 1204200
70
Table III-I. (continued)
Soil Land Use Hectares Soil Land Use Hectares
MF2 R 22917 NB2 U 2083
MF2 M 60418 NB2 R 6250
MF2 I 4167 NB2 I 2083 .
MF2 F 291675 NB2 D 12500
MF3 R 12500 NHl R 102086
MF 3 M 35418 NHl I 4167
MF3 I 2083 NHl D 110420
MF3 F 447929
NH2 R 181396
MF 4 R 4167 NH2 F 148036
MF4 F 100003 NH2 D 8340
MOl R 47918 NH3 R 250007
MOl N 4167 NH3 F 4167
MOl F 127087 NH3 D 16667
M02 R 106253 NH4 R 114587
M02 M 6250 NH4 D 139587
M02 F 354177
NH5 R 433682
M03 R 31250 NH5 I 10417
M03 F 258341 NH5 F 4167
■. NH5 D 122920
M04 R 47918
M04 M 2083 NH6 R 35418
M04 F 177088 NH6 D 4167
M05 R 33334 NPl R 85419
M05 F 43751 NPl F 10417
NPl D 27084
M06 R 31251 ..
M06 F 89586 NP2 , R 133337
NP2 I 2083
NBl R 56252 NP2 F 12500
NBl I 14584 NP2 D 18751
NBl D 60418
NP3 R 25020
71
Table III-I. (continued)
Soil Land Use Hectares Soil Land Use Hectares
NP4 R 10417 PH2 W 4167
NP4 D 20834 PH2 R 143754
PH2 I 12500
NP5 R 154171 PH2 D 100003
NP5. F . 8334
NP5 D 33334 PPl R 91669
PPl D 72919
NP6 R 139587
NP6 . D 316676 PP2 R 181255
PP2 F 8334
OGl R 2083 PP2 D 58335
OGl D 89586
SGl R 154171
0G2 W. 2083 SGl D 983361
0G2 R 233340
0G2 D 206256 SG2 W 2083
SG2 R 789606
0G3 R 187505 SG2 I 4167
0G3 I 4167 SG2 F 2083
0G3 D 202089 SG2 D 1045863
0G4 R 75002 SG3 R 229173
0G4 D 460430 SG3 D 47918
PBl R 14584 TBl R 70835
PBl I ’ 2083 TBl I 10417
PBl D 16667 TBl F 4167
TBl D 12500
PB2 I . 25001
PB2 D 10417 TB2 R 47918
TB2 I 2083
PB3 . R 41668' TB2 D 22917
PB3 I 20834
PB3 D 50001 TB3 U 6250
TB3 R 58335
PHl R 443763 TB3 I . 39584
PHl F 91669 TB3 F 4167
PHl D 6250 TB3 D 12500
72
Table III-I. (continued)
Soil Land Use Hectares Soil Land Use Hectares
THl R 756272 UGl R 506264
THl D 20833 UGl . I 2083
UGl D 108336
TH2 R 477467
TH2 F 154171 UG2 R 100003
TH2 . D 27084 UG2 F 4167
TH3 R 541682 VBl U 2083 ,
TH3 F 43751 VBl R 2083
TH3 D 4167 VBl D 27084
TH4 R 541682 VB2 W 6250
TH4 I 4167 VB2 U 2083
TH4 F 66669 VB2 R 8334
VB2 D 241674
TH5 R 139696
TH5 F 229351 VB3 R 50001
TH6 R 848599 VB4 I 2083
TH6 D 6250 VB4 D 20834
TPl R 68752 VB5 R 102086
TPl D 52085 VB5 D 31254
TP2 R 266882 . VB6 R 4167
TP2 . . F 8334 VB6 D 8334
TP2 D 8334
VHl R 1304203
TP3 R 291675 VHl F 127087
TP3 D 29168 VHl D 75002
TP4 R 152088 VH2 R 58335
TP4 F 39585 VH2 I 2083
TP4 D 8334 VH2 F 2083
VH2 D 6250
TP5 R 356260
TP5 F 14584
TP5 D 129170
73
Table III-I. (continued)
Soil Land. Use Hectares Soil Land Use , Hectares
VH3 R . 533348 WATER 120837
VH3 I 4167 WATER R 27084
VH3 F 6250 WATER . F. 2083
,VH3 D 8334 WATER . D 2083 .
Total = 37,671,904 hectares
Key to Land Use Symbols .
R = Range . M = Alpine and Rock Outcrops
F = Forest U = Urban
I = Irrigated Crops W = Water
D = Dryland Crops N =  Nonsense
APPENDIX IV
MONTANA SOIL ASSOCIATIONS' DESCRIPTIONS AND. 
PERCENT LAND COVER .
75
Montana Soil Associations' Descriptions and 
Percent Land Cover
The data presented in Table IV-I were derived from a pre­
liminary draft of a new bulletin entitled Soils of Montana by the 
National Cooperative Soil Survey in Montana. Being preliminary, 
these results are subject to change and should hot be used without
verification.
Table VI-1. Montana soil associations' descriptions and percent land cover.
Map Soil Surface _____X Land Cover* Representative
Symbol______ Great Groups________________ Landform____________________Parent Material_____ Depth_____ Texture I D F R A _______ Soil Series
Great Plains Section
Apl Torrifluvents Nearly level to strongly 
sloping terraces, fans 
floodplains
alluvium deep l.fsl, ,11 27 7 64 - Glendive, Harlem, Havre 
Bowdoin, Hanley
Ap2 Camborthids, TorrifInvents, 
Torriorthents
Nearly level to strongly 
sloping terraces, fans 
floodplains
alluvium, lacustrine 
sediments
deep I,Slcl1 Sll 16 4 3 77 - Glendive, Harlem, Havre, 
Kaber, Namoc, Busby, 
Lambeth, Lonna
Ap3 Camborthids, Natrargids, 
Torrifluvents
Nearly level to strongly 
sloping terraces, fans 
floodplains
alluvium deep l.sil.sl I 9 I 89 - Alona, Busby, Cambeth, 
Creed, Gerdrum, Harlem, 
Havre, Lonna, Yamac, 
Absher, Glendive, Hanley
Ap4 Ustochrepts, UstifInvents, 
Argiborolls
Nearly level to strongly 
sloping terraces, fans 
floodplains
alluvium deep sil.l.cl
fsl
19 25 55 - Cambert, Cherry, Farland, 
Farnuf, Havrelon, Lohler, 
Trembles, Banks, Hoffman- 
ville, Macar, Ridgelawn, 
Savage
ApS Ustifluvents, Haploboralls, 
Argiborqlls
Nearly level to strongly 
sloping terraces, fans, 
floodplains
alluvium deep I, sil,sicl, 3 59 36 - Farland, F a m u f , Havrelon, 
Lohler, Shambo, Trembles, 
Wabek
Ap6 Haploborolls Nearly level to strongly 
sloping terraces, fans, 
floodplains
gravelly alluvium deep I.si,cl 17 6 78 - Lolo, Nesda, Shawa
Ap 7 HaploborolIs, Argiborolls, 
Ustifluvents
Level to strongly sloping 
terraces, fans, flood- 
plains
alluvium, some 
loess
deep cl,sil,sicl 20 40 40 - Farauf, Frazer, Korchea, 
Savage
Aw Haplaquepts, Torrior- 
thents, Haplaquolls
Nearly level to gently 
sloping basins, terraces, 
floodplains
clay sediments, 
clayey alluvium
deep 4 59 33 - Dimmick, Marvan, McKenzie, 
Nobe
*1 - Irrigated crops, D - Dryland crops, F - Forest, R - Range, A - Alpine and outcrops.
