Abstract ID: 452

Quantifying biomass burning in Amazonia using integrated moderate-to-coarse spatial resolution remote sensing data

Satellite active fire detection data represent one of the most important parameter for quantifying vegetation fire activity in Amazonia. Current satellite products extend beyond the simple binary classification of fire pixels (i.e., fire; no fire) providing additional information on sub-pixel fire characteristics including area, temperature and instantaneous radiative power. In this study, we implemented a comprehensive assessment of fire activity in Brazilian Amazonia using MODIS and GOES data at 1 km and 4 km spatial resolution, respectively. Higher spatial resolution data from ASTER and Landsat ETM+ instruments were used along with additional ancillary data including precipitation and percentage tree cover to substantiate our analyses. A new integrated map of fire activity combining the MODIS Thermal Anomalies (MOD14) and the GOES Wild Fire Automated Biomass Burning Algorithm (WF-ABBA) fire detection products was generated incorporating information on commission and omission errors derived at the pixel level. Initial results using 2005 data are presented and the merits of each of the two fire products analyzed discussed in light of the distinct regional fire activity characteristics.

Session:  Fire - Fire, drought, and changes in vegetation structure and composition

Presentation Type:  Poster