Bare soil evaporation stress determines soil moisture - evapotranspiration coupling strength bias in land surface modeling

Authors: DONG, J. - US Department Of Agriculture (USDA); DIRMEYER, P. - George Mason University; LEI, F. - Mississippi State University; Anderson, Martha; HOLMES, T. - National Aeronautics And Space Administration (NASA); HAIN, C. - Nasa Marshall Space Flight Center; Crow, Wade

Submitted to: Geophysical Research Letters
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/25/2020
Publication Date: 10/17/2020
Citation: Dong, J., Dirmeyer, P., Lei, F., Anderson, M.C., Holmes, T., Hain, C., Crow, W.T. 2020. Bare soil evaporation stress determines soil moisture - evapotranspiration coupling strength bias in land surface modeling. Geophysical Research Letters. 47. e2020GL090391. https://doi.org/10.1029/2020GL090391.
DOI: https://doi.org/10.1029/2020GL090391

Interpretive Summary: Better monitoring of surface soil water availability improves our ability to forecast precipitation and maximum air temperature. However, this improvement is only realized if land surface models accurately link changes in soil moisture to associated changes in land-surface evapotranspiration (ET). Using remotely sensed estimates of both soil moisture and ET, this paper evaluates the ability of existing models to describe the soil moisture/ET relationship within the central United States. Results demonstrate that existing models generally over-estimate the strength of this relationship. We trace this overestimation to poor modeling of bare soil evaporation in land surface models and demonstrate that correcting this overestimation leads to the enhanced estimation of temporal variations in ET. Results of this study will eventually be applied to improve short-term numerical weather prediction within the central United States.

Technical Abstract: A clear trend in the increased frequency and severity of heatwaves has been observed in recent decades and is expected to continue under global warming. In global dry-wet transitional zones, the predictability of such events is enhanced by knowledge of soil moisture (SM) and local evapotranspiration (ET) coupling mechanisms. However, model-based estimates of SM-ET coupling strength (SECS) vary widely and are prone to bias. Here we apply numerical modeling and remote sensing to identify the process-level source of model SECS bias with the goal of improving the fidelity of current Earth system models. In particular, we generate ensembles of off-line land surface model estimates of SECS that capture model structure and parameter uncertainties related to model representation of bare soil evaporation (E) and transpiration (T) soil water stress, photosynthesis, rooting depth, and soil properties. Results illustrate that modeled SECS is most strongly determined by E stress, and (generally positive) SECS modeling bias is attributable to the oversimplification of soil texture impacts on E stress. Based on new remotely sensed estimates of SECS, we demonstrate that removing SECS bias via a single optimized E stress parameter leads to improved ET accuracy and resolves a well-known modeling bias in the partitioning of ET into E and T. As such, we highlight the importance of the stress function relating E and SM, and its central role in regulating land-atmosphere coupling processes impacting climate extremes.