Self Improvement

Unraveling the complex local-scale flows influencing ozone patterns in the southern Great Lakes of North America

Description
Unraveling the complex local-scale flows influencing ozone patterns in the southern Great Lakes of North America
Published
of 21
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Atmos. Chem. Phys., 10, 10895–10915,2010www.atmos-chem-phys.net/10/10895/2010/ doi:10.5194/acp-10-10895-2010© Author(s) 2010. CC Attribution 3.0 License. AtmosphericChemistryand Physics Unraveling the complex local-scale flows influencing ozone patternsin the southern Great Lakes of North America I. Levy 1 , P. A. Makar 1 , D. Sills 2 , J. Zhang 1 , K. L. Hayden 1 , C. Mihele 1 , J. Narayan 1 , M. D. Moran 1 , S. Sjostedt 1 , andJ. Brook 11 Air Quality Research Division, Science and Technology Branch, Environment Canada, 4905 Dufferin Street, Toronto,Ontario, Canada 2 Cloud Physics and Severe Weather Research Section, Environment Canada, 4905 Dufferin Street, Toronto, Ontario, CanadaReceived: 9 July 2010 – Published in Atmos. Chem. Phys. Discuss.: 23 August 2010Revised: 2 November 2010 – Accepted: 15 November 2010 – Published: 22 November 2010 Abstract. This study examines the complexity of variousprocesses influencing summertime ozone levels in the south-ern Great Lakes region of North America. Results from theBorder Air Quality and Meteorology (BAQS-Met) field cam-paign in the summer of 2007 are examined with respect toland-lake differences and local meteorology using a large ar-ray of ground-based measurements, aircraft data, and sim-ulation results from a high resolution (2.5km) regional air-quality model, AURAMS.Analyses of average ozone mixing ratio from the entireBAQS-Met intensive campaign period support previous find-ings that ozone levels are higher over the southern GreatLakes than over the adjacent land. However, there is greatheterogeneity in the spatial distribution of surface ozone overthe lakes, particularly over Lake Erie during the day, withhigher levels located over the southwestern end of the lake.Modelresultssuggestthatsomeoftheseincreasedozonelev-els are due to local emission sources in large nearby urbancenters. While an ozone reservoir layer is predicted by theAURAMS model over Lake Erie at night, the land-lake dif-ferences in ozone mixing ratios are most pronounced duringthe night in a shallow inversion layer of about 200m abovethe surface. After sunrise, these differences have a limitedeffect on the total mass of ozone over the lakes and land dur-ing the day, though they do cause elevated ozone levels in thelake-breeze air in some locations.The model also predicts a mean vertical circulation duringthe day with an updraft over Detroit-Windsor and downdraftover Lake St. Clair, which transports ozone up to 1500 mabove ground and results in high ozone over the lake. Correspondence to: J. Brook (jeff.brook@ec.gc.ca)Oscillations in ground-level ozone mixing ratios were ob-served on several nights and at several ground monitoringsites, with amplitudes of up to 40ppbv and time periods of 15–40min. Several possible mechanisms for these oscilla-tions are discussed, but a complete understanding of theircauses is not possible given current data and knowledge. 1 Introduction Land and sea/lake breeze winds play an important role indetermining many aspects of coastal environments, and airquality in particular. Air pollutants levels in coastal regions,that hold a large part of the world’s population along withemission sources, are the outcome of the combined effectof several factors, such as long-range transport, short-rangetransport, local emissions, photochemical activity, and sur-face properties (Banta et al., 2005). While the main drivingforce of the breeze winds is the time-varying temperature dif-ference between land and sea/lake, other factors have beenshown to interact with these winds, such as large-scale syn-optic flow (e.g., Oh et al., 2006), the urban heat island (e.g.,Ohashi and Kida, 2002), vicinity to a mountain ridge (e.g.,Lu and Turco, 1994), curvature