Final Report, Volume II: Summary and Integration of Results

OTAG Air Quality Analysis Workgroup

Dave Guinnup and Bob Collom, Co-chair

 

May 26, 1997

Table of Contents

INTRODUCTION *

Background *

Air Quality Analysis Work Group Activities *

Overview of Regional Ozone and Transport in the Eastern United States *

INTEGRATED SUMMARY OF WORKGROUP ANALYSES *

Spatial and Temporal Pattern of Ozone in the OTAG Domain *

Transport Analyses *

Development of Control Strategies *

Impact of an 8-Hour Ozone Standard *

Representativeness of OTAG Episodes *

OTAG Model Evaluation *

Recommendations for Additional Analyses *

REFERENCES *

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Cover Image: Counties not meeting the 0.12 and 0.08 ppm standard according to EPA

INTRODUCTION

Background

During the summer months, high ground-level ozone concentrations are observed within and downwind of many of the large urban areas in the eastern US. Peak hourly average concentrations exceed the current National Ambient Air Quality Standard (NAAQS) for ozone of 0.12 ppm and peak 8-hour average concentrations rise above the recently proposed revised ozone NAAQS level of 0.08 ppm. Figure 1 shows counties containing monitors with design values in excess of the current 1-hour NAAQS; Figure 2 shows those areas with design values in excess of the proposed 8-hour standard. In the East, Violations of the 1-hour standard are restricted to the Northeast urban corridor, the Lake Michigan area, and the immediate vicinity of other major urban areas. The number and geographic extent of counties violating the 8-hour standard are significantly greater. A significant feature of ozone concentrations in the East is that, in contrast to other parts of the country, ozone concentrations well in excess of the 30 - 40 ppb tropospheric background level are observed throughout most of the eastern U.S., including locations outside of the major urban nonattainment areas. As a result of these large-scale elevated ozone events and the proximity of nonattainment areas to one another, many states have found that NAAQS attainment cannot be achieved with any reasonable level of local emission control measures and therefore that a regional, multi-state emission control strategy is necessary. The Ozone Transport Assessment Group (OTAG) was formed to deal with this issue.

Air Quality Analysis Work Group Activities

Within OTAG’s Modeling and Assessment Subgroup, the Air Quality Analysis Workgroup (AQAWG) was formed to evaluate and perform analyses of air quality and meteorological data relevant to OTAG’s mission. Data analyses conducted by AQAWG members complement work performed under the auspices of the Regional and Urban Scale Modeling Workgroup. AQAWG members contributed numerous analyses as listed in Table 1. Some of the analyses were based on ozone and precursor data from intensive field measurement programs such as the Nashville/Middle Tennessee Ozone Study and the NARSTO-Northeast study that provide detailed spatial and temporal coverage of ozone, ozone precursor and meteorological conditions both at the surface and aloft during a limited number of individual ozone episode events. Other analyses were based on research-grade ozone and precursor monitoring programs conducted over multi-year periods at a limited number of locations such as the SOS/SCION network. Still other analyses were based on the routine ozone and meteorological data which is collected at regular time intervals at a large number of sampling sites located throughout the OTAG domain. The latter two types of analyses were designed to identify major features of the ozone climatology within the OTAG domain which provide insights into ozone/precursor relationships and transport issues. A major advantage of these results is that, in contrast to modeling studies and analyses based on intensive measurements of individual ozone episodes, they provide information on ozone, ozone/precursor relationships, and ozone and precursor transport over the full range of meteorological conditions in the OTAG domain.

Analyses contributed by AQAWG members were presented and discussed during monthly AQAWG meetings. Comments made during these meetings frequently resulted in significant revisions and additions. Formal written reports are currently available for many of the analyses. Most of these reports have been posted on the AQAWG world wide web site (http://capita.wustl.edu/OTAG) for widespread distribution and additional discussion and review. Based on these reviews, summaries of many of the analyses were prepared. These summaries, which appear in the final section of this volume, include brief discussions of study purpose, methods, results, technical limitations, and the implications of the study’s findings for OTAG. Draft summaries were provided to the study authors for review and comment before being finalized. Information contained in the final reviews was then used as the basis for an integrated assessment of analysis results as presented in the next section.

Unfortunately, due to resource constraints, it was not possible to prepare summaries for all of the analyses which were presented to or discussed by the Workgroup. Table 1 indicates which analyses were summarized and which were not. Although the following discussion is based primarily on the analyses for which summaries were prepared, references are made in some cases to major results from some of the other studies. It should be recognized, however, that these results were not subjected to the same AQAWG review process as those from studies for which summaries are provided here, although in some cases these studies may have been reviewed by other groups. For clarity, references shown in bold in the following sections are among those for which summaries were prepared; these summaries can be found in Volume III, Summaries of Individual AQAWG Analyses.

