US ENVIRONMENTAL PROTECTION AGENCY

SCIENCE ADVISORY BOARD (SAB) STAFF OFFICE

CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE (CASAC)

OXIDES OF NITROGEN PRIMARY NAAQS REVIEW PANEL PUBLIC MEETING

MARRIOTT AT RESEARCH TRIANGLE PARK

4700 Guardian Drive

Durham, North Carolina 27703

OCTOBER 25, 2007

8:36 A.M.

US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 2

Page 2

1	U.S. ENVIRONMENTAL PROTECTION AGENCY

2	CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE

3	PUBLIC MEETING

4	OCTOBER 25, 2007

5	DR. HENDERSON: I want to thank everyone

6  for being so timely in submitted your paragraphs to

7  Angela, and I thank Angela for bringing it all

8  together.  Now that what's being distributed - -

9	SPEAKER: Hello?

10	DR. HENDERSON: Hello.

11	SPEAKER: Rogene, can you, you need to

12  speak into the microphone.

13	DR. HENDERSON: Okay.

14	SPEAKER: Thank you.

15	DR. HENDERSON: What's being passed out

16  is a compilation of what was submitted, and these are

17  all, everything, it's truly a compilation, but I've

18  read it through it, and I compared it with the list,

19  this small list is, these are the points we listed

20  yesterday afternoon that we thought should be included.

21  So, you might, quickly, compare this list with what's

22  in, what you'd submitted to see if we left anything

23  out.

24	But the consideration that we're going to be

25  making is, is this the substance, does this include

Page 4

1	DR. NUGENT: Good morning.

2	DR. LARSON: Tim Larson.

3	DR. NUGENT: Good morning.

4	DR. ULTMAN: Jim Ultman is.

5	DR. NUGENT: Jim?

6	DR. HENDERSON: Jim Ultman.

7	DR. NUGENT: Ultman.  And, Lee Anne, are

8  you on the phone?  Okay, all right, and just a note

9  about public comments.  I'd mentioned yesterday that we

10  were inviting public comments on yesterday's discussion

11  relating to the ISA.  No member of the public has asked

12  me to speak this morning about the ISA.  I'll ask one

13  more time, because we want to be a little structured

14  about how the discussion proceeds.  Are there members

15  of the public who'd like to present some comments?

16	MR. HICE: Angela?

17	DR. NUGENT: Yes?

18	MR. HICE: This is John Hice on the

19  phone.

20	DR. NUGENT: Yes?

21	MR. HICE: I'd like to make a very, very

22  short comment, if I could.

23	DR. NUGENT: Thank you, okay.  I'll write

24  that down, and we'll turn to you in a moment.

25	MR. HICE: Thank you.



Page 3

1  everything you want to say to the Administrator in our

2  letter.  And, it's not the exact words, because Angela

3  and I will have to go through and make it sound like it

4  was written by one person instead of a committee.

5	But, that is, it's not, it's going to be

6  smoothed out, but does it contain the substance of what

7  we want to say?  Can we agree?  Do you feel comfortable

8  with what is written here as a compilation of

9  everything that we want to say to the Administrator in

10  terms of our peer review of the first draft of the ISA?

11  And, while you're reading that, I think with, I'm going

12  to, I have neglected to let Angela do a roll call of

13  who's on the phone.  So, I will turn it back to her

14  while you're reading it.

15	DR. NUGENT: Thank you, Rogene.  As we

16  start this second day, and we complete the discussion

17  of the ISA, and then move ahead to the discussion of

18  the methods document, I wanted to welcome the people on

19  the phone, and make sure everyone in the room knows

20  who's on the phone, and then talk a little bit about

21  the public comment period, here.  So, may I ask,

22  please, what CASAC panel members are on the phone

23  today?  Are the CASAC panel members on the phone right

24  now?

Page 5

1	DR. NUGENT: Let me also mention that we

2  didn't explicitly list on the agenda public comments

3  for the methods document.  And, I would like to know

4  whether, once we complete this ISA discussion, there

5  are members of the public who'd like to present some

6  brief comments on the methods document.  Okay, hearing

7  none, I think we should proceed.  Rogene, John Hice has

8  some remarks, and I propose that we take them now.

9	DR. HENDERSON: Now would be a great

10  time.

11	DR. NUGENT: Thanks.  I think your

12  audio's working well for us here, so please, speak into

13  your phone set, and we'd love, we'd like to hear your

14  comments now, please.

15	MR. HICE: Thank you very much.  I just

16  wanted to reiterate that we'll be providing written

17  comments for the record to EPA by that deadline, it was

18  October 31st.  And I'm sure several other groups will

19  also.  And I would just ask that the CASAC folks take a

20  look through those comments, at their convenience, and

21  add those thoughts to their own as they think about the

22  review of the next draft.  That's all.

23	DR. NUGENT: Thank you.

24	DR. HENDERSON: And I thank you, too.  I,

25  now, can people hear me if I hold the mike up like

US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 3

Page 6

1  this?

2	SPEAKER ON PHONE: It's fine, Rogene, for

3  me.

4	DR. HENDERSON: Okay, that's good.  This

5  is probably easier than dragging that hand mike.  Okay,

6  have, Angela sent the list of substantive material that

7  we want to have in the letter to the Administrator.

8  Have people had a chance to look at it?

9	SPEAKER: Rogene, you talking about the

10  short list or you talking about the big one?

11	SPEAKER: Big one.

12	DR. HENDERSON: I kind of meant the big

13  one.  This list, this small list, that doesn't have a

14  time on it, is just the notes I jotted down when we

15  were talking at the end of the day yesterday, when I

16  said, you know, what are the substantive issues we want

17  to convey to the Administrator.  And these are simply

18  my notes.  We were in agreement yesterday that this

19  list included everything we wanted to say.

20	Now, would, what I'm asking you, now, do you

21  think these were captured in the more formal listing

22  that Angela pulled together from the people who

23  summarized each charge question?   Well, that's a good

24  idea.  Though some of these overlap quite a bit.  We

25  have charge question one that, have you had a chance to

Page 8

1  rates.  I don't know if that information is available.

2  It's not my area quantitatively, but I was looking for

3  it, and I didn't see it in there.

4	DR. HENDERSON: Okay.

5	DR. CRAPO: Also, with respect to

6  question one, I think that this is an appropriate place

7  where we need to ask if there could be a better

8  assessment of issues related to background, and peaks,

9  and variations in exposure, more data about the

10  variations in exposure across groups, so that we know

11  what the, what percent of, or some idea about what

12  fraction of the country are people, or indoors, or

13  outdoors, is exceeding, or not exceeding the current

14  standard, but substantially higher than the current

15  annual average.  So, that the, the focus on an average

16  annual number makes it really hard for me to analyze

17  what the exposures really are.  So, I think we need

18  more data on that side of the table.

19	DR. HENDERSON: Okay.

20	SPEAKER: Well, I think that actually

21  falls directly under question two.

22	DR. CRAPO: Two, that be great.

23	SPEAKER: That's where ambient mon-,

24  concentrations are.  Some of it's there, but maybe just

25  what more might need to be there.



Page 7

1  look at it?  Would you like to have ten minutes just to

2  look at this one, okay.  Gary's nodding his head.

3  Okay, let's, we will just, we're not breaking.  We're

4  just giving you time to read it, because this is

5  important enough.  I'd like for you to have had a

6  chance to look at it carefully.

7  (WHEREUPON, the members read the document.)

8	DR. HENDERSON: I gather from the

9  conversation that is starting that people are

10  approaching the end of their reading.  Are you about

11  ready to move on?  Okay, what I'd like to hear from you

12  first is, is there anything left out of this that

13  should be added?

14	SPEAKER: Are you asking just about

15  question one, or about all of the questions?

16	SPEAKER: Let's go question by question.

17	DR. HENDERSON: You want to go question

18  by question?  Okay.  We'll take charge question one.

19  Is the response written her, does it include everything

20  that you think should be included?  Ron?

21	DR. WYZGA: One of the things that I

22  think could be useful could be, if they could have more

23  quantitative discussion about the rates of

24  transformation of, I guess, emissions into different

25  species of NOx and what are the influences on these

Page 9

1	DR. HENDERSON: Okay, we're looking at it

2  again.  We come to two, does anybody else have things

3  on one, yes, Terry?

4	DR. GORDON: Just the general, when I

5  read over this, I got the feeling that we were, I mean,

6  it's, the ISA is supposed to help us assess things, and

7  seems like some of the things asked to be added were

8  just making it more criteria document like, just making

9  it longer, and not, not helping us decide things.

10	DR. HENDERSON: Well, that's the strug- -

11	DR. GORDON:  Just a caution.

12	DR. HENDERSON: That's the struggle

13  that's going on, and some of this might go in the

14  annex.  I mean, but, we ask that it be condensed, and

15  that the only policy relevant information given.  And

16  what you're saying is, now, we're asking to expand it

17  in - -

18	DR. GORDON: In some areas.

19	DR. HENDERSON: In some areas.  Mary?

20	DR. ROSS: Well, that was a point of

21  clarification I was going to ask for in general.  It

22  says in the ISA's to include material.  Maps, in

23  particular, will make it longer.  So, one of the

24  questions is, can we balance between annexes and have,

25  expand the ISA a little bit more, but add more of the



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 4

Page 10

1  annexes, too, and just a clarification if that's what

2  CASAC means when they say in the ISA, or it really mean

3  in the ISA.