of the shoreline (e.g., Alpertand Getenio, 1988; Levy et al., 2008b), and other topograph-ical features such as vegetation, land use and land cover.The effect of sea/lake breeze winds on air pollution hasbeen shown in many locations around the world, such asAthens (e.g., Kallos et al., 1993; Kambezidis et al., 1998);the Iberian peninsula (Millan et al., 2000); southern France(e.g., Lasry et al., 2005); Israel (e.g., Alper-Siman Tov et al.,1997; Levy et al., 2008a); Taiwan (e.g., Cheng, 2002; Liuet al., 2002); Korea (Oh et al., 2006), southwestern BritishPublished by Copernicus Publications on behalf of the European Geosciences Union.  10896 I. Levy et al.: Unraveling the complex local-scale flowsColumbia (Brook et al., 2004), and Los Angeles (Lu andTurco, 1994, 1995). The large populations living in thesecoastal cities are often exposed to high pollution levels dueto higher emissions and may experience greater health im-pacts due to higher chronic exposures accentuated or eveninduced by the complex local meteorology. Thus, this mete-orology can lead to more complex yet potentially repeatableexposure patterns.The impact of the North American Great Lakes on pollu-tion levels has been studied since the 1960’s (e.g., Mukam-mal, 1965; Lyons and Cole, 1973, 1976; Anlauf et al., 1975;Keeler et al., 1990; Sillman et al., 1993; Dye et al., 1995;Hanna and Chang, 1995; Hastie et al, 1999; Cooper et al.,2001; Fast and Heilman, 2003, 2005). These multi-year ob-servations combined with model output suggest that ozone issystematically higher over the lakes. Dye et al. (1995) ana-lyzed observations from the Lake Michigan Ozone Study in1991, and found the highest ozone concentrations occur ina shallow and cool conduction layer over the lake. Ozoneprecursors emitted during the night and early morning weretrapped in this shallow layer of cool and stable air over thewater to react and produce high concentrations of ozone dur-ing the day. Hastie et al. (1999) reported increases of 30ppbvin ozone levels measured in air masses arriving with the LakeOntario lake-breeze front, measured at two ground stationsnorth of the lake and from an aircraft. The authors postu-late that the srcin of these polluted air masses is emissionsfrom source areas that are entrained over the lake by theland breeze at night. Fast and Heilman (2005) used the PE-GASUS air-quality model to evaluate ozone levels over theGreat Lakes region for two summers (May–September 1999and 2001). Ozone exceedances above 60 and 80ppbv werefound to be higher on average over the southern Great Lakesthan over land. Similar findings were published by Cappset al. (2010) for a 14-day period in August 2002 using theCMAQ air-quality model.The magnitude of the land-lake differences in ozone arenot well characterized due to a lack of routinely availablepollutant data of fine spatiotemporal resolution. For exam-ple, Hastie et al. (1999) reported that ozone levels rangedfrom 60ppbv to 100ppbv over Lake Ontario during mea-surements with an instrumented aircraft on 26 August 1993around noon, while NO x ranged from 1.5ppbv to 6ppbv. Ina subsequent flight about one hour later, ozone levels werewithin the same range, but the location of the maximumozone had changed, suggesting that both spatial and tempo-ral changes, perhaps related to advection patterns, play animportant part.During the Border Air Quality and Meteorology (BAQS-Met) study, which took place in summer 2007 in southwest-ernOntario, Canada, acomprehensivesuiteofair-qualityandmeteorological measurements were made from both fixedand mobile platforms in order to study air pollutant trans-port and transformation in relation to lake and land breezesand to gain more insight into the nature of the high ozoneover and near the lake in particular. The study region is sit-uated between two of the Great Lakes, Lake Erie and LakeHuron, and a third, smaller lake, Lake St. Clair (see Fig. 1).This paper describes some of the results from this study thatare related to ozone levels over Lake Erie and Lake St. Clair.First, average