Signal and noise in Gravity Recovery and Climate Experiment (GRACE) observed surface mass variations

Abstract

[1] The Gravity Recovery and Climate Experiment (GRACE) product used for this study consists of 43 monthly potential coefficient sets released by the GRACE science team which are used to generate surface mass thickness grids expressed as equivalent water heights (EQWHs). We optimized both the smoothing radius and the level of approximation by empirical orthogonal functions (EOFs) and found that 6.25°a nd three modes are able to describe more than 73.5% of the variance. The EQWHs obtained by the EOF method describe all known variations in the continental hydrology, present-day ice sheet melting, and global isostatic adjustment. To assess the quality of the estimated grids, we constructed degree error spectra of EQWHs. We conclude that a significant part of the errors in GRACE can be explained by a scaling factor of 0.85 relative to degree error estimates provided by the GGM02C gravity model but that the present-day errors in the GRACE data are a factor 2 to 5 larger than forecasted by tide model differences and atmospheric pressure differences. Comparison to a network of 59 International GNSS Service (IGS) stations confined the filter parameter settings to three EOF modes and 5°or 6.25°smoothing radius. Residuals that remain after the EOF method do exhibit S2 aliasing errors and a semiannual continental hydrology signal contained in the Global Land Data Assimilation Systems (GLDAS) model. Further analysis of the residual EOF signal revealed alternating track correlation patterns that are partially explained by the GRACE covariance matrix and the handling of nuisance parameters in the GRACE data processing.

Improvement of Global Hydrological Models Using GRACE Data

“…Information on individual storage components will be extremely valuable towards improving hydrological models efficiently for individual process descriptions. Progress in signal separation requires combined analyses using new observations, modelling, inversion techniques (e.g., Ramillien et al 2004) and advanced statistical methods for identifying dominant and significant signal components (e.g., Schrama et al 2007;Schmidt et al 2008a). …”

On the postprocessing removal of correlated errors in GRACE temporal gravity field solutions

“…One category is empirical that does not use any other information (Swenson and Wahr 2006;Chambers 2006;Chen et al 2007;Schrama et al 2007;Wouters and Schrama 2007;Davis et al 2008). The other category makes use of the error variance-covariance matrix of the SCs, but also requires a priori signal covariance information (Kusche 2007;Klees et al 2008).…”

“…Chambers (2006) and Chen et al (2007) followed the principle of Swenson and Wahr (2006), but chose to fit all the even/odd SCs of the same order to be decorrelated using a polynomial. Schrama et al (2007) and Wouters and Schrama (2007) made use of empirical orthogonal functions (EOFs). Their difference is that Schrama et al (2007) performed the computation in the spatial domain, and Wouters and Schrama (2007) performed the computation in the spherical harmonic domain.…”

“…Schrama et al (2007) and Wouters and Schrama (2007) made use of empirical orthogonal functions (EOFs). Their difference is that Schrama et al (2007) performed the computation in the spatial domain, and Wouters and Schrama (2007) performed the computation in the spherical harmonic domain. Davis et al (2008) presented a statistical approach.…”

Contribution of GRACE Satellite Gravimetry in Global and Regional Hydrology, and in Ice Sheets Mass Balance

“…The aliasing and the modelling errors are responsible for the spurious meridional undulations present in the GRACE monthly grids and known as north-south striping. They are associated to spherical harmonics order 15 and its multiples (Swenson & Wahr, 2006a;Schrama et al, 2007;Seo et al, 2008). Thompson et al (2004) showed that the degree error increased by factors ~ 20 due to atmospheric aliasing, ~ 10 due to the ocean model, and ~ 3 due to continental hydrology model.…”