-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON D.C. 20460
PBCfl*
OFFICE OF THE ADMINISTRATOR
SCIENCE ADVISORY BOARD
October 4,2004
EPA-SAB-CASAC-05-001
Honorable Michael O. Leavitt
Administrator
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW
Washington, DC 20460
Subject: Clean Air Scientific Advisory Committee (CASAC) Paniculate Matter (PM)
Review Panel's Ongoing Peer Review of the Agency's Fourth External Review
Draft of Air Quality Criteria for Paniculate Matter (August 2003)
Dear Administrator Leavitt:
EPA's Clean Air Scientific Advisory Committee (CASAC), supplemented by expert
consultants, collectively referred to as the CASAC Particulate Matter (PM) Review Panel
("Panel"), met via a public teleconference meeting on September 20,2004, to discuss matters
related to its ongoing peer review of the two-volume, June 2003 draft document, Fourth External
Review Draft EPA Air Quality Criteria for Particulate Matter (EPA/600/P-99/002, aD, bD). The
current Panel roster is found in Appendix A of this report.
This teleconference meeting was a continuation of the CASAC PM Review Panel's
review of the Fourth External Review Draft of the Air Quality Criteria Document (AQCD) for
PM in the current cycle for reviewing the National Ambient Air Quality Standards (NAAQS) for
PM. The purpose of this teleconference was for the Panel to review the revised Chapter 9
(Integrative Synthesis) of the AQCD for PM, which the Agency provided to the Panel on August
29, 2004.
After an extensive discussion, the Panel concluded that this revised chapter had been
sufficiently improved that it could close on Chapter 9, with the understanding that the Agency's
National Center for Environmental Assessment (NCEA)-RTP will make further revisions as
necessary to address the issues raised both in this report aridin the Panelists' individual review
comments, which are provided in Appendix B of this report^Tiis action completes the Panel's
review of the revised AQCD for PM.
. •.
.*.
-------
1. Background
The CASAC was established under section 109(d)(2) of the Clean Air Act (CAA or
"Act") (42 U.S.C. 7409) as an independent scientific advisory committee, in part to provide
advice, information and recommendations on the scientific and technical aspects of issues related
to air quality criteria and NAAQS under sections 108 and 109 of the Act. Section 109(d)(l) of
the CAA requires that EPA carry out a periodic review and revision, where appropriate, of the air
quality criteria and the NAAQS for "criteria" air pollutants such as PM. The CASAC, which is
administratively located under EPA's Science Advisory Board (SAB) Staff Office, is a Federal
advisory committee chartered under the Federal Advisory Committee Act (FACA), as amended,
5 U.S.C., App.
EPA is in the process of updating, .and revising where appropriate, the AQCD for PM as
issued in 1996. A detailed history of this current, ongoing review is contained in the Background
section of the CASAC PM Review Panel's report on this subject from the public meeting held in
Research Triangle Park (RTF), NC, on November 12-13, 2003 (EPA-SAB-CASAC-04-004,
dated February 18, 2004). The Panel's most recent report on this topic (EPA-SAB-CASAC-04-
008, dated August 16, 2004) was prepared following the public meeting held on July 20-21,
2004. Both of these documents can be found on the EPA Web Site at: http://www.epa.gov/sab.
2. CASAC PM Review Panel's Ongoing Review of the Revised Chapter 9 of the EPA Air
Quality Criteria for Paniculate Matter (Fourth External Review Draft)
The Panel commends the NCEA staff for the revisions that resulted in a shorter and
clearer synthesis of the information. This version of Chapter 9 was greatly improved over the
prior version. There are some remaining issues that will be described in this report and will need
to be addressed in the final set of revisions.
The description of the science with respect to PM2.5 and to a reasonable extent PMio is
well presented, but the information regarding PMio-2.5 is not yet as clearly articulated. There is
the potential for confusion because there are instances when the use of "PM" can be ambiguous
in terms of the size fraction being considered. As part of this discussion, it will be useful to
provide the rationale for the continuing use of PM2.5 rather than other possible cut-points for fine
particles.
There is relatively limited information on PMio-2.s, and those limits need to be made
clear. PMio is not an appropriate substitute for PMi0-2.s since it is possible that some of the
adverse health effects of PMio, most notably annual mortality, reflect only the PM2.5 effects and
that the coarse mode particles add little to health risk. This issue and the data addressing it need
to more clearly be brought out in the document. It would be useful to reduce the discussion of
and make a clearer distinction in terms of extent of information on the various fractions.
Another area of concern was the discussion of thresholds that is focused on PMio data yet
any standards would involve PMzs or PM 10-2.3 or both. The position taken by the Agency is that
"available studies do not provide strong evidence of a clear threshold for the relationship
between PM concentration and mortality" (p. 9-42) instead of the more accepted way of
characterizing dose-response relationships. A more appropriate wording of mis statement on
-------
page 9-42 would be: "The available evidence does not either support or refute the existence of
thresholds for the effects of ambient PM on mortality." Also, the discussion should recognize
that "exposure-response relationships" for one indicator, such as PMjo, may or may not be
applicable to other indicators. The treatment of thresholds is one of the most important drivers
of PM risk estimates. Thus, due attention to this matter is most appropriate.
As part of the discussion of Figure 9.4, a short description is needed of what is included
in this figure such as the single values for results from the National Morbidity, Mortality, and Air
Pollution Study (NMMAPS) and the Canadian Eight Cities Study, as well as appropriate
references to the fuller discussion of these results in Chapter 8.
The Panel felt that the modifications to the welfare section were well done. However, it
will be useful in future discussions of ecosystem effects for the Agency to clearly address the
cumulative nature of ecosystem responses so that new approaches can be considered to ensure
the protection of terrestrial and aquatic ecosystems
It would be useful to include a clear statement to the effect that there is a well-defined
linear relationship between PM^s concentration and observed light extinction. We suggest
wording such as provided in the comments of Dr. Warren White (see Appendix B, page B-39).
Finally, it would also be helpful if the 2001 to 2003 PM2.5 monitoring data could be
summarized in Section 9.1.2.
There are a number of editorial and minor scientific issues that are provided in the
individual panelists' comments. The Panel does not see any of these issues as major problems
and believe they can be addressed with relatively limited modifications to the existing draft. We
are pleased that we have been able to complete the review and achieved closure on the AQCD
for PM and look forward to working with the NCEA staff on the first external review draft of the
air quality criteria document for ozone and other photochemical oxidants in the near future. As
always, tiie CASAC PM Review Panel wishes the Agency well in this very important endeavor.
Sincerely,
/Signed/
Dr. Philip K. Hopke, Chair
Clean Air Scientific Advisory Committee
Appendix A - Roster of the CASAC Particulate Matter Review Panel
Appendix B - Review Comments from Individual CASAC Particulate Matter Review Panelists
-------
Appendix A - Roster of the CASAC Particulate Matter Review Panel
U.S. Environmental Protection Agency
Science Advisory Board (SAB) Staff Office
Clean Air Scientific Advisory Committee
CASAC Particulate Matter Review Panel*
CHAIR
Dr. Philip Hopke, Bayard D. Clarkson Distinguished Professor, Department of Chemical
Engineering, Clarkson University, Potsdam, NY
Also Member: SAB Board
CASAC MEMBERS
Dr. Ellis Cowling, University Distinguished Professor At-Large, North Carolina State
University, Colleges of Natural Resources and Agriculture and Life Sciences, North Carolina
State University, Raleigh, NC
Dr. James D. Crapo, Chairman, Department of Medicine, National Jewish Medical and
Research Center, Denver, CO, and Chief Executive Officer (CEO) of Aeolus Pharmaceuticals,
Inc.
Dr. Frederick J. Miller, Vice President for Research, CUT Centers for Health Research,
Research Triangle Park, NC
Mr. Richard L. Poirot, Environmental Analyst, Air Pollution Control Division, Department of
Environmental Conservation, Vermont Agency of Natural Resources, Waterbury, VT
Dr. Frank Speizer, Edward Kass Professor of Medicine, Charming Laboratory, Harvard
Medical School, Boston, MA
Dr. Barbara Zielinska, Research Professor, Division of Atmospheric Science, Desert Research
Institute, Reno, NV
CONSULTANTS
Dr. Jane Q. Koenig, Professor, Department of Environmental Health, School of Public Health
and Community Medicine, University of Washington, Seattle, WA
Dr. Petros Koutrakis, Professor of Environmental Science, Environmental Health, School of
Public Health, Harvard University (HSPH), Boston, MA
Dr. Allan Legge, President, Biosphere Solutions, Calgary, Alberta
A-l
-------
Dr. Paul J. Lioy, Associate Director and Professor, Environmental and Occupational Health
Sciences Institute, UMDNJ - Robert Wood Johnson Medical School, NJ
Dr. Morton Lippmann, Professor, Nelson Institute of Environmental Medicine, New York
University School of Medicine, Tuxedo, NY
Dr. Joe Mauderly, Vice President, Senior Scientist, and Director, National Environmental
Respiratory Center, Lovelace Respiratory Research Institute, Albuquerque, NM
Dr. Roger O. McClellan, Consultant, Albuquerque, NM
Dr. Gunter Oberdorster, Professor of Toxicology, Department of Environmental Medicine,
School of Medicine and Dentistry, University of Rochester, Rochester, NY
Dr. Robert D. Rowe, President, Stratus Consulting, Inc., Boulder, CO
Dr. Jonathan M. Samet, Professor and Chair, Department of Epidemiology, Bloomberg School
of Public Health, Johns Hopkins University, Baltimore, MD
Dr. Sverre Vedal, Professor of Medicine, National Jewish Medical and Research Center,
Denver, CO
Mr. Ronald White, Research Scientist, Epidemiology, Bloomberg School of Public Health,
Johns Hopkins University, Baltimore, MD
Dr. Warren H. White, Visiting Professor, Crocker Nuclear Laboratory, University of California
- Davis, Davis, CA
Dr. George T. Wolff, Principal Scientist, General Motors Corporation, Detroit, MI
SCIENCE ADVISORY BOARD STAFF
Mr. Fred Butterfield, CASAC Designated Federal Officer, 1200 Pennsylvania Avenue, N.W.,
Washington, DC, 20460, Phone: 202-343-9994, Fax: 202-233-0643 (Mterfieli.ted.@epa,^oy)
(Physical/Courier/FedEx Address: Fred A. Butterfield, III, EPA Science Advisory Board Staff
Office (Mail Code 1400F), Woodies Building, 1025 F Street, N.W., Room 3604, Washington,
DC 20004, Telephone: 202-343-9994)
* Members of this CASAC Panel consist of:
a. CASAC Members: Experts appointed to the statutory Clean Air Scientific Advisory Committee
by the EPA Administrator; and
b. CASAC Consultants: Experts appointed by the SAB Staff Director to serve on one of the
CASAC's National Ambient Air Quality Standards (NAAQS) Panels for a particular criteria air pollutant
A-2
-------
Appendix B - Review Comments from
Individual CASAC Participate Matter Review Panelists
This appendix contains the preliminary and/or final written review comments of
the individual members of the Clean Air Scientific Advisory Committee (CASAC)
Paniculate Matter (PM) Review Panel who submitted such comments electronically.