4	DR. HENDERSON: Okay.  That's a very good

5  point, and if we say ISA, do we really mean the annex,

6  so.

7	DR. WYZGA: And Mary, if these could be

8  cross references to the annex, I think that would be

9  helpful, too.

10	DR. HENDERSON: We'll go on to question

11  two, because that is what James asked for, you think,

12  is that included in this answer to question two?

13	DR. POSTLETHWAIT: Actually, as a present

14  follow up to James' point, I'm wondering about this

15  issue about getting a little better handle on, first of

16  all, exposures.  I don't know what data is available,

17  but even if some relative analyses of speciation of NOx

18  could be included, just to give us a feel.  I mean if

19  NO2 is 95 percent of it, then the rest of it's fairly

20  trivial.  If it's 25 percent of it, then, you know,

21  there's certainly other issues to consider.  And, I

22  think those two things ought to be in the ISA, and not

23  in the annex, so the reader has that, sort of that,

24  visceral feel as he, as they continue on to the health

25  effects portions, et cetera.

Page 12

1  view.

2	So, my recommendation would be much the

3  opposite as, maybe, minimize this, say that monitoring

4  height could be important, but if they're going to

5  address it in the ISA, they should do it in a more

6  conclusive fashion, and look at more monitors where

7  this, where they could pick up this impact.

8	DR. HENDERSON: Okay, I know, Dale, you -

9  -

10	DR. HATTIS: Yeah, I want to slightly

11  disagree with the fact that that's over-emphasized.  I

12  think that's a critical component of the analysis that,

13  if anything, should be extended to an analysis of,

14  actually, what the biases are in, as-, you know, would

15  be, in assuming that the distribution of levels of the

16  existing monitoring sites are representative of outside

17  outdoor levels, because it does mean that you can't

18  really directly com-, without an analysis of that

19  problem, you cannot directly compare the levels

20  inferred from monitors with the, it helps you

21  reconcile,  to a degree that it's possible, the, any

22  concentration response relationships you infer from it,

23  the epidemiological data, with concentration in your

24  response you infer from things like the Australian

25  study and the indoor, other indoor studies, which are



Page 11

1	DR. HENDERSON: Present those, the two

2  points again, so I'll be sure and get them right.

3	DR. POSTLETHWAIT: Well, again, to

4  speciate NOx, whatever's available, and then the issue

5  of what we know about personal exposure, temporal

6  paradigms.

7	DR. HENDERSON: Okay, some of that may

8  not be available in any of the exploits, but we, and I

9  agree with you, then, by healing what you want.  Yes,

10  Ted.

11	DR. RUSSELL: If I might, and this, also,

12  is captured in response to question three.  There's the

13  discussion about the importance of the height of the

14  monitors that shows up both in the last, sort of,

15  section on the response here, as well as in, there's a

16  fairly large bullet in the next one.

17	There's, currently, a pretty large section in

18  the ISA on the impact of monitoring height, and I,

19  actually, found that was much larger than it should be,

20  and maybe even a red herring as such, in terms of how

21  it might be addressed in the ISA.  For one, there's a

22  lot of information out there where you could compare

23  the values between different height monitors, as

24  opposed to just looking at one special study where they

25  did it.  Which, I think, would, gives you a biased

Page 13

1  based on - -

2	DR. RUSSELL: But you now,  tremendously

3  larger impacts horizontally and spatially than

4  vertically, so I think that that's being blown out of

5  proportion, versus where you're placing a monitor close

6  to a road, or you know, four or five hundred meters

7  away from a road in a park, because that's where you're

8  going to have the bigger differences.

9	DR. HATTIS: Well, I think that's, also,

10  an important area, but this is a systematic error, you

11  know, and the other may well be much more - -

12	DR. RUSSELL: No, it's, they're both

13  systematic.

14	DR. HATTIS: The health studies are based

15  upon aggregate exposure, agg-, exposures within big

16  cities, okay.  And those includes both stuff near

17  roadways and not near roadways.  So, essentially, that

18  tends to be biased by the verticality, although, there

19  are, in fact, some sub-populations within cities that

20  are even more exposed, okay, because of their, you

21  know, proximity to roads.  So, I think that the

22  influences are different, even though there may be a

23  bigger overall number, ratio in the near roadway, far

24  roadway.  This other effect, really, is a substantial,

25  why, I say, now, I don't think that it's overemphasized



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 5

Page 14

1  in the existing document.  I would like to see a fuller

2  analysis of both kinds of effects.

3	DR. HENDERSON: I want to be sure I

4  understand what you're saying here.  What is the

5  difference, say, between, you know, vertically.  Does

6  Albuquerque have a different from San Diego, I mean,

7  they, is that what you're talking about, I mean?

8	DR. HATTIS: No, no, no.  This is a

9  matter of the fact that, the monitors for all the

10  cities are high.

11	DR. HENDERSON: Yeah, and people are

12  breathing down low.

13	DR. HATTIS: And people are breathing

14  down low, so that means that, systematically, the epi

15  studies are based upon concentrations that are

16  measured, that are underestimated.

17	DR. HENDERSON: And what is, what is the

18  difference, the degree of difference, I mean, that

19  you're taking?

20	DR. HATTIS: Well, I think that, that

21  from the, you know, the brief discussion that I

22  remember from the ISA, that that difference is two-,

23  three-fold.  But that's different heights of monitors.

24  If you go down to the ground level, it looks like, you

25  know, that could even be a larger factor.  The fact

Page 16

1  level, it's fairly sparse.

2	DR. HATTIS: Well, whatever the best

3  sources of information are to estimate the effect, you

4  know, they need to be used.  The fact of the matter is

5  that, the existing epidemiological studies are based

6  upon, what appear to be, biased measurements of the

7  concentrations people actually receive.  And,

8  therefore, they are not directly comparable with the

9  indoor measured concentrations that led to the

10  observations in the Australian study and in the, well,

11  Australian study.  So, that's a big problem that needs

12  to be addressed.

13	DR. LARSON: I'm unaware of any, or many

14  NO2 EPA monitors that are actually sited on top of it,

15  is that what we're talking about?  I don't think that's

16  true.

17	DR. GORDON: Well, it just seems that,

18  from this discussion, I'm agreeing with Ted, now,

19  'cause I thought there was a big verticality problem.

20  And if there is more data out there, this chapter

21  doesn't get that across to me.  It says there's a big

22  vertical problem, but they might be variable by site,

23  and that's not brought out.  So, maybe Ted's right.  It

24  should be condensed but expanded in other areas.  I

25  mean they both should be discussed.



Page 15

1  that it's a systematic, you know, biases of several

2  fold can have a big impact on what you might infer is a

3  level that was, you know, protective of public health.

4	DR. LARSON: This is Tim Larson.  I think

5  the EPA monitors are sited, in most cases, in such a

6  way that the inlet height biases are not capturing what

7  you're thinking about, which is the vertical

8  distribution in urban areas, primarily, in confined

9  urban areas.  And that, I agree, is a significant

10  gradient that can be threefold.  But, you're not going

11  to see that at most NO2 monitoring sites, because

12  their, the way their sited, they're, they tend to be in

13  open areas.  And the differences in heights of the

14  inlets in those are-, in those open areas just don't

15  capture the kinds of gradients of exposure that are

16  important.

17	So, doing an analysis of all the inlet

18  heights for all the NO2 monitors that EPA has isn't

19  going to really capture that.  And, unfortunately,

20  there's just not a lot of data on the vertical

21  distribution of the heights in the urban areas that are

22  systematically done.  We're doing a big study in New

23  York City right now, trying to capture some of that,

24  and there is some European data on this subject, but

25  compared to the data that's measured at or near street

Page 17

1	DR. HENDERSON: Well, I hear-, what I

2  hear people saying is that in urban areas, there may be

3  a difference in, there may be a problem with the siting

4  of the monitors, as far as the vertical differences in

5  concentrations with NO2, but we don't have much

6  information.  I hear people saying, we don't need them,

7  we don't know if that's true, so would you like, Dale,

8  in the letter to say that this is a potential problem?

9  That should be addressed.

10	DR. LARSON: There is some literature on

11  this.  I mean, if you could cite that, I, there, it's

12  just not a lot of it.

13	DR. HATTIS: Well, whatever the

14  literature is that's relevant to estimating the

15  population exposures, that are true versus the

16  population exposures that are estimated in the

17  epidemiological studies, that's relevant to judging the

18  levels at which you expect how many of X.

19	DR. HENDERSON: Sure.

20	DR. LARSON: But I think I'm on balance

21  on Ted's, come down on, with Ted on this.  I think

22  relative, the NO2 EPA monitors in urban areas, the

23  biggest gradients are horizontal.  And they're not

24  proximity to roadway per se.  They're, actually,

25  proximity to confined roadways, where you can get up



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 6

Page 18

1  factors of three to five times differences relative to

2  the same traffic, of the same distance from a road in

3  an unconfined location.  So, you're not talking about

4  twenty percent here.  You're talking about three to

5  five hundred percent differences.  And similarly,

6  factors two to three in the verticality at those

7  confined locations with height of, those are big

8  effects, none of which are being captured by any of

9  this.

