Particle size characteristics of levoglucosan in ambient ...
Atmospheric Environment 42 (2008) 8300–8308
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Particle size characteristics of levoglucosan in ambient aerosols from rice straw burning James J. Lee a, *, Guenter Engling b, Shih-Chun Candice Lung b, Kuo-Yang Lee a a b
Department of Environmental Engineering and Safety, National Yunlin University of Science and Technology, Douliou, Taiwan Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 March 2008 Received in revised form 15 July 2008 Accepted 21 July 2008
Agricultural residue burning in Asia is an important source of atmospheric aerosol particles. This study investigates the impact of post-harvest burning of rice fields on the air quality in Taiwan, a subtropical island in East Asia. Size-resolved measurements of the anhydrosugar levoglucosan (biomass burning tracer derived from the thermal breakdown of cellulose) were conducted before, during and after an episode of intensive rice straw burning at a rural and suburban site. While substantially enhanced levoglucosan concentrations were observed during the active rice straw burning episode (up to 1400 ng m3), fairly high values of the smoke tracer were measured throughout the entire study period. Moreover, unusually high levoglucosan levels were present in aerosol particles with aerodynamic diameters larger than 10 mm (PM>10), possibly influenced by the ambient atmospheric conditions, such as high relative humidity, in addition to unique properties of rice straw smoke and the specific burning practices of rice fields. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Agricultural biomass Anhydrosugars Atmospheric aerosols Particle size distributions HPAEC-PAD
1. Introduction Atmospheric aerosol particles have substantial influence on regional air quality, global geochemical cycles, and the radiation budget of the earth. While the combustion of biomass contributes significant amounts of aerosols to the atmosphere on a global scale, the Asian continent is a particularly important source region for biomass smoke (Streets et al., 2003). The dominant constituents of most types of biomass are the polysaccharides cellulose and hemicellulose. The anhydrosugar levoglucosan, derived from the thermal degradation of cellulose (Hornig et al., 1985; Simoneit et al., 1999), is the major molecular tracer for biomass burning activities and has been utilized in several recent source apportionment studies (Leithead
* Corresponding author. Department of Environmental Engineering and Safety, National Yunlin University of Science and Technology, 123, University Rd. Section 3, Douliou, Taiwan. Tel.: þ886 5 534 2601; fax: þ886 5 531 2069. E-mail address: [email protected] (J.J. Lee). 1352-2310/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2008.07.047
et al., 2006; Puxbaum et al., 2007; Rinehart et al., 2006; Ward et al., 2006b). Biomass burning practices in Asia include a large variety of burning activities, such as domestic biofuel use (for cooking and heating), post-harvest burning of agricultural fields, and wildfires (Venkataraman et al., 2005). Several anhydrosugar investigations in North and South America as well as Europe have focused on smoke emissions from wood combustion, providing an extensive range of chemical and physical characteristics of wood smoke aerosol. Two major types of wood burning activities can be distinguished: open burning of trees and shrubs in forests (either in form of wildfires or prescribed burns) and combustion of wood under confined conditions, such as in domestic wood stoves and fireplaces. Smoke from forest fires has been investigated by various research groups (Engling et al., 2006b; Otto et al., 2006; Ward et al., 2006a; Zdrahal et al., 2002). Smoke emissions from residential wood burning have been characterized in several studies as well (Gorin et al., 2006; Leithead et al., 2006; Schmidl et al., 2008; Ward et al., 2006b; Yttri et al., 2005).
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
Conversely, fewer studies have investigated smoke emissions from agricultural burning activities. The most common type of agricultural burning, particularly in Asia, is post-harvest combustion of crop residues (Yevich and Logan, 2003). Considering the large population in Asian countries, the production of agricultural commodities and the associated residue burning is expected to result in considerable aerosol particle emissions, likely comparable to that of occasional wildland fires in North America and other parts of the world. Previous studies of anhydrosugars in smoke from agricultural residue burning have been conducted as controlled combustion experiments, including controlled field burning and laboratory (chamber) burns (Dhammapala et al., 2007; Hays et al., 2005; Mazzoleni et al., 2007; Sheesley et al., 2003; Zhang et al., 2007), as well as ambient measurements (Jimenez et al., 2006; Wang et al., 2007). In many cases, the measured ambient smoke originated from the combustion of mixed agricultural residues (He et al., 2006; Park et al., 2006; Wan and Yu, 2007), while a few ambient studies have been carried out on specific crop species, e.g., wheat straw and grass (Dhammapala et al., 2006). Smoke measurements from laboratory-scale burns of agricultural residues were typically conducted on individual crop types, yet only on a small number of species, particularly wheat and rice straw (Dhammapala et al., 2007; Gullett and Touati, 2003; Li et al., 2007). Investigations of smoke emissions from burning of rice straw are limited to several laboratory experiments and a few ambient observations. Various chemical compounds were measured during the chamber combustion of rice straw in some studies (Hays et al., 2005; Mazzoleni et al., 2007; Sheesley et al., 2003), whereas other laboratoryscale investigations focused on selected chemical species contained in rice straw smoke, such as dioxins (Gullett and Touati, 2003), nitrous oxide (Ogawa and Yoshida, 2005), and PAHs (Keshtkar and Ashbaugh, 2007). Ambient observations of rice straw smoke are limited to the measurement of PAHs (Yang et al., 2006), trace gases (Yoshinori and Kanno, 1997), and one recent study of anhydrosugars in ambient PM10 (Viana et al., 2008). No size-resolved measurements of anhydrosugar source tracers from post-harvest burning of rice straw in ambient air are available to date. Biomass burning contributions to the ambient air in Taiwan are predominantly from burning of rice fields following the harvest in each of the two growing seasons (in early summer and early winter). The winter burning season in Taiwan typically occurs in the time frame from late November to early January, coinciding with the dry season in central and southern Taiwan. Yunlin County is one of the major rice production areas in Taiwan; three neighboring counties in South-Central Taiwan (Changhua, Yunlin, and Chiayi counties) contribute 53% of the total rice production in Taiwan and the contributions from Yunlin County in particular are 19%. The ambient conditions and the geographical setting of Yunlin County are most favorable for the investigation of smoke emissions from rice straw burning during the winter season. Rice straw dries rapidly during this time and the burning is not hindered by
8301
wet precipitation, as is the case in the summer harvest/ burning season. As more than 90% of the global rice production occurs in Asia (Kadam et al., 2000), it is obvious that smoke generated from post-harvest burning of rice straw in Taiwan, as well as in the other Asian countries, contributes a sizeable portion of particulate pollution on local and regional scale. Therefore, the goal of this study is to characterize the smoke aerosol from rice field burning activities in Taiwan and its impact on ambient aerosol concentrations, particularly by investigating the size-resolved distributions of the anhydrosugar levoglucosan as molecular source tracer. 2. Experimental 2.1. Sample collection Aerosol samples were collected in Yunlin County, located in South-Central Taiwan (Fig. 1), at two sites during the winter rice field burning season in December 2006, including a short episode of intensive burning as well as periods prior to and following the episode. The first location was termed ‘‘suburban site’’ and was situated on the roof of a four-story building on the campus of the National Yunlin University of Science & Technology, located in the southern part of Douliou, Taiwan (23 420 9100 N, 120 340 17.900 E). A farming area with a large number of rice fields was located within 7–20 km to the north of this sampling site. The second sampling site was located in a rural area of Yunlin County, south of Douliou, and thus it was designated as ‘‘countryside’’ location. Samples were collected at this site on the roof of a two-story building of the Jiouan Elementary School. This site was surrounded by cultivated (agricultural) land, mainly used for rice farming. However, the number of rice fields in the surrounding area of this site was smaller than in the area to the north of the suburban site. Consequently, the suburban site was located directly downwind of the major rice production area within the study region, as the predominant wind direction during the study period was northerly (Table 1). The burning practices of rice straw in Taiwan are unique, as thin layers of straw are typically laid on the ground in an attempt to exterminate pests in the top layer of the soil, rather than burning piles of straw which is another common practice in certain countries. Thus, the straw and soil slowly burn together, resulting in even lower combustion efficiency. Various high-volume (Hi-vol) and medium-volume samplers were utilized. A PM10 Hi-vol sampler (Thermo Andersen) equipped with a 2.5 mm single-stage impactor was used to collect PM2.5 and PM2.5–10 particles for anhydrosugar determination at the suburban site. Three samplers for total suspended particulate matter (TSP) were employed at the two sites: one Thermo Andersen sampler (to collect TSP for PM mass determination) and a Kimoto TSP sampler (for anhydrosugar measurement) were installed at the suburban site, while one Thermo Andersen sampler was used at the countryside location to collect TSP for PM mass determination. In addition, a UAS Sampler (MSP Corporation) was operated at the countryside location to obtain PM2.5 and PM2.5–10 particles for anhydrosugar analysis. PM>10 fractions were calculated by subtraction of
8302
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
Fig. 1. Location of Taiwan in East Asia (A) and Yunlin County in South-Central Taiwan (B).