The comments are included here to provide both a full perspective and a range of
individual views expressed by Panel members during the review process. These
comments do not represent the views of the CASAC PM Review Panel, the CASAC, the
EPA Science Advisory Board, or the EPA itself. The consensus views of the CASAC
PM Review Panel and the CASAC are contained in the text of the report to which this
appendix is attached. Panelists providing comments are listed on the next page, and their
individual comments follow.
B-l
-------
Panelist
Dr. Ellis Cowling
Dr. James D. Crapo
Dr. Frederick J. Miller
Dr. Frank Speizer
Dr. Barbara Zielinska
Dr. Jane Q. Koenig
Dr. Petros Koutrakis
Dr. Allan Legge
Dr. Paul J. Lioy
Dr. Morton Lippmann
Dr. Joe Mauderly
Dr. Roger O. McClellan
Dr. Gtinter Oberdorster
Dr. Robert D. Rowe
Dr. Jonathan M. Samet
Dr. Svefre Vedal
Mr. Ronald H. White
Dr. Warren H. White
Dr. George T. Wolff
Paeeff
.B-3
.B-5
.B-7
.B-ll
..B-13
..B-14
.. B-15
.B-16
B-17
.B-18
.B-20
..B-23
B-33
.B-34
B-36
..B-37
...B-38
...B-39
B-40
B-2
-------
Dr. Ellis Cowling
Cowling Comments on Chapter 9 of the 2004 PM Criteria Document
This newly revised chapter is now very close to closure in ray opinion. The much improved organizational
structure and even more carefully formulated conclusionary statements bring this chapter very close to fulfilling the
promise of its title - Integrative Synthesis.
I hope the mostly editorial comments listed below for section 9.3 - dealing with PM-Related Welfare Effects
where I have some special competence — will help NCEA staff put the final touches on a generally very good redraft
of a very complex concluding chapter for this Criteria Document. This should be possible to achieve before the
mandated deadline of October 29,2004. This revised chapter should provide the integrative synthesis of scientific
background for equally careful policy analysis and interpretation that should serve the nation well over the next few
years.
I was particularly pleased that the Critical Loads concept, the desirability of a visibility-related Secondary
Standard for PM, and the Global Climate Change dimensions of the PM management problems are all dealt with
fairly and with reasonable thoroughness. The discussion of increased circulation of reactive forms of nitrogen and
its many and varied — sometimes beneficial and sometimes detrimental — effects on ecosystem health, productivity,
and stability - was very well done, although some editorial and other changes such as those suggested below will
make for further improvement.
1) Page 9-91 line 16 - John Sienfeld's name is misspelled.
2) As indicated in my earlier comments to Chairman Hopke on August 2,2004, it was disappointing to find no
tables or figures in Chapter 9 dealing with any of the four major PM Related Welfare Effects.
The discussion on page 9-92, lines 15-20 would be improved greatly if one of Molenar's "simulated photographs" of
regional haze were included on this page.
Also, the Nitrogen Cascade discussion on pages 9-98 and 9-97 would be greatly enhanced by including the original
figure of this Cascade as shown below:
Figure 6. The nitrogen cascade illustrates the movement of human-produced reactive nitrogen (Nr) as it cycles
through environmental reservoirs in the atmosphere, terrestrial ecosystems, and aquatic ecosystems (Galloway and
Cowling, 2002).
Indicate. denfcrWoition potential
B-3
-------
3) Page 9-92, line 30 - What is meant by "special areas."
4) Page 9-93, line 27 - Better to say "In addition, similar "acceptable" and "not acceptable"
threshold determinations " rather than "In addition, similar threshold determinations ."
5) Page 9-94, line 6 - Subscript 2.5; lines 8 and 16 - change "levels" to "concentrations";
line 13 - add "Phoenix, AZ, Lake Tahoe, CA, and State of Vermont" after "Denver, CO."
6) Page 9-95, line 2 - better to hyphenate "site-specific"; line 3 - change "different stressors in
PM deposition" to "different chemical constituents in deposited aerosols"; line 29 - add "sulfur
and phosphorus" after "nitrogen".
7) Page 9-96, line 2 - omit "being developed"; line 5 - change "component" to "specific
chemical constituent"; line 26 - change "of nitrogen via the nitrogen cycle controls" to "of
biologically active forms of nitrogen controls"; line 28 - after "roots," add "and by foliar
absorption through leaves,"; line 29 - change "(bacteria, blue-green algae)" to "(including
bacteria and free-living blue-green algae)"; lines 29-30 - omit last sentence.
8) Page 9-97, line 14 - after 'NH3" add "mainly for use as fertilizers"; line 27 - change "get" to
"are"; line 28 - add "are" after "and".
9) Page 9-98 - see suggestion 2 above.
10) Page 9-99, lines 8,13,17,19,20, and 21 - change "nitrogen" to "Nr"; line 9 - omit
"background levels (nitrogen loading)"; line 13 - add "accumulation" after "nitrogen".
11) Page 9-100, lines 2, 6, 8,10,11,14,16,19,20, 23,29, 31 - change "nitrogen" or "N" to
"Nr"; line 23 - add "amounts of after "increased".
12) Page 9-101, lines 1,2,10, 22,24, and 31 - change "nitrogen" to "Nr"
13) Page 9-102, line 2 - change "nitrogen" to "Nr"; lines 14 and 17 - change "acidic" to "acid";
line 12 - add "other" before "acidifying"; change "reduced" to "decreased".
14) Page 9-103, line 1 - change "acidic" to "acid"; lines 8 and 11 - change "nitrogen" to "Nr";
lines 9-10 - change "there is still insufficient data available" to "insufficient data are available".
15) Page 9-104, line 4 - change "(wet, dry, occult)" to "(wet, dry, and occult (cloud and fog
deposition)".
16) Page 9-105, line 9 - change "effects on radiation receipts" to "effects on incoming solar
radiation"; line 10 - add "still" before "poorly"; line 11 change "reducing" to "decreasing".
17) Page 9-107, line 12 - change "particle" to "PM"; line 13 - change "reduces" to "decreases";
change "reduce" to "decrease"; change "exposure level" to "exposures"; change "perceived " to
"observed".
B-4
-------
Dr. James D. Crapo
Review of Chapter 9 PM AQCD
General Comments:
Crapo
I have reviewed the revised Chapter 9. In general this chapter is thorough. It is accurate and
clearly written. I have concerns regarding two general areas which would require some
restructuring and small modifications of the document
1. Assessment of Health Effects of Fine vs. Coarse Mode Particulates
The document presents strong data showing that there are significant health effects
associated with PMz5 and with PMio. It contains numerous statements as to the limited
availability of data regarding PMio-2.5- The critical decision element here is whether or not
coarse mode particles need to be separately regulated. The coarse mode particles are
substantially different particulates with regard to their formation, characteristics, and
distribution. It is likely that they are substantially different in health effects and there appears to
be regional differences in these particles that may have direct impacts on health risk. A critical
question is whether or not PMio can be used as a surrogate for health effects of coarse particles, I
agree with the court ruling which suggests that regulation of coarse mode particles should be
based on data directly on coarse mode particles, not including the PM2.5 fraction as would occur
if one used PMio. It is possible that the adverse health effects of PMio reflect only the PM2.5
effects and that the coarse mode particles add little to health risk. This issue and the data
addressing it need to more clearly be brought out in the document. I would recommend that the
discussion of the PMjo data be reduced and focused more on the regional differences in PMio
health effects and whether or not under certain circumstances PMio can be used as a surrogate for
health effects of the coarse mode particulates. The concern is whether or not health effects of the
PMzs fraction dominate the results from PMio and thus PMio would not provide a rational basis
for separately regulating coarse mode particles. The discussion of the PMio-2.5 results should be
expanded to better address this issue and lay a foundation for a conclusion to either call for more
research or consider regulation of coarse mode particles.
Figures 9.4 and 9.5 would suggest that the health effects of PMio are the same as or
greater than that found with either PM2.s or PMio-is. The discussion suggests that the most
robust definition of adverse health effects is with PM2.5. On page 9-28, lines 17-19, it states that
if the effect estimates were presented per microgram/m3 for each PM mass measure, the effect
estimates for both PM2.s and PMio.2.5 would be larger than those for PMio. It seems to me that
this is the most important subject this document should address. I would recommend changing
Figures 9.4 and 9.5 to express the effect estimates per microgram/m3 for each PM mass measure.
Finally, the discussion of the cut point of 2.5 as noted on pages 9-19 through 9-21 ignores
the data presented at the last CASAC meeting showing that the obvious cut point between fine
and coarse mode particles is one micrometer. The document should contain a more accurate
description of the advantages and disadvantages of a 2.5 micron cut-off and a frank discussion of
B-5
-------
the advantage or disadvantage of a 1 micron cut point. The current document fails to even
address this issue.
