3

Review of the Draft NARSTO Assessment Document

GENERAL COMMENTS ON STRUCTURE, COVERAGE, AND PRESENTATION

In the judgment of this NRC committee, the draft NARSTO assessment document provides a commendable review of selected scientific issues related to the urban and regional ozone problem. The topics covered are important, and the treatment of each is careful and brings to bear critical information, including information that has been developed since the previous reviews were conducted. The document provides a cogent summary of those issues that will be of substantial utility to scientists who already understand the basic character and dimensions of ozone problems and they will benefit from knowledge of developments outside their own fields of expertise.

In developing its recommendations for improving the assessment document, the committee was aware that NARSTO may decide to retain a scientific-review focus for the assessment document, and perhaps address the necessary policy-relevant aspects through some other endeavor. However, to facilitate discussions in this report, the committee presents its recommendations as if the revised assessment document were intended to have the fullest policy-relevant perspective.

The strengths of the NARSTO assessment document as a scientific review are mirrored by its weakness as a document appropriate for an audience of “decision makers and other stakeholders tasked with managing air quality in North America” (Preface to the document). The document does not contain an orderly presentation of the major elements of the ozone problem that starts with policy goals and then lays out the issues to be resolved; nor does it note where scientific knowledge appears to be sufficient, where important uncertainties exist, and where and how further advances would assist in ozone management in North America. Such a framework would add perspective to the specific topics selected for extensive discussion in the document. It would also guide the presentations on each topic, and in particular would direct the authors to explain better why the conclusions are important to the policy-making community. Further, such a framework would focus on what additional information could contribute to regulatory and other governmental decisions and thereby help build a constituency for needed research. As it is, the reader is largely left to infer why topics are presented, without any guidance as to the policy



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 12
REVIEW OF THE NARSTO DRAFT REPORT: AN ASSESSMENT OF TROPOSPHERIC OZONE POLLUTION–A NORTH AMERICAN PERSPECTIVE 3 Review of the Draft NARSTO Assessment Document GENERAL COMMENTS ON STRUCTURE, COVERAGE, AND PRESENTATION In the judgment of this NRC committee, the draft NARSTO assessment document provides a commendable review of selected scientific issues related to the urban and regional ozone problem. The topics covered are important, and the treatment of each is careful and brings to bear critical information, including information that has been developed since the previous reviews were conducted. The document provides a cogent summary of those issues that will be of substantial utility to scientists who already understand the basic character and dimensions of ozone problems and they will benefit from knowledge of developments outside their own fields of expertise. In developing its recommendations for improving the assessment document, the committee was aware that NARSTO may decide to retain a scientific-review focus for the assessment document, and perhaps address the necessary policy-relevant aspects through some other endeavor. However, to facilitate discussions in this report, the committee presents its recommendations as if the revised assessment document were intended to have the fullest policy-relevant perspective. The strengths of the NARSTO assessment document as a scientific review are mirrored by its weakness as a document appropriate for an audience of “decision makers and other stakeholders tasked with managing air quality in North America” (Preface to the document). The document does not contain an orderly presentation of the major elements of the ozone problem that starts with policy goals and then lays out the issues to be resolved; nor does it note where scientific knowledge appears to be sufficient, where important uncertainties exist, and where and how further advances would assist in ozone management in North America. Such a framework would add perspective to the specific topics selected for extensive discussion in the document. It would also guide the presentations on each topic, and in particular would direct the authors to explain better why the conclusions are important to the policy-making community. Further, such a framework would focus on what additional information could contribute to regulatory and other governmental decisions and thereby help build a constituency for needed research. As it is, the reader is largely left to infer why topics are presented, without any guidance as to the policy