PM10 from TSP tracer concentrations. Prior to the study, the sampling precision for the sized-resolved samplers (PM10impactor and UAS) was verified to be within 10%. Samples were collected on pre-fired quartz fiber filters (Pall Corporation) for anhydrosugar determination and on glass fiber filters (Pall Corporation) for PM mass measurement. All sample filters were stored at 20 C until sample analysis. Quality control (QC) procedures also included collection and analysis of blank samples. The study period extended from 18 to 26 December 2006, including an episode of intensive burning (22–24 December) as well as periods prior to and following this episode with continued, yet less intensive burning activities (18–21 and 25–26 December), designated as ‘‘seasonal background’’ and ‘‘post-burning’’ periods (Table 1). The typical sampling length was 24 h, except for two 12-h day/ night sampling periods in each phase of the study. The prevailing meteorological conditions during the frame of the study were characterized by fairly high temperatures (20.0 1.73 C), low to medium wind speeds (3.9 1.3 m s1, predominantly northerly and north-westerly), and relatively dry conditions (no precipitation and R.H. ¼ 74.3 2.9%), as illustrated in Table 1. Ambient concentrations of total suspended particulate matter (TSP)
were 110 (30.3) mg m3 on average for the entire study period, and 87 (4.9), 144 (25.3), and 106 (17.7) mg m3 for the seasonal background, burning, and post-burning periods, respectively. In addition, anhydrosugar measurements were conducted from March to May, 2007, designated as ‘‘off-season background’’, with average TSP mass concentrations of 99.6 (18.9) mg m3 (n ¼ 9). 2.2. Analytical methods The quartz filter samples were extracted with deionized water (>18.2 MU resistivity) under ultrasonic agitation for 60 min, followed by filtration of the extracts (for removal of insoluble components) with a syringe filter containing a pre-combusted quartz filter (0.3 mm pore size). Typical recoveries of levoglucosan were 100.2 (4.7) %, based on spiked filter extractions (n ¼ 3). All sample extracts were stored at 4 C until sample analysis. Samples were analyzed by high-performance anion exchange chromatography (HPAEC) with pulsed amperometric detection (PAD) without further sample concentration or chemical derivatization (Engling et al., 2006a). A Dionex ICS-3000 ion chromatograph (IC) was utilized, operated in pulsed amperometric mode. Therefore, the IC was equipped with
Table 1 Meteorological data for entire sampling period Conditiona
Avg. W.S. (m s1) Predominant W.D. (Contribution, %) Avg. Temp. ( C) Avg. R.H. (%) TSP Mass (mg m3)
Harvest-season background
Biomass burning
Post-burning
Dec. 18
Dec. 19
Dec. 20
Dec. 21
Dec. 22
Dec. 23
Dec. 24
Dec. 25
Dec. 26
3.1 2.3 NW, NNW (44%) 17.7 5.3 74.6 16.3 90.2
2.4 2.1 N, NNW (53%) 19.6 6.1 69.9 11.2 90.1
4.6 2.1 N, NNW (94%) 22.3 5.6 75.2 14.7 79.6
4.9 2.1 N, NNE (69%) 20.2 8.8 72.3 11.4 86.8
4.1 4.3 N, NNW (54%) 18.5 6.2 69.2 17.3 119.1
1.9 2.0 N, NNW (55%) 21.1 7.2 74.1 19.3 169.7
4.3 1.9 N, NNW (67%) 22.3 12.3 75.6 19.6 143.1
6.0 4.4 N, NNW (73%) 18.0 10.8 79.4 15.8 93.4
3.8 2.6 NW, NNW (55%) 20.0 10.6 74.2 16.9 118.5
The year of 2006. a W.S.: wind speed; W.D.: wind direction; R.H.: relative humidity.
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
1500
Levoglucosan Conc. (ng m-3)
an electrochemical detector, amperometric cell, and CarboPak MA1 analytical column (4 250 mm). NaOH (400 mM) was used as eluent with a flow rate of 0.4 mL min1. Method detection limits for levoglucosan were estimated as 0.58 ng m3. Measurement precision was better than 4.9% for levoglucosan (CV; n ¼ 5). The procedural blanks of levoglucosan were below detection limits. PM mass concentrations were determined gravimetrically from the pre-weighed glass fiber filters (after conditioning at 25 C and 50% R.H. for 24 h) on a Mettler–Toledo analytical balance (0.01 mg).
A
8303
TSP
Countryside Suburban
1200 900 600 300 0
3. Results 3.1. Ambient levoglucosan distributions Ambient concentrations of the biomass burning tracer levoglucosan in three particle size ranges (PM2.5, PM10, and TSP) for the entire study period (seasonal background, intensive burning, and post-burning) at the countryside location ranged between 167–687, 199–788, and 334– 1138 ng m3, respectively (Table 2). The corresponding ambient levoglucosan observations at the suburban site for the three particle size classes were 237–1080, 245–1104, and 253–1438 ng m3, respectively. The average TSP levoglucosan concentrations at the countryside and suburban site were 836 and 889 ng m3, respectively, during the intensive burning period, and 573 and 519 ng m3 in the post-burning period, as demonstrated in Fig. 2. The higher tracer concentrations observed at the suburban site (relative to the countryside location) are likely due to this site being situated more directly downwind from the farming area with the majority of rice fields which were burned during the episode. The seasonal background TSP values of levoglucosan were 398 and 331 ng m3, respectively at the two locations, and the offseason background concentration (based on samples collected between March and May, 2007) at the suburban
Table 2 Levoglucosan concentrations as a function of particle size at both sampling locations (countryside and suburban site) Date
Levoglucosan conc. (ng m3) Countryside PM2.5
PM10a
Suburban TSP
PM2.5
PM10a
TSP
Harvest-season Background Dec. 18 195 Dec. 19 245 Dec. 20 237 Dec. 21 167
231 278 285 199
417 442 –b 334
309 371 237 332
316 378 245 342
331 390 253 352
Biomass burning Dec. 22 310 Dec. 23 687 Dec. 24 478
370 788 548
632 1138 737
456 1080 420
468 1104 437
701 1438 528
Post-burning Dec. 25 308 Dec. 26 346
344 395
398 747
252 471
258 485
404 633
a b
Determined by summation of PM2.5 and PM2.5–10 concentrations. Sampling failure.
Levoglucosan Conc. (ng m-3)
1200
B PM2.5
900
600
300
0
Harvest-Season Background
Biomass Burning
Post Buring
Off-Season Background
Fig. 2. Average levoglucosan concentrations in TSP (A) and PM2.5 (B) particles at both sampling locations during the December 2006 rice straw burning season, as well as the off-season background period. Error bars represent concentration ranges in the harvest-season background, burning episode, post-burning and off-season background periods in form of one standard deviation (n ¼ 4, 3, 2, and 12).
site was 133 ng m3 (Fig. 2). While tracer concentrations were notably enhanced (by a factor of 2–3 at the respective sites) during the active burning episode of rice straw relative to the seasonal background, levoglucosan levels in the post-burning period were found to still be rather high, likely because of ineffective dispersion, as well as continued burning of individual rice fields. Likewise, harvest-season background concentrations (measured from 18 to 21 December, 2006) of biomass burning tracers were rather high as well (levoglucosan concentrations were three times higher than the off-season background value, while PM mass concentrations were only 10% higher), but lower than those during the post-burning period. Again, this is most likely due to burning activities by individual rice farmers in the days (or weeks) prior to the intensive burning episode on the weekend of December 22, 2006. A possible reason for the rather small increase in PM mass concentrations (compared to the 3-fold increase in levoglucosan concentrations) between the background period and the episode with intensive burning is the dominant influence of various anthropogenic sources, including traffic emissions and industrial processes, as well as natural sources, on ambient PM levels in the central and southern part of Taiwan. Moreover, the winter months in this part of Taiwan, typically called the ‘‘dry’’ season, are characterized by relatively stagnant meteorological conditions, giving rise to accumulation of aerosol particles. The
8304
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
resulting aged aerosol consequently contains anhydrosugars from a variety of biomass burning activities. It is also noteworthy that levoglucosan is a selective tracer for biomass burning and is not emitted by other combustion processes, such as coal burning or different types of fossil fuel combustion. Thus, ambient levoglucosan concentrations are relatively more responsive to the influence of biomass burning activities compared to total aerosol mass concentrations. On the other hand, PM levels increased drastically (by approximately 70%) during the intensive burning episode, clearly indicating the substantial impact of post-harvest burning on ambient air quality. Fine particle (PM2.5) levoglucosan concentrations at the countryside and suburban site were highly elevated during the intensive burning period with average values of 492 (189) and 652 (371) ng m3, respectively, and 327 (27) and 362 (155) ng m3 in the post-burning period. Harvest-season background values at the two sites were 211 (38) and 312 (56) ng m3, respectively (Table 2). The off-season background concentration of levoglucosan at the suburban site (88 ng m3) was one order of magnitude lower than the observed values during the intensive burning period. The PM2.5 levoglucosan concentrations showed rather large variations during the intensive burning period, while exhibiting the same pattern as the TSP tracer concentrations, with peak values on December 23, 2006, when burning activities were most intense. During the intensive burning period, the day/night levoglucosan concentration ratios in PM2.5 were rather high with values of 3.9 and 1.9 at the countryside and suburban area, respectively (Fig. 3). The corresponding TSP ratios were 2.4 at the countryside and 1.9 in the suburban area. The day/night levoglucosan ratios during the post-burning and seasonal background periods were substantially lower (10 fraction was rather high with 35 (13) %, as shown in Fig. 4A. In contrast, the PM2.5 levoglucosan fraction was predominant in the suburban area, accounting for 81 (13) % of TSP on average (Fig. 4B). The levoglucosan measured during the period prior to the intensive burning episode at the suburban site was almost entirely present in the fine particle mode; 94% of the total levoglucosan concentration was found in PM2.5. The PM>10 fraction drastically increased during the intensive burning episode, both absolutely as well as relative to the fine fraction, from 3% in the background period to 25% of TSP during the burning episode.
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
8305
Table 3 Comparison of ambient fine-particle (PM2.5) levoglucosan concentrations from rice straw burning (this study) with levoglucosan emissions from combustion of other types of biomass (reported in the literature) Biomass type
Location (state/city, season)
Levoglucosan (ng m-3)
Reference
Mean (range) Rice straw Rice straw Wheat stubble Agricultural waste Agricultural waste Agricultural waste Agricultural waste Agricultural waste Forest (wildfire) Forest (wildfire)
(mixed) (mixed) (mixed) (mixed) (mixed)
Suburban
B
Suburban
0.6
1200
Harvest-Season Background
900 0.6
600
Harvest-Season Background
Biomass Burning
Dec. 26
Dec. 25
Dec. 24
Dec. 23
Dec. 22
Dec. 21
Dec. 20
Dec. 19
0
Dec. 18
300
Post Burning
Fig. 4. Timelines of ambient levoglucosan concentrations in three particle size ranges (PM2.5, PM10, and TSP) observed at the countryside (A) and suburban sites (B).
