Modelling oldfield species richness in a mountain area

Abstract

This paper is an attempt, using statistical modelling techniques, to understand the patterns of vascular plant species richness in mountain pastures ensuing from agriculture abandonment. Species richness was measured in 10 × 10 m plots and the explanatory factors used in the models were physical variables calculated from a digital terrain model (DTM) and vegetation, lithology, and administrative regions maps all constructed in a 100 m-resolution GIS environment. Poisson regressions, a particular type of GLM (generalized linear models), were used to construct the models using a forward stepwise procedure. The resulting model included three topo-ecological variables derived from the DTM (mean annual rainfall, specific catchment area and distance from the nearest watercourse) and accounted for 70% of the total variability. Maximum species richness occurred at intermediate rainfall levels, at topographic positions with relatively little drainage or in areas distant from watercourses. The model is discussed in the context of current theories on species richness.