The downward continuation (DWC) method was used to determine the density contrast between the seawater and the ocean bottom topographic mass to estimate accurate bathymetry using the gravity-geologic method (GGM) in two study areas, which are located south of Greenland (Test Area #1: 40 - 50°W and 50 - 60°N) and south of Alaska (Test Area #2: 140 - 150°W and 45 - 55°N). The data used in this study include altimetry-derived gravity anomalies, shipborne depths and gravity anomalies. Density contrasts of 1.47 and 1.30 g cm^-3 were estimated by DWC for the two test areas. The considerations of predicted density contrasts can enhance the accuracy of 3 ~ 4 m for GGM.

The GGM model provided results closer to the NGDC (National Geophysical Data Center) model than the ETOPO1 (Earth topographical database 1) model. The differences along the shipborne tracks between the GGM and NGDC models for Test Areas #1 and #2 were 35.8 and 50.4 m in standard deviation, respectively. Furthermore, these differences were more strongly correlated with gravity anomalies than bathymetry in the test areas. It is shown that an accuracy of under 40 m can be obtained with comparisons to shipborne depths only in Test Area #1.