10	DR. HENDERSON: For any epi studies,

11  there's always the problem of exposure.  I mean, we're

12  never happy with the exposure.  Now, and, I think this

13  is an example of some of the issues that come up.  I

14  think it should be mentioned in the letter.  As I

15  recall in reading through the document, it was

16  discussed quite a bit, but - -

17	DR. LARSON: Well, the inlet height

18  effect of the monitors is discussed, which I'm not sure

19  is the important parameter.

20	DR. HENDERSON: I think in the letter, we

21  do, we confirmed the fact that we are aware that the

22  exposures, there's always a problem with measuring

23  personal exposures in an epi study.

24	DR. HATTIS: Yeah, but this is not just

25  the usual problem.  This is not, the usual problem is a

Page 20

1  types of measurements that are used in some of the, you

2  know, the better direct studies establishing the

3  effects of the NO2.

4	DR. HENDERSON: To point out the

5  uncertainties associated with the other, go ahead,

6  George.

7	DR. THURSTON: Yeah, right, well, yeah, I

8  agree with that last part that the problem is when you

9  go to compare it to, like, indoor measurements and

10  those measurements.  But, it's not a problem with

11  regard to interpreting and the epidemiology, I think we

12  have to keep that clear, and applying it to for

13  standard setting.  Because, ultimately, you know,

14  you're applying the standards at the central site

15  monitors.

16	So, that's what you want to use in the

17  epidemiology, and the fact that, let's say, those

18  levels, let's say, they were fifty percent of what

19  people were actually exposed to, it, then you would

20  take all the numbers, double them, and then when you go

21  to set the standards, divide them by two.  I mean, it

22  would be a waste of time.

23	So, I think that it's a fact, but it's not a

24  problem that there are differences in the absolute

25  levels between what's at the central site monitor, and



Page 19

1  random error.  And we know how to deal with that.

2	DR. HENDERSON: I understand.  You're

3  saying that this is - -

4	DR. HATTIS: This is a systematic error.

5	DR. HENDERSON: - - systematic because

6  the, you think the inlets are consistently higher than

7  the level of - -

8	DR. HATTIS: Look it, all I know is what

9  I read in the ISA, and this seems to be, you know, what

10  the ISA seems to say.  And then I, sort of, believe

11  that they will have located the monitors at elevated

12  levels.  You know, maybe, and if it's not true, then

13  fine, you know, but.

14	DR. HENDERSON: Well, I, our charge is to

15  advise them on how to improve the ISA, and are you

16  saying you'd like the - -

17	DR. HATTIS: I'd like that, and I think

18  that if it's, you know, if the analysis, if the

19  statements in the ISA are correct, then, you know,

20  maybe they need to be modified with, including the

21  information from a larger literature base.  But, you

22  know, if they are, then it's worth an a-, worth some

23  much more quantitative analysis, because it creates a

24  serious difference between the types of measurements

25  that are used as the basis of the epi studies, and the

Page 21

1  what people experience on, at street level.  But, only,

2  the only place where I, you know, I think it is a

3  problem that I can think of, you know, I agree.

4	When you go, if we're going to put some

5  importance on these indoor studies, we ought to

6  remember that those concentrations are not directly

7  comparable to the central site concentrations.  And

8  that's, I think, the key that Dale brought out.

9	DR. WYZGA: And the clinical studies as

10  well.

11	DR. THURSTON: Yeah, and the clinical

12  studies as well, yeah, that's true.  Because the actual

13  concentrations associated with the NO2 exposures that

14  we measure at the central site monitors are, actually,

15  higher.  And so, that might explain some of the

16  differences that we see between the exposure studies,

17  the indoor studies, and the ambient results.  So, it,

18  yeah, so that's going to have some importance later on

19  in interpreting the results, so that is an important

20  point to bring up in that respect.

21	DR. HENDERSON: Yeah, I can see that- -

22	DR. HATTIS: A condition of the central,

23  there is always random error, addition, to effect, you

24  know, and that's also a problem to be analyzed, but.

25	DR. HENDERSON: Well, okay, I see the



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 7

Page 22

1  point that, George, that, you know, you just made, and

2  Dale, too, that there is, the, how something comparing

3  the clinical and the indoor dose response first to what

4  you may see in epidemiology.  That can be, I think,

5  clearly stated.  Yes, George, do you have - -

6	DR. THURSTON: I just have one separate

7  comment.  I guess it, I'm not sure if it goes on two or

8  three, but I think two, that I brought up something

9  that I, in my quick review, I don't see reflected with,

10  yesterday, which was that we need to, more clearly,

11  delineate the difference between personal exposures to

12  all NOx versus personal exposures to ambient NOx, and

13  their respective relationships to outdoor central site

14  monitors.  I didn't see that written in here anywhere,

15  and I did bring that up.  And I hope that that's

16  included.

17	DR. LARSON: Well, we had a bullet in

18  section three on trying to look at the alpha, I guess

19  I, the ratio of the outdoor to personal ambient.

20	DR. THURSTON: Is that what that bullet

21  means?

22	DR. LARSON: Yeah, alpha.

23	DR. THURSTON: I didn't get it.

24	DR. LARSON: Okay, we'll fix it.

25	DR. CRAPO: Could I ask a question for

Page 24

1  want to talk about that in the context of the ISA,

2  fine, or context of the other is fine, too.

3	DR. HENDERSON: Were you proposing that,

4  well, this would be quite a few individual comments,

5  but can you explain it to us what it's saying?

6	DR. HATTIS: Yeah.  What this, I need to

7  get it in front of me.  What this is, is, essentially,

8  plotting the, it is, basically, a lo-, what these are

9  called is log normal probability plots.  And,

10  essentially, what's being plotted is the Z score, which

11  is, essentially, the number of standard deviations that

12  each value represents in the distribution.

13	So, that, for example, the first data point

14  here is, generally, the first per-, is the one

15  percentile level.  The next is the, I think the five

16  percentile level, et cetera.

17	But, plotted on a probability scale, so that,

18  if, in fact, the date corresponded to a log normal

19  distribution, which is the usual expectation, then the

20  points would fall on the straight line.  The regression

21  equation in each case is an estimate of the, the

22  intercept is the log of the geometr-, it's an estimate

23  of the log of the geometric mean, and the slope is the,

24  an estimate of the log of the geometric standard

25  deviation, okay.



Page 23

1  clarification as I listened to the conversation.  When

2  we get an average annual level expressed for us, is

3  that what, is that the average over the whole 24-hour

4  day, then averaged annually.  Or is that the high for

5  the day averaged annually.  Are we talking about - -

6	SPEAKER: Everything.

7	DR. CRAPO: - - everything averaged

8  together, so when NO2's have the peaks during the

9  traffic periods of the day, and it goes down very low

10  at night, you're taking these high levels that occur

11  during the day and averaging it out with twelve to

12  twenty hours with, of low levels and getting a fairly

13  low level out of it.  That's, so, we need a lot more

14  information about the peak, 'cause, probably, the

15  average annual is about the last thing we want to look

16  at to assess this st-, this substance.

17	DR. HENDERSON: You know, I think, that's

18  what I have written down for your, what I wrote down

19  for this.  And we need to remember about the pattern.

20	DR. HATTIS: I've made a series of plots,

21  actually, of the distributions of, for different

22  average-, of NO2 levels for different averaging times

23  from the existing data in one of the annex tables.

24  And, so, we can talk about that later.  Yeah, that's

25  the, yeah, that's the graph, so, essentially, so if you

Page 25

1	So, essentially, these, so these, if these

2  log transformed values were normally distributed, they

3  should fall more or less on the line, and they more or

4  less do.  They are not perfect log normal

5  distributions.  In fact, the actual data have, don't

6  have as fat a, in the tails as they should for a

7  perfect log normal.  But, essentially, what this does

8  is to show the change in the slope is, means,

9  essentially, the longer averaging time or, you know,

10  more tightly distributive than the shorter averaging

11  time.  So, the shorter the averaging time that you

12  take, the data are further spread out, just because of

13  regression of the mean effects.  And this says how,

14  how, what?

15	DR. HENDERSON: I mean that's what you

16  need said, isn't it?

17	DR. HATTIS: Yes, and this, basically,

18  quantifies how much less the dispersed the lo-, the

19  yearly and three averages are relative to the one-hour

20  averages.

21	DR. HENDERSON: And that's what James is

22  saying, that they didn't give us much information, I

23  mean, yes.

24	DR. CRAPO: So, let me just ask a real

25  practical question from a real simple mind.  How many,



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 8

Page 26

1  if I, and instead of doing the average annual and

2  saying that was fifteen parts per billion, what would

3  be, if I took the highest one hour from each day, would

4  that be two hundred parts per billion?

5	DR. HATTIS: Those data are in the table,

6  and I didn't plot them.

7	SPEAKER: Actually, they're in the ISA,

8  too.

9	SPEAKER: They're in the hot spot.

10	DR. CRAPO: Right, so what's the answer?

11	SPEAKER: 201.

12  (WHEREUPON, there was a discussion off the record.)