ozone mixing ratios from ground-level mea-surementsandfromtheAURAMSregionalair-qualitymodelare examined in time and space. Then, a case study is pre-sented to illustrate the influence of local meteorological cir-culations driven by the southern Great Lakes on air qual-ity. Included in the analysis are measurements from a meso-network, a chemistry supersite, an instrumented aircraft, aninstrumented ferry, an instrumented buoy, and a tethersonde.The goal of this paper is to elucidate the complexity of various processes influencing summertime ozone levels inthe southern Great Lakes region of North America. Morespecifically, it examines the spatio-temporal variability inthe levels of surface ozone and related air pollutants, withrespect to precursors emission sources, regional- (synop-tic) and local- (land-lake breeze) scale meteorology, differ-ing processes over land and water, and the vertical structureof ozone. The analyses, which are unique in the extent towhich detailed measurements and model applications havebeen combined and interpreted, provide insight regarding thebehaviour and significance of the elevated ozone levels overthe complex coastal environment. 2 Datasets The BAQS-Met field campaign was conducted during thesummer of 2007 to study the effect of mesoscale meteorol-ogy on air pollution in the southwestern Ontario. Figure 1provides an overview of the study area and measurement sta-tion locations used in this paper. In addition to routine air-quality monitoring stations operated by the Province of On-tario (Thermo Scientific instrumentation with ozone loggedat 1min resolution) and the State of Michigan (1h resolutionfor ozone), spatial detail was enhanced by deploying a meso-network of ozone and meteorological instruments. This wasin operation before, during and after the intensive observingperiod of the study (20 June to 10 July) to obtain 1–5mintime-resolved measurements for the late spring to late sum-mer period. Ozone mixing ratios at the meso-network siteswere measured using low-power monitors from 2B Tech-nologies (Model #202). Ozone instruments were also in-stalled on a buoy in Lake Erie and on a ferry operating inLake Erie between Leamington and Pelee Island (ThermoScientific instrumentation with ozone logged at 1min reso-lution).Additional detailed chemical measurements were madewith research-grade instruments at three chemistry super-sites (Harrow, Bear Creek, and Ridgetown) which operatedfor the intensive period (e.g., Stroud et al., 2010). Thesesites used instruments from Thermo Scientific to measureAtmos. Chem. Phys., 10, 10895–10915,2010 www.atmos-chem-phys.net/10/10895/2010/   I. Levy et al.: Unraveling the complex local-scale flows 10897 !.!.!.!.!.!.!. DetroitBuffaloTorontoLansingColumbusClevelandPittsburgh 0 15075 km ± ####### n " # ^ X ^^ """""! WNWWAROAKDET ALP BEC HARPEIXRGESSWNDECOLIGWHELEAPAQSOMCROPAL Lambton AURAMS 2.5km domainAURAMS sub-regionHighwayFerry n Buoy X Pelee Is. # Rural sites ^ Super Sites " Urban sites DetroitToledoCleveland 0 40 8020 km Sarnia   L a  k e   E r  i e LSC LH  LO  Windsor  Fig. 1. Map of study region, indicating (left) model analysis sub-domain and the locations of chemistry supersites (BEC – Bear Creek, XRG– Ridgetown, HAR – Harrow), monitoring stations, buoy (ECO), ferry route, and main urban areas (gray shaded regions), and (right) the2.5-km AURAMS model domain and larger setting. LSC marks Lake St. Clair; LH marks Lake Huron, and LO marks Lake Ontario. O 3 (Model 49C), SO 2 (Model 43C), CO (Model 48 TLE)and NO/NO x (Model 42C). A standard chemiluminescence-based NO/NO x instrument (Thermo Scientific Model 42C)was modified in-house to conduct continuous measurementsof NO, NO 2 and NO y with 1min time resolution. Gas phasevolatile organic compounds were measured at Harrow with5 min resolution by an Ionicon Proton Transfer Mass Spec-trometer (PTR-MS) (Gouw and Warneke, 2007). An instru-mented Twin Otter aircraft was also deployed for approxi-mately 30h of flight time measuring gas and particle phasepollutants as well as meteorology