Abstract. The probabilistic description of soil moisture dynamics is a relatively new topic in hydrology. The most common ecohydrological models start from the soil water balance, a stochastic differential equation where the unknown quantity is the function of the soil moisture, depending both on spaces and time. Most of existing solutions in literature are obtained in a probabilistic framework and under steady-state condition; even if this last condition allows the analytical handling of the problem, it has considerably simplified the problem by subtracting generalities from it. The steady-state hypothesis, used in many ecohydrological works, appears perfectly applicable in arid and semiarid climatic areas like those of African's or middle American's savannas, but it seems to be no more valid in areas with Mediterranean climate, where, notoriously, the wet season foregoes the growing season, thus recharging the soil moisture. This initial condition, especially for deep rooted vegetation, has a great importance by enabling survival in absence of rainfalls during the growing season and, however, keeping the water stress low during its first period. The aim of this paper is to investigate the soil moisture dynamics using a simple non-steady numerical ecohydrological model. The numerical model is able to reproduce soil moisture probability density function, obtained analytically in previous studies for different climate and soil conditions in steady state conditions. The proposed model gives both the soil moisture time-profile and the vegetation static water stress time-profile. From the former it is possible to extract the probability density function of soil-moisture during the whole growing season, while the latter allows the estimation of the vegetation response to the water stress. Here the differences between the analytical and the numerical probability density functions are presented, showing how the numerical model is able to capture the effects of winter recharge on the soil moisture. The dynamic water stress is numerically evaluated, implicitly taking into account the soil moisture condition at the beginning of the growing season. The model proposed here is applied in the forested river basin of the Eleuterio in Sicily (Italy).
A Novel Limit Distribution for the Analysis of Randomly Mistuned Bladed Disks
