Biomass Burning Emissions From Large and Mega Fires in East Siberia

Catastrophic boreal forest fires of human origin are expected to increase in the future climate, as warmer and dryer conditions in longer fire seasons. The amount of biomass consumed during boreal forest fires can vary significantly because of the large variation in amount of surface organic soil burned. The estimates of boreal forest fire emissions might be evaluated by using atmospheric chemistry transport model (CTM) and satellite observation of carbon monoxide (CO). Even if models agree with the current CO measurements, it would not necessarily guarantee that the models could predict the future fire emission well. However, the model should represent the response of CO to inter-annual changes due to boreal forest fires before extrapolating it to long-term prediction. The ability of models to predict the response of CO to short-term changes can be evaluated in part by examining how ambient concentrations of CO respond to day-to-day variations in East Siberia during two different years of 2002 and 2003 with different fire severity of large and mega fires, respectively. Here we discuss the issues of the representation of boreal fires in an open vegetation burning emission model and chemistry transport model, with a focus on the 2003 Siberian forest fires in contrast with the year 2002. MODIS products of burned area are used to estimate CO emissions from open vegetation burning in conjunction with Dynamic Global Vegetation Model. The combustion factor (CF) and emission factor (EF) for grassland fires are determined from the satellite observation of Normalized Difference Vegetation Index (NDVI), since this accounts for the seasonal and spatial variation in CF and EF. The CF for the organic soil carbon is based on literature values, as a function of the soil moisture and fire severity. Simulations of CTM are performed for daily emissions from boreal forest fires. When the height of a smoke plume is more than the planetary boundary layer (PBL), emissions in the model grid box containing the plume are released to the model layer corresponding to the MISR-derived plume height. Satellite observation of CO from MOPITT is used to evaluate the model performances in simulating temporal and spatial trends in the fire emissions. Both the model result and MOPITT observation capture high mixing ratio of carbon monoxide from Siberian forest fires.