Enhanced concentrations of reactive nitrogen species in wildfire smoke

Atmospheric Environment 148 (2017) 8e15

Contents lists available at ScienceDirect

Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv

Enhanced concentrations of reactive nitrogen species in wildfire smoke Katherine B. Benedict a, Anthony J. Prenni b, Christian M. Carrico c, Amy P. Sullivan a, Bret A. Schichtel d, Jeffrey L. Collett Jr. a, * a

Department of Atmospheric Science, Colorado State University, Campus Delivery 1371, Fort Collins, CO 80523, USA National Park Service, Air Resources Division, Lakewood, CO 80225, USA New Mexico Institute of Mining and Technology, Socorro, NM 87801, USA d National Park Service, Air Resources Division, Fort Collins, CO 80523, USA b c

h i g h l i g h t s þ
NH3, NOx, HTC-RN, NO 3 , and NH4 were measured in fresh smoke from two wildfires.
Enhanced concentrations of all reactive nitrogen species except NHþ 4 were observed.
Ammonia during smoke-impacted periods was enhanced by a factor of 20.
Excess mixing ratios NH3 relative to excess CO were found to be 0.027 ± 0.002.
Excess mixing ratios NOx relative to excess CO were found to be 0.057 ± 0007.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 June 2016 Received in revised form 6 October 2016 Accepted 17 October 2016 Available online 18 October 2016

During the summer of 2012 the Hewlett Gulch and High Park wildfires burned an area of 400 km2 northwest of Fort Collins, Colorado. These fires both came within 20 km of the Department of Atmospheric Science at Colorado State University, allowing for extensive measurements of smoke-impacted air masses over the course of several weeks. In total, smoke plumes were observed at the measurement site for approximately 125 h. During this time, measurements were made of multiple reactive nitrogen compounds, including gas phase species NH3, NOx, and HNO3, and particle phase species NO 3 and NHþ 4 , plus an additional, unspeciated reactive nitrogen component that is measured by high temperature conversion over a catalyst to NO. Concurrent measurements of CO, levoglucosan and PM2.5 served to confirm the presence of smoke at the monitoring site. Significant enhancements were observed for all of the reactive nitrogen species measured in the plumes, except for NHþ 4 which did not show enhancements, likely due to the fresh nature of the plume, the presence of sufficient regional ammonia to have already neutralized upwind sulfate, and the warm conditions of the summer measurement period which tend to limit ammonium nitrate formation. Excess mixing ratios for NH3 and NOx relative to excess mixing ratios of CO in the smoke plumes, DNH3/DCO (ppb/ppb) and DNOx/DCO (ppb/ppb), were determined to be 0.027 ± 0.002 and 0.0057 ± 0.0007, respectively. These ratios suggest that smoldering combustion was the dominant source of smoke during our plume interceptions. Observations from prior relevant laboratory and field measurements of reactive nitrogen species are also briefly summarized to help create a more comprehensive picture of reactive nitrogen and fire. © 2016 Published by Elsevier Ltd.

Keywords: Biomass burning Ammonia Reactive nitrogen Emissions

1. Introduction Biomass burning represents an important source of particles

* Corresponding author. E-mail address: [email protected] (J.L. Collett). http://dx.doi.org/10.1016/j.atmosenv.2016.10.030 1352-2310/© 2016 Published by Elsevier Ltd.

and trace gases to the global atmosphere, with 33e43 million km2 burned annually (Giglio et al., 2010). Although global burn area is dominated by savanna fires in Africa (Giglio et al., 2010; van der Werf et al., 2006), fire carbon emissions largely track burning of forested areas (van der Werf et al., 2006). In temperate North America, approximately 1.5 Mha are burned annually. Large

K.B. Benedict et al. / Atmospheric Environment 148 (2017) 8e15

wildfires (>400 ha or 4 km2) in the western US increased significantly in the mid-1980s (Westerling et al., 2006), and more recent data demonstrate that wildfires are becoming larger in size relative to 20th century fires (www.nifc.gov). Such increases may be tied to a warming climate (Moritz et al., 2012). In the Greater Yellowstone ecosystem, Westerling et al. (2011) showed that a difference in average summer and spring temperature of only ~0.5  C is sufficient to distinguish extreme fire years from most other years. With a warmer future climate predicted, fire activity is expected to increase in the US (Flannigan et al., 2000; Moritz et al., 2012; Spracklen et al., 2007), including fires caused by lightning (Price and Rind, 1994). During years of large acreage burned from wildfires, biomass burning emissions can account for up to 65% of PM2.5 organic carbon (OC) emissions in the western US (Spracklen et al., 2007). Regional variability in OC concentrations throughout the continental US can, in part, be explained by western US wildfire emissions. Additionally transport of air masses across international boundaries in some regions results in significant increases in OC mass with implications for ambient air quality standards, visibility, and cloud radiative properties (Ge et al., 2014; Park et al., 2003; Targino et al., 2013). In addition to carbon, nitrogen compounds are abundant in biomass burning emissions (Yokelson et al., 2008; Wiedinmyer et al., 2006, 2011; Akagi et al., 2011; Andreae and Merlet, 2001). In fact, global anthropogenic reactive nitrogen emissions were dominated by biomass burning until the middle of cek and Posch, 2011). Ammonia (NH3) and the 20th century (Kopa nitrogen oxides (NOx) emissions can account for 20e50% of the nitrogen consumed during burns, with lesser contributions from nitric acid (HNO3) and particulate nitrogen species (McMeeking et al., 2009). Additionally, nitrous acid (HONO), hydrogen cyanide (HCN), and acetonitrile (CH3CN) are important N-containing compounds emitted by biomass burning (Akagi et al., 2011). Emissions ratios (ER) to CO (mol/mol) for a variety of fuel types are reported in Stockwell et al. (2015) for a number of nitrogen containing gases including NH3 (max ER 0.194), HCN (0.022), NO (0.0893), NO2 (0.0588), HONO (0.0119), CH3CN (0.0073), dimethylamine and ethylamine (C2H7N; 0.0024), acetamide (C2H5NO; 0.00998), triethylamine (C3H9N; 0.0028), assorted amides (C4H9NO; 0.000797), assorted amines (C4H11NO; 0.00023), benzonitrile (C7H5N; 0.00022). Additional N-containing compounds were observed but were unable to be quantified, including acrylonitrile, propanenitrile, pyrrole, and pyridine. Observation of many of these nitrogen-containing organic compounds are few but the results from Stockwell et al. (2015) indicate that together they may account for 0.1e8.7% of fuel nitrogen. The ratio of NH3 to NOx has been shown to be strongly related to modified combustion efficiency (MCE), where a higher ratio of NH3 to NOx is associated with a lower MCE (Goode et al., 2000, 1999; McMeeking et al., 2009; Yokelson et al., 1996), suggesting a tradeoff in the form of nitrogen depending on combustion conditions. An average of measurements from more than 19 wildfires suggest that emissions of ammonia are 2% of carbon monoxide (CO) emissions in nitrogen limited environments (Goode et al., 2000; Nance et al., 1993) such as those found in the Rockies (Baron et al., 2000). However, there is evidence that fire environment (e.g. atmospheric conditions, soil moisture content) plays an important factor in the DNH3/DCO emission ratio (Hegg et al., 1988). Ammonia concentrations tend to decrease as a plume ages (Goode et al., 2000), relative to CO concentrations, likely due to deposition and conversion to particulate NHþ 4 , while particulate nitrate concentrations generally increase (Hobbs, 2003) from chemical reactions in the plume such as nitric acid conversion to nitrate. The formation of submicron particulate ammonium nitrate can move both gaseous ammonia and gaseous nitric acid into the particle

