Skip to main content
NCBI home page
EPA Author Manuscripts logoLink to EPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Aug 9.
Published in final edited form as: Sci Total Environ. 2017 Nov 4;618:1665–1676. doi: 10.1016/j.scitotenv.2017.10.008

The impact of the 2016 Fort McMurray Horse River Wildfire on ambient air pollution levels in the Athabasca Oil Sands Region, Alberta, Canada

Matthew S Landis a,*, Eric S Edgerton b, Emily M White c, Gregory R Wentworth d, Amy P Sullivan e, Ann M Dillner f
PMCID: PMC6084447  NIHMSID: NIHMS983401  PMID: 29102183

Abstract

An unprecedented wildfire impacted the northern Alberta city of Fort McMurray in May 2016 causing a mandatory city wide evacuation and the loss of 2,400 homes and commercial structures. A two-hectare wildfire was discovered on May 1, grew to ~157,000 ha by May 5, and continued to burn an estimated ~590,000 ha by June 13. A comprehensive air monitoring network operated by the Wood Buffalo Environmental Association (WBEA) in and around Fort McMurray provided essential health-related real-time air quality data to firefighters during the emergency, and provided a rare opportunity to elucidate the impact of gaseous and particulate matter emissions on near-field communities and regional air pollution concentrations. The WBEA network recorded 188 fire-related exceedances of 1-hr and 24-hr Alberta Ambient Air Quality Objectives. Two air monitoring sites within Fort McMurray recorded mean/maximum 1-hr PM2.5 concentrations of 291/5229 μg m−3 (AMS-6) and 293/3259 μg m−3 (AMS-7) during fire impact periods. High correlations (r2 = 0.83–0.97) between biomass combustion related gases (carbon monoxide (CO), non-methane hydrocarbons (NMHC), total hydrocarbons (THC), total reduced sulfur (TRS), ammonia) and PM2.5 were observed at the sites. Filter-based 24-hr integrated PM2.5 samples collected every 6 days showed maximum concentrations of 267 μg m−3 (AMS-6) and 394 μg m−3 (AMS-7). Normalized excess emission ratios relative to CO were 149.87 ± 3.37 μg m−3 ppm−1 (PM2.5), 0.274 ± 0.002 ppm ppm−1 (THC), 0.169 ± 0.001 ppm ppm−1 (NMHC), 0.104 ± 0.001 ppm ppm−1 (CH4), 0.694 ± 0.007 ppb ppm−1 (TRS), 0.519 ± 0.040 ppb ppm−1 (SO2), 0.412 ± 0.045 ppb ppm−1 (NO), 1.968 ± 0.053 ppb ppm−1 (NO2), and 2.337 ± 0.077 ppb ppm−1 (NOX). A subset of PM2.5 filter samples was analyzed for trace elements, major ions, organic carbon, elemental carbon, and carbohydrates. Sample mass reconstruction and fire specific emission profiles are presented and discussed. Potential fire-related photometric ozone instrument positive interferences were observed and were positively correlated with NO and NMHC.

Keywords: Boreal wildfire, Biomass combustion, Normalized excess emission ratios, Particulate matter, Ozone, Total reduced sulfur

GRAPHICAL ABSTRACT

graphic file with name nihms-983401-f0001.jpg

1. Introduction

Over the last thirty-four years (1983–2016) the number of wildfires in the United States has not significantly changed (Appendix A.1a), while the total acreage burned has significantly (Appendix A.1b) increased (United States National Interagency Fire Center, 2017). Nine of the ten highest acreage burn years have occurred since 2000, including the peak of 4.1 million ha in 2015. This time period coincides with many of the warmest years on record in the United States (United States Environmental Protection Agency, 2016). The combination of increased temperature, drought, earlier snowmelt, and historically high fuel loading (e.g., undergrowth and tree density) generally promotes more aggressive fire behavior (Kitzberger et al., 2007; Littell et al., 2009; United States Department of Agriculture, 2014; United States Department of Agriculture, 2016; Westerling, 2016). In addition to increasing size, the intensity of fires estimated as both moderate (Appendix A.2a) or severe (Appendix A.2b) damage have significantly increased from 1984 to 2014 (United States Wildland Fire Leadership Council, 2016; Miller et al., 2009). In Canada over the last fifty-five years (1961–2015) both the number of fires (Appendix A.3a) and the total acreage burned have significantly (Appendix A.3b) increased (Stocks et al., 2003; Canadian Forest Service, 2016). However, the number of fires in Canada has leveled off over the last two decades (Appendix A.3a). In the Province of Alberta, the number of wildfires has steadily increased (Appendix A.4a) throughout the time period and peaked at 1954 fires in 2006 (Alberta Agriculture and Forestry, 2017). The trend in acreage burned in Alberta is generally stable with typical annual totals < 100,000 ha, punctuated by periodic large catastrophic fires (Appendix A.4b) like the 2011 Richardson Backcountry fire that consumed > 577,000 ha in the north-eastern portion of the Province (Liu et al., 2015).

Smoke from wildland fire is an increasing global public health research focus as information on the relative magnitude of atmospheric emissions and the resulting human health effects are elucidated. The combined contribution from wildland and agricultural fires to the emission of U.S. primary air pollutants such as black carbon (BC), fine particulate matter (PM2.5; median aerodynamic diameter < 2.5 μm), carbon monoxide (CO), and volatile organic compounds (VOCs) constitute 43%, 34%, 32%, and 29% of the annual total in the 2011 national emission inventory (NEI), respectively (United States Environmental Protection Agency, 2015). In addition, wildland fire emissions contribute to the formation of photochemically mediated secondary organic aerosol (Donahue et al., 2009; Ortega et al., 2013; Baker et al., 2016). Wildfire smoke has been shown to have detrimental respiratory and cardiovascular health effects, is associated with all-cause mortality and morbidity, and exhibits lung toxicity and mutagenicity ~60 times greater than diesel exhaust (Lee et al., 2014; Liu et al., 2015; Reid et al., 2016; Adetona et al., 2016; Kim et al., 2017). Smoke from wildfires in Alberta, Saskatchewan, and the Northwest Territories has been shown to affect air pollution concentrations in communities in the Athabasca Oil Sands Region (AOSR) in Alberta, Canada (Bytnerowicz et al., 2016; Landis et al., 2017). The Wood Buffalo Environmental Association (WBEA) Bertha Ganter-Fort McKay community air monitoring site (AMS-1) PM2.5 monthly geometric mean concentrations reflect clearly observable wildfire smoke impacts during the summers of 2011, 2012, 2014, and 2015 (Percy, 2013; Appendix A.5a). Similar wildfire smoke impact trends are observable at the AMS-6 and AMS-7 community monitoring sites in the region (Appendix A.5b–c).

In May 2016 an unprecedented wildfire (Horse River Fire) impacted the northern Alberta city of Fort McMurray in the AOSR, causing a mandatory evacuation of all residents and the loss of over 2400 homes and commercial structures. Weather conditions preceding the fire were un-seasonably hot (daily highs up to 35 °C), windy (sustained 10 m winds reaching 43 km hr−1), and the region was experiencing an El Nino influenced drought. An estimated two-hectare wildfire was first discovered on May 1 by a fire patrol and despite an immediate response quickly jumped the Athabasca River, burned through the northern portion of the city, and grew to ~157,000 ha over the next four days. The fire continued to burn through June 13, 2016 with an estimated total size of ~590,000 ha (Fig. 1), resulting in $3.6 billion (CAD) of insured losses. The community air monitoring network operated by WBEA in and around Fort McMurray operated continuously throughout the fire event and recorded 188 exceedances of 1-hr and 24-hr Alberta Ambient Air Quality Objectives (AAAQOs) for PM2.5, hydrogen sulfide (H2S), ozone (O3), nitrogen dioxide (NO2), and CO during the event in May 2016 (Government of Alberta, 2017; Appendix B.1). The progression of the Horse River Fire through the comprehensively instrumented air monitoring network provided a rare opportunity to elucidate the impact of gaseous and particulate matter emissions on near-field and regional communities’ air pollution concentrations in the vicinity of a northern boreal forest mega fire (Appendix A.6a–b). Canada contains 28% of the worlds boreal zone area (552 million ha), and 75% of Canadian forests and woodlands (307 million ha) are located in the boreal zone (Natural Resources Canada, 2017). The two predominant tree species in the AOSR are Jack Pine (Pinus banksiana) and Black Spruce (Picea mariana).

Fig. 1.

Fig. 1.

Open in a new tab

Map depicting the location of the WBEA air monitoring sites measuring continuous PM2.5, the outline of the Horse River Fire burn area, and the outline of oil sand surface mining area.

This study’s goals were to (i) quantify the relative enhancement of ambient air pollution concentrations due to smoke from the Horse River Fire in surrounding AOSR communities, (ii) calculate emission ratios of fire related pollutants of human and ecological concern, (iii) chemically characterize measured PM2.5 and PM10 constituents, and (iv) summarize reconstructed PM2.5 mass.

2. Methods

2.1. WBEA monitoring network

At the time of the fire, WBEA operated a continuous ambient air monitoring network in the AOSR comprised of six community monitoring sites, eleven industrial monitoring sites, one regional background monitoring site, one comprehensive meteorological site, and one mobile monitoring site covering an area ~350 km (N-S) and ~125 km (EW; Fig. 1; Appendix B.2). The AMS-6 Patricia McInnes-Fort McMurray (56°45′5.00″ N; −111°28′35.75″ W), AMS-7 Athabasca Valley-Fort McMurray (56°44′0.18″ N; −111°23′25.75″ W), and AMS-14 Anzac (56°26′56.19″ N; −111°2′16.84″ W) community ambient air monitoring stations were located within the Horse River Fire burn area (Fig. 1). The fire burn area was bounded to the north by the AMS-1 Bertha Ganter-Fort McKay community site (57°11′21.70″ N; −111°38′26.06″ W) and to the south by the AMS-18 Stony Mountain regional background site (55°37′17.19″ N; −111°10′22.16″ W). The community AMS stations provide near real-time ambient air quality information, including the Air Quality Health Index (AQHI), to their respective populations.