Table IV-I. (continued)
H»p
Symbol
Soil
Great Groups Parent Material
% Land Cover
Depth_____ Texture I D F R A
Representative 
Soil Series
Nbl Argiborolls Nearly level to steep 
fans, benches, terraces
alluvium deep l.d,ell 46 11 43 - Farland, Farnuf, Savage, 
Turner
Nb 2 Haploborolls Nearly level to steep 
fans, benches, terraces
alluvium, eolian
moder­
ately
deep
fsl,sil,cl 9 55 - 36 - Azaar, Lihen, Tally
Nhl Haploborolls, Calciborolls, 
Argiborolls
Strongly sloping to steep 
plains, hills
alluvium, shale, 
clayey sediments
ately
deep
cl,gl,Sicl 2 51 47 - Danvers, Lawther, Regent 
Utica, Windham, Winifred
Nh 2 Ustochrepts, Ustorthents1 
Haploborolls
Strongly sloping to steep 
plains, hills
sedimentary mater­
ials , alluvium
shallow 
to deep
I,cl,c,sic - 2 44 54 - Barkof, Barvon, Bitton, 
Cabba, Dast, Lamedeer,
Macar, Ringling, Searing, 
Wayden
Nh 3 Ustorthents, Lithic Haplo- 
borolls, Argiborolls
Strongly sloping 
plains, hills
to steep sedimentary rock 
beds, alluvium
shallow 
to deep
cl,sil,l,6ic - 6 2 92 - Absarkee, Cabba, Castrer 
Lambert, Wayden, Xavier
Nh 4 Ustorthernts, Ustochrepts Strongly sloping 
plains, hills
to steep sedimentary beds, 
alluvium, glacial 
till
deep to
ately
deep
cl,sil,sl,l - 55 45 - Cambert, Cherry, Dast, 
Dimyaw, Lambert, Macar, 
Zahill
Nh5 Ustochrepts, Ustorthents Strongly sloping 
plains, hills
to steep sedimentary rock 
beds, alluvium
deep to 
shallow
cl,sil,sic,
si
2 22 I 75 - Cabba, Cambert, Cherry, 
Dast, Dimyaw, Lambert, 
Macar
Nh 6 Ustorthents, Haploborolls Strongly sloping 
plains, hills
to steep alluvium, silt- 
stone, eolian
deep to 
shallow
1, cl,sil,sl - 11 89 - Cabbart, Chama, Dast, 
Doney, Lambert, Lihen, 
Shambo, Tinsley, Wayden
Npl Argiborolls, Ustorthents Nearly level to strongly 
sloping plains
sedimentary rock 
beds, alluvium
shallow 
to mod­
erately 
deep
cl,I,Sil - 22 8 69 - Cabba, Doney, Farnuf, 
Reeder, Wayden
Table IV-I. (continued)
Map
Great Groups Landfonn Parent Material
Soil
Depth
Surface
Texture
Np2 Argiborolls, Lithic 
Argiborolls
Nearly level to strongly 
sloping plains
sandstone, shale, 
siltstone ately 
deep to 
shallow
l.cl
Mp3 Ustorthents, Haploborolls, 
Natriborolls
Nearly level to strongly 
sloping plains
sedimentary rock 
beds, allvlum, 
eolian
ately
deep
£»1.1,cl
Np4 Calciborolls, Haploborolls Nearly level to strongly 
sloping plains
alluvium, meta- 
mo rphic rocks, 
sandstone
deep *1.1
Np5 Ustorthents, Argiborolls Nearly level to strongly 
sloping plains
sedimentary rock 
beds, alluvium ately
deep
cl,l,sicl
Np6 Haploborolls, Ustochrept, 
Ustorthents
Nearly level to strongly 
sloping plains
sedimentary rock 
beds, alluvium, 
eolian
deep to 
shallow
•11.I.cl.»1
Ogi ArgiborolIs, Haploborolla 
(loamy)
Undulating to strongly 
rolling plains
alluvium, eolian, 
glacial till
deep I,si,cl,all
OgZ Argiborolls, Haploborolls, 
(clayey)
Undulating to strongly 
rolling plains
glacial till, 
alluvium
deep cl,Sicl1I
Ogi Argiborolls, Ustorthents, 
(thin solum)
Undulating to strongly 
rolling plains
glacial till deep l.cl
Og* Argiborolla, Ustorthents, 
(thick solum)
Undulating to strongly 
rolling plains
alluvium, glacial 
till, lacustrine 
sediments
deep •11.I.Cl
X Land Cover 
I D F R A
I 11 8 80 -
— — — 100 —
— 67 — 33 —
- 17 4 79 -
— 69 - 31 —
- 9 8  — 2 —
— 47 — 53 —
I 51 - 48 -
— 86 — 14 —
Representative 
Soil Series
Absarokee, Amherst, 
Sinnigam, Borky, Darret
Adger, Cabba, Dast, Doney, 
Rhoades, Tally, Vebar1 
Badland
Bitton, Castner, Ipano, 
Twin Creek
Cabba, Doney, Farland, 
Regent, Wayden
Amor, Cambert, Chama, 
Cherry, Dast, Lambert, 
Macar, Shambo, Vebar, 
Blanchard, Cabba, Dlmyaw, 
Lehr, Tally
Dooley, Farnuf, Parshall, 
Shambo, Tally, Williams
Bearpaw, Lawther, Savage, 
Vlda
Farnuf, Vida, Williams, 
Zahill
Bowbells, Farnuf, Grail, 
Lambert, Tinsley, Williams
oo
Table IV-I. (continued)
IUp Soil Surface % Land I Representative
Symbol Great Groups Landform Parent Material Depth Texture I D F R A Soil Series
Oh Ustorthents, Arglborolls Strongly rolling to 
steep plains
glacial till. deep l.cl,ell ■ - 50 - 50 - Lambert, Vida, Williams 
Zahill
Ot Arglborolls, Haploborolls Nearly level to strongly 
sloping terraces
alluvium, glacial 
till, eolian
deep sl,I,all, 
cl
" 90 - 10 - Dooley, Farnuf, Manning, 
Tally, Turner, Wabek
Pbl Arglborolls, Torrlorthents Nearly level to steep 
fans, benches, terraces
alluvium deep I,all 6 50 - 44 - Evanston, Lambeth
