UBC Faculty Research and Publications

Biogenic isoprene in the Lower Fraser Valley, British Columbia Gurren, Kristina 2011

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


Steyn_1998_98JD01214.pdf [ 930.34kB ]
JSON: 1.0041854.json
JSON-LD: 1.0041854+ld.json
RDF/XML (Pretty): 1.0041854.xml
RDF/JSON: 1.0041854+rdf.json
Turtle: 1.0041854+rdf-turtle.txt
N-Triples: 1.0041854+rdf-ntriples.txt

Full Text

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 103, NO. D19, PAGES 25,467-25,477, OCTOBER 20, 1998 Biogenic isoprene in the Lower Fraser Valley, British Columbia Kristina Curten and Terry Gillespie Land Resource Science, University of Guelph, Guelph, Ontario, Canada Douw Steyn Geography, University of British Columbia, Vancouver, British Columbia, Canada Thomas Dann and Daniel Wang Environment Canada, River Road, Ottawa, Ontario, Canada Abstract. Tropospheric ozone is formed by photochemical reactions between nitrogen oxides (NO• and volatile organic ompounds. In some regions, biogenic isoprene may be a significant contributor to the production of tropospheric ozone. The contribution ofbiogenic isoprene is an important aspect of regional ozone chemistry as it represents an ozone precursor that cannot be eliminated through emissions controls. The purpose of this study was to evaluate the contribution of isoprene to the production of tropospheric ozone in the Lower Fraser Valley, British Columbia. Seasonal trends and diurnal profiles were used to examine isoprene's relationship with temperature, to determine its source, and to investigate the chemical and physical factors that limit the ambient levels of isoprene present in the region. Total isoprene levels in the Lower Fraser Valley were low, and evidence suggested that a substantial fraction originated from anthropogenic rather than biogenic sources. Diurnal isoprene profiles were generally flat, and the times of the highest concentrations did not coincide with peak NO x levels nor with the times of optimal ozone- producing meteorological conditions. These results are consistent with those of previously reported studies and suggest that biogenic isoprene may not be as important o the tropospheric ozone chemistry in the Lower Fraser Valley as it is in some southern U.S. cities. 1. Introduction Tropospheric ozone is a major component of urban smog that affects the health of people, crops, and natural vegetation. Tropospheric ozone is formed by photochemical reactions between nitrogen oxides (NO•) and volatile organic ompounds (VOCs). While most NO x originates from anthropogenic sources, VOCs may be emitted from both anthropogenic and biogenic sources. In recent years, ithas become apparent that biogenic hydrocarbons, particularly isoprene (C•Hs), may be significant contributors to the production f tropospheric ozone in some areas. The contributions of biogenic hydrocarbons arean important aspect of regional ozone chemistry as they represent an ozone precursor that cannot be eliminated through emissions controls. Previous tudies that measured biogenic hydrocarbon emission rates from several species of vegetation showed that the Lower Fraser Valley contains few species of agricultural crops, natural plants, or trees that emit isoprene in appreciable amounts [Drewitt, 1996; Drewitt et al., 1998; Curten, 1998]. This suggests that ambient biogenic isoprene l vels in the region should be generally ow. It is possible, however, that there are isoprene-emitting species ofplants present in the area that were not accounted for in the emissions studies and that ambient biogenic isoprene l vels near these sources could be high enough to have a significant impact on the production of tropospheric ozone in the region. This paper describes the methods and results from two analyses of ambient isoprene levels in the Lower Fraser Valley, British Columbia. The first study is an analysis of seasonal isoprene trends in the region. Copyright 1998 by the American Geophysical Union. Paper number 98JD01214. 0148-0227/98/98JD-01214509.00 The purpose of this study was to examine the ambient concentrations of isoprene throughout the year at various ites in the Lower Fraser Valley to determine if significant ambient concentrations f i oprene exist in the region; to examine isoprene's relationship with temperature; and to determine whether isoprene in the region oqginated from anthropogenic or biogenic sources. The second study served to examine the diurnal variation in isoprene concentrations during summer ozone episode and non episode days in the Lower Fraser Valley and investigated the chemical nd physical factors that produce the observed diurnal isoprene profiles through the use of a simple box model. The results of these and previously published studies were used to evaluate the impact of biogenic isoprene in the Lower Fraser Valley on the regional production oftropospheric ozone. 2. Background The Lower Fraser Valley is a roughly triangular valley in southwestern British Columbia that extends from the Strait of Georgia in the west to the Fraser Canyon in the east. The valley is bounded by the Coast Mountains to the north and the Cascade Mountains to the southeast and is inhabited by approximately 2 million people [Steyn et al., 1997]. The bulk of the anthropogenic ozone precursor emissions arises from mobile sources, primarily light-duty vehicles [Steyn et al., 1997]. The region's natural vegetation is largely made up of coniferous trees, including western red cedar •huja plicate), coastal Douglas fir (Psudotsuga menziesii ssp. menziesii), and coastal hemlock (Tsuga rnertensiana) [Drewitt, 1996; Drewitt et al., 1998]. The valley floor has a range of urban and agricultural uses that results in considerable variation in local biogenic emissions [Singleton et al., 1996]. Although ambient isoprene concentrations in rural areas tend to be higher than those in urban areas due to the greater density of isoprene 25,467 25,468 CURREN ET AL.: BIOGENIC ISOPRENE IN THE LOWER FRASER VALLEY emitting plants in rural areas, significant ambient isoprene concentrations have been reported in urban areas. Midday summertime isoprene concentrations as high as 8 ppbv have been recorded at in Atlanta, Georgia, although measurements of the order of 2.5 ppbv are more common [Cardelino and Chameides, 1995]. Ambient isoprene levels measured in Toronto, Ontario, showed that isoprene levels increased uring the summer months in response to increasing emissions from biogenic sources. Summer isoprene concentrations at this site varied, but were consistently less than 1 ppbv [McLaren et al., 1996]. Isoprene concentrations measured at urban and suburban sites in the Lower Fraser Valley are generally less than 0.4 ppbv, with suburban sites showing slightly higher mixing ratios (0.2-0.35 ppbv) than urban sites (<0.2 ppbv) [Canadian Council of Ministers