0.5
B
Biomass Burning
Dec. 26
1500
Countryside
0.7
Dec. 25
0
0.8
Dec. 24
300
A
0.9
Dec. 23
Conc. Ratio PM2.5 / PM10)
1.0
Dec. 22
600
Dec. 21
900
Dec. 20
TSP PM10 PM2.5
Dec. 19
Countryside
A
corresponding ratio at the suburban site, the PM2.5 levoglucosan fraction of PM10 at this site was still dominant (Fig. 5A). The slightly higher fine particle concentrations measured at the suburban site might be due to loss (by dry deposition) of a portion of PM2.5–10 during transport from rice fields north of Douliou through the urban area to the suburban sampling site on the south side of the city. Whereas the PM2.5/PM10 levoglucosan ratios were fairly consistent at both locations throughout the entire campaign, the PM>10/TSP ratios showed a broad variation in all periods (prior to, during, and after the intensive burning episode), particularly at the suburban site, as demonstrated in Fig. 5B. The harvest-season background PM>10/TSP ratios of the ambient levoglucosan concentrations were 0.41 (0.04) and 0.03 (0.01) at
Dec. 18
1200
Levoglucosan Conc. (ng m-3)
(This study) (This study) Viana et al., 2008 Jimenez et al., 2006 Rinehart et al., 2006 Park et al., 2006 Wan and Yu, 2007 Wang et al., 2007 He et al., 2006 Ward et al., 2006a Zdra´hal et al., 2002
Measured in PM10.
In terms of PM10, Fig. 5A illustrates the fine particle ratios (PM2.5/PM10) of the ambient levoglucosan concentrations at the countryside and suburban sites; the average PM2.5/PM10 ratio was 0.87 (0.02) and 0.97 (0.01) at the two sites, respectively. In agreement with the high PM2.5 levoglucosan fraction in TSP (as shown above), fine particle tracer concentrations clearly dominated the atmospheric PM10 composition in the suburban area (PM2.5/ PM10 ¼ 0.97). While the PM2.5/PM10 levoglucosan ratio at the countryside location was somewhat lower than the
Levoglucosan Conc. (ng m-3)
572 (310–1080) 262 (167–371) 74 (58–129)a 78 202 1754 190 (35–489) 480 (120–950) 78 3152 (1726–6091) 2006 (446–4106)
Conc. Ratio (PM> 10 / TSP)
a
Taiwan (Douliou, Burning peak) Taiwan (Douliou, Burning season) Spain (Valencia) U.S.A. (Washington and Idaho) U.S.A. (California/Fresno) Korea (Gwangju) China (Hong Kong, Winter) China (Guangzhou) China (Beijing, Winter) U.S.A. (Montana) Brazil (Rondoˆnia, dry season)
Post Burning
Countryside Suburban
0.4 0.3 0.2 0.1 0.0
Harvest-Season Background
Biomass Burning
Post Burning
Fig. 5. Illustration of PM2.5/PM10 (A) and PM>10/TSP (B) concentration ratios for levoglucosan throughout entire study period. Error bars represent ranges of concentration ratios in the harvest-season background, burning episode, and post-burning periods in form of one standard deviation (n ¼ 4, 3, and 2).
8306
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
the countryside and suburban sites, respectively (Fig. 5B). Surprisingly high levoglucosan concentrations were found in PM>10 at the countryside location during the entire study period. At the suburban site, the PM>10 tracer concentrations were rather low during the period prior to the intensive burning activities, but increased dramatically during the burning episode and continued to be elevated during the following days. Both laboratory and ambient measurements of smoke emissions from biomass burning processes have previously shown anhydrosugars to be predominantly present in the fine particle mode (Engling et al., 2006a; Herckes et al., 2006). On the contrary, the size-resolved measurements during this campaign revealed rather high biomass burning tracer concentrations in particles with aerodynamic diameters larger than 10 mm. As mentioned above, no sizeresolved ambient measurements of biomass burning tracers from rice straw burning have been reported to date. Thus, no direct comparison of such observations with other locations is possible. It is noteworthy that the burning practices of rice straw in Taiwan are different than those in other countries, as thin layers of straw and soil slowly burn together at relatively low temperatures. Such distinctly different burning conditions for rice straw (compared to other types of biomass) may result in soil and ash particles, which are typically more coarse, to be elevated during the burning process by the convective processes over the flame and subsequently be suspended into the air together with the smoke particles. In addition, the higher ambient relative humidity in Taiwan (compared to other regions with biomass burning processes, such as wildfires in the Western U.S. or Savanna fires in Africa) may result in substantial water up-take of rice straw smoke. Smoke particles from rice straw burning have been shown in laboratory experiments to be significantly more hygroscopic than most other types of biomass (Engling, unpublished data), enhancing the water up take potential of rice straw smoke. This unique feature of rice straw smoke particles together with other atmospheric process, including coagulation of smoke particles with ambient aerosols, may partially explain the observation of high PM2.5–10 and PM>10 tracer concentrations during burning of rice straw. Moreover, re-suspension of soil from rice fields by wind activity (wind speeds in Yunlin County can be up to 6–10 m s1) may constitute additional sources of PM2.5–10 and PM>10, containing the source tracers. Preliminary experiments in our laboratory have shown levoglucosan to be present in local soil. For instance, the abundance of levoglucosan in soil from rice fields in Yunlin County during spring (wet season) and winter time (dry season) of 2007 was measured at average values of 1.8 (0.75; n ¼ 6) and 8.7 (7.7; n ¼ 6) mg g1, respectively. Due to the high solubility of anhydrosugar tracers, it is expected that these measurements during the wet season constitute lower estimates of the tracer content in soil. Therefore, further investigations are in progress in order to quantify the tracer content in such particles, specifically during the winter post-harvest burning season, and to estimate the potential ambient contributions from the re-suspension of soil. On the other hand, the PM2.5–10 and PM>10 fractions observed in the suburban area during the background
period were rather low. This might be in part due to dry deposition of the larger particles during transport of aerosol particles from fields in the farming area north of the suburban sampling site. Also, re-suspension of soil did not play an important role in the contribution of PM>10 levoglucosan as was the case at the countryside location which was surrounded by cultivated land. While during the active burning period PM>10/TSP levoglucosan concentration ratios at the countryside were similar to the background period, the fraction of PM>10 levoglucosan increased drastically at the suburban site (from 0.03 to 0.25 0.08) during the burning episode, as demonstrated in Fig. 5B. Thus, the particle size distributions at the two locations (countryside and suburban site) showed closer resemblance during the burning episode. In the post-burning period, levoglucosan concentration ratios (PM>10/TSP) at the countryside and suburban areas were 0.31 (0.23) and 0.30 (0.09), respectively (Fig. 5B), indicating that levoglucosan concentration ratios at the two sites had reached some type of steady condition in the atmosphere. 4. Conclusions Biomass burning in general and agricultural burning activities in particular are important sources of aerosol particles on a global basis affecting local and regional air quality. Distinguishing different combustion source types, e.g., agricultural burning versus forest fires, is crucial, and thus detailed studies of specific burning processes are needed, both in the laboratory and in ambient air. To our knowledge, this is the first investigation of sized-resolved anhydrosugar distributions in ambient air resulting from post-harvest burning of rice fields in Asia. Significantly increased tracer concentrations were observed at two locations (rural and suburban) during the active burning episode within the post-harvest burning season of rice straw in Taiwan. Even higher levoglucosan levels were noticed at the suburban site relative to the countryside, showing the importance of local-scale (as well as regionalscale) transport of smoke aerosol. Interestingly, a vast mass fraction of levoglucosan was present in PM>10, possibly due to a variety of factors, including coagulation of fresh smoke particles, enhanced by the high relative humidity, as well as suspension of soil and ash particles during the combustion process of the rice straw. Fairly high background tracer concentrations were also observed prior to the intensive burning episode as well as outside the post-harvest burning season. This may be in part due to persistent contributions of smoke from small-scale fires and re-suspension of soil, containing smoke particles which have been accumulated over time. This study has further demonstrated the significance of rice straw burning practices on air quality by investigating ambient source tracer concentrations as a function of particle size. Our measurements were also applied to estimate source contributions to ambient PM2.5 based on emission factors from chamber studies. Furthermore, we have shown that the unique conditions associated with agricultural burning practices in subtropical regions of Asia may give rise to distinctly different chemical and physical characteristics of the resulting smoke aerosols compared to
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
chamber studies or ambient observations in other regions. The important new insights gained from this study provide motivation and direction for future research to better characterize biomass burning emissions, particularly from agricultural burning practices, and to generate source profiles that are representative of actual burning activities as well as the associated ambient conditions. Such work is in progress in our laboratory, including additional ambient observations, near-source measurements, and chamber studies. Acknowledgments Thanks are due to all the students from National Yunlin University of Science and Technology who helped collect the filter samples and perform the chemical analyses. The authors also acknowledge the helpful assistance by Mr. Sung-Tien Yang, principal of Jiouan Elementary School, and his colleagues during the sampling process. Financial support for this study came from the Engineering School of the National Yunlin University of Science and Technology (ENG 94-5) as well as the Research Center for Environmental Changes at Academia Sinica. References Dhammapala, R., Claiborn, C., Corkill, J., Gullett, B., 2006. Particulate emissions from wheat and Kentucky bluegrass stubble burning in eastern Washington and northern Idaho. Atmospheric Environment 40, 1007–1015. Dhammapala, R., Claiborn, C., Jimenez, J., Corkill, J., Gullett, B., Simpson, C., Paulsen, M., 2007. Emission factors of PAHs, methoxyphenols, levoglucosan, elemental carbon and organic carbon from simulated wheat and Kentucky bluegrass stubble burns. Atmospheric Environment 41, 2660–2669. Engling, G., Carrico, C.M., Kreidenweis, S.M., Collett Jr., J.L., Day, D.E., Malm, W.C., Hao, W.M., Lincoln, E., Iinuma, Y., Herrmann, H., 2006a. Determination of levoglucosan in biomass combustion aerosol by High Performance Anion Exchange Chromatography with Pulsed Amperometric Detection. Atmospheric Environment 40, S299–S311. Engling, G., Herckes, P., Kreidenweis, S.M., Collett Jr., J.L., 2006b. Composition of the fine organic aerosol in Yosemite National Park during the 2002 Yosemite Aerosol Characterization Study. Atmospheric Environment 40, 2959–2972. Gorin, C.A., Collett Jr., J.L., Herckes, P., 2006. Wood smoke contribution to winter aerosol in Fresno, CA. Journal of the Air & Waste Management Association 56, 1584–1590. Gullett, B.K., Touati, A., 2003. PCDD/F emissions from burning wheat and rice field residue. Atmospheric Environment 37, 4893–4899. Hays, M.D., Fine, P.M., Geron, C.D., Kleeman, M.J., Gullett, B.K., 2005. Open burning of agricultural biomass: physical and chemical properties of particle-phase emissions. Atmospheric Environment 39, 6747–6764. He, L.Y., Hu, M., Huang, X.F., Zhang, Y.H., Tang, X.Y., 2006. Seasonal pollution characteristics of organic compounds in atmospheric fine particles in Beijing. Science of the Total Environment 359, 167–176. Herckes, P., Engling, G., Kreidenweis, S.M., Collett Jr., J.L., 2006. Particle size distributions of organic aerosol constituents during the 2002 Yosemite Aerosol Characterization Study. Environmental Science & Technology 40, 4554–4562. Hornig, J.F., Soderberg, R.H., Barefoot, A.C., Galasyn, J.F., 1985. Wood smoke analysis: vaporization losses of PAH from filters and levoglucosan as a distinctive marker for wood smoke. In: Cooke, M., Dennis, A.J. (Eds.), Polynuclear Aromatic Hydrocarbons: Mechanisms, Methods, and Metabolism. Batelle Press, pp. 561–568. Jimenez, J., Wu, C.-F., Claiborn, C., Gould, T., Simpson, C.D., Larson, T., Liu, S.L.-J., 2006. Agricultural burning smoke in eastern Washington – part I: atmospheric characterization. Atmospheric Environment 40, 639–650. Kadam, K.L., Forrest, L.H., Jacobson, W.A., 2000. Rice straw as a lignocellulosic resource: collection, processing, transportation, and environmental aspects. Biomass & Bioenergy 18, 369–389.