13	DR. CRAPO: Did I guess, I guessed it

14  right on the money?

15	SPEAKER: Yes.

16	DR. HENDERSON: You get a gold star this

17  morning.

18	DR. CRAPO: And the excursion, the high

19  end excursion is from that?  Do we have a significant

20  if the population is exposed 500 ppd?

21	SPEAKER: No, that was an excursion.

22	DR. CRAPO: That is an excursion, I was,

23  that is the excursion, okay.

24	DR. HENDERSON: Well, okay, Ed, go ahead.

25	DR. POSTLETHWAIT: Well, if there are

Page 28

1  proportionally more while I'm breathing more.

2	DR. CRAPO: Ted has just pointed to me

3  that the 200 ppb that we were talking about is probably

4  at fifteen feet up and not at ground level, so that the

5  ground level might twice that level.  What?

6	DR. LARSON: Ron, wait, no way, no way.

7	DR. RUSSELL: More likely at four meters,

8  aren't more of your monitors at four meters than - -

9  (WHEREUPON, there was a discussion off the record.)

10	DR. LARSON: Those kinds of gradients

11  don't exist.

12	SPEAKER: What?

13	DR. CRAPO: You say gradients of that

14  nature don't exist?

15	DR. LARSON: Not that, I mean, not that

16  strong a gradient over three meters.

17	DR. PINTO: Yeah, no, I think you're

18  right, I mean.  I think what I was trying to say was,

19  no, this particular data point, okay, where it was a

20  change in Lakewood, California, downtown Los Angeles,

21  in other words, and is one of the roadside monitors, so

22  you would expect it a, first of all to be very hot; b,

23  you would also expect the inlet to be at, you know, the

24  standard there, at the standard height, which of the

25  order of three meters or so.



Page 27

1  estimates made of personal exposures, won't that

2  capture that issue, as opposed to, when we compare data

3  back just to annual averages, which gets back to the

4  central issue of how much stuff are people really

5  inhaling, versus, you know, what's the average floating

6  around.

7	DR. HATTIS: Not exactly, because it

8  depends upon, on the averaging time for the personal

9  exposures, where you also will have a similar - -

10	DR. POSTLETHWAIT: Well, that's where,

11  that's where temporal plot, if it was possible, would

12  be really useful, even from a qualitative standpoint.

13	DR. HATTIS: Yeah.  But essentially,

14  you'd have to have comparable, you know, different

15  lengths of time averages to be able to compare.  And I,

16  offhand, I don't know whether the internal, the indoor

17  exposures are more variable with time than the outdoor.

18  So, you can have a different, you could have different

19  comparability depending upon the, you know, how, what

20  that looks like.

21	I mean, there's, also, a likely case that the

22  indoor exposures will be correlated with differences in

23  breathing rate.  So, for example, it may well be that

24  while I'm up and about, one of the things I'm doing is

25  cooking on my gas stove, and exposing myself to

Page 29

1	DR. CRAPO: The reason I'm bringing this

2  up is that I think those of us that are really focusing

3  on the health effects are not even in our heads not

4  even correlated to the right thing.  And we're sitting

5  here looking at the ambient levels and thinking 15 ppb

6  average annual, and we're seeing health effects in

7  asthmatics and people living near roadsides.  When, in

8  fact, the people near the roadsides are getting 200

9  ppb.  And our correlation, all these correlation

10  coefficients on the things that we're looking at are,

11  at least, I'm not sure that because we've used a, such

12  a bad metric to correlate what's going on, I don't

13  think that we're thinking correctly on the health

14  effects side.

15	DR. HENDERSON: I think one thing we can

16  emphasize in our letter is the importance of the

17  temporal and spatial variability in the NOx exposures,

18  and how that will vary.

19	DR. CRAPO: Because this makes our

20  biological plausibility, the discussion yesterday,

21  change directions completely.  It puts us, it, we were

22  arguing that we weren't exposing enough to NO2 to get

23  the level.  If these things, if these exposure metrics

24  change, then our whole argument yesterday, the

25  biological plausibility is met, becomes much more



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 9

Page 30

1  likely.

2	DR. RUSSELL: No, maybe Jill wants to go

3  into this further.  I mean, again, the high monitor is

4  Lakewood in downtown Los Angeles.  It's a roadside

5  monitor, so it's going to be high.  You know, actually,

6  some of these monitors are sitting really near

7  freeways.

8	DR. CRAPO: Well, so am I.

9	DR. RUSSELL: Right, but what I'm saying

10  is that, keep in mind, when you're talking about, these

11  monitors, many of these monitors are capturing very

12  much the highest levels that you're going to get,

13  except in a very confined street canyon.

14	DR. CRAPO: Okay, and that's what I'm

15  thinking, is that the high levels that are causing the

16  disease that we're seeing, and we're not understanding

17  who's got that high level, and where it is, and why.

18  As well as, at least, the medical side of us are,

19  because we're not dealing with the numbers in the form.

20  So, I'm just wondering if the, if this is, a large part

21  of our discussion yesterday wasn't confounded by some

22  of us not quite understanding the exposure levels that

23  our sub-populations were being exposed to.  And, you

24  know, the fact that your, having those, no NO2 all

25  night long is irrelevant to the fact that you get up in

Page 32

1  you were, had actually got, a see-, cumulative

2  distribution function, that those are the ones you'd

3  see at the upper tail.  So, I think we are capturing

4  those.  And, actually, in response to your question

5  about the biologic plausibility is, I think it goes

6  both ways, is that in many cases, I think we might be

7  looking at overestimates of what the potential exposure

8  to NO2 is in a general population.  Because a lot of

9  people live out in the suburbs, and you know, again,

10  I'm sort of parochial in knowing at Atlanta, is that,

11  we've got more monitors near busy areas, than we do

12  sort of in the general suburbs.

13	DR. HENDERSON: I, well, again, I think

14  maybe we can cover this by a paragraph discussing the

15  importance of the temporal and spatial variability of

16  the pollutants and this is not special to NOx.  It's

17  always a problem, and that we, that this should be

18  emphasized to discuss in the ISA.  And it is true that

19  we'll extend, but I don't hear anything that's, that

20  couldn't be covered under the importance of temporal

21  and spatial variability, and the, what we listed, as

22  far as monitoring and determining exposures that we, we

23  had discussed this yesterday afternoon, there's the,

24  you know, you have the indoor outdoor exposures, the

25  spatial and temporal variability, the siting of the



Page 31

1  the morning and get a big dose from 8:00 a.m. to 12:00

2  noon or whatever.

3	DR. BALMES: This is John Balmes.  I have

4  to, I need a clarification.  I thought I heard Tim

5  Larson say that most of the site, and I don't have any

6  map in front of me here on the phone.  I heard Tim

7  Larson say that most of the regular monitoring sites

8  are not, they're in open areas and are not necessarily

9  near freeways.  But I just heard that a lot of the

10  monitors are by freeways.  That makes a big difference

11  to me.

12	DR. LARSON: Not a lot of them.  Some of

13  them are, but not a lot of them.

14	DR. BALMES: Right, I think that's

15  important to know.

16	DR. LARSON: Most of them are not.

17	DR. BALMES: Yes.  That was my

18  understanding, too.

19	DR. RUSSELL: Yeah, so I misspoke when I

20  said a lot, but you do have a representative population

21  of ones that are near freeways.

22	DR. BALMES: Right.

23	DR. RUSSELL: And those are the ones, or

24  very heavily traffic roads, and those are the ones that

25  you do see on the one extreme of our population.  If

Page 33

1  monitors, all of that is something that we've already

2  said is very important.  So, I would like to just

3  summarize that in a paragraph in the letter, and

4  emphasize how the differences between the more precise

5  measurement indoor in the clinical studies, and, as

6  opposed to outdoor ambient studies.

7	Does that, would that co-, I mean, we've,

8  you're absolutely ri-, and if you don't have the

9  correct exposure, the response then is, it can't be

10  related to the amount of, precisely  towards lead and,

11  but that stands, that's always a problem with epi

12  studies.  They don't have it for very long exposures.

13	DR. AVOL: Just one small point of

14  information, I think what Joe meant was Lynwood,

15  California not Lakewood.  The Lynwood station is

16  alongside the Long Beach Freeway, and gets several

17  hundred thousand vehicles a day.

18	SPEAKER: Yeah, thanks, Ed.

19	DR. LARSON: Well, as my comments

20  yesterday, and you know, at each of those, you know,

21  two sites, there is a information in the database on

22  distance from major roads.  And you could compile that

23  fairly easily, and probably, compare that with the

24  population, U.S. population at large.

25	DR. HATTIS: I think that'd be a good



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 10

Page 34

1  thing to do to try to see what biases one should expect

2  and what, you know, how do we characterize the

3  variability and the likely exposures in relation to the

4  variability that we see in the monitors.

5	DR. HENDERSON: Is that something, Mary,

6  that could be done?

7	DR. ROSS: We can look into it.

8	DR. PINTO: I mean, perhaps, with help

9  from the program offices, I mean, I tend to think that

10  that sort of effort if, you know, if done well, I mean,

11  could take a bit of time and maybe, even, longer.  I'm

12  thinking in terms of longer than the time scale for

13  setting the next draft to come out.  But we'd have to

14  look into that, Rogene.

