72 
map units corresponded to relevant ground units, there were some areas which 
appear similar on the image but which differ in vegetation types ? amount of 
vegetative cover, and productivity. A typical problem was misinterpretation 
between specific plant associations and areas which had been burned by range 
fires a few years previous to the date of imagery. The only effective means 
of distinguishing the two would be the detection of the linear boundaries 
associated with the previously burned or treated areas. Only the most generalized 
maps of vegetation and/or soils can be prepared in this type of environment when 
interpreting only one date of ERTS imagery. However, the broad stratification 
of terrain types is an important first step when multistage sampling procedures 
are used to estimate resource parameters. 
The vegetation maps prepared through interpretation of high altitude 
color infrared photography (1:30,000) provided information which for many 
resource uses was too detailed for the regional management efforts of resource 
specialists. Certain vegetation communities could be mapped to a two to five 
hectares minimum, due to the high level of resolution of the photography. 
Resource specialists concurred, however, that it was easier to make generaliza 
tions at a later date based upon detailed maps rather than to achieve a higher 
level of specificity from maps which were generalized from the outset. 
The detection and classification of sensitive areas through manual 
photo interpretation provided valuable information for assessing grazing, 
wildlife and recreational potential. For certain classes, e.g. wet and dry 
reservoirs, the sensitive area maps served to update existing information; for 
other classes, e.g. stringer meadows, moist sites, and springs, the sensitive 
area maps provided new information which was not available through other sources. 
Some difficulty was encountered in the classification of certain meadow types 
due to the variable climatic influences throughout the study area and the date 
of photography (August 21, 1973). For example, similar meadow types (according 
to the classification scheme) appeared differently in different regions 
because of the different rate of drying of the herbaceous vegetation. The 
date of photography also posed problems in identifying small reservoirs which 
had subsequently dried, and areas where the herbaceous vegetation had dried 
due to low water holding capacity of the soils. However, the high resolution 
of the photography enabled interpreters to do an accurate job in identifying 
these categories. 
For the inventory, evaluation and mapping of soil resources the mapping 
cell concept was found to be more applicable than the transect method in large, 
inaccessible areas. Due to the roughness of the terrain in the study area and 
lack of inherent diversity, the transect method was not a feasible alternative; 
but certain modifications were utilized within each of the four mapping cells. 
Primarily, these included sampling the soils on various aspects, perpendicular 
to major drainage patterns, and across dominant landforms. Placement of soil 
profile plots was best accomplished through non-random allocation within each 
mapping cell; this is justified since no quantification of soil properties 
was desired for the current project. Each of the mapping cells (7.5 X 15km, 
ground dimensions) contained ten complete profile descriptions or one plot per 
1000 hectares. The low number of profiles was considered adequate due to 
the lack of complexity in the soil-vegetation-terrain resources in the study area. 
Patter 
simila 
aerial 
inforn 
The us 
a mult 
of the 
eroded 
with d 
Interp 
dense 
basin 
Land u 
homoge 
cover 
map th 
with a 
terra! 
variab 
on a r 
techni 
is des 
behave 
have b 
units i 
discus 
documei 
detail* 
though 
simila: 
coales* 
resolu 
approa* 
field ] 
a regie 
no sac: 
level. 
of acc< 
area nc 
effort 
of inf( 
or thr( 
constri 
the maj 
time ai 
withoul 
color ]