at a 1 s time resolution.These measurements are described in detail in Hayden etal. (2010). At the Ridgetown supersite, a Vaisala Tether-sondesystemconsistingofthreeECCozonesondesfromSci-ence Pump (model 6A) and tethered meteorological sondes(model # TTS111/RSS911) made measurements from thesurface up to about 1000 m above ground level (a.g.l.) onmultiple days during the period.To better understand the BAQS-Met observations, outputfrom the GEM/AURAMS regional air-quality modeling sys-tem (Cho et al., 2009; Cˆot´e et al., 1998; Makar et al., 2009),which was run with three nested grids with 42, 15 and 2.5kmhorizontal grid spacing (Fig. 1), are also analyzed in this pa-per. A more detailed description of the model set up anda comprehensive evaluation of the model’s performance forBAQS-Met is given by Makar et al. (2010a, b).Themainpurposeofthemodelapplicationspresentedherewas to obtain a more detailed interpretation of the pollutantbehavior in the study region and to assess the prevalenceand/or validity of the commonly observed and modeled highozone levels over the lakes. Although the model has lim-itations, it also has unique value in providing a complete,physically-consistent, four-dimensional description of the at-mosphere during BAQS-Met that is not possible with mea-surements. Recognizing the model is imperfect, an attemptis made in this paper to use the model predictions in conjunc-tion with available measurements so as to provide broader in-sight in a manner that accentuates the model strengths (e.g.,long term average fields as opposed to specific measurementperiods) and provides complementary information that helpsto interpret the measurements. Our focus here has not beento evaluate the model (cf. Makar et al., 2010a, b) but to useit to help understand the conditions in the region, in light of available measurements. Nonetheless, the reader needs to beaware that what is displayed in several cases is model outputand we have tried to make this distinction clear in the text.Model output was analyzed with the open-source statisticallanguage R (RDCT, 2009) and visualized with the UnidataIntegrated Data Viewer (IDV) (Murray et al., 2003).Diurnal averages of measured ozone were calculated at 11rural sites and 6 urban sites, as well as for the Buoy and PeleeIsland sites in Lake Erie (shown in Fig. 1, listed in Table 1).For consistency between different sites, averages were calcu-lated for the Buoy operational period only (26 June–10 July2007). Diurnal averages were created by calculating 2-h run-ning averages of ozone on a 1-min basis (1-h for Michigansites) for each site during the entire period, then averagingby time of day for each site. The 2-h period was selectedfor consistency with Michigan sites. Composite diurnal av-erages were then obtained by averaging over sites groupedby type, i.e., rural and urban, as described in Table 1. Di-urnal change rates in ozone concentration were calculatedby hour as the difference between the averages of two con-secutive hours. Also, the AURAMS predicted ozone levelsfor the near-surface layer during the buoy operational periodwere extracted over a relevant analysis sub-domain aroundwww.atmos-chem-phys.net/10/10895/2010/ Atmos. Chem. Phys., 10, 10895–10915,2010  10898 I. Levy et al.: Unraveling the complex local-scale flows Table 1. List of sites used for ozone analysis.Name Abbreviation Type Latitude Longitude Network Palmyra PAL Rural 42 ◦ 26.525  N 81 ◦ 44.370  W MESONetCroton CRO Rural 42 ◦ 36.571  N 82 ◦ 04.812  W MESONetSombra SOM Rural 42 ◦ 41.780  N 82 ◦ 25.781  W MESONetPaquette Corners PAQ Rural 42 ◦ 11.717  N 82 ◦ 57.922  W MESONetLeamington LEA Rural 42 ◦ 04.121  N 82 ◦ 36.826  W MESONetWheatley WHE Rural 42 ◦ 07.978  N 82 ◦ 23.731  W MESONetLighthouse Cove LIG Rural 42 ◦ 17.499  N 82 ◦ 31.343  W MESONetEssex ESS Rural 42 ◦ 09.600  N 82 ◦ 50.000  W OME 1 Ridgetown (Supersite) XRG Rural 42 ◦ 27.200  N 81 ◦ 53.268  W OMEHarrow (Supersite) HAR Rural 42 ◦ 01.978  N 82 ◦ 53.603  W EC 2 , OMEBear Creek (Supersite) BEC Rural 42 ◦ 32.153  N 82 ◦ 23.352  