8307
Keshtkar, H., Ashbaugh, L.L., 2007. Size distribution of polycyclic aromatic hydrocarbon particulate emission factors from agricultural burning. Atmospheric Environment 41, 2729–2739. Leithead, A., Li, S.M., Hoff, R., Cheng, Y., Brook, J., 2006. Levoglucosan and dehydroabietic acid: evidence of biomass burning impact on aerosols in the Lower Fraser Valley. Atmospheric Environment 40, 2721–2734. Li, X.G., Wang, S.X., Duan, L., Hao, J., Li, C., Chen, Y.S., Yang, L., 2007. Particulate and trace gas emissions from open burning of wheat straw and corn stover in China. Environmental Science & Technology 41, 6052–6058. Mazzoleni, L.R., Zielinska, B., Moosmuller, H., 2007. Emissions of levoglucosan, methoxy phenols, and organic acids from prescribed burns, laboratory combustion of wildland fuels, and residential wood combustion. Environmental Science & Technology 41, 2115–2122. Medeiros, P.M., Conte, M.H., Weber, J.C., Simoneit, B.R.T., 2006. Sugars as source indicators of biogenic organic carbon in aerosols collected above the Howland Experimental Forest, Maine. Atmospheric Environment 40, 1694–1705. Ogawa, M., Yoshida, N., 2005. Nitrous oxide emission from the burning of agricultural residue. Atmospheric Environment 39, 3421–3429. Otto, A., Gondokusumo, R., Simpson, M.J., 2006. Characterization and quantification of biomarkers from biomass burning at a recent wildfire site in Northern Alberta, Canada. Applied Geochemistry 21, 166–183. Park, S.S., Bae, M.S., Schauer, J.J., Kim, Y.J., Cho, S.Y., Kim, S.J., 2006. Molecular composition of PM2.5 organic aerosol measured at an urban site of Korea during the ACE-Asia campaign. Atmospheric Environment 40, 4182–4198. Puxbaum, H., Caseiro, A., Sanchez-Ochoa, A., Kasper-Giebl, A., Claeys, M., Gelencser, A., Legrand, M., Preunkert, S., Pio, C., 2007. Levoglucosan levels at background sites in Europe for assessing the impact of biomass combustion on the European aerosol background. Journal of Geophysical Research – Atmospheres 112, Art. No. D23S05. doi: 10.1029/2006JD008114. Rinehart, L.R., Fujita, E.M., Chow, J.C., Magliano, K., Zielinska, B., 2006. Spatial distribution of PM2.5 associated organic compounds in central California. Atmospheric Environment 40, 290–303. Schmidl, C., Marr, L.L., Caseiro, A., Kotianova, P., Berner, A., Bauer, H., KasperGiebl, A., Puxbaum, H., 2008. Chemical characterization of fine particle emissions from wood stove combustion of common woods growing in mid-European Alpine regions. Atmospheric Environment 42, 126–141. Sheesley, R.J., Schauer, J.J., Chowdhury, Z., Cass, G.R., Simoneit, B.R.T., 2003. Characterization of organic aerosols emitted from the combustion of biomass indigenous to South Asia. Journal of Geophysical Research 108, Art. No. 4285. doi: 10.1029/2002JD002981. Simoneit, B.R.T., Schauer, J.J., Nolte, C.G., Oros, D.R., Elias, V.O., Fraser, M.P., Rogge, W.F., Cass, G.R., 1999. Levoglucosan, a tracer for cellulose in biomass burning and atmospheric particles. Atmospheric Environment 33, 173–182. Streets, D.G., Yarber, K.F., Woo, J.H., Carmichael, G.R., 2003. Biomass burning in Asia: annual and seasonal estimates and atmospheric emissions. Global Biogeochemical Cycles 17, Art. No. 1099. Venkataraman, C., Habib, G., Eiguren-Fernandez, A., Miguel, A.H., Friedlander, S.K., 2005. Residential biofuels in south Asia: carbonaceous aerosol emissions and climate impacts. Science 307, 1454– 1456. Viana, M., Lopez, J.M., Querol, X., Alastuey, A., Garcia-Gacio, D., BlancoHeras, G., Lopez-Mahia, P., Pineiro-Iglesias, M., Sanz, M.J., Sanz, F., Chi, X., Maenhaut, W., 2008. Tracers and impact of open burning of rice straw residues on PM in Eastern Spain. Atmospheric Environment 42, 1941–1957. Wan, E.C.H., Yu, J.Z., 2007. Analysis of sugars and sugar polyols in atmospheric aerosols by chloride attachment in liquid chromatography/ negative ion electrospray mass spectrometry. Environmental Science & Technology 41, 2459–2466. Wang, Q.Q., Shao, M., Liu, Y., Kuster, W., Goldan, P., Li, X.H., Liu, Y.A., Lu, S. H., 2007. Impact of biomass burning on urban air quality estimated by organic tracers: Guangzhou and Beijing as cases. Atmospheric Environment 41, 8380–8390. Ward, T.J., Hamilton, R.F., Dixon, R.W., Paulsen, M., Simpson, C.D., 2006a. Characterization and evaluation of smoke tracers in PM: results from the 2003 Montana wildfire season. Atmospheric Environment 40, 7005–7017. Ward, T.J., Rinehart, L.R., Lange, T., 2006b. The 2003/2004 Libby, Montana PM2.5 source apportionment research study. Aerosol Science and Technology 40, 166–177. Yang, H.H., Tsai, C.H., Chao, M.R., Su, Y.L., Chien, S.M., 2006. Source identification and size distribution of atmospheric polycyclic aromatic hydrocarbons during rice straw burning period. Atmospheric Environment 40, 1266–1274.
8308
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
Yevich, R., Logan, J.A., 2003. An assessment of biofuel use and burning of agricultural waste in the developing world. Global Biogeochemical Cycles 17, Art. No. 1095. Yoshinori, M., Kanno, T., 1997. Emissions of trace gases (CO2, CO, CH4, and N2O) resulting from rice straw burning. Soil Science and Plant Nutrition 43, 849–854. Yttri, K.E., Dye, C., Slordal, L.H., Braathen, O.A., 2005. Quantification of monosaccharide anhydrides by liquid chromatography combined with mass spectrometry: application to aerosol samples from an urban and a suburban site influenced by small-scale wood
burning. Journal of the Air & Waste Management Association 55, 1169–1177. Zdrahal, Z., Oliveira, J., Vermeylen, R., Claeys, M., Maenhaut, W., 2002. Improved method for quantifying levoglucosan and related monosaccharide anhydrides in atmospheric aerosols and application to samples from urban and tropical locations. Environmental Science & Technology 36, 747–753. Zhang, Y.X., Shao, M., Zhang, Y.H., Zeng, L.M., He, L.Y., Zhu, B., Wei, Y.J., Zhu, X.L., 2007. Source profiles of particulate organic matters emitted from cereal straw burnings. Journal of Environmental Sciences – China 19,167–175.
