Abstract: Abstract Improving the accuracy of Arctic seafloor topographic model is of great significance for the research of Arctic shipping lanes, and the IBCAO V4.2 model occupies an important position in Arctic related research fields with its fusion of multi-source sonar data, and the construction of seafloor topographic model by combining bathymetry data and gravity data has obvious advantages in relying on the construction of the model by relying on the bathymetry data, and the Gravity- Geologic Method (GGM) has a significant advantage in this regard. Geologic Method (GGM) can effectively combine bathymetric and gravity data to construct seafloor topographic models. Aiming at the deformation problem of geographic grid and the key parameter density difference in the construction of seafloor topographic model in the Arctic sea by GGM, this paper proposes a way and method to solve the density difference constant by regression analysis under the polar spherical projection condition, and then invert the seafloor topography by GGM. The method takes the eastern part of the Greenland Sea as the test area, and combines the SDUST_GA_2022 satellite data and part of the shipborne measured bathymetry data to construct a seafloor topographic model covering about 7.6×104km2, which is called the AO1-GGM model. The model was externally validated using shipboard measured bathymetry data, and compared with the IBCAO V4.2 model and the AO1-GGM model constructed by relying only on shipboard measured bathymetry data. The results of the study show that 1) in the region where shipboard measured bathymetry data are lacking, the AO1-GGM model can draw topographic details with the help of ocean gravity anomalies in the polar region, and its performance is better than that of the AO1-Grid model. 2) Compared with the IBCAO V4.2 model, the AO1-GGM is more compatible with the actual situation, and it has better results in the checking of the assessment indexes, such as the MSE, RMSE, etc., with an improvement of the accuracy of about 35.01%, and the degree of dispersion of the checking error is smaller. The above results also show that the accuracy of the Arctic seafloor topography model can be significantly improved by integrating multi-source sonar and satellite altimetry and gravity data.

Key words: satellite altimetry, gravity anomaly, seafloor topography, polar spherical projection, model accuracy, Arctic