A new distributional model coupling environmental and biotic factors

Abstract

Species distribution models (SDM) are the spatial surrogate of the suitability of a species in the biophysical aspect, since they are based on predicting their presence using climatic and environmental indicators. SDMs are satisfactory at regional scales, where biological interactions such as predation and competition do not influence distribution. However, at the local scale, they are incomplete for characterizing the ecology of a species since the algorithms do not include information about biotic variables. In this paper, we present a mathematical model that couples biophysical and biotic interactions in a spatially explicit way. We used a distributional database of 12 species of sigmodontine rodents from Argentine Patagonia as a study case. We performed numerical simulations of the dynamics of each rodent from a stochastic and spatially explicit population model. The biophysical suitability of each species was modeled using Maxent, which generated an indicator of its patch colonization capacity. The vegetation cover of each patch was characterized with remote sensing indices, associating the coverage with the pressure of aerial predation. The effect of interspecific competition was modeled from the assembly rules. The initial occupation conditions for each species were proposed as known sites of occurrence, and the temporal evolution of these systems was compared with that obtained from using random initial occupation conditions. The obtained results not only enrich the characterization of the studied ecosystems but also underscore a remarkable predictive capacity. Aerial predation and competition for resources, in accordance with assembly rules, dynamically modify populations and their distributions, revealing intricate interdependencies among species. This novel modeling tool holds promise for forecasting potential ecological scenarios, providing valuable insights into the intricate web of species interactions and their influence on distribution patterns.