New approach for modeling climatic data with applications in modeling tree species distributions in the states of Jalisco and Colima, Mexico

Abstract

Spatial models of monthly climatic data in the states of Jalisco and Colima, Mexico, are developed using a combination of satellite imagery, topographic data and climatic data from 256 weather stations. The models accounted for 45–85% of the variability in the monthly temperature, precipitation and evaporation. In spite of having highly skewed distributions, cross-validation showed the models to have nominal prediction bias. The monthly climatic models for temperature, precipitation and evaporation were used to define 12 climate zones. Comparing the climatic zones against observed patterns of vegetation showed that the model captured the general placement of arid, semi-arid, temperate and tropical dry forests. Distribution models for four tree species are derived based on climatic constraints controlling their abundance. Site conditions are spatially simulated using Geographic Information Systems and represent resource gradients (temperature, precipitation, evaporation) and indirect variables, which reflect soil properties (topographic data). The application of the climatic zones in modeling the spatial distribution of vegetation types, soil texture and other related properties, identifying habitats suitable for selected endangered and threatened species and land use planning are discussed.