WBEA was also operating a network of ecologically analogous jack pine stand level Forest Health Monitoring sites measuring monthly integrated ambient concentrations of sulfur dioxide (SO2), NO2, O3, ammonia (NH3), and nitric acid (HNO3) via passive sampling (Percy, 2013). The purpose of this remote helicopter-only accessible network is to estimate ecosystem exposures to regional air pollutants, at distances up to 150 km from mining and upgrading operations.

2.2. Study design

The continuous measurement study period was defined as April 1–July 31, 2016; which includes approximately a month both before and after the Horse River Fire. The quality assured 5-min and 1-hr integrated monitoring data from all the WBEA ambient monitoring stations (AMS) measuring continuous PM2.5 concentrations (AMS-1, AMS-6, AMS-7, AMS-13, AMS-14, AMS-15, AMS-16, AMS-17, AMS-18, AMS-21) were included in the study (Wood Buffalo Environmental Association (WBEA), 2016). Continuous monitoring data from the AMS-14 site was interrupted from May 6–14 due to power being turned off to the site as a fire precaution, and continuous PM2.5 measurements were interrupted at AMS-21 from May 5–11 as a result of a pump failure following a short power interruption. The filter based measurement study period was defined as April 1–August 4, 2017 to allow for an additional sampling day to be included to characterize non-fire conditions.

2.3. WBEA AMS continuous measurements

Routine WBEA ambient monitoring data includes continuous (i) PM2.5 mass measured using ThermoScientific (Franklin, MA) Model 5030 Synchronized Hybrid Ambient Real-time Particulate Monitors (SHARP), (ii) nitrogen oxide (NO), NO2, and total oxides of nitrogen (NOx) measured using ThermoScientific Model 42i and Teledyne API (San Diego, CA) Model T200 chemiluminescence analyzers, (iii) SO2 measured using ThermoScientific Model 43i and Teledyne API Model T100 pulsed fluorescence analyzers, (iv) NH3 measured using ThermoScientific Model 17i and Teledyne API Model T201 chemiluminescence analyzers, (v) CO measured using a ThermoScientific Model 48i gas filter correlation analyzer, (vi) O3 measured using ThermoScientific Model 49i and API Model T400 UV absorption analyzers, (vii) H2S measured using ThermoScientific Model 450i and API Model T101 UV fluorescence analyzers, (viii) Total Hydrocarbons (THC) measured using ThermoScientific Model 51i flame ionization detector (FID) instruments, (ix) Non-methane Hydrocarbons (NMHC) and methane (CH4) measured using ThermoScientific Model 55i gas chromatography FID instruments, and (x) total reduced sulfur (TRS) measured using ThermoScientific Model 43i coupled with a CD Nova Ltd. (Surrey, BC) Model CDN101 high temperature thermal oxidizer (Wood Buffalo Environmental Association (WBEA), 2016). All continuous gas analyzers in the WBEA monitoring network are automatically single point span calibrated and zeroed on a daily basis using Teledyne API Model T700 dynamic dilution calibration systems with U.S. EPA protocol certified gas standards. Multi-point span calibrations are conducted on a monthly basis. The SHARP PM2.5 instruments beta gauges are calibrated with certified mass foil sets every six months and the instruments nephelometers are zeroed on a monthly basis.

2.4. WBEA AMS filter based particulate matter measurements

Routine 24-hr integrated filter based measurements of PM2.5 (sites AMS-1, AMS-6, AMS-7, and AMS-14) and PM10 (sites AMS-1, AMS-6, AMS-7, AMS-13, AMS-14, AMS-15, and AMS-16) are carried out using collocated ThermoScientific Model 2000i Partisol U.S. EPA designated Federal Reference Method (FRM) samplers. Measurement Technologies Laboratories (Minneapolis, MN) 47 mm Teflon membrane filters with Teflon support rings samples are robotically weighed for PM2.5 and PM10 mass by Atmospheric Research & Analysis, Inc. (ARA, Morrisville, NC) using a method described by Landis et al. (2017). One of each collocated filter pair is extracted and analyzed for major ions, and the other collocated filter is extracted and analyzed for trace elements as described below. Bracketing the Horse River Fire, a subset of the Teflon filter samples was selected for additional carbohydrate and organic carbon (OC)/elemental carbon (EC) analysis. At the AMS-1 site, an additional 24-hr integrated PM2.5 sample is routinely collected onto a quartz filter for determination of OC/EC. Filter samples are collected according to the Canadian National Air Pollution Surveillance Program (NAPS) 1-in-6-day schedule.

2.5. WBEA forest health monitoring site measurements

WBEA collects tower-based annular denuder samples at four forest health sites (JP104, JP107, JP201 and JP213) to obtain above-canopy measurements of HNO3 and NH3 across the AOSR. JP104 is located approximately 45 km N of Fort McMurray and 25 km SSE of Fort McKay, and JP201, JP107 and JP213 are located 100–150 km W, N and E of JP104, respectively (Fig. 1). Samples are collected at a height of ~25 m above ground level (roughly 5 m above canopy; Appendix A.7a) using solar-power for flow maintenance and data logging, and two URG Corporation (Carrboro, NC) 150 mm multi-channel annular denuders (URG-2000–30×150–3CSS) in series coated with potassium chloride (KCl) and citric acid for capture of HNO3 and NH3, respectively (United States Environmental Protection Agency, 1999a, 1999b). Samples are collected over nominal monthly intervals at a flow rate of 1.25 L min−1. Following exposure, denuders are sealed and returned to the ARA laboratory for analysis. Field blanks are collected with each monthly deployment and used to blank-correct each batch of samples. One of the samplers at JP104 is configured for the collection of PM2.5.This sampler has an in-line PM2.5 cut point cyclone downstream of the KCl and citric acid denuders followed by a two-stage filter pack containing a Teflon pre-filter and a nylon backup filter (Appendix A.7b). The Teflon filter captures non-volatile PM2.5 while the nylon filter captures volatile nitrate lost from the Teflon filter during sampling. After exposure, filter packs are sealed and returned to the ARA laboratory for extraction and analysis. Teflon filters are weighed and then cut in half, with one half used for analysis of major ions and the other half used for trace element analysis.

2.6. Carbon analysis

Quartz filter samples collected at the AMS-1 Fort McKay site were analyzed for OC and EC by Desert Research Institute (Reno, NV) using both the thermal optical transmission (TOT) NIOSH method 5040 (United Stated National Institute for Occupational Safety and Health, Manual of Analytical Methods (NMAM), et al., 2003) and thermal optical reflectance (TOR) IMPROVE_A protocol (Chow et al., 1993; Chow et al., 2007). Routine semi-continuous (5-min) BC measurements were made using Magee Scientific (Berkeley, CA) model AE-33 spectrum Aethalometers (Drinovec et al., 2015) at the AMS-1 and AMS-18 sites. Organic components of biomass combustion emissions have been observed to result in enhanced UV (370 nm) relative to IR (880 nm) wavelength optical absorption, the enhanced absorption is expressed as Delta-C (DC) in units of μg m−3, and calculated as BC370–BC880 (Allen et al., 2004).

2.7. Fourier transform infrared spectrometer analysis for OC and EC

A subset of the routinely collected AMS PM2.5 FRM samples was also analyzed using a Bruker Optics (Billerica, MA) Model Tensor II Fourier transform infrared (FTIR) spectrometer equipped with a liquid nitrogen-cooled wide-band mercury cadmium telluride detector. TOR equivalent OC and EC measurements (Chow et al., 2007) are obtained from the FTIR spectra via a partial least squares regression calibration. Complete method details can be found elsewhere (Weakley et al., 2016).

2.8. Filter extraction and trace element analysis

Following filter post weighing, the primary set of the PM2.5 and PM10 sample filters were microwave digested in an acid-peroxide solution and analyzed for 43 elements using a Perkin-Elmer (Waltham, MA) Model 9000 Elan-II dynamic reaction cell inductively coupled plasma mass spectrometer (DRC-ICPMS) by ARA using a method described by Landis et al. (2017). Instrument drift and suppression, or enhancement of instrument response caused by the sample matrix, was corrected by internal standardization (Edgerton et al., 2012). Target isotopes and method detection limits (MDLs) are presented in Landis et al. (2017).

2.9. Majorions

Following filter post weighing, the collocated set of PM2.5 and PM10 sample filters were extracted with 30 mL of ASTM Type I (18.2 MΩ·cm) ultra-pure water and analyzed for major ions (CL‒,F‒,NO3−,SO42−,PO43−,NH4+,Ca2+,Mg2+,k+,Na+) using ion chromatography (IC) (United States Environmental Protection Agency, 1997). Forest health tower annular denuders were extracted with 10 mL of ASTM Type I ultra-pure water and analyzed for major ions (NO3−,SO42−,NH4+) using IC. A Dionex (Sunnyvale, CA) Model ICS-3000 dual channel instrument equipped with a Dionex Model AS40 autosampler was utilized. Ion separations were performed with a 100 μL sample loop injection onto a Dionex IonPac© AS18 analytical column (anions) and a 25 μL sample loop injection onto a Dionex IonPac© CS16 analytical column (cations). Each channel operates a Dionex self-regenerating SRS-ULTRA suppressor in external regeneration mode, a CD20 conductivity detector, and a GP50 gradient pump with built-in degassing. The system uses Dionex eluent cartridges and laboratory-produced ASTM Type I ultra-pure water for automated generation of eluents: potassium hydroxide for anions and methanesulfonic acid for cations. The IC system was calibrated with NIST-traceable mixed standard solutions with concentrations of each analyte in the range of 0.05–5.0 mg L−1. Following calibration of the IC, independent quality control solutions were analyzed to confirm accuracy. Multiple secondary source standards are also used to routinely assess precision and accuracy.

2.10. Carbohydrates

A subset of the AMS PM2.5 and PM10 FRM sample aqueous extracts during the study period was also analyzed for levoglucosan, mannosan, galactosan, and other carbohydrate smoke markers. The carbohydrate smoke markers were determined using high-performance anion-exchange chromatography-pulsed amperometric detection (HPAECPAD) using a Dionex DX-500 series ion chromatograph with a Dionex CarboPac PA-1 column (4 × 250 mm) employing a sodium hydroxide gradient. Full details of the method can be found in Sullivan et al. (2011a, 2011b, 2014).

2.11. Statistical analysis

Data processing and all statistical analyses were performed using SAS v.9.4 (SAS Institute, Cary, NC). Parametric statistics used in this analysis included a t-test for independent samples, and one-way analysis of variance (ANOVA). The assumptions of the parametric procedures were examined using residual plots, skewness and kurtosis coefficients, Shapiro-Wilk test, and the Brown-Forsythe test. If assumptions of the parametric procedures were violated, then the Wilcoxon and Kruskal-Wallis non-parametric procedures were used. One-sided tests and a level of significance of α = 0.05 were used for all statistical procedures unless otherwise stated. The SAS REG and ORTHOREG procedures were used for least square general linear model regressions.

3. Results and discussion

3.1. Impact of fire emissions on hourly average air pollution concentrations

In the past, researchers have used a fixed PM2.5 concentration cut off of 25 μg m−3 in the AOSR (Bytnerowicz et al., 2016) and in Finland (Saarnio et al., 2010) to categorize fire impacted sampling periods due to smoke from boreal forest wildfires. In this case, the proximity of the Horse River Fire to the WBEA ambient monitoring network sites, the high time resolution air quality measurements, and the magnitude of impact provided the opportunity to definitively delineate the smoke impact events. Hourly integrated WBEA continuous data were categorized as fire impacted during the study period if PM2.5 concentrations were elevated relative to pre-fire concentrations and were coincident with multiple elevated indicators of biomass combustion such as CO, NMHC, THC, TRS, H2S, and NH3. Fig. 2 presents a time series of PM2.5, CO, NMHC, and TRS at the AMS-7 Athabasca Valley community monitoring site during the study period highlighting the Horse River Fire impact periods. Similar time series plots are presented for the other AMS sites in Appendix A.8a–i. The AMS-7 community site was located within the fire burn area and was the only AMS site monitoring CO, which is considered a good near-field indicator of wildland fire emissions (Urbanski, 2013). Ten exceedances of the hourly AAAQO for CO (13 ppm) were recorded at the AMS-7 site during the Horse River Fire (Appendix B.1). During the study period, CO concentrations at AMS-7 had significant (p < 0.0001) positive correlations with PM2.5 (0.91), THC (0.99), NMHC (0.99), CH4 (0.97), TRS (0.97), and NO2 (0.77) demonstrating the efficacy of their use in conjunction with PM2.5 as indicators of biomass combustion impact at other sites not measuring CO.

Fig. 2.

Fig. 2.

Open in a new tab

Times series of AMS-7 Athabasca Valley community monitoring site PM2.5, CO, TRS, and NMHC concentrations during the study period (grey highlighting indicates periods categorized as Horse River Fire impacted).

A statistical summary of the AMS site fire impacted and non-impacted hourly integrated continuous data are presented in Table 1. There is no regulatory AAAQO for 1-hr PM2.5, but the non-regulatory Alberta Ambient Air Quality Guideline is 80 μg m−3 (Government of Alberta, 2017). The mean ± standard deviation of hourly ambient PM2.5 concentrations at the AMS sites during the study period in the absence of fire impact ranged from 4.3 ± 4.1 μg m−3 at the AMS-21 Conklin community site to 7.1 ± 6.1 μg m−3 at the AMS-16 industrial monitoring site. During fire impacted periods, the mean hourly ambient PM2.5 concentrations at the AMS sites ranged from 80 ± 67 μg m−3 at the AMS-18 regional background site to 293 ± 507 μg m−3 at the AMS-7 community monitoring site, representing an average increase of a factor of 18 and 47 at the sites, respectively. The PM2.5 concentration enhancements at the WBEA community monitoring sites during fire impact periods ranged from a factor of 19 at AMS-21 (Conklin) to 54 at AMS-6 (Fort McMurray) with associated mean hourly concentrations of 82 ± 62 and 291 ± 673 μg m−3, respectively. WBEA reported 653 hourly PM2.5 concentrations in the community sites during the Horse River Fire impact period above the 1-hr PM2.5 Alberta Ambient Air Quality Guideline of 80 μg m−3.

Table 1.

Summary of mean ± standard deviation hourly ambient air pollution concentrations (April 1–July 31, 2016) categorized by fire impact.

AMS-1 AMS-6 AMS-7 AMS-13 AMS-14 AMS-15 AMS-16 AMS-17 AMS-18 AMS-21
No Fire Fire No Fire Fire No Fire Fire No Fire Fire No Fire Fire No Fire Fire No Fire Fire No Fire Fire No Fire Fire No Fire Fire
Number (n) 2737 191 25 413 2279 399 2772 156 2476 252 2793 135 2700 228 2746 182 2759 169 2570 164
PM2.5 (μg M‒3) 6.6 ± 6.0 233 ± 585 5.4 ± 5.5 291 ± 673 6.3 ± 5.0 293 ± 507 5.1 ± 4.2 282 ± 711 4.9 ± 4.4 149 ± 403 6.6 ± 6.7 146 ± 307 7.1 ± 6.1 259 ± 826 4.6 ± 4.4 196 ± 219 4.4 ± 3.9 80 ± 67 4.3 ± 4.1 82 ± 62
CH4 (ppm) 2.0 ± 0.1 2.1 ± 0.2 2.0 ± 0.1 2.2 ± 0.4 1.9 ± 0.1 2.1 ± 0.4 2.0 ± 0.1 2.2 ± 0.9 1.9 ± 0.1 1.9 ± 0.1 1.9 ± 0.1 1.9 ± 0.1
NMHC (ppm) 0.0 ± 0.1 0.2 ± 0.3 0.0 ± 0.0 0.3 ± 0.5 0.0 ± 0.0 0.4 ± 0.7 0.0 ± 0.0 0.2 ± 1.2 0.0 ± 0.0 0.1 ± 0.1 0.0 ± 0.0 0.0 ± 0.0
NO (ppb) 0.7 ± 1.9 0.4 ± 0.8 0.7 ± 1.4 0.8 ± 1.1 1.0 ± 2.4 2.2 ± 4.2 0.7 ± 1.9 0.5 ± 1.1 0.3 ± 1.5 2.3 ± 15.9 1.0 ± 3.8 0.5 ± 1.5 4.9 ± 9.7 3.8 ± 9.9 0.4 ± 0.6 0.1 ± 0.6 0.1 ± 0.1 0.1 ± 0.2 0.2 ± 0.9 0.1 ± 0.2
NO2 (ppb) 3.4 ± 4.2 5.4 ± 7.0 2.7 ± 2.9 5.8 ± 7.9 3.5 ± 2.9 7.7 ± 8.3 3.0 ± 3.8 4.9 ± 6.5 1.1 ± 1.1 4.7 ± 19.2 3.1 ± 4.1 6.1 ± 7.6 6.2 ± 6.5 7.0 ± 8.1 1.1 ± 1.7 5.1 ± 4.6 0.8 ± 0.5 2.7 ± 1.6 1.0 ± 0.9 2.3 ± 1.4
NOX (ppb) 4.0 ± 5.4 5.7 ± 7.6 3.4 ± 3.8 6.5 ± 8.3 4.4 ± 4.1 9.8 ± 10.6 3.7 ± 5.1 5.3 ± 7.4 1.5 ± 2.1 6.5 ± 35.1 4.1 ± 7.0 6.6 ± 8.6 11.2 ± 15.2 10.9 ± 15.5 1.3 ± 2.0 4.4 ± 4.5 0.8 ± 0.6 2.8 ± 1.6 1.2 ± 1.5 2.4 ± 1.5
O3 (ppb) 27 ± 13 34 ± 16 33 ± 11 39 ± 26 28 ± 11 41 ± 41 25 ± 14 31 ± 16 31 ± 12 38 ± 101 31 ± 12 39 ± 12 38 ± 10 43 ± 11 31 ± 14 32 ± 15
SO2 (ppb) 0.9 ± 3.6 1.8 ± 4.2 1.0 ± 3.0 1.0 ± 1.7 0.9 ± 2.1 1.7 ± 4.0 0.9 ± 3.6 1.7 ± 4.2 0.5 ± 0.5 1.8 ± 5.2 0.5 ± 1.9 1.1 ± 1.3 0.8 ± 2.4 1.5 ± 3.3 0.5 ± 1.3 1.4 ± 1.6 0.3 ± 0.3 1.8 ± 1.6 0.3 ± 0.3 0.3 ± 0.2
THC (ppm) 2.0 ± 0.2 2.3 ± 0.6 2.0 ± 0.1 2.4 ± 0.9 1.9 ± 0.1 2.6 ± 1.1 2.3 ± 0.4 2.8 ± 1.2 2.0 ± 0.1 2.4 ± 2.1 2.2 ± 0.3 2.4 ± 0.5 2.4 ± 0.4 2.7 ± 0.5 2.1 ± 0.1 2.5 ± 0.4 2.0 ± 0.1 2.0 ± 0.1 1.9 ± 0.1 1.9 ± 0.1
TRS (ppb) 0.5 ± 0.2 1.2 ± 1.3 0.2 ± 0.1 1.3 ± 2.4 0.3 ± 0.2 2.2 ± 2.6 0.2 ± 0.3 0.9 ± 1.5 0.2 ± 0.1 1.4 ± 3.3 0.3 ± 0.7 0.7 ± 1.0 0.3 ± 0.0 0.5 ± 0.2 0.4 ± 0.2 0.4 ± 0.2
BC (μg m−3) 0.3 ± 0.4 3.3 ± 6.4 0.1 ± 0.2 2.5 ± 2.2
Delta-C (μg m-3) 0.1 ± 0.2 20 ± 162 0.1 ± 0.3 8.0 ± 8.2
NH3 (ppb) 1.4 ± 1.2 25 ± 39 3.3 ± 4.1 136 ± 221
CO (ppm) 0.1 ± 0.0 2.8 ± 3.7
H2S (ppb) 0.1 ± 0.1 0.5 ± 0.5
Open in a new tab

Significant enhancements of biomass combustion related pollutants NMHC, NH3, BC, TRS, NOx, and H2S were also observed. While NMHC, BC, NOx, and reduced nitrogen compounds such as NH3 have routinely been measured from wildland fires (Nance et al., 1993; Worden et al., 1997; Dennis et al., 2002), this is the first observation of measurable reduced sulfur compound (TRS/H2S) emissions. We observed significant correlations between hourly background-corrected NH3 (ΔNH3) and background-corrected TRS (ΔTRS) at AMS-6 (r2 = 0.948; Appendix A.9a) and AMS-1 (r2 = 0.868; Appendix A.9b) during smoke impacted periods.

