18 Use of Satellite Remote Sensing Data for Modeling Carbon Emissions from Fires: A Perspective in North America

Accurate accounting of carbon cycling is paramount to understanding and modeling global climate change. At present, a considerable amount of global carbon uptake (~2 Gt/year) remains unaccounted for in the carbon budget. It has been argued that the missing carbon may be absorbed in the terrestrial biomes of the Northern Hemisphere (Tans et al., 1990), in particular the temperate and boreal forests in North America (NA), which could account for the bulk (1.7 Gt/year) of the missing carbon (Fan et al., 1998). Fire is a driving factor controlling the carbon dynamics in NA, which affects both the sign and magnitude of the carbon budget According to the modeling results of Chen et al. (2000), boreal forests in NA have undergone tremendous fluctuations in its carbon budget over the last 200 years (Fig. 18.1). Around 1880 when widespread severe fires released a huge amount of carbon into the atmosphere, the forests were a very large source of carbon (~140 GtC/year) while around 1940, the forests became a very large sink of carbon (~200 GtC/year) due to fast forest regeneration that absorbed a large quantity of atmospheric carbon. The net carbon exchange is now so small that it is being debated whether the boreal forest is currently a sink (Chen et al., 2000) or a source (Kurz et al., 1995). The close correlation between the carbon budget and fire activity demonstrates the importance of the accurate estimation of carbon emissions from fires. So far, the continental-scale estimates of carbon emission were made mainly for fires occurring over the forest ecosystem using ground-based fire datasets (French et al., 2000). Few attempts have been made employing remote sensing data from coast to coast. While ground-based data are valuable, they have certain limitations that can be overcome by remote sensing. Ground-based fire data are mainly restricted to total burned area with their quality and completeness varying from year to year and region to region. Remote sensing is capable of providing additional spatial and temporal fire information to improve fire emission estimations.