2. Threshold and the use of a linear concentration-response model with no threshold.
I strongly disagree with the discussion of concentration-response relationships contained
in Section 9.2.2.5. I know of no chemical for which the biologic health effects show a linear
concentration no-threshold model when adequately and properly studied. It is extremely
unlikely that particulates truly exhibit a no-threshold model. The problem is in extrapolating to
zero from highly variable data at relatively high concentrations. The high variability of the data
helps create a good fit for a linear model. The current data does not suggest that there is no
threshold. In the absence of solid reproducible data at very low concentrations, we should not
conclude that there is no threshold. We should, instead, criticize the data for not providing
evidence of where that threshold is. We should not say that we do not provide strong evidence
of a clear threshold, rather, it should be stated that we do not have adequate studies to define the
threshold.
I believe the above issues can be easily addressed and provide a more scientifically rigorous
document that would better enable the next stage of decision making for setting air quality
standards.
B-6
-------
Dr. Frederick J. Miller
Chapter 9. Integrative Synthesis
General Comments
This version of Chapter 9 is significantly improved over the version reviewed at the My 2004
CAS AC meeting. The organization and flow of the chapter is good. With due attention to
comments offered by CASAC reviewers, I believe this chapter does not need to be seen again by
CASAC.
The authors have done a better job of summarizing current knowledge about the health effects of
PM2.5 than they have for PMio-2.5- The case for PMjo-2.s is presented more as "riding on the
coattails of PM in general" rather than an effective presentation of the state of our knowledge
about the potential for PM 10-2.5 to cause health effects or to exacerbate existing disease. Since the
courts have ruled that PMio reflects double jeopardy if one has separate standards for PM2.s, the
integrated synthesis chapter needs to do a better job of treating PMio-zs directly and to that
extent, a lessening of the space devoted to describing PMio effects would be warranted.
The other area still of concern to this reviewer has to do with the material concerning thresholds.
All of the discussion of thresholds is focused around PMio data yet any standards would involve
PMzs or PMjo-2.5 or bom. The position taken by the Agency is "available studies do not provide
strong evidence of a clear threshold for the relationship between PM concentration and
mortality" (p. 9-42) instead of the more accepted way of characterizing dose-response
relationships. If that were done, the statement on p. 9-42 would be "available studies do not
provide strong evidence mat there is not a threshold for the relationship between PM
concentration and mortality." Currently, Chapter 9 comes across as a clear bias by the Agency in
favor of no threshold models. Yet, for the curves in Figure 8-30, which form the basis of a
discussion of thresholds, 7 of the 9 curves are clearly nonlinear to the naked eye, as noted in
previous comments by this reviewer. How thresholds are or are not treated will be the single
most important driver of PM risk estimates. Thus, due attention to this matter is most
appropriate. In my specific comments below, threshold aspects are further commented upon.
Specific Comments
p. 9-2,1. 28 For the syntax of this sentence and grammar, the decreases should not be listed as
negative values.
p. 9-5,1. 3 It would be useful to make reference back to the chemical composition chapters
for additional information on the coarse mode composition.
p. 9-12,1. 2 The discussion of the inclusion of high wind blown dust events is an important
aspect that needs to be brought forward in the final interpretation of potential
PM2.s effects as how such days are treated can have an impact on the statistical
form of the standard.
B-7
-------
p. 9-15 Figure 9-2 is misleading. The x axis is labeled in intervals and yet the data are
connected by a series of line segments. To more accurately reflect the interval
nature of the data, a bar graph should be used.
p. 9-15,1. 9 It would be worth clarifying what is meant by non-ambient components of
personal exposure so that the reader understands these are indoor generated
materials and are not the ambient components that have penetrated into the house
as a function of the penetration factor.
p. 9-20,1.18 Why the distinction thoracic coarse particles as indexed by PMio-2.s? The particles
in the size range between 2.5 and 10 have significant deposition throughout the
respiratory tract and for many of the particle sizes have a maximum in deposition
in the head. The bottom line is that there is no value in using nomenclature to
imply thoracic coarse particles compared to non-thoracic coarse particles. This
nomenclature is a holdover from the use of PMio where the concern was for
particles capable of penetrating to the thorax, but the use of this modifier at this
point is questionable to this reviewer.
p. 9-27,1. 7 The statements about PM 10-2.5 confidence intervals relative to hospitalization for
cardiovascular effects is worded in a biased manner by indicating that some but
not all of the associations are statistically significant. A more accurate
representation of the data would be to say that very few of the associations are
statistically significant.
p. 9-28,1. 17 The conclusion or statement here seems counterintuitive because another way to
phrase the result is that the whole is less than the sum of its part in that PMio
effect estimates are indicated to be less than the combined effect estimates for
either of the two fractions comprising it. Do the data really support the statement
here? Or are we seeing a lack of precision due to decreased statistical power from
most of the studies that looked at a component of PMio compared to those studies
that only used PMio as the indicator variable?
p. 9-35,1. 13 The conclusion here, while correct, is somewhat misleading in that the statement
is a generality in terms of PM, as opposed to the various subdivisions of PM (i.e.,
PM2.5 or PMio-xs) that would have to be used in revising any existing standards or
proposing new ones.
p. 9-39,1. 30 Don't we know the direction of bias in this case for the single day lag with the
largest effect? If so, this should be indicated because currently the reader does not
know if the bias is to increase or decrease the estimate of the true effect.
p. 9-40,1. 25 The statement here is very important and should carry forward relative to the
averaging time for any particular indicator variable as it may relate to desired
protection against a specific adverse health outcome.
B-8
-------
p. 9-41,1.11 This paragraph comes across as a weak attempt to justify the assumption of a no
threshold model for PM effects. The basic tenet is that it would be difficult to
establish a population threshold due to individual differences in susceptibility or
other aspects and that it would be difficult to interpret an observed population
threshold biologically. One could just as easily approach this paragraph by saying
that it is difficult mathematically to demonstrate convincingly that a clear
threshold does not exist in population studies. This latter statement of the situation
is more in line with the concepts of sigmoidal curves for dose response whereby
extremely low dose effects may be approached with low dose linear
approximations, but in the case of PMio would be counter indicated because the
Agency is arguing there is no nonlinearity in the overall response curve, a position
mat is not supported by the curves contained in Figure 8-30. Also worth noting is
the fact that the Clear Air Act does not require protection of every single
individual, thereby acknowledging that individual thresholds do in fact exist, but
that sensitive subpopulations that can be identified need to be protected against
adverse health outcomes. It is difficult for this reviewer to see how the Agency
can argue both sides of the coin.
p. 9-54,1. 5 The aggregation of the smaller particles in the accumulation and Aitken modes is
a red herring relative to re-suspended vs. real world ambient human exposures.
This is comparing apples and oranges and needs to be reworded because the
surface area comparison that is alleged here has no basis in reality of making re-
suspended material equivalent to original ambient inhaled material.
p. 9-55,1. 2 Don't the authors mean combustion and non-combustion source particles?
p. 9-55,1.11 The authors should refer to this reviewer's comments relative to page 7-166,1.10
of Chapter 7 whereby it was previously commented on the caveats that are needed
for the particular comparisons being made. These comments were submitted as
part of the July review of Chapters 7 and 8 of the PM Criteria Document. Taking
into account those comments, this paragraph needs to be modified.
p. 9-74,1.1 Given the results of Figure 9-5 concerning hospital admissions for respiratory
disease for the PMio-zs fraction, this statement that coarse thoracic particles are
likely contributing to exacerbation of various respiratory conditions is worded too
strong. In the opinion of this reviewer, the case for PMio-2.5 effects is a weak one
given the limited data available and the nature of their results.
p. 9-81,1.20 The section on vulnerability appears to be a late addition as well as a late aspect
of interpretation for PM effects. The variables discussed in terms of
socioeconomic are typically accounted for in models that adjust for these as
potential concomitant variables and other aspects such as greater exposure near
roadways would be reflected in dose response considerations. Thus, the rationale
for this section is not immediately apparent. Workplace exposure or tobacco
smoking can be considered as vulnerability factors. The introduction to this
B-9
-------
section should clarify why only some vulnerability factors are being discussed
here.
p. 9-89 Section 9.3 on welfare effects still lacks an integrative punch. This reviewer
cannot tell the increment in knowledge, whether there is concern that secondary
standards should be changed because new evidence is available for effects at
lower levels, whether the 6-8 hour averaging time on visibility should carry
forward, and the list goes on. In short, as opposed to the health effects described
in the integrated synthesis chapter, the welfare effects are described in much
broader detail and without any clear implication for the relevance of new findings
in relationship to PM welfare effects. Thus, the imbalance in the bottom line for
health vs. welfare is a striking difference to this reviewer.
B-10
-------
Dr. Frank Speizer
Review of Chapter 9 submitted by Frank Speizer
I think this is a much improved chapter on which I am prepared to sign off on. I have taken the 5
basic questions outlined in section 9.2 and offered for each a brief commentary. Mostly the
suggestions can be ignored as I am sure they will all be taken up again in the Staff paper. Were
minor changes are suggested staff can consider them in the context that they might offer some
further clarity.
The Section 9.2 synthesis of PM-related health effects information is organized around five key
issues:
(1) Consideration of fine and coarse thoracic particles,as separate subclasses of PM pollution,
taking into account atmospheric science, exposure, and dosimetry information;
Most of the discussion of the particle subclasses is theoretical. The limited data that discuss
penetration and mode accumulation are important Although there are regional data what is not
pointed out is that there are relatively few data sets that can provide relationships between the
modes for broad regions of the US. It is not likely that the early chapters create such a catalogue.
This should be indicated as a potential weakness, particularly as related to setting national
standards.
(2) Assessment of strengths and limitations of the epidemiological evidence for associations
between health effects and fine and coarse thoracic PM within the mix of ambient air pollutants;
The presentation seems convincing that there is a consistent (enough) relationship for total
mortality, cardiovascular mortality and respiratory mortality related to PM10 and PM2.5 to
affirm the findings for 1996. The assessment of PM 10-25 is clearly hampered by the lack of
sufficient independent measures of the course fraction to assess the separation from PM2.5.
However, the studies that are presented do indeed show positive associations, although generally
with wider confidence intervals and generally overlapping the null. An issue that might be
discussed further is the appropriateness of using 25ug for PM2.5 and PM10-2.5 and 50ug for
PM10. Is there a more accurate delta that could be considered in the comparison? (I think this is
covered in chapter 3 and some words might be transferred here.