15	DR. HENDERSON: Okay, well, let's, I

16  think we've had a good discussion of this issue, which

17  is a, certainly, an important one.  Can we look now

18  beyond the first three charge questions to going to the

19  health, unless there's anybody else has something else

20  on the first three charge questions.  The next four

21  charge questions relate to the health effects.  And,

22  was there, were there things that were left out or

23  that, yes, Ed.

24	DR. POSTLETHWAIT: As part of the charge

25  four things, and this just may be simply an issue of

Page 36

1  Balmes, again.  I think the last sentence in the bullet

2  here, is, contains key information about that the ISA

3  would be improved if a plan or process for integration

4  and study selection is clearly laid out.  So, that it

5  would be clear to some, to a reader, such as Ed, why

6  studies were included.

7	DR. HENDERSON: That's a good point.

8  Yeah.  Are there other things?

9	DR. WYZGA: Rogene, I had a couple of

10  things on five, but.

11	DR. HENDERSON: Okay, Ron, and then

12  Joyce, go ahead.

13	DR. WYZGA: Okay.  I guess, first of all,

14  I'm flattered that my name is mentioned, but I would

15  also mention that John, in number five, John Balmes

16  mentioned some toxicological studies that weren't

17  included.  And I would change the wording to say that,

18  instead of several of the latest NOx human field

19  studies is, basically, several recent epidemiological

20  studies that examine the association between health

21  outcomes and NO2, were, either, not included nor

22  studied correctly, and say, especially, in describing

23  the impacts of other pollutants on the NO2 health

24  associations.

25	DR. HENDERSON: Certainly, and nobody's



Page 35

1  verbiage.  On that third line, where it says only the

2  key studies that support an NAAQS should be included.

3  I wasn't sure what that meant, in terms of, to support

4  what we have now; to support a new one, I mean.  And

5  when you read that, it almost sounds like you, as we

6  discussed yesterday, you could, inadvertently,

7  introduce selection bias on what studies you were

8  reporting, positive versus negative.

9	DR. HENDERSON: Well, this still is from

10  the, has a little history behind it.  And that's what

11  I, how I interpret it.  If you looked at the CD, it

12  includes everything from, you know, a 500 ppm exposure

13  of a toad frog to, you know, something at ambient

14  levels.  And you're right.  How do you choose the key

15  studies.  But I think the meaning of this statement is

16  that, that chapter three could be condensed to even

17  more to make it less like a CD, and more, just includes

18  studies that are relevant for setting a standard.  I

19  think that's the meaning of it, but you're bringing up

20  a problem which we have discussed, CASAC has discussed,

21  and who chooses, you know.  But we came down that it

22  was more beneficial for us doing this review to have

23  the Agency choose what they felt were the key relevant

24  studies.

25	DR. BALMES: But that, this is John

Page 37

1  saying it's going to be mentioned in here, I mean.

2	DR. WYZGA: Okay.

3	DR. HENDERSON: You know, I mean, as I

4  said, I mean, this will not be covered - -

5	DR. WYZGA: But, I guess, part of it is,

6  it's broader than simply the epidemiological is because

7  there's some toxicological studies as well.

8	DR. HENDERSON: Okay, yeah, okay.  And,

9  is that ment-, it's a tox study, yeah, Balmes, oh,

10  yeah.  No, we, it's mentioned there.  Sure, Ron, never

11  in my wildest dreams would I mention names.

12	DR. COTE: This is an opportunity,

13  though, for me to ask people if you, if there are

14  specific papers that you're aware of that we don't

15  have, please give us the references.  Because, you

16  know, we've done this careful lit search, for whatever

17  reason those papers have not popped up.  So, you know,

18  if it's a flaw in the keywords or whatever, so please

19  help us by giving us the specific references rather

20  than.

21	DR. BALMES: So, I will include, this is

22  John Balmes.  I'll include the ones that I referred to

23  yesterday in my written comments.  But the nitrogen

24  dioxide will get the one paper published in 2005.  You

25  don't have to get fancy with the keywords.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 11

Page 38

1	DR. COTE: Thank you.

2	DR. HENDERSON: Okay, so, Ron and John

3  will provide, you will provide those in your written

4  individual comments.  Then that, then in the letter, we

5  can refer to it.  We can, see individual comments of,

6  with them, okay.  Anything about question six that

7  people have problems with or would like to add.

8	DR. THURSTON: Well, as to the question

9  five.

10	DR. HENDERSON: You back to five, okay.

11	DR. THURSTON: Yeah, well, I mean,

12  actually, the reason I put Ron's name in there was to,

13  so that I knew he would respond to that and clarify

14  that sentence with it.  I figured that, otherwise, he

15  might get ignored and he would make sure it was

16  correct.  So, we could just write members also pointed

17  out, or something, instead of put, naming names here.

18  The other thing is, in the iterations, I don't know,

19  either I didn't have it in the beginning, or it got

20  left out or something, but I would add at the, in the

21  last sentence, just, well, I'm just going to get this,

22  finally, examining the epidemiology results, and I

23  would say, after results, add the words across outcomes

24  as a function of.  Because, the whole idea was to look,

25  not just individually, but look across outcomes and

Page 40

1  you say that yesterday?

2	DR. SHEPPARD: I don't think I was the

3  one that brought that up.

4	DR. WYZGA: Okay.

5	DR. SHEPPARD: Probably is an important

6  point.

7	DR. HENDERSON: Okay, are there more

8  comments on the substance of answers to question five?

9  Go on to six, then.

10	DR. AVOL: I have a question on six.

11  This Ed Avol.  My question is this.  About seven lines

12  in, there's a comment about sensitive populations.

13  There's no comment about genetic susceptibility in

14  that, and I just have a question for whoever wrote

15  this, if that was a conscious exclusion because they

16  don't believe it's sufficient - -

17	DR. CRAPO: It wasn't conscious.  It was

18  late at night in the middle of a bad Rockies game.

19	SPEAKER: That was a good game.

20	DR. CRAPO: So, let's add that, let's

21  just add that.

22	DR. HENDERSON: So, what line, at what

23  line, yeah, where is it, what line is it, that - -

24	DR. AVOL: It's about seven lines down,

25  the last just in, I'm sorry, there's evidence of



Page 39

1  look for coherence.  That's it.

2	DR. HENDERSON: Okay, and I will remind

3  you, as far as wordsmithing, when we finish here this

4  morning, Angela and I will be drafting the actual

5  letter, based on the substance of these comments.  And

6  you will be receiving it for concurrence and review.

7  So, small wordsmithing, you can take care of at that

8  point if you want.  Can we go on to charge question

9  six, then?

10	DR. WYZGA: One, on question five.

11	DR. HENDERSON: Five, I'm still on five.

12	DR. WYZGA: And, is Lianne on the phone?

13	DR. HENDERSON: Lianne wasn't coming on

14  till when, 9:00.  What's Angela's time.

15	DR. NUGENT: Rogene?

16	DR. HENDERSON: Yes.

17	DR. NUGENT: Is Lianne on the phone?  She

18  said she would be on the phone.

19	DR. SHEPPARD: Yeah, I'm here.

20	DR. WYZGA: Okay, Lianne, this is Ron

21  Wyzga.  You said something yesterday, number five,

22  something about, and I wanted to see if I could capture

23  it, about looking more systematically or in a better

24  organized way at the, how one deals with co-pollutants

25  and interpret studies using co-pollutants.   Did I hear

Page 41

1  adverse health effects in sensitive populations such

2  as.

3	DR. CRAPO: And how would you say that,

4  though.  The only ones that have really been studied

5  are the ones that are mentioned here, really.  I mean,

6  genetic is a theoretic thing, but there's no hard study

7  that says this is a gene that creates susceptibility to

8  NO2.

9	DR. HENDERSON: You could add a sentence

10  - -

11	DR. CRAPO: I mean, I believe it's true.

12  I just don't think, I can't think of a study that would

13  prove it.

14	DR. HENDERSON: You could add a sentence

15  saying, genetic polymorphisms may also influence the

16  response.

17	DR. AVOL: I mean, there is some

18  published information from our lab and others on GST,

19  and the sensitive to oxidative stress mechanistic

20  pathways.

21	DR. HENDERSON: We can put into the,

22  genetically perception.

23	DR. POSTLETHWAIT: Will that fall under,

24  stuff under question seven.  I mean, does, six and

25  seven are, essentially, addressing the same chapter in



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 12

Page 42

1  the ISA.  And there is this issue of defining

2  susceptible populations.

3	DR. AVOL: Yeah, that's, I mean, I think

4  that's fine.  I'd be happy to put it into seven.  I

5  just pointed it out there, because it seemed like it

6  was - -

7	DR. POSTLETHWAIT: Right, yeah, I mean,

8  this becomes redundant, just like, you know, I said

9  it's redundant.

10	DR. HENDERSON: Okay, I noted that that

11  should be mentioned.  Are there other things for six or

12  seven?

13	DR. RUSSELL: Rogene, before going on,

14  I'm actually curious.  On the last sentence, that, I

15  mean, to me that's a rather important sentence.