OCR for page 12
REVIEW OF THE NARSTO DRAFT REPORT: AN ASSESSMENT OF TROPOSPHERIC OZONE POLLUTION–A NORTH AMERICAN PERSPECTIVE significance of the findings (or uncertainties) reported. Although the document often refers to the policy relevance of specific sections, it does so in general terms rather than in a structured way that would enable the reader to understand the implications for action. In this regard, the committee notes that the NARSTO internal review comments (provided to the committee) relating to the failure to address policy questions were not yet adequately addressed by the document 's authors. In short, despite its strengths in synthesizing scientific information on important topics, the document falls short in achieving the objective of being “timely and useful for regulatory initiatives and mid-course corrections” (NARSTO 1997). The second objective for the NARSTO assessment document, according to the executive plan, was to provide guidance for setting NARSTO priorities for longer-term research. The document contributes to this objective by, for example, forcefully directing attention to the shortcomings of current knowledge of emissions and of instrumentation used for monitoring. It clearly identifies the deficiencies of the air-quality modeling systems (AQMSs) used to devise control strategies. It describes the conundrum posed by the diverse geographic impacts of volatile organic compound (VOC) and oxide of nitrogen (NOx) controls in an urban setting. The document does not, however, use these assessments of the science to make recommendations for research priorities, either for the short or the long term. It states that it explicitly chose not to do so and that this task is left to other NARSTO bodies. By focusing on the difficulties of the ozone-control issue without more clearly indicating what might be done to make progress, however, the NARSTO assessment document will leave the reader (especially the policy-oriented reader) with the sense that there is little hope for progress in ozone control. Although the NARSTO assessment document does not examine alternative approaches to achieving ozone compliance, the information summarized will be useful as inputs for those who would attempt to devise such alternatives. An example of such useful information is the discussion of inferences from observations (observation based modeling) as a complementary approach to AQMS for determining the efficacy of VOC and NOx control. The NARSTO objective concerning ozone-precursor chemistry and its integration with related-pollutant issues, is not discussed in context except for a brief discussion in the document of fine particulate formation and growth. In summary, the NARSTO assessment document is a scientific assessment addressed to scientists, and succeeds for the most part in meeting this objective. It addresses, to a greater or lesser degree, the scientific questions posed to it by NARSTO. (Although the document is not organized around those questions, in a helpful appendix, it cross-references sections of the NARSTO assessment document to them.) Unfortunately, the document does not meet the needs of the broader community seeking to know what help science now (or in the future) could provide in dealing with the persistently intractable problem of tropospheric ozone's detrimental impact on ambient air quality. REVIEW OF THE CONTENT OF THE NARSTO ASSESSMENT DOCUMENT As noted above, the major strength of the NARSTO assessment document is that it presents a good review of current understanding of important topics in the atmospheric science of urban and regional ozone. However, the document does little to explain why ozone pollution is a serious problem. A short summary of health and vegetation effects, with reference to major reviews, in the introduction would improve the impact of the document by reminding the reader why it is important to solve the ozone problem. The document should also indicate that a better understanding of tropospheric ozone would allow for a better understanding of ozone effects.

OCR for page 12
REVIEW OF THE NARSTO DRAFT REPORT: AN ASSESSMENT OF TROPOSPHERIC OZONE POLLUTION–A NORTH AMERICAN PERSPECTIVE There should be more discussion of the need for exposure data to be used as inputs for modeling the health and environmental responses. Although the NARSTO assessment document stresses the regional nature of the ozone problem, this is hardly a new finding, and was a key component of the NRC report (1991). The NARSTO assessment document does well to emphasize the fact that background concentrations of ozone in summer are 30 to 40 parts per billion (ppb); this background is likely influenced by emissions from across North America as well as from other continents. The data shown from Staehelin et al. (1994) in Figure 1.7 of the document, taken with recent data from Europe (Scheel et al. 1997) and North America, suggest that summer concentrations of ozone have grown by more than the factor of 2 (as noted in the document) since the nineteenth century. In fact, based on Volz and Kley (1988), background concentrations have grown by a factor of 3 to 4. In addition, the mean summer afternoon concentrations in rural areas of the United States (Logan 1988) and Europe (Scheel et al. 1997) have grown by a factor of 4 to 6. As stated in the document, the 8-hr ozone standard will be harder to meet than the 1-hr standard, because, relative to the 1-hr standard, the 8-hr standard is closer to background concentrations and to concentrations commonly found in rural locations in eastern North America in summer. Regarding the “Key Findings in the Science of Ozone Pollution,” Section III of the NARSTO assessment document focuses almost exclusively on the situation in the United States, with little discussion about Canada, except in the context of observed trends; the unique case of the aggregation of urban communities into Mexico City receives scant attention. Also, there is no discussion of relevant scientific findings from Europe or Asia. Section III clearly articulates the complexity of the VOC-NOx issue, the role of natural emissions, and the deficiencies in instrumentation used for monitoring ozone precursors. However, the section does not adequately address the seventh policy question listed in the appendix of the NARSTO assessment document (also shown in Appendix D of this report). That question asks if efforts to manage ozone will help or hinder efforts to mitigate other environmental problems. A key issue addressed in Section III.1 of the NARSTO assessment document is the question of ozone and precursor trends and whether they are consistent with expectations. The discussion of trends stresses, correctly, that most of the ozone data are from urban locations, and that many of the sites designated as rural might not be truly rural (this applies also to the Fiore et al. (1998) analysis). Although the document correctly states that the average trend as given by the U.S. EPA is downward (Figure 1.3 of the document), and that urban and suburban areas in Canada generally show increases, it does not state that published studies show that about 80% of the Aerometric Information Retrieval System (AIRS) sites in the United States show no statistically significant trend (EPA 1996; Fiore et al. 1998). The document should emphasize that trends and time series in ozone measurement should be given for periods longer than 10 years in the United States. The same method has been used to measure ozone since about 1980, and control measures have been in place for more than 20 years. The discussion of trends in ozone precursors is inadequate, because it does not discuss whether observed trends in VOCs and NO x in Los Angeles and in the northeast United States are consistent with predicted trends in emissions for these regions. Comparisons to trends in U.S. national emissions are hardly relevant. Nor are reasons given for the quoted 25 to 30% decrease in VOCs in 5 years. The reader is left wondering how such a large decrease could happen in only 5 years, given the slow turnover of the motor vehicle fleet. This section also fails to come to any definitive conclusion on the question of consistency.