of the of the Environment (CCME), 1997]. Isoprene is generally considered to originate from biogenic sources. Studies in Louisiana and California have shown that the temperature dependence of ambient isoprene concentrations was consistent with laboratory-measured temperature dependence of biogenic emissions [National Research Council (NRC) , 1991], that isoprene concentrations were negatively correlated with VOCs normally associated with anthropogenic mobile sources [Chameides et al., 1992], and that the diurnal profiles of isoprene were temporally out of phase with those of anthropogenic compounds [Chameides et al., 1992]. At some sites, however, isoprene may have an anthropogenic source. Detectable levels of isoprene were measured at York University during the winter months when biogenic sources were dormant under snow cover. An analysis of this data showed a significant correlation between isoprene and 1,3 butadiene, a well known combustion product, indicating that the isoprene levels observed in the winter months originated from anthropogenic mobile sources [McLaren et al., 1996]. Chemical mass balance modeling of this data indicated that the observed winter isoprene concentrations could be accounted for by vehicle exhaust containing isoprene at a level of 0.0018% of nonmethane hydrocarbon mass in the exhaust [McLaren et al., 1996]. ß Similarly, an anthropogenic source of isoprene has been suggested for the Lower Fraser Valley. A correlation between ambient isoprene concentrations at several of the sampling sites in the Lower Fraser Valley and the corresponding isoprene mission rates predicted by PC-BEIS [Pierce and Waldruff, 1991 ] showed a general trend of increasing concentrations with increasing ambient emissions [Singleton et al., 1996], but isoprene concentrations at a site in the urban center of Vancouver appeared to be anomalously high. Possible explanations proposed to explain the reading from this site included confounding effects of very local biogenic sources or that some of the measured isoprene originated from vehicle exhaust [McLaren et al., 1996; Singleton et al., 1996]. PC-BEIS does not include anthropogenic isoprene inthe vehicle exhaust profile [Singleton et al., 1996]. In addition to isoprene's absolute concentration, the diurnal patterns of isoprene emissions must be considered when evaluating its contribution to the production of tropospheric ozone. The diurnal variations in isoprene and other ozone precursor concentrations are important considerations because all precursors may not be present in appreciable concentrations in the ambient atmosphere at the same time. Data collected at several urban sites in the southern United States showed that anthropogenic hydrocarbons and NO• both peaked during the early morning and evening rush hours [Chameides et al., 1992]. Similarly, total nonmethane hydrocarbons and NO x mixing ratios measured in the Lower Fraser Valley showed minima in the afternoon due to enhanced mixing and peaks during the evening and morning rush hours when emissions were high and the mixed layer was shallow [CCME, 1997]. Isoprene mitted from biogenic sources, on the other hand, may show a different diurnal profile from anthropogenic compounds. Isoprene mixing ratios measured at rural forest sites, where areal emission rates are high, are generally low in the early morning, rise sharply at sunrise, and continue to increase throughout the day in response toincreasing temperatures [Trainer et al., 1987; Fehsenfelt et al., 1992]. Isoprene profiles measured at urban sites in the Atlanta, Georgia, region showed a different shape from those at forest sites, with isoprene mixing ratios at all but one of these sites roughly constant hroughout most of the day [Cardelino and Chameides, 1995]. Three of the sites howed slightly elevated levels of isoprene in the morning, possibly due to active biogenic sources combined with the relatively shallow mixing layer, while two of the sites showed dramatic increases in isoprene levels late in the afternoon, probably as the result of the collapse of the mixing layer and decreasing concentrations of compounds that act as isoprene sinks. The profile measured at the Fort McPherson site was similar to profiles over forests at rural sites, with isoprene levels higher than 5 ppbv for most of the day. No explanation was given for the consistently high levels of isoprene measured at this site [Cardelino and Chameides, 1995]. 3. Methods 3.1 Data Collection The data used in this analysis were obtained from the national field sampling program database maintained by the Pollution Measurement Division of the Environment Protection Service, Environment Canada [Dann et al., 1994; CCME, 1997]. Systematic year round measurements at several urban sites and one suburban site in the Lower Fraser Valley began in 1989, with several nonurban sites added to the program in later years (Table 1). Sites were classified as urban, urban street (close to street level in the urban core), or suburban (in an urban area but outside the urban core), and note was made of sites that were potentially impacted by industrial emissions. No data were available from rural sites in the region. Field samples were collected by provincial or municipal environment departments participating in the National Air Pollution Surveillance program. Samples were normally collected over a 24 hour period once every 6 days, with more intensive sampling programs conducted at Rocky Point Park during the summers of 1992 and 1993. In addition to the 24 hour samples, eight sequential 3 hour samples were collected at various sites in the Lower Fraser Valley on selected days in 1992 and 1993 [CCME, 1997]. Whole air samples were collected in 6 L electropolished stainless teel canisters and were analyzed in Environment Canada's Ottawa laboratory. Studies have shown that VOC mixtures can be stored for up to 30 days in these canisters with little change in composition [Oliver et al., 1986]. The whole air samples were analyzed using a cryogenic preconcentration technique and a gas chromatograph coupled with a flame ionization detector or a mass selective detector. The analytical system was based Table 1. Ambient NMHC Measurement Sites in the Lower Fraser Valley Site Site Industrial Staa Date Classification Source Kensington Park urban refinery Jan. 4, 1989 Rocky Point Park urban refinery Jan. 4, 1989 Robson Street urban street Jan. 28, 1989 Surrey East suburban Jan. 10, 1989 Richmond South urban Jan. 22, 1989 Mahon Park urban May 23, 1990 Shellmount Street suburban pipeline March 18, 1990 transfer Lan[•ley suburban July 12, 1993 CURREN ET AL.: BIOGENIC ISOPRENE IN THE LOWER FRASER VALLEY 25,469 on the methods described by Winberry et aL [1988] and is further described by Dann et al. [1994] and Bottenheim et al. [1997]. Typical detection limits for this method were 0.02-0.04 ppbv; species values that were below detection level were set to zero [CCME, 1997]. The reported precision of the method for analytes with concentrations greater than 0.25 ppbv was in the range of 10-15% [Dann et al., 1994]. 