Wildfire is common across the pan-Arctic tundra. Tundra fires exert significant impacts on terrestrial carbon balance and ecosystem functioning. Interactions between fire and climate change can enhance their impacts on the Arctic. However, the driving mechanisms of tundra fire occurrences remain poorly understood. This study focuses on identifying key environmental factors controlling fire occurrence in Arctic tundra of Alaska. Our random forest models, considering ignition source, fuel, fire weather, and topography, have shown a strong predictive capability with an overall accuracy above 91%. We found cloud-to-ground (CG) lightning probability by far the dominant driver controlling tundra fire occurrence. Warmer and drier near-surface weather was required to support burning, while fuel composition and topography have modest impacts on fire occurrence. Our results highlight the critical role of CG lightning in driving tundra fires and that incorporating lightning modeling is essential for fire monitoring, forecasting, and management in the Arctic.

Wildland fire smoke exposure affects a broad proportion of the U.S. population and is increasing due to climate change, settlement patterns and fire seclusion. Significant public health questions surrounding its effects remain, including the impact on cardiovascular disease and maternal health. Using atmospheric chemical transport modeling, we examined general air quality with and without wildland fire smoke PM2.5. The 24-hour average concentration of PM2.5 from all sources in 12-km gridded output from all sources in California (2007–2013) was 4.91 μg/m3. The average concentration of fire-PM2.5 in California by year was 1.22 μg/m3 (~25% of total PM2.5). The fire-PM2.5 daily mean was estimated at 4.40 μg/m3 in a high fire year (2008). Based on the model-derived fire-PM2.5 data, 97.4% of California’s population lived in a county that experienced at least one episode of high smoke exposure (“smokewave”) from 2007–2013. Photochemical model predictions of wildfire impacts on daily average PM2.5 carbon (organic and elemental) compared to rural monitors in California compared well for most years but tended to over-estimate wildfire impacts for 2008 (2.0 µg/m3 bias) and 2013 (1.6 µg/m3 bias) while underestimating for 2009 (−2.1 µg/m3 bias). The modeling system isolated wildfire and PM2.5 from other sources at monitored and unmonitored locations, which is important for understanding population exposure in health studies. Further work is needed to refine model predictions of wildland fire impacts on air quality in order to increase confidence in the model for future assessments. Atmospheric modeling can be a useful tool to assess broad geographic scale exposure for epidemiologic studies and to examine scenario-based health impacts.