9

phase. Ammonium nitrate formation, which is reversible, is favored when temperatures are low and relative humidities are high. In US emission inventories, agricultural practices such as livestock manure management and application of chemical fertilizers contribute 80% to national ammonia emissions, while emissions from vehicles that have catalytic converters contribute 7% (Reis et al., 2009). Biofuels also may be a significant source of ammonia to the atmosphere in regions such as Africa, India, and the Amazon (Bertschi, 2003; Christian et al., 2007). Because emissions inventories are designed from regulatory data, wildfire emissions are often not included, and for those inventories that include wildfire emissions of ammonia, attempts to validate them do not always show consistent results (Whitburn et al., 2015). Globally, biomass burning is thought to account for emission of 5.9e10.3 Tg yr1 of NH3 (Andreae and Merlet, 2001; Bouwman et al., 1997; Hegg et al., 1988), with burning in natural ecosystems accounting for 3.2 Tg N yr1 (Asman et al., 1998). In the US the 2011 EPA National Emissions Inventory (NEI) lists wildfires as the third largest source of ammonia nationally, contributing ~5% of ammonia emissions; when all fires (wildfires, prescribed burns, and agricultural fires) are considered, fires make up nearly 9% of total ammonia emissions. Increases in ambient NH3 and other reactive nitrogen species concentrations due to fires have been observed in the laboratory (Christian et al., 2007; Levin et al., 2010; McMeeking et al., 2009; Yokelson et al., 2008, 1996), from ground-based (Benedict et al., 2013a; Prenni et al., 2014) and aircraft-based field studies (Andreae et al., 1988; Goode et al., 2000; Hegg et al., 1988; Hobbs, 2003), from monitoring networks (Chen et al., 2014) and from satellite observations (Coheur et al., 2009; R'Honi et al., 2013; Van Damme et al., 2014; Whitburn et al., 2015). Here we report measurements of ammonia and other reactive nitrogen species during smoke impacts from two wildfires in northern Colorado, the Hewlett Gulch and High Park fires. We also summarize laboratory and field measurements of unspeciated reactive nitrogen species collected from biomass burning events from previous studies by our research group.

2. Experimental 2.1. Colorado wildfires The Hewlett Gulch fire began on May 14, 2012 and was human caused. The fire was approximately 20 km northwest of Fort Collins, CO and ultimately grew to 3100 ha (inciweb.nwcg.gov/ incident/2863/). Grasses and shrubs (mountain mahogany and bitter brush) were the dominant fuels on the southerly facing slopes while fuels on the north facing slopes were timber, primarily Ponderosa pine with some Douglas fir and white fir at higher elevation (Richardson, 2012). Measurements at our laboratory at Colorado State University began May 17 and continued through June 4, well after the fire was contained and when smoke impacts were no longer observed. The High Park fire (inciweb.nwcg.gov/incident/2904/) was significantly larger, covering 35,300 ha (Walker et al., 2012), and contributing significantly to aerosol loading over a large vertical extent in northern Colorado (Val Martin et al., 2013). The lightning-caused fire began in earnest on June 9, and this fire came within 4 km of our laboratory. Measurements from the High Park fire were carried out from June 9 through June 30. At lower elevations during the High Park fire, fuels included lodge pole pine, ponderosa pine, and Douglas-fir (Walker et al., 2012). Upper elevations consisted of forests transitioning to Engelmann spruce and subalpine fir. Pockets of aspen existed throughout forested areas.

10

K.B. Benedict et al. / Atmospheric Environment 148 (2017) 8e15

2.2. Methods During the Hewlett Gulch and High Park fires, measurements of opportunity were made continuously and in real time, with instrumentation housed in the National Park Service Mobile Laboratory, which was located in the Department of Atmospheric Science parking lot at Colorado State University. Continuous measurements included gaseous ammonia (NH3; 5 s data), nitrogen oxides (NOx: NO and NO2; 1 min data), carbon monoxide (CO; 1 min data), an additional reactive nitrogen component (1 min data) and PM2.5 mass (6 min data) which were averaged to hourly data. The experimental setup has been described in detail (Prenni et al., 2014) for most of the measurements. Briefly, a Teledyne 201E chemiluminescence instrument was used to measure NOx (precision ¼ 0.5%) and an additional reactive nitrogen measurement (precision ¼ 1%), described below. Gas phase ammonia measurements were made with a Picarro G1103 cavity ring down spectrometer (precision ¼ 1%). Potential interferences in the Picarro NH3 measurement are discussed in the next section. A trace level CO analyzer (Teledyne-API Inc., Model 300EU) was operated in parallel with the NOx and NH3 instruments which has a precision of 20 ppb for CO less than 1 ppm and 0.5% for higher concentrations. Calibrations of all gas phase instruments were done after the study. In addition to the gas phase measurements, real time measurements of particle mass concentrations were obtained using a tapered element oscillating microbalance (TEOM 1405-DF, Thermo Scientific) (Val Martin et al., 2013). For the additional reactive nitrogen measurement, ambient air passed through a PTFE filter to remove particles and then through a URG annular denuder coated with 10% (w/v) phosphorous acid solution to remove ammonia, amines (Kallinger and Niessner, 1999), and any other rapidly diffusing basic gas phase species. This denuded sample was sent to a stainless steel and platinum converter heated to 825  C (Teledyne 501NH). This high temperature catalyst allows for additional conversion of nitrogen species that were not converted at 315  C, the conversion temperature for the collocated NOx measurements. The NOx concentration was then subtracted from the total measurement, effectively subtracting any nitrogen compounds present that can be converted at 315  C. With particles, ammonia and amines removed, and NOx subtracted, the resulting measurement represents gas phase nitrogen species other than ammonia and amines which are converted at high temperature, but not at 315  C. Collectively, we designate these compounds as high temperature conversion reactive nitrogen (HTC-RN; Prenni et al., 2014). This measurement is thought to include a variety of reduced organic nitrogen compounds; however, due to the lack of selectivity of the measurement, other compounds may also be present (see discussion below). In addition to the real time measurements, 24 h integrated samples were collected using annular denuder (gases)/filter-pack (particles) samplers with a PM2.5 cyclone (University Research Glassware; URG) from 10:00 a.m. to 10:00 a.m. local time during both fires. Gaseous ammonia and nitric acid were collected in denuders upstream of a nylon filter (PALL Nylasorb, 1 mm pore size) that collected PM2.5 aerosol. It is possible for collected fine particle ammonium to be volatilized from the nylon filter so another denuder operated downstream of the filter to capture any ammonium that was lost from the filter (Yu et al., 2006). Any fine particle nitrate that volatilizes from the nylon filter is retained on the filter (as nitric acid) and still measured as nitrate (Yu et al., 2006). Extracted samples were analyzed for inorganic gas (NH3, HNO3,  þ þ 2þ 2þ  SO2) and particulate species (NHþ 4 , NO3 , Na , K , Mg , Ca , Cl , 2 SO4 ) using ion chromatography (IC). Sample collection and analysis procedures followed those described by past studies (Benedict et al., 2013a). Filter extracts were analyzed for levoglucosan using

high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) following the method of Sullivan et al. (2011).