3.2 Seasonal Data Analysis Seasonal trends in isoprene mixing ratios were examined by combining all available data for each station. The relationships between isoprene and the daily mean temperature measured at the Abbotsford airport in summer and winter were examined for each station, and regression equations were calculated where appropriate. For the purposes of this analysis, "winter" was defined as the period between November and March and "summer" as the period between May and September. The transitional months of April and October were excluded from the temperature analysis [McLaren et al., 1996]. A large number of data points were available for Rocky Point Park (n - 306), with considerably fewer data available for other stations. Regressions between ambient concentrations of isoprene and trans-2-pentene and cis-2-butene, VOCs normally associated with anthropogenic mobile sources, were performed totest he hypothesis that isoprene in the Lower Fraser Valley originates atleast partly from anthropogenic sources. The data were logarithmically transformed for the regressions to ensure constant variance throughout the range of measured concentrations. I  order to perform the transformations, zero readings were replaced with concentrations equal to half the detection limit (0.02 ppbv). Separate r gressions were performed for the winter and summer months. Again, considerable data were available for Rocky Point Park (n -- 359), while far fewer data were available for the other sites. 3.3 Diurnal Data Analysis Diurnal isoprene profiles at two sites, Rocky Point Park and Langley, were constructed fornon episode days and for selected days during the mild ozone episode of 1993. Average non episode day profiles were created by combining data for days with similar daily maximum temperatures, while profiles for the episode days represent a single day's data. 3.4 Box Model Diurnal isoprene profiles in the Lower Fraser Valley were reproduced using a simple box model. For the purposes of this analysis, i oprene advection, deposition, and entrainment from above the mixing layer were assumed to be zero. Under these assumptions, the concentration of isoprene within the 1 km x 1 km box was dependent upon the initial concentration f isoprene at the start of the day, the areal anthropogenic and biogenic isoprene source strengths, the depth of the mixing layer, and the strength of atmospheric isoprene sinks. Each of these factors will be discussed below in turn. 3.4.1. Initial concentration of isoprene. The diurnal isoprene profiles at Rocky Point Park and Langley indicated that low concentrations of isoprene xisted in the ambient atmosphere during pre dawn hours. Isoprene concentrations measured at 0300 hours (representing the average mixing ratio from 0300-0600 hours) were used as the initial concentration within the box at the start of the day. 3.4.2. Areal isoprene source strength. Biogenic and anthropogenic sources of isoprene were included in the model. Biogenic isoprene mission rates are primarily dependent upon temperature and illumination. The length of day and the temperature profile for the day were calculated using the model of Parton and Logan [1981], which assumes a sinusoidal temperature trend from sunrise to sunset and an exponential decay from sunset o the following sunrise. The radiation profile was assumed to be sinusoidal, increasing from zero at sunrise to a maximum of 1600 !amol m '2 s 'l at noon and returning to zero at sunset [Oke, 1978]. Cloud cover was assumed to be zero for the non episode days and for 2 of the 3 episode days (August 2, 1993 (Langley), and August 5, 1993 (Rocky Point Park)). In aexordanee with recorded meteorological data during Pacific '93, observations of cloud cover for August 4, 1993 (Langley) were used to reduce the illumination at that site. Temperature- and illumination-dependent isoprene mission rates were calculated using the following algorithms from Guenther et al. [1991]: E = S x T'x L' (1) where E is the isoprene mission rate (tag m '2 h'l); S is the mean isoprene emission rate at 301 K (tag m '2 h4); T' is the correction factor for temperature; and L' is the correction factor for illumination. T '= exp [T• (T•_- TsYR T, T_•s] 1 + exp [T 2 (T L - T3)/R T L Ts] (2) where Ti• is the leaf temperature = air temperature; T is •the normalizing temperature (301 K); R = 8.314 JK 4 mo14; T• = 95,100 J mol't; T 2 = 231,000 J mo14; and T3 = 311.83 K. L '= x- (x 2 - 4xfxlxL1). (3) 2L• where x = fxI + L• + L2; f is the fraction of light absorbed by chloroplasts (= 0.385); I is the irradiance (tamol m '2 s'l); L• = 105.6 tamol m '2 s'l; and L 2 = 6.12 tamol m '2 s 't. Correction factors for humidity and CO 2 mixing ratios were neglected as these corrections are very small. The mean areal isoprene emission rate (S) for the Lower Fraser Valley at 301 K has been estimated at226 tag m '2 h 4 [Drewitt, 1996]. Visual inspection of a satellite image of the region revealed that the density of foliage in the vicinity of the Langley site was greater than that surrounding Rocky Point Park. Accordingly, the mean areal biogenic isoprene emission rate for Langley was set at 226 tag m '2 h 't for this analysis, while the rate for Rocky Point Park was set at 169.5 tag m '2 h 'l. Hourly estimates of anthropogenic isoprene missions were made by assuming that anthropogenic soprene is emitted only by mobile sources (automobiles). Based on the results of the seasonal isoprene analysis, the total amount of anthropogenic isoprene mitted in the region over the course of a day was assumed to be 25% of the value of the total biogenic isoprene emitted in the region on a day of moderate temperature. Thus the gross amount of anthropogenic isoprene mitted in the Lower Fraser Valley was assigned a value of 9 x 10 •ø tag d 'l, or 225 tag m '2 d 4. This total amount was partitioned into hourly emissions using the relative allocation factors from the Pacific '93 VOC emission inventory [Levelton, 1996; Jiang et al. 1996]. Relative allocation factors were available for July 31 to August 6, 1993. For the non episode days, averaged relative allocation factors were used to calculate hourly anthropogenic emission rates, while the factors from specific days were used for the episode days. 3.4.3 Depth of mixing layer. The depth of the modelled mixing layer varied with time of day. The nighttime depth of the mixing layer was 200 m. The mixing layer began to grow shortly after sunrise and followed a sinusoidal curve until its maximum depth was reached at 1500 hours. The mixing layer collapsed 1 hour before sunset o a