About project 4

Optimizing GRACE data for a better understanding of global freshwater system

Global terrestrial freshwater plays an important role in many respects, such as sea level raise, ecosystems and climate dynamics. However, current global hydrological models produce valuable information containing great uncertainty. An efficient way to reduce the uncertainty of model output is using optimized geodetic and satellite data. In our study, we use GRACE (Gravity Recovery and Climate Experiment) data to improve hydrological modelling.

The overall roles of P4 are (1) to provide the optimized GRACE data including uncertainty information for calibration and data assimilation (C/DA approach) and (2) to validate the results from C/DA approach against in-situ data and regional high resolution models.

GRACE, a joint mission by NASA and the German Aerospace Center, DLR (German: Deutsches Zentrum für Luft- und Raumfahrt e.V.), was launched 17 March 2002, and contains two twin satellites. Gravity is measured by referring to the distances between the two satellites. Gravity influences the moving speeds of both satellites and thus the distance to each other is constantly changed. The precision of the accuracy is within the width of a hair.

The traditional method to implement geodetic GRACE data is the use of spherical harmonics solutions. However, several studies have pointed out that biases and noises could be introduced by these solutions. In the recent years, an alternative solution MASCON was proposed but has not yet been deeply studied.

Spherical harmonics

Spherical harmonics were first presented by Hofmann-Welenhof and Moritz in 2006. They are the expression of harmonic functions, defining the surface of a sphere. Harmonic functions are the solutions for the divergence of the gradient potential, which can be expressed by ∆V=0. The biggest drawback of spherical harmonics is the required introduction of filtering to reduce the errors, which also leads to a reduction of the wanted signals.

MASCON (Mass Concentration blocks)

Unlike spherical harmonics, MASCON has no need for filtering. It is developed based on a regional parameterization. However, the information of errors caused by MASCON solutions is not clear yet. Hence there is a disagreement about which approach is the best.

At a later stage, we will use ground based in-situ data to validate the hydrological model results of spherical harmonics and mascon solutions.