A stochastic model for vegetation water stress

Soil water limitations cause water‐stressed conditions in vegetation because of temporary and/or permanent damage to plant tissues. Vegetation water stress is here assumed to increase when soil moisture is below a threshold level, s *, and to decrease otherwise. The crossing properties of s * allow one to describe the temporal evolution of water stress, Φ, as a dichotomic Markov process. The available analytical solutions for dichotomic processes are used to determine the probability density function of Φ and the mean first‐passage times (MFPT) of a water stress threshold. MFTP is used as a measure of the variability of water stress. The use of MFPT in conjunction with the average stress levels allows us to provide a more complete characterization of health/stress conditions in vegetation. We investigate the influence on the dynamics of vegetation water stress of different climatic, pedological, and plant physiological parameters. From these analyses, we find, for example, that resistant species are favoured in relatively humid environments, whereas resilient shallow‐rooted plants have an advantage in drier conditions.