depth of 200 m. In addition to temporal variation, the depth of the mixing layer over a coastal region also varies spatially. Air flowing from offshore to onshore is modified by changes in temperature contrasts, which produce a thermal or convective internal boundary layer. The growth of the convective internal boundary layer can be represented bythe following equation [Venkatram, 1977; Hsu, 1986]: 25,470 CURREN ET AL.: BIOGENIC ISOPRENE IN THE LOWER FRASER VALLEY h = (2C•0_,•:•0 •)X) ø's (4) (7(1- 2B) ø's where h isthc depth ofthe mixing layer (m); C d is the drag coefficient, 0• is the potential ir temperature over land (ø C); 0 m is the potential air temperature over sea (øC); X is the distance downwind from the shoreline (m); ¾ is the lapse rate above the boundary layer ( ø C/m); and F is the entrainment coefficient. Equation (4) was used to calculate the maximum depth of the mixing layer for the modeled nonepisode ays. The distances ofRocky Point Park and Langley downwind from the shoreline were estimated using the simplified Lower Fraser Valley shoreline depicted by Steyn and Oke [1982, Figure 5], and assuming a west-southwest airflow. The values of the parmneters entered into equation (4) were as follows Cd= 0.012, 0,• = 16-17øC, X = 29.4 km (Rocky Point Park) and 32 krn (Langl•), ¾ = 0.065 øC/m, and F = 0.2. The resulting calculated maximum mixing layer depths varied between 650 and 900 m. The maximum mixing layer depths entered into the model for the ozone episode days of August 2 (Langley), August 4 (Langley), and August 5 (Rocky Point Park) were taken from the measured epths reported by Hayden et al. [1997]. In general, the measured epth of the mixing layer from August 1-5, 1993, ranged from 500-800 m in the center of the valley. 3.4.4. Isoprene sinks. The oxidation of isoprene in the ambient atmosphere is initiated by reaction with OH and lqO3 radicals and with 03. For the pu• of this model, OH radicals were considered as the primary sink for isoprene during the day, and NO3 radicals were considered as the primary sink for isoprene during the night. The reaction between isoprene and ozone is much slower than that between isoprene and OH or NO3, and O3 levels are generally low in the Lower Fraser Valley during nonepisode ays and very low at night; thus the reaction between isoprene and O3 was neglected in this analysis. The OH profile used in this analysis was sinusoidal, with concentrations increasing from zero after sunrise, reaching a peak concentration at oon, and retuming to zero shortly before sunset. The shape of the OH profile was chosen to correspond with the radiation profile as OH is formed by the photolysis of03. OH concentrations are difficult to measure, and no data were available for this analysis. For nonepisode ays, peak concentrations of OH radicals were taken to be l x106 radicals cm 4 (N. Bunce, University of Guelph, personal communication, 1998). Since O3 is required for the production of OH radicals and NO2 is a sink for OH, OH concentrations may be depressed under conditions of moderate O 3 and elevated NOx. The 1993 ozone episode was mild, with peak O3 concentrations at Rocky Point Park and Langley ranging from 54-72 ppbv and peak daytime NOx levels ranging from 3047 ppbv. For these days, the peak OH concentration was taken to be l x10 s radicals em 4 (N. Bunee, University of Guelph, personal communications, 1998). Peak concentrations of NO3 radicals and the shape of night time NO3 profiles can vary substantially from night to night [Platt et al., 1981; Finlayson-Pitts and Pitts, 1986]. For this simulation, NO3 was increased linearly from zero at sunset and reached its peak concentration of 10 ppt 2 hours after the onset of darkness. This profile was chosen as it roughly matched the shape of a measured profile at Deuselbaeh, West Germany, where NO3 levels rose at 1900 hours and reached a peak at 2100 hours [Platt et al., 1981]. In addition, NO3 is likely to be formed from NO2 emitted from the evening rush hour. Peak concentrations of NO3 were maintained until midnight, whereupon they decreased sinusoidally, reaching a concentration of0 ppt at dawn. 4. Results and Discussion 4.1 Seasonal Analysis Mean daily isoprene concentrations a d temperatures throughout the year at Rocky Point Park are shown in Figure 1. In general, isoprene mixing ratios were low, with a mean of 0.16 ppbv (n = 162) 0.6 3O 0.5 0.4 0.3 0.2 0.1 i i i i 0 50 100 150 200 250 300 350 Julian Day [-o-- Isoprene -13-- Tmean I 25 2O 5 m 0 -10 Figure 1. Seasonal isoprene and temperature t nds at Rocky Point Park, 1989-1995 CURREN ET AL.: BIOGEN!C ISOPRENE IN THE LOWER FRASER VALLEY 25,471 Table 2. Mean and Maximum Isoprene Mixin• Ratios Summer Winter Station Mean, Maximum, Mean, Maximum, ppbv ppbv ppbv ppbv Rocky Point Park 0.16 0.44 0.17 0.58 (Urban) (n=162) (n=144) Kensington 0.15 0.33 0.14 0.24 (Urban) (n=28) (n-20) Mahon Park 0.16 0.43 0.09 0.22 (Urban) (n=15) (n=15) Richmond South 0.15 0.46 0.21 0.47 (Urban) (n=22) (n=14) Robson Street 0.10 0.19 0.14 0.77 (Urban - Street) (n=23) (n= 16) Shellmount Street 0.15 0.49 0.09 0.32 (Suburban) (n=41) (n=52) Surrey East 0.10 0.29 0.04 0.18 (Suburban) (n=19) (n=22) Langley 0.22 0.64 0.02 0.05 (Suburban) (n=lO) (n=8) in the summer and 0.17 ppbv (n = 144) in the winter. Peak summer isoprene concentrations of 0.36 ppbv or higher were recorded on five occasions, generally on warm days; the highest summer concentration of isoprene recorded was 0.44 ppbv (August 12, 1992). Winter concentrations of 0.36 ppbv or more were recorded on seven occasions; the highest winter concentration recorded was 0.58 ppbv (December 27, 1993). These results contrast with those at York University, where summer concentrations of isoprene were higher than winter concentrations [McLaren et al., 1996]. Table 2 shows the mean and maximum summer and winter isoprene mixing ratios for all stations. Although the number of data points at stations other than Rocky Point Park was limited, some trends can be discerned. Mean summer isoprene mixing ratios were similar at most sites, with slightly higher averages recorded at Langley, the most easterly and least urban site. Mean winter isoprene mixing ratios were similar to or slightly higher than summer values for the urban sites, but lower than summer values at suburban sites. Maximum summer isoprene concentrations were less than 0.5 ppbv at all stations, except for Langley where the maximum isoprene measured was 0.64 ppbv. Maximum winter isoprene concentrations were also less than 0.5 ppbv for all stations, except for Rocky Point Park and Robson Street. The reading of 0.77 ppbv at Robson Street on January 28, 1989, was the hig,hcst recorded in the region and supports the hypothesis that anthropogenic sources and a shallow mixing depth can occasionally result in very high ambient isoprcne concentrations. The