2.2.1. Evaluation instrument sensitivity to potential interferences To identify species that might generate significant response in the NH3 and HTC-RN instruments, we conducted laboratory tests to determine the sensitivity of the instruments to several compounds other than ammonia. Test species were chosen for their atmospheric significance and prevalence in specific sources (e.g, biomass burning). For these tests, the Teledyne and Picarro instruments measured concentrations of the selected species, generated using permeation tubes and a VICI Metronics Dynacalibrator. A Particle Measuring Systems Air Sentry II NH3 monitor was also tested, as this instrument has been used as part of previous studies of our group (Prenni et al., 2014). The Air Sentry II and the Teledyne chemiluminescence instruments are known to be sensitive to species other than NH3. The Air Sentry II measures amines, while the Teledyne chemiluminescence instrument, which is meant to provide a quantitative measure of ammonia, likely measures a variety of species (Dunlea et al., 2007; Parrish et al., 1990). Here sensitivity is reported as the calibrated instrument response divided by the supplied concentration, in percent. For acetonitrile, the only biomass burning-specific compound studied, significant sensitivity was observed in the Teledyne chemiluminescence instrument (70% ± 10%), while the Picarro cavity ringdown and Air Sentry II ion mobility showed negligible (<1%) responses. In contrast, methylamine showed high sensitivities in both the Teledyne and Air Sentry II measurements at 77% ± 9% and 82% ± 5%, respectively, while the Picarro instrument exhibited only slight sensitivity (4.7% ± 1.4%). For non-nitrogen containing VOCs, there was little response in any of the instruments. Acetaldehyde and apinene showed no response (<1%), while isoprene showed minimal sensitivity (3.5%) in the Teledyne chemiluminescence instrument only. In other work summarized by Andreae and Merlet (2001) emission ratios for isoprene and terpenes are approximately 10% that of ammonia from various types of biomass burning. The lower emissions of both isoprene and terpenes compared to ammonia, and the small signal of biogenic VOCs in the Teledyne chemiluminescence instrument, suggest the VOC artifact is low in this measurement for measurement of biomass burning smoke. Interference from VOCs like isoprene and other terpenes might be more problematic during non-smoke measurements when isoprene concentrations are similar or greater in concentration than ammonia.

3. Results and discussion Measurements of NH3, NOx, CO, HTC-RN, and particulate matter began on May 17, 2012, shortly after the start of the Hewlett Gulch fire, and continued through June 30, 2012, when the fire was fully contained. Denuder/filter-pack measurements were made over a similar timespan, but were not continuous. Fig. 1 shows a timeline of the real-time measurements collected from the NPS Mobile Laboratory. Periods which were impacted by emissions from the fires show large increases in all of the measured species except PM2.5 ammonium. For example, CO concentrations increased by more than a factor of 25 when impacted by smoke relative to background concentrations, and PM2.5 concentrations increased by over a factor of 100. Smoke impacts most often occurred in the morning hours before the boundary layer broke up, when downslope drainage flows brought smoke to the lower elevations where the Mobile Lab was located.

K.B. Benedict et al. / Atmospheric Environment 148 (2017) 8e15

11

Fig. 1. Hourly averaged measurements of NOx, HTC-RN, PM2.5, and CO concentrations during the Hewlett Gulch and High Park fires are shown in panel a and b. In panel c the hourly  average measurements of NH3 are shown with the 24-hr integrated measurements of NH3 on the left axis and HNO3, NHþ 4 , NO3 and levoglucosan (a smoke tracer) are shown on the right axis. Ammonia concentrations are reported as mg/m3 to be consistent with the particle phase data. In panel d the AMoN bi-weekly ammonia concentrations are shown for 2012 and compared to ammonia concentrations at the same site during other years. Fire periods are designated in the gray shaded regions, based on the dates when the fires started and were subsequently designated fully contained.

3.1. NH3 and NOx During the smoke-impacted periods, hourly averaged concentrations of all gas phase nitrogen containing species (NH3, NOx, HTC-RN) increased (Fig. 1). Following previous work (Goode et al., 2000; Yokelson et al., 2009), we determined excess mixing ratios for NH3, NOx and HTC-RN relative to excess mixing ratios for CO for measurements during the fires. Excess mixing ratios were calculated by subtracting background concentrations collected during a non-fire impacted period (May 28eJune 4, 2012) from the smoke impacted periods. Average background concentrations were 154 ppb for CO, 5 ppb for NH3, 5 ppb for NOx, and 1.8 ppb for HTCRN. Smoke impacted periods were designated as those that had CO concentrations that were at least double the background measurements from May 28-June 4 (i.e., >308 ppb). In the following figures, an orthogonal distance regression with the intercept forced through zero was implemented to estimate emission ratios and determine the 95% confidence interval in the regression given the error in both the CO and nitrogen species measurement. Excess ammonia (DNH3) versus excess CO (DCO) is plotted in Fig. 2, resulting in DNH3/DCO ¼ 0.0274 ± 0.002 (ppb/ppb). The emission ratio is sensitive to the background value used so a range of emission ratios were calculated based on the mean ± 1 standard deviation to determine the upper and lower concentration limits observed during the background period. DNH3/DCO ranged from 0.025 to 0.030 for the maximum and minimum NH3 concentrations measured during the background period. This range suggests that

the derived emission ratio is likely more sensitive to the background concentration than to the uncertainty in the measurement. The emission ratio values observed here are in very close agreement with the value estimated from the 2011 NEI emissions of NH3

Fig. 2. Excess NH3 versus excess CO during smoke-impacted periods for both fires. The dashed lines represents the sensitivity (0.021e0.032) in the emission ratio to the background concentration used.

