Atmospheric Environment35(2001)3217–3228
北京地铁路线图高清版Trends in exceedances of the ozone air quality standard in the continental United States,1980–1998 C.-Y.Cynthia Lin1,Daniel J.Jacob*,Arlene M.Fiore
Di v ision of En g ineerin g and Applied Sciences and Department of Earth and Planetary Sciences,Har v ard Uni v ersity,
Pierce Hall,29Oxford St.,Cambrid g e,MA02138,USA
Received24September2000;accepted18February2001
Abstract
In1997,the United States National Ambient Air Quality Standard(NAAQS)for ozone was revised from a1-h average of0.12parts per million(ppm)to an8-h average of0.08ppm.Analysis of ozone data for the ensemble of the contiguous United States and for the period1980–1998shows that the average number of summer days per year in exceedance of the new standard is in the range8–24in the Northeast and in Texas,and12–73in southern California. The probability of exceedance increases with temperature and exceeds20%in the Northeast for daily maximum temperatures above305K.We present the results of several different approaches to analyzing the long-term trends in the old and new standards over the continental United States from1980to1998.Daily temperature data are used t
o resolve meteorological variability and isolate the effects of changes in anthropogenic emissions.Significant negative trends are found in the Northeast urban corridor,in the Los Angeles Basin and on the western bank of Lake Michigan. Temperature segregation enhances the detection of negative trends.Positive trends occur at isolated sites,mostly in the Southeast;a strong positive trend is found in Nashville(Tennessee).There is some evidence that,except in the Southwest,air quality improvements from the1980s to the1990s have leveled offin the past decade.#2001Published by Elsevier Science Ltd.
Keywords:Air quality;Air quality standard;Ozone;Ozone trends;Ground-level ozone;Surface ozone
1.Introduction
High levels of surface ozone are responsible for most
violations of the National Ambient Air Quality Stan-
dards(NAAQS)in the United States(US Environ-
mental Protection Agency(EPA),2000).Set in1979,the
original NAAQS for ozone was a1-h average of0.12
parts per million(ppm),not to be exceeded more than
three times in3years.In July1997,based on its review上海有几个
机场 of scientific evidence linking ozone exposure to adverse
health and welfare effects at levels allowed by the1-h
standard,the EPA revised the standard to an8-h
average of0.08ppm,with a form based on the3-year
average of the annual fourth-highest daily maximum8-h
average ozone concentrations measured at each monitor
within an area(EPA,1998).Because ozone concentra-
tions are generally measured to the nearest part per
billion(ppb),the smallest concentrations in exceedance
of the old and new standards are125and85ppb,
respectively.As of1998,51million persons in the
United States lived in areas violating the old standard,
while130million persons lived in areas violating the new
standard(EPA,2000).In a study of rural sites in the
eastern United States,Saylor et al.(1998)found that2–
12%of the sites violated the old standard and30–50%
of the sites violated the new standard for ozone during
1993–1995.A May1999federal appeals court ruling
has 1Present address:Department of Economics,Harvard
University,Cambridge,MA02138,USA.
*Corresponding author.Tel.:+1-617-495-1794;fax:+1-
617-495-4551.
E-mail addresses:cclin@fas.harvard.edu(C.-Y.C.Lin),
djj@io.harvard.edu(D.J.Jacob).国力仁和
1352-2310/01/$-see front matter#2001Published by Elsevier Science Ltd.
长沙3日游旅游攻略PII:S1352-2310(01)00152-2 questioned the constitutionality of the process by which the new standard was enacted;enforcement of this standard must wait until the issue is settled in the courts (American Trucking Associations vs.EPA,1999).
We examine in this paper the long-term trends in exceedances of the new and old ozone standards in the United States over the past two decades.Considerable effort has been expended during this p
eriod to control anthropogenic emissions of ozone ,nitrogen oxides (NO x =NO+NO 2)and hydrocarbons.These efforts have been hindered by population growth (31%from 1970to 1997)and increased vehicle miles traveled (127%over the same period)(EPA,1998).It is estimated that anthropogenic emissions of hydrocarbons in the United States decreased by 30%from 1980to 1998,while NO x emissions remained constant nationally to within a few percent (EPA,1998,2000).
