Dryness stress is a crucial limitation on vegetation growth (Liu et al. 2020). It can change vegetation water use efficiency, drive tree dieback and affect carbon cycle of terrestrial ecosystem (Yang et al. 2016; Humphrey et al. 2021; Zemp et al. 2017; Green et al. 2019). High vapour pressure deficit (VPD; a measure of atmospheric dryness) and low soil moisture (SM) are two main drivers of dryness stress on vegetation.
However, the mechanism of vegetation response to drought is still under debate, leading to the uncertainties in estimating the impact of climate change on ecosystem carbon flux. On the one hand, increased vapour pressure deficit (VPD; a measure of atmospheric dryness) triggers stomatal closure, a mechanism for reducing water loss, leading to a decline in photosynthesis. On the other hand, low soil moisture (SM) makes plants unable to have sufficient water to support their transpiration (usually accompanied by carbon assimilation), resulting in low photosynthetic rate. More importantly, the two variables are strongly coupled through land atmosphere interaction (Humphrey et al. 2021; Lansu et al. 2020). Under the condition of low soil water content, land evapotranspiration is limited, and more solar radiation is partitioned into sensible heat to warm the surface, causing an increase in air temperature, a decline of relative humidity and an increase of VPD. The subsequent drought will further reduce the soil water content and form a positive feedback (Figure 1). The relative role of SM and VPD in limiting vegetation productivity is still under debate. This is also reflected in the model structures. For instance, one version of the P model (Stocker et al. 2020) includes soil moisture stress function while others do not have this part.
At present, researchers try to disentangle the respective SM and VPD limitations through satellite observation, field experiment and model simulations. Based on global gridded satellite observations and climate data sets, Liu et al. (2020) used control experiments to disentangle the effects of SM and VPD in the climate space of SM and VPD. They examined the sensitivity of photosynthesis to dryness stress and found that SM, rather than VPD, dominates dryness stress in vegetation production. This main finding is counter to some field experiments at leaf level. For example, Collatz et al. (1991) found that increased VPD led to the decrease of stomatal conductance, consequently resulting in a low carbon flux. In addition, Humphrey et al. (2021) used earth system model simulations to separate the effects of soil moisture and soil moisture–atmosphere feedback. They found that the feedback amplifies humidity (VPD) anomalies and enhances the direct effects of soil water stress.
It is therefore important to account for feedback between soil and atmospheric dryness when estimating the response of vegetation to dryness. Furthermore, if the models do not correctly consider the SM-VPD coupling, they may not be able to adequately predict the dryness stress on ecosystems and the long-term response of carbon cycle to a warming climate.