12

K.B. Benedict et al. / Atmospheric Environment 148 (2017) 8e15

and CO from wildfires (0.026), and falls within the range of previously measured values (e.g. Goode et al., 2000 and references therein; Stockwell et al., 2015). This enhancement ratio is expected to drop rapidly during the first 12 h of plume transport due to the short atmospheric lifetime of NH3 compared to CO (Coheur et al., 2009; Goode et al., 2000), as considerable NH3 will either be converted to particulate NHþ 4 or removed by deposition to terrain located downwind. Although the measurements are consistent with literature values, they fall on the higher end of reported measurements (Bertschi, 2003; Christian et al., 2007; Goode et al., 2000; Hobbs, 2003). Higher NH3 emissions may be tied to an enhanced nitrogen fuel content, as nitrogen content often controls nitrogen emissions (Andreae and Merlet, 2001; Lobert et al., 1990; Yokelson et al., 1996), or it may be tied to the combustion phase during the fire (Hegg et al., 1988). Ammonia, amines and nitriles are typically associated with smoldering combustion, while NOx and N2O are released primarily during flaming combustion (Andreae and Merlet, 2001; Goode et al., 2000; Lobert et al., 1990; Yokelson et al., 1996). Observed enhancements in NOx were significantly lower than observed for NH3, with DNOx/DCO ¼ 0.0057 ± 0.0007 (ppb/ppb), ranging from 0.004 to 0.007 depending on the background concentration used. The relative amounts of NH3 to NOx measured during these two fires suggest that the emissions were primarily due to smoldering combustion (Carrico et al., 2016; McMeeking et al., 2009), although there were certainly contributions from both flaming and smoldering combustion. Unfortunately there are no CO2 data available in these measurements of opportunity to calculate modified combustion efficiency, which is commonly used to separate observations of flaming and smoldering combustion, for the fire period. In Fig. 3 we consider the diurnal profile of the ratio of NH3 to NOx measured during the fires collected throughout the time period shown in Fig. 1. Note that these data are for the measured concentrations, rather than excess concentrations, to avoid the very high values obtained when DNOx is low. Further, the smoke plumes were detected at our site primarily in the evening and early morning hours, and there is not an equal distribution of data throughout the day. From the figure,

Fig. 3. Diurnal profile of mean values for the ratio of NH3/NOx (ppb/ppb) plotted versus the hour of day for fire impacted periods (red) and the background period (blue). Shaded regions indicate minimum and maximum hourly values. Areas without shaded regions have only 1 h of data. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

there is a clear pattern in the NH3 to NOx ratio during the fire periods, with higher values during the evening and early morning hours, and lower values during mid-day. This pattern is the opposite of what is observed during the non-fire periods, when this ratio peaks in the early afternoon hours. The large values for NH3:NOx in the evenings and early mornings during these fires suggest that the emissions are likely more impacted by smoldering combustion at these times, while during the hotter, drier hours of mid-day, flaming combustion is more prominent. Limited aircraft measurements collected during the High Park fire indicated a mixture of flaming and smoldering combustion for an air mass sampled at higher altitudes (Apel et al., 2015) while limited ground-based measurements suggested greater smoldering influence during the early morning hours relative to mid-day (Carrico et al., 2016). This is also consistent with typical biomass burning events, where the minimum in fire radiative energy is typically between 00:00 and 08:00 (Vermote et al., 2009) and tends to be the result of meteorological conditions at night favoring increased atmospheric stability, lower temperatures, and high relative humidity (Rothermel, 1983). Despite the large amounts of ammonia measured when the measurement site was impacted by smoke for shorter periods of time, longer time-integrated measurements (2 week) from passive ammonia sampling (site CO13) by the AMoN network (Puchalski et al., 2015) only show a factor of two increase for the most intense fire period (Fig. 1c), indicating the difficulty of using timeintegrated passive NH3 sampling to determine the influence of biomass burning on ammonia concentrations at a national level. The ammonia concentrations from MayeJuly are very similar from 2009 to 2014 with the exception of 2012 when the site was impacted by the High Park fire. No obvious increases in NH3 are observed in the AMoN Fort Collins data from the period of the Hewlett Gulch fire. This issue is exacerbated locally by the fact that NH3 has a variety of strong regional sources, most notably from agriculture. 3.2. Nitric acid, ammonium and nitrate Additional measurements were made at 24-hr time resolution to quantify ammonia, nitric acid, particulate ion concentrations and levoglucosan (during the High Park fire only) which is used as a smoke marker (Sullivan et al., 2008). These data are consistent with the real-time measurements, with marked increases in inorganic nitrogen and levoglucosan during fire-impacted periods (Fig. 1b). For PM2.5 nitrate, enhancements in concentrations were 3e5 times greater than background concentrations, reaching a maximum concentration of 1.5 mg/m3. Gaseous ammonia (from the denuders) and nitric acid increased 2e3 times above background concentrations with maximum concentrations of 18 and 4.7 mg/m3, respectively. PM2.5 ammonium showed almost no increase above background during smoke periods. This is consistent with observations from FLAME when little to no increase in NHþ 4 concentration was observed during open combustion from laboratory when fresh smoke was measured (McMeeking et al., 2009). Sufficient ammonia is generally already present in the regional atmosphere to fully neutralize fine particle sulfate and the warmer temperatures typical of late spring/summer are not conducive to formation of fine particle ammonium nitrate even downwind of the fire. To examine composition of particle nitrogen, the ions measured from the filter were paired according to aerosol thermodynamics to look at the neutralized salts that were likely to be present in the  2 þ aerosol. In this case we considered only NHþ 4 , NO3 , SO4 , and K . Although nitrate can also form salts through reaction with sea salt and soil dust (Lee et al., 2008), we did not consider Naþ due to the absence of local sea salt sources while Ca2þ was omitted because it

K.B. Benedict et al. / Atmospheric Environment 148 (2017) 8e15

is generally associated with coarse particle soil dust while our measurements here were of PM2.5. We assumed that NHþ 4 was associated with SO2 4 until it was fully neutralized; any remaining  unbound ammonium was then paired with NO 3 . If any NO3 remained Kþ was paired to form KNO3. The particle chemistry during the most intense smoke periods suggest the nitrogen in PM2.5 is found as KNO3, (NH4)2SO4, and NH4NO3 (in order of abundance). Similar results have been found in other studies that have examined the particle anion-cation pairs in biomass burning plumes (Ma et al., 2003; Song et al., 2005). PM2.5 levoglucosan concentrations increased ~40 times compared to non-smoke-influenced periods. Ratios of levoglucosan to organic carbon are used in modeling and measurements to attribute smoke emissions to different types of fuels (Munchak et al., 2011; Sullivan et al., 2011, 2008). During periods when levoglucosan concentrations were greater than 2 mg/m3 the levoglucosan to organic carbon (mg/mg C) ratio ranged from 0.083 to 0.131 which is consistent with the fuels burned and laboratory measurements of similar fuels (Sullivan et al., 2008).

13

Fig. 5. Measurements using two ammonia instruments at Rocky Mountain National Park in 2010. Teledyne ‘NH3’ is the uncorrected data from the Teledyne chemiluminescence instrument. The difference between that measurement and the Air Sentry ammonia measurement is HTC-RN.

