, CA, 9-11 Nov. 1999 
AND VALIDATION 
ome base camp were surveyed 
d GPS and laser altimetry in 
and 5). 
Coffee Can Station 
0 
Skidoo SPS surveys 
Laser altimetry su "eus 
  
  
  
2 4 6 
n surveys at Siple Dome, West 
as skidoo) surveys are used to 
le for comparison with laser 
bias was found for the laser 
systematic timing error in laser 
aser range during processing. 
i are converted into a local, 
n where the x-axis points to 
he center of the Earth, and the 
tem. The position of the GPS 
the SOAR tent was selected as 
Equation 3 describes the 
ind 5 show how the reference 
iteO 2): R, {pont tee, : 
)) (Egn.3) 
r is not perfectly aligned with 
, as defined by the INS. The 
d as the angular difference 
he laser axes. The estimated 
es in pitch and -0.3 degrees in 
son (SOAR field notes). Pitch 
vely flat test field are used to 
d surface shown in Figure 5 is 
| no mounting bias corrections 
d dp to minimize the surface 
(km) 
  
   
International Archives of Photogrammetry and Remote Sensing, Vol. 32, Part 3W14, La Jolla, CA, 9-11 Nov. 1999 
deformation the mounting bias can be estimated. In this case 
the initial estimate of -1.2 degree in pitch and - 0.3 degree in 
roll was confirmed by using a simple grid search. 
PA TE RUE AN ST STL TE TNT 
eT" 
  
pei LE eri cb ee Be 
  
-450[1.. dod i i 
0 2 4 6 
l | | 
10 12 14 
  
eL 
Distance (km) 
Fig. 4. Surface derived from repeat laser altimetry (grey lines) and 
snowmobile-mounted GPS (thin black lines) along Skiwayl. 
53. System Accuracy 
After correction of range and angular biases the SOAR system 
performed well over short distances. Repeat flights over 
Skiwayl show a 2.8 cm bias with an overall RMS of 10.3 cm. 
The bias is attributed to surface slope, because the repeat flight 
lines were approximately 25 m apart. The RMS difference is 
measurement of surface roughness caused by sastrugi. Sub- 
decimeter accuracy was also found when laser derived 
elevations were compared to elevations measured with the 
snowmobile surveys. 
BIT T T T T T 
  
—— — — — Surface from skidoo mounted GPS 
a Surface from Laser altimetry before 
rêtru after 
  
  
  
Fig. 5. Comparison of measured elevations along Skiway 2 surveyed 
With snowmobile-mounted GPS and laser altimetry. The two laser 
profiles are derived from the same survey before and after the removal 
of the laser mounting bias. The magnitude of the pitch maneuvers are 
about 8 degrees off nadir. 
6. REPEATABILITY AND ACCURACY 
6.1. Crossing flight lines 
Laser surveys that cross nearly perpendicular to one another are 
evaluated to determine the repeatability of the procedure over 
long baselines. ^ Crosses are evaluated by finding laser 
measurements within a 10 m radius of the cross-over point. The 
average elevation difference of crosses on Ice Stream C are 
shown in Table 1. Table 1 also shows how crosses compare if 
both laser surveys were conducted during the same or 
independent GPS surveys. The agreement is better for crosses 
measured during the same flight. 
Elevations produced from Ice Stream C survey flight number 26 
were found to have large biases of up to 84 cm. This large bias 
is 8 times greater than the reported GPS RMS for that flight, 
which suggests that the phase ambiguities were not resolved 
throughout the entire flight. Poor GPS initialization practices 
(ie. the aircraft started and ended at different base camps over 
280 km apart) are the most likely source of the problem. The 
end result is that the flight is of limited use for change detection 
on à two year time-scale, and topographic maps made with this 
flight could have an error of nearly a meter. Average elevation 
differences are given for Ice Stream C with and without this 
flight (Table 1). 
  
Surveys Number Number of Mean 
Considered of Laser Elevation 
Crosses | Measurements Difference 
Between 
Surveys 
All surveys 8 34 0.36 m 
w/o 26: 4 16 0.25 m* 
Same GPS 2 8 0.18 m* 
Survey 
Independent 2 8 0.32 m* 
GPS survey 
  
  
* Values do not include measurements from survey flight 26. 
Table 1. Cross-over comparisons for Ice Stream C 
6.2. | Coffee-can comparisons 
Laser surveys go near distant coffee can stations 
providing a means for evaluating accuracy over long baselines 
(Fig. 6). The coffee-can markers are surveyed using precision 
GPS methods and are used to study ice sheet mass balance 
(Hamilton et al., 1998). GIPSY is used to determine marker 
positions with a vertical error of less than 0.01 m. The laser 
flights missed the coffee can site on Ice Stream C by 840 m,