The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
downward continuation process. Using Stokes' formula, the 
geoid undulation is determined from the downward continued 
gravity anomaly Ag . The downward continuation can be 
applied to either the airborne gravity anomaly or the disturbing 
potential at the acquisition altitude. As shown in Figure 7, the 
global geopotential model is used as a reference for the local 
gravity while the airborne gravity measurements from the 
airborne inertial gravity system are used to densify the local 
gravity information, i.e. provide the high frequency gravity 
variations. For the geoid determination, the terrain correction to 
the gravity anomaly is needed. This can be done either using 
available terrain models or from the interferometric DTE. The 
details of Intermap airborne gravity and geoid mapping system 
can be found in Wei and Tennant, 1999&2000. 
Figure 7: Diagram of STARGRAV process 
3.2 Airborne Gravity and Geoid Mapping Results 
To assess the accuracy of the airborne gravity anomaly 
estimates, independent upward continued ground gravity data 
were compared to the airborne gravity anomaly at the flight 
altitude. Independent ground gravity anomalies in the 
Washington DC area are available through the NGS US and are 
terrain corrected free air anomalies of 2'x2' grid. Figure 8 shows 
the survey site in the Washington area. To evaluate the 
performance of the Intermap AIGS solution, the NGS ground 
gravity anomaly is first upward continued to the flight altitude 
by using the Poisson’s integral. The terrain correction is applied 
to the airborne gravity data. 
Figure 8: Airborne gravimetry test area 
Figure 9 shows the airborne gravity anomaly filtered by a 120- 
second filtering compared to the upward continued ground 
gravity anomaly. The standard deviations of the differences 
between the measured airborne gravity anomaly and the upward 
continued gravity anomaly are about 1 - 2 mGal (1 <7 ) for 16 
flight lines. Similar airborne gravity results from the STAR 
AIGS in other survey areas can be found in Wei and Tennant 
(1999). 
Using Stokes’ integral the residual geoid undulation is 
computed from the downward continued residual gravity 
anomaly. The precise geoid undulation is determined by 
applying the restore procedure to the residual geoid undulation. 
The local geoid determined by the airborne gravity data is 
compared to the geoid G96SSS from NGS. Figure 10 shows the 
difference of the local geoid determined by the airborne gravity 
data and the geoid G96SSS from NGS. The standard deviation 
of the difference of the geoid undulation is 5.6 cm, as shown in 
Table 3. 
number of data 
Figure 9: Comparison to upward continued ground gravity data 
'-50 0 50 
S-coordinate (km) 
Figure 10: Difference of geoid undulation (m) 
Type of Comparison 
Mean 
Std. 
Airborne gravity anomaly (mGal) 
-0.33 
2.86 
Ground gravity anomaly (mGal) 
-0.16 
2.63 
Geoid undulation (meter) 
-0.164 
0.056 
Table 3: Statistics of airborne gravity and geoid results 
3.3 Integrated Airborne Mapping System 
Using IFSAR technology, Intermap's STAR systems provide a 
new generation of digital elevation models (DEMs) and 
orthorectified radar image (ORRI) maps. Because the Global 
Positioning System (GPS) is used as the position reference for 
STAR system, the terrain heights provided by STAR system are 
referenced to an ellipsoid. They must then be referenced to an 
orthometric or mean sea level height, to be consistent with