3.3. High temperature conversion reactive nitrogen Measurements of additional unspeciated nitrogen compounds using a high temperature converter showed significant concentrations of HTC-RN during the Hewlett Gulch and High Park fires, as shown in Figs. 1 and 4. These measurements indicate concentrations that exceed those of NOx in the smoke plumes, with ammonia present at even higher concentrations. Apel et al. (2015) reported aircraft measurements of select organic nitrogen compounds for one flight during the High Park fire in which biomass burning emissions were sampled. Hydrogen cyanide (HCN) and acetonitrile excess mixing ratios of 7.0 ± 0.5 and 2.3 ± 0.5 pptv ppbv1, relative to CO, were observed in these measurements, which could account for approximately half of the HTC-RN that is measured in the current study, assuming ~70% conversion of these species in the Teledyne instrument (discussed above). Although the HTC-RN measurement does not speciate all of the compounds that were present in the fire plume, they do indicate that a significant amount of gas phase reactive nitrogen other than NH3 and NOx was produced during the two fires. We have also observed HTC-RN in emissions from other fires.

During the Rocky Mountain Atmospheric Nitrogen and Sulfur studies (Beem et al., 2010; Benedict et al., 2013a, 2013b), smoke from the Fourmile Canyon Fire near Boulder, CO reached the sampling site in Rocky Mountain National Park (ROMO) on Sept 6e8, 2010, as shown in Fig. 5. The presence of smoke was indicated by large increases in CO concentrations and increases in the scattering coefficient (bsp) measured by an Optec NGN-2 nephelometer. The NH3 Teledyne 201E chemiluminescence instrument with a 501NH converter was operated to measure additional reactive nitrogen without removing ammonia. Ammonia was separately measured with an Air Sentry II Ion Mobility Spectrometer. During the smoke episodes, both instruments show peaks, with the Teledyne instrument reaching values nearly six times greater than the Air Sentry II. The strong response by the Teledyne instrument indicates measurement of abundant compounds other than ammonia and amines in the smoke plume. Although the specific compounds measured were not identified, possible nitrogen containing compounds include acetonitrile and hydrogen cyanide (Yokelson et al., 1996), as well as nitriles (Andreae and Merlet, 2001; Yokelson et al., 1996), methyl nitrate (Hobbs, 2003) and isocyanic acid (Roberts et al., 2011). In the Grand Teton Reactive Nitrogen Deposition Study (GrandTReNDS) in 2011, we again used a range of ammonia and reactive nitrogen measurement techniques (Prenni et al., 2014). In that study, the Teledyne chemiluminescence instrument was operated in the same manner as the current study, giving a direct measure of HTC-RN. Biomass burning episodes provided the only occurrences of HTC-RN during the GrandTReNDS campaign, coincident with elevated levels of NH3 from the more selective Air Sentry II ion mobility and Picarro cavity ringdown instruments (Prenni et al., 2014). 4. Conclusions

Fig. 4. Excess HTC-RN versus excess CO during smoke-impacted periods for both fires. The dashed lines represents the sensitivity (0.011e0.016) in the emission ratio to the background concentration used.

The Hewlett Gulch and High Park fires presented a unique opportunity to measure reactive nitrogen concentrations in fresh smoke plumes from two different fires over several weeks, with approximately 125 h of measurements in the smoke plumes. Enhancements in nitrogen-containing species were observed for nearly all species during these fires with the exception of fine particle ammonium. Gaseous ammonia during smoke-impacted periods was enhanced by a factor of 20 at hourly time resolution. For these fires DNH3/DCO was determined to be 0.027 ± 0.002 (ppb/

14

K.B. Benedict et al. / Atmospheric Environment 148 (2017) 8e15

ppb) in smoke plumes, which is consistent with previously reported values. DNOx/DCO was lower at 0.0057 ± 0.0007 (ppb/ppb). Combined, these observations indicate that the observed smoke was likely from predominantly smoldering combustion. The NH3 to NOx ratio changed over the course of a day; on average, lower ratios were observed during the day compared to the night, consistent with expected changes in fire behavior related to diurnal temperature and relative humidity shifts. Nitrate concentrations increased by 3e5 times in smoke plumes, while nitric acid increased by 2e3 times compared to non-smoke periods. Concentrations of HTC-RN were also elevated during both fires. During the High Park and Hewlett Gulch fires concentrations of ammonia were greater than HTC-RN which, in turn, exceeded NOx concentrations, while observations during a fire observed in Rocky Mountain National Park in 2010 indicated that HTC-RN concentrations can sometimes exceed ammonia (Fig. 5). Laboratory testing of the chemiluminescence instrument's sensitivity to VOC and nitrogencontaining compounds indicate significant sensitivity to acetonitrile, which is expected to be abundant in smoke plumes. There is a plethora of evidence to indicate reactive nitrogen emissions from biomass burning can be significant. The results presented here focus on a case study that highlights the different components of the emitted reactive nitrogen and their relative contribution. However, more research is needed to address the impact of reactive nitrogen on ecosystem health downwind of fires, and the relevance to atmospheric chemistry and air quality of the compounds that comprise HTC-RN. Acknowledgment The authors would like to thank Arsineh Hecobian and Xi (Doris) Chen, for their help collecting and analyzing the denuder and filter samples. Support for this work was provided by the National Park Service. The assumptions, findings, conclusions, judgments, and views presented herein are those of the authors and should not be interpreted as necessarily representing the National Park Service. References Akagi, S.K., Yokelson, R.J., Wiedinmyer, C., Alvarado, M.J., Reid, J.S., Karl, T., Crounse, J.D., Wennberg, P.O., 2011. Emission factors for open and domestic biomass burning for use in atmospheric models. Atmos. Chem. Phys. 11, 4039e4072. http://dx.doi.org/10.5194/acp-11-4039-2011. Andreae, M.O., Browell, E.V., Garstang, M., Gregory, G.L., Harriss, R.C., Hill, G.F., Jacob, D.J., Pereira, M.C., Sachse, G.W., Setzer, A.W., Dias, P.L.S., Talbot, R.W., Torres, A.L., Wofsy, S.C., 1988. Biomass-burning emissions and associated haze layers over Amazonia. J. Geophys. Res. Atmos. 93, 1509e1527. http://dx.doi.org/ 10.1029/JD093iD02p01509. Andreae, M.O., Merlet, P., 2001. Emission of trace gases and aerosols from biomass burning. Glob. Biogeochem. Cycles 15, 955e966. http://dx.doi.org/10.1029/ 2000GB001382. Apel, E.C., Hornbrook, R.S., Hills, A.J., Blake, N.J., Barth, M.C., Weinheimer, A., Cantrell, C., Rutledge, S.A., Basarab, B., Crawford, J., Diskin, G., Homeyer, C.R., Campos, T., Flocke, F., Fried, A., Blake, D.R., Brune, W., Pollack, I., Peischl, J., Ryerson, T., Wennberg, P.O., Crounse, J.D., Wisthaler, A., Mikoviny, T., Huey, G., Heikes, B., O'Sullivan, D., Riemer, D.D., 2015. Upper tropospheric ozone production from lightning NO x -impacted convection: smoke ingestion case study from the DC3 campaign. J. Geophys. Res. Atmos. 120, 2505e2523. http:// dx.doi.org/10.1002/2014JD022121. Asman, W.A.H., Sutton, M.A., Schjørring, J.K., 1998. Ammonia: emission, atmospheric transport and deposition. New Phytol. 139, 27e48. http://dx.doi.org/ 10.1046/j.1469-8137.1998.00180.x. Baron, J.S., Rueth, H.M., Wolfe, A.M., Nydick, K.R., Allstott, E.J., Minear, J.T., Moraska, B., 2000. Ecosystem responses to nitrogen deposition in the Colorado front range. Ecosystems 3, 352e368. http://dx.doi.org/10.1007/s100210000032. Beem, K.B., Raja, S., Schwandner, F.M., Taylor, C., Lee, T., Sullivan, A.P., Carrico, C.M., McMeeking, G.R., Day, D., Levin, E., Hand, J., Kreidenweis, S.M., Schichtel, B., Malm, W.C., Collett, J.L., 2010. Deposition of reactive nitrogen during the Rocky mountain airborne nitrogen and Sulfur (RoMANS) study. Environ. Pollut. 158, 862e872. http://dx.doi.org/10.1016/j.envpol.2009.09.023. Benedict, K.B., Carrico, C.M., Kreidenweis, S.M., Schichtel, B., Malm, W.C., Collett, J.L., 2013a. A seasonal nitrogen deposition budget for Rocky Mountain National Park. Ecol. Appl. 23, 1156e1169. http://dx.doi.org/10.1890/12-1624.1.

