Feedbacks Among Forestry, Agriculture, and Fire in Amazonia

As the rural economies of the Amazon Basin become tied to world markets, new pulses of logging and deforestation will follow. Potential consequences of this trend include a reduction of forest-dependent rainfall, the spread of savanna-like scrub vegetation, substantial losses of biodiversity, and increased greenhouse gas emissions. Greater investments in governance of new highway corridors and the strengthening of forest-based economies could favor a more equitable and sustainable developmental pathway that maintains the forest ecosystem. Policy interventions that could foster this alternative pathway are poorly understood, however, in part because of the complex interactions among ecological, economic, and climatic systems operating across multiple scales spatial and temporal scales in the Amazon region. The interdisciplinary research project will develop quantitative, spatially explicit models of Amazon land-use change, fire, vegetation dynamics, and carbon flux for a major new highway corridor and for the entire basin. The ecosystem model will describe fire risk as a function of drought, land use, and disturbance history, and will be coupled to the economic models through land-use effects on the fire regime and through fire-risk effects on the behavior of land users. The ecosystem model will track vegetation structure and recovery with associated carbon pools to estimate the carbon fluxes and flammability associated with land-use scenarios. The economic models will first calculate the potential profitability of each pixel for competing agricultural and forestry land uses as a function of infrastructure and biophysical factors determined using satellite data and a large GIS database. Land uses that actually occupy each pixel will depend on population and urbanization variables that are captured within an econometric model of municipal-level census data and spatial associations that are represented within a cellular automata model. Ecological and economic modeling will be supported by field research already underway along two of the five major highway corridors that are being paved in Amazonia. The proposed research would develop the first spatially explicit simulation model for a tropical frontier region that integrates the rent-seeking behavior of land users with the ecological and climatic processes that shape this behavior. The model will be designed to assess the responses of regional economies and ecosystems to infrastructure investments, regulatory and fiscal interventions, macroeconomic factors, and climatic change, and may therefore increase the scientific quality of the regions policy-making process. This research provides the foundation for a program of interdisciplinary education that supports the training of at least 16 doctoral and 15 masters-level students. Interdisciplinary graduate field courses, a tri-lingual textbook, and an educational simulation model will be developed in collaboration with an NSF IGERT program. This project is supported by an award resulting from the FY 2004 special competition in Biocomplexity in the Environment focusing on the Dynamics of Coupled Natural and Human Systems.