- Email: [email protected]
Aerosol ion concentration dependence on atmospheric conditions in Chicago
ARTICLE IN PRESS
Atmospheric Environment 40 (2006) 6638–6649 www.elsevier.com/locate/atmosenv
Aerosol ion concentration dependence on atmospheric conditions in Chicago Tinamarie Fosco, Martina Schmeling Loyola University Chicago, 6525 N. Sheridan Rd. Chicago, IL 60626, USA Received 13 September 2005; accepted 18 May 2006
Abstract This study seeks to determine the influence of precursor trace gases and local meteorology, including lake breeze events, on the concentrations of secondary aerosol species in Chicago. For this, two particulate air samples per day were collected onto quartz fiber filters at the Loyola University Chicago Air Station (LUCAS) during the summer months of 2002 and 2003 and subsequently analyzed by ion chromatography for sulfate, nitrate and oxalate. In parallel, mixing ratios of ozone (O3) and nitrogen oxides (NO and NO2QNOx) were monitored and weather parameters were recorded. In addition, backward trajectories were obtained to estimate air mass transport to Chicago. Ozone and NOx mixing ratios as well as sulfate, nitrate and oxalate concentrations varied substantially throughout the study, but three situations could be distinguished based on meteorology and chemistry. Case one had the lowest ozone, NOx and ion levels due to wind directions constantly from Lake Michigan. Case two comprised of days showing the highest pollutant levels because of to predominantly southwestern air currents and warm temperatures and case three experienced an air stagnation situation in the morning leading to high NOx mixing ratios and a subsequent lake breeze event. In the last case, elevated ozone mixing ratios and ion concentrations were observed after lake breeze onset indicating pollutant transport. r 2006 Elsevier Ltd. All rights reserved. Keywords: Ion chromatography; Lake breeze; Urban air quality; Secondary aerosol formation; Pollutant transport
1. Introduction Chicago as a major metropolitan center in the Midwest presents a significant source region of emissions from traffic and industry and experiences regularly air pollution events during hot summer days. In addition, Chicago’s location at the southwestern end of Lake Michigan exposes the city to lake breezes where the major air current originates Corresponding author. Tel.: +1 773 508 3124, fax: +1 773 508 3086. E-mail address: [email protected] (M. Schmeling).
1352-2310/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2006.05.061
from Lake Michigan. It was found that these events might play a major role in the pollution pattern of the region (see for example: Bloomfield et al., 1996; Gatz, 1975; Lee et al., 1993; Lestari et al., 2003; Scheff et al., 1984). In order to investigate these pollution patterns in more detail, aerosol samples were collected during the summer months of 2002 and 2003 and analyzed for their ionic content. The major ions of interest comprised of sulfate, nitrate and oxalate, because these ions are considered secondary pollutants formed as a result of reactions by reactive trace gases such as sulfur dioxide (SO2), nitrogen oxides (NO, NO2) and volatile organic
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
compounds (VOCs) with ozone (O3) and hydroxyl radicals (dOH). Studies have found that the formation of secondary pollutants does not only depend on emission sources, but is also influenced by local meteorology. For example, high ozone episodes have been observed in areas downwind of pollution sources with high temperatures and low wind speeds (Aneja et al., 1999; Bloomfield et al., 1996; Chow et al., 1998). Also, studies have linked elevated ozone levels to sea or lake breezes in regions close to large bodies of water (Cass and Shair, 1984; Gusten et al., 1988; Hastie et al., 1999; Lyons and Cole, 1976; Nair et al., 2002; Pont and Fontan, 2000). Both situations for formation of secondary aerosols have to be considered in Chicago with major emission sources primarily located in the south (S) and southwestern (SW) part of the city and lake breezes occurring at about 20% of summer days. A lake breeze is the result of a temperature gradient between land and lake surfaces. During hot summer days, the land surface heats faster than the water body creating a temperature and pressure gradient between these two. The difference in pressure leads to air mass transport from the high pressure area over the lake (cooler air) to the lower pressure area over the land (warmer air). During the night, the opposite so-called land-breeze can occur, which transports cooler polluted air from the city towards and above Lake Michigan. An air mass experiencing a lake or land breeze event is confined to a relatively small volume with little mixing and trace gases and particulates can become highly concentrated. A detailed description of the lake breeze climatology with respect to Chicago has been published by Lyons (1972). The study presented here seeks to investigate the conditions leading to formation of the secondary aerosol species, sulfate, nitrate and oxalate, in conjunction with ozone and NOx mixing ratios as well as local meteorology including the lake breeze in Chicago. 2. Methods 2.1. Sample collection Atmospheric aerosol samples were collected at Loyola University Chicago’s air station (LUCAS). LUCAS is situated 60m high atop a residence hall on Loyola University Chicago’s lake shore campus located approximately 13 km north of downtown
6639
Chicago and 200m west of Lake Michigan (Fig. 1). Sampling took place in 2002 (13 days) and 2003 (20 days) during the summer because the highest frequency of lake breeze events and the highest possible ozone mixing ratios and ion concentrations are expected at this time of the year. Particulates (PM2.5) were collected at a flowrate of 2 m3 h 1 onto quartz fiber filters (QMA, 47 mm diameter, Whatman International Ltd, Maidstone, UK) housed inside a Teflon filter holder. Samples were taken twice a day, from 07:00 h to 10:00 h (collection A), and 11:00 h to 13:00 h (collection B) local time (LT). Observations indicated that a lake breeze most likely establishes between 10:00 and 11:00 h; hence, the collection times selected allowed for the determination of particulate concentrations before and during the onset of a lake breeze. Immediately after collection, filters were frozen until analysis. In addition to air particulate collection, continuous 24 h trace gas measurements of NOx (NO and NO2) and ozone were made using chemiluminescence for NOx (Thermo Environmental Instruments, Inc. (TEI) Model 42C; US EPA designated method RFNA-1289-074, US EPA, 2005) and UV photometry (TEI Model 49C; US EPA designated method EQOA-0880-047, US EPA, 2005) for ozone, respectively. Both units were calibrated regularly with aid of the local Illinois Environmental Protection Agency’s bureau of air. Weather conditions were also recorded during the sampling time from 07:00 to 13:00 h. 2.2. Sample analysis Before analysis, filters were extracted in 5 ml of nanopure water for 15 min by ultrasonication. The obtained solution was analyzed subsequently for anions and low molecular weight organic acids by ion chromatography (IC) with chemical suppression and conductivity detection (Metrohm 761 Compact IC). The analysis procedure followed closely previously described ones with slight adaptations (Granby et al., 1997; Grosjean, 1988; Kadowski, 1986; Kato et al., 1999; Lin, 2002; Solomon et al., 1992; Tolocka et al., 2001). A Metrosep Supp5 column with 1.0mM NaHCO3 and 3.5mM Na2CO3 as eluent was used to determine fluoride, acetate, formate, chloride, nitrite, bromide, nitrate, phosphate, benzoate, sulfate, succinate, and oxalate. Sulfate, nitrate and oxalate accounted for the most abundant and continuously detected anions in the aerosol samples.
ARTICLE IN PRESS 6640
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
Fig. 1. Regional map of the metropolitan Chicago area (Mapquest.com, 2006). The measurement station (LUCAS) is located north of downtown Chicago in vicinity of Lake Michigan and indicated as a black dot. The majority of industrial sources are present in the southwest and south of the city. These sources include brick and cement works, steel manufacturing and metal processing as well as paper mills. One major airport (O’Hare International Airport) is located about 15 km to the west and a second major airport (Midway Airport) is located about 15 km to the southwest of LUCAS.
Clean quartz fiber filters were analyzed to determine the filter blank. It was observed that concentrations for some ions were high and variable on the blank filters, whereas some blanks were below detection limit. For instance, the blank concentrations for sulfate and nitrate were 12.8715.7 and 1.0670.702 mg/filter, respectively, but oxalate was found below detection limit on the blank filters. Blank values were subtracted from the ion concentrations in the samples and reported as 7standard deviation of the filter blank per volume of air sampled during collection. A detailed description of the method validation can be found in Fosco and Schmeling, 2006 and Fosco, 2006. All values presented here are based on a student t-test with 95% confidence level (Skoog et al., 2000).
2.3. Data analysis Sulfate, nitrate and oxalate concentrations varied substantially between collection days such that three different situations describing their formation could be distinguished. The collection days for each case experienced comparable ozone and NOx mixing ratios, and meteorology, but these parameters differed substantially between the cases. Also, the frequency of lake breeze events and its influence on pollutant transport varied among the cases. A lake breeze was identified by an initial wind direction from the land (N, NW, W, SW, S) and a later shift towards wind directions from Lake Michigan (NE, E, SE).
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
Also, backward air trajectories were calculated using the NOAA HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) model (Draxler and Rolph, 2003) to estimate the long-range transport of air masses to Chicago. However, the HYSPLIT model could not always identify the local scale lake breeze phenomenon because of its large grid size. 3. Case study results and discussion 3.1. Description of relevant atmospheric chemistry In order to account for the ion concentrations determined in each case, a brief description is given
6641
of the relevant chemical reactions. Details regarding these reactions and their kinetics can be found in Finlayson-Pitts and Pitts (2000), Chameides et al. (1992), Jenkin and Clemitshaw (2000), Jacob (2000), Hoffman, 1986, and Walcek et al. (1997). The reaction sequence leading to nitrate and sulfate aerosol formation is seen in Fig. 2 and involves the production of ozone (a) via photolysis of NO2, with NO2 being directly emitted into the atmosphere by industrial activity and vehicular traffic. Ozone production predominates during the day and in the summer when solar intensity is greatest. Without presence of other species a photostationary state is established cycling ozone, NO and NO2 according to reaction sequences (a) and (b). Ozone can be
Fig. 2. Relevant tropospheric chemistry reactions.