A statistical summary of the fire impacted ambient data from AMS-7 is presented in Table 2, and similar summary tables for the other sites are presented in Appendix B.3a–i. The magnitude of the impact related to the Horse River Fire emissions are highlighted at many of the AMS sites where the mean observed concentrations of fire related pollutants (e.g., PM2.5, TRS, NH3, NO, NO2, NOX, CH4, NMHC, SO2) exceed their 75th percentile value concentrations. The maximum hourly fire impacted PM2.5 concentrations at the community sites ranged from 357 μg m−3 (AMS-21) to 5229 μg m−3 (AMS-6), representing extremely high and unhealthy acute exposures to firefighting personnel working in these communities and citizens not heeding evacuation orders. The ambient PM2.5 concentrations measured at the AMS community sites during fire impact periods are on the same order as PM2.5 concentrations that have been reported for wildland firefighter occupational exposures (Naeher et al., 2006; Miranda et al., 2010; Adetona et al., 2011), and in some cases higher by > 2 orders of magnitude above downwind community monitoring sites where observable public health effects were documented (Lee et al., 2014; Liu et al., 2015; Reid et al., 2016; Adetona et al., 2016; Kim et al., 2017). Ambient PM2.5 measurements made as part of past routine community monitoring programs and emergency response operations in many cases have not reported concentrations above 800–1000 μg m−3 due to inherent limitations of the continuous monitoring instruments (Bytnerowicz et al., 2016; Schweizer et al., 2016) and filter loading related flow restrictions on integrated samplers (Landis et al., 2017). Unfortunately, carbon dioxide (CO2) measurements were not part of the WBEA AMS routine monitoring plan in 2016, so modified combustion efficiencies (Urbanski, 2013) could not be calculated; and the measured ambient concentrations could not be categorized in terms of fire combustion regime (e.g., flaming, smoldering).

Table 2.

Statistical summary of AMS-7 site hourly ambient air pollution concentrations (April 1–July 31, 2016) during Horse River Fire impact periods.

Species Unit Mean 25th percentile Median 75th percentile Maximum
CO ppm 2.8 0.4 0.8 4.0 17.1
PM2.5 µg m−3 293 33 91 308 3259
THC ppm 2.6 1.9 2.0 3.0 6.6
CH4 ppm 2.1 1.9 1.9 2.3 3.7
NMHC ppm 0.4 0.0 0.0 0.7 3.0
NO ppb 2.2 0.3 1.0 2.2 35.2
NO2 ppb 7.7 2.5 4.9 10.0 54.3
NOx ppb 9.8 3.1 5.7 12.5 61.0
TRS ppb 2.2 0.5 0.8 3.3 12.6
SO2 ppb 1.7 0.2 0.7 1.4 36.1
O3 ppb 41 20 30 45 358
Open in a new tab

The reported elevated “O3” concentrations reached a maximum hourly concentration of 1586 ppb at AMS-14, and appeared to be a smoke (volatile organic compound) induced measurement artifact from the WBEA UV photometric instruments, as have been previously reported (Kleindienst et al., 1993; United States Environmental Protection Agency, 1999a, 1999b; Spicer et al., 2010). Elevated “O3” concentrations were observed at night (22:00–05:00) in the absence of photochemistry and were positively correlated with NO and NMHC. The significant relationships between nighttime “O3”, NO, and NMHC at sites AMS-14 (May 5) and AMS-6 (May 4) are presented in Appendices A.10a–b, respectively. Nighttime 5-min “O3” concentrations reached 7077 ppb at AMS-14 and 361 ppb at AMS-6. These observations raise concerns that regulatory O3 measurements utilizing UV photometric Federal Equivalent Method (FEM) instruments may be reporting false positive results, and that the reported hypothesis that rapid photo-chemical processing was responsible for observed elevated O3 concentrations reported in downwind wildfire plumes may be influenced analytical artifacts (e.g., Liu et al., 2017). EPA is currently evaluating Federal Reference Method (FRM) and photometric FEM instruments for positive O3 artifact response in wildland fire smoke.

3.2. Emissionratios

Emissions of air pollutants from wildfires are commonly expressed as normalized emission ratios (ER) where the excess in each analyte X in a smoke plume above its background value (ΔX = Xplume – Xbackground) is divided by a simultaneously measured excess mixing ratio (ΔY = Yplume – Ybackground) of a relatively inert tracer gas like CO (Yokelson et al., 1999; Yokelson et al., 2009; Akagi et al., 2011; Hecobian et al., 2011; Hornbrook et al., 2011; Simpson et al., 2011; O’Shea et al., 2013; Urbanski, 2013; Collier et al., 2016; Liu et al., 2017). In this case, the ERX/CO were calculated for all air pollutants measured at the AMS-7 monitoring site through the linear regression of hourly averaged background corrected ambient concentrations where the regression intercept was forced through zero (Yokelson et al., 1999). The background normalization for each hourly integrated data point categorized as smoke impacted was calculated by subtracting the median site concentration during periods categorized as non-smoke impacted during the fire event period (Table 3). The AMS-7 background for CO was 93 ± 2 ppb (median ± variance) which compared quite well to the values of 105 ± 9 ppb reported by O’Shea et al. (2013) for Eastern Canada and 97 ± 6 ppb reported by Cofer et al. (1998) for the Canadian Northwest Territories.

Table 3.

Statistical summary of AMS-7 normalized excess mixing ratios (NEMRs) of air pollutants during Horse River Fire impacted hourly time periods calculated using normal linear regression (all p < 0.0001).

Compound Backgrounda NEMRb r2
CO 0.093 ± 0.002 ppm n/a n/a
PM2.5 5.08 ± 14.88 μg m−3 149.87 ± 3.37 μg m−3 ppm−1 0.8402
THC 1.855 ± 0.007 ppm 0.274 ± 0.002 ppm ppm−1 0.9782
NMHC 0.169 ± 0.001 ppm 0.169 ± 0.001 ppm ppm−1 0.9777
CH4 1.855 ± 0.003 ppm 0.104 ± 0.001 ppm ppm−1 0.9587
TRS 0.694 ± 0.007 ppb 0.694 ± 0.007 ppb ppm−1 0.9590
SO2 0.167 ± 1.761 ppb 0.519 ± 0.040 ppb ppm−1 0.3062
NO 0.641 ± 14.356 ppb 0.412 ± 0.045 ppb ppm−1 0.1774
NO2 1.750 ±7.514 ppb 1.968 ± 0.053 ppb ppm−1 0.7764
NOX 2.530 ± 25.965 ppb 2.337 ± 0.077 ppb ppm−1 0.7012
Open in a new tab
a

Background = Median ± variance of non-fire impacted ambient hourly sample concentrations during fire period (number of observations ranged from 723 to 726). Median value used to normalize each hourly observation to calculate ΔX.

b

NEMR = slope of linear regression ΔX and ΔCO when intercept forced through zero ± standard error.

The ERΔX/ΔCO regressions are presented in Fig. 3a–i and summarized in Table 3 for PM2.5 (149.87 ± 3.37 μgm−3 ppm−1 ; slope ± standard error), THC (0.274 ± 0.002 ppm ppm−1), NMHC (0.169 ± 0.001 ppm ppm−1), CH4 (0.104 ± 0.001 ppm ppm−1), TRS (0.694 ± 0.007 ppb ppm−1), SO2 (0.519 ± 0.040 ppb ppm−1), NO (0.412 ± 0.045 ppb ppm−1), NO2 (1.968 ± 0.053 ppb ppm−1), and NOX (2.337 ± 0.077 ppb ppm−1). While the ERΔPM2.5/ΔCO could include secondary PM2.5 formed after emission, given the proximity of the wildfire it was likely that primary PM2.5 was dominant. The ER regressions for THC, NMHC, CH4, and TRS were extremely well fit over a wide range of smoke impacts representing ~400 h of measurements resulting in relatively low standard errors. The ER regressions for PM2.5, SO2, NO, NO2, and NOX demonstrated more variance that may be related to changing burn conditions (e.g., flaming, smoldering) during the fire impact periods (Urbanski, 2013) resulting in larger standard errors. The five AMS-7 PM2.5 observations that were N2500 μg m−3 and fall outside the regression 95% confidence internal (Fig. 3a) were all observed on May 7, 2016 from 06:00–10:00, however no other associated observations during this time period were similarly documented as regression outliers.

Fig. 3.

Fig. 3.

Open in a new tab

Scatter plots of background-corrected analytes versus ΔCO for (a) ΔPM2.5, (b) ΔTHC, (c) ΔNMHC, (d) ΔCH4, (e) ΔTRS, (f) ΔSO2, (g) ΔNO, (h) ΔNO2, and (i) ΔNOX for Hourly AMS-7 Horse River Fire impacted samples.