Overview of Regional Ozone and Transport in the Eastern United States

The regional nature of elevated ozone episodes in the eastern US has long been recognized (see, e.g, NRC, 1991; OTC, 1994; SOS, 1995; Husar, 1996a). A review of measurement studies and statistical data analyses conducted by the National Research Council (NRC, 1991) identified elevated ozone events on spatial scales exceeding 600,000 km2 lasting from a few days to as long as a week with concentrations exceeding at least 80 ppb for a major portion of each day. The most wide spread regional episodes were found to be associated with slow moving high pressure weather systems that produced high temperatures, light winds, limited vertical mixing, and mostly clear skies. Air movement around the high, the slow (typically eastward or northeastward) migration of the high, and the influence of transient weather features on the fringes of the high can result in transport of elevated ozone and precursor concentrations over significant distances. Within this regional "sea" of elevated ozone, imbedded urban ozone plumes consisting of significantly higher concentrations can be found. Surface and aloft observations have shown that these urban plumes can be distinguished over a period of 12 hours covering an area of up to 8,000 km2 (NRC, 1991)

Given the observational evidence cited above, it is clear that transport mechanisms acting to produce the large-scale regional ozone excess above tropospheric background in the eastern U.S. have the potential to contribute significantly to exceedances of the current 1-hour or proposed 8-hour ozone standards. The direction and spatial extent of transport and the relative contribution of transported ozone and precursors to individual ozone exceedances is highly variable. On the one hand, some episodes, such as the July 11-15, 1995 Nashville episode analyzed by Meagher (1996) may be entirely local and not involve any transport. During this episode, ozone concentrations in downtown Nashville reached 138 ppb, more than 55 ppb above the daily maximum value at any surrounding site and about 80 ppb above values near the boundary of the metropolitan area. On the other hand, as indicated by Blumenthal et al. (1997), a number of one day and multi-day transport mechanisms have been observed along the coastal plain in the Northeast, some or all of which may be involved in any given ozone exceedance episode. These mechanisms include:

1. Near surface flows acting within the first few hundred meters above ground level where surface frictional, radiative and sensible heating processes are important. In coastal areas, this includes offshore, over water, and onshore flows; orographic effects (drainage flows, blocking, channeling) are also important in areas with significant terrain.

2. Boundary layer synoptic flows occurring within the daytime planetary boundary layer but above the surface layer at heights (typically 800 - 1500 m) where surface influences are less important and synoptic scale pressure gradient forces are relatively more important.

3. Channeled nighttime flows in low level jets which form under stable conditions just above the nocturnal boundary layer (typically 200 - 800 m above sea level under episode conditions in the Northeast) and are enhanced by the channeling effects of major terrain features (such as the Appalachians in the Northeast).

Within any given episode, each of the above flow features may be characterized by different directions, average speeds, and temporal and spatial variations. Surface flows are typically more sluggish and more localized than the synoptic flows and exhibit a strong diurnal pattern. In most cases, winds are calm or near calm during the nighttime and early morning hours under episode conditions. Surface flows are thus primarily involved in near-field transport whereas the synoptic flows are capable of transporting material for two or more days over considerably larger distances. The relatively high speed channeled nighttime flows can move material long distances overnight close to the surface where it is available to be mixed down to the surface as the mixing depth increases the next morning. Based on surface and upper air observations of winds and back trajectory calculations during Northeastern ozone episodes, Blumenthal and co-workers conclude that near surface transport along the urban corridor can cover distances of 50 - 250 km from morning until evening, boundary layer synoptic flows can transport material 200 - 600 km across the Appalachians in a 24-hour period, and channeled nighttime flows such as those observed by Ray et al. (1997) can transport material 200 - 400 km overnight. While not all of the three basic flow regimes described above will be important at all locations, some or all may play important roles in ozone and precursor transport in various nonattainment regions throughout the OTAG domain. Thus, the factors contributing to transport are quite complex. The AQAWG conducted analyses designed to identify the relative contributions of these transport mechanisms to ozone exceedances at different locations in the OTAG domain.

INTEGRATED SUMMARY OF WORKGROUP ANALYSES

Analyses conducted by the AQAWG provide information in six key areas:

An integrated summary of AQAWG analysis results as they pertain to these six key areas is provided in the following sections. This is followed by a discussion of recommendations for future analyses. Summaries of individual analyses are provided in the final section of this volume.

Spatial and Temporal Pattern of Ozone in the OTAG Domain

To better understand the temporal and spatial patterns of regional ozone levels associated with transport impacts in the eastern U.S., an integrated set of ozone monitoring data from both the primarily population-oriented AIRS network (consisting of the National Ambient Monitoring Station and the State and Local Ambient Monitoring Station networks) and several rural networks (CASTNet, IMPROVE, SCION, EMEFS) was constructed (Husar and Husar, 1996). Analyses of this database (Husar, 1996a,b,c) revealed the following key features:

Although no effort was made to estimate the flux of ozone and precursors across the OTAG domain boundary, these observations suggest that, with the possible exception of the international border along the Quebec-Windsor corridor, air masses entering the OTAG domain contain little or no ozone burden above tropospheric background levels (and precursor fluxes into the domain are likely to be equally negligible). Thus, ozone concentrations above approximately 35 ppb observed within the OTAG domain must be generated by sources within the domain. Due to the pervasive influence of anthropogenic activities, present-day monitoring data cannot provide any information on the likely range of concentrations which would be observed in the OTAG domain in the absence of any anthropogenic influences. While seasonal mean daily maximum concentrations along some parts of the OTAG boundary where anthropogenic influences are minimized are in the 30 - 40 ppb range as noted above, climatic conditions and biogenic inventories within the central portion of the domain differ markedly from those in these boundary regions and may conspire to produce somewhat higher "natural" background ozone concentrations. On the other hand, some modeling estimates of pre-industrial ozone levels suggest that lower values might be expected in the absence of any anthropogenic influence (e.g., Levy et al., 1997) but these have not been reviewed for this report.