16	DR. HENDERSON: Which question are you -

17  -

18	DR. RUSSELL: Oh, six.  That we concur on

19  the findings and, et cetera, directly result in adverse

20  impacts.  And this comes, you know, I was sitting here

21  a little uncomfortably yesterday about the use of the

22  word likely causal when we, they say the strongest new

23  evidence comes from epidemiologic studies of ED visits

24  and hospitalization.  And I'm an air quality person, so

25  I, the medical end is somewhat beyond me, but,

Page 44

1  whether, in verbiage that, while it may not be the

2  chemical species NO2, per se, certainly, there's a

3  linkage between NO2 in the air and all the other

4  goodies and these adverse health outcomes.

5	DR. CRAPO: I don't remember exactly what

6  I said.  I can tell you that I, last night, I

7  deliberately wrote this sentence very strong, 'cause I

8  wanted to make us talk about it.  I think we have to

9  decide, I mean, this is, I think, the heart of the

10  paper right here in that one sentence.  It tell them

11  whether we really agree or don't agree with the

12  fundamental conclusion of the document.  The, when I

13  get, at the end of the day, I'm impressed that the,

14  it's not, we have to cite the few studies that were

15  negative; but in fact, this is a, it's a ten to one

16  vote in favor of positive, but it's not an equal, half

17  were positive and half were negative.

18	These are, the most of the studies that come

19  out are showing strong effects.  And I do think that

20  it's likely the effects are tracking primarily the

21  products of combustion, which the paragraph says.  The,

22  but the, but there are, overwhelmingly, strong data

23  showing an association that we haven't really dealt

24  with.  And I'm concerned that the biggest problem is in

25  our exposure metric.  I think that our correlation



Page 43

1  recognizing that the epi studies that we've talked

2  about the problems with the monitoring and the spatial

3  variability.  There are some that find associations.

4  There's others that don't.  And, it almost struck me

5  that, I was sitting here going, the strongest new

6  evidence, and you have some studies that go the other

7  direction, and I recognize EPA was, sort of, sitting on

8  the fence on this one, too.  Directly result may just

9  overstate how I think I feel on this.  And I'm

10  wondering how others feel, too.

11	DR. HENDERSON: I think that's a strong

12  statement considering our discussion.  I agree with

13  you.  I don't find, I think we're all trying to figure

14  out is there something here or not, and - -

15	DR. SHEPPARD: Yeah, I agree as well.  I

16  think that's pretty strong.

17	DR. AVOL: I think, in fact, that

18  yesterday in the discussion, John Samet challenged the

19  consistent coherent issue.

20	DR. POSTLETHWAIT: James, yesterday, you

21  and I were talking, and you came up with a really great

22  multi-word descriptor, you know, that might soften this

23  a bit.  And I'm trying to remember what you said.  It

24  was something about, you know, NO2 appears to be a key

25  player, I mean, you said it better than that.  And

Page 45

1  needs with, needs to be with the highest one-hour

2  average.

3	Not, like ozone was before it changed to an

4  eight hour averaging time.  And if we converted all of

5  our data to one-hour averaging time, we might have a

6  lot more confidence in this conclusion.  But I don't,

7  but I can't walk away from the strength of the data

8  that's summarized in the ISA.  It's a very strong

9  document with studies from every dimension from every,

10  from lots of different countries, consistently finding

11  associations with products of combustion that metric

12  was within all.

13	DR. BALMES: So, the, this is John

14  Balmes.  I agree with you, Jim, James, that the

15  epidemiologic data taken as a whole from a

16  stratospheric level are pretty impressive.  My problem

17  is that I don't think the coherence is, necessarily,

18  there with the toxicologic data.  We talked about that

19  yesterday.  And I don't, actually, personally, have a

20  problem with the Agency moving ahead with a new

21  standard if that's the ultimate outcome, based on

22  epidemiologic data.  I do epidemiology.

23	I appreciate its value.  But I think it's, I

24  don't think the toxicology is really there to support

25  the epidemiologic findings.  And it's often that



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 13

Page 46

1  toxicology is behind epi.

2	So, I think we should recognize that, I mean,

3  if we have to say that.  Because, otherwise, people

4  will criticize the document for not, sort of, fairly

5  representing the literature.  And I think, with regard

6  to respiratory infection risk, I think the toxicology

7  is there.

8	But I don't think that we understand why NO2

9  causes, or is associated with a kind of lung function

10  decrements that the children's health study found.  I

11  think that's a very important finding that should be

12  very strongly emphasized in the document.  But I don't,

13  I, certainly, don't understand how that occurs.

14	DR. CRAPO: Would you say the toxicology

15  does support it if we're speaking about 200 ppb instead

16  of 15 ppb?

17	DR. BALMES: Uh - -

18	DR. CRAPO: 'Cause I've changed my mind

19  on that one.

20	DR. BALMES: Well, I think that's a

21  greyer area, but I'm not sure that 200 ppb, the

22  toxicology supports 200 ppb.

23	DR. LARSON: James, Tim Larson, again.

24  15 ppb is your annual average.  The 200 ppb is your

25  one-hour max.  So - -

Page 48

1	DR. LARSON: I'm not disagreeing with

2  your general conclusion, but I'm just saying, to be

3  more precise, those epidemiology studies are probably

4  looking at 24 hour time series.  And those 24 hour

5  averages are certainly at a max greater than 15 ppb.

6	DR. CRAPO: Yeah, and I'm, but I'm also

7  saying that is a, I think that NO is not driven, NO's

8  health effects are not driven by the daily average.

9  That's probably driven by the peak, and - -

10	DR. LARSON: Right, but - -

11	DR. CRAPO: - - and the, and we never

12  looked at the peak in terms of comparison of.

13	DR. LARSON: Right, because a relevant

14  comparison would be the 24 hour versus daily max hourly

15  average.  Because the type, and the epi are based,

16  primarily, I believe, on the 24 hour, but I'm just

17  saying, those are the two numbers to compare.

18	DR. CRAPO: Well, it is except that we're

19  looking at fairly profound health effects, and, I mean,

20  in many of these studies, and - -

21	DR. LARSON: Well, and I'm agreeing with

22  you.  I'm just saying that the change from one day to

23  the next is greater that 15 ppb.  And, you know, in

24  some cases, the one-hour max could be several hundred,

25  and the change from one hour to the next could be



Page 47

1	DR. CRAPO: Well, that's one-hour max in

2  monitors on certain places.  It's not the one-hour

3  personal max.

4	DR. LARSON: Right, but what's the, what

5  is the annual average at that monitor.  I mean, that,

6  in terms of, if there isn't, I mean, there's two things

7  going on here.  One of them is spatial, you know,

8  proximity to roads, et cetera.  The other one is the

9  annual average versus the one-hour average.  Both cause

10  differences in these numbers.  But, when you say 200

11  versus 15, one's a chronic exposure and one of them's

12  acute exposure.

13	DR. CRAPO: Well, I know that, but the

14  toxicology's almost all acute.  And the peaks are

15  acute.  And I, see, yesterday, I was sitting here

16  thinking about the 15 ppb and saying, I've got an order

17  of magnitude or two orders of magnitude difference in

18  my toxicology and my epidemiology.  But, in fact, I

19  don't.  It's, they're coming together.  The lowest

20  threshold effects for NO are, you know, are in some, a

21  few hundred ppb.  We saw that yesterday, where they

22  looked at the lower limits of toxicology having

23  effects.  And now, we've got peak levels at fifteen

24  feet up in the atmosphere, in certain locations,

25  pushing the same levels.

Page 49

1  fairly large, too.  So, I mean, the 15 ppb, it seems

2  like, is, it's not even in the range of what we're

3  talking about.  I mean, that's an annual average, and

4  it just gets washed out, as you say, by all the

5  seasonally flow seasons and all the midnights and

6  everything else, so.

7	DR. AVOL: This is Ed Avol.  Not to go

8  back to discussion of the health effects, but since the

9  lung function changes in the children health study we

10  brought up, let me just point out one perspective.  And

11  that is that in, of course, in looking at lung function

12  growths or decrements in lung function growth among

13  children, we're looking at long-term changes of

14  children that are moving around their communities, and

15  we're looking at those annual averages from those

16  central site monitors in those areas.  And so, while it

17  may be true that close to roadways or at traffic peaks,

18  there are several hundred parts per billion

19  concentrations, in fact, our relationships with those

20  changes in lung function are with the annual average.

21	DR. CRAPO: Well, I understand that

22  factor.  I'm just hypothesizing, I mean, I'm saying

23  that the health effects appear to be real.  And I,

24  biologically, couldn't explain them with the annual

25  average, but then it occurs to me that what's likely



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 14

Page 50

1  happening, maybe I only need to hit my kid with 200 ppb

2  once a week in order to cause adverse changes in his

3  growth of his lung.  And I take him on my freeway for

4  three or four hours a week while I'm driving various

5  places.  So, it could be that that's the problem, and

6  we have-, but my point is, we haven't even analyzed it

7  that way.

