I certainly have had the impression that both agriculture and natural ecosystems are becoming more water scarce due to global warming, and that this is going to be a big problem at some point. We hear that the Amazon (River basin) may be tipping into an arid ecosystem, which has implications for the entire global climate and food supply, for example. But this article in Water Resources Research suggests there may be some feedback loops being overlooked. If I can try to summarize in a couple sentences, the concern is that higher temperatures cause greater evaporation from soil and transpiration (evaporation of soil moisture through the pathway of plant roots and leaves), and this will lead to drying both in agricultural and natural ecosystems. But plants have mechanisms to resist this loss of moisture, specifically by closing stomata which are the openings in leaves through which transpiration takes place. The mechanism can offset some but not all of the increased drying effect.

The CO₂ Balancing Act: Why Global Warming and Greening Don’t Dry Earth as Much as We Thought

While air warming and vegetation greening are widely assumed to intensify terrestrial drying through enhanced evapotranspiration, rising atmospheric CO₂ concentration ([CO₂]) may counteract this effect by inducing stomatal closure and reducing water depletion. However, the complex interplay between these factors has obscured their net impact on global terrestrial drying. Here, we develop a model that physically and effectively quantifies the relationships among evapotranspiration, [CO₂], and climate and vegetation changes, which can explicitly reflect how CO₂-mediated stomatal regulation interacts with climate and vegetation changes to modulate evapotranspiration. We find that, globally, the drying effects of warming and greening are largely offset by CO₂-induced reductions in surface conductance (69.4% ± 16.9%) and associated meteorological feedbacks. This compensatory mechanism is overlooked in traditional drying indicators, that is, the Palmer Drought Severity Index (PDSI) overestimates trends in drought-affected area (63.2% ± 10.1%), drought duration (58.7% ± 9.5%), and drought intensity (43.9% ± 7.7%) during 1982–2014 by ignoring CO₂-vegetation-climate interactions, and similarly, potential evapotranspiration-based aridity index underestimates wetting trends by 66.1% ± 3.5%. Our results reveal a systematic bias in current global drying assessments, which exaggerate drying in aridifying regions while underestimating wetting trends elsewhere. These findings reinterpret the hydrological impacts of global change, demonstrating that [CO₂] rise acts as a critical buffer against terrestrial drying. The study provides a mechanistic framework to reconcile observed greening with hydrological trends, offering transformative insights for ecohydrological modeling and water resource management in a high-[CO₂] climate.