um-scale 
verflown 
le viations 
field was 
/ with the 
tography. 
our maps 
the radar 
3 and 4). 
ine | was 
°hromatic 
he photos 
averaged about 1:2700 being flown about 820 m above the ground. The scale of the 
models was controlled by the radar altimeter and by ground control at the ends of the 
1 km transect. Model restitution was further aided by data from a tip and tilt indicator. 
  
^ » NR | à » DS A 
= 3 à $ à 9 3 
i 2 
PHOTO NUMBER 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
| | TREE CANOPY PROFILE 
  
  
GROUND PROFILE 
FIGURE 2. Copy of original output of radar ground and tree canopy profiles. 
The tree profile is offset to the left by one division. 
The profiles were drawn by plotting from independent stereo pairs mounted 
on a Zeiss Stereocord plotter (Figure 5). The Stereocord enables the profile to be 
digitized (x, y and px) and fed to a programmable calculator which carries out the model 
restitution and drives an xy graphic plotter. The graphic plotter coordinates were scaled 
to match the radar strip chart record counterpart. The photo profile was built up by 
joining adjacent models through photo control points. The forest canopy profile could 
then be plotted along the flight path of the corresponding radar altimeter flight; the 
ground profile was plotted wherever the ground or ground vegetation could be seen on or 
near the flight path just mentioned. The dotted and dashed lines in Figure 6 are 
respectively, the photogrametrically generated terrain and forest profiles. The broken 
and solid lines were copied from the strip chart record of the radar altimeter and 
represent the terrain and forest canopy respectively. This profile covers the same line as 
the profile in Figure 3, however, the vertical scale is different. 
Species identification was treated as a secondary objective and thus was 
restricted in scope. Constraints were imposed both by the amount of time available for 
collecting species data on the ground and limited access to existing data. A second 
limitation was imposed by the photo interpreters having no experience or knowledge 
related to tropical tree species. Nevertheless, the good-quality, large-scale photos taken