(3) Integration of epidemiological and experimental (e.g., dosimetric and toxicological) evidence
supporting judgments about the extent to which causal inferences can be made about observed
associations between health endpoints and various indicators or constituents of ambient PM,
acting alone and/or in combination with other pollutants.
Section 9.3.2.1 seems to be generically written for the most part and is a useful general
description of what the potential for how to interpret. It seems to me too long, but I would not
B-ll
-------
change it at this point. In general when the specific outcomes are presented there is a reasonable
correlation between the toxicological results and the epidemiological findings. This is
particularly true for cardiovascular endpoints where the 4 candidate mechanisms are discussed in
detail along with the appropriate animal and in vitro studies, with rather short term insults more
highly correlated. Respiratory symptoms and pulmonary function appear to be correlated with
toxicological findings of slightly longer durations of exposure than cardiovascular effects
(several hours vs. minutes). Probably need to indicate in greater detail that there are few if any
[toxicological] studies of PM10-2.5. Although there are additional toxicological studies that link
mutagenecity. with particles, since we really do not have a model for lung cancer it seems to me
that this needs to be admitted up front and then indicate that the mutagenecity is the closest we
can get. With regard to fetal development the epidemiology data are so weak and inconsistent
that it seems inappropriate to try to show consistency with toxicological studies. In fact the
section simply does not mention any [toxicological] studies, and my suggestion would be to
simply leave it out. This is acknowledged on page 9.74, but is mixed up with a susceptibility
question with regard to effects in children, which probably belongs in the next section.
4) Characterization of susceptible and vulnerable subpopulations potentially at
Increased risk for PM-related health effects;
Good general summary but need to acknowledge that there are insufficient data to say much
about particle size and specific susceptibilities.
(5) Discussion of potential public health impacts (including newly emerging evidence for
adverse cardiovascular effects) of human exposures to ambient PM in the United States.
At the end of Ihe first paragraph on 9.88 there is a statement that may be true but needs better
documentation with more quantification, if that is possible. This will be an important issue to
take up in the Staff paper. It is probably ok to leave it as it is here.
Chapt9.cretique0916/04
B-12
-------
Dr. Barbara Zielinska
Comments on revised Chapters 9 of the PM AQCD September 16,2004
Barbara Zielinska
In my opinion, this revised version of Chapter 9 is significantly improved over the previous
version. The addition of Section 9.1.2 (Trends in US PM Air Quality) provides a proper context
for the subsequent discussion of exposures and health effects. I have only a few minor
comments:
1. Page 9-4, line 7: analytical limitations are due not only to the polar nature of some of the
organic compounds, but also due to the presence of oligomeric and/or polymeric
substances in ambient fine PM (i.e. biopolymers, mimic-like substances).
2. Page 9-10,1. 23-25. The sentence "Fine particles include metals, black or elemental
carbon, organic carbon (primary PM) and sulfate, nitrate, ammonium and hydrogen ions,
and organic compounds (secondary PM)" doesn't make sense. Organic compounds
constitute organic carbon and they could be both primary and secondary. The
composition of fine PM was discussed in Section 9.1.2, so it is redundant here (and not
correct).
3. Page 9-10,1. 27. The composition of coarse PM was also discussed in Section 9.1.2 on
page 9-5, line 3-9. To say that the composition of coarse PM includes only primary
minerals and organic materials is not correct. It looks like these sections were written by
two different persons.
4. Page 9-16, Table 9-2 and p. 9-20,1.10: The infiltration factor for accumulation mode
particles depends on ventilation conditions of a building.
5, Page 9-46,1. 12-13: I'm not sure what authors are trying to say in this sentence: "PM
exist as a complex air pollution mixture that includes other pollutants" Which other
pollutants?
6. Page 9-49,1. 20-22: A prolonged storage of filters in inappropriate conditions leads not
only to volatilization of semi-volatile material, but also to chemical alterations of reactive
compounds and possible growth of mold, bacterial contamination, etc.
7, Page 9-57,1. 5: Secondary components of submicron aerosol include also organic
species, not only inorganic.
8. I support Warren White suggestion to substitute two paragraphs on page 9-90 - 9-91.
B-13
-------
Dr. Jane Q. Koenig
Comments on revised Ch 9
9-18-04
Koenig
It is gratifying to see that the document has been shortened although certainly not by
60%.
I agree with most of the comments that the other reviewers have presented. I do want to
second the point that James Crapb raised about PM1.0. We spent considerable time discussing
this in July and I do believe the CD needs to include a summary of that discussion.
I support Warren's additions re visibility.
I think the discussion on indicator, averaging time, numerical level and statistical form
belong in the Staff paper. Although I certainly agree with Roger that this chapter cant be labeled
a summary!! (114 pages!!).
I am in favor of including PM10-2.5 data (as few as they are). I do not believe the health
studies allow us to ignore the coarse fraction.
I apologize for being so late with my remarks. Time flies!
B-14
-------
Dr. Petros Koutrakis
Chapter 9
Integrative Synthesis
Reviewer: Petros Koutrakis
The Integrative Synthesis chapter is now more concise and reads well. I am satisfied with the
quality of this chapter. Below please find a few minor comments:
9.1.2-Paee 8. Line 21:
I am surprised mat the CD relies mostly on the Lipfert paper. I was expecting that there is more
information on this important issue.
9.1.2-Page 9. Line 19:
As I said in my previous review, this issue has been a little bit exaggerated. Also this is not the
place to comment on particle toxicity.
Table 9-2; Page 21:
Table 9-2 gives a nice comparison of the three modes. My only comment is that the infiltration
factors can vary with home ventilation. So they are not always high or low etc.
9.2.2.3-Page 44. Line 5:
Climatic conditions should also be included, since they affect the home infiltration rates, air
condition use etc.
9.2.3.1.1 - Page 54. Line 8:
CAPs studies have not used overly high exposures.
9.2.3.1.1 - Page 55. Line 2:
I guess you are talking here about the fine particle concentrators because the ultrafine can
concentrate ultrafine particles.
9.2.3.2.2. - Page 69. Line 29:
To my knowledge there are one or two studies showing ultrafine effects. I think the word "a few"
is very generous here.
9.2.5.2. - Page 87. Line 28:
Fix typo, PM10-2.5
B-15
-------
Dr. Allan Legge
PM CD, August/2004, Revised Chapter 9
Allan Legge
September 17, 2004
Review Comments:
The revisions to Chapter 9 ('Integrative Synthesis') have been well done and better
reflect a balance in the available scientific information. Section 9.3.2 Effects of Ambient PM on
Vegetation and Ecosystems, has drawn together the key pieces of our current knowledge and
understanding. It is time to reach closure. That being said, it is hoped that the 'Staff Paper' picks
up on the cumulative nature of ecosystem responses and the need for a new approach to ensure
the protection of terrestrial and aquatic ecosystems.
The following are two specific comments:
1. Page 9-95, line 11.
This should read: "These effects were usually the result ."
2. Paged 9-102, Section 9.3.2.2.2 Acidification from PM Deposition, para. 2.
The important role of Ca in ecosystems is noted. The specific reference to Bondietti and
McLaughlin (1992) on line 24 relates to Ca in wood. This is fine as per the text. An additional
and more current reference with respect to Ca in ecosystems is found in McLaughlin and
Wimmer (1999).
McLaughlin, S.B. and Wimmer, R. 1999.
Calcium physiology and terrestrial ecosystem processes. Tansley Review No.
104.
New Phytologist 142: 373-417.
B-16
-------
Dr. Paul J. Lioy
Integrative Synthesis
Chapter 9 - PM Criteria Document
Fourth External Review draft
Comments of Dr. Paul J. Lioy
General
The Agency has made a very good effort in addressing my concerns, and those presented
by others. At this point in time we should close on the Chapter, and approve the Criteria
Document. I have a few suggestions for improvement.
1. Page 9.4 There are values presented on abundances for some of the trace elements. To be
; consistent, the agency needs to consider adding some information, e.g. > or < X ug/m3 for
the compounds or compound classes mentioned on lines 11 through 18.
2. Page 9.43, Line 14, The paragraph correctly states that there is a link between short term
exposure, hours to days, and cardiovascular and respiratory mortality and morbidity.
However, on pages 9.23 etc, the statements about short term effects < day are vague, if
non-existent. Please introduce a sentence or two to support this important observation.
The current version of Chapter 9 does have a statement related to peak exposures on page
9-39 in subsection 9.2.2.4 - Temporality and the Question of Lags. This should be
supported by a more specific statement on or about 9.23.
3. General: Exposure - response relationships for cardiovascular and other effects over
periods that are on the order of hours have been observed in research studies. These
results require the development of sampling protocols that will collect and analyze
continuous PM data at critical urban locations in the US. Thus, a statement to mis end is
required somewhere in the chapter. Clearly, there will be more studies designed to reduce
uncertainties about the relationship between short term exposures and cardiovascular
disease. Consequently, a national'data base should be developed for PM found in the
ambient air of US cities to establish the frequency and PM concentration ranges for
periods of 2 to 8 hours. Consideration should be given primarily to continuous
measurements of PM2.5.
September 16, 2004
B-17
-------
Dr. Morton Lippmann
REVIEW COMMENTS
M. Lippmann
Chapter 9 of PM AQCD
August 2004 Draft
General Comments
This revised draft has recognized and successfully addressed the significant concerns of
the Panel members at the last review session. In particular, the text on pp. 9-8 and 9-9 and on pp.
9-20 and 9-21 provides the technical and policy bases for choosing PMio and PM2.5 cut sizes for
PM NAAQS that was lacking in the previous draft. Also, the text on p. 9-10 now addresses the
importance of particle bound water (PBW), and Figure 9-1 on p. 9-11 provides a much better
summary representation of the typical volume modes of the ambient aerosol and their size
distributions.
The paragraph on p. 9-9 (lines 16-22) on the study opportunities created by routine PMzs
monitoring for improved understanding of the associations between ambient air PM and human
health should be amplified by a forward looking statement such as: "We can anticipate that the
Speciation Site network and the increasing availability of continuous monitors for PM^s will
provide further opportunities to identify PM components and their sources most responsible for
PM health effects, and for elucidating the role of peak exposures within a day that may cause
acute responses."
