Gravity Recovery and Climate Experiment (GRACE) alias error from ocean tides

Gravity Recovery and Climate Experiment (GRACE) is able to observe interannual and longer‐period mass redistribution driven by climate changes. However, alias errors due to inaccurate ocean tide models may contaminate these signals. To quantify this problem, we estimate alias contamination due to expected errors in eight tidal constituents (Q 1 , O 1 , P 1 , K 1 , N 2 , M 2 , S 2 , and K 2 ) for the period August 2002 to February 2006. Differences between GOT00.2 and TPXO6.2 tide models are used as estimates of tide model errors. The alias estimate is based on an approximate method employing least squares fits of spherical harmonics to potential differences between the two GRACE satellites. Results from this method are similar to previous estimates, confirming this is a useful and efficient way to study the alias problem. New findings are (1) the period of the M 2 tide alias is about 140 d, much longer than the predicted period (13.5 d) because of monthly sampling of the 13.5‐d period; (2) there are two alias periods for K 1 (one about 90 d and one longer than 7 years); and (3) the slow decay in the GRACE orbit causes tide component aliases to differ from pure frequencies as the orbit evolves. We also find that aliases can contaminate in both space and time and thus may affect estimates of mass changes over land. Finally, we compare the simulated S 2 alias with the estimated S 2 alias in GRACE data. They are similar, but our approximate method appears to overestimate alias error over high‐latitude oceans and along certain coastlines while underestimating error over low‐latitude oceans.