The committee finds that the draft assessment is a good draft with respect to the atmospheric-science topics discussed and the level of detail. However, it needs improvements to meet its goal of communicating well with decision-makers. Of greatest importance is the committee’s recommendation that a framework for understanding the complexities of atmospheric PM and its management be more clearly and consistently presented to provide the reader with a context for the assessment’s subject matter. The assessment also needs substantial improvements in presentation. It is long, repetitive, and in some places inconsistent. Many figures and tables are in poor condition. In the following sections, the committee provides overarching comments directed at the assessment’s contents and presentation.
The idea of a conceptual model, typically a simplified and general description of complex processes, is used in a number of ways throughout the PM assessment. The loose definition and multiple applications of the term create confusion for the reader. For example, Figure 10.11, “Elements of a conceptual model for PM adopted in this study,” is presented in Chapter 10 and in the response to policy question 3. The figure provides a highly simplified schematic of the factors that affect PM. Figure 6.8 in Chapter 6 shows another type of conceptual model, which is said to be useful for evaluating the effect of reducing sulfur in Canadian gasoline; it provides specific information linking sources and ambient concentrations. A third form of conceptual model is defined and presented in Chapter 10 for each of the regional case studies; the idea of a “conceptual description,” which is defined as an incomplete conceptual model, is introduced here. The distinction seems unnecessary, given that conceptual models are typically not expected
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 11
Review of the Narsto Draft Report: Narsto Assessment of the Atmospheric Science on Particulate Matter 2 OVERARCHING ISSUES The committee finds that the draft assessment is a good draft with respect to the atmospheric-science topics discussed and the level of detail. However, it needs improvements to meet its goal of communicating well with decision-makers. Of greatest importance is the committee’s recommendation that a framework for understanding the complexities of atmospheric PM and its management be more clearly and consistently presented to provide the reader with a context for the assessment’s subject matter. The assessment also needs substantial improvements in presentation. It is long, repetitive, and in some places inconsistent. Many figures and tables are in poor condition. In the following sections, the committee provides overarching comments directed at the assessment’s contents and presentation. PM CONCEPTUAL MODEL AND FRAMEWORK FOR INFORMING AIRBORNE-PM MANAGEMENT The idea of a conceptual model, typically a simplified and general description of complex processes, is used in a number of ways throughout the PM assessment. The loose definition and multiple applications of the term create confusion for the reader. For example, Figure 10.11, “Elements of a conceptual model for PM adopted in this study,” is presented in Chapter 10 and in the response to policy question 3. The figure provides a highly simplified schematic of the factors that affect PM. Figure 6.8 in Chapter 6 shows another type of conceptual model, which is said to be useful for evaluating the effect of reducing sulfur in Canadian gasoline; it provides specific information linking sources and ambient concentrations. A third form of conceptual model is defined and presented in Chapter 10 for each of the regional case studies; the idea of a “conceptual description,” which is defined as an incomplete conceptual model, is introduced here. The distinction seems unnecessary, given that conceptual models are typically not expected 1 References to figures and tables are to those in the draft NARSTO PM assessment unless specified as being “in the present report” or the like.
OCR for page 11
Review of the Narsto Draft Report: Narsto Assessment of the Atmospheric Science on Particulate Matter to be complete. Other definitions and examples of conceptual models are provided in Sections 1.8 and 9.2.3. Given the multiple uses of “conceptual model” in the assessment, the committee struggled to determine how NARSTO intended to use the term. Looking to the goals and specific objectives that NARSTO set out for the assessment (see Box 1–1 in Chapter 1 of the present report), the committee notes that the second objective is to “Provide a comprehensive conceptual model of aerosol formation and particulate matter distribution for science-policy analysts and air quality decision makers.” However, although conceptual models may be comprehensive in the categories they include, they are by definition not comprehensive in detail, so this goal would be rather difficult to reach. A single conceptual model explicitly linking sources, chemical and physical processes in the atmosphere, and the environmental distribution of PM would be useful, but it is not identified in the assessment. What is provided in the assessment is a set of regionally specific “conceptual descriptions” that are neither comprehensive in the regions evaluated nor complete in the processes described. A single general conceptual model for airborne PM would go a long way toward meeting NARSTO’s second objective as well as its third objective, to provide a plain-language conceptual description of PM air quality for the public. The committee recommends that a single general conceptual model for airborne-PM burden be explicitly introduced in the beginning of the assessment. The conceptual model could readily include emissions, atmospheric processing, and loss processes for airborne PM, all of which can be influenced by meteorologic factors and ultimately affect the distribution of PM in the atmosphere. To alleviate confusion, “conceptual model” should be clearly defined and consistently used throughout the document, and “conceptual descriptions,” “conceptual understanding,” and similar uses of “conceptual” should be eliminated. It should be made clear in the introduction of the conceptual model that it includes generalized concepts of major factors involved in the generation and distribution of PM. NARSTO’s second objective also calls for the model to “accommodate changing knowledge about