Litter decomposition in Mediterranean ecosystems: Modelling the controlling role of climatic conditions and litter quality.

A new process-based model of litter decomposition, characterized by detailed climatic data input and simple litter quality parameters, is proposed. Compared to existing litter carbon models, specific implementations for temperature and moisture limiting effects have been adopted. The model is capable to represent decomposition processes in Mediterranean ecosystems, with summer drought slowing down, even at optimal temperatures, the litter decay rates of sclerophyll plants whose leaf masses are rich in structural compounds and low in N content. The model was calibrated by a best fitting procedure of two different datasets. First, unpublished results of litterbag experiments on leaf litter of 9 Mediterranean species, decomposing under controlled and not limiting temperature and water conditions, have been used to estimate the decay rate dependency from litter quality that was defined by only three initial C pools (labile, stable and recalcitrant compounds) instead of traditional N-based indices. Second, a set of published data from three medium-term field experiments on a single species, Phillyrea angustifolia, decomposing under different climatic conditions, have been used to estimate the limiting effects of temperature and moisture. The model was then validated against published data on seven other species and showed a correct reproduction of the major patterns of litter mass loss during decomposition processes of other seven different Mediterranean species. The model simulations, satisfactory for different litter types under a wide range of climatic conditions, suggest that factors which were not taken into account, such as initial litter N contents, microclimatic variations related to stand structure, soil chemistry and texture, and microbial communities, are not very significant for assessing decomposition dynamics in Mediterranean ecosystems. The minimal requirements of input data, the simple structure, and the easiness of parameterisation make our model, among the many other available litter carbon models, an attractive alternative for different research purposes, at least for Mediterranean ecosystems.