Benedict, K.B., Day, D., Schwandner, F.M., Kreidenweis, S.M., Schichtel, B., Malm, W.C., Collett, J.L., 2013b. Observations of atmospheric reactive nitrogen species in Rocky Mountain National Park and across northern Colorado. Atmos. Environ. 64, 66e76. http://dx.doi.org/10.1016/j.atmosenv.2012.08.066. Bertschi, I.T., 2003. Trace gas emissions from the production and use of domestic biofuels in Zambia measured by open-path Fourier transform infrared spectroscopy. J. Geophys. Res. 108 http://dx.doi.org/10.1029/2002JD002158. Bouwman, A.F., Lee, D.S., Asman, W.A.H., Dentener, F.J., VanderHoek, K.W., Olivier, J.G.J., 1997. A global high-resolution emission inventory for ammonia. Glob. Biogeochem. Cycles 11, 561e587. http://dx.doi.org/10.1029/97GB02266. Carrico, C.M., Prenni, A.J., Kreidenweis, S.M., Levin, E.J.T., McCluskey, C.S., DeMott, P.J., McMeeking, G.R., Nakao, S., Stockwell, C., Yokelson, R.J., 2016. Rapidly evolving ultrafine and fine mode biomass smoke physical properties: comparing laboratory and field results. J. Geophys. Res. Atmos. 121 http:// dx.doi.org/10.1002/2015JD024389. Chen, X., Day, D., Schichtel, B., Malm, W., Matzoll, A.K., Mojica, J., McDade, C.E., Hardison, E.D., Hardison, D.L., Walters, S., Van De Water, M., Collett, J.L., 2014. Seasonal ambient ammonia and ammonium concentrations in a pilot IMPROVE NHx monitoring network in the western United States. Atmos. Environ. 91, 118e126. http://dx.doi.org/10.1016/j.atmosenv.2014.03.058. Christian, T.J., Yokelson, R.J., Carvalho, J.A., Griffith, D.W.T., Alvarado, E.C., Santos, J.C., Neto, T.G.S., Veras, C.A.G., Hao, W.M., 2007. The tropical forest and fire emissions experiment: trace gases emitted by smoldering logs and dung from deforestation and pasture fires in Brazil. J. Geophys. Res. 112 http://dx.doi.org/10.1029/ 2006JD008147. Coheur, P.-F., Clarisse, L., Turquety, S., Hurtmans, D., Clerbaux, C., 2009. IASI measurements of reactive trace species in biomass burning plumes. Atmos. Chem. Phys. 9, 5655e5667. http://dx.doi.org/10.5194/acp-9-5655-2009. Dunlea, E.J., Herndon, S.C., Nelson, D.D., Volkamer, R.M., San Martini, F., Sheehy, P.M., Zahniser, M.S., Shorter, J.H., Wormhoudt, J.C., Lamb, B.K., Allwine, E.J., Gaffney, J.S., Marley, N.A., Grutter, M., Marquez, C., Blanco, S., Cardenas, B., Retama, A., Ramos Villegas, C.R., Kolb, C.E., Molina, L.T., Molina, M.J., 2007. Evaluation of nitrogen dioxide chemiluminescence monitors in a polluted urban environment. Atmos. Chem. Phys. 7, 2691e2704. Flannigan, M.D., Stocks, B.J., Wotton, B.M., 2000. Climate change and forest fires. Sci. Total Environ. 262, 221e229. http://dx.doi.org/10.1016/S0048-9697(00)005246. Ge, C., Wang, J., Reid, J.S., 2014. Mesoscale modeling of smoke transport over the Southeast Asian Maritime Continent: coupling of smoke direct radiative effect below and above the low-level clouds. Atmos. Chem. Phys. 14, 159e174. http:// dx.doi.org/10.5194/acp-14-159-2014. Giglio, L., Randerson, J.T., van der Werf, G., Kasibhatla, P.S., Collatz, G.J., Morton, D.C., DeFries, R.S., 2010. Assessing variability and long-term trends in burned area by merging multiple satellite fire products. Biogeosciences 7, 1171e1186. Goode, J.G., Yokelson, R.J., Susott, R.A., Ward, D.E., 1999. Trace gas emissions from laboratory biomass fires measured by open-path Fourier transform infrared spectroscopy: fires in grass and surface fuels. J. Geophys. Res. Atmos. 104, 21237e21245. http://dx.doi.org/10.1029/1999JD900360. Goode, J.G., Yokelson, R.J., Ward, D.E., Susott, R.A., Babbitt, R.E., Davies, M.A., Hao, W.M., 2000. Measurements of excess O-3, CO2, CO, CH4, C2H4, C2H2, HCN, NO, NH3, HCOOH, CH3COOH, HCHO, and CH3OH in 1997 Alaskan biomass burning plumes by airborne fourier transform infrared spectroscopy (AFTIR). J. Geophys. Res. Atmos. 105, 22147e22166. http://dx.doi.org/10.1029/ 2000JD900287. Hegg, D.A., Radke, L.F., Hobbs, P.V., Riggan, P.J., 1988. Ammonia emissions from biomass burning. Geophys. Res. Lett. 15, 335e337. Hobbs, P.V., 2003. Evolution of gases and particles from a savanna fire in South Africa. J. Geophys. Res. 108 http://dx.doi.org/10.1029/2002JD002352. Kallinger, G., Niessner, R., 1999. Laboratory investigation of annular denuders as sampling system for the determination of aliphatic primary and secondary amines in stack gas. Microchim. Acta 130, 309e316. Kop a cek, J., Posch, M., 2011. Anthropogenic nitrogen emissions during the Holocene and their possible effects on remote ecosystems. Glob. Biogeochem. Cycles 25. http://dx.doi.org/10.1029/2010GB003779. Lee, T., Yu, X., Ayres, B., Kreidenweis, S., Malm, W., Collettjr, J., 2008. Observations of fine and coarse particle nitrate at several rural locations in the United States. Atmos. Environ. 42, 2720e2732. http://dx.doi.org/10.1016/ j.atmosenv.2007.05.016. Levin, E.J.T., McMeeking, G.R., Carrico, C.M., Mack, L.E., Kreidenweis, S.M., Wold, C.E., Moosmüller, H., Arnott, W.P., Hao, W.M., Collett, J.L., Malm, W.C., 2010. Biomass burning smoke aerosol properties measured during Fire Laboratory at Missoula Experiments (FLAME). J. Geophys. Res. 115 http://dx.doi.org/10.1029/ 2009JD013601. Lobert, J., Scharffe, D., Hao, W., Crutzen, P., 1990. Importance of biomass burning in the atmospheric budgets of nitrogen-containing gases. Nature 346, 552e554. http://dx.doi.org/10.1038/346552a0. Ma, Y., Weber, R.J., Lee, Y.-N., Orsini, D.A., Maxwell-Meier, K., Thornton, D.C., Bandy, A.R., Clarke, A.D., Blake, D.R., Sachse, G.W., Fuelberg, H.E., Kiley, C.M., Woo, J.-H., Streets, D.G., Carmichael, G.R., 2003. Characteristics and influence of biosmoke on the fine-particle ionic composition measured in Asian outflow during the Transport and Chemical Evolution over the Pacific (TRACE-P) experiment. J. Geophys. Res. Atmos. 108, 8816. http://dx.doi.org/10.1029/ 2002JD003128. McMeeking, G.R., Kreidenweis, S.M., Baker, S., Carrico, C.M., Chow, J.C., Collett, J.L., Hao, W.M., Holden, A.S., Kirchstetter, T.W., Malm, W.C., Moosmüller, H.,

