ReWildNet: A Spatial Modeling Approach for Evaluating Rewilding Outcomes in Altered Ecosystems

Abstract

As a restoration approach gaining increasing attention from conservationists, rewilding has been proposed to reverse the loss of biodiversity in degraded lands and ecosystems. Nevertheless, the prediction of spatial and ecological effects of a particular rewilding project is difficult to be made, mainly because of the complexity of interactions between species and habitats and landscape level connectivity. In order to predict rewilding results, the present study proposes ReWildNet, which is a spatial model framework allowing the simulation and evaluation of the effects of rewilding interventions in altered ecosystems. The model uses information about spatial connectivity, species distributions and landscape to predict the ecological impact of different rewilding projects and compare several types of interventions, such as habitat corridors restoration and reintroductions, before their application. Based on the findings, the importance of landscape-level spatial connectivity and heterogeneity in the success of rewilding initiatives is confirmed, as well as limited effectiveness of partial rewilding approaches in comparison with others. This model provides decision-making tool for conservation practitioners to select those intervention strategies that promote ecological resilience and potential of species recovery. This study contributes to the literature on spatial ecology by showing the role of predictive modeling in developing evidence-based plans for rewilding in fragmented and anthropogenic environments.