Biomass burning emission is one of the largest uncertainties in modeling the spatial and temporal variations of aerosols and their optical properties. We present in this study a method of estimating day-to-day variations of aerosol emission in the North American boreal forest based on (1) the MODIS fire counts, (2) the ecosystem-specific biomass availability that includes the above ground, peatland, and surface soil organic matter carbon storage, and (3) the severity of the fire (low, medium, and high) that is determined by the atmospheric stability and dryness.

The total emission of black carbon and organic carbon in July 2004 estimated in this study are 0.078 and 0.623 Tg/mon in the North American boreal forest, when intense fires occurred in Alaska and Canada. Such biomass burning emission estimated for every day of the month has been used in the Goddard Chemistry Aerosol Radiation and Transport (GOCART) model to simulate aerosol distributions over North America in July 2004. The GOCART model predicts the total aerosol burden and global distribution of the major aerosol types: black carbon (BC), organic carbon (OC), dust, sea salt, and sulfate, including its precursors DMS, SO2 and MSA. For this study, only BC, OC, and sulfate were considered. The model results are compared with the previous model runs, where the monthly estimate of aerosol emissions were obtained from the Global Fire Emissions Database version 2(GFEDv2). Model-derived aerosol optical depth was also compared with that measured by the satellite instrument MODIS and with concentrations and vertical distributions measured by aircraft in the Intercontinental Chemical Transport Experiment – North America (INTEX-NA), phase A experiments that took place in the summer of 2004.