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Particle size characteristics of levoglucosan in ambient aerosols from rice straw burning James J. Lee a, *, Guenter Engling b, Shih-Chun Candice Lung b, Kuo-Yang Lee a a b
Department of Environmental Engineering and Safety, National Yunlin University of Science and Technology, Douliou, Taiwan Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 March 2008 Received in revised form 15 July 2008 Accepted 21 July 2008
Agricultural residue burning in Asia is an important source of atmospheric aerosol particles. This study investigates the impact of post-harvest burning of rice fields on the air quality in Taiwan, a subtropical island in East Asia. Size-resolved measurements of the anhydrosugar levoglucosan (biomass burning tracer derived from the thermal breakdown of cellulose) were conducted before, during and after an episode of intensive rice straw burning at a rural and suburban site. While substantially enhanced levoglucosan concentrations were observed during the active rice straw burning episode (up to 1400 ng m3), fairly high values of the smoke tracer were measured throughout the entire study period. Moreover, unusually high levoglucosan levels were present in aerosol particles with aerodynamic diameters larger than 10 mm (PM>10), possibly influenced by the ambient atmospheric conditions, such as high relative humidity, in addition to unique properties of rice straw smoke and the specific burning practices of rice fields. Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Agricultural biomass Anhydrosugars Atmospheric aerosols Particle size distributions HPAEC-PAD
1. Introduction Atmospheric aerosol particles have substantial influence on regional air quality, global geochemical cycles, and the radiation budget of the earth. While the combustion of biomass contributes significant amounts of aerosols to the atmosphere on a global scale, the Asian continent is a particularly important source region for biomass smoke (Streets et al., 2003). The dominant constituents of most types of biomass are the polysaccharides cellulose and hemicellulose. The anhydrosugar levoglucosan, derived from the thermal degradation of cellulose (Hornig et al., 1985; Simoneit et al., 1999), is the major molecular tracer for biomass burning activities and has been utilized in several recent source apportionment studies (Leithead
* Corresponding author. Department of Environmental Engineering and Safety, National Yunlin University of Science and Technology, 123, University Rd. Section 3, Douliou, Taiwan. Tel.: þ886 5 534 2601; fax: þ886 5 531 2069. E-mail address: [email protected] (J.J. Lee). 1352-2310/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2008.07.047
et al., 2006; Puxbaum et al., 2007; Rinehart et al., 2006; Ward et al., 2006b). Biomass burning practices in Asia include a large variety of burning activities, such as domestic biofuel use (for cooking and heating), post-harvest burning of agricultural fields, and wildfires (Venkataraman et al., 2005). Several anhydrosugar investigations in North and South America as well as Europe have focused on smoke emissions from wood combustion, providing an extensive range of chemical and physical characteristics of wood smoke aerosol. Two major types of wood burning activities can be distinguished: open burning of trees and shrubs in forests (either in form of wildfires or prescribed burns) and combustion of wood under confined conditions, such as in domestic wood stoves and fireplaces. Smoke from forest fires has been investigated by various research groups (Engling et al., 2006b; Otto et al., 2006; Ward et al., 2006a; Zdrahal et al., 2002). Smoke emissions from residential wood burning have been characterized in several studies as well (Gorin et al., 2006; Leithead et al., 2006; Schmidl et al., 2008; Ward et al., 2006b; Yttri et al., 2005).
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
Conversely, fewer studies have investigated smoke emissions from agricultural burning activities. The most common type of agricultural burning, particularly in Asia, is post-harvest combustion of crop residues (Yevich and Logan, 2003). Considering the large population in Asian countries, the production of agricultural commodities and the associated residue burning is expected to result in considerable aerosol particle emissions, likely comparable to that of occasional wildland fires in North America and other parts of the world. Previous studies of anhydrosugars in smoke from agricultural residue burning have been conducted as controlled combustion experiments, including controlled field burning and laboratory (chamber) burns (Dhammapala et al., 2007; Hays et al., 2005; Mazzoleni et al., 2007; Sheesley et al., 2003; Zhang et al., 2007), as well as ambient measurements (Jimenez et al., 2006; Wang et al., 2007). In many cases, the measured ambient smoke originated from the combustion of mixed agricultural residues (He et al., 2006; Park et al., 2006; Wan and Yu, 2007), while a few ambient studies have been carried out on specific crop species, e.g., wheat straw and grass (Dhammapala et al., 2006). Smoke measurements from laboratory-scale burns of agricultural residues were typically conducted on individual crop types, yet only on a small number of species, particularly wheat and rice straw (Dhammapala et al., 2007; Gullett and Touati, 2003; Li et al., 2007). Investigations of smoke emissions from burning of rice straw are limited to several laboratory experiments and a few ambient observations. Various chemical compounds were measured during the chamber combustion of rice straw in some studies (Hays et al., 2005; Mazzoleni et al., 2007; Sheesley et al., 2003), whereas other laboratoryscale investigations focused on selected chemical species contained in rice straw smoke, such as dioxins (Gullett and Touati, 2003), nitrous oxide (Ogawa and Yoshida, 2005), and PAHs (Keshtkar and Ashbaugh, 2007). Ambient observations of rice straw smoke are limited to the measurement of PAHs (Yang et al., 2006), trace gases (Yoshinori and Kanno, 1997), and one recent study of anhydrosugars in ambient PM10 (Viana et al., 2008). No size-resolved measurements of anhydrosugar source tracers from post-harvest burning of rice straw in ambient air are available to date. Biomass burning contributions to the ambient air in Taiwan are predominantly from burning of rice fields following the harvest in each of the two growing seasons (in early summer and early winter). The winter burning season in Taiwan typically occurs in the time frame from late November to early January, coinciding with the dry season in central and southern Taiwan. Yunlin County is one of the major rice production areas in Taiwan; three neighboring counties in South-Central Taiwan (Changhua, Yunlin, and Chiayi counties) contribute 53% of the total rice production in Taiwan and the contributions from Yunlin County in particular are 19%. The ambient conditions and the geographical setting of Yunlin County are most favorable for the investigation of smoke emissions from rice straw burning during the winter season. Rice straw dries rapidly during this time and the burning is not hindered by
8301
wet precipitation, as is the case in the summer harvest/ burning season. As more than 90% of the global rice production occurs in Asia (Kadam et al., 2000), it is obvious that smoke generated from post-harvest burning of rice straw in Taiwan, as well as in the other Asian countries, contributes a sizeable portion of particulate pollution on local and regional scale. Therefore, the goal of this study is to characterize the smoke aerosol from rice field burning activities in Taiwan and its impact on ambient aerosol concentrations, particularly by investigating the size-resolved distributions of the anhydrosugar levoglucosan as molecular source tracer. 2. Experimental 2.1. Sample collection Aerosol samples were collected in Yunlin County, located in South-Central Taiwan (Fig. 1), at two sites during the winter rice field burning season in December 2006, including a short episode of intensive burning as well as periods prior to and following the episode. The first location was termed ‘‘suburban site’’ and was situated on the roof of a four-story building on the campus of the National Yunlin University of Science & Technology, located in the southern part of Douliou, Taiwan (23 420 9100 N, 120 340 17.900 E). A farming area with a large number of rice fields was located within 7–20 km to the north of this sampling site. The second sampling site was located in a rural area of Yunlin County, south of Douliou, and thus it was designated as ‘‘countryside’’ location. Samples were collected at this site on the roof of a two-story building of the Jiouan Elementary School. This site was surrounded by cultivated (agricultural) land, mainly used for rice farming. However, the number of rice fields in the surrounding area of this site was smaller than in the area to the north of the suburban site. Consequently, the suburban site was located directly downwind of the major rice production area within the study region, as the predominant wind direction during the study period was northerly (Table 1). The burning practices of rice straw in Taiwan are unique, as thin layers of straw are typically laid on the ground in an attempt to exterminate pests in the top layer of the soil, rather than burning piles of straw which is another common practice in certain countries. Thus, the straw and soil slowly burn together, resulting in even lower combustion efficiency. Various high-volume (Hi-vol) and medium-volume samplers were utilized. A PM10 Hi-vol sampler (Thermo Andersen) equipped with a 2.5 mm single-stage impactor was used to collect PM2.5 and PM2.5–10 particles for anhydrosugar determination at the suburban site. Three samplers for total suspended particulate matter (TSP) were employed at the two sites: one Thermo Andersen sampler (to collect TSP for PM mass determination) and a Kimoto TSP sampler (for anhydrosugar measurement) were installed at the suburban site, while one Thermo Andersen sampler was used at the countryside location to collect TSP for PM mass determination. In addition, a UAS Sampler (MSP Corporation) was operated at the countryside location to obtain PM2.5 and PM2.5–10 particles for anhydrosugar analysis. PM>10 fractions were calculated by subtraction of
8302
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
Fig. 1. Location of Taiwan in East Asia (A) and Yunlin County in South-Central Taiwan (B).