Pb2 Calclborolls, Calclorthlds Nearly level to steep 
fans, terraces, benches
alluvium deep gl.el 71 29 - - - Binna, Crago, Rothiemay
Pb3 Haploborolls, Camborthlds, 
Arglborolls
Nearly level to steep 
fans, benches, terraces
alluvium, lacustrine deep 
sediments, sedimen­
tary rock beds
B i d , I,Cl, 
SlC
19 44 - 37 /bor, Dutton, Ethridge, 
Evanston, Kobar, Kremlin
Phl Torrlorthents (shallow), 
Arglborolls, Camborthlds
Strongly sloping to 
steep plains, hills
sedimentary rock 
beds, alluvium, 
lacustrine sedi­
ments
deep to 
shallow
l,sicl,sil,
cl
I 17 82 Cabbart, Cambeth, Delpoint1 
Dllts, Ethridge, Evanaton, 
Lisam, Yamac, Yawdim
Ph2 Torrlorthents (shallow), 
Llthlc Arglborolls
Strongly sloping to 
steep plains, hills
sedimentary rock 
beds, alluvium
shallow 
to deep
I,ale,cl,
al
5 38 - 55 - Cabbart, Glendive, Havre, 
Lisam, Rentsac, Yawdim
ppi Arglborolls, Haploborolls, 
Llthlc Arglborolls
Nearly level to strongly 
sloping plains
sandstone, alluvium, 
eolian
shallow 
to deep
cl,I 44 56 Beenom, Boxwell, Chinook, 
E m e n , Marmarth, Sinnigam, 
Tanna
Pp2 Arglborolls, Haplargids, 
Lithic Torriorthents
Nearly level to strongly 
sloping plains
shale, sandstone deep to 
shallow
I,»11,cl - 24 3 73 - Bonfri, Brusett, Marmarth, 
Rentsac, Tanna, Yawdlm
Sgl Argiborolls, Haploborolle Nearly level to hilly 
glaciated plains
glacial till, allu­
vium
deep I.cl,el - 86 - 14 - Chinook, Joplin, Kevin, 
Kremlim, Scobey, Telstad
Table IV-I. (continued)
Map Soil Surface % Land Representative
Symbol Great Groups Landform Parent Material Depth Texture I D F R A Soil Series
Sg2 Arglborolls, Torrlorthents, 
Natrarglds
Nearly level to hilly 
glaciated plains
glacial till deep l.cl <1 57 <1 43 - Elloam, Kevin, Phillips, 
Scobey, Sunburst,
Telstad, Thoeny
Sg3 Arglborolls, Torrlorthents, 
Natrarglds
Nearly level to hilly 
glaciated plains
glacial till deep I,Cl,C 17 - 83 Elloam, Hillon, Kevin, 
Scobey, Sunburst, 
Telstad, Thoeny
Sh Torriothents, Natrargids, 
Torrlorthents (shallow)
Strongly sloping to 
steep glaciated plains
glacial till, 
sedimentary rock 
beds
deep to 
shallow
C,1 2 - 98 Cabbart, Elloam, Lisam, 
Sunburst
St Arglborolls, Haploborolls Nearly level to moder­
ately steep outwash and 
stream terraces
alluvium, eolian, 
lacustrine sedi­
ments
deep l.sil.cl,
si
I 83 - 16 Asslnniboine, Attewan, 
Chinook, Ethridge, Evanston, 
Floweree, Kremlin
Tbl Haplargids, HaploboralIs, 
Arglborolls
Nearly level to steep 
fans, terraces, benches
alluvium, lacus­
trine sediments, 
eolian
deep l.sil.cl 11 13 4 72 Attewan, Ethridge, Evanston, 
Floweree, Fort Collins, 
Keiser, Kremlin
Tb 2 Natrargids, Haplarglds Nearly level to steeply 
sloping fans, terraces, 
benches
alluvium deep I.cl.,1.mil 3 31 - 66 Absher, Creed, Fort Collins, 
Gerdum, Reiser
Tb 3 Camborthids Nearly level to steeply 
sloping fans, terraces, 
benches
alluvium, lacus­
trine, sediments
deep I,sic,sil 35 10 3 51 Kobar, Lonna, Yamac
Thl Torriorthents, Camborthlds Strongly sloping to steep 
plains, hills
sedimentary rock 
beds, alluvium 
shale
deep I,Cl1Sil 3 - 97 Bascovy, Cambeth, Delpoint, 
Lambeth, Yamac
Th2 Torriothents (shallow) 
Camborthids, Torriorthents
Strongly sloping to 
steep plains, hills
sedimentary rock 
beds, alluvium, 
lacustrine sedi-
deep I,si,sic, 
sil
'
4 23 73
‘
Birney, Cabbart, Cambeth, 
Delpoint, Kirby, Kobar, 
Lambeth, Riedel, Twilight, 
Yamac, Yawdim
Table IV-I. (continued)
Mmp
Symbol______ Great Groups________________Landform
Soli Surface ______% Land Cover Representative
Parent Material Depth Texture I D F R A _______ Soil Series
Th 3 Torriorthents (shallow), 
Torriorthents, Camborthids
Strongly sloping 
plains, hills
to steep sedimentary rock 
beds, alluvium, 
glacial till
Th4 Torriorthents, Torriorth- 
enta (shallow), 
Camborthids
Strongly sloping 
plains, hills
to steep sedimentary bed­
rock, alluvium
ThS Torriorthents, Llthlc 
Torriorthents, Camborthids
Strongly sloping 
plains, hills
to steep sandstone, silt- 
stone, shale, 
alluvium
Th6 Torriorthents, Natrargids Strongly sloping 
plains, hills
to steep alluvium, sedi­
mentary bedrock
Tpl Camborthids, Haploborolls, 
Argiborolls
Nearly level to moderately 
steep plains
alluvium, sedi­
mentary bedrock
deep I,«11,*1,cl - I 7 92 - Busby, Cabbart, Cambeth, 
Delpoint, Lambeth, Lonna, 
Riedel, Scroggin, Sun­
burst, Yamac, Yawdim
deep to 
shallow