atmospheric processes, emission sources, emission control technology, exposure, and human health and environmental impacts.” Similarly, the fifth objective calls for the assessment to “provide a framework for atmospheric scientists which relates their work to standards, implementation and air quality management, and to health, exposure, and environmental impact research for standard setting.” Thus, effectively connecting the conceptual model of airborne-PM burden to the impacts and policy implications of increased PM requires placing the conceptual model in a larger framework for informing airborne-PM management. Figure 1.6 in the draft assessment illustrates part of the interaction between science and policy, but it is not detailed enough to meet the objectives. The committee recommends that a clear framework for informing airborne-PM management, of which the conceptual model for airborne-PM burden is an important component, be explicitly introduced in Chapter 1 and near the beginning of the executive summary. The framework should be sufficiently fundamental for its components and their implications to be readily understood by lay readers (that is, it should be largely self-explanatory) and to serve as a basic paradigm for understanding variations that apply to any region over time. An example of such a framework is provided in Figure 2–1 of the present report, and the committee recommends that this framework or something similar of NARSTO’s choosing be used in the assessment. The
OCR for page 11
Review of the Narsto Draft Report: Narsto Assessment of the Atmospheric Science on Particulate Matter framework in Figure 2–1 of this report includes on the left side (first three columns) a single general conceptual model for airborne-PM burden; the right side includes a fourth column devoted to assessment activities that seek to understand the sources of observed PM concentrations, the impacts of PM, and the expected outcomes of various management strategies. The information gathered in the assessment column feeds into the management of PM, as shown in the fifth column. The committee recognizes that this proposed framework extends beyond the scope that NARSTO defined for this assessment—for example, the framework includes impacts on acid deposition, climate, and ecosystems—and is not suggesting that discussion of these topics be added to the current assessment. Nonetheless, providing a broader context by using a more complete framework would be useful for the intended audience of the assessment. In introducing the framework and discussing it throughout the assessment, the authors should emphasize that the framework, especially the conceptual model for airborne-PM burden, can be adapted for specific cases, such as those in Chapter 10 of the assessment, by obtaining appropriate information for each of the components. Variations of specific factors over time and space will cause the different components to be more or less important for a given case. Moreover, although example cases can be presented, the actual combinations of factors at play allow for essentially infinite variations. The committee views the establishment of a single general framework for informing airborne-PM management as critical to the success of the assessment, which depends on how readily the reader can assimilate the detailed information presented throughout the document into a single, consistent conceptual structure. Establishing a sound framework early in the document allows sections to refer to it to show the relevance of expanded information. The committee recommends that each technical chapter begin by positioning its subject matter in the framework for airborne-PM management. Likewise, the information on regionally specific processes could be presented using the same basic figure as the framework but with information in each box reflecting regional characteristics. Such a figure for each region that brings together all the factors affecting its PM problem may be more effective than the tables now presented in Chapter 10, where the information is grouped by factor (such as sources, meteorology, and policy implications). OTHER CONTENT ISSUES One of the stated charges to the team that prepared the draft assessment was to assess the “state of scientific understanding of the atmospheric aerosol as it relates directly to policy questions and program management associated with implementing any new PM standards.” However, the authors do not appear to have made a clear and consistent distinction between public policy and policy-relevant science, and sometimes they unnecessarily shy away from interpretive statements, perhaps out of concern that they may be misconstrued as policy recommendations. One way to characterize this crucial distinction is that public policy is the articulation of the hierarchy of preferred decision options, the range of instruments that can be used to implement options, and the rationales for choosing among the options, whereas policy-relevant science informs or constrains the nature of the options. The committee recognizes that NARSTO’s objective is not to promote specific policies but to summarize the policy-relevant
OCR for page 11
Review of the Narsto Draft Report: Narsto Assessment of the Atmospheric Science on Particulate Matter FIGURE 2–1 Framework for informing airborne-PM management. science. The assessment, however, often lacks a clear discussion of how the material presented could be useful for informing PM policy decisions. The committee recommends that each technical chapter (Chapters 2–9) conclude with a synthesis of how the scientific information presented could be used in managing PM. Such a “policy implications” section should follow the technical summary at the end of each chapter. Another issue is that the assessment focuses much less on airborne PM in Mexico than on that in Canada and the United States. The assessment includes only one Mexican location—the Mexico City area—in its regional descriptions, although information on PM in other areas of
OCR for page 11
Review of the Narsto Draft Report: Narsto Assessment of the Atmospheric Science on Particulate Matter Mexico is available. Likewise, only five of 45 interviews in the development of the eight policy questions were with Mexican decision-makers; this potentially compromises the usefulness of the questions and responses to Mexican policy-makers. Expanding the participation and representation of Mexico in the assessment would improve its applicability to PM management there and help to fulfill NARSTO’s goal of developing a North American perspective on the PM problem. Two NARSTO special journal issues have relevant studies in Mexico that can be referenced (Watson et al. 2001; Chow et al. 2002). In addition, several recently completed or current studies focus on key Mexican-U.S. border regions, and the data from and implications of these studies would be appropriate and pertinent; the studies include Mejia-Velazquez and Rodriguez-Gallegos (1997), Mukerjee (2001), Molina and Molina (2002), and those listed on the Website of the Southwest Center for Environmental Research and Policy (SCERP 2002). The need to understand the organic component of PM is a general theme that runs through the document. The committee agrees that a better understanding of this component of PM is appropriate, given that it is a large portion of ambient PM and that major uncertainties are associated with it. However, the authors need to motivate this focus better from the beginning of the report, perhaps through the use of a text box in Chapter 1, by summarizing why improving the characterization of organic carbon PM is critical for managing PM. In particular, the challenges posed by the large number of organic species and the present inability to resolve organic matter content completely should be explained. Another issue that needs more careful attention in the assessment is spatial variability in PM2.5 composition and concentrations. Epidemiologists have increasing evidence that health effects of PM2.5 and its components vary over spatial scales smaller than that resolved by current monitoring networks. The committee recognizes that the available monitoring network limits the extent to which smaller-scale variations in PM2.5 can be presented in the assessment. However, the draft could leave the impression that PM2.5 concentrations and composition tend to be rather homogeneous across regions in the size range of those highlighted in Chapter 10. Furthermore, the potential need for more fine-scale PM2.5 measurements is not noted. The committee recommends that discussion in the assessment reflect the heterogeneity in the composition and concentrations of PM2.5 within a region, the potential effect of this heterogeneity on variability in exposure and health effects, and the possible need to obtain additional measurements to support health-effects research. Finally, the extent to which the draft assessment represents the views of the North American atmospheric-science community is not clear. It is uncertain what sort of guidance was provided to chapter authors and whether reviewers with particular technical expertise were recruited to conduct scientific reviews of the chapters. In addition, it is unclear to what extent the research recommendations provided in Chapter 11 represent the views of the NARSTO community as a whole. Given the importance of those recommendations for defining research objectives, it is important for the authors to discuss how they were developed. PRESENTATION ISSUES The committee recommends that the assessment document be thoroughly edited to reduce repetition and ensure that terms and concepts are used consistently throughout. Numerous
OCR for page 11
Review of the Narsto Draft Report: Narsto Assessment of the Atmospheric Science on Particulate Matter examples of those problems are presented in this review’s comments on the draft’s chapters. In addition, although some chapters are well written, others would benefit greatly from copyediting to improve distracting grammatical and usage errors. Besides a thorough editing, the authors should also strongly consider using a text box near the beginning of each chapter to introduce and define important terms, abbreviations, and concepts. Such boxes, which are more immediately accessible than a glossary, should supplement definitions given in the glossary. In addition, consistent definitions should be provided in the text as terms are introduced. It is important that the definitions used at different points be the same and that the glossary be complete. The committee finds the presentation of major research recommendations in text boxes confusing. There is no discussion in the draft’s introduction alerting the reader that important research recommendations are summarized in this manner and that the numbering scheme used for the text boxes refers to the numbering used in Chapter 11. The specific text of the recommendations given in the chapters often does not match the text in Chapter 11. Many of the cross references to report sections used in Chapter 11 to support the recommendations are to nonexistent or irrelevant sections. The committee recommends that these problems be corrected. Using text boxes to highlight only research recommendations may give the reader the impression that the authors give more weight to these recommendations than to summarizing policy-relevant PM science. In fact, the authors could consider the potential value of using such boxes to highlight the policy recommendations presented in Chapter 10. In general, the figures and tables are in poor condition. Many have inadequate captions or titles, missing labels, and poor reproduction. In addition, because they typically have been excerpted from other published materials, they are often very technical and inadequately explained in the text. The committee recommends that each figure and table be examined to ensure that its caption or title is adequate, that all text and lines are visible, and that it is sufficiently supported in the text of the assessment. Specific figures and tables that are particularly poor in presentation or are not well integrated into the discussion are noted in Chapters 3 and 4 of this report. A consistent level of citations to the relevant technical literature should be adopted. The committee agrees that the PM assessment should not be an exhaustive review of prior work, but several chapters would benefit from additional references to key publications and reviews. Overall, the committee finds that inadequate attention has been given to editorial details and the presentation of the draft report. It was disappointing to note that many of the committee’s specific editorial comments had already been conveyed by chapter authors and members of the NARSTO community but were not addressed before the draft was provided to the committee.