K.B. Benedict et al. / Atmospheric Environment 148 (2017) 8e15 Sullivan, A.P., Wold, C.E., 2009. Emissions of trace gases and aerosols during the open combustion of biomass in the laboratory. J. Geophys. Res. 114 http:// dx.doi.org/10.1029/2009JD011836. Moritz, M.A., Parisien, M.-A., Batllori, E., Krawchuk, M.A., Van Dorn, J., Ganz, D.J., Hayhoe, K., 2012. Climate change and disruptions to global fire activity. Ecosphere 3, 1e22. http://dx.doi.org/10.1890/ES11-00345.1. Munchak, L.A., Schichtel, B.A., Sullivan, A.P., Holden, A.S., Kreidenweis, S.M., Malm, W.C., Collett, J.L., 2011. Development of wildland fire particulate smoke marker to organic carbon emission ratios for the conterminous United States. Atmos. Environ. 45, 395e403. http://dx.doi.org/10.1016/j.atmosenv.2010.10.006. Nance, J.D., Hobbs, P.V., Radke, L.F., Ward, D.E., 1993. Airborne measurements of gases and particles from an Alaskan wildfire. J. Geophys. Res. Atmos. 98, 14873e14882. http://dx.doi.org/10.1029/93JD01196. Park, R.J., Jacob, D.J., Chin, M., Martin, R.V., 2003. Sources of carbonaceous aerosols over the United States and implications for natural visibility. J. Geophys. Res. Atmos. 108, 4355. http://dx.doi.org/10.1029/2002JD003190. Parrish, D., Hahn, C., Fahey, D., Williams, E., Bollinger, M., Hubler, G., Buhr, M., Murphy, P., Trainer, M., Hsie, E., Liu, S., Fehsenfeld, F., 1990. Systematic variations in the concentration of NOx (NO plus NO2) at Niwot-Ridge, Colorado. J. Geophys. Res. Atmos. 95, 1817e1836. http://dx.doi.org/10.1029/ JD095iD02p01817. Prenni, A.J., Levin, E.J.T., Benedict, K.B., Sullivan, A.P., Schurman, M.I., Gebhart, K.A., Day, D.E., Carrico, C.M., Malm, W.C., Schichtel, B.A., Collett, J.L., Kreidenweis, S.M., 2014. Gas-phase reactive nitrogen near Grand Teton National Park: impacts of transport, anthropogenic emissions, and biomass burning. Atmos. Environ. 89, 749e756. http://dx.doi.org/10.1016/j.atmosenv.2014.03.017. Price, C., Rind, D., 1994. The impact of a 2  CO2 climate on lightning-caused fires. J. Clim. 7, 1484e1494. Puchalski, M.A., Rogers, C.M., Baumgardner, R., Mishoe, K.P., Price, G., Smith, M.J., Watkins, N., Lehmann, C.M., 2015. A statistical comparison of active and passive ammonia measurements collected at Clean Air Status and Trends Network (CASTNET) sites. Env. Sci. Process. Impacts 17, 358e369. http://dx.doi.org/ 10.1039/C4EM00531G. Reis, S., Pinder, R.W., Zhang, M., Lijie, G., Sutton, M.A., 2009. Reactive nitrogen in atmospheric emission inventories. Atmos. Chem. Phys. 9, 7657e7677. http:// dx.doi.org/10.5194/acp-9-7657-2009. R'Honi, Y., Clarisse, L., Clerbaux, C., Hurtmans, D., Duflot, V., Turquety, S., Ngadi, Y., Coheur, P.-F., 2013. Exceptional emissions of NH3 and HCOOH in the 2010 Russian wildfires. Atmos. Chem. Phys. 13, 4171e4181. http://dx.doi.org/10.5194/ acp-13-4171-2013. Richardson, T., 2012. Summary of Management Activities: Hewlett Fire [WWW Document]. http://gacc.nifc.gov/rmcc/dispatch_centers/r2ftc/documents/ Hewlett%20Summary%20of%20Management%20Activities.pdf (Accessed 5 June 2016). Roberts, J.M., Veres, P.R., Cochran, A.K., Warneke, C., Burling, I.R., Yokelson, R.J., Lerner, B., Gilman, J.B., Kuster, W.C., Fall, R., de Gouw, J., 2011. Isocyanic acid in the atmosphere and its possible link to smoke-related health effects. Proc. Natl. Acad. Sci. U. S. A. 108, 8966e8971. http://dx.doi.org/10.1073/pnas.1103352108. Rothermel, R.C., 1983. How to Predict the Spread and Intensity of Forest and Range Fires. United States Department of Agriculture Forest Service, Intermountain Forest and Range Experiment Station, Ogden, UT, p. 84401. Song, C.H., Ma, Y., Orsini, D., Kim, Y.P., Weber, R.J., 2005. An investigation into the ionic chemical composition and mixing state of biomass burning particles recorded during TRACE-P P3B Flight#10. J. Atmos. Chem. 51, 43e64. http:// dx.doi.org/10.1007/s10874-005-5727-9. Spracklen, D.V., Logan, J.A., Mickley, L.J., Park, R.J., Yevich, R., Westerling, A.L., Jaffe, D.A., 2007. Wildfires drive interannual variability of organic carbon aerosol in the western U.S. in summer. Geophys. Res. Lett. 34, L16816. http:// dx.doi.org/10.1029/2007GL030037. Stockwell, C.E., Veres, P.R., Williams, J., Yokelson, R.J., 2015. Characterization of biomass burning emissions from cooking fires, peat, crop residue, and other fuels with high-resolution proton-transfer-reaction time-of-flight mass spectrometry. Atmos. Chem. Phys. 15, 845e865. http://dx.doi.org/10.5194/acp-15845-2015. Sullivan, A.P., Frank, N., Onstad, G., Simpson, C.D., Collett, J.L., 2011. Application of high-performance anion-exchange chromatographyepulsed amperometric