8	DR. LARSON: Right.

9	DR. HENDERSON: Well, there is no - -

10	DR. CRAPO: But the health effects are

11  real.  That's what this sentence says.  The health

12  effects are real.  I don't know why yet, but I, there,

13  and it might not, and it might be a surrogate, but

14  they're, but they are so uniform across so many

15  studies, that we have to take them serious.

16	DR. HENDERSON: Now, as far as - -

17	DR. BALMES: This is John Balmes.  I

18  agree with that.  It's, but I, so the, the coherence in

19  the epi, I have, I'm comfortable with.  It's, if people

20  are assuming from that statement that we mean coherence

21  with the toxicology, I don't think we're really there

22  yet.  That's all I'm trying to say.

23	DR. THURSTON: Could I say?

24	DR. HENDERSON: George?

25	DR. THURSTON: Yeah, I don't really think

Page 52

1  supports that, too.

2	DR. CRAPO: Well, I'll say it, 'cause,

3  actually, I wrote it with, we said it several times

4  yesterday, but when, every time you do the study, you

5  find that there's an adverse health effect, which you

6  can link it, like, to the roadway, or to the children's

7  study with an open fuel on their furnace in the home,

8  the powerful correlations that go with this, the fact

9  the primary source of NOx is combustion.  So, it seemed

10  like an obvious to me.

11	DR. RUSSELL: Yeah, that one I have no

12  problems with.

13	DR. HENDERSON: I have no problems with

14  that, either.  I think that's a solid statement, but

15  George, would you repeat your modification of the last

16  - -

17	DR. THURSTON: Yes, I may have it here.

18  CASAC concurs that the epidemiologic findings indicate,

19  we'll see, that current ambient, is directly, no, are

20  associated with.  That's what epidemiology tells us,

21  associations.  Are associated with adverse impacts on

22  the public health, comma, but that the document needs

23  to better - -

24	SPEAKER: Articulate?

25	DR. THURSTON: Well, you could say the



Page 51

1  we're at with, we may end up where this last sentence

2  is, but I don't think we're there yet, in my opinion.

3  So, I mean, I would just say, concur that the

4  epidemiologic findings, you know, indicate, let me see.

5  I had it written down here, too.  Yeah, concurs that

6  the epidemiologic findings indicate that current

7  ambient NO2 exposures are associated with adverse

8  effects on the public health.  But the document needs

9  to better document, or better, you know, lay out the

10  plausibility, a consistency in coherence.  I think that

11  work, that needs to tightened up, and that's where we

12  ought to be focusing this next iteration.  I don't

13  think we're done yet, and this gives the impression

14  we're done.

15	DR. HENDERSON: That we're done.  No, I

16  would agree with that, and I'll let Mary talk on it.

17  Would you write down your modified sentence so that.

18	DR. ROSS: While you're on that subject,

19  I just want to draw your attention to a sentence a few

20  sentences earlier.  CASAC recognizes that the primary

21  associations are between products of combustion and

22  adverse health effects.  That's also a strong

23  conclusion that will have policy implications.  And

24  just wanted to make sure if that's something you agreed

25  with.  It's helpful if you provide why, you know, what

Page 53

1  ISA needs to, yeah, the ISA needs to better document

2  that these findings are plausible, consistent and

3  coherent with, now  do we want to say with toxicology

4  or something with other evidence?

5	DR. CRAPO: Well, to me, I would say, in

6  looking at lots of different medical kinds of issues, I

7  see more consistency of this data than anything,

8  virtually, anything that I make medical decisions on.

9  A lot of consistency across broad settings, where the

10  issue has been plausibility or coherence with the

11  toxicology.  And that was a dose issue.  It was not,

12  there's plenty of toxicology at high dose.  There's no

13  question about if you're talking about 10 ppb, I mean,

14  10 ppm, there's no question it correlates.  So, the

15  whole issue is, to me, the only issue is dose.

16	SPEAKER: Well, yeah, I mean, that is a

17  big - -

18	DR. THURSTON: Well, I just think it

19  needs to be better.  I think you're probably right.

20  There is, having read it, you know, there is a lot of

21  that evidence, but it hasn't been laid out in a way

22  that makes it obvious that where the coherencies are.

23	DR. BALMES: Well, one thing that might

24  be useful is to look at the relationship between these

25  one-hour peak exposures and the annual averages.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 15

Page 54

1	DR. HENDERSON: I thought that was done

2  in there, but I - -

3	DR. HATTIS: We do have direct evidence

4  in a table in the ISA on that point, that essentially,

5  the 99th percentile of the one-hour maximum is like 72

6  ppb.  It's not the highest is like 200, but that's the

7  highest of 288,000 measurements.  So, I mean, you're a

8  little bit far out on the scale there with the 200.

9  But, certainly, the 99th percentile is about, is 72

10  parts per billion whereas the 99th percentile, the one

11  hour of the yearly averages is 33 parts per billion.

12  So, you have a couple fold there, which gets a little

13  closer to the toxicology, but.

14	DR. HENDERSON: Mary?

15	DR. ROSS: You know, when we talk about -

16  -

17	DR. HATTIS: But that's again, for the

18  monitors that, some of them, which may be close to

19  roadways, but they're still a little high up, so it

20  may, may still be some additional distortions.

21	DR. HENDERSON: Mary?

22	DR. ROSS: We tried to evaluate the

23  short-term exposure studies that looked at different

24  indices, and there's a small discussion on page 5-5 of

25  24 hour studies versus one-hour max studies.  And they

Page 56

1  tended to look at NO2.

2	DR. CRAPO: Yeah, so I mean, you could

3  say it either way, but since they interconvert, you're

4  a little bit wondering what it really is you're.  I

5  mean, is the NO2 a surrogate for NO-, for all the other

6  species.  And so that's why I use them interchangeably,

7  without being very discretionary.  I knew, I need, you

8  should use the same term.

9	DR. HENDERSON: So, you're saying, I

10  mean, what you're suggesting, Ron, is that NOx can be a

11  significant factor?

12	DR. WYZGA: Yes.

13	DR. HENDERSON: Does anybody have an

14  objection, I mean, can we agree on can be?

15	DR. CRAWFORD-BROWN: Now, is that can be

16  in the sentence of can be under some circumstances?

17	DR. WYZGA: Well, we've avoided making

18  definitive conclusion in the last sentence, and we're

19  saying, you know, we're waiting for the document to,

20  basically, organize and, you know, give us a redraft.

21  And it seems to me to make that conclusion that it is,

22  it can be, and I think that's one of the things we're

23  waiting on, you know, the next round, to see whether or

24  not the document supports, you know, it is a

25  significant factor.



Page 55

1  don't find, there's not a lot of difference in the

2  epidemiologic.  Now, that is one-hour max on a given

3  day, but, you know, we did try to evaluate that, and

4  we'll look at if there are any further studies.

5	DR. HENDERSON: Okay, Ron?

6	DR. WYZGA: Rogene, I would say, I guess

7  in the spirit of what we said, if you look at the

8  previous sentence, we, basically, say NOx is a

9  significant factor, and I wonder, given what we said

10  later, if we could change the is to can be.

11	DR. HENDERSON: Yeah, I know, I see

12  (WHEREUPON, Dr. Henderson reviewed the document.)

13	DR. ROSS: And can I ask one more, I'm

14  sorry to keep bothering you but, when you say NOx, do

15  you mean NO2 or NOx, and it's one of the things we

16  battle with all the time is selecting the term.

17	DR. CRAPO: I use them  interchangeably,

18  because we did it yesterday.  I don't think we know,

19  exactly, what the species is, but NO2 seems to be a

20  good surrogate for it, so you could use NO2, but, in

21  fact, you're measuring the, you're measuring the, well,

22  you're using it as a surrogate for NOx, so probably, I

23  think NOx is your better term, because you don't really

24  know it's NO2, do you?

25	DR. WYZGA: Except the studies have

Page 57

1	DR. HENDERSON: Yes, that's the sense

2  that I understood.  Okay, did we, for the NOx issue, we

3  decided to keep the same term.  What did you decide on

4  the - -

5	DR. CRAPO: I like NOx better, because it

6  - -

7	DR. KENSKI: I actually would prefer NO2

8  in the, you know, all of the epi stuff is based on NO2

9  measurements, and the tox stuff is NO2 measurements,

10  and you know, yes, they are, they do interconvert, but

11  you know, the peak, peak, I mean, you know, what we

12  measure as NOx is, what we measure as NO2 is, you know,

13  the difference between NOx and NO-, so I, I don't know.

14  I just think it's better to be consistent and keep

15  that, you know, link with NO2.

16	DR. HENDERSON: The toxicity data is

17  based on NO2, I mean.

18	DR. KENSKI: Right.

19	DR. HENDERSON: Yeah, I mean, the

20  clinical and the - -

21	DR. KENSKI: And what's repor-, and

22  what's reported, granted, it's not, you know,

23  absolutely pure, you know, true NO2, but it's as close

24  to it - -

25	DR. CRAPO: Well, I wouldn't be, I'd be



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 16

Page 58

1  happy to accept either one, as long as that, someplace

2  in it, you defined that three was interconversion and

3  that NO2 is as, is a critical species in the sequence.

4  So, you just make, as long as you define what you're

5  using, that's fine to use the other term, as far as I'm

6  concerned.