OCR for page 12
REVIEW OF THE NARSTO DRAFT REPORT: AN ASSESSMENT OF TROPOSPHERIC OZONE POLLUTION–A NORTH AMERICAN PERSPECTIVE Section 111.4 addresses biogenic precursor emissions, and asks, “What Is Nature's Role?” In addition to biogenic emissions, this section should discuss explicitly nature's role with respect to ozone and the dynamic factors, such as variations in the extent of horizontal dispersion, vertical mixing, and temperature, that influence ozone variation in the troposphere. It is not apparent why meteorology is discussed separately from this section on nature's role. Section 111.5 deals with the role of meteorology in a relatively brief and less comprehensive way than the coverage given to chemical issues. Because meteorology is an important factor contributing to air quality (e.g., NRC 1991), the authors should place greater emphasis on meteorology in the abstract of key findings for this section. For example, this could be done in the Executive Summary of the document under the scientific question that asks, “What role does meteorology play in regional and urban ozone pollution episodes?” The single finding states simplistically that “both stationary and transported air masses contribute to the ozone pollution problem.” Figure 111.22 of the NARSTO assessment document states, “Note that the transport regimes shown here represent upper level large scale flow.” Does “upper level” refer to 850 mb, 700 mb, or 500 mb? Is it reasonable to depict upper-level flow with 90th percentile daily maximum ozone concentrations that come from a surface-observing network? The discussion of VOC and NOx control illustrates well the complexity of the issue, but it needs to clarify one point. The strategy of VOC-only control might lower ozone in the urban core of a large metropolitan area, but the lack of NOx control will have negative implications for downwind communities. This section of the NARSTO assessment document describes new empirical findings regarding ozone production efficiency (OPE). It states that these findings suggest that “NOx-based strategies...will require larger emissions reductions than previously estimated.” Because these empirical findings have not, to the committee's knowledge, been used in the United States to determine emissions reductions, it is unclear what the text is referring to. The chemical mechanisms used in air-quality models are discussed only in the context of smog chamber studies. The document suggests that observations of the OPE should also be used to test mechanisms. The document should also discuss more extensively the use of high-quality field observations of ozone precursors and radicals to test such mechanisms (e.g., Hidy In press). Weaknesses in emissions inventories have been identified in previous assessments of ozone pollution (e.g., NRC 1991). The NARSTO assessment document discusses progress (or lack thereof) in quantifying emissions from mobile sources, with the discussion mostly about tunnel studies. Little mention is made of the utility of cross-road measurements. There is no discussion of advances in quantifying area sources of VOCs. The discussion of improvements in emissions modeling seems entirely disconnected from the discussion of underlying problems with actually quantifying emissions. For example, the first conclusion about emissions inventories in the Executive Summary of the document indicates that emissions models have improved, rather than a statement about the reliability of current emissions estimates. The NARSTO assessment document is rightly critical of the instrumentation used for monitoring NOx, because of its poor detection limit, and states explicitly the difficulty of making high-quality VOC measurements. However, the document avoids commenting on the techniques in use at the photochemical assessment monitoring stations (PAMS) sites, although it is clear from Figure 111.26 of the document and from the text that there are likely to be serious problems with data quality. More explicit comments on the PAMS instrumentation and on the quality of the PAMS data are needed. The discussion of improvements needed in monitoring networks (Section 111.6C) appears to be inconsistent with the preceding discussion of problems with current instrumentation. Nevertheless, this section makes important points about the need