ARTICLE IN PRESS 6642
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
photolytically destroyed as seen in reaction sequence (c), presenting a major pathway for hydroxyl radical formation in urban areas. NOx and hydroxyl radicals also form due to reactions with VOCs as highlighted in reactions (d) and (e). Hydroperoxy radicals produce hydrogen peroxide, another major oxidant in the atmosphere, as in reaction (f). During the day, nitrogen dioxide is removed from the reaction cycle forming ozone (a) by oxidation with hydroxyl radicals yielding nitric acid (g). Ozone can also react with sulfur dioxide in a heterogeneous reaction on water droplets to form sulfuric acid (h). Another possible pathway, though of minor importance, for sulfuric acid formation presents the oxidation of sulfur dioxide with hydrogen peroxide. Finally, ammonium sulfate (or ammonium bisulfate) and ammonium nitrate particles can result when gas phase ammonia is present to neutralize these species. Carboxylic acids, such as oxalic acid, found in urban aerosols are formed as a result of complex and not yet fully understood oxidation pathways involving VOCs, ozone and hydroxyl radicals (Chebbi and Carlier, 1996; Jacobson et al., 2000; Khwaja, 1995). Therefore the presence of these species is mentioned in the following discussion, but no reaction scheme suggested. 3.2. Case one: clean air conditions The first case is characterized by days with the lowest NOx and ozone mixing ratios, as well as the lowest sulfate, nitrate and oxalate concentrations, therefore can be identified as ‘‘all low case’’. Five
days representative for case one are discussed in the following section and Table 1 shows the weather parameter and trace gas mixing ratios during the aerosol collection times for these days. As can be seen, NOx mixing ratios were very low with values less than 10 ppb and ozone mixing ratios were less than 45 ppb and remained relatively constant with no distinguishable diurnal pattern. Fig. 3(a) displays a typical photochemical pattern, while Fig. 4(a) shows a typical air mass pattern in the form of the HYSPLIT backward trajectory. Air masses originated from the north of LUCAS and traveled a substantial time across rural areas and water. Not surprising, the case one days also experienced the lowest temperatures of all three cases, averaging only 20 1C. The average wind speed (4.8 m s 1) in case one was moderate, while the average relative humidity (67.1%) ranked highest when compared to the other two cases. Fig. 5(a) displays the mean concentrations with standard deviations (Student ttest, 95% confidence level) of sulfate, nitrate and oxalate concentrations for case one days. The overall average values were 6.1973.14, 1.1570.14, and 0.11870.0 mg m 3 for the three respective ions. The low trace gas mixing ratios and ion concentrations observed can likely be attributed to the wind direction and low temperature on these days. The samples were collected from air masses arriving from the northeast where pollution sources are scarce and any long-range pollution would have traveled a substantial distance across Lake Michigan and be most likely deposited into the lake. This is most pronounced when north or northeast winds are predominant as on 18 July 2003 for example. On
Table 1 Case one weather data averaged for each collection (A: 07:00–10:00 h, B: 11:00–13:00 h), and ozone and NOx mixing ratios at start and end of each collection Day
24 July 2002
2 August 2002
6 August 2002
7 August 2002
18 July 2003
Wind direction
ESE/E (10:30 h) A: 4.7 B: 2.8 A: 19.8 B: 20.4 A: 57 B: 63 A: 5, 6 B: 5, 4 A: 35, 30 B: 34, 35
NNE
E
ESE
A: 7.1 B: 6.2 A: 21.7 B: 22.0 A: 76 B: 66 A: 3, 3 B: 3, 3 A: 45, 38 B: 37, 36
A: 6.1 B: 4.7 A: 19.0 B: 19.7 A: 70 B: 68 A: 2, 4 B: 3, 1 A: 18, 24 B: 26, 28
A: 4.0 B: 2.6 A: 19.4 B: 20.3 A: 58 B: 59 A: 2, 5 B: 5, 2 A: 25, 28 B: 27, 34
N-NE Lake Breeze 11:00 h A: 4.0 B: 5.7 A: 18.5 B: 19.4 A: 77 B: 77 A: 1, 3 B: 3, 2 A: 29, 30 B: 31, 42
Average wind speed (m s 1) Average temperature (oC) % Average % relative humidity NOx at times A: 07:00 h, 10:00 h B: 11:00 h, 13:00 h (ppb) O3 at times A: 07:00 h, 10:00 h B: 11:00 h, 13:00 h (ppb)
All times are shown as local times (LT).
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
6643
200 O3
180
200
NOx
180
160
120
100
ppb
80
100
Time
22:00
20:00
18:00
16:00
14:00
12:00
10:00
8:00
6:00
0:00
22:03
20:03
18:02
16:02
14:02
12:02
10:01
8:01
6:00
0 4:00
20
0 2:00
40
20 0:00
60
40
4:00
80
60
2:00
ppb
NOx
140
120
(a)
O3
160
140
Time
(b)
200 O3
180
NOx
160 140 ppb
120 100 80 60 40 20
(c)
22:02
20:02
18:02
16:02
14:01
12:01
10:01
8:01
6:01
4:00
2:00
0:00
0
Time
Fig. 3. 24-hour ozone and NOx mixing ratios on: 7 August 2002 (a); 25 June 2003 (b); and 14 July 2003 (c).
three of the five days described here under case one, namely 24 July, 6 August, and 7 August 2002, substantially higher sulfate and nitrate concentrations were observed during the second collection period. The percent change between collection A and B based on the mean values, were +125% (sulfate) and +256% (nitrate) on 24 July +58% (sulfate) and +272% (nitrate) on 6 August and +54% (sulfate) and +261% (nitrate) on 7 August 2002. In contrast the changes between collections A and B on 2 August 2002 and 18 July 2003 were +57% (sulfate) and +7%(nitrate) as well as 72% (sulfate) and +183% (nitrate), respectively. Weather data for the first three days indicate that the air masses did not originate from the northeast, but rather from the southeast and hence were exposed to the industrial area around Gary, IN and the southern Michigan shoreline leading to the elevated ion concentrations observed. Interestingly, the
backtrjectories for the first three days mentioned show a similar pattern with air masses arriving from the east and originating north of Michigan as shown in Fig. 4(a). In contrast, the air masses on the latter two days did have their origin west of the station with subsequently moving onto Lake Michigan in Wisconsin, thus circumventing the metropolitan Chicago area before swinging back towards the city with easterly air currents as seen in Fig. 4(b). The pattern these two days followed corresponds to low emissions as the region traversed is mostly rural and agricultural. Nevertheless, in contrast to cases two and three the average ion concentrations were much lower in case one, which might be explained by the following pattern: With the lack of pollution sources, NOx is limited and ozone destruction (reaction (c) in Fig. 2) is favored (Walcek et al., 1997). Also, VOCs are expected to be low causing a further restriction of secondary NOx formation in
ARTICLE IN PRESS 6644
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
Fig. 4. Three-day backward isentropic air trajectories from an altitude of 100m at Chicago starting at 07:00 h (CDT): (a) 7 August 2002 (b) 18 July 2003 (c) 25 June 2003 (d) 1 July 2003.
reaction (d) and ozone production in (a). Correspondingly, formation of the secondary aerosols, sulfate, oxalate and nitrate is suppressed yielding the low concentrations observed. Moderate wind speeds kept the air diluted and cool temperatures, further limited NOx photolysis and, thus ozone and ionic aerosol formation. Only one lake breeze day was recognized in case one. Weather data indicates that on 18 July 2003, an air mass arrived from the north during the early morning hours followed by a shift in wind direction at 11:00 h to northeast winds. Since major pollution sources are lacking, air masses following these
directions are typically clean. However, higher nitrate concentrations were observed on 18 July 2003, compared to the other days in case one. This can likely be attributed to the return of pollutants with the lake breeze and will be discussed further in case three. 3.3. Case two: pollution episodes The second case is characterized by days with the highest ozone mixing ratios, and sulfate and oxalate concentrations and can be classified as the ‘‘high pollutant case’’. The weather parameter, and trace
ARTICLE IN PRESS 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
sulfate nitrate oxalate ug/m^3
ug/m^3
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
A
A
B
A
B
A
B
A
B
July 24, August 2, August 6, August 7, July 18, 2002 2002 2002 2002 2003
ug/m^3
(a)
B
(c)
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
sulfate nitrate oxalate
A
B
July 30, 2002
(b)
6645
A
B
A
B
A
B
August 1, August 12, June 24, 2003 2002 2002
A
B
June 25, 2003
sulfate nitrate oxalate
A B August 9, 2002
A B July 1, 2003
A B July 14, 2003
A B July 30, 2003
A B July 31, 2003
Fig. 5. Sulfate, nitrate and oxalate concentrations for: case one (a); case two (b); and case three (c). Oxalate concentrations were increased by a factor of 15 for visual purposes. (collection A: 07:00–10:00 h, collection B: 11:00–13:00 h).
