Constraining CMIP6 soil water storage with GRACE satellite observations

Abstract:

Soil moisture is a key component of the hydrological cycle, and plays an important role in the surface energy budget and in water-carbon cycle interactions. Accurate representation of soil moisture in Earth System Models is necessary for the prediction of future hydrological extremes, freshwater availability, and terrestrial carbon uptake. Total soil moisture is determined by the interplay of sources (precipitation) and sinks (evapotranspiration, runoff). The 6th iteration of the Coupled Model Intercomparison Project (CMIP6) shows divergent future projections of soil moisture between the models.

Here, we use empirical orthogonal functions to compare patterns of monthly soil moisture variability in CMIP6 models to terrestrial water storage changes obtained from the GRACE satellites. Using this methodology, we identify which models best match the EOF patterns derived from GRACE and analyze what structural model characteristics may explain differences between the simulated and observed patterns of water storage variability. We investigate whether observed total water storage can be used as a constraint to reduce the spread of soil water storage projections in CMIP6 models.