15

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 http://dx.doi.org/10.1029/2010JD014166. Sullivan, A.P., Holden, A.S., Patterson, L.A., McMeeking, G.R., Kreidenweis, S.M., Malm, W.C., Hao, W.M., Wold, C.E., Collett, J.L., 2008. A method for smoke marker measurements and its potential application for determining the contribution of biomass burning from wildfires and prescribed fires to ambient PM 2.5 organic carbon. J. Geophys. Res. 113 http://dx.doi.org/10.1029/ 2008JD010216. Targino, A.C., Krecl, P., Johansson, C., Swietlicki, E., Massing, A., Coraiola, G.C., Lihavainen, H., 2013. Deterioration of air quality across Sweden due to transboundary agricultural burning emissions. Boreal Env. Res. 18, 19e36. Val Martin, M., Heald, C.L., Ford, B., Prenni, A.J., Wiedinmyer, C., 2013. A decadal satellite analysis of the origins and impacts of smoke in Colorado. Atmos. Chem. Phys. 13, 7429e7439. http://dx.doi.org/10.5194/acp-13-7429-2013. Van Damme, M., Clarisse, L., Heald, C.L., Hurtmans, D., Ngadi, Y., Clerbaux, C., Dolman, A.J., Erisman, J.W., Coheur, P.F., 2014. Global distributions, time series and error characterization of atmospheric ammonia (NH3) from IASI satellite observations. Atmos. Chem. Phys. 14, 2905e2922. http://dx.doi.org/10.5194/ acp-14-2905-2014. van der Werf, G.R., Randerson, J.T., Collatz, G.J., Kasibhatla, P.S., Arellano, A.F., 2006. Interannual variability in global biomass burning emissions from 1997 to 2004. Atmos. Chem. Phys. 6, 3423e3441. Vermote, E., Ellicott, E., Dubovik, O., Lapyonok, T., Chin, M., Giglio, L., Roberts, G.J., 2009. An approach to estimate global biomass burning emissions of organic and black carbon from MODIS fire radiative power. J. Geophys. Res. Atmos. 114, D18205. http://dx.doi.org/10.1029/2008JD011188. Walker, P.B., Williams, T., Mohr, A., La Rue, D., 2012. High Park Fire Burned Area Emergency Response (BAER) Report [WWW Document]. http://www.larimer. org/highparkfire/bear_report.pdf (Accessed 15 February 2016). Westerling, A.L., Hidalgo, H.G., Cayan, D.R., Swetnam, T.W., 2006. Warming and earlier spring increase Western U.S. Forest wildfire activity. Science 313, 940e943. http://dx.doi.org/10.1126/science.1128834. Westerling, A.L., Turner, M.G., Smithwick, E.A.H., Romme, W.H., Ryan, M.G., 2011. Continued warming could transform Greater Yellowstone fire regimes by mid21st century. Proc. Natl. Acad. Sci. U. S. A. 108, 13165e13170. http://dx.doi.org/ 10.1073/pnas.1110199108. Whitburn, S., Van Damme, M., Kaiser, J.W., van der Werf, G.R., Turquety, S., Hurtmans, D., Clarisse, L., Clerbaux, C., Coheur, P.-F., 2015. Ammonia emissions in tropical biomass burning regions: comparison between satellite-derived emissions and bottom-up fire inventories. Atmos. Environ. 121, 42e54. http:// dx.doi.org/10.1016/j.atmosenv.2015.03.015. Wiedinmyer, C., Akagi, S., Yokelson, R., Emmons, L., Al-Saadi, J., Orlando, J., Soja, A., 2011. The fire INventory from NCAR (FINN): a high resolution global model to estimate the emissions from open burning. Geosci. Model Dev. 625e641. Wiedinmyer, C., Quayle, B., Geron, C., Belote, A., McKenzie, D., Zhang, X., O'Neill, S., Wynne, K.K., 2006. Estimating emissions from fires in North America for air quality modeling. Atmos. Environ. 40, 3419e3432. http://dx.doi.org/10.1016/ j.atmosenv.2006.02.010. Yokelson, R.J., Christian, T.J., Karl, T.G., Guenther, A., 2008. The tropical forest and fire emissions experiment: laboratory fire measurements and synthesis of campaign data. Atmos. Chem. Phys. 8, 3509e3527. http://dx.doi.org/10.5194/ acp-8-3509-2008. Yokelson, R.J., Crounse, J.D., DeCarlo, P.F., Karl, T., Urbanski, S., Atlas, E., Campos, T., Shinozuka, Y., Kapustin, V., Clarke, A.D., Weinheimer, A., Knapp, D.J., Montzka, D.D., Holloway, J., Weibring, P., Flocke, F., Zheng, W., Toohey, D., Wennberg, P.O., Wiedinmyer, C., Mauldin, L., Fried, A., Richter, D., Walega, J., Jimenez, J.L., Adachi, K., Buseck, P.R., Hall, S.R., Shetter, R., 2009. Emissions from biomass burning in the Yucatan. Atmos. Chem. Phys. 9, 5785e5812. http:// dx.doi.org/10.5194/acp-9-5785-2009. Yokelson, R.J., Griffith, D.W.T., Ward, D.E., 1996. Open-path Fourier transform infrared studies of large-scale laboratory biomass fires. J. Geophys. Res. 101, 21067. http://dx.doi.org/10.1029/96JD01800. Yu, X., Lee, T., Ayres, B., Kreidenweis, S., Malm, W., Collettjr, J., 2006. Loss of fine particle ammonium from denuded nylon filters. Atmos. Environ. 40, 4797e4807. http://dx.doi.org/10.1016/j.atmosenv.2006.03.061.