PM10 from TSP tracer concentrations. Prior to the study, the sampling precision for the sized-resolved samplers (PM10impactor and UAS) was verified to be within 10%. Samples were collected on pre-fired quartz fiber filters (Pall Corporation) for anhydrosugar determination and on glass fiber filters (Pall Corporation) for PM mass measurement. All sample filters were stored at 20 C until sample analysis. Quality control (QC) procedures also included collection and analysis of blank samples. The study period extended from 18 to 26 December 2006, including an episode of intensive burning (22–24 December) as well as periods prior to and following this episode with continued, yet less intensive burning activities (18–21 and 25–26 December), designated as ‘‘seasonal background’’ and ‘‘post-burning’’ periods (Table 1). The typical sampling length was 24 h, except for two 12-h day/ night sampling periods in each phase of the study. The prevailing meteorological conditions during the frame of the study were characterized by fairly high temperatures (20.0 1.73 C), low to medium wind speeds (3.9 1.3 m s1, predominantly northerly and north-westerly), and relatively dry conditions (no precipitation and R.H. ¼ 74.3 2.9%), as illustrated in Table 1. Ambient concentrations of total suspended particulate matter (TSP)
were 110 (30.3) mg m3 on average for the entire study period, and 87 (4.9), 144 (25.3), and 106 (17.7) mg m3 for the seasonal background, burning, and post-burning periods, respectively. In addition, anhydrosugar measurements were conducted from March to May, 2007, designated as ‘‘off-season background’’, with average TSP mass concentrations of 99.6 (18.9) mg m3 (n ¼ 9). 2.2. Analytical methods The quartz filter samples were extracted with deionized water (>18.2 MU resistivity) under ultrasonic agitation for 60 min, followed by filtration of the extracts (for removal of insoluble components) with a syringe filter containing a pre-combusted quartz filter (0.3 mm pore size). Typical recoveries of levoglucosan were 100.2 (4.7) %, based on spiked filter extractions (n ¼ 3). All sample extracts were stored at 4 C until sample analysis. Samples were analyzed by high-performance anion exchange chromatography (HPAEC) with pulsed amperometric detection (PAD) without further sample concentration or chemical derivatization (Engling et al., 2006a). A Dionex ICS-3000 ion chromatograph (IC) was utilized, operated in pulsed amperometric mode. Therefore, the IC was equipped with
Table 1 Meteorological data for entire sampling period Conditiona
Avg. W.S. (m s1) Predominant W.D. (Contribution, %) Avg. Temp. ( C) Avg. R.H. (%) TSP Mass (mg m3)
Harvest-season background
Biomass burning
Post-burning
Dec. 18
Dec. 19
Dec. 20
Dec. 21
Dec. 22
Dec. 23
Dec. 24
Dec. 25
Dec. 26
3.1 2.3 NW, NNW (44%) 17.7 5.3 74.6 16.3 90.2
2.4 2.1 N, NNW (53%) 19.6 6.1 69.9 11.2 90.1
4.6 2.1 N, NNW (94%) 22.3 5.6 75.2 14.7 79.6
4.9 2.1 N, NNE (69%) 20.2 8.8 72.3 11.4 86.8
4.1 4.3 N, NNW (54%) 18.5 6.2 69.2 17.3 119.1
1.9 2.0 N, NNW (55%) 21.1 7.2 74.1 19.3 169.7
4.3 1.9 N, NNW (67%) 22.3 12.3 75.6 19.6 143.1
6.0 4.4 N, NNW (73%) 18.0 10.8 79.4 15.8 93.4
3.8 2.6 NW, NNW (55%) 20.0 10.6 74.2 16.9 118.5
The year of 2006. a W.S.: wind speed; W.D.: wind direction; R.H.: relative humidity.
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
1500
Levoglucosan Conc. (ng m-3)
an electrochemical detector, amperometric cell, and CarboPak MA1 analytical column (4 250 mm). NaOH (400 mM) was used as eluent with a flow rate of 0.4 mL min1. Method detection limits for levoglucosan were estimated as 0.58 ng m3. Measurement precision was better than 4.9% for levoglucosan (CV; n ¼ 5). The procedural blanks of levoglucosan were below detection limits. PM mass concentrations were determined gravimetrically from the pre-weighed glass fiber filters (after conditioning at 25 C and 50% R.H. for 24 h) on a Mettler–Toledo analytical balance (0.01 mg).
A
8303
TSP
Countryside Suburban
1200 900 600 300 0
3. Results 3.1. Ambient levoglucosan distributions Ambient concentrations of the biomass burning tracer levoglucosan in three particle size ranges (PM2.5, PM10, and TSP) for the entire study period (seasonal background, intensive burning, and post-burning) at the countryside location ranged between 167–687, 199–788, and 334– 1138 ng m3, respectively (Table 2). The corresponding ambient levoglucosan observations at the suburban site for the three particle size classes were 237–1080, 245–1104, and 253–1438 ng m3, respectively. The average TSP levoglucosan concentrations at the countryside and suburban site were 836 and 889 ng m3, respectively, during the intensive burning period, and 573 and 519 ng m3 in the post-burning period, as demonstrated in Fig. 2. The higher tracer concentrations observed at the suburban site (relative to the countryside location) are likely due to this site being situated more directly downwind from the farming area with the majority of rice fields which were burned during the episode. The seasonal background TSP values of levoglucosan were 398 and 331 ng m3, respectively at the two locations, and the offseason background concentration (based on samples collected between March and May, 2007) at the suburban
Table 2 Levoglucosan concentrations as a function of particle size at both sampling locations (countryside and suburban site) Date
Levoglucosan conc. (ng m3) Countryside PM2.5
PM10a
Suburban TSP
PM2.5
PM10a
TSP
Harvest-season Background Dec. 18 195 Dec. 19 245 Dec. 20 237 Dec. 21 167
231 278 285 199
417 442 –b 334
309 371 237 332
316 378 245 342
331 390 253 352
Biomass burning Dec. 22 310 Dec. 23 687 Dec. 24 478
370 788 548
632 1138 737
456 1080 420
468 1104 437
701 1438 528
Post-burning Dec. 25 308 Dec. 26 346
344 395
398 747
252 471
258 485
404 633
a b
Determined by summation of PM2.5 and PM2.5–10 concentrations. Sampling failure.
Levoglucosan Conc. (ng m-3)
1200
B PM2.5
900
600
300
0
Harvest-Season Background
Biomass Burning
Post Buring
Off-Season Background
Fig. 2. Average levoglucosan concentrations in TSP (A) and PM2.5 (B) particles at both sampling locations during the December 2006 rice straw burning season, as well as the off-season background period. Error bars represent concentration ranges in the harvest-season background, burning episode, post-burning and off-season background periods in form of one standard deviation (n ¼ 4, 3, 2, and 12).
site was 133 ng m3 (Fig. 2). While tracer concentrations were notably enhanced (by a factor of 2–3 at the respective sites) during the active burning episode of rice straw relative to the seasonal background, levoglucosan levels in the post-burning period were found to still be rather high, likely because of ineffective dispersion, as well as continued burning of individual rice fields. Likewise, harvest-season background concentrations (measured from 18 to 21 December, 2006) of biomass burning tracers were rather high as well (levoglucosan concentrations were three times higher than the off-season background value, while PM mass concentrations were only 10% higher), but lower than those during the post-burning period. Again, this is most likely due to burning activities by individual rice farmers in the days (or weeks) prior to the intensive burning episode on the weekend of December 22, 2006. A possible reason for the rather small increase in PM mass concentrations (compared to the 3-fold increase in levoglucosan concentrations) between the background period and the episode with intensive burning is the dominant influence of various anthropogenic sources, including traffic emissions and industrial processes, as well as natural sources, on ambient PM levels in the central and southern part of Taiwan. Moreover, the winter months in this part of Taiwan, typically called the ‘‘dry’’ season, are characterized by relatively stagnant meteorological conditions, giving rise to accumulation of aerosol particles. The
8304
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
resulting aged aerosol consequently contains anhydrosugars from a variety of biomass burning activities. It is also noteworthy that levoglucosan is a selective tracer for biomass burning and is not emitted by other combustion processes, such as coal burning or different types of fossil fuel combustion. Thus, ambient levoglucosan concentrations are relatively more responsive to the influence of biomass burning activities compared to total aerosol mass concentrations. On the other hand, PM levels increased drastically (by approximately 70%) during the intensive burning episode, clearly indicating the substantial impact of post-harvest burning on ambient air quality. Fine particle (PM2.5) levoglucosan concentrations at the countryside and suburban site were highly elevated during the intensive burning period with average values of 492 (189) and 652 (371) ng m3, respectively, and 327 (27) and 362 (155) ng m3 in the post-burning period. Harvest-season background values at the two sites were 211 (38) and 312 (56) ng m3, respectively (Table 2). The off-season background concentration of levoglucosan at the suburban site (88 ng m3) was one order of magnitude lower than the observed values during the intensive burning period. The PM2.5 levoglucosan concentrations showed rather large variations during the intensive burning period, while exhibiting the same pattern as the TSP tracer concentrations, with peak values on December 23, 2006, when burning activities were most intense. During the intensive burning period, the day/night levoglucosan concentration ratios in PM2.5 were rather high with values of 3.9 and 1.9 at the countryside and suburban area, respectively (Fig. 3). The corresponding TSP ratios were 2.4 at the countryside and 1.9 in the suburban area. The day/night levoglucosan ratios during the post-burning and seasonal background periods were substantially lower (10 fraction was rather high with 35 (13) %, as shown in Fig. 4A. In contrast, the PM2.5 levoglucosan fraction was predominant in the suburban area, accounting for 81 (13) % of TSP on average (Fig. 4B). The levoglucosan measured during the period prior to the intensive burning episode at the suburban site was almost entirely present in the fine particle mode; 94% of the total levoglucosan concentration was found in PM2.5. The PM>10 fraction drastically increased during the intensive burning episode, both absolutely as well as relative to the fine fraction, from 3% in the background period to 25% of TSP during the burning episode.
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
8305
Table 3 Comparison of ambient fine-particle (PM2.5) levoglucosan concentrations from rice straw burning (this study) with levoglucosan emissions from combustion of other types of biomass (reported in the literature) Biomass type
Location (state/city, season)
Levoglucosan (ng m-3)
Reference
Mean (range) Rice straw Rice straw Wheat stubble Agricultural waste Agricultural waste Agricultural waste Agricultural waste Agricultural waste Forest (wildfire) Forest (wildfire)
(mixed) (mixed) (mixed) (mixed) (mixed)
Suburban
B
Suburban
0.6
1200
Harvest-Season Background
900 0.6
600
Harvest-Season Background
Biomass Burning
Dec. 26
Dec. 25
Dec. 24
Dec. 23
Dec. 22
Dec. 21
Dec. 20
Dec. 19
0
Dec. 18
300
Post Burning
Fig. 4. Timelines of ambient levoglucosan concentrations in three particle size ranges (PM2.5, PM10, and TSP) observed at the countryside (A) and suburban sites (B).
