Article pubs.acs.org/est
Impacts of Siberian Biomass Burning on Organic Aerosols over the North Pacific Ocean and the Arctic: Primary and Secondary Organic Tracers Xiang Ding,† Xinming Wang,*,† Zhouqing Xie,‡ Zhou Zhang,† and Liguang Sun‡ †
State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, People’s Republic of China, ‡ School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, People’s Republic of China S Supporting Information *
ABSTRACT: During the 2003 Chinese Arctic Research Expedition (CHINARE2003) from the Bohai Sea to the high Arctic (37°N− 80°N), filter-based particle samples were collected and analyzed for tracers of primary and secondary organic aerosols (SOA) as well as water-soluble organic carbon (WSOC). Biomass burning (BB) tracer levoglucosan had comparatively much higher summertime average levels (476 ± 367pg/m3) during our cruise due to the influence of intense forest fires then in Siberia. On the basis of 5-day back trajectories, samples with air masses passing through Siberia had organic tracers 1.3−4.4 times of those with air masses transporting only over the oceans, suggesting substantial contribution of continental emissions to organic aerosols in the marine atmosphere. SOA tracers from anthropogenic aromatics were negligible or not detected, while those from biogenic terpenenoids were ubiquitously observed with the sum of SOA tracers from isoprene (623 ± 414pg/m3) 1 order of magnitude higher than that from monoterpenes (63 ± 49 pg/m3). 2-Methylglyceric acid as a product of isoprene oxidation under high-NOx conditions was dominant among SOA tracers, implying that these BSOA tracers were not formed over the oceans but mainly transported from the adjacent Siberia where a high-NOx environment could be induced by intense forest fires. The carbon fractions shared by biogenic SOA tracers and levoglucosan in WSOC in our ocean samples were 1−2 orders of magnitude lower than those previously reported in continental samples, BB emissions or chamber simulation samples, largely due to the chemical evolution of organic tracers during transport. As a result of the much faster decline in levels of organic tracers than that of WSOC during transport, the trace-based approach, which could well reconstruct WSOC using biogenic SOA and BB tracers for continental samples, only explained ∼4% of measured WSOC during our expedition if the same tracer-WSOC or tracer-SOC relationships were applied.
■
condensation nuclei and influence the cloud droplet number.8 More complex is that biogenic SOA (BSOA) over oceans can be not only produced by the oxidation of BVOCs from oceanic emissions, but also, more importantly, transported from the adjacent continents. It is an interesting topic to distinguish transported BSOA from those locally formed over the marines. Specific SOA tracers can provide insight on precursors and processes influencing SOA productions. For example, although isoprene has the largest emission compared to other VOCs,4 SOA formation from isoprene photo-oxidation had long been thought to be negligible due to its low SOA yield 9 until its SOA tracers, 2-methyltetrols, were first identified in the Amazonian rain forest aerosols.1 At the moment, the global
INTRODUCTION Secondary organic aerosols (SOA) affect the earth’s radiation balance and regional air quality. They are produced by homogeneous 1 and heterogeneous 2 reactions of volatile organic compounds (VOCs) as well as aging of primary aerosols.3 Biogenic VOCs (BVOCs), such as isoprene and monoterpenes, are important SOA precursors with their global emissions 4 1 order of magnitude higher than those of anthropogenic sources.5 Oceanic emissions of BVOCs, however, were considered to be very minor as compared to the global burdens 4 and therefore were supposed to have no significant contributions to marine aerosols.6 Nevertheless, global oceanic emissions of BVOCs are subjected to large uncertainties. For instance, estimated global oceanic α-pinene emissions differed 3 orders of magnitude between “bottom-up” and “top-down” methods.7 A study in the Southern Ocean also found that SOA from phytoplankton-produced isoprene could profoundly affect chemical compositions of marine cloud © 2013 American Chemical Society
Received: Revised: Accepted: Published: 3149
September 13, 2012 February 17, 2013 February 26, 2013 February 26, 2013 dx.doi.org/10.1021/es3037093 | Environ. Sci. Technol. 2013, 47, 3149−3157
Environmental Science & Technology
Article
Table 1. Summary of Tracers (pg/m3) and WSOC during CHINAREN2003 average
median
min
max
SDa
LEVO
476
421
57
2131
367
MTHB1 MTHB2 MTHB3 MGA MTL1 MTL2
2.3 8.7 36 430 59 87 47 145 623
ndb 5.6 21 305 50 73 29 127 508
nd nd 2.6 66 14 14 2.6 30 115
15 46 215 1258 208 336 232 544 1494
3.5 9 41 362 41 65 48 101 414
43 19 1.1 63
35 8.8 nd 48
6.4 nd nd 9.0
159 145 18 250
29 28 3.7 49
2.4 1164 222 0.117 0.149 0.023 0.001 0.290
0.6 1065 179 0.091 0.142 0.019 0.001 0.291
nd 246 43 0.033 0.044 0.005 nd 0.100
30.6 3767 975 0.391 0.320 0.074 0.011 0.689
5.4 707 178 0.095 0.078 0.016 0.002 0.153
abbrev biomass burning tracer levoglucosan isoprene SOA tracers cis-2-methyl-1,3,4-trihydroxy-1-butene 3-methyl-2,3,4-trihydroxy-1-butene trans-2-methyl-1,3,4-trihydroxy-1-butene 2-methylglyceric acid 2-methylthreitol 2-methylerythritol C5−alkene triolsc 2-methyltetrolsd sum of isoprene SOA tracers monoterpene SOA tracers cis-pinonic acid 3-hydroxyglutaric acid 3-methyl-1,2,3-butanetricarboxylic acid sum of monoterpene SOA tracers β-caryophyllene SOA tracer β-caryophyllenic acid sum of all organic tracers WSOC (ngC/m3) levoglucosan/WSOC (%)e isoprene SOA tracers/WSOC (%) monoterpeneSOA tracers/WSOC (%) β-caryophyllenic acid/WSOC (%) total tracers/WSOC (%)
MTLs
PNA HGA MBTCA
CARY
a
SD is one standard deviation. bnd means not detect. cSum of cis-2-methyl-1,3,4-trihydroxy-1-butene, 3-methyl-2,3,4-trihydroxy-1-butene and trans2-methyl-1,3,4-trihydroxy-1-butene. dSum of 2-methylthreitol and 2-methylerythritol. eCarbon fraction.
Both SOA and BB aerosols contain large amounts of oxygenated and polar compounds and are thus closely associated with water-soluble organic carbon (WSOC). Previous study found that WSOC could be fully explained by the sum of BB-induced WSOC (WSOC BB ) and the unapportioned OC by chemical mass balance model 23 which were typically regarded as secondary organic carbon (SOC).24 Both SOC and WSOCBB can be estimated by tracer-based approach12,13 assuming that in the ambient these organic tracers are stable and the tracer-SOC or tracer-WSOC relationships remain the same as those from chamber simulations or in BB emissions. In the very aged air with residence time over several days, such a tracer-based approach should be challenged, since chemical evolution during transport would change not only the tracer concentrations,but also the tracer−OC relationships. In the current study, we took advantage of a research expedition aboard the icebreaker Xuelong from the Bohai Sea to the North Pole area (37°N−80°N) during July−September 2003, and collected filter-based particle samples to study the chemical compositions and spatial distribution of SOA and BB tracers over a large latitudinal range. Particular concerns were put on the enhancement of both primary and secondary OA over the oceans by the Siberian forest fires and chemical evolution of organic tracers during transport.
SOA production from isoprene is estimated at 19.2TgC/yr, accounting for 79% of total SOA.10 Large-scale survey of these SOA tracers can provide important information about the spatial distribution and origins of SOA. Until the present, SOA tracers available in the marine air were only recently retrieved from samples collected onboard around-the-world cruise during October 1989 to March 1990.11 While data of SOA tracers are accumulating in the continents,1,12−15 more information is also needed for their occurrence over oceans. Biomass burning (BB) is an important source of organic aerosols. On the global scale, it contributed more than 90% of organic carbon (OC) released to the atmosphere.16 Previous studies found that forest fires in Russia contributed substantially to the Arctic aerosols.17 In 2003, as satellites recorded, there were great forest fires in Siberia 18 and smoke from these fires had been streaming across eastern Russia and out over the Pacific Ocean off and on for months, and reaching the Arctic.19 Ding et al. 20 confirmed that BB was the major source of polycyclic aromatic hydrocarbons in the marine aerosols over the North Pacific Ocean during summer 2003. As BB can directly produce large amounts of VOCs and oxidants,21 fire events could indirectly enhance the formation of SOA in the atmosphere.22 Yet there is quite limited knowledge about the enhancement of SOA formation by BB. The intense forest fires in Siberia during 2003 should have significant impact on the marine aerosols over the adjacent North Pacific Ocean, which could be investigated by ship-based observation then over the North Pacific, as the research expedition aboard the icebreaker Xuelong from Bohai Sea to the Arctic during July−September 2003.
■
EXPERIMENTAL SECTION
Field Sampling and Chemical Analysis. During the Xuelong expedition, 49 total suspended particles (TSP) samples as well as three field blanks were collected using quartz fiber filters (QFF) between Bohai Sea and the Arctic (37.78 oN−
3150
dx.doi.org/10.1021/es3037093 | Environ. Sci. Technol. 2013, 47, 3149−3157
Environmental Science & Technology
Article
Figure 1. Comparisons between ocean and land origin samples.
as field samples. Target organic tracers were not detected in the blanks. Recoveries of the target compounds in six spiked samples (authentic standards spiked into solvent with prebaked quartz filter) were 104 ± 2% for cis-pinonic acid, 78 ± 13% for levoglucosan, 62 ± 14% for erythritol, and 78 ± 10% for octadecanoic acid. The relative differences for target compounds in paired duplicate samples (n = 6) were all
