Impact of biomass burning on a metropolitan area in the Amazon during the 2015 El Niño: The enhancement of carbon monoxide and levoglucosan concentrations

Extreme droughts associated with changes in the climate have occurred every 5 years in the Amazon during the 21st century, with the most severe being in 2015. The increase in biomass burning (BB) events that occurred during the 2015 drought had several negative socioeconomic and environmental impacts, one of which was a decrease in the air quality. This study is an investigation into the air quality in the Manaus Metropolitan Region (MMR) (central Amazon, Brazil) during the dry (September to October) and wet (April to May) seasons of 2015 and 2016. A strong El Niño event began during the wet season of 2015 and ended during the wet season of 2016. Particulate matter samples were collected in the MMR during 2015 and 2016, and analyses of the satellite-estimated total carbon monoxide (CO) column and observed levoglucosan concentrations were carried out. Levoglucosan has been shown to be significantly correlated with regional fires and is a well-established chemical tracer for the atmospheric particulates emitted by BB, and CO can be treated as a gaseous-phase tracer for BB. The number of BB events increased significantly during the El Niño period when compared to the average number during 2003-2016. Consequently, the total CO column and levoglucosan concentration values in the MMR increased by 15% and 500%, respectively, when compared to the normal conditions. These results indicate that during the period that was analyzed, the impacts of BB were exacerbated during the strong El Niño event as compared to the non-El Niño period. In this study, we provided evidence that the air quality in the MMR will degrade in the future if droughts and BB occurrences continue to increase.

Photochemical model evaluation of 2013 California wild fire air quality impacts using surface, aircraft, and satellite data

The Rim Fire was one of the largest wildfires in California history, burning over 250,000 acres during August and September 2013 affecting air quality locally and regionally in the western U.S. Routine surface monitors, remotely sensed data, and aircraft based measurements were used to assess how well the Community Multiscale Air Quality (CMAQ) photochemical grid model applied at 4 and 12 km resolution represented regional plume transport and chemical evolution during this extreme wildland fire episode. Impacts were generally similar at both grid resolutions although notable differences were seen in some secondary pollutants (e.g., formaldehyde and peroxyacyl nitrate) near the Rim fire. The modeling system does well at capturing near-fire to regional scale smoke plume transport compared to remotely sensed aerosol optical depth (AOD) and aircraft transect measurements. Plume rise for the Rim fire was well characterized as the modeled plume top was consistent with remotely sensed data and the altitude of aircraft measurements, which were typically made at the top edge of the plume. Aircraft-based lidar suggests O₃ downwind in the Rim fire plume was vertically stratified and tended to be higher at the plume top, while CMAQ estimated a more uniformly mixed column of O₃. Predicted wildfire ozone (O₃) was overestimated both at the plume top and at nearby rural and urban surface monitors. Photolysis rates were well characterized by the model compared with aircraft measurements meaning aerosol attenuation was reasonably estimated and unlikely contributing to O₃ overestimates at the top of the plume. Organic carbon was underestimated close to the Rim fire compared to aircraft data, but was consistent with nearby surface measurements. Periods of elevated surface PM_2.5 at rural monitors near the Rim fire were not usually coincident with elevated O₃.

Contributions of Kansas rangeland burning to ambient O3: Analysis of data from 2001 to 2016

Prescribed range/pasture burning is a common practice in Kansas to enhance the nutritional value of native grasses and control invading weeds, trees, and brush. A major concern associated with the burning is the contribution of smoke to elevated ground level ambient ozone (O₃). The objective of this study is to estimate contributions of Kansas rangeland burning to ambient O₃ mixing ratios through regression analysis (1) between observed O₃ data and available satellite burn activity data from 2001 to 2016; and (2) between observed O₃ data and the smoke contributions to PM_2.5 which were resolved from receptor modeling. Positive correlations were observed between ambient O₃ levels and the acres burned each year estimated from satellite imagery. When burned acres in April were larger than or equal to 1.9 million, O₃>70ppb occurred at least at one of the ten monitoring sites in Kansas. Statistical regression models of daily maximum 8-hour O₃ mixing ratios were developed at each of the ten monitoring sites using meteorological predictors. The O₃ model residuals that were not explained by the meteorological effect models were affected by PM_2.5 contributors including sulfate/industrial sources and emissions that generated secondary organic particles, such as rangeland burning, which were derived from receptor modeling. The average O₃ model residual on the high O₃ days in April was 21±9ppb, which was likely associated with smoke emissions from burning. Research will continue to obtain daily satellite burn activity data and to correlate burn data with daily O₃ data, so that modeling of O₃ levels can be improved under influences of daily burn activities. Less frequency of high O₃ days was observed in April since 2011, which may be partly due to implementation of the Flint Hills Smoke Management Plan which promoted better timing of burns.