Interpreting the Earth's time varying geopotential as observed from space and comparisons to global models of hydrologic transport

Date

2004

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Measurements of temporal changes in Earthís gravitational field were measured using six years of satellite laser ranging (SLR) to Lageos-1 and Lagoes-2 and the results were compared to geophysical models of mass variability for the atmosphere, ocean, and continental hydrology. Annual estimates of spherical harmonic gravity coefficients (degree and order four expansion) derived from the SLR observations when compared to combinations of the mass models had degree correlations that generally exceeded the 90% confidence limit and agreed to about the 1 mm level in terms of geoid height anomaly. The Gravity Recovery and Climate Experiment (GRACE) is measuring Earthís gravitational field approximately every month at spatial scales of a few hundred kilometers. In order to achieve smaller temporal and spatial scales, it is necessary to account for the effects of short period, non-tidal, mass variability which was not previously included in other gravity determinations. Orbital simulations of GRACE showed that the highest degrees were impacted the most by unmodeled variability in the atmosphere, oceans, and continental hydrology (a factor of ~20 increase in degree error in the case of the atmosphere). The use of approximate models gave the greatest reduction in aliasing error for the mid-degrees and higher; however, the lowest degrees (~2-5) were dominated by the sensitivity of the GRACE processing system to systematic error. GRACE data processing that used a combined atmosphere-ocean de-aliasing (AOD) model showed improvement in the gravity estimates consistent with the simulations: the shorter spatial wavelengths (higher degrees) were improved while the longest spatial wavelengths (particularly important for time-variable gravity studies) were relatively unaffected. Monthly gravity solutions from GRACE resolved features on the order of 2-3 mm geoid height anomaly when smoothed to 400-km spatial scales. Comparisons with the Global Land Data Assimilation System (GLDAS) terrestrial water storage model indicated a high degree of correlation up to spatial wavelengths of 600 km or larger; a significant improvement over the spatial and temporal scales obtained with SLR observations. However, temporal variability in the degree 2 coefficients, particularly the zonal, seemed to be better resolved by SLR observations than by GRACE observations.

Description

text

Keywords

Citation