The ERs observed during the Horse River Fire generally compared well to other boreal forest fire results reported in the literature as summarized in Table 4. The Horse River Fire ERs for CH4 (0.104 ± 0.001 ppm ppm−1), SO (0.519 ± 0.040 ppb ppm−1), and NOX (2.337 ± 0.077 ppb pm−1) agree within reported uncertainties with “fresh biomass burning” plumes (n = 18) measured from aircraft at altitudes 2.0 ± 1.3 km reported by Singh et al. (2010). The Horse River Fire ER for CH4 agrees within reported uncertainties, the ER for NO2 (1.968 ± 0.053 ppb ppm−1) is approximately a factor of three lower, and the ER for NO (0.412 ± 0.045 ppb ppm−1) is approximately a factor of six lower than whole air sampling of Canadian boreal forest fire plumes (n = 7) measured from ground and aircraft platforms as reported by Simpson et al. (2011). The Horse River Fire ER for NOX agrees within reported uncertainties, and the ER for CH4 is approximately a factor of four lower than whole air samples of forest fire plumes measured from seventeen aircraft flights reported by Hecobian et al. (2011) from Canadian boreal and California fires. The Horse River Fire ERs for CH4 is approximately 30% higher than the 0.079 ± 0.001 ppm ppm−1 reported by O’Shea et al. (2013) measured from fifteen aircraft flights of eastern Canadian forest fires.

Table 4.

Comparison of this study’s NEMRs to other literature values for North American boreal forest wildfires.

Compound Unit This study O’Sheaetal. (2013) Hecobianet al. (2011) Simpson et al. (2011) Singh et al. (2010)
PM2.5 µg m−3ppm−1 149.87 ±3.37
THC ppmppm−1 0.274 ± 0.002
NMHC ppmppm−1 0.169 ± 0.001
CH4 ppmppm−1 0.104 ±0.001 0.079 ± 0.001 0.478 ± 0.199 0.072 ± 0.044 0.08 ± 0.03
TRS ppbppm−1 0.694 ± 0.007
SO2 ppbppm−1 0.519 ± 0.040 1.1 ± 0.6
NO ppbppm−1 0.412 ± 0.045 2.4 ± 0.1
NO2 ppbppm−1 1.968 ±0.053 5.6 ± 0.3
NOX ppbppm−1 2.337 ± 0.077 1.721 ±2.196 3.5 ± 4.2
Open in a new tab

3.3. Impact of fire emissions on forest health site monthly integrated concentrations

HNO3 and NH3 concentrations for calendar year 2016 (CY2016) are presented in Fig. 4a–b. Data are segregated into three time periods corresponding to non-fire impacted samples (n = 10–11 per site), fire impacted samples (1 per site), and the annual average. Due to variable sample durations, non-fire impacted and annual concentrations are calculated as time-weighted averages. For HNO3 (Fig. 4a) CY2016 concentrations are highest at JP104 and JP107 followed by JP213 and then JP201, and there is virtually no difference between fire impacted samples and non-fire impacted samples. Inspection of PM2.5 nitrate data from site JP104 shows that concentrations during the fire period were roughly 4 times higher than the non-fire period, and roughly 3 times higher than HNO3 during the fire period. This suggests that the majority of HNO3 produced during, or subsequent to, the fire partitioned into the particulate phase, due to reaction with abundant NH3 or relatively alkaline pyrogenic PM. For NH3, the order of CY2016 concentrations is somewhat different, with JP104 > JP213 > JP107 > JP201. More importantly, there are striking NH3 concentration differences between fire and non-fire impacted samples at all sites (Fig. 4b). For JP104 and JP213, NH3 concentrations in fire samples are roughly 14 times higher than non-fire samples. At JP107 and JP201, fire samples are roughly 7 and 2 times higher than non-fire samples. Expressed another way, the sample collected during the Horse River Fire event contributed about 65% of the annual time-weighted average NH3 at JP104 and JP107, 80% at JP213, and 25% at JP201.

Fig. 4.

Fig. 4.

Open in a new tab

Forest health monitoring monthly sampling results for (a) HNO3, (b) NH3, and (c) PM2.5 and trace elements quantifying the Horse River Fire impact.

These observations show that short-term emissions from wildland fires can have a marked impact on annual concentrations of NH3 as well as a potentially significant impact on nutrient loadings to terrestrial and aquatic ecosystems in the AOSR. This finding is consistent with Benedict et al. (2017) who observed hourly NH3 enhanced by a factor of ~20 in wildfire smoke-impacted periods in Fort Collins, Colorado. However, the same study found a more moderate wildfire enhancement (~2–3 times) for daily integrated denuder NH3 as well as HNO3 measurements. The difference in magnitude between the NH3 enhancement observed in the AOSR as a result of the Horse River Fire and the results reported by Benedict et al. (2017) are likely the result of differences in NH3 emission as observed by the ΔNH3/ΔNOX at AMS-6 (21.0; Appendix A.11a), AMS-1 (3.6; Appendix A.11b), and ~2.5–3.0 reported by Benedict et al. (2017) (Fig. 4b). It is possible the Benedict et al. (2017) observed increase in HNO3 was due to the larger distance from the flame front, allowing time for the oxidation of NOx into HNO3, or less partitioning to the particulate phase than was observed in this study.

The PM2.5 mass and trace element results (JP104 only) show two interesting patterns that include the majority of measured analytes (Fig. 3c). For refractory elements (e.g., Si, Mo, Ni, V, Ce, Th, U), we observed significantly lower concentrations during the Horse River Fire event than the rest of the year. This likely reflects the temporary shutdown of oil sands mining and bitumen upgrading operations during and immediately after the fire, with mining having been previously shown to be a dominant source of fugitive dust in the region (Landis et al., 2012; Landis et al., 2017). In addition, upgrader emissions and “mixed sources” have been identified as major contributors to Mo, V, and Ni at AMS-5, AMS-11, and AMS-13, whereas K and Cd were found to be tracers for biomass burning (Phillips-Smith et al., 2017), consistent with Fig. 4c. The second group includes volatile elements and oxides that are markedly enhanced during the fire. These include Rb, Tl, and K (3–4 times higher), Zn, As and Sn (6–10 times higher) and Sb, Cd, Pb (15–27 times higher). The wildland fire impact on Sb, Cd and Pb more than doubled their respective annual time-weighted concentrations, which suggests that inter-annual variability of these elements (like NH3) may be driven by local and regional wildland fires.

3.4. AMS site filter-based PM2.5 mass reconstruction

Table 4 summarizes the PM2.5 mass reconstruction from each of the AMS sample filters that were analyzed for carbohydrates, OC, and EC (including both fire impacted and non-fire impacted days). Carbon was integrated into the PM2.5 composition analysis by using the TOR OC and EC values from the AMS-1 quartz filter samples, and the FTIR OC and EC measurements from the AMS-6/AMS-7 Teflon filter samples. OC results were multiplied by 1.76 to obtain total organic matter (OM) which accounts for associated oxygen and hydrogen mass (Aiken et al., 2008). OM was the major constituent on both fire impacted and non-impacted sample days. On August 4, 2016, a sampling day with no fire impact, PM2.5 concentrations at the AMS-1, AMS-6, and AMS-7 community sites was 8.0 ± 0.8 μg m−3 (mean ± standard deviation) and OM represented ~60% of the total mass with ~1.5% from the sum of all carbohydrates (ΣSugar; Appendix A.12a). On fire impacted days at sites AMS-6 and AMS-7, the mean PM2.5 concentration was 125 μg m−3 and OM represented ~90% of mass with ~25% from ΣSugar (Table 5). Levoglucosan is an anhydrosugar produced from the combustion of cellulose and a well-established tracer for biomass combustion (Sullivan et al., 2011a; Sullivan et al., 2011b; Sullivan et al., 2014). During the Horse River Fire levoglucosan was strongly correlated with pooled PM10 and PM2.5 mass (r2 = 0.954; Appendix A.13a), and other anhydrosugars mannosan (r2 = 0.993; Appendix A.13b) and galactosan (r2 = 0.980; Appendix A.13c) across the WBEA network.

Table 5.

Reconstructed PM2.5 mass for filter samples with enhanced analytical results.

Component (μg m‒3) AMS-1 AMS-6 AMS-7
Fire impacted No fire Fire impacted No fire Fire impacted No fire
May 18 May 24 August 4 May 18 August 4 May 12 August 4
PM2.5 Mass 25.8 42.2 7.1 139.0 8.4 110.5 8.4
ΣSugar OC 1.0 (3.8%) 5.6 (13.3%) 0.1 (1.9%) 38.5 (27.7%) 0.1 (1.3%) 25.5 (23.0%) 0.1 (1.3%)
Other organic mattera 13.6 (52.7%) 26.2 (62.1%) 4.7 (66.0%) 84.6 (60.9%) 4.1 (49.0%) 69.6 (63.0%) 4.8 (57.4%)
Crustalb 8.34 (32.3%) 6.31 (14.9%) 1.98 (27.9%) 0.85 (0.6%) 1.61 (19.2%) 0.93 (0.8%) 1.52 (18.1%)
Elemental carbon 0.30 (3.6%) 1.66 (4.0%) 0.32 (4.5%) 16.6 (11.9%) 0.30 (3.6%) 15.7 (14.2%) 0.40 (4.8%)
Sulfate 0.38(1.5%) 0.87 (2.1%) 0.38 (5.4%) 0.62 (0.8%) 0.86 (10.3%) 0.84 (0.8%) 1.08 (12.8%)
Nitrate 0.16 (0.6%) 0.30 (0.7%) 0.07 (1.0%) 0.20 (0.1%) 0.14 (1.8%) 0.06 (0.0%) 0.10 (1.3%)
Ammonium 0.13 (0.5%) 0.19 (0.5%) 0.11 (1.6%) 0.50 (0.4%) 0.24 (2.9%) 0.39 (0.4%) 0.31 (3.7%)
TEOc 0.66 (2.6%) 0.76 (1.8%) 0.10 (1.5%) 0.82 (0.6%) 0.10 (1.3%) 0.28 (0.3%) 0.11 (1.4%)
Unexplained mass 1.01 (3.9%) 0.22 (0.5%) 0.70 (9.8%) −4.02 (−2.9%) 0.89 (10.6%) −2.94 (−2.7%) −0.07 (−0.8%)
Open in a new tab
a

Other Organic Matter = (OC * 1.76) - ΣSugar OC.

