-4- 
It is however possible to compute corrections to eliminate this error 
influence in the adjustment phase of aerial triangulation. 
With the development of LASER-APR, applying a extremely narrow pencil of 
rays, as announced in the U.So, considerable improvement can be expected 
with respect to the accuracy and reliability of radar altimeter data. 
Rot only will the actual accuracy of the radar distances be improved, but 
also the unfavourable influence of certain terrain-types will be eliminated 
to a large extent. This will probably increase the possible applications of 
APR equipment since there will be less dependence on the terrain characte 
ristics. 
However, one must not overlook the fact that this new equipment is still 
based on hypsometer registration. This means that although the radar 
component is considerably more accurate, the total combination, representing 
the absolute terrain heights, cannot be better than the information obtained 
by the hypsometer component. 
The accuracy of APR-data 
Different evaluations of APR data (conventional type) have resulted in a 
standard deviation of 2-3 m under optimum conditions and 3-6 m under less 
favourable conditions. There is a clear indication that the accuracy 
decreases with increasing flying height. 
As an example, results of an investigation carried out at the ITC (ref.[ 17] ) , 
which can be considered as quite representative, are given below. 
High altitude APR 
(H = 5600 m) 
Low altitude APR 
(H = 2000 m) 
No. of points 
m • s • 6 0 
class 
a 
125 
3.0 m 
class 
b 
201 
3.9 m 
class 
c 
352 
4*4 m 
class 
d 
555 
7.9 m 
All points 
1233 
6. 1 m 
class 
a 
69 
2.4 m 
class 
b 
128 
2.2 m 
class 
c 
289 
2.8 m 
class 
d 
388 
4»2 m 
All points 
874 
3.4 m 
Due to the fact that individual APR registrations contain certain random 
errors, their value for the control of aerial triangulation can be improved, 
by using the mean of groups of APR points.