posteriori compensation, percentage after eliminating system- 
atic errors by a posteriori compensation is also computed 
Statistically, ; 
n= }- Cs 6% 
n= number of percentage 
Gs is systematic errors of residuals for coor- 
dinates in blocks after applying a posteriori compensation 
6; is systematic errors of residuals for coor- 
dinates in blocks without avplving a posteriori compensation. 
Statistical results are shown in Table 2. 
From Table 2 it can be seen that, after a posteriori com- 
pensation has been used, individual lowering in accuracy on 
the contrary occured, this is because while interporating a 
certain kind of systematic error, the important thing is to 
make planimetry or height to produce svstematic distortion, 
but under the process of refinement, as it is not controlled 
properly, that is, among 3 statistical measurements Tx,Ty,Tz, 
if any of them is larzer than 1 or equal to 1, a correction 
of systematic error for 3 coordinates must be made respec- 
tively. so as to render coordinate accuracy of T «1 lower. 
Owing to the fact that the addition of Systematic error 
(1)» (2), makes the block to yield much larger planimetric 
SvStematie error, but height systematic error tends to be 
much smaller(with regard to this fact, it can also be con- 
Sidered that planimetric Systematic error has influence on 
height systematic error), therefore, while Statistical tables 
are made for the average increase of rate of multiple and 
average percentage of 6:9 and Óz , no consideration as above 
Will be taken: under the similar circumstances, in systematic 
error (3) distortion of planimetric SyStem appears to be much 
Smaller, in statistical table of Ge», and Gp » no consideration 
will also be taken (in Table 1 and 2, within the square frame 
rounded by thick lines are the principal parts where system- 
atic errors derive ). 
Relationship between benefit of a posteriori compensation 
and Signal-to-noise ratio is shown in Table 3, Fig.4 and 
Fige5. 
- 496 - 
i .... 
AUR. À 3