Two additional summary variables were calculated including a soil or crustal variable calculated as the sum of oxides of crustal elements (Chow et al., 2015), and trace element oxides (TEO) which was calculated as the sum of other atmospheric metal oxides (Landis et al., 2001). On average, the PM mass collected in the AOSR were well attributed with <5% absolute mass left unexplained. Horse River Fire source profiles were developed from each of the fire impacted samples in Table 5 by dividing the chemical concentration for each sample by the total mass on the filter. The resulting profiles (Appendix A.14) were fairly uniform across the network, are in good agreement with the profiles previously observed during the 2011 Richardson Backcountry Fire reported by Landis et al. (2012), and can be used as a “fingerprint” for fire influence in future studies in northern Alberta.

4. Conclusions

Smoke from the 2016 Horse River Fire had a major air quality impact on the city of Fort McMurray and other communities in the AOSR resulting in 188 exceedances of 1-hr and 24-hr AAAQOs, and Air Quality Health Index (AQHI) “very high” health risk designations. The ambient PM2.5 concentrations measured at the AMS community sites during fire impact periods are on the same order as PM2.5 concentrations that have been reported for wildland firefighter occupational exposures, and in some cases higher by >2 orders of magnitude above ambient community concentrations where public health effects from wildland fire smoke were documented. The PM2.5 concentration enhancements at the six WBEA community monitoring sites during fire impact periods ranged from a factor of 19 at the AMS-21 (Conklin) site to 54 at the AMS-6 (Fort McMurray) site with associated mean/maximum hourly concentrations of 82/357 μg m−3 and 291/5229 μg m−3, respectively. The highest maximum hourly PM2.5 concentration of 6106 μg m−3 was observed at the AMS-16 Muskeg River Industrial site (Appendix B.3f).

Significant enhancements of biomass combustion related pollutants NMHC, NH3, BC, Delta-C, TRS, NOx, and H2S were also observed across the WBEA network; this is the first observation of measurable reduced sulfur compound (TRS/H2S) emissions from a wildfire. Normalized excess emission ratios were calculated from ~400 hourly Horse River Fire impacted observations at the AMS-7 for PM2.5 (149.87 ± 3.37 μg m−3 ppm−1), THC (0.274 ± 0.002 ppm ppm−1), NMHC (0.169 ± 0.001 ppm ppm−1), CH4 (0.104 ± 0.001 ppm ppm−1), TRS (0.694 ± 0.007 ppb ppm−1), SO2 (0.519 ± 0.040 ppb ppm−1), NO (0.412 ± 0.045 ppb ppm−1), NO2 (1.968 ± 0.053 ppb ppm−1), and NOX (2.337 ± 0.077 ppb ppm−1). The emission of NH3 from the Horse River Fire also had a pronounced impact on annual concentrations measured at the WBEA forest health monitoring sites (up to ~14 times higher than non-fire impacted periods), as well as the resulting potential nutrient loadings to terrestrial and aquatic ecosystems in the AOSR. Filter-based PM2.5 samples impacted by Horse River Fire emissions exhibited enhancements in smoke related constituents such as OM, representing ~90% of mass with ~25% of that mass from ΣSugar (primarily levoglucosan, mannosan, and galactosan). The Horse River Fire source profiles were fairly uniform across the network, and can be used as a “fingerprint” for fire influence in future studies in northern Alberta.

Supplementary Material

Supp

HIGHLIGHTS.

  • Horse River Fire had a major air quality impact on the city of Fort McMurray.

  • PM2.5 enhancements at the community monitoring sites ranged from a factor of 19–54.

  • Significant enhancements of NMHC, NH3, BC, DC, TRS, NOx, and H2S were observed.

  • First observations of reduced sulfur compounds (TRS/H2S) emissions from a wildfire

  • Fire PM2.5 profiles were uniform across the network, can be used as a fingerprint.

Acknowledgements

This work was funded by the WBEA (AD113–16). The content and opinions expressed by the authors do not necessarily reflect the views of the WBEA or of the WBEA membership. The EPA through its Office of Research and Development contributed to this research as part of a study into emissions from wildland fires and their impact on ambient air quality. The views expressed in this paper are those of the authors and do not necessarily reflect the views or policies of EPA. It has been subjected to Agency review and approved for publication. Mention of trade names or commercial products do not constitute an endorsement or recommendation for use. We thank WBEA personnel for managing the ambient monitoring network and data quality assurance; Brad Edgerton (ARA) filter weighing, Mike Fort (ARA) for filter extraction and IC/DRC-ICPMS analysis, and Andrew Weakley (UC Davis) for FTIR analysis.

Footnotes

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.scitotenv.2017.10.008.