OCR for page 12
REVIEW OF THE NARSTO DRAFT REPORT: AN ASSESSMENT OF TROPOSPHERIC OZONE POLLUTION–A NORTH AMERICAN PERSPECTIVE for timely access to data, and for evaluation of data from the monitoring networks. The NARSTO assessment document advocates the use of observation-based methods for examining control strategies, and the committee concurs that these methods are a valuable complement to AQMSs for examining policy options. Given the potential value of these methods, coupled with the fact that they are fairly new, the text in Section 111.7 of the document is far too brief to allow the reader to grasp how these methods can provide useful information. The discussion of air-quality models is excellent in many respects, and is candid about the shortcomings of these models. It is more comprehensive than those for most other topics addressed in the NARSTO assessment document, and it is written in a highly technical manner. Because these are the primary tools used to devise control strategies, a less-technical synopsis of the key points and many challenges could be presented for policy-oriented readers. The committee supports the need for openness in the modeling community, and much can be learned from the use of “Cautionary Tales.” The document should stress even more strongly the need to take a broader view of model verification. In addition to predictions of ozone, model performance should be compared with data for ozone precursors as well as other photochemically generated species—a point that has appeared in previous assessments (e.g., NRC 1991). It should also discuss the errors in emissions inventories as one of the limitations of AQMSs. This is not clear from the chapter, although it is stated in the Executive Summary of the document. The suggestion in Text Box IV.7 that the calculated change in ozone is much better defined than the calculation of ozone seems at odds with the discussion in Text Boxes IV.6 and IV.8. The document, in Section V, addresses the issue of accountability, using trends in atmospheric lead concentrations as an example. NRC (1991) similarly stressed the need for measurement testing to determine whether control strategies were, in fact, reducing emissions of VOCs and NOx. That objective could be met by using Los Angeles as an example for ozone-control strategies, because this airshed has had more ambient measurements, more inventory work, and more air-quality modeling than other regions of the United States. A quantitative investigation is needed of whether estimated emissions reductions in NOx and VOCs are consistent with observations, and whether the observed reduction in ozone in Los Angeles is consistent with expectation of air-quality models; information that is lacking could be noted. An effort of this sort is a high priority, and if not met in the revised assessment document, should be an early target of future NARSTO work. It is odd to read Text Box 111.2 on the question of what would have happened in Los Angeles without mobile-source emissions controls, absent a discussion of the consistency question, given the fact that there have been controls. It is this type of synthesis that is lacking in the document. It should also mention that the lack of a trend in 80% of the AIRS sites might reflect the fact that NOx emissions have been almost constant. SUGGESTIONS FOR STRENGTHENING THE NARSTO ASSESSMENT DOCUMENT To be fully successful, the NARSTO assessment document should be revised to achieve more-effective communication with its diverse audience. As noted above, the document is a scientific report, written by scientists and apparently addressed to scientists. It does not communicate a “synthesis and evaluation” of its scientific findings to policy-makers responsible for devising “regulatory initiatives and mid-course corrections.” This could be partially remedied if the document delineated the implications of its scientific review on current regulatory practice. For example, how the implications of long-range transport of the ozone affect the need for local,