gas mixing ratios for the five representative days of case two are shown in Table 2. Fig. 3(b) displays the typical trace gas mixing ratios for comparison to the other two cases. NOx mixing ratios were moderate with 30 and 60 ppb maxima between 06:00 to 09:00 h. Ozone mixing ratios were highest with 100 ppb maxima just after 13:00 h and had a distinct diurnal pattern. The average temperature for case 2 days (26.9 1C) was the highest observed for all cases; average wind speed was slightly higher than for case one with 5.62 m s 1 and relative humidity was below case one, averaging 59.7%. The wind direction was predominantly southwest except for 30 July 2002 experiencing initial winds from the northwest and later from the southeast. A typical air mass pattern is depicted in Fig. 4(c). In Fig. 5(b), ion concentrations for case two are shown in comparison with the two other cases discussed. Sulfate and oxalate concentrations were highest for all three cases
averaging 15.873.2 and 0.30170.0 mg m 3, respectively. Nitrate concentrations were found to be 7.2570.14 mg m 3 in average and thus, ranked between case one and case three. The high pollution conditions found for case two are not surprising when considering that the main wind direction was southwest on almost all days. Most industrial activities are concentrated in the southwest of Chicago, and reactive trace gases (NOx, VOCs and SO2) emitted in this area are transported by southwest air currents across the metropolitan area to the measurement station. Therefore, sufficient precursor species can be present and the reaction patterns described in Fig. 2 are not restricted as in case one. Also, the high temperatures experienced and the abundance of solar radiation in summer enhance the photolysis of NO2 and thus, ozone production via reaction (a). During the air mass transport time, photochemical
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
6646
Table 2 Case two weather data averaged for each collection (A: 07:00–10:00 h, B: 11:00–13:00 h), and ozone and NOx mixing ratios at start and end of each collection Day
30 July 2002
1 August 2002
12 August 2002
24 June 2003
25 June 2003
Wind direction
NW-ESE Lake breeze 10:30 h A: 2.0 B: 3.5 A: 24.4 B: 26.1 A: 71 B: 60 A: 32, 17 B: 11, 4 A: 15, 55 B: 76, 98
SW
SW
SW-SSE Lake breeze 11:30 h
SW
A: 7.4 B: 8.6 A: 26.7 B: 30.4 A: 69 B: 57 A: 30, 18 B: 17, 13 A: 35, 56 B: 63, 80
A: 6.9 B:7.2 A: 23.7 B: 28.4 A: 75 B: 55 A: 33, 22 B: 14, 12 A: 15, 44 B: 69, 90
A: 5.4 B: 3.4 A: 25.5 B: 28.1 A: 52 B: 51 A: 40, 28 B: 39, 11 A: 38, 67 B: 57, 98
A: 6.7 B: 5.1 A: 25.5 B: 29.9 A: 55 B: 52 A: 40, 24 B: 18, 13 A: 40, 72 B: 91, 100
Average wind speed m s
1
Average temperature oC % Average % relative humidity NOx at times A: 07:00, 10:00 B: 11:00, 13:00 h (ppb) O3 at times A: 07:00, 10:00 h B: 11:00, 13:00 h (ppb)
All times are shown as local times (LT).
and other reactions, such as pattern (g) and (h) can take place to form the secondary aerosols detected at LUCAS. On some subsequent days, for example 24 June 2003 and 25 June 2003, very high ozone mixing ratios of 90 ppb were encountered and are most likely the result of a stagnant situation with consistent southwest winds combined with high temperature and solar radiation. Because of this the sulfate concentrations were found to be increased as well and ranked highest for all days. In case of 25 June 2003 almost no variations between collection A and B was recorded (+2% change) which most likely can be attributed to the stagnant air current. Two Lake Breeze days were observed (30 July 2002 and 24 June 2003) where wind directions changed towards the SE later in the morning. These days clearly differed from other days of case two with wind speeds being much lower during collection A and ion concentrations showing a greater increase from collection A to collection B. For 30 July 2002 the mean sulfate concentration increased by +36%, the mean nitrate concentration by +120% and the mean oxalate concentration by +31%. On 24 June 2003 the increase of the mean concentrations were found to be of roughly the same order for sulfate (+35%) and nitrate (+115%), but higher for oxalate (+96%). The reason for the higher oxalate increase might be a difference in original wind direction and in relative humidity making oxalate more available in the particulate (solid) phase and less dissolved in the liquid phase on 24 June 2003.
3.4. Case 3: lake breeze pollutant transport The third case discussed here can be characterized with respect to high NOx mixing ratios and high nitrate concentrations. Table 3 shows the weather parameter, and NOx and ozone mixing ratios for five representative days grouped into case three. As seen in Fig. 3(c) for 14 July 2003, NOx mixing ratios were found to be very high in the morning between 06:00 and 09:00 h. Despite of these consistently high NOx mixing ratios in the morning, ozone mixing ratios were in average lower then the ones observed for case two. The same is valid for sulfate and oxalate concentrations, which also ranked lower than in case two. Fig. 5(c) shows the ion concentrations for case three with averages of 10.173.4, 9.571.5, and 0.18570.0 mg m 3 for sulfate, nitrate and oxalate, respectively. Temperatures in this case were moderate, averaging 24 1C. However, wind speed readings (3.5 m s 1 average) and relative humidity (51.6% average) were the lowest of all three cases with wind speeds increasing from collection A to collection B on all, but one day. All days originally had south or southwesterly wind directions, characterized as being high in pollutants, and then shifted to the east or southeast, thus establishing a lake breeze (Fig. 4(d)). The observations discussed for case three indicate that days experiencing high NOx mixing ratios in the morning also are likely to show a lake breeze event, and both phenomena appear to occur under the same atmospheric conditions. Conditions are favorable for a
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
6647
Table 3 Case three weather data averaged for each collection (A: 07:00–10:00 h, B: 11:00–13:00 h), and ozone and NOx mixing ratios at start and end of each collection Day
9 August 2002
1 July 2003
14 July 2003
30 July 2003
31 July 2003
Wind direction
SW-SE Lake breeze 10:00 h A: 3.1 B: 6.9 A: 21.7 B: 22.3 A: 53 B: 59 A: 90, 44 B: 30, 10
SW-NE-E Lake breeze 09:30, 10:30 h A: 1.4 B: 1.2 A: 24.4 B: 25.8 A: 54 B: 41 A:226, 90 B: 93, 25
S-SE Lake breeze 11:00 h A: 2.6 B: 3.9 A: 23.3 B: 26.7 A: 52 B: 48 A:150, 20 B: 17, 16
SSW-SE Lake breeze 12:50 h A: 3.7 B: 4.3 A: 21.6 B: 25.5 A: 53 B: 31 A:140, 36 B: 30, 25
SW-SE Lake breeze 08:45 h A: 3.8 B: 4.1 A: 23.1 B: 25.3 A: 65 B: 60 A: 85, 16 B: 20, 11
A: 10, 30 B: 35, 60
A: 7, 25 B: 30, 75
A: 20, 38 B: 36, 60
A: 9, 49 B: 52, 60
A: 10, 60 B: 57, 82
Average wind speed (m s 1) Average temperature (1C) Average % relative humidity NOx at times A: 07:00, 10:00 h B: 11:00, 13:00 h (ppb) O3 at times A: 07:00, 10:00 h B: 11:00, 13:00 h (ppb)
All times are shown as local times (LT).