The issue of identifying a population threshold from observational studies, which is
addressed on p. 9-41 remains a concern to scientists who deal primarily with toxicological study
data. It may be advisable to refer, in this section on concentration-response relationships, to
explicitly refer to the following section on natural experiment studies, which generally indicate
B-18
-------
that significant reductions in ambient pollution are associated with improvements in health .
status. Also, this section should introduce the Kunzli et al. (2001) concept of frailty, which is
cited in Section 9.2.5.2 (on p. 9-88) as a factor that could account for the lack of a threshold in
population studies.
Specific Comments
Line(s)
11
10-12
30-31
6
18
9-9
9-32
9-34
9-58
9-71
Comment
insert "the health risks of* before "thoracic"
These statements improperly ignore the 10 year
prospective study of children in Southern California.
insert "to" before "have"
insert "mortality" before "data"
B-19
-------
Dr. Joe Mauderly
Review of Chapter 9 of PM CD
Mauderly
General Comments
Overall, this is a reasonable integrative synthesis. I think that with minor editing, it can be
accepted. I like the approach that avoids citing lots of references. I'd still like a last section in
which the key advances since the last CD, and those considered most critical to the standard, are
listed clearly in bulletized form.
Specific Comments
P 9-1, L 21: "Remaining uncertainties that remain" is redundant.
P 9-2, L 25: Not clear whether this is geographical national average or average population
exposure.
P 9-3, L 8: Same question as above.
P 9-3, L20-21: The point about different monitoring methods needs explaining - a short, single-
sentence expansion would suffice.
P 9-8, L 12-13: This sentence doesn't make sense. Of course particle size is the basis for
distinguishing between fine and coarse collection, but I don't think that's what you mean.
P 9-10, L 19: "PBW" is already defined.
P 9-12, L 17: Use the definition, "PBW".
P 9-30, L 17: It should be "the magnitudes of the associations are -".
P 9-39, L 12-15: This sentence isn't a useful summary for the section. The reader perceives that
you are trying to make some distinction between "numerous locations" and "various places",
which, I don't think, is your point. Your point either isn't clear or this is a very unremarkable
summary sentence. It doesn't say anything.
P 9-47, L 26: It should be "—area are summarized -".
P 9-49, L 14-22: Probably should also mention here the possibility for biological growth on the
samples during uncontrolled storage (bacteria, molds).
P 9-49, L 28: It should be "-—CAPs varies across—".
B-20
-------
P 9-53, L 30: Delete "exercising" - the sentence as written is redundant.
P 9-54, L 7-10: This isn't the important point. The more important point is that you can't do the
simulation even with high doses. Yes, you can get a "dose" of PM surface into the lung, but it
wouldn't bear any resemblance to the surface encountered by humans, or the relationship of
surface to mass. It is very misleading to suggest that jacking up a dose of larger PM to simulate
die surface dose of smaller PM is a legitimate comparison. The notion itself is corrupt.
P 9-55, L 2: The sentence isn't clear. Combustion source PM is part of ambient PM. What's
the point you are trying to make?
P 9-55, L 11-20: Unless the exposure materials were shown to be identical (and I don't recall
that they were), this is not a legitimate comparison. It suggests that all CAPs is alike, which it
isn't. Including this comparison is misleading.
P 9-55, L 24: Have you defined "MO"?
P 9-56, L 3: It should be "Comparisons in Chapter 7 of rodent -".
P 9-56, L 28-29: This sentence is so full of jargon as to be obscure. How is the reader supposed
to know what "metastable species" are, or what "dead" components are? If there is an important
thought here, put it in English.
P 9-62, L 6: Two animal studies, one human experimental exposure, and one epidemiology
study have shown Factor VII to be decreased in relation to exposure to PM. What is your
evidence that it should be increased, other than our incorrect preconceived notions of a few years
ago?
P 9-62, L 15-16: Several locations are cited below. What does "within a given location" mean?
P 9-63, L 28-30: What in the world could "metal-associated ROFA" mean? Presumably you
mean ROFA containing high concentrations of metals. You have already defined the
abbreviations given in line 30.
P 9-64, L10-11: The sentence implied that fibrinogen was increased at 69 u,g/m3. I doubt that
was the case. We tried to correct sentences like this in earlier chapters.
P 9-65, L 23: I think you have a stray hyphen here.
P 9-68, L 18: "Follow-up" is usually hyphenated.
P 9-70, L 18-19: As I recall, the 1996 AQCD did not state that bioaerosols were unlikely to
"contribute to" effects. I think the statement was that they were unlikely to "account for" PM
effects. That's two different things. In fact, that language carried over into the first draft of this
CD, and was corrected. The main point is that now, it is considered likely that biomaterials do
contribute to PM effects.
B-21
-------
P 9-70, L 27-28: We need to focus on different components. That doesn't necessarily mean
from different regions. "Regions" is the flyer here - the focus should be on components. The
fact that different regions have different average compositions is true, but not the relevant point
here. It would be unreasonable to suggest that the key way to get at the importance of different
components would be to study different regions. Region is a very crude surrogate for
composition.
P 9-72, L 10: Tumorigenic" is misspelled.
P 9-77, L 21-24: Two currently-used models ought to be included in this list: experimental
coronary artery blockage, and apo E atherosclerosis-prone mice.
P 9-78, L 2-3: Two "roles" are redundant.
B-22
-------
Dr. Roger O. McClelLan
Roger O. McClellan, DVM, DABT, DABVT
Advisor, Toxicology and Human Health Risk Analysis
13701 Quaking Aspen Place N.E.
Albuquerque, NM 87111-7168
Tel: 505-296-7083
E-mail: roeer.o.mccleiian@att.net
September 22, 2004
' Review Comments on Revised Chapter 9; Integrative Summary (August 2004)
The August 2004 draft of Chapter 9, Integrative Summary of the Paniculate Matter
Criteria Document, is improved over the earlier version of the Chapter. However, in my
professional judgment, the Chapter still does not reach the standard I would like to see the
U.S.E.P.A. reach in summarizing the current scientific information that will inform the
Administrator's decisions on the four elements of a National Ambient Air Quality Standard for
Paniculate Matter; (1) indicator, (2) averaging time, (3) numerical level, and (4) statistical form.
Nonetheless, recognizing the-court-ordered time constraints faced by the Agency, I reluctantly
concur in a decision for CASAC to close on the Chapter and move on to consideration of the
Staff Paper. As reflected in my comments below, in my professional judgment, the Integrative
Summary over-states the certainty of the scientific evidence relevant to re-affirming or revising a
PM2.5 standard and very substantially over-states the certainty of the scientific evidence to be
considered in making a decision on the potential promulgation of a PMio-2.5 standard.
I offer the following comments for consideration by the EPA staff:
!»
General Comments
s
1. The present draft is not an adequate "Integrative Summary." It is excessively
long and, thus, is clearly not a summary. In some places it comes close to integrating the
available data but the integration is obscured by excessive detail. I suspect a hard-nosed science
editor could readily reduce the length of the Chapter by a factor of two, most of the reduction
could occur by reducing the discussion on pgs 9-46 to 9-73 which has limited bearing on the
setting of aNAAQS for any PM indicator.
B-23
-------
2. The CD, including this Chapter, is still handicapped by the artificial division
between the science covered in the 1996 CD and the more recent evidence. The NAAQS for any
PM indicator must be based on all the available science irrespective of when it was developed
and published. The document, and Chapter, would be easier to read and understand if it focused
on all the important science. The present CD, including this summary, labors in many places to
include an excessive amount of new details. Most of the new details presented on pages 9-46 to
9-73 will not inform the setting of a standard and could be readily summarized in a table and a
few pages of text.
3. The "Integrative Summary" is not a balanced presentation of the science that will
inform the setting of the NAAQS for PM. The tone in many places attempts to justify past
actions in the setting of the PM2.s standard. In other places, it strives to provide a basis for
promulgating a standard based on a PMl0-2.s indicator. This chapter does not always summarize
critical uncertainties in key data analyses and interpretations that were ultimately added to earlier
chapters. The resulting tone in this chapter reflects a higher degree of certainty in evidence than
is warranted.
4. I was pleased to see the inclusion of a section (9.1.2) on "Trends United States
PM Air Quality." I suggest that the time period covered be extended by both considering
information pre-1992 (see the ACS study and Harvard Six Cities study for some early time
period monitoring data) and for 2001-2003.
5. I was pleased to see the section (9.2.5.1) on health statistics although I would not
have titled it as "Magnitude of Susceptible Groups." A title such as "Baseline Health Statistics"
would be more appropriate. I urge that this specific section be moved to the Introduction as
Section 9.1.3 right after the "Trends in United States PM Air Quality." Tables 9.4 and 9.5,
which consider morbidity indices, should be complemented by a Table on mortality, i.e. causes
of death. It is important to recall that data on increased mortality from cardiovascular and
respiratory disease associated with increased levels of ambient PM 9see Figure 9.4) plays a
major role in decisions on the NAAQS for PM. In discussing Tables 9.4 and 9.5 and the new
table on causes of death, a new paragraph should be added indicating the dominant role of
cigarette smoking in causing cardiovascular and respiratory disease. I was reminded of this
again today when I noted the local freeway billboard showing 318,000 deaths in 2004 through
September 20th attributed to cigarette smoking.
B-24
-------
6. Section 9.2.1 (Fine and Coarse Particles as Separate Sub-classes of PM Pollution)
should be rewritten with a stronger and more accurate historical orientation. The revision should
more clearly indicate the direct linkage between what has been monitored and the conduct of
subsequent epidemiological investigations. In short, epidemiological studies can only be
conducted using the indicators that have been monitored. Over time these measurements have
progressed from metrics such as smog days to black smoke to total suspended particulate (TSP)
matter to PMio to PM^s to PMjo-2.5 and particle number. PMi.o and PMis were only monitored
for a brief period of time in the 1980s. This section needs to be forthright in acknowledging that
EPA's regulatory compliance monitoring program (TSP -> PMio -> PMu) has dictated the kind
of data available for conduct of epidemiological studies. The lack of monitoring data on PMi.o
and PM 10-2.5 and the related lack of epidemiological data for these indicators is a direct result of
past EPA actions. One might hypothesize that PMi.o and PMio-i.o would be the better indicators
for fine and coarse particles man PMaj and PMio-is- However, this will remain a hypothesis
until monitoring data and related epidemiological results on these indicators are available and the
findings compared to epidemiological evaluations of PM2.s and PMio-2.5 at the same locations.