concentrations of isoprene observed at Rocky Point Park and other sites during the winter are probably also enhanced by the shallow mixing layer, but these results indicate active local sources of isoprene during the winter months, particularly in urban areas. The source is unlikely to be a biogenic one, as most plants in the region are dormant during this time. Although some plant species in the Vancouver area remain green throughout the mild winter, temperatures are too low for substantial biogenic emissions. Figures 2 and 3 show the relationship between mean daily isoprene mixing ratios and temperatures during the summer and winter months at Rocky Point Park. Figure 2 shows a clear relationship between isoprene and temperatures greater than 15ø(2 in the summer, while Figure 3 shows a complete lack of correlation between the two in the winter. Biogenic emissions increase with increasing temperature; thus these results uggest that the isoprene measured at Rocky Point Park in the winter months does not originate from a biogenic source. Similar relationships between ambient isoprene concentrations and temperature during summer and winter were observed at the other sites (data not shown). The relationship between isoprene and temperature in the summer falls into two groupings. Isoprene mixing ratios at daily mean 0.7 0.6 0.5 0.3 0.2 ' 0.1 10 15 20 25 30 Mean Daily Temperature (C) Figure 2. The relationship between isoprene and temperature at Rocky Point Park during summer. 25,472 CURREN ET AL.: BIOGENIC ISOPRENE IN THE LOWER FRASER VALLEY 0.6 0.5 0.4 0.3 0.2 0.1 00 ß ß ß 0 , , 4• , , - 0 -5 0 5 10 15 Mean Daily Temperature (C) Figure :5. The Relationship Between Isoprene and Temperature atRocky Point Park during -winter. temperatures less than approximately 15 øC arc variable (likely as the result of varying meteorological onditions on specific days), but show no correlation-with temperature. Biogenic emissions at daily mean temperatures below 15øC (corresponding to a daily maximum temperature of about 20øC) should be very small. The isoprenc measured on these days probably originates from the same anthropogenic source as the isoprenc measured uring the -winter months, while the isoprenc measured on-warmer days is a mixture of isoprenc from both biogcnic and anthropogenic sources. Table 3 shows the average summer isoprenc mixing ratios on days with mcan temperatures less than 15 øC for all Lower Fraser Valley sites. Although data are limited at sites other than Rocky Point Park, these values give rough indications of the background anthropogenic level of isoprenc oncentrations during the summer at each of these sites. Isoprene concentrations at Rocky Point Park on warm summer days (T.• >20øC, corresponding to T.• >27øC) ranged from 0.14-0.44 ppbv,-while the background level of isoprene on cool days (T•,. <]5øC) was 0.l ] ppbv. Background anthropogenic isoprene may therefore account for 25-78% of the total observed isoprene load at Rocky Point Park. Although Guenther et al. [1991] calculated an exponential equation for biogenic isoprene mission rates with increasing Table 3. Mean Summer Isoprene Mixing Ratios for Days with Mean Temperatures <15 ø C Site Mean Isoprene Mixins Ratio ppbv n Rocky Point Park 0.11 54 Kensington Park 0.08 11 Mahon Park 0.14 7 Richmond South 0.11 5 Robson Street O. 10 10 Shellmount Street 0.08 22 Surrey East 0.04 6 Lan•ey 0.10 2 temperature, the relationship between ambient isoprene.concentrations and temperature is not necessarily exponential due to the effects of changing meteorological conditions and the changing strengths of isoprene sinks. There-were sufficient summer data for Rocky Point Park to perform aregression between daily mean isoprene mixing ratio and temperature. The increase in isoprene concentration with temperature for daily mean temperatures greater than 15øC -was approximately inear. Thus the daily mean isoprene mixing ratio at Rocky Point Park can be characterized by equation (5): Isoprene (ppbv) = 0.11, Tm•,• < 15øC, (5) = 0.023 x Tm• - 0.24, Tm• > 15 ø (2, r • = 0.50 Correlation coefficients for regressions between isoprene, trans-2-pentene, and cis-2-butene for all sites are shown in Table 4. Isoprene-was positively correlated with both cis-2-butene and trans-2-pentene at all sites in both summer and -winter. Positive correlations-were also observed between the two anthropogenic compounds. These results are in contrast to those of Chameides et al. [1992] for southern U.S. cities and indicate that an anthropogenic source of isoprene exists in the Lower Fraser Valley during both summer and -winter. Seasonal differences in correlations between isoprene and the anthropogenic hydrocarbons-were variable. At Rocky Point Park, the station with the most data available, the correlations -were higher in the winter than in the summer. This result can be interpreted to indicate that anthropogenic isoprene accounted for a higher fraction of the total isoprene load in the -winter than in summer. This trend-was not consistently observed at other sites, but this may be the result of limited data at these sites. 4.2. Diurnal Analysis 4.2.1. Ambient diurnal isoprene measurements. Average diurnal isoprene profiles for Rocky Point Park and Langley are shown in Figure 4. Data for nonepisode days-with similar maximum CURREN ET AL.: BIOGENIC ISOPRENE IN THE LOWER FRASER VALLEY 25,473 Table 4. Correlation Coefficients for Isoprene, cis-2-Butene, and trans-2-Pentene Station Compound Summer Winter r n r n Rocky Point Park (Urban) Kensington (Urban) Richmond South (Urban) Robson Street (Urban) Surrey East (Suburban) Langley (Suburban) cis-2-butene/isoprene 0.64 192 0.73 157 trans-2-pentene/Iioprene 0.68 192 0.81 157 cis-2-butene/trans-2- 0.82 192 0.75 157 pentene cis-2-butene/isoprene O. 76 29 O. 63 22 trans-2-pentene/isoprene 0.62 29 0.8 22 cis-2-butene/trans-2- 0.84 29 0.82 22 pentene cis-2-butene/isoprene 0.82 24 0.72 16 trans-2-pentene/isoprene 0.75 24 0.58 16 cis-2-butene/trans-2- 0.73 24 0.72 16 pentene cis-2-butene/isoprene 0.79 24 0.39 17 trans-2-pentene/is oprene O. 86 24 O. 59 17 cis-2-butene/trans-2- O. 89 24 0.24 17 pentene cis-2-butene/isoprene O. 17 21 O. 49 23 trans-2-pentene/isoprene 0.57 21 0.82 23 cis-2-butene/trans-2- 0.37 21 0.37 23 pentene cis-2-butene/isoprene 0.73 12 0.62 10 trans-2-pentene/isoprene 0.81 12 0.68 10 cis-2-butene/trans-2- 0.90 12 0.69 10 pentene temperatures at each site were grouped, and the average profiles were graphed. The times indicated on the graphs mark the first hour of the 3 hour sample. In general, both Rocky Point Park and Langley showed flat profiles with low concentrations of isoprene on nonepisode days with maximum temperatures less than 25øC (Figure 4a). Isoprene profiles at other sites (data not shown) were similar to those depicted in Figure 4a. Daytime concentrations