References

  1. Adetona O, Dunn K, Hall DB, Achtemeier G, Stock A, Naeher LP, 2011. Personal PM2.5 exposure among wildland firefighters working at prescribed forest burns in Southeastern United States. J. Occup. Environ. Hyg 8 (8):503–511. 10.1080/15459624.2011.595257. [DOI] [PubMed] [Google Scholar]
  2. Adetona O, Reinhardt TE, Domitrovich J, Broyles G, Adetona AM, Kleinman MT, Ottmar RD, Naeher LP, 2016. Review of the health effects of wildland fire smoke on wildland fire fighters and the public. Inhal. Toxicol 28 (3):95–139. 10.3109/08958378.2016.1145771. [DOI] [PubMed] [Google Scholar]
  3. Aiken AC, DeCarlo PF, Kroll JH, Worsnop DR, Huffman JA, Docherty KS, Ulbrich IM, Mohr C, Kimmel JR, Sueper D, Sun Y, Zhang Q, Trimborn A, Northway M, Ziemann PJ, Canagaratna MR, Onasch TB, Alfarra MR, Prevot ASH, Dommen J, Duplissy J, Metzger A, Baltensperger U, Jimenez JL, 2008. O/C and OM/OC ratios of primary, secondary, and ambient organic aerosols with high resolution time-of-flight aerosol mass spectrometry. Environ. Sci. Technol 42:4478–4485. 10.1021/es703009q. [DOI] [PubMed] [Google Scholar]
  4. Akagi SK, Yokelson RJ, Wiedinmyer C, Alvarado MJ, Reid JS, Karl T, Crounse JD, Wennberg PO, 2011. Emission factors for open and domestic biomass burning for use in atmospheric models. Atmos. Chem. Phys 11:4039–4072. 10.5194/acp-11-4039-2011. [DOI] [Google Scholar]
  5. Alberta Agriculture and Forestry, 2017. Historical Wildfire Database Last Accessed September 19 http://wildfire.alberta.ca/resources/historical-data/historical-wildfire-database.aspx.
  6. Allen GA, Babich P, Poirot RL, 2004. Evaluation of a new approach for real-time assessment of wood smoke PM. Proceedings of the Regional and Global Perspectives on Haze: Causes, Consequences, and Controversies, Air and Waste Management Association Visibility Specialty Conference, Asheville, NC paper #16. [Google Scholar]
  7. Baker KR, Woody MC, Tonnesen GS, Hutzell W, Pye HOT, Beaver MR, Pouliot G, Pierce T, 2016. Contribution of regional-scale fire events to ozone and PM2.5 air quality estimated by photochemical modeling approaches. Atmos. Environ 140:539–554. 10.1016/j.atmosenv.2016.06.032. [DOI] [Google Scholar]
  8. Benedict KB, Prenni AJ, Carrico CM, Sullivan AP, Schichtel BA, Collett JL Jr., 2017. Enhanced concentrations of reactive nitrogen species in wildfire smoke. Atmos. Environ 148:8–15. 10.1016/j.atmosenv.2016.10.030. [DOI] [Google Scholar]
  9. Bytnerowicz A, Hsu YM, Percy K, Legge A, Fenn ME, Schilling S, Fraczek W, Alexander D, 2016. Ground-level air pollution changes during a boreal wildland mega-fire. Sci. Total Environ 572:755–769. 10.1016/j.scitotenv2016.07.052. [DOI] [PubMed] [Google Scholar]
  10. Canadian Forest Service, 2016. National Fire Database – Agency Fire Data Natural Resources Canada, Northern Forestry Centre, Edmonton, Alberta. [Google Scholar]
  11. Chow JC, Watson JG, Pritchett LC, Pierson WR, Frazier CA, Purcell RG, 1993. The DRI Thermal/Optical Reflectance carbon analysis system: description, evaluation and applications in U.S. air quality studies. Atmos. Environ 27A (8):1185–1201. 10.1016/0960-1686(93)90245-T. [DOI] [Google Scholar]
  12. Chow JC, Watson JG, Chen LWA, Chang MCO, Robinson NF, Trimble D, Kohl S, 2007. The IMPROVE-A temperature protocol for thermal/optical carbon analysis: maintaining consistency with a long-term database. J. Air Waste Manage. Assoc 57:1014–1023. 10.3155/1047-3289.57.9.1014. [DOI] [PubMed] [Google Scholar]
  13. Chow JC, Lowenthal DH, Chen LWA, Wang X, Watson JG, 2015. Mass reconstruction methods for PM2.5: a review. Air Qual. Atmos. Health 8:243–263. 10.1007/s11869-015-0338-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cofer WR III, Winstead EL, Stocks BJ, Goldammer JG, Cahoon DR, 1998. Crown fire emissions of CO2, CO, H2, CH4, and TNMHC from a dense jack pine boreal forest fire. Geophys. Res. Lett 25:3919–3922. 10.1029/1998GL900042. [DOI] [Google Scholar]
  15. Collier S, Zhou S, Onasch TB, Jaffe DA, Kleinman L, Sedlacek AJ III, Briggs NL, Hee J, Fortner E, Shilling JE, Worsnop D, Yokelson RJ, Parworth C, Ge H, Xu J, Butterfield Z, Chand D, Dubey MK, Pekour MS, Springston S, Zhang Q, 2016. Regional influence of aerosol emissions from wildfires driven by combustion efficiency: insights from the BBOP campaign. Environ. Sci. Technol 50:8613–8622. 10.1021/acs.est.6b01617. [DOI] [PubMed] [Google Scholar]
  16. Dennis A, Fraser M, Anderson S, Allen D, 2002. Air pollutant emissions associated with forest, grassland, and agricultural burning in Texas. Atmos. Environ 36:3779–3792. 10.1016/S1352-2310(02)00219-4. [DOI] [Google Scholar]
  17. Donahue NM, Robinson AL, Pandis SN, 2009. Atmospheric organic particulate matter: from smoke to secondary organic aerosol. Atmos. Environ 43:94–106. 10.1016/j.atmosenv.2008.09.055. [DOI] [Google Scholar]
  18. Drinovec L, Močnik G, Zotter P, Prévôt ASH, Ruckstuhl C, Coz E, Rupakheti M, Sciare J, Müller T, Wiedensohler A, Hansen ADA, 2015. The “dual-spot” Aethalometer: an improved measurement of aerosol black carbon with real-time loading compensation. Atmos. Meas. Tech 8:1965–1979. 10.5194/amt-8-1965-2015. [DOI] [Google Scholar]
  19. Edgerton ES, Fort JM, Baumann K, Graney JR, Landis MS, Berryman S, Krupa S, 2012. Method for extraction and multi-element analysis of Hypogymnia Physodes samples from the Athabasca Oil Sands Region In: Percy Kevin (Ed.), Alberta Oil Sands: Energy, Industry and the Environment Elsevier, Oxford, England. [Google Scholar]
  20. Government of Alberta, 2017. Alberta Ambient Air Quality Objectives and Guidelines Summary, AEP, Air Policy, 2016, No. 2, June 29, 2017 Last accessed September 19 http://aep.alberta.ca/air/legislation/ambient-air-quality-objectives/documents/AAQO-Summary-Jun29-2017.pdf.
  21. Hecobian A, Liu Z, Hennigan CJ, Huey LG, Jimenez JL, Cubison MJ, Vay S, Diskin GS, Sachse GW, Wisthaler A, Mikoviny T, Weinheimer AJ, Liao J, Knapp DJ, Wennberg PO, Kürten A, Crounse JD, St. Clair J, Wang Y, Weber RJ, 2011. Comparison of chemical characteristics of 495 biomass burning plumes intercepted by the NASA DC-8 aircraft during the ARCTAS/CARB-2008 field campaign. Atmos. Chem. Phys 11:13325–13337. 10.5194/acp-11-13325-2011. [DOI] [Google Scholar]
  22. Hornbrook RS, Blake DR, Diskin GS, Fried A, Fuelberg HE, Meinardi S, Mikoviny T, Richter D, Sachse GW, Vay SA, Walega J, Weibring P, Weinheimer AJ, Wiedinmyer C, Wisthaler A, Hills A, Riemer DD, Apel EC, 2011. Observations of nonmethane organic compounds during ARCTAS – part 1: biomass burning emissions and plume enhancements. Atmos. Chem. Phys 11:11103–11130. 10.5194/acp-11-11103-2011. [DOI] [Google Scholar]
  23. Kim YH, Warren S, Krantz QT, King C, Jaskot R, Preston WT, George BJ, Hays MD, Landis MS, Higuchi M, DeMarini D, Gilmour MI, 2017. Mutagenicity and lung toxicity of smoldering versus flaming emissions from various biomass fuels: implications for health effects from wildland fires Environ. Health Perspect (In Press). [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kitzberger T, Brown PM, Heyerdahl EK, Swetnam TW, Veblen TT, 2007. Contingent Pacific–Atlantic Ocean influence on multi-century wildfire synchrony over western North America. P. Natl. Acad. Sci. USA 104 (2):543–548. 10.1073/pnas.0606078104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kleindienst TE, Hudgens ER, Smith DF, McElroy FF, Bufalini JJ, 1993. Comparison of chemiluminescence and ultraviolet ozone monitor responses in the presence of humidity and photochemical pollutants. Air Waste 43:213–222. 10.1080/107316X.1993.10467128. [DOI] [PubMed] [Google Scholar]
  26. Landis MS, Norris GA, Williams RW, Weinstein JP, 2001. Personal exposures to PM2.5 mass and trace elements in Baltimore, MD, USA. Atmos. Environ 35:6511–6524. 10.1016/S1352-2310(01)00407-1. [DOI] [Google Scholar]
  27. Landis MS, Pancras JP, Graney JR, Stevens RK, Percy KE, Krupa S, 2012. Receptor modeling of epiphytic lichens to elucidate the sources and spatial distribution of inorganic air pollution in the Athabasca Oil Sands Region. In: Percy KE (Ed.), Alberta Oil Sands: Energy, Industry and the Environment. Elsevier, Oxford, England, p. 427. [Google Scholar]
  28. Landis MS, Pancras JP, Graney JR, White EM, Edgerton ES, Legge A, Percy KE, 2017. Source apportionment of ambient fine and coarse particulate matter at the Fort McKay community site, in the Athabasca Oil Sands Region, Alberta, Canada. Sci. Total Environ 584–585:105–117. 10.1016/j.scitotenv.2017.01.110. [DOI] [PubMed] [Google Scholar]
  29. Lee GE, Breysse PN, McDermott A, Eftim SE, Geyh A, Berman JD, Curriero FC, 2014. 2014 Canadian forest fires and the effects of long-range transboundary air pollution on hospitalization among the elderly. ISPRS Int. Journal of Geo-Inf 3 (2):713–731. 10.3390/ijgi3020713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Littell JS, McKenzie D, Peterson DL, Westerling AL, 2009. Climate and wildfire area burned in western U. S. ecoprovinces, 1916–2003. Ecol. Appl 19:1003–1021. 10.1890/07-1183.1. [DOI] [PubMed] [Google Scholar]
  31. Liu JC, Pereira G, Uhl SA, Bravo MA, Bell ML, 2015. A systematic review of the physical health impacts from non-occupational exposure to wildfire smoke. Environ. Res 136:120–132. 10.1016/j.envres.2014.10.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Liu X, Huey LG, Yokelson RJ, Selimovic V, Simpson IJ, Müller M, Jimenez JL, Campuzano-Jost P, Beyersdorf AJ, Blake DR, Butterfield Z, Choi Y, Crounse JD, Day DA, Diskin GS, Dubey MK, Fortner E, Hanisco TF, Hu W, King LE, Kleinman L, Meinardi S, Mikoviny T, Onasch TB, Palm BB, Peischl J, Pollack IB, Ryerson TB, Sachse GW, Sedlacek AJ, Shilling JE, Springston S, St. Clair JM, Tanner DJ, Teng AP, Wennberg PO, Wisthaler A, Wolfe GM, 2017. Airborne measurements of western U.S. wildfire emissions: Comparison with prescribed burning and air quality implications. J. Geophys. Res.-Atmos 122:6108–6129. 10.1002/2016JD026315. [DOI] [Google Scholar]
  33. Miller J, Safford H, Crimmins M, Thode A, 2009. Quantitative evidence for increasing forest fire severity in the Sierra Nevada and Southern Cascade Mountains, California and Nevada, USA. Ecosystems 12:16–32. 10.1007/s10021-008-9201-9. [DOI] [Google Scholar]
