GRACE-Based Hydrological Metrics for Improving Earth System Models

Abstract:

From a hydrological modeling perspective, terrestrial water storage changes are often viewed as the consequence of both changes in atmospheric forcing (such as precipitation, temperature, radiation, etc.) and local conditions (e.g. soil texture, vegetation type, ...).

However, in fully coupled Earth system models used for climate simulations, land hydrology also exerts a strong feedback on the atmosphere and the biosphere. For instance, droughts play a key role in generating surface temperature extremes and low relative humidity, thus amplifying the overall response of the vegetation to dry conditions. Used in combination with other satellite sensors and long-term climate datasets, GRACE/GRACE-FO measurements can be used to derive high-order metrics (such as residence time, plant rooting depth, and carbon flux sensitivities to water storage anomalies) to evaluate and better constrain such relationships in Earth system models.

Here we focus on the response of photosynthesis and carbon fluxes to hydrological droughts. Using outputs from the Coupled Model Intercomparison Project (CMIP), we show how shortcomings in the representation of water storage variability can affect long-term freshwater trends and lead to an over- or under-estimation of the vegetation response to droughts.