Close Window

Estimating Interannual Variability of Carbon Emissions from Land Use-Related Fires in the Southern Amazon

Ruth DeFries, University of Maryland, College Park, rdefries@mail.umd.edu (Presenting)
Yosio Shimabukuro, INPE, yosio@dsr.inpe.br
Guido Van der Werf, Vrije Universiteit Amsterdam, guido.van.der.werf@falw.vu.nl
G.J. Collatz, Goddard Space Flight Center, jcollatz@ltpmail.gsfc.nasa.gov
Doug Morton, University of Maryland College Park, Douglas.C.Morton.99@Alum.Dartmouth.ORG
James Randerson, University of California Irvine, jranders@uci.edu

Fires from deforestation and pasture maintenance are major pathways for carbon emissions to the atmosphere. There are large uncertainties in the total carbon flux from land-use related fires, contributions from different types of land use, and interannual variability in emissions. We combine MODIS-derived information on the fate of deforestation (pasture or cropland) and active fire observations in an ecosystem model (DECAF) to estimate carbon emissions from land-use fires for 2001 to 2005 for the state of Mato Grosso. Model results allow partitioning of fire emissions according to fate of deforestation and provide annual estimates of emissions from deforestation fire and decomposition of remaining slash. Initial results indicate that fires for pasture clearing are the largest source of emissions in Mato Grosso overall but deforestation for cropland results in higher emissions on a per area basis. Fire emissions continue to occur several years after detection of initial deforestation due to repeated burning for clearing. Because of this offset between carbon emissions from deforestation fires and reported deforestation rates, we estimate that 2005 has relatively high emissions although deforestation rates dropped. Model estimates at 250m MODIS resolution in Mato Grosso provide insights into approaches for scaling across the entire basin at coarser resolution.

Science Theme:  LC (Land Use and Land Cover Change)

Session:  3C: Vegetation Structure and Disturbance

Presentation Type:  Oral

Abstract ID: 24

View presentation

Close Window