Additional findings from the spatial and temporal pattern analyses include:

Transport Analyses

Several analyses conducted by the AQAWG focussed on identifying the spatial and temporal extent and magnitude of the transport of ozone and precursors. Measurements of ozone, precursors, and meteorological conditions conducted as part of both the 1995 SOS Nashville/Middle Tennessee ozone study, and the 1995 NARSTO Northeast field study were analyzed for this purpose (Edgerton and Hartsell, 1996; Edgerton, 1997a; Ray et al., 1997; Vukovich, 1996; Blumenthal et al., 1997; Hudischewskyj and Douglas, 1997). In addition, long-term data sets of routine air quality and meteorological measurements were used to examine spatial ozone and temperature correlation patterns and to compute air parcel trajectories (Poirot and Wishinski, 1996a,b,c; Wishinski and Poirot, 1996; Schichtel and Husar, 1996, 1997). Daily back trajectories ending at 23 monitoring sites located throughout the OTAG domain were computed using seven summers (1989 - 1995) of three-dimensional gridded wind fields and variations in the trajectory patterns with ozone concentrations at the receptor sites were analyzed. Principal findings of these analyses are summarized below.

One day transport of ozone within plumes from urban areas and major power plants has been demonstrated using aircraft data (NRC, 1991 and references therein). In addition, impacts of distinct power plant plumes and urban plumes have been noted at several rural sites in the Southeast (Edgerton and Hartsell, 1996). Aircraft measurements of NOx, NOz, and O3 made within the Cumberland and Johnsonville power plant plumes located near Nashville (Meagher, 1996) indicated that, during these summer daytime measurement periods, essentially all of t

he NOx is converted to NOz within 30 - 100 km of the source. Thus, ozone production due to NOx in the plume ceases within 100 km of the power plant. These results differ somewhat from those the earlier Tennessee Plume Study in which aircraft measurements of the Cumberland plume conducted in August, 1978 indicated ozone production continuing beyond 110 km to at least 160 km (Gillani and Pleim, 1996). Differences in the distance estimates between these two studies may be related to differences in mean plume transport wind speeds, as well as other meteorological and air quality factors. It is important to note that, under the right conditions, power plant plumes may travel relatively long distances overnight with little loss of NOx and thus be available to participate in photochemical reactions at distant locations on the following day. Trainer (1993) shows that ambient O3/NOz concentration ratios provide a measure of the efficiency of ozone production, i.e. essentially the amount of ozone produced during the lifetime of each NO2 molecule. Planetary boundary layer measurements of O3/NOz ratios from aircraft and surface sites made within the power plant plumes and the Nashville urban emissions plume indicate that ozone formation efficiency as measured by this ratio is 65 - 70 percent greater in the urban plume as compared to the power plant plumes (Meagher, 1996). Thus, on a per unit emissions basis, urban area source NOx emissions may be more important to same day ozone formation than elevated point source NOx emissions.

Conditions during ozone episodes are conducive to longer transport distances over one to two day periods in the northeastern portion of the OTAG domain (coinciding with higher average wind speeds and a more organized flow pattern) than in the southeastern portion of the domain where average wind speeds are lower and the flow pattern is less well organized. Evidence of this can be found in the "source regions of influence" and residence time analyses conducted by Schichtel and Husar (1997, 1996), and in the residence time analysis, autocorrelations, and lagged inter-regional correlations computed by Poirot and Wishinski (1996a,b,c).

Analyses of the July 11-15 ozone episode in Nashville (Meagher, 1996) indicate that this episode was primarily a result of local emissions - the area of high ozone (with concentrations up to 80 ppb above a 60 - 70 ppb regional background) was restricted to the immediate metropolitan area and did not extend away from it in any direction. However, the frequency of such "home grown" episodes vs. regional episodes was not examined.