  34. Miranda AI, Martins V, Cascão P, Amorim JH, Valente J, Tavares R, Borrego C, Tchepel O, Ferreira AJ, Cordeiro CR, Viegas DX, Ribeiro LM, Pita LP, 2010. Monitoring of firefighters exposure to smoke during fire experiments in Portugal. Environ. Int 36:736–745. 10.1016/j.envint.2010.05.009. [DOI] [PubMed] [Google Scholar]
  35. Naeher LP, Achtemeier GL, Glitzenstein JS, Streng DR, MacIntosh D, 2006. Real-time and time-integrated PM2.5 and CO from prescribed burns in chipped and non-chipped plots: firefighter and community exposure and health implications. J. Expo. Sci. Environ. Epidemiol 16 (4):351–361. 10.1038/sj.jes.7500497. [DOI] [PubMed] [Google Scholar]
  36. Nance JD, Hobbs PV, Radke LF, Ward DE, 1993. Airborne measurements of gases and particles from an Alaskan Wildfire. J. Geophys. Res 98:14873–14882. 10.1029/93JD01196. [DOI] [Google Scholar]
  37. Natural Resources Canada, 2017. 8 Facts about Canada’s boreal forest Last accessed October 18, 2017 www.nrcan.gc.ca/forests/boreal/17394.
  38. Ortega AM, Day DA, Cubison MJ, Brune WH, Bon D, de Gouw JA, Jimenez JL, 2013. Secondary organic aerosol formation and primary organic aerosol oxidation from biomass-burning smoke in a flow reactor during FLAME-3. Atmos. Chem. Phys 13:11551–11571. 10.5194/acp-13-11551-2013. [DOI] [Google Scholar]
  39. O’Shea SJ, Allen G, Gallagher MW, Bauguitte SJ-B, Illingworth SM, Le Breton M, Muller JBA, Percival CJ, Archibald AT, Oram DE, Parrington M, Palmer PI, Lewis AC, 2013. Airborne observations of trace gases over boreal Canada during BORTAS: campaign climatology, air mass analysis and enhancement ratios. Atmos. Chem. Phys 13:12451–12467. 10.5194/acp-13-12451-2013. [DOI] [Google Scholar]
  40. Percy KE, 2013. Ambient air quality and linkage to ecosystems in the Athabasca Oil Sands, Alberta. Geosci. Can 40:182–201. https://doi.org/10.12789/ geocanj.2013.40.014. [Google Scholar]
  41. Phillips-Smith C, Jeong C-H, Healy RM, Daebk-Zlotorzynska E, Celo V, Brook JR, Evans G, 2017. Sources of particulate matter in the Athabasca Oil Sands Region: investigation through a comparison of trace element measurement methodologies. Atmos. Chem. Phys. Discuss 10.5194/acp-2016-966 (in review). [DOI]
  42. Reid CE, Brauer M, Johnson FH, Jerrett J, Balmes JR, Elliott CT, 2016. Critical review of health impacts of wildfire smoke exposure. Environ. Health Perspect 124:1334–1343. 10.1289/ehp.1409277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Saarnio K, Aurela M, Timonen H, Saarikoski S, Teinila K, Makela T, Sofiev M, Koskinen J, Aalto PP, Kulmala M, Kukkonen J, Hillamo R, 2010. Chemical composition of fine particles in fresh smoke plumes from boreal wild-land fires in Europe. Sci. Total Environ 408:2527–2542. 10.1016/j.scitotenv.2005.01.024. [DOI] [PubMed] [Google Scholar]
  44. Schweizer D, Cisneros R, Shaw G, 2016. A comparative analysis of temporary and permanent beta attenuation monitors: the importance of understanding data and equipment limitations when creating PM2.5 air quality health advisories. Atmos. Pollut. Res 7:865–875. 10.1016/j.apr.2016.02.003. [DOI] [Google Scholar]
  45. Simpson IJ, Akagi SK, Barletta B, Blake NJ, Choi Y, Diskin GS, Fried A, Fuelberg HE, Meinardi S, Rowland FS, Vay SA, Weinheimer AJ, Wennberg PO, Wiebring P, Wisthaler A, Yang M, Yokelson RJ, Blake DR, 2011. Boreal forest fire emissions in fresh Canadian smoke plumes: C1–C10 volatile organic compounds (VOCs), CO2, CO, NO2, NO, HCN and CH3CN. Atmos. Chem. Phys 11:6445–6463. 10.5194/acp-11-6445-2011. [DOI] [Google Scholar]
  46. Singh HB, Anderson BE, Brune WH, Cai C, Cohen RC, Crawford JH, Cubison MJ, Czech EP, Emmons L, Fuelberg HE, Huey G, Jacob DJ, Jimenez JL, Kaduwela A, Kondo Y, Mao J, Olson JR, Sachse GW, Vay SA, Weinheimer A, Wennberg PO, Wisthaler A, 2010. Pollution influences on atmospheric composition and chemistry at high northern latitudes: boreal and California forest fire emissions. Atmos. Environ 44:4553–4564. 10.1016/j.atmosenv.2010.08.026. [DOI] [Google Scholar]
  47. Spicer CW, Joseph DW, Ollison WM, 2010. A re-examination of ambient air ozone monitor interferences. J. Air Waste Manag. Assoc 60:1353–1364. 10.3155/1047-3289.60.11.1353. [DOI] [PubMed] [Google Scholar]
  48. Stocks BJ, Mason JA, Todd JB, Bosch EM, Wotton BM, Amiro BD, Flannigan MD, Hirsch KG, Logan KA, Martell DL, Skinner WR, 2003. Large forest fires in Canada, 1959–1997. J. Geophys. Res 108, D08149 10.1029/2001JD000484. [DOI] [Google Scholar]
  49. Sullivan AP, Frank N, Kenski DM, Collett JL Jr., 2011a. Application of high-performance anion-exchange chromatography – pulsed amperometric detection for measuring carbohydrates in routine daily filter samples collected by a national network: 2. Examination of sugar alcohols/polyols, sugars, and anhydrosugars in the upper Midwest. J. Geophys. Res 116, D08303 10.1029/2010JD014169. [DOI] [Google Scholar]
  50. Sullivan AP, Frank N, Onstad G, Simpson CD, Collett JL Jr., 2011b. Application of high-performance anion-exchange chromatography - pulsed amperometric detection for measuring carbohydrates in routine daily filter samples collected by a national network: 1. Determination of the impact of biomass burning in the Upper Midwest. J. Geophys. Res 116, D08302 10.1029/2010JD014166. [DOI] [Google Scholar]
  51. Sullivan AP, May AA, Lee T, McMeeking GR, Kreidenweis SM, Akagi SK, Yokelson RJ, Urbanski SP, Collett JL Jr., 2014. Airborne-based source smoke marker ratios from prescribed burning. Atmos. Chem. Phys 14:10535–10545. 10.5194/acp-14-10535-2014. [DOI] [Google Scholar]
  52. United Stated National Institute for Occupational Safety and Health, Manual of Analytical Methods (NMAM), O’Connor PF, Schlecht PC, 2003. Monitoring of Diesel Particulate Exhaust in the Workplace, Chapter Q, Third Supplement to NMAM, 4th Edition, NIOSH, Cincinnati, OH: DHHS (NIOSH) Publication No. 2003–154. [Google Scholar]
  53. United States Department of Agriculture, 2014. United States Forest Resource Facts and Historical Trends, FS-1035: Washington, DC: https://www.fia.fs.fed.us/library/brochures/docs/2012/ForestFacts_1952-2012_English.pdf. [Google Scholar]
  54. United States Department of Agriculture, 2016. Forest Health Monitoring: National Status, Trends, and Analysis 2015 Forest Service Research & Development, Southern Research Station, SRS-213, Asheville, NC. [Google Scholar]
  55. United States Environmental Protection Agency, 1997. Method 300.1 Determination of Inorganic Anions in Drinking Water by Ion Chromatography http://water.epa.gov/scitech/drinkingwater/labcert/upload/met300.pdf Last accessed September 19, 2017.
  56. United States Environmental Protection Agency, 1999a. Laboratory Study to Explore Potential Interferences to Air Quality Monitors. Office of Air Quality Planning and Standards, EPA-454/C-00–002: Research Triangle Park, NC: https://www3.epa.gov/ttnamti1/archive/files/ambient/criteria/reldocs/finalreport.pdf Last accessed September 19, 2017. [Google Scholar]
  57. United States Environmental Protection Agency, 1999b. Compendium Method IO-4.2: Determination of Reactive acidic and basic gases and strong acidity of atmospheric fine particles (<2.5 μm) Office of Research and Development, Compendium of Methods for the Determination of Inorganic Compounds in Ambient Air, EPA/625/R-96/010a: Cincinnati, OH: https://www3.epa.gov/ttnamti1/files/ambient/inorganic/mthd-4-2.pdf Last accessed September 19, 2017. [Google Scholar]
  58. United States Environmental Protection Agency, August, 2015. 2011 National Emissions Inventory (NEI) Technical Support Document, Office of Air Quality Planning and Standards, Research Triangle Park, NC, https://www.epa.gov/sites/production/files/2015-10/documents/nei2011v2_tsd_14aug2015.pdf Last accessed September 19, 2017. [Google Scholar]
  59. United States Environmental Protection Agency, 2016. Climate Change Indicators in the United States: 2016 Fourth Edition, EPA 430-R-16–004: Washington, DC: https://www.epa.gov/sites/production/files/2016-08/documents/climate_indicators_2016.pdf Last accessed September 19, 2017. [Google Scholar]
  60. United States Wildland Fire Leadership Council, 2016. Monitoring Trends in Burn Severity (MTBS) Project Database Last accessed September 19, 2017 http://mtbs.gov/dataaccess.html.
  61. United States National Interagency Fire Center, 2017. Historical Wildland Fire Information Last accessed October 18 https://www.nifc.gov/fireInfo/fireInfo_statistics.html.
  62. Urbanski SP, 2013. Combustion efficiency and emission factors for wildfire-season fires in mixed conifer forests of the northern Rocky Mountains, US. Atmos. Chem. Phys 13:7241–7262. 10.5194/acp-13-7241-2013. [DOI] [Google Scholar]
  63. Weakley AT, Takahama S, Dillner AM, 2016. Ambient aerosol composition by infrared spectroscopy and partial least-squares in the chemical speciation network: Organic carbon with functional group identification. Aerosp. Sci. Technol:1096–1111 10.1080/02786826.2016.1217389. [DOI] [Google Scholar]
  64. Westerling AL, 2016. Increasing western U.S. forest wildfire activity: Sensitivity to changes in the timing of spring. Philos. Trans. R. Soc. B 371 10.1098/rstb.20150178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Wood Buffalo Environmental Association (WBEA), 2016. Annual Report – Volume I Continuous Monitoring, March 17 Wbea.org/resources/reports-and-publications/ambient-air-monitoring-reports/ambient-annual-report Last accessed September 19, 2017.
  66. Worden H, Beer R, Rinsland CP, 1997. Airborne infrared spectroscopy of 1994 western wildfires. J. Geophys. Res 102:1297–1299. 10.1029/96JD02982. [DOI] [Google Scholar]
  67. Yokelson RJ, Goode JG, Ward DE, Susott RA, Babbitt RE, Wade DD, Bertschi I, Griffith DWT, Hao WM, 1999. Emissions of formaldehyde, acetic acid, methanol, and other trace gases from biomass fires in North Carolina measured by airborne Fourier transform infrared spectroscopy. J. Geophys. Res 104:30109–30125. 10.1029/1999JD900817. [DOI] [Google Scholar]
  68. Yokelson RJ, Crounse JD, DeCarlo PF, Karl T, Urbanski S, Atlas E, Campos T, Shinozuka Y, Kapustin V, Clarke AD, Weinheimer A, Knapp DJ, Montzka DD, Holloway J, Weibring P, Flocke F, Zheng W, Toohey D, Wennberg PO, Wiedinmyer C, Mauldin L, Fried A, Richter D, Walega J, Jimenez JL, Adachi K, Buseck PR, Hall SR, Shetter R, 2009. Emissions from biomass burning in the Yucatan. Atmos. Chem. Phys 9:5785–5812. 10.5194/acp-9-5785-2009. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp
NIHMS983401-supplement-Supp.pdf (3.7MB, pdf)

RESOURCES