7. In many places the document uses what I term a "PM blanket" approach,
analogous to the blanket Limis in the cartoons, or my kids, carried around. In short, when the
evidence on a particular indicator does not exist or is uncertain the authors attempt to make up
for it by relating evidence from another PM indicator or resort to describing PM evidence in
general. This approach is inappropriate. The evidence for each indicator needs to be considered
independently. It is not appropriate to attempt to make up for a lack of evidence for the PMio-zs
indicator by using evidence on PM^s.
8. The Chapter would benefit from a concluding section on the key science that will
inform decisions on the four elements, as noted above, of a NAAQS for PM. Beyond
consideration of the indicators as noted above, the Chapter would benefit from a more thorough
presentation of the evidence on the relationship between ambient concentrations (PMjo, PMio-2.s
and PMzs) and increased adverse health effect outcomes at levels typically found in the U.S.
today. The failure of the "Integrative Summary" to adequately address this critical issue
abdicates consideration of this matter to the Staff Paper. This is a science matter that should
have been more adequately covered in the CD and especially this Chapter. The issue of ambient
concentration-response to relationships is a matter that goes well beyond the cursory coverage of
B-25
-------
"thresholds" in the CD and this Chapter. The CD, and this Chapter, also fails to address the
scientific information that under-girds consideration of a statistical form for the NAAQS for PM.
Specific Comments
Pg 9-1, line 25: Insert a paragraph noting the science reviewed in this chapter is intended to
inform decisions on the four key elements of a National Ambient Air Quality Standard
(NAAQS): (1) indicator, (2) averaging time, (3) numerical level, and (4) statistical form.
In this last chapter of the CD which, up to this point, is largely an exposition on the "world
of PM and its health and welfare effects" needs to acknowledge that the CD has a specific
purpose-inform decision on the four elements of the NAAQS for PM.
Pg 9-8, line 23 - pg 9-9, line 4: This long sentence conveys the view that a conscious decision
was made to select PMj.s fim as a size cut in the last review over selection of some other
size metric. This is ridiculous. The decision had been made much earlier when EPA, in
my view, inappropriately selected PM2.5 Jim as the cut-point in the design of new
monitoring instrumentation. If a science-based approach had been taken, monitoring data
would have been collected on both PMi.o um and PM2.5 um cut-points and epidemiological
studies conducted on both indicators. This would have provided a basis for the CD
considering both PMi.o and PM2.s um cut-points. EPA should not attempt to revise history
and ignore its traditional "regulatory compliance" orientation to collecting monitoring data.
Unfortunately, the monitoring data available for conducting epidemiological studies is that
collected under the "regulatory lamp post."
Pg 9-11, line 4: 1996 PM AQCD, not 1966
Pg 9-12, line 19: This sentence may not be technically correct as I noted in my comments in the
last draft of Chapter 9. Yes, formaldehyde and other volatile compounds may become
associated with particles. However, this particle associated fraction may not be as
biologically significant as the associated gas phase concentrations of the chemicals. By
and large this is an important hypothesis that should guide future research. However, the
available data are so limited that it is of limited value in setting a NAAQS for PM.
Moreover, the hypothesis does not warrant discussion more than once in this summary
chapter.
Pg 9-15, line 8 on: The discussion needs to emphasize the substantial geographic differences in
"infiltration factors" as influenced by temperature, humidity and local ventilation practices.
B-26
-------
What is true for Boston is certainly not true quantitatively for other cities, for example,
Birmingham, AL and San Diego, CA.
Pg 9-16, line 20: Insert "calculated" as in "The calculated fractional depositions — " Later it
should be noted that detailed actual measured deposition values are available primarily
over the size range of 0.5 to 5 um.
Pg 9-19, line 16: The sentence will confuse many readers. The earlier discussion (Table 9.1)
emphasizes that the fine mode consists of an ultrafine and accumulation mode. The "story
line" needs to be consistent.
Pg 9-20, line 5-9: This sentence is confusing and adds nothing. The behavior of ultrafine and
coarse particles is fundamentally different. The strained linkage here is inappropriate and
creates confusion.
Pg 9-20, line 20 on: This paragraph needs to be rewritten to more clearly convey that what is
important is the "integrated" deposition of all particles within a given mode (i.e. ultrafine,
fine and coarse), not whether there are differences in the deposition of particles of a
specific size, such as 0.1, 0.5,1.0,2,0, 3.0, 5.0 or 10.0 um. The tone of the text suggests
that mathematical modelers are losing touch with biological reality.
Pg 9-23, line 4-6: This very important sentence should be rewritten - "the impact of ambient air
pollution exposure on these outcomes maybe is very small in comparison to that of other
risk factors (e.g. smoking and diet)." I find it curious that the EPA staff can be so
conclusive elsewhere about PM effects but here when it draws a comparison with the well-
established major risk factor, smoking, it uses the word may, apparently to heighten
concern for PM effects and reduce concern for smoking effects.
Pg 9-26, line 7; Figure 9-4 and the related discussion here is MISLEADING. The data from
Dominici et al (2003a), the National Mortality and Morbidity Air Pollution Study on 90
(88?) cities, and the Burnett et al (2003) study of 8 Canadian cities are presented as single
estimates in contrast to all the other single city estimates. The result is to mislead the
reader as to the consistency of the observations. The fact of the matter is that "Some of the
associations between PMioond total mortality are positive and some are statistically
significant. However, many associations are not statistically significant especially when
cities are studied using the same methodology." This needs to be acknowledged in the
text and also reinforced by including a figure illustrating the NMMAPS results for multiple
B-27
-------
cities. In my opinion, the NMMAPS results represent a critical finding that should inform
decisions on the NAAQS for PM. It is not scientifically appropriate to focus on the
composite estimate to the exclusion of the individual city estimates. If for policy or
ideological reasons the decision is to ignore the individual city estimates, EPA needs to
clearly articulate the rationale for the approach.
Pg 9-26, line 8: The Chapter would convey a more scientifically credible tone if it would simply
described the evidence for a given size fraction. This serves as a good example in which
the discussion of PMzs starts by relating it to PMio- The approach taken is one that again
tries to wrap every size fraction in the "PM blanket" The authors should describe the
evidence and avoid attempting to lead the reader.
Pg 9-26, line 9: The same is true for PMio-2.s, although here the "PM blanket" is even more moth
eaten. Leave out the first sentence. The evidence is — "There are a few effect estimates for
PMto.2.5, most are positive and similar in magnitude to those reported for PM^.s and PMio,
but few reach statistical significance."
Pg 9-26, line 11-14: This represents another example of "bundling" or using the "PM blanket."
In earlier sections, apparently written by a different author, the document emphasizes the
differences between the various indicators. Now the authors reverse course and elect to
bundle PMio and PM2.5 together. THE DATA FOR EACH INDICATOR, PMio, PM2.5
AND PMio-2.5, NEEDS TO BE DESCRIBED SEPARATELY. It is obvious that PM2.5 and
PMio-2.5 are a part of PMio. However, if we did not have some evidence of differences it
would be appropriate to just use a single indicator, PMio.
Pg 9-27, line 8: It would be more accurate to end - "and most of the associations are not
statistically significant." .
Pg 9-28, line 10-13: I submit that this statement concerning PMio effects in the western United
States is an unproven hypothesis. It should be stated as such or much more substantial
evidence should be presented for the western United States. It is important to recall the
diverse nature of the western United States ranging from Los Angeles to Houston to Dallas
to Albuquerque to Phoenix to Seattle to Salt Lake City. This is a blatant example of how
the CD tries to selectively relate science to apparently meet some pre-conceived regulatory
notations. In this case, the apparent attempt is to provide the basis for a PMio-2.s standard
B-28
-------
based on alleged associations with mortality and morbidity. In this case, the science does
not exist.
Pg 9-30, line 15-30: In this summary paragraph the authors again attempt to wrap PMi0-2.5 in the
"PM blanket." It would be more scientifically accurate to say "there is limited scientific
evidence for an association between PM 10-2.5 ambient concentrations and mortality,"
Pg 9-31, line 25-31: It is unfortunate that the recent paper from the Southern California school
children study cannot be cited in the CD. My read of it is that it identifies an association
between elevated air pollution (using multiple pollutant indices) and retarded lung growth.
However, the authors were careful to not focus on PM to the exclusion of other pollutants.
Pg 9-32 to 9-42: This entire section needs to be rewritten and each topic addressed on an
individual indicator basis. The present style of writing in which the authors sometimes
describe the evidence for PM and at other times for PMio, PMzs or PMio-2.s is misleading.
The nature of the evidence for each PM indicator is different and, in some cases, there is
every reason to hypothesize that the nature of the evidence for each indicator will be
different. An example is concentration-response relationships.
Pg 9-33, line 27 (Confounding): The section on confounding is written with a bias toward
emphasizing a PM effect and downplaying the effects of other pollutants.
Pg 9-36, line 19: The use of phrases like - "would be expected" serve as a tip-off that the
authors are ready to offer a conclusion that would be better stated as a "hypothesis."
Ultimately, standards are to be set based on scientific conclusions, not scientific
hypotheses.
Pg 9-36, line 23 on: The next 3 pages provide a discourse on consistency and variability. In the
interest of scientific completeness it would be appropriate to note that some studies that fail
to show an association between various PM indices and health outcomes may in fact reflect
an absence of a PM effect. Throughout the document, studies that do not show a positive
association of a health outcome with a PM indicator or fail to be statistically significant are
described as though the effect was there; it just has not yet been found.