0.5
B
Biomass Burning
Dec. 26
1500
Countryside
0.7
Dec. 25
0
0.8
Dec. 24
300
A
0.9
Dec. 23
Conc. Ratio PM2.5 / PM10)
1.0
Dec. 22
600
Dec. 21
900
Dec. 20
TSP PM10 PM2.5
Dec. 19
Countryside
A
corresponding ratio at the suburban site, the PM2.5 levoglucosan fraction of PM10 at this site was still dominant (Fig. 5A). The slightly higher fine particle concentrations measured at the suburban site might be due to loss (by dry deposition) of a portion of PM2.5–10 during transport from rice fields north of Douliou through the urban area to the suburban sampling site on the south side of the city. Whereas the PM2.5/PM10 levoglucosan ratios were fairly consistent at both locations throughout the entire campaign, the PM>10/TSP ratios showed a broad variation in all periods (prior to, during, and after the intensive burning episode), particularly at the suburban site, as demonstrated in Fig. 5B. The harvest-season background PM>10/TSP ratios of the ambient levoglucosan concentrations were 0.41 (0.04) and 0.03 (0.01) at
Dec. 18
1200
Levoglucosan Conc. (ng m-3)
(This study) (This study) Viana et al., 2008 Jimenez et al., 2006 Rinehart et al., 2006 Park et al., 2006 Wan and Yu, 2007 Wang et al., 2007 He et al., 2006 Ward et al., 2006a Zdra´hal et al., 2002
Measured in PM10.
In terms of PM10, Fig. 5A illustrates the fine particle ratios (PM2.5/PM10) of the ambient levoglucosan concentrations at the countryside and suburban sites; the average PM2.5/PM10 ratio was 0.87 (0.02) and 0.97 (0.01) at the two sites, respectively. In agreement with the high PM2.5 levoglucosan fraction in TSP (as shown above), fine particle tracer concentrations clearly dominated the atmospheric PM10 composition in the suburban area (PM2.5/ PM10 ¼ 0.97). While the PM2.5/PM10 levoglucosan ratio at the countryside location was somewhat lower than the
Levoglucosan Conc. (ng m-3)
572 (310–1080) 262 (167–371) 74 (58–129)a 78 202 1754 190 (35–489) 480 (120–950) 78 3152 (1726–6091) 2006 (446–4106)
Conc. Ratio (PM> 10 / TSP)
a
Taiwan (Douliou, Burning peak) Taiwan (Douliou, Burning season) Spain (Valencia) U.S.A. (Washington and Idaho) U.S.A. (California/Fresno) Korea (Gwangju) China (Hong Kong, Winter) China (Guangzhou) China (Beijing, Winter) U.S.A. (Montana) Brazil (Rondoˆnia, dry season)
Post Burning
Countryside Suburban
0.4 0.3 0.2 0.1 0.0
Harvest-Season Background
Biomass Burning
Post Burning
Fig. 5. Illustration of PM2.5/PM10 (A) and PM>10/TSP (B) concentration ratios for levoglucosan throughout entire study period. Error bars represent ranges of concentration ratios in the harvest-season background, burning episode, and post-burning periods in form of one standard deviation (n ¼ 4, 3, and 2).
8306
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
the countryside and suburban sites, respectively (Fig. 5B). Surprisingly high levoglucosan concentrations were found in PM>10 at the countryside location during the entire study period. At the suburban site, the PM>10 tracer concentrations were rather low during the period prior to the intensive burning activities, but increased dramatically during the burning episode and continued to be elevated during the following days. Both laboratory and ambient measurements of smoke emissions from biomass burning processes have previously shown anhydrosugars to be predominantly present in the fine particle mode (Engling et al., 2006a; Herckes et al., 2006). On the contrary, the size-resolved measurements during this campaign revealed rather high biomass burning tracer concentrations in particles with aerodynamic diameters larger than 10 mm. As mentioned above, no sizeresolved ambient measurements of biomass burning tracers from rice straw burning have been reported to date. Thus, no direct comparison of such observations with other locations is possible. It is noteworthy that the burning practices of rice straw in Taiwan are different than those in other countries, as thin layers of straw and soil slowly burn together at relatively low temperatures. Such distinctly different burning conditions for rice straw (compared to other types of biomass) may result in soil and ash particles, which are typically more coarse, to be elevated during the burning process by the convective processes over the flame and subsequently be suspended into the air together with the smoke particles. In addition, the higher ambient relative humidity in Taiwan (compared to other regions with biomass burning processes, such as wildfires in the Western U.S. or Savanna fires in Africa) may result in substantial water up-take of rice straw smoke. Smoke particles from rice straw burning have been shown in laboratory experiments to be significantly more hygroscopic than most other types of biomass (Engling, unpublished data), enhancing the water up take potential of rice straw smoke. This unique feature of rice straw smoke particles together with other atmospheric process, including coagulation of smoke particles with ambient aerosols, may partially explain the observation of high PM2.5–10 and PM>10 tracer concentrations during burning of rice straw. Moreover, re-suspension of soil from rice fields by wind activity (wind speeds in Yunlin County can be up to 6–10 m s1) may constitute additional sources of PM2.5–10 and PM>10, containing the source tracers. Preliminary experiments in our laboratory have shown levoglucosan to be present in local soil. For instance, the abundance of levoglucosan in soil from rice fields in Yunlin County during spring (wet season) and winter time (dry season) of 2007 was measured at average values of 1.8 (0.75; n ¼ 6) and 8.7 (7.7; n ¼ 6) mg g1, respectively. Due to the high solubility of anhydrosugar tracers, it is expected that these measurements during the wet season constitute lower estimates of the tracer content in soil. Therefore, further investigations are in progress in order to quantify the tracer content in such particles, specifically during the winter post-harvest burning season, and to estimate the potential ambient contributions from the re-suspension of soil. On the other hand, the PM2.5–10 and PM>10 fractions observed in the suburban area during the background
period were rather low. This might be in part due to dry deposition of the larger particles during transport of aerosol particles from fields in the farming area north of the suburban sampling site. Also, re-suspension of soil did not play an important role in the contribution of PM>10 levoglucosan as was the case at the countryside location which was surrounded by cultivated land. While during the active burning period PM>10/TSP levoglucosan concentration ratios at the countryside were similar to the background period, the fraction of PM>10 levoglucosan increased drastically at the suburban site (from 0.03 to 0.25 0.08) during the burning episode, as demonstrated in Fig. 5B. Thus, the particle size distributions at the two locations (countryside and suburban site) showed closer resemblance during the burning episode. In the post-burning period, levoglucosan concentration ratios (PM>10/TSP) at the countryside and suburban areas were 0.31 (0.23) and 0.30 (0.09), respectively (Fig. 5B), indicating that levoglucosan concentration ratios at the two sites had reached some type of steady condition in the atmosphere. 4. Conclusions Biomass burning in general and agricultural burning activities in particular are important sources of aerosol particles on a global basis affecting local and regional air quality. Distinguishing different combustion source types, e.g., agricultural burning versus forest fires, is crucial, and thus detailed studies of specific burning processes are needed, both in the laboratory and in ambient air. To our knowledge, this is the first investigation of sized-resolved anhydrosugar distributions in ambient air resulting from post-harvest burning of rice fields in Asia. Significantly increased tracer concentrations were observed at two locations (rural and suburban) during the active burning episode within the post-harvest burning season of rice straw in Taiwan. Even higher levoglucosan levels were noticed at the suburban site relative to the countryside, showing the importance of local-scale (as well as regionalscale) transport of smoke aerosol. Interestingly, a vast mass fraction of levoglucosan was present in PM>10, possibly due to a variety of factors, including coagulation of fresh smoke particles, enhanced by the high relative humidity, as well as suspension of soil and ash particles during the combustion process of the rice straw. Fairly high background tracer concentrations were also observed prior to the intensive burning episode as well as outside the post-harvest burning season. This may be in part due to persistent contributions of smoke from small-scale fires and re-suspension of soil, containing smoke particles which have been accumulated over time. This study has further demonstrated the significance of rice straw burning practices on air quality by investigating ambient source tracer concentrations as a function of particle size. Our measurements were also applied to estimate source contributions to ambient PM2.5 based on emission factors from chamber studies. Furthermore, we have shown that the unique conditions associated with agricultural burning practices in subtropical regions of Asia may give rise to distinctly different chemical and physical characteristics of the resulting smoke aerosols compared to
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
chamber studies or ambient observations in other regions. The important new insights gained from this study provide motivation and direction for future research to better characterize biomass burning emissions, particularly from agricultural burning practices, and to generate source profiles that are representative of actual burning activities as well as the associated ambient conditions. Such work is in progress in our laboratory, including additional ambient observations, near-source measurements, and chamber studies. Acknowledgments Thanks are due to all the students from National Yunlin University of Science and Technology who helped collect the filter samples and perform the chemical analyses. The authors also acknowledge the helpful assistance by Mr. Sung-Tien Yang, principal of Jiouan Elementary School, and his colleagues during the sampling process. Financial support for this study came from the Engineering School of the National Yunlin University of Science and Technology (ENG 94-5) as well as the Research Center for Environmental Changes at Academia Sinica. References Dhammapala, R., Claiborn, C., Corkill, J., Gullett, B., 2006. Particulate emissions from wheat and Kentucky bluegrass stubble burning in eastern Washington and northern Idaho. Atmospheric Environment 40, 1007–1015. Dhammapala, R., Claiborn, C., Jimenez, J., Corkill, J., Gullett, B., Simpson, C., Paulsen, M., 2007. Emission factors of PAHs, methoxyphenols, levoglucosan, elemental carbon and organic carbon from simulated wheat and Kentucky bluegrass stubble burns. Atmospheric Environment 41, 2660–2669. Engling, G., Carrico, C.M., Kreidenweis, S.M., Collett Jr., J.L., Day, D.E., Malm, W.C., Hao, W.M., Lincoln, E., Iinuma, Y., Herrmann, H., 2006a. Determination of levoglucosan in biomass combustion aerosol by High Performance Anion Exchange Chromatography with Pulsed Amperometric Detection. Atmospheric Environment 40, S299–S311. Engling, G., Herckes, P., Kreidenweis, S.M., Collett Jr., J.L., 2006b. Composition of the fine organic aerosol in Yosemite National Park during the 2002 Yosemite Aerosol Characterization Study. Atmospheric Environment 40, 2959–2972. Gorin, C.A., Collett Jr., J.L., Herckes, P., 2006. Wood smoke contribution to winter aerosol in Fresno, CA. Journal of the Air & Waste Management Association 56, 1584–1590. Gullett, B.K., Touati, A., 2003. PCDD/F emissions from burning wheat and rice field residue. Atmospheric Environment 37, 4893–4899. Hays, M.D., Fine, P.M., Geron, C.D., Kleeman, M.J., Gullett, B.K., 2005. Open burning of agricultural biomass: physical and chemical properties of particle-phase emissions. Atmospheric Environment 39, 6747–6764. He, L.Y., Hu, M., Huang, X.F., Zhang, Y.H., Tang, X.Y., 2006. Seasonal pollution characteristics of organic compounds in atmospheric fine particles in Beijing. Science of the Total Environment 359, 167–176. Herckes, P., Engling, G., Kreidenweis, S.M., Collett Jr., J.L., 2006. Particle size distributions of organic aerosol constituents during the 2002 Yosemite Aerosol Characterization Study. Environmental Science & Technology 40, 4554–4562. Hornig, J.F., Soderberg, R.H., Barefoot, A.C., Galasyn, J.F., 1985. Wood smoke analysis: vaporization losses of PAH from filters and levoglucosan as a distinctive marker for wood smoke. In: Cooke, M., Dennis, A.J. (Eds.), Polynuclear Aromatic Hydrocarbons: Mechanisms, Methods, and Metabolism. Batelle Press, pp. 561–568. Jimenez, J., Wu, C.-F., Claiborn, C., Gould, T., Simpson, C.D., Larson, T., Liu, S.L.-J., 2006. Agricultural burning smoke in eastern Washington – part I: atmospheric characterization. Atmospheric Environment 40, 639–650. Kadam, K.L., Forrest, L.H., Jacobson, W.A., 2000. Rice straw as a lignocellulosic resource: collection, processing, transportation, and environmental aspects. Biomass & Bioenergy 18, 369–389.