Porter et al. (1996) present spatial correlation results for the short-term (deseasonalized) ozone component that suggest a spatial scale on the order of 560 - 640 km (350 - 400 miles) for same-day formation and transport. Schichtel and Husar’s source regions of influence analysis indicates a scale of 300 - 600 km for one day transport times and 450 - 1000 km for a two day transport times in the northern half of the OTAG domain. In the southern half of the domain, one day scales were found to be somewhat less: 200 - 400 km (one day) and 300 - 800 km (two day). Similarly, analysis of typical transport distances and trajectory residence times for days in the upper 20 percent of daily maximum ozone in each portion of the OTAG domain show relatively large transport distances and short residence times in the Northeast and short transport distances and long residence times in the South. Interpretation of these results is subject to two important caveats:

1. Given the somewhat weak but nevertheless significant correlations of Porter’s short-term ozone component with temperature and the likely influence of other meteorological factors on day-to-day variations in the short-term ozone component, it is not possible to distinguish the degree to which the spatial scales implied by this analysis are representative of actual ozone and precursor transport or merely representative of correlations in meteorological conditions between locations.

2. The spatial scales obtained from Schichtel and Husar’s source regions of influence analysis represent the typical distances air parcels are estimated to travel over one and two day intervals and p7rovide no information on the relative contributions of precursor sources to ozone concentrations at the given downwind distances.

Trajectory model results do not show actual spatial transport scales but do indicate potential source regions along the pathways of air parcels associated with above average ozone at receptor sites. Back trajectory analyses conducted by Poirot and Wishinski (1996a,b,c) indicate that air parcels originating outside of the OTAG domain are associated with below average ozone at various locations throughout the OTAG domain. In contrast, air parcels associated with above average ozone concentrations at these receptor sites predominantly originate from the heart of the OTAG domain. Remarkably, this was found to be the case no matter where in the domain the receptor site is located. Of course, it should be noted that the trajectory analysis does not take into consideration atmospheric chemical processes, the injection of precursor emissions, or the deposition of material along the trajectory path. Therefore, while these results provide consistent circumstantial evidence that emissions from the central portion of the domain contribute in some way to above average ozone events throughout the domain, the relative contribution of precursor emissions along various portions of the trajectory paths leading to the receptor sites (i.e., emissions originating near the upwind end of the trajectory vs. emissions originating near the receptor site) are not known.

Development of Control Strategies

Comparisons of historical trends in ozone and precursor concentrations with trends in precursor emission levels can provide insights into the relative effectiveness of control strategies. Unfortunately, trends in the annual extreme ozone concentrations of interest are sensitive to the influences of interannual variations in meteorological conditions, movement of monitoring sites, missing data, and random fluctuations in concentrations which can make identification of consistent trends difficult. Furthermore, high quality data on ambient precursor trends is extremely limited and estimates of precursor emissions are subject to potentially significant biases. Despite these difficulties, several important features can be noted in the trends of ozone and precursor emissions between the early to mid-1980s and the mid-1990s as described below. Many of the analyses presented here were summarized by Morris, 1996.

The above results do not lend themselves to formulation of a definitive conclusion regarding causal relationships between changes in VOC and NOx emissions and ambient ozone concentrations. When looking at emissions and ozone trends from the mid-1980s to the mid-1990s for various ozone trend measures at various locations, the most consistently (although by no means universally) identifiable feature is the simultaneous decline in peak ozone concentrations and VOC emissions between 1988 and 1991. This result by itself, however, does not mean that the observed decrease in ozone is attributable in whole or even in part to the drop in VOC, although it is reasonable to conclude that VOC decreases contributed at least partially to the ozone decline outside of biogenic dominated southern locations. It should be noted that comprehensive analyses of ambient NOx or VOC concentration trends which might validate the emission trends have not been conducted. Rich Poirot of the Vermont Department of Environmental Conservation has noted examples of apparent discrepancies between ambient NO and NO2 concentration trends, nitrate deposition trends, and NOx emission trends which warrant further investigation. In particular, EPA (1995) reported a three percent increase in national total NOx emissions between 1985 and 1994 while over the same period, the national composite NO2 concentrations decreased by nine percent.

It should also be noted that future ozone/precursor trend relationships may differ from those of the past. For example, if the ozone peak downwind of an urban area results from the superposition of an ozone plume generated under VOC limited conditions by the urban area with a regional background "cloud" of elevated ozone formed largely under NOx limited conditions, then urban VOC reductions, such as those which occurred between the mid-1980s and mid-1990s, may reduce the total ozone peak by reducing the urban area’s contribution. However, as the peak is lowered, the regional, NOx limited contribution to the total peak becomes relatively more important. Thus, future VOC reductions may be less effective at reducing ozone concentrations than in the past. This effect would be magnified for exceedances of an 8-hour average ozone standard as compared to a 1-hour standard since the 8-hour averages are less sensitive to sharp 1-hour concentration peaks downwind of urban areas.

Impact of an 8-Hour Ozone Standard

EPA’s proposed revisions to the ozone NAAQS is based on an 8-hour average daily maximum concentration instead of the 1-hour daily maximum measure used in the current standard. Although closely related, spatial and temporal patterns of peak 8-hour averages, the role of transport, and the response of 8-hour values to control strategies are somewhat different than for one hour averages. AQAWG analyses of these issues produced the following principal findings:

No analyses of the relative effectiveness of various control strategies for 1-hour vs. 8-hour ozone have been conducted by the AQAWG. Relative effectiveness is dependent on the complex interaction of many factors and can be expected to vary significantly from one situation to the next.

