Do Models Mis-Represent Evaporative Regimes in Arid and Semi-Arid Regions?
Zhu, L., Simpson, I. R.. (2025). Do Models Mis-Represent Evaporative Regimes in Arid and Semi-Arid Regions?. , doi:https://doi.org/10.5065/kmpe-rw70
| Title | Do Models Mis-Represent Evaporative Regimes in Arid and Semi-Arid Regions? |
|---|---|
| Genre | Manuscript |
| Author(s) | Lilian Zhu, Isla R. Simpson |
| Abstract | Observed atmospheric humidity trends since 1980 disagree with global climate models in arid and semi-arid regions of the world (Simpson et al., 2024). Specifically, model simulations depict an increase in humidity, which has not been observed. This discrepancy is present in both climate models of the Couple Model Intercomparison Project, phase 6 (CMIP6) and a high resolution (4 km) CONUS404 WRF simulation that is additionally constrained by ERA5 at the boundaries. One potential cause of this discrepancy is a misrepresentation of water availability from the land surface from the perspective of evaporative regimes, i.e., how evaporative fraction (EF) relates to soil moisture. If models either tend to exist more in a radiation-limited regime, or tend to show a weaker dependence of evapotranspiration on soil moisture within the transitional regime, this could lead to evapotranspiration declining more in reality than in models as soil moisture declines with warming. Here, we will present an assessment of the climatological representation of evaporative regimes through consideration of daily timescale variability. Observed relationships between EF and soil moisture from Ameriflux tower sites in the U.S. Southwest will be compared against CONUS404 and the CESM2 Large Ensemble (LENS2). In order to quantify differences in the evaporative fraction and soil moisture relationship, we use the segmented regression approach of Hsu and Dirmeyer (2022) to determine the slope of the EF-soil moisture relationship and the fraction of days spent in the transitional regime to assess the similarity of models relative to observations. We find that there is no systematic difference between models and observations across the nine sites considered in the relationship between soil moisture and evapotranspiration. Observations generally sit within the spread of LENS2, and there is no consistent bias between CONUS404 and observations across the nine stations. This indicates that whatever the cause of the discrepancy in long-term trends may be, it is not apparent through the consideration of high frequency variability or the soil moisture and evaporative fraction relationship. |
| Publication Title | |
| Publication Date | Aug 1, 2025 |
| Publisher's Version of Record | https://doi.org/10.5065/kmpe-rw70 |
| OpenSky Citable URL | https://n2t.net/ark:/85065/d7n58rv3 |
| OpenSky Listing | View on OpenSky |
| CGD Affiliations | CAS |