lake breeze when initial wind speeds are low in the morning with clear and sunny skies. Such situation causes air stagnation, and therefore, promotes pollution build-up with subsequent NOx photolysis leading to ozone and secondary aerosol formation. The high NOx mixing ratios observed in the early morning on these days are the result of pollutants trapped in the nocturnal boundary layer, which did not mix well due to calm conditions. Mixing is slow even after sunrise leading to the elevated NOx mixing ratios detected. Fig. 3(c) shows an example on 14 July 2003. Photographs taken during such days show a distinct brown layer above the metropolitan Chicago area extending also onto Lake Michigan. As the boundary layer continues to expand and mixing occurs, NOx mixing ratios decrease in the later morning hours (09:00–11:00 h). Despite the high NOx mixing ratios during the morning hours, the observed ozone mixing ratios were lower than for case two. The reason could be that at very high NOx mixing ratios, insufficient hydroperoxy radicals from reaction (d) are present to catalyze ozone formation as NO2 reacts with hydroxyl radicals to form HONO2, thus removing hydroxyl radicals from the reaction cycle (Walcek et al., 1997). Such reaction pattern would explain the very high nitrate concentrations determined for collection A compared to the other cases. Once the high NOx level subsides, reaction sequence (d) leading to secondary NOx formation and ozone production in (a) could take place and, subse-
quently, ozone levels begin to rise at 10:00 h. Until this point in time, the samples were taken from air currents of southerly direction transporting urban pollution towards LUCAS and also above Lake Michigan. On 14 July 2003 the wind direction changed towards southeast at around 11:00 h establishing a lake breeze and bringing with it the polluted air residing above Lake Michigan towards LUCAS. A peak in NOx mixing ratios at 11:00 h indicates this passing of the lake breeze front before leveling off later on. Such situation occurred on other lake breeze days as well, however, at different times, owing to the somewhat different atmospheric conditions. The return of the polluted air mass with the lake breeze, causes in many cases a noticeable increase in sulfate, oxalate, and also nitrate from collection A to collection B, because chemical reactivity is high due to abundant solar radiation leading to the formation of secondary pollutants in that air mass. However, under some circumstances either sulfate or nitrate concentrations did increase during the second collection. This might be the result of a combination between differing strength of emission sources and variations in meteorological conditions. 4. Conclusion and summary Airborne particulates were collected and reactive trace gases (NOx and ozone) were recorded at LUCAS in Chicago during the summer months of
ARTICLE IN PRESS 6648
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
2002 and 2003. Ion chromatography was used to determine the concentrations of sulfate, nitrate and oxalate in the airborne particulate samples. It was found that the ion concentrations depended on local atmospheric conditions, including the lake breeze. Based on data obtained from the different species as well as meteorology, three cases were distinguished and considered representative for air quality conditions present in Chicago. Air currents arriving north and northeast from Lake Michigan showed low values for both gaseous precursors and ions because pollution sources are scarce in these areas. In contrast, southwesterly air currents allowed for transport of pollutants emitted from industrial sources in the southwest of Chicago towards LUCAS. In addition, the high solar intensity experienced on such days lead to the detection of elevated secondary pollutants, in form of ozone and sulfate. During a third situation, a stagnant air mass was encountered during the early morning hours leading to high local pollution and a lake breeze later in the day. Despite pollutants leveled off as a result of better mixing at midmorning (10:00 h) the onset of a lake breeze increased mixing ratios of NOx and secondary pollutants, especially nitrate temporarily. Acknowledgements Part of this study was supported by the Department of Energy/Argonne National Laboratory Grant 2F-02341. We would like to thank the anonymous reviewers for their valuable comments.
References Aneja, V.P., Oommen, R.G., Riordan, A.J., Arya, S.P., Wayland, R.J., Murray, G.C., 1999. Ozone patterns for three metropolitan statistical areas in North Carolina, USA. Atmospheric Environment 33, 5081–5093. Bloomfield, P., Royle, A.J., Steinberg, LJ., Yang, Q., 1996. Accounting for meteorological effects in measuring urban ozone levels and trends. Atmospheric Environment 30, 3067–3077. Cass, G.R., Shair, F.H., 1984. Sulfate accumulation in a sea breeze/land breeze circulation system. Journal of Geophysical Research 89 (D1), 1429–1438. Chameides, W.L., Fehsenfeld, F., Rodgers, M.O., Cardelino, C., Martinez, J., Parrish, D., Lonneman, W., Lawson, D.R., Rasmussen, R.A., Zimmerman, P., Greenberg, J., Middleton, P., Wang, T., 1992. Ozone precursor relationship in the ambient atmosphere. Journal of Geophysical Research— Atmospheres 97 (D5), 6037–6055.
Chebbi, A., Carlier, P., 1996. Carboxylic acids in the troposphere, occurrences, sources and sinks: a Review. Atmospheric Environment 30, 4233–4249. Chow, J.C., Watson, J.G., Lowenthal, D.H., Egami, R.T., Solomon, P.A., Thuillier, R.H., Maglianos, K., Ranzieri, A., 1998. Spatial and temporal variations of particulate precursor gases and photochemical reaction products during SJVAQS/AUSPEX ozone episodes. Atmospheric Environment 32, 2835–2844. Draxler, R.R., Rolph, G.D., 2003. HYSPLIT (HYbrid SingleParticle Lagrangian Integrated Trajectory) Model access via NOAA ARL READY Website (http://www.arl.noaa.gov/ ready/hysplit4.html). NOAA Air Resources Laboratory, Silver Spring, MD. Finlayson-Pitts, B.J., Pitts Jr., J.N., 2000. Chemistry of the Upper and Lower Atmosphere: Theory, Experiments and Applications. Academic Press, San Diego, CA, 970pp. Fosco, T., 2006. Ion concentrations in atmospheric aerosols: the influence of NOx and ozone mixing ratios, local meteorology, and the lake breeze. Ph.D. Thesis, Loyola University Chicago. Fosco, T., Schmeling M., accepted 2006. Determination of watersoluble atmospheric aerosols using ion chromatography. Environmental Monitoring and Assessment. Gatz, D.F., 1975. Relative contributions of different sources of urban aerosols: application of a new estimation method to multiple sites in Chicago. Atmospheric Environment 9, 1–18. Granby, K., Egelov, A.H., Nielson, T., Lohse, C., 1997. Carboxylic acids: seasonal variation and relation to chemical and meteorological parameters. Journal of Atmospheric Chemistry 28, 195–201. Grosjean, D., 1988. Aldehydes, carboxylic acids and inorganic nitrate during NSMCS. Atmospheric Environment 22 (8), 1637–1648. Gusten, H., Heinrich, G., Cvitas, T., Klasinc, L., Ruscic, B., Lala, D.P., 1988. Photochemical formation and transport of ozone in Athens, Greece. Atmospheric Environment 22, 1855–1861. Hastie, D.R., Narayan, J., Schiller, C., Niki, H., Shepson, P.B., Sills, D.M.L., Taylor, P.A., Moroz, Wm.J., Drummond, J.W., Reid, N., Taylor, R., Roussel, P.B., Melo, O.T., 1999. Observational evidence for the impact of the lake breeze circulation and ozone concentrations in Southern Ontario. Atmospheric Environment 33, 323–335. Hoffman, M.R., 1986. On the kinetics and mechanism of oxidation of aquated sulfur dioxide by ozone. Atmospheric Environment 20, 1145–1154. Jacob, D.J., 2000. Heterogeneous chemistry and tropospheric ozone. Atmospheric Environment 34, 2131–2159. Jacobson, M.C., Hansson, H.C., Noone, K.J., Charlson, R.J., 2000. Organic atmospheric aerosols: review and state of the science. Reviews of Geophysics 38, 267–294. Jenkin, M.E., Clemitshaw, K.C., 2000. Ozone and other secondary photochemical pollutants: chemical processes governing their formation in the planetary boundary layer. Atmospheric Environment 34, 2499–2527. Kadowski, S., 1986. On the nature of atmospheric oxidation processes of SO2 to sulfate and of NO2 to nitrate on the basis of diurnal variations of sulfate, nitrate and other pollutants in an urban area. Environmental Science and Technology 20 (12), 1249–1253.
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649 Kato, S., Koyama, Y., et al., 1999. Composition of aerosols in an industrialized area of northeastern China. Journal of Environmental Science Health A 32 (5), 1075–1092. Khwaja, H.A., 1995. Atmospheric concentrations of carboxylic acids and related compounds at a semiurban site. Atmospheric Environment 29, 127–139. Lee, S.K., Wadden, R.A., Scheff, P.A., 1993. Measurement and evaluation of acid air pollutants in Chicago using an annular denuder system. Atmospheric Environment 27A, 543–553. Lestari, P., Oskouie, A.K., Noll, K.E., 2003. Size distribution and dry deposition of particulate mass, sulfate and nitrate in an urban area. Atmospheric Environment 37, 2507–2516. Lin, J.J., 2002. Characterization of water-soluble ion species in urban ambient particles. Environment International 28, 55–61. Lyons, W.A., 1972. The climatology and prediction of the Chicago Lake Breeze. Journal of Applied Meteorology 11, 1259–1270. Lyons, W.A., Cole, H.S., 1976. Photochemical oxidant transport: mesoscale Lake Breeze and synoptic-scale aspects. Journal of Applied Meteorology 15, 733–743. Mapquest.com, 2006. www.mapquest.com. Nair, P.R., Chand, D., Lal, S., Modh, K.S., Naja, M., Parameswaran, K., Ravindran, S., Venkataramani, S., 2002. Temporal variations in surface ozone at Thumba (8.6 N,
6649
77 E)—a tropical coastal site in India. Atmospheric Environment 36, 603–610. Pont, V., Fontan, J., 2000. Local and regional contributions to photochemical atmospheric pollution in southern France. Atmospheric Environment 34, 5209–5223. Scheff, P.A., Wadden, R.A., Allen, R.J., 1984. Quantitative assessment of Chicago air pollution through analysis of covariance. Environmental Science and Technology 18, 1623–1631. Skoog, D.A., West, D.M., Holler, F.J., Crouch, S.R., 2000. Analytical Chemistry: An Introduction. Saunders College Publishing, Fort Worth, TX, 155pp. Solomon, P.A., Salmon, L.G., Fall, T., Cass, G.R., 1992. Spatial and temporal distribution of atmospheric nitric acid and particulate nitrate concentrations in the Los Angeles area. Environmental Science and Technology 26, 1594–1601. Tolocka, M.P., Solomon, P.A., et al., 2001. East versus West in the US: chemical characteristics of PM2.5 during the winter 1999. Aerosol Science and Technology 34, 88–96. United States Environmental Protection Agency, 2005. List of designated reference and equivalence methods. http://www. epa.gov/ttnamti1/files/ambient/criteria/ref1005.pdf Walcek, C.J., Yuan, H., Stockwell, W.R., 1997. The influence of aqueous-phase chemical reaction on ozone formation in polluted and nonpolluted clouds. Atmospheric Environment 31, 1221–1237.