I.»11.»1,cl I - 11 88 - Bainville, Busby, Cabbart 
Delpoint, Lambeth, Yamac, 
Yawdim
deep to
ately
deep
I,«11.cl 62 38 - Bainville, Delpoint, 
Lambeth, Renstac, Yamac
deep I,«11,cl,C - I 99 - Absher, Archin, Benz, 
Creed, Gerdrum, Lambeth, 
Lisam, Marvan, Yawdim
deep to
ately
deep
I,«11,«1,cl - 43 51 - Boxwell, Busby, Cambett, 
Chanta, Chinook, Delpolnt, 
Evanston, Kremlin, Lonna, 
Marmarth
Tp2 Camborthids, Torrior- 
thents (shallow)
Tp3 Camborthlds
Nearly level to moderately 
steep plains
Nearly level to moderately 
steep plains
sedimentary bed- deep
rock, alluvium
alluvium, sedi- deep
mentary bedrocks
I,SleSicl,cl - 3 3 94
I,sil,sl,cl - 9 - 9 1
- Birney, Busby, Cabbart, 
Delpoint, Twilight, Yamac
- Busby, Cambeth, Delpoint, 
Lonna, Yamac
Tp4 Camborthlds, Torrlor-
thents
Nearly level to moderately sedimentary bed- 
steep plains rock, alluvium
TpS Haplorgids, Camborthids 
Natrarglds
Nearly level to moderately alluvium, sedi- 
steep plains mentary bedrock
deep to si,I,cl,sic
ately
deep
deep I,sll,cl
4 20 76
26 3 71
Armells, Busby, Delpoint, 
Kobar, Riedel, Twilight, 
Yomac
Absher, Bonfri, Cambeth, 
Delpoint, Gerdrum, Lonna 
Yamac
C O
H
Table IV-I. (continued)
Map
Symbol Great Groups Landform Parent Material
Soil
Depth
Surface
Texture I
% Land Cover 
D F R A
Representative 
Soil Series
Ugl Paleargids1 Natrargids, 
Argiborolls
Undulating to strongly 
rolling plains
glacial till deep I,cl,si I 18 - 82 - Creed, Elloam, Kevin, 
Phillips, Scobey, Thoeny
Ug2 Natrargids, Camborthide, 
Torrlorthents
Undulating to strongly 
rolling plains
glacial till, sedi­
mentary bedrock
deep C 1Cl1I1SiC - 4 96 - Bascovy1 Creed, Dilts1 
Elloam, Sunburst, Thoeny
Vbl Argiborolls Nearly level to strongly 
rolling terraces, fans, 
benches, basins
glacial till, lacus- deep 
trine sediments, 
alluvium
cl,Slcl
■
87 13 - Gerber, Linnet, Scobey
Vb 2 Camborthlds, Torriorthents, 
Argiborolls
Nearly level to strongly 
rolling terraces, fans, 
benches, basins
Lacustrine sedi­
ments, alluvium, 
glacial till
deep Sicl1I1Cl1C 94 4 - Ethridge, Gerber, Hillon, 
Kobar, Linnet, Marias, 
Pondray, Scobey
Vb 3 Camborthlds, Natrargids Nearly level to strongly 
rolling terraces, fans.
Alluvium deep C 1I1Sl1Sil - - - 100 - Bickerdyke, Gerdrum
Vb4 Natrargids1 Torriorthents
VbS Torriorthents1 Natrargids1 
Torrifluvents
Vb6 Haplaquepts
Vhl Torriorthents (shallow), 
Camborthids1 Natrargids
benches, basins
Nearly level to strongly 
sloping terraces, fans, 
benches, basins
Nearly level to strongly 
sloping terraces, fans, 
benches, basins
Nearly level to strongly 
sloping terraces, fans, 
benches, basins
Strongly sloping to steep 
plains
alluvium, glacial 
till, shale
alluvium, lacus­
trine sediments, 
sedimentary bedrock
calcareous clay 
sediments
shale, alluvium
I,cl,c,sic
I,C1Sic1Sil 
cl
9 91 -
- 23 - 77 -
Absher, Npbe1 Vanda
Absher, Gerdrum, Harlem, 
Havre, Marias, Marian, 
Thebo, Vaeda, Vanda
- 67 - 33 - McKenzie
shallow sic,c,l,8l 
to deep
- 5 8 87 Abor1 Archin1 Bascovy1 
Creed, Bilts1 Gerdrum, 
Kobar, Lieam, Neldore1 
Yawdim
OOro
Table IV-I. (continued)
Map
Symbol Great Groups Landform
Vh2 Ustorthents, Ustorthents 
(shallow), Haploborolls
Strongly sloping to steep 
plains
Vh3 Torriorthents (shallow) 
Torriorthents
Strongly sloping to steep 
plains
X Badlands, Torriorthents
Intermountain and Piedmont Section
Steep to very steep river 
breaks
Avl Fluvaquents, Torriorth- 
ents, Calclorthlds
Nearly level to strongly 
sloping fans, terraces, 
floodplains
Av2 Xerochrepts, Xerofluvents, 
Haplaquolls
Nearly level to strongly 
sloping terraces, fans, 
footslopes
Av3 Ustochrepts, Ustofluvents, 
Haplaquolls
Nearly level to strongly 
sloping terraces, fans, 
footslopes
Parent Material
Soil
Depth
Surface
Texture
% Land Cover 
D F R A
Representative 
Soil Series
Av4 Argiborolls, Haploborolls, Nearly level to gently 
Fluvaquents sloping fans, terraces,
footslopes
Haploborolls, Argiborolls, Nearly level to strongly 
Camborthids sloping outwash plains
deep to cl,sic,c,I 
shallow
shale, alluvium, 
clayey sediment 
beds
claystone, shale, 
clayey, alluvium
claystone, shale, 
alluvium
stratified alluvium shallow I
3 9 3 85 - Benz, Dimyaw, Ellsac,
Lawther, Marias, Norbert, 
Shambo, Vanda, Wayden, 
Winifred
shallow c,sic I I I 97 - Dilts, Lisam, Marvan,
to deep Thebo
shallow c,sic,si I 2 37 69 - Alona, Cabbart, Dilts
to mod­ Lisam, Riedel, Thebo
erately
deep
(calcareous) to deep 
?
