A description and evaluation of WRF/Chem prediction of biomass burn emissions

A versatile and operational WRF/Chem smoke forecast system has been developed for Alaska during recent years. Type- and distribution characteristics of wildfire emissions were estimated using a biomass burn emission model and a time dependent plumerise model embedded inline in WRF/Chem. Model evaluation and examples of atmospheric effects of wildfires were derived from WRF/Chem runs including emissions from Alaska wildfires, which consume typically 400,000 hectares of mostly boreal forest annually. System evaluation with local particulate measurements and satellite imagery showed clearly the need for accurate fire source data, which were derived through a combination of Alaska Interagency Coordination Center fire information data and from MODIS MOD14 thermal anomalies. Comparison of the plume rise model with lidar measurement data revealed partly good coincidence of the coupled modeling system. Further system advantages were the demonstrated improved performance of the atmospheric host model by taking into account the impact of smoke particulates on atmospheric radiation transfer and humidity.