Pg 9-41, "Concentration-Response Relationships": This section, which should summarize
information that will be key in setting the "numerical level" of any NAAQS for a PM
indicator, is totally inadequate and misleading. This section needs to describe the evidence
related to the title of the section. It is not adequate to just discuss the "threshold concept"
B-29
-------
The discussion needs to acknowledge the limitations in the statistical tools and the data for
describing concentration-response relationships at ambient concentrations of PMio, PMxs
or PMio-2.5 as experienced in most areas of the United States today. This would include
linear versus non-linear relationships and not just thresholds.
Pg 9-42, line 4-10: This concluding paragraph is "one-sided." It should acknowledge the
difficulty in evaluating ambient concentration-health response relationships for a "weak"
risk factor, such as various PM indicators, when ambient concentrations are low. I would
argue that if statistically significant associations are not found, as in the case of many cities
in NMMAPS that stands as evidence of a "practical threshold." As a scientist I do not
understand the unwillingness of EPA to acknowledge that there are areas in the United
States today where there may be an absence of a PM effect or, indeed, on air pollution
effect.
Pg 9-42, line 14-24: Reverse the flow of the paragraph to correspond with the strength of the
evidence starting with PMio, then PM^s and concluding with PMio-2.5-
pg 9-43, line 8-13: Eliminate the paragraph.
Pg 9-44, line 7: Why not offer a separate statement on PMio-2.5 and let go of the "PM security
blanket" For example, - "There is only very weak and uncertain evidence for an
association between PM 10-2.5 and mortality/morbidity outcomes."
Pg 9-44, line 13-15: Another blatant attempt to try to rationalize a lack of evidence for PMio-2.5
effects. This kind of "pleading" should be avoided in the CD. The CD should lay out the
science as it exists, not attempt to re-interpret it to apparently support a regulatory agenda.
Pg 9-45, line 4: To use the word "causal" to modify the "associations" is an over-statement. The
epidemiological evidence supports an association with a "weak signal" for PMio and
related to baseline mortality and morbidity.
Pg 9-45, line 4-11: Describe the evidence for each indicator. DO NOT attempt to cloak
with PMio evidence or PMio-2.5 with PMio or PM^s evidence. Remember these are
different indicators and it is not surprising that the associations between excess risk and
ambient concentrations will be different.
Pg 9-45, line 11-12: The one-liner is a bold statement. Give a better description of the nature of
the evidence for the association between excess health risk and increased ambient
B-30
-------
concentrations of each indicator (PMi0) PM2.5, PM 10-2.5). DO NOT use the "PM blanket" to
dodge this critical issue.
Pg 9-46 to 9-73: This section can be substantially shortened. I recognize that it reports the
results of an enormous amount of research. However, calling a "spade a spade" this section
only offers modest support for the epidemiological associations between increased ambient
concentrations of various PM indicators and an increase in health outcomes. Most of the
information could be summarized in one or two tables and a few paragraphs. One
paragraph could describe the challenges and experimental limitations of linking the "weak"
stochastic signals found in some epidemiological studies with experimental approaches
dependent upon relating "deterministic" measures of effects. It could conclude with a
statement that the experimental evidence lends support to a causal association between
exposure to high levels of some PM indicators and an increase in adverse health outcomes.
The nature of the evidence is so fragmentary that it is not always apparent as to how it
relates the various indicators. A revision of the section needs to clearly describe the nature
of the integrated evidence for the individual indicators, PMio, PM^s and PMio-zs and
especially the latter two indicators since they are likely to be the focus of the NAAQS-PM
standard setting. In the scientific communities zeal to focus on PM^s little attention has
been directed to obtaining experimental evidence on the PM 10-15 fraction. This lack of
experimental evidence on PM 10-2.5 needs to be acknowledged in Chapter 9.
Pg 9-54, end of page: It would be appropriate to add - "Moreover, the extensive modeling
results presented in Appendix 7A provide a number of hypotheses for experimental testing.
There will be increased confidence in the models and results when additional experimental
data are obtained and the models subjected to rigorous validation with new data other than
that used to develop the models."
Pg 9-71, line 18-21: If the 2002 extended ACS analysis is cited as evidence of an association
between exposure to fine particles and lung cancer at least a brief mention should be made
of the lack of long-term smoking histories in the ACS study. As I recall the smoking
history of individuals in the cohort were only ascertained at the beginning of the study.
Differential shifts in the various sub-populations such as those from different socio-
economic levels could have influenced the outcome.
B-31
-------
Pg 9-84, Susceptible Groups: I find it interesting that the authors cannot state the obvious -
"The largest susceptible groups for PM associated effects are cigarette smokers and
former smokers. The vast majority of the respiratory and cardiovascular endpoints
summarized in Tables 9.4 and 9.5 reflect effects observed in cigarette smokers and former
. smokers." I suggest including a paragraph on cigarette smoking trends in the U.S.
B-32
-------
Dr. Gunter Oberdorster
To: -Drs. Butterfield/Hopke
Review of Chapter 9 of PM Air Quality Criteria Document (G. Oberdorster)
In my view, EPA has. responded well to the comments of CASAC. I think the Chapter,
although still long, represents a good approach for an integrative summary and, with the
additional suggestions made during Monday's conference call, it is ready for closure. Some
additional editorial/factual comments/suggestions are as follows:
Page 16, Table 9:2: The infiltration factor for ultrafme particles is described here as
"generally low". However, since it is actually high between 60 and 100 nm (see Figure 9-2), I
suggest to state "low to high (size dependent)".
. Page 20, line 2: I suggest to add "mass" to the word "concentrations" since this
statement does not apply to number concentrations.
Page 48, lines 24 through 26: A sentence should be added that such models of
compromised animals should be relevant to the human condition as much as possible with
respect to the pathophysiology of a specific condition. The first example of a compromised
animal model given here, i.e., monocrotaline-treated rats, is in my view not a good example for
the human condition. Thus, the emphasis on relevancy would be important to point out here.
Page 50, line 9: Although CAPs concentrators can include gaseous co-pollutants, gases
will not be concentrated like the particles, so CAPs exposures are still different from the real
world.
Page 52, line 14: Emphasize that for in vitro studies always several doses should be
included and that a "real-world" dose should always be included in such studies.
Page 55, lower paragraph, and page 56, upper paragraph, inhibition of phagocytosis:
Overwhelming alveolar macrophages by the amount of PM is very unlikely for ambient particles,
this is an overload issue. Also, there will be a compensatory AM influx rather than a decrease in
AM numbers. I am also not sure that the differences in inhibition of phagocytosis between
humans and hamsters on the one hand and mouse on the other are real, given that primary cells
are compared here with a cell line. Also, the smaller size of mouse macrophages vs. those of
humans should not be of significance for ambient PM exposures, since we are not dealing with
overload situations here. Thus, I am not sure that the alleged differences in inhibition of
phagocytosis pointed out here are real; at best one could say that there may be differences
between species.
Page 80, line 25: Replace "homogeneity" with "regional efficiency".
B-33
-------
Dr. Robert D. Rowe
September 10,2004
Robert Rowe comments on August, 2004 PM CD Chapter 9.
Visibility section
> Page 9-93, lines 9-10. .I'm not comfortable with the sentence as is -1 don't think it makes
clear the core issue with the studies is about the exact dollar values (which varies with
variations in the method(s) and is not unusual in science). Mavbe something more like
"Results vary across studies and uncertainties remain about the specific dollar values
estimated." The studies can also provide information about whether people believe different
levels of visibility impairment are sufficiently adverse to warrant doing something or paying
something, which the existing sentence indirectly but inappropriate puts in question as well.
> Page 9-93 lines 13,14,26; page 9-94 line 19. 'Values" is used with the perception/adversity
judgment studies in several places. Broadly, the responses are "value statements." However, I
worry that here "value" will be confused with monetary valuation. I think phrasing it as
"'perceptions and attitudes about visibility impairment,' including what is adverse' (as on
page 9-94 lines 10-11) is more precise as to what theses studies provide and that what they
provide ties nicely to the concept of adversity in the CAA.
> As noted in the Committee letter, "it would be helpful to indicate the approximate PMb.5
mass concentrations that correspond to the various visual range values that are discussed" in
the perception/adversity judgment studies.
Ecosystems section. Page 9-95, line 11. Syntax needs editing "These effects were the usually ."
Climate section. Page 9-105, line 2. "impairment" would be slightly preferred to "degradation"
for consistency with terminology in the visibility section.
Materials.
> This section relies heavily on 1996 AQCD, which is OK. But, there needs to be a sentence as
to why - which is that there has been limited applicable research since that time that
fundamentally changes the conclusions, or some such statement to justify not saying a thing
about work since 1996 (for example, see climate section page 9-105 line 11). The last
sentence (page 9-107 lines 17-20) alludes to the same concept without being explicit.
> Page 107, lines 15-16. This sentence misses several endpoints, as identified in the committee
letter. "Available data indicate that airborne particles can result in increases in the frequency
of cleaning, maintenance, or replacement of exposed materials surfaces, as well as reduced
usefulness and enjoyment of injured soiled materials." (Its not just surfaces, its not just
soiling, its not just cleaning).
> Page 107, lines 17-20. Its not just perceptions ("perception thresholds of damages and
soiling") but both perception and functionality, or welfare impact. For example, upon close
B-34
-------
inspection a material may be clearly altered but may have no function or welfare impact
requiring cleaning, maintenance or replacement
B-35
-------
Dr. Jonathan M. Samet
September 17,2004
To: PhilHopke
From: Jon Samet
Subject: PM AQCD Chapter 9
Phil, I have read the revised chapter and in general, I find it to be satisfactory and ready
for closure. Staff have responded to my concerns on the prior draft, particularly those related to
the interpretation of the epidemiological evidence. The chapter is more concise and even
readable. I have a few specific comments as follows:
* In discussing PM^s and PMio-2.s, the chapter addresses issues relate4d to composition and
dosimetry and covers the scant epidemiological data. The chapter should integrate the
information on site of deposition and composition to infer whether different adverse
health effects would be anticipated for the two size fractions. This section could be a bit
stronger on mechanisms and health outcomes.
