NASA GSFC mascon solutions
|GSFC_global_mascons_v02.2||HDF5 ASCII|| Standard solution:
|GSFC_global_mascons_v02.2-ICE6G||HDF5 ASCII|| ICE6G GIA model removed [Peltier et al., 2015]
|GSFC_global_mascons_v02.2-GeruoA||HDF5 ASCII|| GeruoA GIA model removed [A et al., 2013]
|GSFC_global_mascons_v02.2+ECMWF+MOG2D||HDF5 ASCII||ECMWF+MOG2D atmosphere and ocean de-aliasing model restored||Global|
|GSFC_ocean_mascons_v02.2_OBP-GeruoA||HDF5 ASCII|| GAD restored and GeruoA GIA model removed
|GSFC_ocean_mascons_v02.2_SLA-GeruoA||HDF5 ASCII|| GAD restored; mean atmospheric pressure removed; GeruoA GIA model removed
Previous mascon versions: v01.0/v01.1
HDF5 format documentation: GSFC_mascons_HDF5_format.pdf
Forward models and the GSFC solution strategy:
Since the beginning of the GRACE mission it has been standard practice to apply atmosphere and ocean de-aliasing products when processing the Level 1B data in order to directly remove these high-frequency signals from the inter-satellite measurements and the distributed gravity solutions. [Luthcke et al., 2006] and [Sabaka et al., 2010] demonstrated the benefit of also modeling global hydrology towards a further reduction in the measurement residuals and the mitigation of signal leakage. The GSFC mascon estimation procedure leverages this strategy by modeling global hydrology (GLDAS/Noah), glacial isostatic adjustment (GIA) [Peltier et al., 2015], and the largest co-seismic events during the span of the mission [Han et al., 2013]. In practice what we actually estimate is a "Delta" to these forward models, meaning that the fundamental resolution of GRACE (300-400 km) is only limited to the portion of the gravity field that is not captured by the various models and datasets. We also employ solution iteration [Luthcke et al., 2013], where the estimated updates to the mascons are applied as corrections to the forward model for a subsequent processing of the Level 1B data. This iterative process maximizes the role of the GRACE data on the final solution and minimizes the possible influence of the signal covariance design. The final solution is equal to the sum of the iterated “Delta” and the global hydrology model that was applied in the initial processing step. Over land and ice the mascons describe the total variability in water content, and over the ocean we provide separate solutions for ocean bottom pressure and sea level anomalies. The global solutions are available with two different GIA model corrections, and we have elected to not restore the co-seismic signal.
De-aliasing products and ocean-only mascons:
The GSFC ocean bottom pressure (OBP) and sea level anomaly (SLA) products make use of the GAC and GAD, which are products distributed by the GRACE Project. Over the ocean, both the GAC and GAD contain the sum of the OMCT non-tidal ocean model and the ECMWF atmospheric variability. They differ in that the GAC contains the vertically integrated atmospheric mass variability required for precise orbit determination, while the GAD applies the atmospheric surface pressure as required to relate GRACE measurements to ocean bottom pressure. The definitions of GSM, GAC and GAD products are described by Bettadpur  and Flechtner et al. . Ocean bottom pressure from GRACE Project spherical harmonics is computed by: GSM+GAD-GIA. The GSFC procedures currently apply a different ocean model (MOG2D) in the Level-1B data processing than the GRACE Project (OMCT). In order to provide ocean products that are consistent with other solutions, we must restore our own de-aliasing model (ECMWF+MOG2D) and then remove the GRACE Project de-aliasing model (GAC) prior to restoring the GAD. To produce a product (SLA) that is comparable to steric-corrected altimetry measurements of sea level anomalies, we must remove the spatial mean of atmospheric surface pressure over the ocean, which is equal to the mean of the GAD as the non-tidal ocean component conserves mass.
When using this data please cite:
Luthcke, S.B., T.J. Sabaka, B.D. Loomis, et al., 2013, Antarctica, Greenland and Gulf of Alaska land ice evolution from an iterated GRACE global mascon solution, J. Glac. 59(216), 613-631, doi:10.3189/2013JoG12J147.