全球变化知识资源中心

Cumulus clouds make a significant contribution to the Earth's energy balance and hydrological cycle and are a major source of uncertainty in climate projections. Reducing uncertainty by expanding our understanding of the processes that drive cumulus convection is vital to the accurate identification of future global and regional climate impacts. Here we adopt an interdisciplinary approach that integrates interrelated scales from plant physiology to atmospheric turbulence. Our explicit simulations mimic the land-atmosphere approach implemented in current numerical weather prediction, and global climate models enable us to conclude that neglecting local plant dynamic responses leads to misrepresentations in the cloud cover and midtropospheric moisture convection of up to 21% and 56%, respectively. Our approach offers insights into the key role played by the active vegetation on atmospheric convective mixing that has recently been identified as the source of half of the variance in global warming projections (i.e., equilibrium climate sensitivity).

Plain Language Summary Our research paves the way to deepen our understanding of the essential role of biophysical spatiotemporal scales and the need of accurately representing them in atmospheric models. By applying first biophysical principles, our coupled plant and cloud-resolved simulations shows that an inaccurate representation of plant responses yields strong overestimations in regional cloud cover and in-cloud transport. This gives rise to uncertainties in weather predictions and climate projections. The underlying cause is found in a feedback driven by the cloud partitioned incoming direct and diffuse light. In combination with an asymmetric plant stomatal response, dynamic and heterogeneous patterns arise in the vegetation. This directly affects the surface energy balance that subsequently controls the overlying atmosphere, which results in reduced convection rates.