Plants are able to survive and function under extremely variable environmental conditions, including dramatic changes in soil water availability and atmospheric evaporative demand. This could not be achieved without powerful regulatory mechanisms at a variety of organizational and time scales, which allow plants to modulate water use in response to those changes. Drought-induced mortality can be interpreted as a failure of these regulatory mechanisms. Despite a large research effort, however, we lack a coherent framework able to accommodate the variety of plant responses to reduced water availability. Our first objective was to develop a new theoretical model to describe plant responses to drying soil conditions based on the relationship between two commonly measured ecophysiological variables: midday and predawn leaf water potentials. We then applied this scheme to a newly compiled global database of leaf water potentials including data from 103 species, in order to characterize their drought responses and the phylogenetic signal of the fitted model parameters, as well as their relationships with other key hydraulic properties at the species level.
Results/Conclusions
Under steady-state conditions, leaf water potential is a relatively simple function of soil water potential. This relationship depends on the functional responses of xylem hydraulic conductivity and stomatal conductance to soil water potential. If these responses are characterized using Weibull functions a new simplified equation is obtained that relates leaf to soil water potentials. This equation depends on two parameters: an intercept that measures transpiration capacity per unit of xylem water transport capacity (L) and a slope describing the relative sensitivity of stomatal conductance and xylem hydraulic conductivity to declining soil water potential (Cgk). The latter parameter can be interpreted as a quantitative measure of stomatal behaviour (isohydric vs anisohydric). Our new model was able to capture the variability of midday water potential as a function of predawn water potential in our dataset, both within and between species (overall R2 = 0.90). The obtained Cgk estimates imply that most species reduce stomatal conductance well before appreciable losses in hydraulic conductivity occur. Parameters L and Cgk showed a significant degree of phylogenetic conservatism and were negatively related, suggesting the possibility of a trade-off. We also found that species with tighter stomatal control (relative to their hydraulic vulnerability) were more vulnerable to xylem embolism. We believe that this new framework offers a promising tool to study drought responses of plants and to predict their vulnerability to drought-induced mortality.