Representativeness of OTAG Episodes

An issue of concern in the interpretation of control strategy modeling results is the degree to which the set of episodes selected for modeling represent the full range of conditions under which ozone episodes occur. If certain types of episodes are not well represented by the selected episodes, model results may give a misleading picture of control strategy impacts. The AQAWG examined the representativeness of episodes modeled by the OTAG Regional and Urban Scale Modeling Workgroup ("OTAG episodes") from two perspectives:

1. How does the spatial distribution of ozone on OTAG episode days compare with that observed on high ozone days in general?

2. How do forward and backwards trajectories computed as part of the "source regions of influence" and "trajectory residence time" analyses discussed above compare between OTAG episode days and high ozone days in general?

While the general spatial patterns of daily maximum ozone concentrations during the modeled episodes in 1988, 1991, 1993, and 1995 were similar to the pattern of 90th percentile daily maximum concentrations for all days during the 1991-1995 ozone seasons, there are clear differences from episode to episode (see Figure 3). Generally speaking, the 1991 and 1995 episodes are more similar to each other than the 1993 episode which was characterized primarily by high ozone levels in the South with lower values elsewhere. Similar results are seen in the comparisons of transport vectors and normalized residence times between the modeled episodes in each year (Figure 4). Each episode is characterized by different regions of high residence times, with the 1993 episode differing markedly from the other two with high residence times in the South and low residence times in the North.

OTAG Model Evaluation

Although the primary model evaluation studies for OTAG were conducted by the Regional and Urban Scale Modeling Workgroup (RUSMWG), the AQAWG also conducted several diagnostic comparisons of observed and predicted ozone and ozone precursor concentrations which shed light on the suitability of the OTAG photochemical model results for evaluation of regional control strategy impacts (Hartsell and Edgerton, 1996; Edgerton, 1997b; Morris et al., 1997; Imhoff, 1996; Lurmann et al., 1997). The principal findings of the RUSMWG and AQA analyses are discussed below.

Ozone: Comparisons of observed surface ozone concentrations with model predictions for all four modeled episodes (1988, 1991, 1993, and 1995) were conducted under EPA sponsorship (OTAG, 1996). Results of this analysis indicated "generally good agreement between simulated and observed values; [with] no large positive or negative biases." (ibid., p. 6-1). Comparisons of observed and predicted concentration fields indicate that the overall spatial and temporal ozone patterns appear to have been properly simulated (OTAG, 1996; Husar and Schichtel, 1996). However, there was a regional bias in the model results with under prediction of concentrations on average in the northern half of the domain and over prediction in the southern half of the domain. Furthermore, for some episodes and in some regions, a "bias creep" was observed in which concentrations were under predicted on average at the beginning of the episode and over predicted at the end of the episode.

Comparisons of observed surface ozone concentrations during the July, 1995 episode with concentrations predicted by the OTAG UAM-V runs for a suburban and a rural site near Nashville revealed a fair amount of scatter but no bias. However, comparisons aloft indicated that the model consistently under predicted ozone concentrations above the mixed layer. This result is consistent with under predictions aloft during the morning hours along the western boundary of the Northeast urban corridor noted by Lurmann, 1997. There are several possible reasons for this and the current analyses do not allow us to distinguish among them.

Comparisons of observed and modeled ozone for nine rural sites in the South and Northeast for the July 1995 episode revealed a tendency for the model to over-estimate the frequency and duration of ozone values in the 80-100 ppb range for these rural sites. If confirmed by analyses for a larger group of sites, this finding would indicate a potential problem with using OTAG results to provide boundary conditions for urban scale modeling.

Isoprene: Comparisons of observed mid-afternoon isoprene concentrations at monitoring sites located in various parts of the OTAG domain with predicted surface layer isoprene concentrations for corresponding model grid cells revealed a tendency for the model to over predict isoprene (Morris et al., 1997; Edgerton, 1997b). More than three-fourths of the sites examined in these studies exhibited biases greater than ±25% and nearly half the sites exhibited mean biases of greater than ±50% of mean observed values. Nearly all the biases over ±50% were over predictions. No geographic pattern in the magnitude or sign of the bias is evident in the results and the magnitude of the over predictions greatly exceed the level of bias one might expect from comparing a near surface concentration with a (50 m) vertically averaged value for the model grid cell. These results point to a potentially significant problem with the UAM-V/BEIS2 modeling system: if isoprene emissions in the OTAG inventory are significantly overestimated (for whatever reason), the over representation of biogenic VOCs would cause the model to underestimate the expected response to anthropogenic VOC controls. In addition, the model would tend to be overly sensitive to NOx controls in all but the largest urban areas.