Atmospheric Environment 40 (2006) 6638–6649 www.elsevier.com/locate/atmosenv
Aerosol ion concentration dependence on atmospheric conditions in Chicago Tinamarie Fosco, Martina Schmeling Loyola University Chicago, 6525 N. Sheridan Rd. Chicago, IL 60626, USA Received 13 September 2005; accepted 18 May 2006
Abstract This study seeks to determine the influence of precursor trace gases and local meteorology, including lake breeze events, on the concentrations of secondary aerosol species in Chicago. For this, two particulate air samples per day were collected onto quartz fiber filters at the Loyola University Chicago Air Station (LUCAS) during the summer months of 2002 and 2003 and subsequently analyzed by ion chromatography for sulfate, nitrate and oxalate. In parallel, mixing ratios of ozone (O3) and nitrogen oxides (NO and NO2QNOx) were monitored and weather parameters were recorded. In addition, backward trajectories were obtained to estimate air mass transport to Chicago. Ozone and NOx mixing ratios as well as sulfate, nitrate and oxalate concentrations varied substantially throughout the study, but three situations could be distinguished based on meteorology and chemistry. Case one had the lowest ozone, NOx and ion levels due to wind directions constantly from Lake Michigan. Case two comprised of days showing the highest pollutant levels because of to predominantly southwestern air currents and warm temperatures and case three experienced an air stagnation situation in the morning leading to high NOx mixing ratios and a subsequent lake breeze event. In the last case, elevated ozone mixing ratios and ion concentrations were observed after lake breeze onset indicating pollutant transport. r 2006 Elsevier Ltd. All rights reserved. Keywords: Ion chromatography; Lake breeze; Urban air quality; Secondary aerosol formation; Pollutant transport
1. Introduction Chicago as a major metropolitan center in the Midwest presents a significant source region of emissions from traffic and industry and experiences regularly air pollution events during hot summer days. In addition, Chicago’s location at the southwestern end of Lake Michigan exposes the city to lake breezes where the major air current originates Corresponding author. Tel.: +1 773 508 3124, fax: +1 773 508 3086. E-mail address: [email protected] (M. Schmeling).
1352-2310/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2006.05.061
from Lake Michigan. It was found that these events might play a major role in the pollution pattern of the region (see for example: Bloomfield et al., 1996; Gatz, 1975; Lee et al., 1993; Lestari et al., 2003; Scheff et al., 1984). In order to investigate these pollution patterns in more detail, aerosol samples were collected during the summer months of 2002 and 2003 and analyzed for their ionic content. The major ions of interest comprised of sulfate, nitrate and oxalate, because these ions are considered secondary pollutants formed as a result of reactions by reactive trace gases such as sulfur dioxide (SO2), nitrogen oxides (NO, NO2) and volatile organic
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
compounds (VOCs) with ozone (O3) and hydroxyl radicals (dOH). Studies have found that the formation of secondary pollutants does not only depend on emission sources, but is also influenced by local meteorology. For example, high ozone episodes have been observed in areas downwind of pollution sources with high temperatures and low wind speeds (Aneja et al., 1999; Bloomfield et al., 1996; Chow et al., 1998). Also, studies have linked elevated ozone levels to sea or lake breezes in regions close to large bodies of water (Cass and Shair, 1984; Gusten et al., 1988; Hastie et al., 1999; Lyons and Cole, 1976; Nair et al., 2002; Pont and Fontan, 2000). Both situations for formation of secondary aerosols have to be considered in Chicago with major emission sources primarily located in the south (S) and southwestern (SW) part of the city and lake breezes occurring at about 20% of summer days. A lake breeze is the result of a temperature gradient between land and lake surfaces. During hot summer days, the land surface heats faster than the water body creating a temperature and pressure gradient between these two. The difference in pressure leads to air mass transport from the high pressure area over the lake (cooler air) to the lower pressure area over the land (warmer air). During the night, the opposite so-called land-breeze can occur, which transports cooler polluted air from the city towards and above Lake Michigan. An air mass experiencing a lake or land breeze event is confined to a relatively small volume with little mixing and trace gases and particulates can become highly concentrated. A detailed description of the lake breeze climatology with respect to Chicago has been published by Lyons (1972). The study presented here seeks to investigate the conditions leading to formation of the secondary aerosol species, sulfate, nitrate and oxalate, in conjunction with ozone and NOx mixing ratios as well as local meteorology including the lake breeze in Chicago. 2. Methods 2.1. Sample collection Atmospheric aerosol samples were collected at Loyola University Chicago’s air station (LUCAS). LUCAS is situated 60m high atop a residence hall on Loyola University Chicago’s lake shore campus located approximately 13 km north of downtown
6639
Chicago and 200m west of Lake Michigan (Fig. 1). Sampling took place in 2002 (13 days) and 2003 (20 days) during the summer because the highest frequency of lake breeze events and the highest possible ozone mixing ratios and ion concentrations are expected at this time of the year. Particulates (PM2.5) were collected at a flowrate of 2 m3 h 1 onto quartz fiber filters (QMA, 47 mm diameter, Whatman International Ltd, Maidstone, UK) housed inside a Teflon filter holder. Samples were taken twice a day, from 07:00 h to 10:00 h (collection A), and 11:00 h to 13:00 h (collection B) local time (LT). Observations indicated that a lake breeze most likely establishes between 10:00 and 11:00 h; hence, the collection times selected allowed for the determination of particulate concentrations before and during the onset of a lake breeze. Immediately after collection, filters were frozen until analysis. In addition to air particulate collection, continuous 24 h trace gas measurements of NOx (NO and NO2) and ozone were made using chemiluminescence for NOx (Thermo Environmental Instruments, Inc. (TEI) Model 42C; US EPA designated method RFNA-1289-074, US EPA, 2005) and UV photometry (TEI Model 49C; US EPA designated method EQOA-0880-047, US EPA, 2005) for ozone, respectively. Both units were calibrated regularly with aid of the local Illinois Environmental Protection Agency’s bureau of air. Weather conditions were also recorded during the sampling time from 07:00 to 13:00 h. 2.2. Sample analysis Before analysis, filters were extracted in 5 ml of nanopure water for 15 min by ultrasonication. The obtained solution was analyzed subsequently for anions and low molecular weight organic acids by ion chromatography (IC) with chemical suppression and conductivity detection (Metrohm 761 Compact IC). The analysis procedure followed closely previously described ones with slight adaptations (Granby et al., 1997; Grosjean, 1988; Kadowski, 1986; Kato et al., 1999; Lin, 2002; Solomon et al., 1992; Tolocka et al., 2001). A Metrosep Supp5 column with 1.0mM NaHCO3 and 3.5mM Na2CO3 as eluent was used to determine fluoride, acetate, formate, chloride, nitrite, bromide, nitrate, phosphate, benzoate, sulfate, succinate, and oxalate. Sulfate, nitrate and oxalate accounted for the most abundant and continuously detected anions in the aerosol samples.
ARTICLE IN PRESS 6640
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
Fig. 1. Regional map of the metropolitan Chicago area (Mapquest.com, 2006). The measurement station (LUCAS) is located north of downtown Chicago in vicinity of Lake Michigan and indicated as a black dot. The majority of industrial sources are present in the southwest and south of the city. These sources include brick and cement works, steel manufacturing and metal processing as well as paper mills. One major airport (O’Hare International Airport) is located about 15 km to the west and a second major airport (Midway Airport) is located about 15 km to the southwest of LUCAS.
Clean quartz fiber filters were analyzed to determine the filter blank. It was observed that concentrations for some ions were high and variable on the blank filters, whereas some blanks were below detection limit. For instance, the blank concentrations for sulfate and nitrate were 12.8715.7 and 1.0670.702 mg/filter, respectively, but oxalate was found below detection limit on the blank filters. Blank values were subtracted from the ion concentrations in the samples and reported as 7standard deviation of the filter blank per volume of air sampled during collection. A detailed description of the method validation can be found in Fosco and Schmeling, 2006 and Fosco, 2006. All values presented here are based on a student t-test with 95% confidence level (Skoog et al., 2000).
2.3. Data analysis Sulfate, nitrate and oxalate concentrations varied substantially between collection days such that three different situations describing their formation could be distinguished. The collection days for each case experienced comparable ozone and NOx mixing ratios, and meteorology, but these parameters differed substantially between the cases. Also, the frequency of lake breeze events and its influence on pollutant transport varied among the cases. A lake breeze was identified by an initial wind direction from the land (N, NW, W, SW, S) and a later shift towards wind directions from Lake Michigan (NE, E, SE).
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
Also, backward air trajectories were calculated using the NOAA HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) model (Draxler and Rolph, 2003) to estimate the long-range transport of air masses to Chicago. However, the HYSPLIT model could not always identify the local scale lake breeze phenomenon because of its large grid size. 3. Case study results and discussion 3.1. Description of relevant atmospheric chemistry In order to account for the ion concentrations determined in each case, a brief description is given
6641
of the relevant chemical reactions. Details regarding these reactions and their kinetics can be found in Finlayson-Pitts and Pitts (2000), Chameides et al. (1992), Jenkin and Clemitshaw (2000), Jacob (2000), Hoffman, 1986, and Walcek et al. (1997). The reaction sequence leading to nitrate and sulfate aerosol formation is seen in Fig. 2 and involves the production of ozone (a) via photolysis of NO2, with NO2 being directly emitted into the atmosphere by industrial activity and vehicular traffic. Ozone production predominates during the day and in the summer when solar intensity is greatest. Without presence of other species a photostationary state is established cycling ozone, NO and NO2 according to reaction sequences (a) and (b). Ozone can be
Fig. 2. Relevant tropospheric chemistry reactions.
