Kateryna Lapina

Evidence of significant large‐scale impacts of boreal fires on ozone levels in the midlatitude Northern Hemisphere free troposphere

Summertime observations of O 3 and CO made at the PICO-NARE station during 2001, 2003, and 2004 are used to assess the impact of boreal forest fires on the distribution of O 3 mixing ratios in the midlatitude Northern Hemisphere (NH) lower free troposphere (FT). Backward trajectories were used to select measurements impacted by outflow from high-latitude regions. Measurements during these periods were segregated into two subsets: those obtained during periods with and without apparent significant upwind fire emissions. Periods affected by fire emissions were identified based on enhanced CO levels confirmed by global simulations of fire emissions transport. During fireimpacted periods, O 3 was shifted toward higher mixing ratios, with medians significantly higher than in periods without detectable upwind fire impacts. This implies a significant impact of boreal wildfires on midlatitude lower FT background O 3 during summer. Predicted future increases in boreal wildfires may therefore affect summertime O 3 levels over large regions.

Late summer changes in burning conditions in the boreal regions and their implications for NOx and CO emissions from boreal fires

task with significant uncertainties in the methods used. In this work, we assess the impact of seasonal trends in fuel consumption and flaming/smoldering ratios on emissions of species dominated by flaming combustion (e.g., NOx) and species dominated by smoldering combustion (e.g., CO). This is accomplished using measurements of CO and NOy at the free tropospheric Pico Mountain observatory in the central North Atlantic during the active boreal fire seasons of 2004 and 2005. DNOy/DCO enhancement ratios in aged fire plumes had higher values in June-July (7.3 � 10 �3 mol mol �1 ) relative to the values in August-September (2.8 � 10 �3 mol mol � 1 ), indicating that NOx/CO emission ratios declined significantly as the fire season progressed. This is consistent with our understanding that an increased amount of fuel is consumed via smoldering combustion during late summer, as deeper burning of the drying organic soil layer occurs. A major growth in fuel consumption per unit area is also expected, due to deeper burning. Emissions of CO and NOx from North American boreal fires were estimated using the Boreal Wildland Fire Emissions Model, and their long-range transport to the sampling site was modeled using FLEXPART. These simulations were generally consistent with the observations, but the modeled seasonal decline in the DNOy/DCO enhancement ratio was less than observed. Comparisons using alternative fire emission injection height scenarios suggest that plumes with the highest CO levels at the observatory were lofted well above the boundary layer, likely as a result of intense crown fires.