Data Assimilation in the Earth System

Numerical models have contributed significantly to improve our understanding of the dynamics of the
global water cycle. However, these models indicate limitations due to the uncertainty of input data,
boundary conditions, parameterisation, and imperfect model structure. Earth Observation (EO)
satellite missions provide invaluable estimates of atmospheric and hydrologic variables, which often
cover the entire globe. From these EO missions, the Gravity Recovery And Climate Experiment
(GRACE) and its follow-on mission (GRACE-FO) measurements can be used to estimate Terrestrial
Water Storage Changes (TWSC), i.e. a vertical summation of surface and sub-surface water storage
changes. In addition, various satellite missions provide multi-decadal Surface Soil Moisture (SSM)
estimates, as well as Land Surface Temperature (LST). These missions include SMOS, SMAP,
MODIS, and Sentinel, which typically measure electromagnetic radiance emitted by the Earth surface
or sample waveforms returned from radar pulses. However, the relationship between the measured
radiance or waveforms and the quantities of interest might be incredibly complex. Therefore, a
strategic step for the Earth sciences involves merging data and models via data assimilation and model
parameter calibration techniques that is the research focus of the Data Assimilation in the Earth
System group.