siltstone, alluvium moder­
ately
loamy, alluvium
loamy alluvium
moder- gl,l,sicl
ately
deep to
deep
moder- gl,l 
ately 
deep to 
deep
55 4 4 36 - Amesha, Mussel, Musselshell,
Scavo, Villy
63 - 28 10
28 - 39 33 - Chereete, Hagstadt, Kiwanis,
Larry, Totelake
31 6 4 59 - Beaverton, Farnuf,
Grantsdale, Villy
20 15 - 65 Attewan, Beaverell, Bercoil,
Grantsdale, Straw
OO
W
Table IV-I. (continued)
Map
Symbol Great Groups Landfonn Parent Material
Soil
Depth
Surface
Texture I
% Land 
D F
Cover
R A
Representative 
Soil Series
B Cryaquolls, Cryoborolls Nearly level, cold, wet
basins
alluvium deep cl 56 - 5 37 - Fumlss, Gallatin, Gapo 
Leavitt
Ctl Haploxerolls, Xerochrepts 
Xeropsamments
Nearly level to moderately 
sloping on outwash terraces 
fans, benches
? deep ? 50 I 26 21 I 7
Ct2 Haploxerolls, Natrarglds Nearly level to moderately 
sloping outwash terraces, 
fans, benches
? deep ? 80 20 7
Dtl Haploxerolls, Arglxerolls Nearly level to strongly 
sloping lacustrine 
terraces
lacustrine sediments deep t 13 2 79 6 7
Dt 2 Natrlxerolls, Arglxerolls Nearly level to strongly 
sloping lacustrine terraces
lacustrine sediments deep ? 76 - 4 20 - 7
Et Xerochrepts, Xerorthents, 
Ustochrepts
Nearly level to strongly 
sloping outwash terraces, 
fans
? deep ? 93 7 7
Fg Haploxerolls, Natrlxerolls Undulating to strongly 
rolling soils on glacial 
moraines
glacial moraine deep ? 33 - - 67 - 7
Gbl Arglborolls, Haploborolls Nearly level to steep fans, 
benches, terraces
loess deep Sll 20 59 20 Amsterdam, Bozeman
Gb 2 Arglborolls, Llthlc 
Haploborolls, Haploborolls
Nearly level to steep 
slopes on fans, benches, 
terraces
alluvium, sandstone
quartlzlte,
argillite
shallow 
to deep
cl 10 9 7 74 Castner, Grantsdale, 
Hllger, Perma, Sweetgrass, 
Turner
Gb 3 Arglborolls, Eutroborolls, 
Haplaquolls
Nearly level to strongly 
sloping uplands, fans and 
terraces
alluvium, loess 
glacial till, 
siltstone
moder­
ately 
deep to 
deep
Cl 9 27 64 Cohen, Crow, Danvers, Ekah, 
Lalny, Lubrecht, MartIns- 
dale, Savage
Table IV-I. (continued)
Hsp
Symbol Great Groups Parent Material
Soil
Depth
Surface
Texture
X Land Cover_________  Representative
I D F R A _______ Soil Series
GbA Arglborolls Nearly level to strongly 
sloping fans, terraces
loamy alluvium ■oder- l,cl,gl
ately
deep to
deep
8 29 3 61 - Beaverton, Fairfield, 
Farnuf, Judith, Martins- 
dale, Windham
Gb 5 Arglborolls, Calclborolls Nearly level to strongly 
sloping fans, terraces
loamy alluvium 
(calcareous)
shallow cl,I 
to deep
8 A3 I AS - Danvers, Fairfield, Farnuf, 
Judith, Martinsdale, 
Windham
Gb6 Calclborolls Nearly level to strongly 
sloping terraces
limestone alluvium shallow l,gl 6 13 - 78 - Judith, Utica, Windham
Gfl Arglborolls, Llthlc 
Arglborolls
Moderately sloping to 
steep slopes on foothills
shale, sandstone shallow I,cl 
to moder­
ately 
deep
6 22 72 - A3der, Darret, Doughty, 
Sinnigam
Gf2 Arglborolls, Haploborolls Moderately sloping to 
steep slopes on foothills
shale, sandstone, 
silt stone, 
alluvium
shallow cl,I 
to deep
3 5 92 - Big Timber, Darret, Fergus, 
Litlmber, Terrad, Tlmberg
Gf 3 Calclborolls, Haploborolls Nearly level to strongly 
sloping uplands, fans, 
terraces
alluvium, limestone shallow sic, I 
to mod­
erately 
deep
10 90 - Armington, Lap, Perlsta, 
Windham
GfA Haploborolls, Ustorthents Undulating to steep 
sloping uplands
shale shallow I 3 97 - Abac, Arnegard, Cabba, 
Eltsac, Norbert, Perista, 
Perma, Roundup
GfS Haploborolls, Llthlc 
Haploborolls
Gently sloping to steep 
foothills
igneous rock, 
sandstone
shallow I 
to deep
- 2 2 98 - Belaln, Castner, Hedoes
00
Ul
Table IV-I. (continued)
Map
Symbol______ Great Groups________________Landform
Soil Surface _____ % Land Cover Repreaentative
Parent Material_____ Depth_____ Texture_____I D F R A _______ Soil Series
Ggl Argiborolla, Haploborolla
Gg2 Arglborolla, Haploborolla,
Calciborolle
Undulating to steep pied­
mont glacial moraines
Undulating to steep pied­
mont glacial moraines
alluvium sediments deep
alluvium, glacial deep
till
7 - 40 53 -
— 4 — 95 —
Chinook, Coben, Ekah, 
Grantedale, Martinsdale
Farnuf, Judith, Martins- 
dale, Perma, Vida, 
Vllllama, Windham, Zahl
Hfl Haploborolls, Argiborolls, 
Cryoborolls
Moderately sloping to 
steep foothills
loamy alluvium deep I,all
Hf 2 Haploborolls, Cryoborolls, 
Paleboralfs
Moderately sloping to 
steep foothills
alluvium
ately
deep
I.Si
Hf 3 Arglborolls, Haploborolls, 
Cryoborolls
Moderately sloping to 
steep foothills
alluvium deep I
Hf4 Uetiorthents, Lithlc Haplo- Moderately sloping to 
borolla, Cryoborolle steep foothills
shale, alluvium shallow I,c 
to mod­
erately 
deep
Ibl Cryoborolls Nearly level to steeply 
eloping fans, benches, 
terraces
alluvium, sandstone moder­
ately
deep
I
Ib2 Cryoborolle, Cryaquolls Nearly level to steeply 
sloping fans, terraces, 
benches
alluvium, sedimen­
tary rock, igneous 
rock
deep I
6 2 25 68 - Adel, Boxwell, Bridger,
Coben, Danvers, Ekah, 
Grantedale, Martlnsdale, 
Phillpsburg, Quigley, 
Savage
I 4 24 72 - Bltton, Castle, Castner,
Hanson, Teton, Tlgeron
I I 17 81 - Abearokee, Brldger, Hilger,
Perma, Work
- I I 98 Bridger, Cobba, Castner,
Elteac
15 4 4 77 Bridget, Passcreek, Rooset
22 - 13 65 Adel, Bridget, Gapo, 
Philipeburg
Ib 3 Cryoborolle Nearly level to steeply alluvium, glacial moder- I
sloping fane, terraces. till, sedimentary ately
benches rock deep
5 - 95 Fifer, Michelson, Redchlef
OO
. 2
Table IV-I. (continued)
Soil Surface I Land Cover Representative
Great Groups Landform Parent Material Depth Texture I D F R A Soil Series
Ifl Cryoborolla, Llthlc, 
Cryoborolla
Moderately sloping to 
steep foothills
alluvium, igneous 
rock, sandstone
deep to 
shallow
I - 5 44 51 - Bridger, Cheadle, Ess
If2 Cryoborolls Moderately sloping to 
steep foothills.
