Marine gravity gradient tensor is one important gravity over global ocean. The model in the manuscript is derived from GM data of 5 altimetry satellite missions. The model can be used in marine geodesy and geophysics.

• Line 6: Here what are literatures?
• Line 41: Some altimetry satellite missions consider their GM phase in their start or middle life of satellites, not all end of life.
• Line 49: Here should be geoid model instead of geoid.
• Line 58: Here what are literatures?
• Section 2: Only GM data of 5 satellite missions are used in the study. Why not use the other satellite altimeter data include ERM data? What errors are corrected in SSHs? How to assess the precision of SSHs? How to distinguish SSHs’ frequencies?
• Line 93: What earth gravity field model is used in the MDT model? Not egm08??
• Line 110: What are x and y?
• Line 112: How to precisely determine weight?
• Line 119: Why to use 2 degrees?
• Line 125: How to define the local reference frame in detail? One local reference frame is only used in one small area.
• Line 128: Here normal gravity should be normal gravity in geoid.
• Line 143: How to calculate λ? How about λ=0?
• (11): In eq. (9), normal gravity is a function with respective to latitude.
• Line 159: How about north and east components of DOV?
• Line 165: How to construct the eq.?
• Line 173: Jason GM?
• Line 178: What are both weighting approaches?
• Line 185: SIO can provide DOV model. But DTU not provide directly DOV model.
• (17): δg is generally for gravity disturbance, not gravity anomaly.
• Line 130: The Laplacian equation only holds true outside the earth.
• Line 235: The number of decimal separator may be more.
• Line 255: How to distinguish wave lengths?
• Figure 1: Here should be GMs.
• Figure 3: The resolution of the figure is low.
• Figure 5: The resolution of the figure is low.