Comparison of two different ocean tide models especially with respect to the GRACE satellite mission

Abstract

The GRACE dual satellite mission (launched in March, 2002) offers the possibility of computing monthly highly accurate mean gravity fields over an expected lifetime of five years. Unfortunately, the quality of these monthly gravity field products does not yet reach the pre-launch expectations. Possible error sources might be insufficient instrument data processing and parameterization or modeling of short-term atmospheric and oceanic mass variations. Another candidate is the ocean tide model. Especially, incomplete subtraction (de-aliasing) of short period tides may be partially aliased into the monthly gravity field solutions. Therefore, we analyzed the difference of two ocean tide models (FES2004 and CSR 4.0) which are used at the GRACE Science Data System level-2 processing centers at GFZ Potsdam and CSR (Center for Space Research, Austin), respectively, and which may serve as a measure of the ocean tide model error. We have computed: a) straightforward monthly means of tidal elevation differences and b) simulations of tidal elevation differences at footpoints of GRACE A. The results of a) represent the differences of the monthly means of both tide models with respect to an uniform sampling (grid). The results of b) include the influence of spatially uneven sampling distribution (only along the orbit) and show that for the S2 and K2 tidal constituents, aliasing causes effects which cannot be neglected with respect to the presently achievable GRACE measurement accuracy for degrees n.le.7 (S2) and n.le.8 (K2).