ARTICLE IN PRESS 6642
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
photolytically destroyed as seen in reaction sequence (c), presenting a major pathway for hydroxyl radical formation in urban areas. NOx and hydroxyl radicals also form due to reactions with VOCs as highlighted in reactions (d) and (e). Hydroperoxy radicals produce hydrogen peroxide, another major oxidant in the atmosphere, as in reaction (f). During the day, nitrogen dioxide is removed from the reaction cycle forming ozone (a) by oxidation with hydroxyl radicals yielding nitric acid (g). Ozone can also react with sulfur dioxide in a heterogeneous reaction on water droplets to form sulfuric acid (h). Another possible pathway, though of minor importance, for sulfuric acid formation presents the oxidation of sulfur dioxide with hydrogen peroxide. Finally, ammonium sulfate (or ammonium bisulfate) and ammonium nitrate particles can result when gas phase ammonia is present to neutralize these species. Carboxylic acids, such as oxalic acid, found in urban aerosols are formed as a result of complex and not yet fully understood oxidation pathways involving VOCs, ozone and hydroxyl radicals (Chebbi and Carlier, 1996; Jacobson et al., 2000; Khwaja, 1995). Therefore the presence of these species is mentioned in the following discussion, but no reaction scheme suggested. 3.2. Case one: clean air conditions The first case is characterized by days with the lowest NOx and ozone mixing ratios, as well as the lowest sulfate, nitrate and oxalate concentrations, therefore can be identified as ‘‘all low case’’. Five
days representative for case one are discussed in the following section and Table 1 shows the weather parameter and trace gas mixing ratios during the aerosol collection times for these days. As can be seen, NOx mixing ratios were very low with values less than 10 ppb and ozone mixing ratios were less than 45 ppb and remained relatively constant with no distinguishable diurnal pattern. Fig. 3(a) displays a typical photochemical pattern, while Fig. 4(a) shows a typical air mass pattern in the form of the HYSPLIT backward trajectory. Air masses originated from the north of LUCAS and traveled a substantial time across rural areas and water. Not surprising, the case one days also experienced the lowest temperatures of all three cases, averaging only 20 1C. The average wind speed (4.8 m s 1) in case one was moderate, while the average relative humidity (67.1%) ranked highest when compared to the other two cases. Fig. 5(a) displays the mean concentrations with standard deviations (Student ttest, 95% confidence level) of sulfate, nitrate and oxalate concentrations for case one days. The overall average values were 6.1973.14, 1.1570.14, and 0.11870.0 mg m 3 for the three respective ions. The low trace gas mixing ratios and ion concentrations observed can likely be attributed to the wind direction and low temperature on these days. The samples were collected from air masses arriving from the northeast where pollution sources are scarce and any long-range pollution would have traveled a substantial distance across Lake Michigan and be most likely deposited into the lake. This is most pronounced when north or northeast winds are predominant as on 18 July 2003 for example. On
Table 1 Case one weather data averaged for each collection (A: 07:00–10:00 h, B: 11:00–13:00 h), and ozone and NOx mixing ratios at start and end of each collection Day
24 July 2002
2 August 2002
6 August 2002
7 August 2002
18 July 2003
Wind direction
ESE/E (10:30 h) A: 4.7 B: 2.8 A: 19.8 B: 20.4 A: 57 B: 63 A: 5, 6 B: 5, 4 A: 35, 30 B: 34, 35
NNE
E
ESE
A: 7.1 B: 6.2 A: 21.7 B: 22.0 A: 76 B: 66 A: 3, 3 B: 3, 3 A: 45, 38 B: 37, 36
A: 6.1 B: 4.7 A: 19.0 B: 19.7 A: 70 B: 68 A: 2, 4 B: 3, 1 A: 18, 24 B: 26, 28
A: 4.0 B: 2.6 A: 19.4 B: 20.3 A: 58 B: 59 A: 2, 5 B: 5, 2 A: 25, 28 B: 27, 34
N-NE Lake Breeze 11:00 h A: 4.0 B: 5.7 A: 18.5 B: 19.4 A: 77 B: 77 A: 1, 3 B: 3, 2 A: 29, 30 B: 31, 42
Average wind speed (m s 1) Average temperature (oC) % Average % relative humidity NOx at times A: 07:00 h, 10:00 h B: 11:00 h, 13:00 h (ppb) O3 at times A: 07:00 h, 10:00 h B: 11:00 h, 13:00 h (ppb)
All times are shown as local times (LT).
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
6643
200 O3
180
200
NOx
180
160
120
100
ppb
80
100
Time
22:00
20:00
18:00
16:00
14:00
12:00
10:00
8:00
6:00
0:00
22:03
20:03
18:02
16:02
14:02
12:02
10:01
8:01
6:00
0 4:00
20
0 2:00
40
20 0:00
60
40
4:00
80
60
2:00
ppb
NOx
140
120
(a)
O3
160
140
Time
(b)
200 O3
180
NOx
160 140 ppb
120 100 80 60 40 20
(c)
22:02
20:02
18:02
16:02
14:01
12:01
10:01
8:01
6:01
4:00
2:00
0:00
0
Time
Fig. 3. 24-hour ozone and NOx mixing ratios on: 7 August 2002 (a); 25 June 2003 (b); and 14 July 2003 (c).
three of the five days described here under case one, namely 24 July, 6 August, and 7 August 2002, substantially higher sulfate and nitrate concentrations were observed during the second collection period. The percent change between collection A and B based on the mean values, were +125% (sulfate) and +256% (nitrate) on 24 July +58% (sulfate) and +272% (nitrate) on 6 August and +54% (sulfate) and +261% (nitrate) on 7 August 2002. In contrast the changes between collections A and B on 2 August 2002 and 18 July 2003 were +57% (sulfate) and +7%(nitrate) as well as 72% (sulfate) and +183% (nitrate), respectively. Weather data for the first three days indicate that the air masses did not originate from the northeast, but rather from the southeast and hence were exposed to the industrial area around Gary, IN and the southern Michigan shoreline leading to the elevated ion concentrations observed. Interestingly, the
backtrjectories for the first three days mentioned show a similar pattern with air masses arriving from the east and originating north of Michigan as shown in Fig. 4(a). In contrast, the air masses on the latter two days did have their origin west of the station with subsequently moving onto Lake Michigan in Wisconsin, thus circumventing the metropolitan Chicago area before swinging back towards the city with easterly air currents as seen in Fig. 4(b). The pattern these two days followed corresponds to low emissions as the region traversed is mostly rural and agricultural. Nevertheless, in contrast to cases two and three the average ion concentrations were much lower in case one, which might be explained by the following pattern: With the lack of pollution sources, NOx is limited and ozone destruction (reaction (c) in Fig. 2) is favored (Walcek et al., 1997). Also, VOCs are expected to be low causing a further restriction of secondary NOx formation in
ARTICLE IN PRESS 6644
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
Fig. 4. Three-day backward isentropic air trajectories from an altitude of 100m at Chicago starting at 07:00 h (CDT): (a) 7 August 2002 (b) 18 July 2003 (c) 25 June 2003 (d) 1 July 2003.
reaction (d) and ozone production in (a). Correspondingly, formation of the secondary aerosols, sulfate, oxalate and nitrate is suppressed yielding the low concentrations observed. Moderate wind speeds kept the air diluted and cool temperatures, further limited NOx photolysis and, thus ozone and ionic aerosol formation. Only one lake breeze day was recognized in case one. Weather data indicates that on 18 July 2003, an air mass arrived from the north during the early morning hours followed by a shift in wind direction at 11:00 h to northeast winds. Since major pollution sources are lacking, air masses following these
directions are typically clean. However, higher nitrate concentrations were observed on 18 July 2003, compared to the other days in case one. This can likely be attributed to the return of pollutants with the lake breeze and will be discussed further in case three. 3.3. Case two: pollution episodes The second case is characterized by days with the highest ozone mixing ratios, and sulfate and oxalate concentrations and can be classified as the ‘‘high pollutant case’’. The weather parameter, and trace
ARTICLE IN PRESS 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
sulfate nitrate oxalate ug/m^3
ug/m^3
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
A
A
B
A
B
A
B
A
B
July 24, August 2, August 6, August 7, July 18, 2002 2002 2002 2002 2003
ug/m^3
(a)
B
(c)
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0
sulfate nitrate oxalate
A
B
July 30, 2002
(b)
6645
A
B
A
B
A
B
August 1, August 12, June 24, 2003 2002 2002
A
B
June 25, 2003
sulfate nitrate oxalate
A B August 9, 2002
A B July 1, 2003
A B July 14, 2003
A B July 30, 2003
A B July 31, 2003
Fig. 5. Sulfate, nitrate and oxalate concentrations for: case one (a); case two (b); and case three (c). Oxalate concentrations were increased by a factor of 15 for visual purposes. (collection A: 07:00–10:00 h, collection B: 11:00–13:00 h).