Other precursors: An analysis of CO/NOY and VOC/NOY ratios at a Nashville urban site during 1995 found good agreement between measured and modeled ratios, indicating consistency between the modeled and ambient CO/NOX and VOC/NOX emission ratios around this site (Imhoff, 1996). The intercepts at zero NOY from these analyses are indicative of the background concentrations of CO and VOC: the analyses indicated that the model tended to under estimate the CO background and over estimate the VOC background at this site. The discrepancy between the apparent background for VOC of 90 ppb in the model compared to near zero in the ambient data possibly indicates that a source of VOC (but not CO and NOY) is overestimated in the model. The tendency for the model to under estimate the CO background at this urban site is surprising given that the model had a tendency to over predict CO at two rural sites in Georgia and Tennessee: however, these findings are not necessarily inconsistent. The tendency to over predict CO at the rural sites may be indicative of emissions or atmospheric mixing problems in this region.

Chemistry: Comparisons of observed and predicted ozone production efficiencies based on O3/NOZ ratios showed good agreement at rural site near Nashville in 1995; both observed and predicted O3/NOZ ratios were about 6 suggesting ozone formation was NOX limited (Imhoff, 1996). At a Nashville suburban site, there was more variability in both the observed and (particularly) modeled O3/NOZ ratios, making it difficult to draw conclusions about model performance for this location. The average observed O3/NOZ ratio (about 3) was lower for the suburban site than the rural site, suggesting ozone formation was relatively less NOX limited and more VOC limited at the suburban site. Similarly, a comparison of modeled and observed O3/NOY relationships revealed good agreement across 9 rural sites in the South and Northeast in 1995; both observed and predicted O3/NOY ratios were about 9 suggesting ozone formation was NOX limited (Hartsell and Edgerton, 1996). The good performance of the model chemistry for O3/NOY (and O3/NOZ) relationships under NOX limited conditions supports the use of the model for evaluating regional control strategies. This finding is particularly relevant to OTAG because ozone formation is NOX limited over much of the OTAG domain.

 

 

Recommendations for Additional Analyses

Further work is recommended in a number of different areas as outlined below. The infrastructure consisting of the community of analysts brought together by the OTAG process and associated communications capabilities (regular meetings, conference calls, web sites, mailing lists) has proven to be a valuable resource for conducting policy-relevant research and efforts should be made to maintain this infrastructure in the future. Specific areas recommended for future inquiry include:

Several AQAWG members also suggested that the current ozone, ozone precursor, and meteorological monitoring network be reconfigured and enhanced to better address the issue of regional elevated ozone events, ozone and precursor transport, and photochemical model evaluation. Currently, there is a lack of routine air quality and meteorological data suitable for analysis of transport over a broad range of episode conditions. In particular, the current monitoring network is primarily focussed on the larger metropolitan areas whereas spatial coverage in less populated areas or along boundaries of political jurisdictions is quite limited. The current monitoring network is also primarily geared towards monitoring ozone attainment status and trends. Measurements of ozone aloft or of ozone precursors are quite limited. Such measurements are needed both for transport assessments and to better evaluate the performance of photochemical models used to estimate control strategy impacts.

 

Table 1. Technical analyses considered by the Air Quality Analysis Workgroup.

 
 

Summary?

Written

Report

Available?

WWW

Document?1

1. Spatial and temporal pattern of ozone and precursors

     

Spatial Pattern of Daily Maximum Ozone (Husar, 1996a)

Y

Y

Y

Spatial Pattern of Daily Maximum 8-Hour Ozone (Husar, 1996b)

Y

Y

Y

Weekly Pattern of Ozone (Husar, 1996c)

Y

Y

Y

Spectral Decomposition of O3 time Series (Porter et al., 1996)

Y

Y

Y

Seasonal Pattern of Ozone over the OTAG Region. (Husar, 1997)

N

Y

Y

Historical Perspective on the Climatological Potential for "Local" Pollution Episodes (Poirot and Wishinski, 1997)

N

Y

Y

Representativeness of 1995 Ozone Season (Chinkin et al, 1996)

Y

Y

---

Review of Ozone Trend Studies (Morris, 1996)

Y

Y

---

Nashville/Middle TN Ozone Study (Meagher, 1996)

Y

Y

---

 

2. Transport of ozone and precursors

 

 

 

 

 

 

Air Trajectory Analysis of Long-Term Ozone Climatology (Poirot and Wishinski, 1996a,b,c; Wishinski and Poirot, 1996)

Y

Y

Y

Source Regions of Influence (Schichtel and Husar, 1997, 1996)

Y

Y

Y

Ozone/Tracer/NOy Relationships at Three SOS SCION Sites (Edgerton and Hartsell, 1996)

Y

Y

Y

Analysis of Ozone, NOy, and Tracer Data from a Site in South-Central PA (Edgerton, 1997a)

Y

Y

Y

Analysis of low-level jets using NARSTO Northeast data (Ray et al., 1997)

N

Y

---

Intra-Annual and High Frequency Variations at SOS-SCION sites (Vukovich, 1996)

N

Y

---

Transport and Mixing Phenomena Related to Ozone Exceedances in the Northeast U.S. (Blumenthal et al., 1997)

N

Y

Y

Classification of Ozone Episodes for Four Southern Cities According to Transport Characteristics (Hudischewskyj and Douglas, 1997)

N

Y

Y

 

3. Model performance evaluation

 

 

 

