FIGURE 4.—The center member of a stereo model illustrating how unique images are selected for identification 
and description. Two or more examples of each image are obtained on vegetation and soil conditions represented 
sary are checked on the ground, and identifying data are circled as potential sample areas. As many as neces- 
by: each image examined. The record is organized into a training aid and reference set from which trainees may 
learn the image-subject relationships and may develop initial skill as photo interpreters of rangeland subjects. 
The legend symbols are expansible, starting 
with notations of information that can be 
reliably derived from interpretation. As ground 
checking is done, symbols are added to convey 
the kinds of information that can be deter- 
mined only by ground examinations. Where 
the interpreters are highly skilled and can 
make statements about the resource from ob- 
served image characteristics and from conver- 
gent and associated evidence with reasonable 
186 
reliability, the symbol system provides for a 
different notation than where the same infor- 
mation is acquired on the ground. Thus, users 
of the information are automatically advised of 
the probable reliability level of data on each 
parcel. 
Resource analyses of this type provide infor- 
mation on landforms and soils characteristics 
that are relevant to what rangeland produces, 
to plant communities and vegetation, and to 
