Presentation description

Smoke poses significant respiratory health risks. Smoke transport models help us understand when and how severe bad air quality days will be, allowing us to minimize these risks. Fire emissions inventories (FEIs) provide estimates of fire emissions, but rely on satellite fire detection, resulting in data gaps. This study evaluates two FEI improvements from a previous study for daily smoke transport modeling in the western U.S. for three representative fire years (2013, 2016, 2018). Starting with the Wildland Fire Emissions Inventory System (WFEIS) Moderate Resolution Imaging Spectroradiometer (MODIS) product for daily temporal resolution, previous improvements include incorporating Monitoring Trends In Burn Severity (MTBS) burned area measurements and interpolating during short data gaps. The MTBS product has increased spatial resolution, useful for small fire detection. Interpolation during short data gaps helps model fire emissions when obscured from MODIS fire detections (e.g., cloud cover). This study simulates forward trajectories for each fire using the Hybrid Single-Particle Lagrangian Integrated Trajectory model (HYSPLIT). The North American Mesoscale (NAM) model is used for meteorological inputs to HYSPLIT every 3 hours at a 12km horizontal resolution. In R, daily average smoke concentrations were calculated for each unique location and fire. The data were summed over four vertical levels, and the columnar sum and the surface layer were used for data analysis. We find that the FEI changes produced demonstrably different distributions and concentrations of smoke in HYSPLIT, appearing to fill gaps on days with missing fire detections, and increasing the overall amount of modeled smoke. Satellite smoke observations offered unclear results when compared with the HYSPLIT results, and more analysis is needed to determine the accuracy of the models. These FEI changes offer a promising way to improve smoke models, potentially leading to better predictions of bad air quality events. Results from this project will help inform a larger smoke exposure modeling study.