Hydrocarbon
Fig.1.Average values for 1980–1998of (a)the annual fourth-highest daily maximum 8-h average ozone concentration (ppb)and (b)the annual second-highest daily maximum 1-h average ozone (ppb).The mean over all years is first calculated for each of the 522individual sites meeting the data density requirement described in the text.The data are averaged for display over 0.58Â0.58(roughly 50km Â50km)grid squares.
C.-Y.C.Lin et al./Atmospheric Environment 35(2001)3217–3228
3218
emission controls have been credited for abating the most extreme ozone pollution events,but a more general linkage between trends in precursor emissions and trends in ozone has yet to be demonstrated(National Research Council(NRC),1991;Sillman,1999;EPA, 2000).
Long-term trends or interannual variability of me-teorological variables could contribute to or obfuscate ozone trends.Meteorological conditions influence ozone by affecting the biogenic emissions of hydrocarbons,the atmospheric circulation,and the photochemical envir-onment(Rao et
al.,2001).In particular,temperature is found to be strongly correlated with ozone concentra-tions on a day-to-day basis(Wolffand Lioy,1978;Clark and Karl,1982;NRC,1991).Higher temperatures enhance biogenic hydrocarbon emissions and increase the abundance of NO x by thermal decomposition of the peroxyacetyl nitrate(PAN)reservoir(Olszyna et al., 1997).Temperature is also a surrogate for clear skies and stagnation(Jacob et al.,1993;Olszyna et al.,1997; Smith and Adamski,1998).As a result,it is found that the correlation with temperature can account for most of the influence of meteorological variables on ozone (Jacob et al.,1993).
Many previous studies have examined long-term trends of ozone concentrations in selected regions of the United States(Rao et al.,2001,and references therein).In an examination of nationwide trends in medians and90th percentiles of summer afternoon ozone for1980–1995,Fiore et al.(1998)found that trends were insignificant over most of the continental United States;decreasing trends were clustered in the New York City,Los Angeles and Chicago metropolitan areas.We extend here on the above studies by presenting a trend analysis targeted at exceedances of both the1-and8-h ozone standards for the entire contiguous United States and for nearly two decades(1980–1998). We also attempt to isolate the contributions of anthro-pogenic emissions to the trends by using temperature as
a proxy for meteorological influences.
2.Data and methods
Hourly ozone concentration data from1980to1998 were extracted from the EPA’s Aerometric Information Retrieval System(AIRS).Ozone measurements at the AIRS sites are subject to frequent calibration and consistency checks.We chose to begin the analysis in 1980,a year after the EPA used an ultraviolet photometric method to uniformly calibrate all ozone measurements(Chock,1989).The AIRS database includes over2000sites across the United States that have monitored ozone at one time or another over the 19years of our data set,but many of these sites have sparse records.For cases in which data for sequential years were available from sites a few kilometers apart, usually because the monitoring station had moved,the ozone data were merged.
In accordance with the protocol for the new standard (EPA,1997),running8-h averages,indexed by thefirst hour,were calculated for each8-h interval with at least 6h of data,and the daily maximum8-h average was stored for each day for each site.Similarly,daily maximum1-h average ozone concentrations were also stored for each day for each site.
Fig.1shows the average values over the19years of the data set of the annual fourth-highest daily maximum 8-h average ozone and the annual second-highest daily maximum1-h average ozone for sites with at least 15years fulfilling the minimum data density requirement of4and2days per year,respectively.For both the8-h and1-h analyses,there were522sites that fulfilled the data density requirement.The19-year mean was calculated for each individual site and then averaged for display over0.58Â0.58(roughly50kmÂ50km)grid squares.For both statistics,the highest values are in southern California,along the Northeast coast,in Atlanta,and in Houston.More sites are in violation of the new8-h standard,as denoted by the green,
yellow, Fig.2.Average fraction of days per month whose daily maximum8-h average exceeds85ppb,out of all days per month with ozone data,for the analysis period1980–1998and for four quadrants of the United States partitioned at368N, 97.58W(northern Oklahoma).