W OMEPelee Island PEI Lake 41 ◦ 46.767  N 82 ◦ 40.220  W OMELake Erie Buoy ECO Lake 42 ◦ 02.005  N 82 ◦ 59.006  W IADN 3 Windsor Downtown WND Urban 42 ◦ 18.983  N 83 ◦ 02.664  W OMEWindsor West WNW Urban 42 ◦ 17.567  N 83 ◦ 04.400  W OMEAllen Park ALP Urban 42 ◦ 13.700  N 83 ◦ 12.000  W MI 4 Detroit DET Urban 42 ◦ 18.250  N 83 ◦ 06.000  W MIOak Park OAK Urban 42 ◦ 27.784  N 83 ◦ 10.998  W MIWarren WAR Urban 42 ◦ 30.800  N 83 ◦ 00.000  W MI 1 OME - Ontario Ministry of the Environment; 2 EC – Environment Canada; 3 IADN - Integrated Atmospheric Deposition Network; 4 MI – Michigan State . southwestern Ontario, Lake St. Clair, Detroit, and the north-western quadrant of Lake Erie (Fig. 1), and model grid cellswere grouped by different surface type (i.e., rural, lake, andurban).Throughout the paper, Eastern Daylight Time (EDT) isused to present the data, which is one hour ahead of East-ern Standard Time (EST) and four hours behind CoordinatedUniversal Time (UTC). 3 Results and discussion3.1 Diurnal pattern of surface ozone To examine the diurnal behavior of ground-level ozone overthe lakes as compared to land, 2-h averages of ozone andtheirhourlychangerateswerecalculatedfortheBuoy(repre-senting the lake), Pelee Island (representing lake influencedconditions), and the groups of 11 rural and six urban sites(Fig. 2). Ozone measurements over Lake Erie were alsomade by a monitor installed for BAQS-Met on the Leam-ington – Pelee Island ferry, but due to its routine move-ment it was not possible to compute relevant diurnal aver-ages. The ferry measurements were therefore averaged bytrip times (approximately 90min for each trip), for four sail-ings that were scheduled for the same departure times on allweekdays (Fig. 2a). Thus only its over-lake measurementswere combined for comparison with and to further supportthe buoy and Pelee Island measurements. Although differ-ences in ozone concentrations between land and lake weresuggested in previous studies, this is the first time a detaileddiurnal pattern and change rate are available at high temporalresolution and compared between lake, rural and urban sites.Figure2showstheaveragedozonediurnalpattern(2a)andchange rates (2b) measured at the different site types, as wellas the average ozone diurnal pattern (2c) and change rates(2d) predicted for various surface types for the same periodby the AURAMS model. The nighttime (00:00–06:00EDT)ozone mixing ratio over the lake (buoy) is ∼ 5ppbv higherthan the Pelee Island site, ∼ 15ppbv higher than the rural-site average, and close to 25ppbv higher than the urban-siteaverage. At 10:00EDT, average ozone levels at the rural andurban sites match those of the buoy. Higher ozone concentra-tionsoverthelakeatnightareduetoalackoffreshNOemis-sions, particularly in comparison to urban areas, and lowerdeposition rates to water surface than land surfaces (e.g., Dyeet al., 1995; Sillman et al., 1993). Examination of the morn-ing (09:00EDT) rate of increase in ozone at the buoy, shownin Fig. 2b, indicates that it is slower over water than overland (e.g., 1.5 vs. 3.4, 5.8, and 7.5ppbvh − 1 , for the buoy,Pelee Island, rural, and urban sites, respectively). These dif-ferences are due to greater vertical mixing over land whichquickly replenishes the ozone at the surface in the morning,particularly over the urban areas where ozone experiencesAtmos. Chem. Phys., 10, 10895–10915,2010 www.atmos-chem-phys.net/10/10895/2010/   I. Levy et al.: Unraveling the complex local-scale flows 10899 706050403020100    M  e  a  s  u  r  e   d  o  z  o  n  e   (  p  p   b  v   ) 06:0012:0018:0000:0006:00 BuoyPelee IslandRural sitesUrban sitesFerry 706050403020100    M  o   d  e   l  e   d  o  z  o  n  e   (  p  p   b  v   ) 06:0012:0018:0000:0006:00 Entire regionLand surfaceLake surfaceUrban surface -12-8-404812    M  e  a  s  u  r  e   d  c   h  a  n  g  e  r  a   t  e   (  p  p   b  v   h  r   -   1    ) 06:0012:0018:0000:0006:00Time of day (EDT) BuoyPelee IslandRural sitesUrban sites -12-8-404812    M  o   d  e   l  e   d  c   h  a  n  g  e  r  a   t  e   (  p  p   b  v   h  r   -   1    ) 06:0012:0018:0000:0006:00Time of day (EDT) Entire regionLand surfaceLake surfaceUrban surface a)b)c)d) Fig. 2. (a) 2-h running averages of surface ozone for the Buoy, Pelee Island, rural-site composite, urban-site composite, and the average of ferry measurements for trips scheduled on all weekdays; (b) running change rate between two consecutive hours of the measured data; (c) AURAMS diurnal averages at surface level over the 2.5-km grid sub-domain in Fig. 1 grouped by lakes, land and urban areas; (d) runningchange rates between two consecutive hours of the model data. the greatest depletion at night. This behavior suggests thatthe nighttime differences between lake and land are limitedto a shallow layer above the surface since within a few hoursafter sunrise mixing from aloft results in a minimum in spa-tial variability in surface ozone from 10:00–11:00EDT.Previous studies have shown that the break up of the sur-face inversion and vertical mixing with the ozone residuallayer aloft has a significant impact on morning build up of ground-level ozone, whereas photochemical production pro-cesses augment surface ozone levels later in the day (Zhangand Rao, 1999). Zhang and Rao (1999) found compara-ble summer mean ozone build-up rates at a rural site, withthe highest rates ( ∼ 6.5ppbvh − 1 ) at 10:00EDT, compared to10ppbvh − 1 at 11:00EDT at an urban site in New York City.Kleinman et al. (2002) used a chemical box model to esti-mate ozone build-up rates at five large cities in the UnitedStates. The authors found the highest rates of 11.3ppbvh − 1 in Philadelphia, PA and Houston TX, compared to 6, 4.3,and 3.5ppbvh − 1 in Nashville, TN, New York City, NY, andPhoenix, AZ, respectively. In this study over southwest-ern Ontario, measured rates of change of ozone mixing ra-tio at the urban sites (Fig. 2b) are between 8ppbvh − 1 (at09:00EDT) and − 6ppbvh − 1 (at 21:00EDT).To compare the measured diurnal cycles of ozone to thosepredicted by AURAMS, an analysis sub-domain of the 2.5-km model grid was selected over the study region (Fig. 1).The AURAMS mean ozone diurnal pattern for the lakes ver-sus land surfaces in the region (Fig. 2c, d) is similar tothe measurements, with a ∼ 15ppbv difference at night andbuild-up rates of 2.7 and 6ppbvh − 1 over lake and land sur-faces, respectively, at 09:00EDT. Urban areas (i.e., mainlyDetroit) over the modeled region show a higher build-up rateof 10.8ppbvh − 1 in the morning (10:00EDT) compared tothe maximum rate of 7.5ppbvh − 1 measured at the urbansites an hour earlier, as well as faster loss rates at night( − 13ppbvh − 1 at 20:30EDT) compared to measurements( − 6ppbvh − 1 at 21:00 EDT). These differences might bedue, at least in part, to the six urban sites being located ator close to the city center, which are thus more influencedby traffic NO x emissions compared to the part of the modeldomain that is assigned to the “urban” group (cf. Fig. 1).From 23:00 to 06:00EDT Fig. 2b shows a general trendof decreasing measured ozone loss rates. In contrast, themodel predicts an almost constant loss rate with a magni-tude of about 2ppbvh − 1 . The result of these differences innighttime change rates is lower predicted ozone mixing ra-tios in the morning (06:00EDT) compared to measurements.Theselowermodelvalues atsunrisearethencompensatedbyhigher predicted versus measured change rates in the morn-ing, particularly over rural (8 vs. 6ppbvh − 1 , respectively)and urban (11 vs. 8ppbvh − 1 , respectively) areas. As a re-sult, the predicted daily maxima in the afternoon are higherthan measured for urban areas (63 vs. 50ppbv, respectively),but are in good agreement for rural areas ( ∼ 58ppbv).Given the limited ability to obtain systematic three-dimensional measurements, particularly over the lakes, andafter showing reasonable agreement between measured andpredicted average ozone levels near the surface in the studywww.atmos-chem-phys.net/10/10895/2010/ Atmos. Chem. Phys., 10, 10895–10915,2010
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks