Sensitivity analysis and validation with experimental data of a land surface exchange model, for use in drought and desertification studies

As it is generally known, drought and desertification represent not only a climate problem, but they are also related to the ecosystem, in particular to all those processes taking place among atmosphere, vegetation and soil. In order to get a quantitative description of the above mechanisms, it is necessary to develop a mathematical model in which the climatic, hydrological and even biological processes interact with each other. In the past few years, this kind of modelshave been designed (Deardorff 1978, Dickinson 1981 and 1985, Sellers 1985, Lin 1985) and, more recently, one of the authors (Ji er al., 1989) has proposed a simple land process model (LPM). In the frame of the Drought and Desertification Project Land 2 n. 305, DMP Subproject (modeling), a sensitivity analysis and validation of this last has been carried out, with the aim to parametrize exchange of physical properties (heat, moisture, momentum, etc.) from ground surface to the lowest atmosphere, by using observation data gathered in different sites, belonging to different climatic regimes. This report discusses the main results which have been obtained in the present study with reference to the following topics: 1) improvement of soil structure parametrisation allowing an extended class of terrain types to be included in the model; 2) assessment and validation of model performance by using experimental data observed in real time; 3) sensitivity of model outputs to changes in vegetation cover and surface albedo which, to a certain extent, relate to human activities; 4) simulations of surface energy budgets in arid and other climatic zones.