OCR for page 12
REVIEW OF THE NARSTO DRAFT REPORT: AN ASSESSMENT OF TROPOSPHERIC OZONE POLLUTION–A NORTH AMERICAN PERSPECTIVE regional, national, and international actions should be spelled out. Pollutants other than ozone and fine particles should be considered. Another gap that the authors should address is the absence of adequate attention to the situation in Canada and Mexico. For example, explicit consideration should be given to the degrees to which findings are, and are not, applicable to situations existing in those countries. Also, the document requires an Executive Summary that is written for a nonscientist interested in policy implications. The NARSTO synthesis team should address the relatively small number of scientific issues raised in this report. In doing so, it should be mindful of the need to organize all of the scientific information presented so that it addresses and informs the policy issues facing environmental managers and regulators to the extent possible. One practical way to start this process is to take to heart the many policy-relevant questions raised by NARSTO reviewers of the initial draft. To the extent practicable, the additional analyses required to be responsive to these comments should be performed. In addition, previous integrated assessments of other problem domains, such as the National Acid Precipitation Assessment Program (NAPAP), should be examined for relevant insights.1 The NAPAP Oversight Review Board (1991) provides specific insights on how scientific information can be mobilized for policy purposes. The synthesis team should also revisit their decision to omit discussion of research priorities. Although they rightly conclude that it is not their task to set priorities, few are better positioned to inform the priority-setting process. The document should include a section addressing the needs for key future studies to fill knowledge gaps, which reflect the authors' explicit opinion on priorities, coupled with their reasoning in reaching them. The priorities should be tied to the decision-making process, that is, how will the new knowledge help make better decisions to improve air quality. It will be essential to have a list of key recommendations if NARSTO ever expects to begin the more difficult task of designating priorities for long-term research that will address outstanding questions. The document can initiate and contribute to this process. Without accepted, well-motivated, policy-relevant research objectives or clear criteria for setting them, there is little chance that NARSTO will be able to acquire either the political support or funding it needs to fulfill its goals. In Chapter 4 of this report, the committee suggests, for the NARSTO assessment document authors' consideration, unmet needs to be addressed by policy-relevant tropospheric ozone research in the hope of aiding NARSTO in this critical task. The NARSTO assessment document would be further strengthened by explicit consideration of how the implications of recent advances in scientific understanding could apply to new approaches to reducing ambient ozone concentrations. Although devising such approaches is beyond the document 's scope, evaluation of elements found in such approaches against scientific understanding would be helpful in placing alternatives on or off the table. It is important that the document show how NARSTO 's research objectives could contribute to the identification and design of new and better control strategies. For example, this could include a discussion of the measurement and monitoring needs for compliance assessment, as well as the need for baseline data sets. It could also identify the ambient data requirements to assess whether implemented emissions management strategies were actually having the predicted effects. A final suggestion for strengthening the body of the document is to enhance its treatment of the interconnection among air-quality issues. This is particularly important, given the fact that NARSTO recently agreed to widen its scope to address the formation and atmospheric distribution of fine particles. The document does briefly address the interconnection between ozone and fine particulate precursors and, as NARSTO evolves, this issue will undoubtedly 1 See, for example, NAPAP's 1990 Integrated Assessment Report (NAPAP 1991).

OCR for page 12
REVIEW OF THE NARSTO DRAFT REPORT: AN ASSESSMENT OF TROPOSPHERIC OZONE POLLUTION–A NORTH AMERICAN PERSPECTIVE receive additional, deeper analysis. The draft is largely silent on the interconnections between tropospheric ozone pollution and either climate change or acid deposition. Because the committee believes that these connections are both subtle and important, it suggests that they be addressed in a revised document. Turning to the Executive Summary of the NARSTO assessment document, the point has been made that, at a minimum, this portion of the document should be made more accessible to the policy-oriented target audience. Beyond this, other changes would be useful as discussed below. Most of the findings summarized in the Executive Summary of the document relate to the synthesis of scientific issues. There are very few findings that directly correspond to recommendations of any kind. Notable exceptions to this are the role of monitoring and the discussion of the impact of more-stringent ozone standards. Specifically, page x of the Executive Summary of the document includes a key finding that “Feasible changes and additions to network operations could improve national ozone monitoring systems.” Unlike most of the rest of the document, that section goes on to recommend what feasible improvements to network operation would make national monitoring networks more effective. The stated recommendations include a systematic network appraisal process, a real-time central archive of air-quality data, and an international “harmonization” of the three NARSTO nations' air-quality networks. The committee strongly supports the document in taking this type of stand and in offering key recommendations along with key findings. What is missing is an analysis of whether these are adequate to meet the needs of assuring compliance. Is there a need for new instrumentation? How well are such efforts linked to the air-quality models and their needs for verification? Another recommendation can be found on page xi of the Executive Summary of the document: “Promulgation of ozone standards and objectives based on a lower concentration averaged over a longer time-period will require expansion and improvement in the scientific and technical infrastructure intended to support the policymaking process.” The discussion immediately following that recommendation addresses the influence of new ozone standards and goes on to specify the likely effect of these standards. They would, for example, enhance the need to account for “background” ozone concentrations and require a greater emphasis on rural and regional ozone pollution. Although the recommendations in the Executive Summary are less than explicit recommendations, they come closer than most other sections in suggesting some action or direction. Dr. George Hidy (a NARSTO representative) posed a set of specific questions to the NRC committee related to possible improvements to the NARSTO assessment document. These questions, and the committees ' responses, are presented in Appendix E.