Testing new potential evaporation formulations for identifying soil moisture deficiency in agricultural areas under global warming

Anthropogenic global warming shortens the residence time of soil moisture and raises atmospheric evaporative demand (E_p), posing significant threats to agricultural productivity. Conventional E_p estimates—such as the reference evapotranspiration (E_ref) derived from the Penman–Monteith equation—assume fixed vegetation properties, thereby neglecting dynamic physiological responses to elevated CO₂ (eCO₂) and vapor pressure deficit (VPD). In this work, we extended recent E_p formulations by transitioning from a big‐leaf to a two‐source framework that separately represents vegetation and bare soil, explicitly capturing the effects of Earth's greening and dynamic stomatal responses to eCO₂ and VPD. Using global datasets, we generated ratios of actual evapotranspiration (E) to E_p for historical (1985–2020) and future (2021–2100) periods under multiple CO₂ emission scenarios. The E/E_p ratios were converted into the Evaporative Stress Index (ESI) and compared with the standardized soil moisture index (SSI). Results showed that, in agricultural regions, the refined E_p formulations exhibited gentler upward trends compared to E_ref—primarily due to the moderating influence of increased stomatal resistance (r_s) under rising VPD. Future projections further suggested that r_s responses to eCO₂ and VPD might partially offset increases in E_p, implying that future plant water stress may be less severe than conventional indices predicted. Nonetheless, persistent discrepancies between ESI and SSI projections underscore the critical role of non-meteorological drivers in shaping future evaporative demand and highlight the need for further research to resolve the remaining uncertainties.