gas mixing ratios for the five representative days of case two are shown in Table 2. Fig. 3(b) displays the typical trace gas mixing ratios for comparison to the other two cases. NOx mixing ratios were moderate with 30 and 60 ppb maxima between 06:00 to 09:00 h. Ozone mixing ratios were highest with 100 ppb maxima just after 13:00 h and had a distinct diurnal pattern. The average temperature for case 2 days (26.9 1C) was the highest observed for all cases; average wind speed was slightly higher than for case one with 5.62 m s 1 and relative humidity was below case one, averaging 59.7%. The wind direction was predominantly southwest except for 30 July 2002 experiencing initial winds from the northwest and later from the southeast. A typical air mass pattern is depicted in Fig. 4(c). In Fig. 5(b), ion concentrations for case two are shown in comparison with the two other cases discussed. Sulfate and oxalate concentrations were highest for all three cases
averaging 15.873.2 and 0.30170.0 mg m 3, respectively. Nitrate concentrations were found to be 7.2570.14 mg m 3 in average and thus, ranked between case one and case three. The high pollution conditions found for case two are not surprising when considering that the main wind direction was southwest on almost all days. Most industrial activities are concentrated in the southwest of Chicago, and reactive trace gases (NOx, VOCs and SO2) emitted in this area are transported by southwest air currents across the metropolitan area to the measurement station. Therefore, sufficient precursor species can be present and the reaction patterns described in Fig. 2 are not restricted as in case one. Also, the high temperatures experienced and the abundance of solar radiation in summer enhance the photolysis of NO2 and thus, ozone production via reaction (a). During the air mass transport time, photochemical
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
6646
Table 2 Case two weather data averaged for each collection (A: 07:00–10:00 h, B: 11:00–13:00 h), and ozone and NOx mixing ratios at start and end of each collection Day
30 July 2002
1 August 2002
12 August 2002
24 June 2003
25 June 2003
Wind direction
NW-ESE Lake breeze 10:30 h A: 2.0 B: 3.5 A: 24.4 B: 26.1 A: 71 B: 60 A: 32, 17 B: 11, 4 A: 15, 55 B: 76, 98
SW
SW
SW-SSE Lake breeze 11:30 h
SW
A: 7.4 B: 8.6 A: 26.7 B: 30.4 A: 69 B: 57 A: 30, 18 B: 17, 13 A: 35, 56 B: 63, 80
A: 6.9 B:7.2 A: 23.7 B: 28.4 A: 75 B: 55 A: 33, 22 B: 14, 12 A: 15, 44 B: 69, 90
A: 5.4 B: 3.4 A: 25.5 B: 28.1 A: 52 B: 51 A: 40, 28 B: 39, 11 A: 38, 67 B: 57, 98
A: 6.7 B: 5.1 A: 25.5 B: 29.9 A: 55 B: 52 A: 40, 24 B: 18, 13 A: 40, 72 B: 91, 100
Average wind speed m s
1
Average temperature oC % Average % relative humidity NOx at times A: 07:00, 10:00 B: 11:00, 13:00 h (ppb) O3 at times A: 07:00, 10:00 h B: 11:00, 13:00 h (ppb)
All times are shown as local times (LT).
and other reactions, such as pattern (g) and (h) can take place to form the secondary aerosols detected at LUCAS. On some subsequent days, for example 24 June 2003 and 25 June 2003, very high ozone mixing ratios of 90 ppb were encountered and are most likely the result of a stagnant situation with consistent southwest winds combined with high temperature and solar radiation. Because of this the sulfate concentrations were found to be increased as well and ranked highest for all days. In case of 25 June 2003 almost no variations between collection A and B was recorded (+2% change) which most likely can be attributed to the stagnant air current. Two Lake Breeze days were observed (30 July 2002 and 24 June 2003) where wind directions changed towards the SE later in the morning. These days clearly differed from other days of case two with wind speeds being much lower during collection A and ion concentrations showing a greater increase from collection A to collection B. For 30 July 2002 the mean sulfate concentration increased by +36%, the mean nitrate concentration by +120% and the mean oxalate concentration by +31%. On 24 June 2003 the increase of the mean concentrations were found to be of roughly the same order for sulfate (+35%) and nitrate (+115%), but higher for oxalate (+96%). The reason for the higher oxalate increase might be a difference in original wind direction and in relative humidity making oxalate more available in the particulate (solid) phase and less dissolved in the liquid phase on 24 June 2003.
3.4. Case 3: lake breeze pollutant transport The third case discussed here can be characterized with respect to high NOx mixing ratios and high nitrate concentrations. Table 3 shows the weather parameter, and NOx and ozone mixing ratios for five representative days grouped into case three. As seen in Fig. 3(c) for 14 July 2003, NOx mixing ratios were found to be very high in the morning between 06:00 and 09:00 h. Despite of these consistently high NOx mixing ratios in the morning, ozone mixing ratios were in average lower then the ones observed for case two. The same is valid for sulfate and oxalate concentrations, which also ranked lower than in case two. Fig. 5(c) shows the ion concentrations for case three with averages of 10.173.4, 9.571.5, and 0.18570.0 mg m 3 for sulfate, nitrate and oxalate, respectively. Temperatures in this case were moderate, averaging 24 1C. However, wind speed readings (3.5 m s 1 average) and relative humidity (51.6% average) were the lowest of all three cases with wind speeds increasing from collection A to collection B on all, but one day. All days originally had south or southwesterly wind directions, characterized as being high in pollutants, and then shifted to the east or southeast, thus establishing a lake breeze (Fig. 4(d)). The observations discussed for case three indicate that days experiencing high NOx mixing ratios in the morning also are likely to show a lake breeze event, and both phenomena appear to occur under the same atmospheric conditions. Conditions are favorable for a
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
6647
Table 3 Case three weather data averaged for each collection (A: 07:00–10:00 h, B: 11:00–13:00 h), and ozone and NOx mixing ratios at start and end of each collection Day
9 August 2002
1 July 2003
14 July 2003
30 July 2003
31 July 2003
Wind direction
SW-SE Lake breeze 10:00 h A: 3.1 B: 6.9 A: 21.7 B: 22.3 A: 53 B: 59 A: 90, 44 B: 30, 10
SW-NE-E Lake breeze 09:30, 10:30 h A: 1.4 B: 1.2 A: 24.4 B: 25.8 A: 54 B: 41 A:226, 90 B: 93, 25
S-SE Lake breeze 11:00 h A: 2.6 B: 3.9 A: 23.3 B: 26.7 A: 52 B: 48 A:150, 20 B: 17, 16
SSW-SE Lake breeze 12:50 h A: 3.7 B: 4.3 A: 21.6 B: 25.5 A: 53 B: 31 A:140, 36 B: 30, 25
SW-SE Lake breeze 08:45 h A: 3.8 B: 4.1 A: 23.1 B: 25.3 A: 65 B: 60 A: 85, 16 B: 20, 11
A: 10, 30 B: 35, 60
A: 7, 25 B: 30, 75
A: 20, 38 B: 36, 60
A: 9, 49 B: 52, 60
A: 10, 60 B: 57, 82
Average wind speed (m s 1) Average temperature (1C) Average % relative humidity NOx at times A: 07:00, 10:00 h B: 11:00, 13:00 h (ppb) O3 at times A: 07:00, 10:00 h B: 11:00, 13:00 h (ppb)
All times are shown as local times (LT).
lake breeze when initial wind speeds are low in the morning with clear and sunny skies. Such situation causes air stagnation, and therefore, promotes pollution build-up with subsequent NOx photolysis leading to ozone and secondary aerosol formation. The high NOx mixing ratios observed in the early morning on these days are the result of pollutants trapped in the nocturnal boundary layer, which did not mix well due to calm conditions. Mixing is slow even after sunrise leading to the elevated NOx mixing ratios detected. Fig. 3(c) shows an example on 14 July 2003. Photographs taken during such days show a distinct brown layer above the metropolitan Chicago area extending also onto Lake Michigan. As the boundary layer continues to expand and mixing occurs, NOx mixing ratios decrease in the later morning hours (09:00–11:00 h). Despite the high NOx mixing ratios during the morning hours, the observed ozone mixing ratios were lower than for case two. The reason could be that at very high NOx mixing ratios, insufficient hydroperoxy radicals from reaction (d) are present to catalyze ozone formation as NO2 reacts with hydroxyl radicals to form HONO2, thus removing hydroxyl radicals from the reaction cycle (Walcek et al., 1997). Such reaction pattern would explain the very high nitrate concentrations determined for collection A compared to the other cases. Once the high NOx level subsides, reaction sequence (d) leading to secondary NOx formation and ozone production in (a) could take place and, subse-
quently, ozone levels begin to rise at 10:00 h. Until this point in time, the samples were taken from air currents of southerly direction transporting urban pollution towards LUCAS and also above Lake Michigan. On 14 July 2003 the wind direction changed towards southeast at around 11:00 h establishing a lake breeze and bringing with it the polluted air residing above Lake Michigan towards LUCAS. A peak in NOx mixing ratios at 11:00 h indicates this passing of the lake breeze front before leveling off later on. Such situation occurred on other lake breeze days as well, however, at different times, owing to the somewhat different atmospheric conditions. The return of the polluted air mass with the lake breeze, causes in many cases a noticeable increase in sulfate, oxalate, and also nitrate from collection A to collection B, because chemical reactivity is high due to abundant solar radiation leading to the formation of secondary pollutants in that air mass. However, under some circumstances either sulfate or nitrate concentrations did increase during the second collection. This might be the result of a combination between differing strength of emission sources and variations in meteorological conditions. 4. Conclusion and summary Airborne particulates were collected and reactive trace gases (NOx and ozone) were recorded at LUCAS in Chicago during the summer months of
ARTICLE IN PRESS 6648
T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649
2002 and 2003. Ion chromatography was used to determine the concentrations of sulfate, nitrate and oxalate in the airborne particulate samples. It was found that the ion concentrations depended on local atmospheric conditions, including the lake breeze. Based on data obtained from the different species as well as meteorology, three cases were distinguished and considered representative for air quality conditions present in Chicago. Air currents arriving north and northeast from Lake Michigan showed low values for both gaseous precursors and ions because pollution sources are scarce in these areas. In contrast, southwesterly air currents allowed for transport of pollutants emitted from industrial sources in the southwest of Chicago towards LUCAS. In addition, the high solar intensity experienced on such days lead to the detection of elevated secondary pollutants, in form of ozone and sulfate. During a third situation, a stagnant air mass was encountered during the early morning hours leading to high local pollution and a lake breeze later in the day. Despite pollutants leveled off as a result of better mixing at midmorning (10:00 h) the onset of a lake breeze increased mixing ratios of NOx and secondary pollutants, especially nitrate temporarily. Acknowledgements Part of this study was supported by the Department of Energy/Argonne National Laboratory Grant 2F-02341. We would like to thank the anonymous reviewers for their valuable comments.
