Projected Impacts of Wildfire Emissions on Air Quality by Mid-century in the U. S. Southeast

Wildfires can severely impair the health of ecosystems, life forms and regional economies. In the rapidly changing U. S. Southeast, both climate and socioeconomic factors (e.g., population and income) drive wildfires, and need to be represented in wildfire inventories to assess the air quality (AQ) impacts and health risks of wildfires long-term. This motivated the development of a wildfire emissions projection methodology leveraging published models of annual areas burned (AAB) based on county-level socioeconomic and climate projections for 2011-2060. It is applied to project two sets of AAB, each with a different climate downscaling approach, to estimate wildfire emissions for 2010 and four mid-century years. Competing climate and socioeconomic factors result in 7% - 32% lower projected AAB than 18-year historical mean AABs, and 13% - 62% lower fine particulate matter (PM_2.5) emissions than estimated from the historical AAB in the selected years, with climate driving their temporal variability.

Evaluation of the two sets of emissions projections in air quality (AQ) simulations against those using the National Emissions Inventory (NEI), and network observations for 2010 show little difference among the projection methods in ozone (0.08% - 0.93%) and PM_2.5 (1% - 8%). Larger, comparable biases relative to observations in all three methods for secondary species, especially in winter, are found attributable to non-wildfire emissions or secondary chemical production. The projection methods predict primary wildfire PM better than the NEI, providing confidence that they can assess current wildfire AQ impacts, while enabling longer-term AQ assessments unachievable with static inventories.

AQ simulations using the projected wildfire emissions, and projected emission reductions in SO_x and NO_x from energy and transportation (by up to 80% at mid-century) show peak periods and locations of wildfire impacts on ozone and PM shifting from autumn in Midwestern locations in 2010, to warmer and drier summers east and south by mid-century, following the AAB spatiotemporal patterns. Although considerably lower than 2010 levels, summertime PM_2.5 increases by 4%-5% over the 2040-2060 period in this emission scenario, driven by increases in OC and unspeciated other PM.