8307
Keshtkar, H., Ashbaugh, L.L., 2007. Size distribution of polycyclic aromatic hydrocarbon particulate emission factors from agricultural burning. Atmospheric Environment 41, 2729–2739. Leithead, A., Li, S.M., Hoff, R., Cheng, Y., Brook, J., 2006. Levoglucosan and dehydroabietic acid: evidence of biomass burning impact on aerosols in the Lower Fraser Valley. Atmospheric Environment 40, 2721–2734. Li, X.G., Wang, S.X., Duan, L., Hao, J., Li, C., Chen, Y.S., Yang, L., 2007. Particulate and trace gas emissions from open burning of wheat straw and corn stover in China. Environmental Science & Technology 41, 6052–6058. Mazzoleni, L.R., Zielinska, B., Moosmuller, H., 2007. Emissions of levoglucosan, methoxy phenols, and organic acids from prescribed burns, laboratory combustion of wildland fuels, and residential wood combustion. Environmental Science & Technology 41, 2115–2122. Medeiros, P.M., Conte, M.H., Weber, J.C., Simoneit, B.R.T., 2006. Sugars as source indicators of biogenic organic carbon in aerosols collected above the Howland Experimental Forest, Maine. Atmospheric Environment 40, 1694–1705. Ogawa, M., Yoshida, N., 2005. Nitrous oxide emission from the burning of agricultural residue. Atmospheric Environment 39, 3421–3429. Otto, A., Gondokusumo, R., Simpson, M.J., 2006. Characterization and quantification of biomarkers from biomass burning at a recent wildfire site in Northern Alberta, Canada. Applied Geochemistry 21, 166–183. Park, S.S., Bae, M.S., Schauer, J.J., Kim, Y.J., Cho, S.Y., Kim, S.J., 2006. Molecular composition of PM2.5 organic aerosol measured at an urban site of Korea during the ACE-Asia campaign. Atmospheric Environment 40, 4182–4198. Puxbaum, H., Caseiro, A., Sanchez-Ochoa, A., Kasper-Giebl, A., Claeys, M., Gelencser, A., Legrand, M., Preunkert, S., Pio, C., 2007. Levoglucosan levels at background sites in Europe for assessing the impact of biomass combustion on the European aerosol background. Journal of Geophysical Research – Atmospheres 112, Art. No. D23S05. doi: 10.1029/2006JD008114. Rinehart, L.R., Fujita, E.M., Chow, J.C., Magliano, K., Zielinska, B., 2006. Spatial distribution of PM2.5 associated organic compounds in central California. Atmospheric Environment 40, 290–303. Schmidl, C., Marr, L.L., Caseiro, A., Kotianova, P., Berner, A., Bauer, H., KasperGiebl, A., Puxbaum, H., 2008. Chemical characterization of fine particle emissions from wood stove combustion of common woods growing in mid-European Alpine regions. Atmospheric Environment 42, 126–141. Sheesley, R.J., Schauer, J.J., Chowdhury, Z., Cass, G.R., Simoneit, B.R.T., 2003. Characterization of organic aerosols emitted from the combustion of biomass indigenous to South Asia. Journal of Geophysical Research 108, Art. No. 4285. doi: 10.1029/2002JD002981. Simoneit, B.R.T., Schauer, J.J., Nolte, C.G., Oros, D.R., Elias, V.O., Fraser, M.P., Rogge, W.F., Cass, G.R., 1999. Levoglucosan, a tracer for cellulose in biomass burning and atmospheric particles. Atmospheric Environment 33, 173–182. Streets, D.G., Yarber, K.F., Woo, J.H., Carmichael, G.R., 2003. Biomass burning in Asia: annual and seasonal estimates and atmospheric emissions. Global Biogeochemical Cycles 17, Art. No. 1099. Venkataraman, C., Habib, G., Eiguren-Fernandez, A., Miguel, A.H., Friedlander, S.K., 2005. Residential biofuels in south Asia: carbonaceous aerosol emissions and climate impacts. Science 307, 1454– 1456. Viana, M., Lopez, J.M., Querol, X., Alastuey, A., Garcia-Gacio, D., BlancoHeras, G., Lopez-Mahia, P., Pineiro-Iglesias, M., Sanz, M.J., Sanz, F., Chi, X., Maenhaut, W., 2008. Tracers and impact of open burning of rice straw residues on PM in Eastern Spain. Atmospheric Environment 42, 1941–1957. Wan, E.C.H., Yu, J.Z., 2007. Analysis of sugars and sugar polyols in atmospheric aerosols by chloride attachment in liquid chromatography/ negative ion electrospray mass spectrometry. Environmental Science & Technology 41, 2459–2466. Wang, Q.Q., Shao, M., Liu, Y., Kuster, W., Goldan, P., Li, X.H., Liu, Y.A., Lu, S. H., 2007. Impact of biomass burning on urban air quality estimated by organic tracers: Guangzhou and Beijing as cases. Atmospheric Environment 41, 8380–8390. Ward, T.J., Hamilton, R.F., Dixon, R.W., Paulsen, M., Simpson, C.D., 2006a. Characterization and evaluation of smoke tracers in PM: results from the 2003 Montana wildfire season. Atmospheric Environment 40, 7005–7017. Ward, T.J., Rinehart, L.R., Lange, T., 2006b. The 2003/2004 Libby, Montana PM2.5 source apportionment research study. Aerosol Science and Technology 40, 166–177. Yang, H.H., Tsai, C.H., Chao, M.R., Su, Y.L., Chien, S.M., 2006. Source identification and size distribution of atmospheric polycyclic aromatic hydrocarbons during rice straw burning period. Atmospheric Environment 40, 1266–1274.
8308
J.J. Lee et al. / Atmospheric Environment 42 (2008) 8300–8308
Yevich, R., Logan, J.A., 2003. An assessment of biofuel use and burning of agricultural waste in the developing world. Global Biogeochemical Cycles 17, Art. No. 1095. Yoshinori, M., Kanno, T., 1997. Emissions of trace gases (CO2, CO, CH4, and N2O) resulting from rice straw burning. Soil Science and Plant Nutrition 43, 849–854. Yttri, K.E., Dye, C., Slordal, L.H., Braathen, O.A., 2005. Quantification of monosaccharide anhydrides by liquid chromatography combined with mass spectrometry: application to aerosol samples from an urban and a suburban site influenced by small-scale wood
burning. Journal of the Air & Waste Management Association 55, 1169–1177. Zdrahal, Z., Oliveira, J., Vermeylen, R., Claeys, M., Maenhaut, W., 2002. Improved method for quantifying levoglucosan and related monosaccharide anhydrides in atmospheric aerosols and application to samples from urban and tropical locations. Environmental Science & Technology 36, 747–753. Zhang, Y.X., Shao, M., Zhang, Y.H., Zeng, L.M., He, L.Y., Zhu, B., Wei, Y.J., Zhu, X.L., 2007. Source profiles of particulate organic matters emitted from cereal straw burnings. Journal of Environmental Sciences – China 19,167–175.