Sandstone, granite, 
alluvium
moder­
ately
deep
I
'
- 19 81 - Cheadle, Hanson, Orofino, 
Skaggs, Teton
If3 Cryoborolls, Cryochrepts Moderately sloping to 
steep foothills
metamorphic rock, 
igneous rock, 
alluvium
deep I I I 20 78 - Adel, Bridget, Gorlet, 
Phlllpsburg
If4 Cryoborolls, Cryochrepts, 
Ustochrepts
Moderately sloping to 
steep foothills
alluvium, colluvium, 
glacial till, sedi­
mentary rock
deep I1Sil 2 - 52 47 - Duncom, Garlet, Leavitt, 
Sebad, Whitecow, Whitore, 
Winkler
IR Cryoborolls Undulating to steeply 
sloping piedmont moraines
deep I
Jb Arglborolls, Calclorthlds Nearly level to strongly 
sloping fans, terraces 
benches
alluvium, eolian, 
stratified sedi­
ments
deep I
- 2 2  78 - Adel, Babb, Hanaon
20 6 7 68 - Aoeaha, Brocko, Coben,
Crago, Ekah, Evanaton, 
Martlnadale, Muaeelahell, 
Raderaburg, Sapplngton
Jf
Rbl
■%2
Itl
Arglborolla, Haplarglda, 
Torrlorthenta
Moderately eloping to 
steep soils on foothills
alluvium, sandstone shallow 
Igneous rock to deep
IeSleCl I 7 18 75 - Attewan, Evanston, Nlppt, 
Nuley, Riedel
Calclorthlda, Torrlorthente Nearly level to moderately alluvium 
steep fans, terraces, 
benches
deep I 14 16 2 66 Ameaha, Crago, Mussel, 
Musselshell
Calciorthids, Haplargids, 
AgiborolIs
Nearly level to moderately 
steep fans, terraces, 
benches
7 moder- I 
stely 
deep to 
deep
21 1 2 - 6 7 -
Calciorthids, Argiborolls, 
Camborthids
Nearly level to moderately 
steep fans, terraces, 
benches
Metamorphic rock, 
igneous rock, 
alluvium
deep I,si - - - 100 -
Musselshell, Nlppt, Nuley
Attewan, Colono, Museeshell, 
Nuley, Shurley
OO
Table IV-I. (continued)
Soil Surface % Land Cover Representative
Great Groups Landform Parent Material Depth Texture I D F R A Soil Series
n>4 Calclorthlds Nearly level to moderately 
steep terraces, fans, 
benches
calcareous loamy 
alluvium ately 
deep 
to deep
gi 6 2 92 - Crago, Williams
Kf Calclorthlds, Arglborolls Moderately sloping to 
steeply sloping foothills
alluvium deep I - 12 18 64 - Amesha, Crago, Evanston, 
Musselshell, Nuley
Kt Camborthlds, Natrlarglds Nearly level to moderately 
steep lacustrine terraces
lacustrine sedi­
ments, alluvium
moder­
ately
deep
sic,slcl, 26 U  - 63 - Lonepine, Ronan, Round 
Butte
Lgl Cryochrepts, Eutroboralfs, 
Eutrochrepts
Undulating to rolling 
foothills of valleys and 
glacial moraines
clayey alluvium, 
clayey colluvium
deep T I - 92 6 — Crow, Haugen, Inez, Lubrecht 
Trapps, Whltore, Wlnfall
LgZ Cryoboralfs, Cryoborolla Undulating to rolling 
foothills of valleys.
glacial till, 
alluvium, colluvium
moder­
ately
l.gl.cl - — 86 14 - Babb, Hanson, Loberg
Lg3 Cryochrepts, Cryoboralfs
Lt Cryochrepts, Cryorthents
Ma Cryochrepts, Cryandepts
Mel Cryochrepts, Cryoboralfs,
Me2 Cryochrepts, Cryoboralfs
glacial moraines
Undulating to rolling 
foothllla of valleys, 
glacial moraines
Nearly level outwash 
terraces
Steep to very steeply 
sloping mountain slopes
deep 
to deep
colluvium g to atl
- - 93
loess, ash cap over deep 
argillltic realdlum
Moderately sloping to very Limestone, colluvli* deep 
steep mountain slopes various rock
Moderately sloping to 
very steep mountain slope
colluvium, metamor- deep 
phlc rock, sandstone
?