 

 

 

Model/Measure comparisons at nine regional sites (Hartsell and Edgerton, 1996)

Y

Y

 

Y

Ambient Monitoring Sites for OTAG (time series) Model Evaluation (Poirot, 1996)

Y

Y

Y

Observed/Predicted isoprene comparison (Edgerton, 1997b; Morris et al., 1997)

Y

Y

Y

Comparison of SOS Nashville Data to OTAG 1995 Base Model (Imhoff, 1996)

Y

---

---

Evaluation of the UAM-V Model Performance in the Northeast Region for OTAG Episodes (Lurmann et al., 1997)

N

Y

Y

 

4. Episode representativeness

 

 

 

 

 

 

Episode "Representativeness" (preliminary view from backward trajectory perspective) (Poirot and Schichtel, 1996)

N

---

---

Comparison of ozone distributions during OTAG modeled episodes with climatological distributions (Husar, 1997b)

N

---

---

 

 

1. A "Y" in this column indicates documentation of the study is available on the AQAWG world wide web page at the following URL: http://capita.wustl.edu/OTAG.

Figure 1: Areas not in attainment of the current 1-hour National Ambient Air Quality Standard for ozone.

Figure 2. Areas not in attainment of the proposed 8-hour, 0.08 ppm national ambient air quality standard for ozone

 

Figure 3. Comparison of average daily maximum ozone concentrations for days in the 1988, 1991, 1993, and 1995 OTAG modeled episodes with 90th percentile daily maximum values from the 1991 - 1995 ozone seasons (Source: Husar, 1997).

 

Figure 4. Transport vectors and normalized residence times over the Eastern U.S. for pollutants with one day lifetimes during the a) 1991, b) 1993, c) 1995 OTAG modeled episodes and d) the combined ‘91, ‘93, and ‘95 episodes (Source: Schichtel and Husar, 1997).

REFERENCES

Blumenthal, D.L., F.W. Lurmann, N. Kumar, T.S. Dye, M.E. Korc, 1997. Assessment of transport and mixing and OTAG model performance for northeast U.S. ozone episodes: Summary of results. STI-996133-1710/1716-S, Sonoma Technology, Inc., Santa Rosa, CA, March, 1997.

Edgerton, E.S., 1997a. Analysis of Ozone, NOY and Tracer Data from a Site in South-Central Pennsylvania. (http://capita.wustl.edu/otag/Reports/Reports.html)

Edgerton, E.S., 1997b. Comparison of Modeled versus Observed Isoprene Concentrations at Rural and Suburban Sites Across the Eastern U.S. (http://capita.wustl.edu/OTAG/Reports/Isoprene/ISOPRENE.html)

Edgerton, E.S. and B.E. Hartsell, 1996. Ozone/tracer/noy Relationships at Three Sos-scion Sites. (http://capita.wustl.edu/otag/Reports/Reports.html)

EPA, 1994. National air pollutant emission trends, 1900-1994. EPA 454/R-95-007.

EPA. 1995. National Air Quality and Emissions Trends Report, 1994. EPA 454/R-96-005, October, 1995.

Gillani, N. V. and J. E. Pleim, 1996. Sub-grid-scale Features of Anthropogenic Emissions of NOX and VOC in the Context of Regional Eulerian Models. Atmos. Environ., 30, 2043-2059.

Harstell, B.E. and E.S. Edgerton, 1996. A comparison of modeled and measured ozone, NOy and CO at nine regional monitoring stations during the 1995 OTAG episode. 24 November 1996 (http://capita.wustl.edu/otag/Reports).

Hudischewskyj, A.B. and S.G. Douglas, 1997. Classification of ozone episodes for fourj southern cities according to transport characteristics. SYSAPP-97/03, Systems Applications International, San Rafael, CA, March, 1997.

Husar R.B., 1996a. Spatial Pattern of Daily Maximum Ozone Over the OTAG Region, 16 September 1996 (http://capita.wustl.edu/otag/Reports/otagspat/otagspat.html).

Husar R.B., 1996b. Pattern of 8-Hour Daily Maximum Ozone and Comparison with the 1-Hour Standard. 16 September 1996 (http://capita.wustl.edu/otag/Reports/).

Husar R.B., 1996c. Weekly Pattern of Ozone Over the OTAG Region, 16 September 1996 (http://capita.wustl.edu/otag/Reports/otagweek/otagweek.html).

Husar, R.B., 1997. Seasonal pattern of ozone over the OTAG region. 16 February 1997 (http://capita.wustl.edu/otag/Reports/otagseas/otagseas.html).

Husar, J.D. and R.B. Husar, 1996. Preparation of ozone files for data analysis and model comparison. 20 September 1996 (http://capita.wustl.edu/otag/Reports/dataanalysis/Analysis.html)

Husar, R. and B. Schichtel, 1996. OTAG Model - Data Comparison Files and Animations. http://capita.wustl.edu/otag/data/ModO3Cmp/ModelDataComp.html

Levey, H., P.S. Kasibhatla, W.J. Moxim, A.A. Klonecki, A.I. Hirsch, S.J. Oltmans, W.L. Chameides, 1997. The global impact of human activity on tropospheric ozone. Geophysical Research Letters, V24, N7, pp. 791-794.

