Smoke from fire is a local, regional and often international issue that is growing in complexity as competition for airshed resources increases. BlueSky is a smoke modeling framework designed to help address this problem by enabling simulations of the cumulative smoke impacts from fires (prescribed, wildland, and agricultural) across a region. Versions of BlueSky have been implemented in prediction systems across the contiguous US, and land managers, air-quality regulators, incident command teams, and the general public can currently obtain BlueSky-based predictions of smoke impacts for their region. A highly modular framework, BlueSky links together a variety of state-of-the-art models of meteorology, fuels, consumption, emissions, and air quality, and offers multiple model choices at each modeling step. This modularity also allows direct comparison between similar component models. This paper presents the overall model framework Version 2.5 – the component models, how they are linked together, and the results from case studies of two wildfires. Predicted results are affected by the specific choice of modeling pathway. With the pathway chosen, the modeled output generally compares well with plume shape and extent as observed by satellites, but underpredicts surface concentrations as observed by ground monitors. Sensitivity studies show that knowledge of fire behavior can greatly improve the accuracy of these smoke impact calculations.
Estimating population exposure to particulate matter during wildfires can be difficult because of insufficient monitoring data to capture the spatiotemporal variability of smoke plumes. Chemical transport models (CTMs) and satellite retrievals provide spatiotemporal data that may be useful in predicting PM2.5 during wildfires. We estimated PM2.5 concentrations during the 2008 northern California wildfires using 10-fold cross-validation (CV) to select an optimal prediction model from a set of 11 statistical algorithms and 29 predictor variables. The variables included CTM output, three measures of satellite aerosol optical depth, distance to the nearest fires, meteorological data, and land use, traffic, spatial location, and temporal characteristics. The generalized boosting model (GBM) with 29 predictor variables had the lowest CV root mean squared error and a CV-R2 of 0.803. The most important predictor variable was the Geostationary Operational Environmental Satellite Aerosol/Smoke Product (GASP) Aerosol Optical Depth (AOD), followed by the CTM output and distance to the nearest fire cluster. Parsimonious models with various combinations of fewer variables also predicted PM2.5 well. Using machine learning algorithms to combine spatiotemporal data from satellites and CTMs can reliably predict PM2.5 concentrations during a major wildfire event.
Plume injection height influences plume transport characteristics, such as range and potential for dilution. We evaluated plume injection height from a predictive wildland fire smoke transport model over the contiguous United States (U.S.) from 2006 to 2008 using satellite-derived information, including plume top heights from the Multi-angle Imaging SpectroRadiometer (MISR) Plume Height Climatology Project and aerosol vertical profiles from the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP). While significant geographic variability was found in the comparison between modeled plumes and satellite-detected plumes, modeled plume heights were lower overall. In the eastern U.S., satellite-detected and modeled plume heights were similar (median height 671 and 660 m respectively). Both satellite-derived and modeled plume injection heights were higher in the western U.S. (2345 and 1172 m, respectively). Comparisons of modeled plume injection height to satellite-derived plume height at the fire location (R2 = 0.1) were generally worse than comparisons done downwind of the fire (R2 = 0.22). This suggests that the exact injection height is not as important as placement of the plume in the correct transport layer for transport modeling
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.