ell,I
gl,ell,si,I - 
gl,cl,etel, I
Garlet, Worock
7 - No reference series
94 I 5 Buckhouse, Coerock, Felan,
Holloway, Trueflssure
92 2 5 Crow, Gambler, Holloway,
Loberg, Lubrecht, Trapps, 
Tropal, Whltore, Wlnfall, 
Worock
97 I I Garlet, Holloway, Loberg,
Tenrag, Worock
OO
OO
Table IV-I. (continued)
Hap
Symbol______ Great Groups Parent Material
Soil Surface ___
Depth_____ Texture_____I
% Land Cover 
D F R A
Representative
Me 3 Cryochrepts, Cryoboralfs, 
Cryoborolls
Moderately sloping to very sedimentary rock, 
steep mountain slopes igneous rock,
colluvium, glacial 
till
deep I,SllVgl - 94 4 2 Adel, Elkner, Garlet, 
Sebud, Worock
Me4 Cryochrepts, Cryoboralfs, 
Lithic Cryoborolls
Moderately sloping to very colluvium, igneous 
steep mountain slopes rock, sedimentary
rock
shallow 
to deep
I.Sll1Cl - 91 4 5 CheadIe, Duncom, Gambler, 
Garlet, Loberg, Whitore, 
Worock
MeS Cryochrepts, Rock Outcrop Moderately sloping to very colluvium, alluvium 
steep mountain slopes
deep siI, chan- 
nery I
- - 83 - 17 Garlet, Whitore
Mfl Crfochrepts, Ustochrepts, 
Eutroboralfs
Moderately sloping to very colluvium, alluvium, 
steep mountain slopes calcareous rock
material
deep gl.sll.s+sl I - 96 2 .5 Crow, Garlet, Holloway, 
Macmeal, Repp, Trapps, 
Whitore, Winkler
Mf2 Cryochrepts, Ustochrepts, 
Cryorthents
Moderately sloping to very granite 
steep mountain slopes
shallow 
to deep
Sl1I1Sll I - 77 6 16 Ambrant, Elkern, Stecum
Mf 3 Cryochrepts, Ustochrepts, 
Cryoboralfs
Moderately sloping to sedimentary rock,
steep mountain slopes glacial till
deep gl.sll.l I - 90 3 7 Teton, Mollman, Repp, 
Trapper, Whltore
Mf4 Cryochrepts, Ustochreptst 
Lithlc Cryoborolls
Moderately sloping to mixtures of sedi-
steep mountain slopes mentary, igneous,
metamorphic rock
shallow 
to deep
gl.scsl - 96 4 Cheadle, Garlet, Winkler
Mol Cryoborolls, Cryochrepts, 
Cryoboralfs
Gently sloping to very ?
steep mountain slopes
deep T - - 71 27 - No reference pedons
Mo 2 Cryoborolls, Cryochrepts, 
Ustochrepts
Gently sloping to very alluviisn, colluvium,
steep mountain slopes sedimentary rock
moder­
ately 
deep 
to deep
l.gl.sll - 76 23 I Hanson, Skaggs, Whiteco*:, 
Whltore
00
VO
Table IV-I. (continued)
Map
Symbol Great Groups Landform Parent Material
Soil
Depth
Surface 
Texture I
% Land Cover 
D F R A
Representative 
Soil Series
Mo 3 Cryoborolle, Cryoboralfe, 
Cryochrepte
Gently sloping to very 
steep mountain slopes
colluvium, alluvium, 
igneous rock 
residuum
deep l,gl,sll,g«l - - 89 11 - Bridger, Cowood, Ess, 
Garlet, Loberg, Teton, 
Horock
MoA Cryochrepte, Cryoborolle, 
Haploborolla
Gently sloping to very 
steep mountain slopes
alluvium, igneous 
rock, sedimentary 
rock
shallow 81.1 78 21 I Belain, Bridger, Castner, 
Cheadle, Cowood, Hanson, 
Libeg, Peraa, Tropal
Mo 5 Cryoborolle Gently sloping to very 
steep mountain slopes
shale, limestone shallow 
to deep
• 11.C - 57 43 - Duncom, Mayflower, Tarrete
Mo6 Eutroboralfe, Agriborolle1 
Llthlc Cryoborolls
Gently sloping to very 
steep mountain slopes
alluvium, igneous 
rock
deep all,I,cl - 74 26 - Cheadle, Farnuf, Hilger, 
MicmeaI, Yourame
R Rock outcrop, Tallus, 
Cryoborolls, Cryochrepte
Mountain peaks, gently 
sloping to very steep 
alpine grasslands
alluvium, colluvium, 
glacial till, 
various rock types
shallow 
to deep
I,»11,»1.gel I 43 I 56 Blackleed, Cheadle, Cowood, 
Garlet, Raynesford, Sebud, 
Tropal, Whitore
E t -
j : m
' L l r
£ T t
C  ;
to S>
*
- '
f  *• -
>r. ,v
V T x- ^  v-
«, i
- -  -
%
I
f 6
' - Li" I
- B
m - .L. _ _. .
5? v fcft tr it) ■> «r»- #- -ft c ft s- »• I
■ u^ V
—  - -‘L ■’ »• I
- t  ,-n
- • V - -X -f- :n - X- V X T» . 'J- n ■:<** {*, I: rt -xr tr
•O' «r ■=' s— -'t <•! t f
xr - ,m
•'tJii a
* Ar
M <r> »*r AL
- m L- •#» ! IT
vi»
J
- — ■- ■■ -■'- ....  —J#% n - •? -. -o ST #, v '*»: rf: •»; ft [sr *? 9* v v- ^ ^
.
r uI T  ^  ’ >-  -
: -  ; "  ^
V *• L> OH ‘iX V
- ' '•'- - 3 l^ 
.  :■ - -,I
.
V- ! X\ W>
j.>7 nr- jv r,. V !V
- F i
'•> i i
*- *r ef
rIF
F  F -  * '  ^
H i  ^  c / i
-Ji 'XT'
- X J
X M -O f t  XI tf)
I--:-- I -:--- - ---- L-P
V ' t
' t
- $ & r _  -
T l -
< X- V V#
•y> i* ro m a-
XT xf V X- A
V Xf- V
r- - - —* i——
P t
5 • -
-JX
-o. IT
m m  m yi -ii> v - LtzAUi
B  T m ,
; 'v  :•' llU v F
J  U' Fi
r  v  ■* ■'■
L  XF »2 n  j- x
p ;w -» .
—— -- --
'SH v
r  ^ -
j------—
i'
;—   —A — - -----------------------*
I *  *  t ,  r. w  -I
!  ’ F C - ,  j
r •- t; «- v f». i -
— , [ ••- — i— i 1 —  --
^  i  ; ^  ^  ^  y- ; ^  ^  k  '-T-
* > •.'■ ,  ^ f r
, —  r '
fVm ETETV
L i  L - V  - -  r .  r _  " U r - J j  
Ti --"I. LpT !
R F -Li
L -x. --___ _ ?
J ’ -  —
□  IRG ATE
* < u
r Li
"tT i
3".