Lurmann, F.W., N. Kumar, R. Londergan, and G. Moore, 1997. Evaluation of the UAM-V Model Performance in the Northeast Region for OTAG Episodes. STI-996133-1716, March, 1997.

Meagher, 1996. Slides from presentation given by Jim Meagher of the Tennessee Valley Authority to the OTAG AQAWG meeting, Washington D.C., July 1996.

Morris, 1996. Review of Recent Ozone Measurement and Modeling Studies in the Eastern United States. ENVIRON International Corp., 21 February 1996.

Morris, R.E., K. Lee, and G. Yarwood, 1997. Phase I Comparison of OTAG UAM-V/BEIS2 Modeling Results with Ambient Isoprene and Other Related Species Concentrations. ENVIRON International Corp., Novato, CA 31 January 1997 (http://capita.wustl.edu/OTAG/Reports/morris/EXECSUM.html).

NRC, 1991. Rethinking the ozone problem in urban and regional air pollution. National Research Council, National Academy Press, Washington, DC, 1991.

OTAG, 1996. Evaluation of the UAM-V Model Performance in OTAG Simulations: Summary of Performance Against Surface Observations. Draft report, Ozone Transport Assessment Group, October 25, 1996.

OTC, 1994. The long-range transport of ozone in the ozone transport region. Ozone Transport Commission, Washington, DC, January, 1994.

Poirot, R. and P. Wishinski, 1996a. VT DEC Air Trajectory Analysis of Long-Term Ozone Climatology: Status Report to OTAG Air Quality Analysis Workgroup, 12/3/96. 3 December 1996 (http://capita.wustl.edu/otag/reports/Status_Dec96/Status_Dec96.html)

Poirot and P. Wishinski, 1996b. VT DEC Air Trajectory Analysis of Long-Term Ozone Climatology: Status Report to OTAG Air Quality Analysis Workgroup, 11/7/96. 7 November 1996 (http://capita.wustl.edu/otag/Reports/VTTRAJ5/StatusReport1.html)

Poirot and P. Wishinski, 1996c. VT DEC Air Trajectory Analysis of Long-Term Ozone Climatology. Status Report to OTAG AQA Workgroup: 8/15/96. 15 August, 1996 (http://capita.wustl.edu/otag/Reports/vtdecair/vtdecair.html)

Poirot, R. and P. Wishinski, 1997. Historical perspective on the climatological potential for "local" pollution episodes. Vermont Department of Environmental Conservation, 3 February 1997 (http://capita.wustl.edu/otag/Reports)

Rao, S.T., E. Zalewsky, and I.G. Zurbenko, 1995. Determining temporal and spatial variations in ozone air quality. J. Air and Waste Management Assoc., V45, pp.57-61.

Ray, S.E., M.E. Korc, M. Byars, T.S. Dye, C.G.Lindsey, T. Haste, P.T. Roberts, 1997. Analysis of nocturnal low-level jets in the Northeastern United States during the summer of 1995. STI-996440-1706-DFR, Sonoma Technology, Inc., Santa Rosa, CA, January 31, 1997.

Schichtel, B. and P. Wishinski, 1996. HY-SPLIT and CAPITA Monte Carlo Model Back Trajectory Comparison. Summary Draft, 1 July 1996 (http://capita.wustl.edu/otag/Reports/Trajcomp/trajcomp.html)

Schichtel, B. and R.B. Husar, 1996. Source Regions of Influence for High and Low Ozone conditions in the Eastern U.S. 8/08/96 Summary Draft. (http://capita.wustl.edu/otag/Reports/Reports.html)

Schichtel, B. and R.B. Husar, 1997. Update on the Characterization of Transport over the Eastern US. February 1, 1997 (http://capita.wustl.edu/otag/Reports/Reports.html).

SOS, 1995. The state of the southern oxidants study: Policy-relevant findings in ozone pollution research, 1988-1994. Souther Oxidants Study, College of Forest Resources, North Carolina State Univ., Raleigh, NC, April, 1995.

Trainer, M. et al., 1993. Correlation of O3 with NOy in photochemically aged air. J. Geophys. Res., 98, 2917-2925, 1993.

Vukovich, F.M., 1996. A study of the intra-annual and high frequency variations of ozone and its precursors in the southeastern United States. Synoposis of the OTAG AQAWG presentation made August, 1996; Science Applications International Corp., Raleigh, NC.

Wishinski, P. and R. Poirot, 1996. Air Trajectory Residence Time Analysis Investigation of Ozone Transport Pathways: 1989-95. 20 February 1996 Summary Draft (http://capita.wustl.edu/otag/Reports/Restime/Restime.html)

Wolff, G.T. and P.E. Korsog, 1994. Statistical trends in ozone in the eastern U.S. Presented at A&WMA Specialty Conference on Tropospheric Ozone, Orlando FL, May 1994.