of isoprene on nonepisode days with Tm• < 25øC were less than 0.3 ppbv at both sites. Profiles at Langley on warmer days (Tm• > 25øC) displayed slightly higher concentrations throughout the day (0.2-0.5 ppbv). Increases in isoprene.concentrations wereobserved late in the aftemoon on some days, likely as a result of the collapsing boundary layer. Rocky Point Park tended to have higher isoprene concentrations than Langley on cool days (Tm• < 20øC), probably due to higher anthropogenic emissions in the urban area. Conversely, Langley tended to have higher isoprene concentrations on warm days (Tm•> 20øC) as a result of its suburban location and higher biogenic emissions. On mild ozone episode days (August 2-5, 1993; Tm,x > 30øC), both Rocky Point Park and Langley showed rather different isoprene profiles (Figure 4b). Isoprene concentrations atLangley on these days reached magnitudes of 0.8 ppbv or more, while levels at Rocky Point Park peaked at 0.6 ppbv. The 3 episode days at Langley and Rocky Point Park showed increases in isoprene concentrations after dawn and elevated concentrations throughout he day. All three days exhibited small moming isoprene peaks, possibly as a result of anthropogenic moming rdsh-hour emissions. At Rocky Point Park, this peak occurred at 0600 hours (representing the average mixing ratio from 06004)900 hours). At Langley, the morning peak occurred at 0900 hours on both days. The later time of the morning peak at Langley may reflect the amount of time required for the urban emissions of the Vancouver core to reach this suburban area. It should be noted that the isoprene profiles observed on the 1993 episode days may not be typical of diurnal profiles during major ozone events. The ozone episode of 1993 was mild, and OH concentrations were likely suppressed due to the high levels of NO x. 4.2.2. Box mode[ To determine if the box model could account for typical isoprene concentrations and profiles observed inrelatively simple systems, the model was run using data representative of a rural forest. For this simulation, a higher base mission rate was required. Lamb et al. [1985] estimated that the isoprene flux over a northeastern U.S. deciduous forest was 8000 •tg m '2 h 4 at 30 ø C. This value for S was entered into the model, along with the temperature profile at Rocky Point Park on August 5, 1993 (Tm• = 29øC) and a maximum mixing layer depth of 800 m. Peak values of OH and NO 3 were lx106 radicals cm '3 and 10 ppt, respectively. The predicted profile is shown in Figure 5. The model produces an isoprene profile similar to those measured over forests and published in the literature [e.g., Lamb et al., 1985; Fehsenfeld et al., 1992]. The predicted isoprene concentrations ri e sharply after dawn, continue to rise throughout the day, and peak at a value greater than 10 ppbv in the early evening. The peak in the early evening is the result of decreasing concentrations of OH radicals and the collapse of the mixing layer. Figure 6a shows the predicted isoprene profiles on a representative nonepisode ay at Rocky Point Park. For the nonepisode ays, the box model successfully reproduced the relatively flat profile and low concentrations at both sites. In general, the predicted isoprene concentrations showed avery slight peak in the morning due to the combined effects of the shallow mixing layer and low concentrations of OH radicals, followed by a daytime minimum at noon that corresponded with the OH peak. Concentrations of isoprene rose slightly again in the late aftemoon in response to decreasing OH concentrations and the collapse of the mixing layer. Rising levels of NO3 radicals were responsible for the decrease in isoprene after sunset. Figure 6b shows the predicted isoprene profile at Langley on the episode day August 2, 1993. As discussed previously, the measured isoprene profiles on these days were variable, and the box model was less successful at reproducing the observed pattems, although, in general, it made reasonable stimations of the magnitude of the isoprene mixing ratios. The most notable difference between days was the behavior of the isoprene profile in the late aftemoon and evening. On August 2 (Langley), observed isoprene concentrations began to rise in the late afiemoon and continued to rise throughout the evening. On August 4 (Langley), concentrations rose in the late aftemoon, but fell sharply in the evening. Finally, on August 4 (Rocky Point Park), isoprene concentrations rose until midafternoon and then decreased throughout the rest of the day and evening. The box model, on the other hand, predicted similar profiles on all 3 days, with predicted isoprene concentrations rising in the moming and continuing to increase throughout the day. The high predicted concentrations from noon to late in the aftemoon were the result of active isoprene emissions increasing with temperature, combined with decreasing concentrations f OH radicals. Predicted levels of isoprene fell sharply in the evening in response tothe modeled increasing concentrations of NO3. Them are two factors that may explain the observed afternoon and evening variations in isoprene profiles on episode days and why the model had difficulty reproducing these profiles. First, the time of the collapse of the boundary layer is important in determining whether isoprene concentrations show a substantial increase (other than the increase due to decreasing OH radical concentrations) late in the afternoon. If the boundary layer collapses before sunset while light levels are still sufficient for isoprene mission, ambient concentrations will increase as a result of the reduced dilution. The modeled profiles 25,474 CURREN ET AL.: BIOGENIC I$OPRENE IN THE LOWER FRASER VALLEY 1.6 1.4 1.2 1.o 0.8 0.6 0.4 0.2 0.0 Langley .... Rocky Point >25C (n:9) •J 20-25C (n=14) d,....•• __..•'•'1 . .I ...... ß .. o .. . . . •.•... •. ø' (n:7) _ ..... ' ..... -I- -I ...... '' '•.2.2•] <20c l' -1 • (n=3) 6 12 18 Time (hr) Figure 4a. Isoprene profiles on averaged non episode days. 20-25C (n=8) 24 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 i 12 18 Time (hr) Langley, Aug 2 -' Langley, Aug 4 + Rocky Point, Aug 5 ] Figure 4b. Isoprene profiles on episode days. 24 CURREN ET AL.: BIOGENIC ISOPRENE 1N THE LOWER FRASER VALLEY 25,475 14 12 10 I I 6 12 18 Time (h) Figure 5. Predicted isoprene profile over a rural forest. 24 assumed that the boundary layer collapsed at the same time that the biogenic sources topped emitting, thus negating this effect. Second, the concentration and profile of nighttime NO3 concentrations i  critical to explaining the behavior of isoprene after sunset. Measurements taken at experimental sites in West Germany showed that NO3 concentrations were very variable and that profiles changed substantially from night to night. NOa concentrations at these sites ranged from not detectable to 280 ppt [Platt et al., 1981]. No NOa data from the Lower Fraser Valley were available for this analysis. NOa profiles were assumed to rise quickly at sunset and to maintain a peak concentration of 10 ppt for several hours. A slower increase in NOa concentrations after sunset would result in higher evening isoprene concentrations. 