References Aneja, V.P., Oommen, R.G., Riordan, A.J., Arya, S.P., Wayland, R.J., Murray, G.C., 1999. Ozone patterns for three metropolitan statistical areas in North Carolina, USA. Atmospheric Environment 33, 5081–5093. Bloomfield, P., Royle, A.J., Steinberg, LJ., Yang, Q., 1996. Accounting for meteorological effects in measuring urban ozone levels and trends. Atmospheric Environment 30, 3067–3077. Cass, G.R., Shair, F.H., 1984. Sulfate accumulation in a sea breeze/land breeze circulation system. Journal of Geophysical Research 89 (D1), 1429–1438. Chameides, W.L., Fehsenfeld, F., Rodgers, M.O., Cardelino, C., Martinez, J., Parrish, D., Lonneman, W., Lawson, D.R., Rasmussen, R.A., Zimmerman, P., Greenberg, J., Middleton, P., Wang, T., 1992. Ozone precursor relationship in the ambient atmosphere. Journal of Geophysical Research— Atmospheres 97 (D5), 6037–6055.
Chebbi, A., Carlier, P., 1996. Carboxylic acids in the troposphere, occurrences, sources and sinks: a Review. Atmospheric Environment 30, 4233–4249. Chow, J.C., Watson, J.G., Lowenthal, D.H., Egami, R.T., Solomon, P.A., Thuillier, R.H., Maglianos, K., Ranzieri, A., 1998. Spatial and temporal variations of particulate precursor gases and photochemical reaction products during SJVAQS/AUSPEX ozone episodes. Atmospheric Environment 32, 2835–2844. Draxler, R.R., Rolph, G.D., 2003. HYSPLIT (HYbrid SingleParticle Lagrangian Integrated Trajectory) Model access via NOAA ARL READY Website (http://www.arl.noaa.gov/ ready/hysplit4.html). NOAA Air Resources Laboratory, Silver Spring, MD. Finlayson-Pitts, B.J., Pitts Jr., J.N., 2000. Chemistry of the Upper and Lower Atmosphere: Theory, Experiments and Applications. Academic Press, San Diego, CA, 970pp. Fosco, T., 2006. Ion concentrations in atmospheric aerosols: the influence of NOx and ozone mixing ratios, local meteorology, and the lake breeze. Ph.D. Thesis, Loyola University Chicago. Fosco, T., Schmeling M., accepted 2006. Determination of watersoluble atmospheric aerosols using ion chromatography. Environmental Monitoring and Assessment. Gatz, D.F., 1975. Relative contributions of different sources of urban aerosols: application of a new estimation method to multiple sites in Chicago. Atmospheric Environment 9, 1–18. Granby, K., Egelov, A.H., Nielson, T., Lohse, C., 1997. Carboxylic acids: seasonal variation and relation to chemical and meteorological parameters. Journal of Atmospheric Chemistry 28, 195–201. Grosjean, D., 1988. Aldehydes, carboxylic acids and inorganic nitrate during NSMCS. Atmospheric Environment 22 (8), 1637–1648. Gusten, H., Heinrich, G., Cvitas, T., Klasinc, L., Ruscic, B., Lala, D.P., 1988. Photochemical formation and transport of ozone in Athens, Greece. Atmospheric Environment 22, 1855–1861. Hastie, D.R., Narayan, J., Schiller, C., Niki, H., Shepson, P.B., Sills, D.M.L., Taylor, P.A., Moroz, Wm.J., Drummond, J.W., Reid, N., Taylor, R., Roussel, P.B., Melo, O.T., 1999. Observational evidence for the impact of the lake breeze circulation and ozone concentrations in Southern Ontario. Atmospheric Environment 33, 323–335. Hoffman, M.R., 1986. On the kinetics and mechanism of oxidation of aquated sulfur dioxide by ozone. Atmospheric Environment 20, 1145–1154. Jacob, D.J., 2000. Heterogeneous chemistry and tropospheric ozone. Atmospheric Environment 34, 2131–2159. Jacobson, M.C., Hansson, H.C., Noone, K.J., Charlson, R.J., 2000. Organic atmospheric aerosols: review and state of the science. Reviews of Geophysics 38, 267–294. Jenkin, M.E., Clemitshaw, K.C., 2000. Ozone and other secondary photochemical pollutants: chemical processes governing their formation in the planetary boundary layer. Atmospheric Environment 34, 2499–2527. Kadowski, S., 1986. On the nature of atmospheric oxidation processes of SO2 to sulfate and of NO2 to nitrate on the basis of diurnal variations of sulfate, nitrate and other pollutants in an urban area. Environmental Science and Technology 20 (12), 1249–1253.
ARTICLE IN PRESS T. Fosco, M. Schmeling / Atmospheric Environment 40 (2006) 6638–6649 Kato, S., Koyama, Y., et al., 1999. Composition of aerosols in an industrialized area of northeastern China. Journal of Environmental Science Health A 32 (5), 1075–1092. Khwaja, H.A., 1995. Atmospheric concentrations of carboxylic acids and related compounds at a semiurban site. Atmospheric Environment 29, 127–139. Lee, S.K., Wadden, R.A., Scheff, P.A., 1993. Measurement and evaluation of acid air pollutants in Chicago using an annular denuder system. Atmospheric Environment 27A, 543–553. Lestari, P., Oskouie, A.K., Noll, K.E., 2003. Size distribution and dry deposition of particulate mass, sulfate and nitrate in an urban area. Atmospheric Environment 37, 2507–2516. Lin, J.J., 2002. Characterization of water-soluble ion species in urban ambient particles. Environment International 28, 55–61. Lyons, W.A., 1972. The climatology and prediction of the Chicago Lake Breeze. Journal of Applied Meteorology 11, 1259–1270. Lyons, W.A., Cole, H.S., 1976. Photochemical oxidant transport: mesoscale Lake Breeze and synoptic-scale aspects. Journal of Applied Meteorology 15, 733–743. Mapquest.com, 2006. www.mapquest.com. Nair, P.R., Chand, D., Lal, S., Modh, K.S., Naja, M., Parameswaran, K., Ravindran, S., Venkataramani, S., 2002. Temporal variations in surface ozone at Thumba (8.6 N,
6649
77 E)—a tropical coastal site in India. Atmospheric Environment 36, 603–610. Pont, V., Fontan, J., 2000. Local and regional contributions to photochemical atmospheric pollution in southern France. Atmospheric Environment 34, 5209–5223. Scheff, P.A., Wadden, R.A., Allen, R.J., 1984. Quantitative assessment of Chicago air pollution through analysis of covariance. Environmental Science and Technology 18, 1623–1631. Skoog, D.A., West, D.M., Holler, F.J., Crouch, S.R., 2000. Analytical Chemistry: An Introduction. Saunders College Publishing, Fort Worth, TX, 155pp. Solomon, P.A., Salmon, L.G., Fall, T., Cass, G.R., 1992. Spatial and temporal distribution of atmospheric nitric acid and particulate nitrate concentrations in the Los Angeles area. Environmental Science and Technology 26, 1594–1601. Tolocka, M.P., Solomon, P.A., et al., 2001. East versus West in the US: chemical characteristics of PM2.5 during the winter 1999. Aerosol Science and Technology 34, 88–96. United States Environmental Protection Agency, 2005. List of designated reference and equivalence methods. http://www. epa.gov/ttnamti1/files/ambient/criteria/ref1005.pdf Walcek, C.J., Yuan, H., Stockwell, W.R., 1997. The influence of aqueous-phase chemical reaction on ozone formation in polluted and nonpolluted clouds. Atmospheric Environment 31, 1221–1237.