651 - 
Lterpretation 
.al. , 1968, 
e promising 
rectional 
of the appli 
ed to the 
actively. 
3 ctions all 
y character- 
HORMANN, 
ry is of par- 
can then be 
ss, and nu- 
relief ana- 
answer. An- 
of a number 
1 in conjunction 
le elaboration 
•oduction of 
age (see also 
function gives 
rawn by hand 
;he automatic 
normally 
c. Since the 
ised for the 
:o should be 
-parallax 
photos but re 
duced, and 
;ant deforma- 
. Another 
approach to stereoscopic study of orthophotos has been developed by the Canadian 
Research Council (BLACHUT & van WUK, 1970). A pseudo-orthophoto to be used 
as a stereo mate is produced which is distorted by artificial horizontal parallaxes 
that are proportional to the height differences in the terrain. A normal 3-dimen 
sional view can thus be obtained. 
An evident advantage of orthophotographs is that the measurement of surface area, 
of stream length, of entrance angle, etc. , just mentioned in connection with quan 
titative relief analysis, can be carried out much more precisely and conveniently 
using orthophotography since relief displacement, scale differences, etc. are 
eliminated. Contourmaps and maps showing slope steepness classes can also be 
easily obtained. The dropped line plots can either be used directly, or may serve 
as a basis for the construction of profiles, including serial, projected, and/or 
superposed profiles. Limitations of the method are the scanning direction which 
is restricted to the Y-direction only and the predetermined height zones of the 
dropped line plots which cause generalization of the profiles and interfere with 
the recording of features considered to be of special interest, such as breaks of 
slopes, etc. 
An interesting possibility is the valley count, which in effect amounts to half the 
number of slope reversals. The problem that the direction of profiling cannot be 
selected at random but is invariably parallel in one (Y) direction, limits the ap 
plication to areas where the effect of structure on the drainage net is negligible 
and to sample areas of sufficiently large size where the direction of the drainage 
elements can be assumed to be random. The matter has been elaborated upon by 
GUSTAFSON (1971, 1973), who also mentions an interesting relation between 
valley count and streamlength, given in the formula: 
L = (V ) (K) (OSR) 
p c 
in which Lp is the predicted total stream length per basin, is the valley count; 
K is a constant related to the angle of intersection and OSR is the orthophoto scale 
reciprocal. Through this formula, valley count can, amongst others, be related 
to parameters like drainage density, etc. 
Another possibility of the dropped line plot is the establishment of the relief- 
amplitude per profile or per stereomodel. In view of the intervals at which the 
code (thickness) of the drop line changes, the values for relief amplitude thus ob 
tained will invariably be too low. If the sample studied is sufficiently large, how 
ever, statistical methods of data handling can correct for this. The combination 
of relief amplitude with valley count or the drainage density derived from it, will 
result in indications of ruggedness, etc. The advantage of the dropped line approach 
is that it lends itself readily to automatic or semi-automatic data handling.