C.-Y.C.Lin et al./Atmospheric Environment35(2001)3217–32283219
orange,and red squares,than are in violation of the old 1-h standard,as denoted by the dark yellow,orange and red squares.
In addition to rank order,an alternate type of statistic that can be used to analyze exceedances of the ozone air quality standard is the frequency of occurrence of daily maximum 1-or 8-h average concentrations in excess of the standards.Seasonal selection of the data is appropriate here to avoid any bias caused by site-to-site differences in the yearly extent of data coverage;most sites do not operate year-round.Fig.2shows the average fraction of days in exceedance of the 8-h standard by month for the four quadrants of the United States,as partitioned at 368N,97.58W (in northern Oklahoma).For each month and for each year,the fraction of exceedance days out of
all
Fig.3.Average number of summer days per year (a)whose daily maximum 8-h average exceeds 85ppb and (b)whose daily maximum 1-h average exceeds 125ppb.The analysis period is 1980–1998.The mean over all years is first calculated for each individual site meeting the data density requirement described in the text.The data are averaged for display over 0.58Â0.58grid squares.A total of 462sites were used for (a);463sites were used for (b).
C.-Y.C.Lin et al./Atmospheric Environment 35(2001)3217–3228
3220
days with data was calculated for each site;the result-ing values for each month were then averaged across all years and for all sites in each quadrant.Wefind that the summer months(June–August)account for81%of all the exceedances in the Northeast,60%in the Southeast,70%in the Northwest,and58%in the Southwest.
We therefore focus our trend analysis of the fraction of summer days in exceedance of the new and old standards,and require that sites have at least16days of data per month for all three summer mon
ths and for at least15of the19summers.For ease of comprehension,the fractions are expressed as the number of summer days in exceedance per year via multiplication by92.Figs.3a and b show the average values of the number of summer days in exceedance of the8-and1-h standards over1980–1998,respecti-vely.The greatest frequency of exceedances is
in Fig.4.Linear trend in ppb/year over1980–1998of(a)the annual fourth-highest daily maximum8-h average ozone,and(b)the annual second-highest daily maximum1-h average ozone.Trends are calculated for individual sites and then averaged for display over 0.58Â0.58grid squares.Sites with statistically insignificant trends enter this averaging as a trend of zero.Grid squares whose average trend is zero and therefore insignificant are shown in gray.
C.-Y.C.Lin et al./Atmospheric Environment35(2001)3217–32283221
southern California,where exceedances occur 12–73times per summer for both standards.Many exceedances also occur in the Northeast and eastern Texas (8–24and 2–8occurrences per summer for the 8-and 1-h standards,respectively).In general,sites experience more exceedances of the 8-hour standard than of the 1-h standard.
3.Results
3.1.Trends in exceedances of the ozone standard Fig.4shows the 1980–1998trends in the annual fourth-highest daily maximum 8-h average ozone concentration and the annual second-highest中国晚上10点
俄罗斯是几点
婺源十大
旅游景点daily
Fig.5.Linear trend in days/year over 1980–1998in the number of days per summer (a)whose daily maximum 8-h average exceeds 85ppb and (b)whose daily maximum 1-h average exceeds 125ppb.Trends are calculated for individual sites and then averaged for display over 0.58Â0.58grid squares.Sites with statistically insignificant trends enter the calculation as a trend of zero.Grid squares whose average trend is zero and therefore insignificant are shown in gray.Circles indicate sites with no exceedance days for any of the summers over the 1980–1998period.
C.-Y.C.Lin et al./Atmospheric Environment 35(2001)3217–3228
3222
