Abstract: Objectives: Depth contour is one of the key factors in the expression of submarine geomorphology. With the increasingly urgent need of exploring the ocean, how to quickly and accurately generate the depth contour of a large area has become an urgent problem to be solved. The traditional ocean depth contour generation algorithm converts the data to the plane, then building regular grids for tracking processing. The expansion of the ocean area needs to be processed by framing, and the map splicing is complex, especially in the high latitude areas where the projection deformation is large; Or the triangulated irregular network can be directly constructed for the original data, but the algorithm is complex and too slow, which is not suitable for large area and large data processing. Methods: The discrete global grid system is a multi-resolution grid with consistent spatial datum and seamless global coverage. It supports the processing and analysis of wide area geospatial data structurally, of which polyhedral grid are most commonly used. In the polyhedral grid, the icosahedral hexagonal discrete grid has a more uniform global distribution, and its geometric properties are more ideal than others. The hexagonal global discrete grid system is applied to the generation of depth contour. Firstly, the original data are sampled to the hexagonal grid with appropriate resolution, when the depth contour are traced using the hexagonal grid and the grid crossed by the depth contour is processed using the hexagonal vertex symbol method. Two vertices of a grid are interpolated on the spherical arc. Finally, the depth contour generated by ArcGIS are used as the reference depth contour with the original data, and the hexagonal discrete global grid and the plane hexagonal grid with the same resolution are respectively constructed. Results: From this, two sets of experimental groups of depth contour are generated, where the average and standard deviation of the spherical distance of the offset reference depth contour are used as the quality evaluation indicators. The comparison results show that, under the same resolution, each quality evaluation index of the hexagon global discrete grid depth contour is better than the plane hexagon grid depth contour, and it is visually close to the reference depth contour, and the degree of deviation of the depth contour at different latitudes is uniform. Conclusions: This algorithm can uniformly process global ocean areas without considering the problem of depth contour joining caused by map mosaic. What’s more, we verify the accuracy advantage of it compared to traditional planar grid depth contour generation algorithms.