Toward integrated analysis of human impacts on forest biodiversity: lessons from Latin America.

TLDR: In this paper, the authors summarize the experience gained by an international collaborative research effort spanning more than a decade, focusing on the tropical montane forests of Mexico and the temperate rain forests of southern South America, both of which are global conservation priorities.

Abstract: Although sustainable forest management (SFM) has been widely adopted as a policy and management goal, high rates of forest loss and degradation are still occurring in many areas. Human activities such as logging, livestock husbandry, crop cultivation, infrastructural development, and use of fire are causing widespread loss of biodiversity, restricting progress toward SFM. In such situations, there is an urgent need for tools that can provide an integrated assessment of human impacts on forest biodiversity and that can support decision making related to forest use. This paper summarizes the experience gained by an international collaborative research effort spanning more than a decade, focusing on the tropical montane forests of Mexico and the temperate rain forests of southern South America, both of which are global conservation priorities. The lessons learned from this research are identified, specifically in relation to developing an integrated modeling framework for achieving SFM. Experience has highlighted a number of challenges that need to be overcome in such areas, including the lack of information regarding ecological processes and species characteristics and a lack of forest inventory data, which hinders model parameterization. Quantitative models are poorly developed for some ecological phenomena, such as edge effects and genetic diversity, limiting model integration. Establishment of participatory approaches to forest management is difficult, as a supportive institutional and policy environment is often lacking. However, experience to date suggests that the modeling toolkit approach suggested by Sturvetant et al. (2008) could be of value in such areas. Suggestions are made regarding desirable elements of such a toolkit to support participatory-research approaches in domains characterized by high uncertainty, including Bayesian Belief Networks, spatial multi-criteria analysis, and scenario planning.