F -  F -
F j !FufT
k?i 39926
F' • 'L '
M ox
Of '. _.  , fB— ,  --- >--- - r—
r W 1Xi I x- re « >r "=• ♦•‘ ‘ *'•; - 1—‘
F  ! ' T
T  P T
'i
B E T U  U „ . ‘
- :  $
- ------^ : r~
•> rrj O •%
SB
' .L S
<n -9 f ji) '"> I T 'i L1 >.•
i x* *  Xf x» xr •; ... ., V
•ti T m -) ^
m
Znn- crt <y a> r. ,
x,
-E B
— - ,
" - r r • .—  -V T--- I— r *— I
*  v  h' -n
k , i
■jtrfrr-i r
3 -  B•9__7L_Lr-*"
' • 9 t •  Lit •>{■_
r r> x- *i -7 <• iv> x r  J
-- : - '— — -
___
EL
S d T
-rI #  <30 o- <f> IT-
-r "I CO — I 'T CL
T IT
>- to O
T l :  . T E
iO- f^S I' ■
t«- _
ta
E ' %
Xv ^  , n; n U
. .a , a  u
TL it, rt,
•c- rx. I- x«- -
L i u ; - -
T E -  L
A 'O CO V* K
ft Vj
T- v ;
"  ,1V V V <-,■ -O :A
; i .  J r -
K^ j—- —
PT' T JtT 
ZT J' T&i
---X-1 ------------:----J
1X: 'en A C»?
T
T  E F  : : 
• Li
. iJl ; " ' i
n y. *
T
=f <r
c  co
-y
L I  Tl
L-
'
—-------------4-— — -4 X.w - » OV - -----
;n »  <r .*• .< .-«- '  m •' . , t
U, - rn , '
. CO T
T . T
TTi it JI
J7
■v < ^ V
Eu ?•' -K  >V CM ^ ^  *0
n  *r- (c Cf, <n r> ^
S
J3• 9 i
-
,--- »- U"
L  CO m CX r ,  T J
Cl Cl TT
" # L >
B•9 *-
.
S r
T
Tl
i . T - "
jCrt <T* '7
■O *17 CO
tt» ^  ^
x> r mJl
dLB ES
<*» (* A
<n ret
V  o> ? «5*1 CO
r  -
v-
ErT F ;liZp t
T T  T  
r EIrXtU-
*1  *  f» j  i"1
10  ■»:■ <0
FT
t ix
K fS H H H iH
5. T- I Tuu " '
H H H H jL E fT i
: PPT
FDFEL “
/ >  I i T  i  iIvLT v /  - M
Z
r? CA i
Eg tr ! T tT  i D  m ,1
f  th UrZ kJ> • 'rt ! ?f.X  -jT-'
v  V  F  / Lv
T ir-
F ■ Es
F
-Z—  ;—
CO «■
x ; ? .6 73
m <0
CT
-  -  j
T ' 'E- X
L  -  -  I
I PHHt-
-  C OP
fc
T•v 07 «r-
T-S  „S_____________
B i
ro ” 1 TQ
I-L
jf i  T  _  _  A
V v - F  j.
JTTl
1-------------—  ,  - . . - - •  -C :
L - -  J X  "  ■'
■ >« eg
I -  j J  «r.
M  X i  r -  — T  L
F** *rt CO > A
ISO v. 0 u> ,
I "  ! ------------------ '- —  •; L--------------—-------
-- - : «>•••* r- n <r> ' - - ^rf---1 ---- ; j— I ----
|«0 '0 ■> I of -r r f  <* O "V
L3
•rt f)  #0
- '•
rtrt rt (v> r)
g i  F 1 L X «
. I rr
FJ
r_ JZf
ffi
L  *■
L—«—
rry ^  Trt rtt
-------T Xf
-  X- --• -
11-t
.. ~
X
-Cl
-  --■ .
n
mu 
c_:r
S  i r t r  I F X  - C 1
F- -L ,.*  -- —  - i — '
'!O C  »  I "
I f H H H l
E
LiE
,-0 *
<« O
' R FiT ^ X .. ,- j
Hf
jJFp T
■ p*—i  SEY EL
. - S :i j 
R T c
r
X F  . I
C i -f S 7H  e  E T
p‘
m
B R
X TP F
Lu*.
C NG n:
--
i _ 0:
Es a' '• K
«-» -- —
-  _■
Iio <r> -* » I
- o 'T l  T T
' CL XE X  L
n  -r
n I
F T
-> §
ER
mi
«*• nO
- A - A  ■'. ri ;V ,
[ J
T
j
Hi
CA R- % <V
..
T
—— — t— . j
' ' J '
or. <r
T L '  I -T
A y C fift T T
L I
R«■> iv
L  H F  f  T
- a # r.
X e
-x—— -f—
c"11 - L L_ a
'•-: r L hfi L
ir
F7S F
y <r ' r  m ♦> s. . <• s»v ••*» -t#
- U t
'C' A  e r
■ 3
11 *# -
S
RFRX 3
if  ■ v  tr
FV Xl
0 S
----------L ------ r—J
A ^  IU I,
rci
tJt I
tr tt w
Sn
T 1 J
— -----------------------
LC Vrt -A -rt- *A Xj- r?
m m *?  A
_______________f c :  H  f c
J1 ES SB .
FUrrT '
", RiJT- a
— 3 ' -T> -*>
T
A EMT NS
f -  (ra w Tf
• VJ :,to
Ifi D  T  f> !
i  -------------------------- •
r- " —
f  "I on x<
-RT 0
z to X- I -  ^
JrCf
■> AvJW tm a  ^ *  «"
r
L
rt* A $f>" ?•» u> r*
y, :
0
0
s E
El STFru ---1
fi (*! .,
R
S T i
L  L J
STTl XJl La S i -
L j — .«
,----- • "X --C ------}
I ^  =O *■ ] .-a to
, % —
-
- R i i  T i T
B T x  u -------------------------p
• . -  j — - 05 ,■» m <D di (D j
B'
I : ' /  R  L
ir
ti
T
1-------------. — ---------H ]
;.t Pt ■ ‘ {Or. Crt 4A - I \  — ; X
1— -J r— — ‘ —-
«
S i  L i
XJl '
- F ' 3
L1--L
_3 J
J
,*• x) r  !
f  RlTrT t  U r "Lr-U I
L R :  R  •
•  ' r - ! - - '  » X
! _  ^  I -  f i t
B prT
I----- U . -  L  —  )------—  ;Rz m  to
A
m  wi r>
v rt or, W  A t - ,  _ V i -f*
-V) .*> •*; JT.'_ '
a'  3 F r I
! X r t i -  J T
•-tv .rt' X
TT
LZ
1 Cf re
e
------
5 r*> ..» <*$
f I
'C1»  m
1 V  S  ; p
L 0 ' Rl I
e <rt- M e,
!
UA 1 A  it)
<r p <f T
&T
—
Z f
-V Z
5 ® C r. :c
1"-
---
S R
■*
I hR ’ c .. -
rt
.
f
— - 1
.
HCS-RRMS .
SCSJT T  I T C O T T l
.
R-Fr KlJTt TC iC  CS"
MONTANA STATE UNIVERSITY LIBRARIES
stks N378.Sm634@Theses
Development o f  a Montana land cover map
3 1762 00169414 8