5. Conclusions The seasonal impme study described in this paper examined mean daily impme concentrations at several monitoring stations within the Lower Fraser Valley. As predicted by previous tudies which found 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 I I 0 6 12 18 24 Time (h) [-*- Measured + Predicted I Figure 6a. Predicted and measured isoprene profile at Rocky Point Park; averaged ays with 20øC<Tm•<25 øC. 25,476 ½URREN ET AL.: BIOGENIC ISOPRENE IN THE LOWER FRASER VALLEY 1.4 1.2 1.0 0.6 0.4 0.2 0.0 I 0 6 12 18 24 Time (h) ['•- Measured + Predicted l Figure 6b. Predicted and measured isoprene profile at Langley, August 2, 1993. few species of isoprene-cmitting vegetation i  the region [Drewitt, 1996; Drewitt et al., 1997; Curren, 1998), daily mean isoprene mixing milos at all of the urban and suburban sites examined were low throughout the year. Mean summer isoprene concentrations were approximately equal in magnitude to mean winter concentrations at ud•m sites and were slightly higher than mean winter concentrations at suburban sites. Although winter isoprene concentrations were enhanced by the shallow mixing layer of this season, the presence of isoprene in the atmosphere during the season when most plants are dormant indicates the existence ofactive local nonbiogenie sources of isoprene atthe sites tudied. Winter concentrations did not show any relationship with temperature, supporting the hypothesis of a nonbiogenie winter isoprene source. Detectable l vels of isoprene were present at all sites on cool summer days when biogenie sources were relatively inactive, indicating a background anthropogenic level of isoprene during this season as well. Summer isoprene concentrations at temperatures greater than 15 oC showed a dear relationship with temperature. These results suggest that winter isoprene mixing ratios are the result of anthropogenic emissions of isoprene, while summer concentrations are a mixture of both anthropogenic and biogenie isoprene. The measured levels ofisoprene showed strong positive correlations with cis-2-butene and trans-2-.pentene, hydrocarbons ormally associated with anthropogenic mobile sources, af all sites in both seasons. This is in contrast to the results reported for southern U.S. cities and indicates that a substantial portion of both the summertime and wintertime ambient isoprene load originates from anthropogenic mobile sources. Concentrations of isoprene on cool summer days (when biogenie sources are likely to be inactive) suggest that background anthropogenic isoprene may account for 25-78% of the total observed isoprene load at Rocky Point Park. The diurnalisoprene study examined the daily variation i  isoprene concentrations in the Lower Fraser Valley and investigated the physical and chemical factors that defined the observed profiles through t e use of a simple box model. On average, isoprene l vels on nonepisode days at urban sites showed flat profiles and low concentrations ofisoprene throughout the day. Higher isoprene l vels were measured uring the mild ozone episode days of 1993 and showed profiles that increased after sunrise and remained elevated throughout theday. There was some vidence ofa morning isoprene peak due to rush hour elated anthropogenic isoprene emissions on these days. A simple box model was written to investigate the physical nd chemical factors that determine the shape and the magnitude of isoprene profiles in the Lower Fraser Valley and to explain why the observed urban profiles were so much flatter in shape than those measured at rural sites over forests. The model included temperature- and light-dependent biogenie isoprene emissions, estimated anthropogenic isoprene emissions, a time-dependent mixing depth, daytime OH sinks, and nighttime NO3 sinks. The model successfully predicted both isoprene concentrations a dprofile shapes aturban sites on nonepisode days. and over forests, and was partially successful in estimating isoprene tr nds on chemically complex pisode ays. The results ofthe model indicate that isoprene concentrations ver deciduous forests are consistently high throughout the day as a result of the very high local areal emission rates that successfully compete with the effects of dilution due to a growing boundary layer and destruction by OH radicals. Isoprene mission rates at urban and suburban sites in the Lower Fraser Valley, on the other hand, are an order of magnitude lower than those at rural forest sites. At urban sites, daytime isoprene emissions fight a losing baffle with the effects of a deepening mixing layer and destruction byOH radicals; low isoprene concentrations and flat diurnal profiles are the result. The flat shape of the daytime [soproe profiles and low daytime concentrations lessen the potential impact of biogenie isoprene upon regional ozone production i  the Lower Fraser Valley as the times of its highest concentrations (during the early evening) do not coincide with peak NO x concentrations nor with the times of optimal ozone-producing meteorological conditions. The weight of evidence from this and previously published research suggests hat biogenic isoprene is not a major contributor to the production of ozone in the Lower Fraser Valley. Few species ofcrops, CURREN ET AL.: BIOGENIC ISOPRENE IN THE LOWER FRASER VALLEY 25,477 natural vegetation, and trees in the region emit isoprene in appreciable amounts. As a result, ambient concentrations of total isoprene are correspondingly low throughout the day and in all seasons. Measurable concentrations ofisoprene in winter and on cool summer days, as well as the positive correlation of isoprene with hydrocarbons known to be emitted from anthropogenic mobile sources, indicate that isoprene inthe Lower Fraser Valley originates from both biogenic and anthropogenic sources; thus biogenic isoprene constitutes only a fraction of the small total isoprene load. An analysis of isoprene's contribution to the total reactivity of nonmethane hydrocarbons in the Lower Fraser Valley showed that approximately 10% or less of the ozone produced at Langley results from reactions with isoprene [Bottenheirn et al., 1997]. Isoprene's contribution tothe formation of ozone was expected to be even less in more urban areas as a result of its lower abundance at those sites. This value is in good agreement with the analysis of Biesenthal et al. [ 1997], who used a separate and independent method to determine that approximately 13% of the ozone formed at another site in the Lower Fraser Valley was the product of reactions with isoprene. Together with the conclusions presented above, research suggests hat biogenic isoprene may not be as important to the tropospheric ozone chemistry in the Lower Fraser Valley as it is in southern U.S. cities. Acknowledgements. Funding for this research was provided by •e Natural Sciences and Engineering Research Council and Environment Canada. Thanks to Weimin Jiang• National Research Council Canada, for providing temporal anthropogenic VOC profiles, and to Douw Steyn, University of British Columbia, and Nigel Bunce, University of Guelph, for advice and comments on this manuscript. References Biesenthal, T.A.