• The discussion on pages 9-19 and 9-20 has several confusing elements. How does
consideration of the ultrafme mode add to understanding of PM^s health effects? The
comparison of ultrafine and coarse particles on page 9-20 is not helpful and should be
deleted.
* At the bottom of page 9-26, having invoked exposure error and other issues, what are the
net consequences for interpretation of the evidence? A sweeping concern is raised
without follow-up.
B-36
-------
Dr. Sverre Vedal
PM CD (chapter 9)
Sverre Vedal
September 20,2004
1. General.
This revised draft has dealt with nearly every one of the specific issues I had raised in
July about the previous version. I will not detail these here. Any concerns that I have with the
chapter, at this point, are largely due to differences in opinion as to how the evidence should be
interpreted rather than concerns about fair representation of the findings or about factual errors.
Overall, I find the current organization of the chapter to be acceptable, even though I considered
the earlier organization around 5 questions to also have been a reasonable approach. The current
organization of topics in fact corresponds to the earlier one, but with different topic headings.
2. Consistency of findings.
This discussion of consistency has been expanded and improved, although the main point
now seems to be one of explaining why inconsistency is to be expected, rather than maintaining
that consistency is present. It is argued, based on figures 9-4 (p.9-24) and 9-5 (p.9-25), that
studies with more power generally show more consistently significant effects. While this is
probably true, I would point out mat for the important outcome of cardiovascular mortality
(figure 9-4), studies with less power have generally larger and significant effect estimates. I
would drop the argument relating power and statistical significance based on findings reported in
the 1996 CD (p.9-38, L16-18). This reliance on the 1996 CD suggests to me, perhaps unfairly,
that findings based on studies reported subsequently do not make the same point.
On a small point, chapter 8 of the CD appropriately qualifies the lack of statistical
evidence for heterogeneity of effects in NMMAPS by noting the limited statistical power of the
test for heterogeneity in this setting; no such qualifying of this conclusion is included here (p.9-
37, L7 & 21).
3. Coherence.
The chapter now focuses the discussion of coherence as pertaining to that between the
observational findings and the experimental and lexicological findings, which I believe is
appropriate. Arguing strongly for coherence based on types observational outcomes is fraught
with pitfalls and is probably best avoided at this time.
4. Cohort studies.
I continue to maintain that the relatively strong effect of S02 in the ACS cohort study
should not merely be dismissed as being difficult to interpret due to the fact that SO2 is a
precursor of sulfate (p.9-35, L3-4). This finding in the ACS study should have prompted a more
thoughtful discussion of residual confounding in the cohort studies, particularly in light of the
questionable consistency of findings in the cohort studies as a whole.
B-37
-------
Mr. Ronald H. White
Comments of Ronald White, M. S.T.
Revised Draft Paniculate Matter Criteria Document - Chapter 9: Integrated Synthesis
September 20,2004
General Comments
This revision of Chapter 9 of the fourth draft Particulate Matter Criteria Document (PM CD) has
adequately responded to all of the major issues and editorial changes requested by the CAS AC
PM panel at its My 2004 meeting. The discussion of particle size, composition and respiratory
tract deposition is improved from the previous version, as is the discussion and use of the
concepts of consistency and coherence in interpreting the results of health effects studies. The
document has been considerably condensed and extraneous discussion and references to specific
studies that are not essential to the major points of the document have generally been eliminated.
The integration of animal toxicology and controlled human study results.with the
epidemiological evidence in the discussion of coherence of results and potential biological
mechanisms in Section 9.2.3 finally achieves the key objective for an integrated synthesis
chapter.
The elimination of Tables 9A1-3 should be reconsidered, however, as they provided important
summary information from the key U.S. and Canadian studies on the mortality and morbidity
effects estimates associated with daily and long-term PM exposures. The inclusion of
information on the mean and range of PM concentrations measured in each study is especially
useful in providing a context for the levels of PM associated with the effects estimates. While I
appreciate the need to limit the size of Chapter 9 so that it is a readable synthesis of the scientific
information, dropping these valuable tables is a case of "penny-wise, pound-foolish".
Alternatively, this information could be integrated into the summary tables in Appendices 8A
and 8B, though with the result of adding complexity to already dense tables.
With the presumption that only relatively minor editorial changes which are likely to be
suggested by the CAS AC PM panel as part of the current review will be satisfactorily addressed
by EPA in the final version of Chapter 9,1 recommend closure for this chapter.
Specific Comments
Pg. 9-26, Line 6: As there is frequent reference in this and subsequent sections to "more precise"
study results, an operational definition for the criteria used (i.e. confidence interval size) to
determine which study results are considered "more precise" than others should be included.
Pg. 9-28, Line 17-19: This is an important point that bears further discussion in the conclusions
section of the review of health effects science.
Pg. 9-40, Lines 3-4: To support the concept that distributed lag models provide a more complete
effects estimate than single-day lags, reference should be made to key studies such as Schwartz
(2000) and Braga et al. (2001) that have examined various single-day as well as distributed lags.
B-38
-------
Dr. Warren H. White
PM AQCD Chapter 9, rev. 8/04 Comments by Warren H. White, 9/12/04
Visibility
I urge something like the following in place of the existing text from line 15, page 9-90 to line 2,
page 9-91:
Airborne particles degrade visibility by scattering and absorbing light. A particle's optical
impact is described by its effective cross section for these interactions, which in general is a function
of its size and composition. The local intensity of haze is quantified by the concentration of these
cross sections in the air (area per volume), called the extinction coefficient (dimension inverse length,
or fractional attenuation per distance). The cross sections of distinct particles add, so the extinction
coefficient produced by a given distribution of particle sizes and compositions is strictly proportional
to the particle mass concentration.
Because the extinction cross section associated with a given mass is much higher for fine
particles than for coarse, it is fine-particle mass concentrations that tend to drive extinction
coefficients in polluted air. The main source of variation in observed extinction per measured PMzs
is the loss of particle-bound water from ambient particles when they are dried for weighing, hi arid
regions such as the Southwest, where this effect is minimized, observed extmction/PM2.5 ratios can
vary within surprisingly narrow ranges, typically remaining near 2Vi +/- V4 m2/g. In more humid
areas such as the East, observed cross sections per measured mass are generally higher.
The point is that visibility doesn't just "depend1' on the mass concentration - it is linear with it,
and that is a much stronger statement. (Consider how much blood has been spilled over the
linearity question for health effects and acid deposition!) Moreover, it is not the "efficiency" that
depends on particle mass "as modulated by ", but the extinction itself. The extinction
efficiency, whether referenced to mass, volume, or geometric cross-section, is essentially
independent of particle mass concentration, and that is why it is a useful concept.
As noted in the Committee letter, "it would be helpful to indicate the approximate PM2.s mass
concentrations that correspond to the various visual range values that are discussed" in the
perception/adversity judgment studies. Using the nominal value of 2Vi +/- Vz m2/g given above
for the Southwest, it could be noted that the Denver standard corresponds to short-term PM2.5
concentrations of no more than 20-30 ug/m3.
The distinction of "fine" and "coarse"
The introductory sentence beginning on line 12 of page 9-8 would be clearer as:
The evidence available in the last review strongly focused on particle size as the basis for
distinguishing two essentially different classes of particles.
I would urge also that "naturar and "a fundamental distinction" be dropped from the
introductory sentence beginning on line 1 of 9-8.
B-39
-------
Dr. George T. Wolff
Comments on Chapter 9
By
George T. Wolff
(9/20/04)
While I think the chapter has improved, it still reflects the Agency's biases for overstating the
evidence for PM effects and downplaying the uncertainties.
The document does contain the relevant science except for an objective discussion of the latest
information on models specification. I acknowledge the fact that a conscious decision was made
at the July meeting to exclude parts of this issue from the discussion. However, as we move on
to the Staff Paper, I do not see how we can avoid this discussion.
Starting on page 9-22,1 think it is misleading to state that many recent studies "built upon what
was previously known" without saying that many, if not most, of the new studies cannot be
considered because of the GAM fiasco.
On page 9-23, the top line should be re-worded to say the GAM-related has led to smaller effects
in "most" rather than "some" cases.
On the same page, in the bottom paragraph, it should be pointed out that the single-pollutant
analysis tends to overestimate the effects because it does not consider the contributions from
other factors.
Figure 9-4 presents a distorted view of the risk estimates by hiding the heterogeneity that is
present in the NMMAPS results. As a result, the accompanying discussion on 9-26 to 9-30 does
not reflect the uncertainty that exist in the data because of the heterogeneity issue.
It should be pointed out that the studies cited on the bottom of page 9-28 and 9-29 that implicate
traffic are based mostly on studies that used Pb as a tracer and may have no relevance to current
vehicle emissions.
I still disagree with the Agency's interpretation of the long-term exposure studies and their
decision to largely ignore the negative results in the VA and AHSMOG studies and the
inexplicable results in the ASC and 6 Cities studies. As I pointed out in my previous comments,
there are more individuals in the four long-term studies that exhibit no PM effects than those that
experience an effect.
Model specification discussion on pages 9-32 and 9-33 - The components of this issue are
mentioned, but not put into context. This section has no conclusions. It does not convey the
seriousness of this issue.
B-40
-------
P 9-43, line 30 to p 9-44, lines 1 and 2 -1 do not agree with this sentence. If it remains, it needs
a reference. As far as I am aware, the HEI commentary on the revised studies is the most recent
statement on this subject and it is in conflict with this sentence.
B-41
-------
NOTICE
This report has been written as part of the activities of the Environmental
Protection Agency's (EPA) Clean Air Scientific Advisory Committee (CASAC), a
Federal advisory committee administratively located under the EPA Science Advisory
Board Staff that is chartered to provide extramural scientific information and advice to
the Administrator and other officials of the EPA. The CAS AC is structured to provide
balanced, expert assessment of scientific matters related to issue and problems facing the
Agency. This report has not been reviewed for approval by the Agency and, hence, the
contents of this report do not necessarily represent the views and policies of the EPA, nor
of other agencies in the Executive Branch of the Federal government, nor does mention
of trade names or commercial products constitute a recommendation for use. CAS AC
reports are posted on the SAB Web site at: http://www.epa.gov/sab.
B-42
------- |