Tropical deforestation is historically one of the largest drivers of biodiversity loss and carbon emissions globally. The growing demand for food, fiber and biofuels along with market’s globalization is expected to add further pressure on tropical deforestation in the coming decades. In this sense, a number of models have been proposed to explore future deforestation trends, particularly in the Amazon. However, none of these models plausibly captured the general trajectory of land cover change that has been observed in this region. This thesis provides evidence that previous modeling approaches were not able to consistently represent the forces that shape land use dynamics in the Amazon. In general they are restricted by either global or regional drives of land cover change. Therefore, an alternative modeling approach should be taken to explore cross-scale interactions such as the world demand for resources and land use regulations. The main objective of this thesis is to explore an innovative modeling approach for the Amazon which allows simulating how the global demand for agricultural commodities and different regional land use policies could affect future deforestation trends inside and outside the Brazilian Amazon, paying special attention to leakage effects over the Cerrado. A global economic model was taken to integrate supply and demand factors at both global and regional scales. Then a spatially explicit land-use model is used to explore future patterns of land cover change over the Brazilian Amazon and Cerrado biome. Leakage effects are simulated in two different ways, regarding land demand and land allocation. In the first case, leakage effects are determined by changes on the relative land rents of different land use types mediated by changes on regional land use policies. In the second case, leakage effects are simulated based on Spatial Lag technique for land demand allocation which accounts for the spatial dependence of the deforestation. Based on this approach six contrasting multi-scale scenarios are explored focusing on deforestation rates and spatial pattern analysis for both Amazon and Cerrado. Our results revealed that Amazon conservation might not be the end of deforestation in Brazil once it can lead 43% increase over the Cerrado cleared area up to 2050. Massive land cover changes would be expected throughout the Cerrado biome, especially on the Midwest region and over the emerging agricultural frontier of MATOPIBA (acronym formed by the first letters of the Maranhão, Tocantins, Piauí and Bahia Brazilian states). Biofuels targets compliance can further press land cover changes over this region revealing that productivity gains will be decisive for both Amazon and Cerrado conservation. In summary, biodiversity conservation and emissions reduction in Brazil will depend on broader land use policies and land use efficiency. Otherwise, managing a transition towards a more sustainable land use can become utopian.
Disponível em: mtc-m21b.sid.inpe.br/col/sid.inpe.br/mtc-m21b/2014/05.23.11.59/doc/publicacao.pdf