• Q. Wu, P.B. Shepson, H.A. Wiebe, K.G. Anlauf, and G.I. MacKay, A study of relationships between isoprene, its oxidation products and ozone in the Lower Fraser Valley, B.C., Atmos. Environ. 31(14), 2049- 2058, 1997. Bottenheim, J.W., P.C. Brickell, T.F. Dann, D.K. Wang, F. Hopper, A•J. Gallant, K.G. Anlauf, and H.A. Wiebe, Non-methane hydrocarbons and CO during Pacific '93, Atmos. Environ. 31 (14), 2079-2087, 1997. Canadian Council of Ministers of the Environment (CCME), Canadian 1996 No•OC science assessment: Ground-level ozone and its precursors, report of the data anal. working group, Ottawa, Ont., Canada, 1997. Cardelino, C./c, and W.L. Chaincities, An observation-based model for analyzing ozone precursor relationships inthe urban atmosphere, J. Air Waste Manage. Assoc. 45, 161-180, 1995. Chameides, W.L. et al., Ozone precursor elationships in the ambient atmosphere, J. Geophys. Res., 97(D5)6037-6056, 1992. Cu• K., Biogenic isoprene and the production of tropospheric ozone in the Lower Fraser Valley, BritiSh Columbia, Ph.D. thesis, Land Resour. Sci., Univ. of Guelph, Guelph, *Ontario, 1998. Dann, T., D. Wang A. Steenkamer, R. Halman, and M. Lister, Volatile organic compound measurements in the Greater Vancouver Regional District (GVRD), 1989-1992, Rep. PMD 94-1, Environ. Technol. Cent., Environ. Can., Ottawa, Ont., Jan. 1994. Drewitt, G., Measurement ofbiogenic hydrocarbon emissions from vegetation in the Lower Fraser Valley, British Columbia, M. Sc. thesis, 121 pp., Vancouver., B.C., Canada, Dep. of Geogr., Lhaiv. of B.C., 1996. Drewitt, G.B., K. Curreft,. D.G. Steyn, T.J. Gillespie, and H. Niki, Measurement ofbiogenic hydrocarbon emissions from vegetation in the Lower Fraser Valley, B.C.,Atmos. Environ., in press, 1998. Fehsenfeld, F. et al., Emissions ofvolatile organic ompounds from vegetation and the implications for atmospheric chemistry, Global Biochem. Cycles, 6(4), 389-430, 1992. Finlayson-Pitts, B.J., and J.N. Pitts, Atmospheric Chemistry: Fundamentals and Experimental Techniques, Wiley-Interscience New York, 1986. Guenther, A.B., R.K. Monson, and R. Fall, Isoprene and monoterpene emission rate variability: Observations with eucalyptus and emission rate algorithm development, d. Oeophys. Res., 96(D6), 10,799-10,808, 1991. Hayden, K.L., K.G. Anlauf, R.M. Hoff, J.W. Strapp, J.W. Bottenheing H.A. Wiebe, F.A. Frouric, and J.B Martin, The vertical chemical and meteorological structure of the boundary lay .er in the Lower Fraser Valley during Pacific •)3 ,Atmos. Environ., 31(14), 2089-210;5, 1997. Hsu, S.A., Anote on estimating the height of the convective internal boundary layer near shore, Boundary LayerJl/leteorol., 35, 311-316, 1996. Jiang, W., D.L. Singleton, A. Dorkalam, S. Bohme, and M. Hedley, Processing the Lower Fraser Valley Pacific 93 emission inventory for UAM-V applications: Area and mobile sources, Rep. PET- ]3•5-965, Nat. Res. Counc.,Ottawa, Ont., Canada, Dec. 12, 1996. Lamb, B., H. Westberg, and G. Allwine, Biogenic hydrocarbon emissions from deciduous and coniferous trees in the United States,J. Geophys. Res., 90(D1), 2380-2390, 1985. LeveRon, B.H., Pacific 93 air emissions inventory draR report, B.H. Levelton & Assoc. Ltd., Richmond, B.C., Canada,'July 1996. McLaren, P•, D.L. Singleton, J.Y.K. Lai, B. Khouw, E. Singer, Z. Wu, and H. Niki, Analysis of motor vehicle sources and their contribution to ambient hydrocarbon distributions at urban sites in Toronto during the southern Ontario oxidants study, Atmos. Environ., 30(12), 2219-2232, 1996. National Research Council (NRC), Committee on Tropospheric Ozone Formation and Measurement, Rethinking the Ozone Problem in Urban and Regional Air Pollution, N at, Acad, Press, Washington, D.C., 1991. Oke, T.R., Boundary Layer Climates, Methuen, New York, 1978. Oliver, K.D., J.D. Pleil, and W.A• McClenny, Sample integrity of trace level volatile organic compounds in ambient air stored in summa- polished canisters, Atmos. Environ., 20, 1403, 1986. Parton, W.J., and J.A. Logan, A mode! for diurnal variation in soil and air temperature, Agric. Meteorol., 23, 205-216, 1981. Pierce, T.E., and P.S. Waldruff, PC-BEIS: A personal computer version ofthe Biogenic Emissions Inventory System/. Air Waste Manage. Assoc., 41,937-941, 1991. Platt, U., D. Perner, J. Schr6der, C. Kessler, and A• Toennissen, The diurnal variation of NO3, d. Geophys. Res., 86(C12), 11,965-11,970, 1981. Singleton, D., M. Hedley, W. Jiang, R. McLaren, T. Dann, and P.B. Shepson, Evaluation of isoprene emission and chemistry for photochemical and ozone modelling in the Lower Fraser Valley, British Columbia, paper presented at the 89th Annual Meeting and Exhibition, Air and Waste Manage. Assoc., Nashville, Tenn., June 23-28, 1996. Steyn, D.G., and T.R. Oke, The depth of the daytime mixed layer at two coastal sites: A model and its validation, Boundary Layer Jl/leteorol., 24,161-180, 1982. Steyn, D.G., J.W. Bottenheim, and R.B. Thomson, Overview of tropospheric ozone in the Lower Fraser Valley and the Pacific '93 study, Almos. Environ., 3](14), 2025-2035, 1997. Trainer, M., E.J. Williams, D.D. Parrish, M.P. Buhr, E.J. Allwine, H.H. Westberg. F C. Fehsenfeld, and S.C Liu, Models and observations ofthe irrq•ct ofnatural hydrocarbons on rural ozone, Nature, 329, 705-707, 1987. Venkatram, A., A model for internal boundary layer development, Boundary Layer Jl/leteorol., ] 1,, 419-437, 1977. Winberry, W.A, Jr., N.T. Murphy, and R.M. Priggan, Method TO-14 and Method TO-13, in Compendium of Methods for the Determination f Toxic Organic Compounds in Ambient Air, Rep. EPA-600/4-89-O! 7, U.S. Environ. Prot. Agency, Research Triangle Park, N. C., June 19gg. IC Cu•en and T. Gillespie, Land Resource Science, University of Guelph, Guelph, ontario, Canada N1G 2W1. (e-mail: tgillesp•lrs.uoguelph. ca). T. Dann and D. Wang ETC, Environment Canada, River Road, Ottawa, Ontario, Canada K1A 0H3. (Received November 25, 1997; revised March 31, 1998; accepted April 7, 1998.)


Citation Scheme:


Usage Statistics

Country Views Downloads
China 8 25
Japan 3 0
City Views Downloads
Beijing 8 0
Tokyo 3 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}


Share to:


Related Items