Modeling a decade of carbon gross emissions from forest fires in the Amazon: Conciliating the bottom-up and top-down views of the problem

Abstract

Fires in tropical forests can be considered as one of the most important environmental problems of the XXI century. In the world's largest rainforest, the Amazon, the frequency of fires increased over the last decade as a result of climate variability (drought in 1998, 2005 and 2010) and anthropogenic activities (Aragão and Shimabukuro, 2010, Science). During the 2010 drought, 0.48±0.18 petagrams of carbon per year (PgCyr1) was released to the atmosphere by fires (e.g. Gatti et al. 2014, Nature). However a question is still unanswered: What are the main sources of fire emissions? Are these emissions a result of fires from fast turnover grassland vegetation, with minor impact on the overall atmospheric CO2 accumulation, or are they related to slow turnover forest vegetation, with potentially high impact on the atmospheric CO2 accumulation? It is well known that Amazonian fires can spread to undisturbed forests, causing forest fires and possibly contributing significantly to land use and land cover change component of carbon emissions. Despite being an important component of the global carbon balance, however, the magnitude of forest fire impacts on the long-term stability of carbon stocks are still poorly quantified.In the proposed fellowship the Post-doctoral researcher will use and refine the first systematic bottom-up model for quantifying annual gross carbon emissions from forests affected by fires at the Amazon biome scale. The model called Fire-Associated Transient Emissions from fires in Amazonia (FATE-Amazon) developed by a multidisciplinary research team coordinated by Dr Aragão at the National Institute for Space Research (INPE), was recently applied by Anderson et al. (2015). The model consist in a spatially explicit bookeeping model approach based on remote sensing data, for extracting information on biomass, burned area and land cover types, and empirically derived forest biomass loss due to fires. This model allows disentangling the sources of carbon released to the atmosphere from fires. These estimates will support a comprehensive interpretation and cross-validation of the atmospheric-based analysis of the Amazon carbon cycle.