7	DR. KENSKI: It just might be good to add

8  a sentence, you know, up front saying that, you know,

9  we acknowledge that, you know, the NO2 that we measure

10  is not, you know, true a hundred percent - -

11	DR. WYZGA: But I think the other thing

12  is that, that we use NO2 with the relationship between

13  NO2 and the other components of NOx may change

14  temporally and spatially, and we don't really have

15  enough evidence to say that it's consistent.  And if

16  that relationship were consistent, then, I think, we

17  could jump to NOx, but it's not consistent.

18	DR. KENSKI: Right, and we're asking for

19  a better, you know, definition of some of those.

20	DR. HENDERSON: Can we go on to seven and

21  eight.  I put in on seven about the genetics.  I

22  thought, I don't know who wrote eight, but I thought

23  that was well written.  It was very clearly written.

24  Somebody wrote that.  It's Doug.  Oh, we, we'll get a

25  my kudos to Doug.  I thought that was well written.

Page 60

1  inherently increased exposure represent susceptibility.

2	DR. HENDERSON: Oh, okay, or   whether it

3  goes in the exposure.

4	DR. COTE: Just as a point of

5  clarification on that.  We tend to talk about

6  susceptible and vulnerable populations, and susceptible

7  is a more innate quality.  And vulnerable being people

8  at increased risk, or individuals at increased risk for

9  some not intrinsic attribute.  So, exposure would be

10  increased vulnerability.

11	DR. POSTLETHWAIT: Then the first bullet

12  needs to be changed to incorporate, not only defining

13  susceptible, but defining vulnerable.  And then, depend

14  upon what the panel feels, you can leave the high

15  exposure in there, or not.

16	DR. HENDERSON: Okay, I see, you, there

17  is the, the people near the roadway are vulnerable

18  because of the high exposure.  That makes sense.  I,

19  and that's what the question asks, susceptible or

20  vulnerable.

21	DR. ROSS: And to expand on that, the

22  vulnerable population includes the two sub-categories,

23  other than the biological, the socio-economic and the

24  geographic were, generally, extrinsically sensitive.

25  So, we could split that vulnerability up into two



Page 59

1  The, we've talked about the multi-pollutant aspects,

2  and I'm, hope-, I think Ellis, probably, you

3  contributed that.  And, that will be, can be worked

4  into the letter as a major point that, you know, after

5  all our discussions, we still have this problem of the

6  multi-pollutant aspects for, when we try to assess the

7  risk of air pollutants, particularly the different

8  oxidant pollutants.  But, let's see if there's any big

9  changes in seven and eight that we want to make.

10  Particularly, anything we want to add.

11	DR. BALMES: I thought I heard yesterday

12  that some people were uncomfortable with the idea of

13  defining a susceptible group relative based on their

14  where they live.  That's the first bullet in, uh - -

15	DR. HENDERSON: Page seven.

16	DR. AVOL: I think Tim's right.  There

17  was some discussion about moving the issues of high

18  exposure locations and near roadway into exposure.

19	DR. POSTLETHWAIT: Yes, and when I  put

20  this together during the Rockies game - -

21	DR. CRAWFORD-BROWN: I don't think the

22  Rockies actually had a game.

23	DR. POSTLETHWAIT: That's what I was

24  thinking.  Yeah, I mean, that's, it can go wherever.

25  The question is is whether it does, does increased,

Page 61

1  components, and that would address, I think, Dr. Avol's

2  questions.

3	DR. AVOL: Yeah, but I think there is a

4  interaction here in the sense that vulnerable

5  populations, those are the high exposure alongside

6  roadways, are likely, are disproportionately likely to

7  be lower SES and get into issues of environmental

8  justice.  And then, they may have biological in that

9  sense, be the former susceptible.  They may also fall

10  into the susceptible population as well.  So, they get,

11  sort of, a double whammy.  But I think that it is true

12  that there are susceptible and vulnerable sub-

13  categories here.

14	DR. HENDERSON: Okay.  Any more for

15  charge question seven?

16	DR. COTE: If you have time, I had a

17  quick question.  The two on this, on the page five, the

18  partial bullet at the top, the last sentence, the

19  chapter did not address biologic plausibility with

20  regard to specific populations, thus it's difficult to

21  attribute health outcomes to direct causal.  I think

22  we're all in agreement when you have biologic

23  plausibility, you're much better off.  Or is this

24  intended to mean, though, if you don't have mechanism

25  of action information, then you can't say there's a



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 17

Page 62

1  causal outcome?

2	DR. POSTLETHWAIT: Well, my intent was

3  to, sort of, throw that open for discussion that, would

4  it make the document more robust to have, why would an

5  asthmatic be more susceptible to NO2 than a normal.

6  For example, what is it about the biology of NO2 that

7  induces susceptibility in a specific subset of the

8  population.  Yes?

9	DR. COTE: Yeah, I think we can make that

10  stronger.  I'm not sure we can actually, and in each

11  case, can be successful, do we not.

12	DR. CRAPO: The asthmatic has a more

13  responsive airway, and greater responsive inflammation,

14  and the NO2 is an irritant, so it could easily be an

15  oxidant.  So, it could easily be - -

16	DR. POSTLETHWAIT: That one's easy.  Some

17  of the others may not be so easy.  And we can't - -

18	DR. LARSON: Are we doing, dealing with a

19  likely causal, and that term.  I mean - -

20	DR. POSTLETHWAIT: Yeah - -

21	DR. LARSON: - - causal is a diff-, is a

22  higher standard than likely causal.  My understanding

23  was, you don't need the biological plaus-, I mean, you

24  don't need the detailed mechanism to go to likely

25  causal.

Page 64

1  you've got injury, repair, growth, and development.

2	DR. COTE: Okay.

3	DR. POSTLETHWAIT: And, which is not, two

4  of those factors are, don't occur in the adult.

5	DR. HATTIS: Yeah, if you want to back

6  off from unique, you might say distinctive.

7	DR. POSTLETHWAIT: Sure.

8	DR. COTE: Thank you.

9	DR. HENDERSON: Okay.  Did you get that,

10  Angela, distinctive.  Let's look at charge question

11  eight and, uh - -

12	DR. AVOL: Could we just go back.

13	DR. HENDERSON: Oh, sure.

14	DR. AVOL: I'm actually, I mean, I think

15  that Ed is right.  It is unique because of the growth

16  aspect.  The tissues are in the period of growth and

17  are more sensitive.  And I think that is a unique

18  attribute.  But it's not a unique population, anything

19  in the population that makes them unique, susceptibles

20  population.

21	DR. HENDERSON: I think we can work that

22  in.  Let's see.

23	DR. LARSON: The first part of the

24  sentence refers to the children's health, California

25  health studies.  The second part refers to children in



Page 63

1	DR. POSTLETHWAIT: Well, in a perfect

2  world, the detailed mechanisms would be wonderful, but

3  I'm not sure we're there yet, or even close.

4	DR. HENDERSON: And regulations have to

5  be decided on, you know, even in the absence of

6  mechanisms for sure.

7	DR. COTE: The second question I had is

8  in the next bullet, the word unique.  I wasn't and sure

9  what was intended.  So, it says a unique and probably

10  susceptible - -

11	DR. POSTLETHWAIT: Hey, Jim Ultman,

12  you're up.

13	DR. HENDERSON: Jim, are you there.  You

14  were there.  Not answering.

15	DR. POSTLETHWAIT: He sent me this.  I

16  cut and pasted it in, so it's his fault.

17	DR. HENDERSON: Oh, okay.

18	DR. COTE: That's o-, it's not, it's not

19  a deal breaker either way, so that's okay.

20	DR. HENDERSON: Well, I - -

21	DR. COTE: Thank you anyway.

22	DR. HENDERSON: Okay.  We can find out.

23	DR. POSTLETHWAIT: I mean, there is some

24  uniqueness in children because of the superimposition

25  of exposure on top of growth and development.  So,

Page 65

1  general.

2	DR. HENDERSON: Yes.

3	DR. COTE: I thought the concept of the

4  injury, repair, growth, development was what was useful

5  for me.

6	DR. HENDERSON: Yes, I, and, maybe, we

7  can put that in, that children, I don't know, are

8  unique in that, you know, injury, growth and repair.

9	DR. POSTLETHWAIT: Actually, you could

10  throw in a fifth variable, which would be dose, 'cause

11  their running around breathing harder, we hope.

12	DR. HENDERSON: Okay, now can we go on to

13  eight, and I, Doug wrote this.  It just seemed like it

14  was very clear and captures many of the concerns that

15  the committee had.  The, a multi-pollutant aspect, I

16  think, it will be something we'll bring up at the end

17  of the letter as a, you know, general concern we have

18  for all airborne pollutants, and maybe we'll suggest

19  the need for, in the future, striving to address, you

20  know, multi-pollutants, rather than one pollutant at a

21  time, a one atmosphere approach, which the Agency is

22  trying to take anyway.

23	DR. LARSON: But, I also thought, based

24  on the discussions yesterday, that there was a general

25  scientific consensus that nitric oxide was not a



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 18

Page 66

1  confounder either in the palliative, its palliative

2  effects, or its, you know,  irritant effects.  There

3  was no, really no information at these concentrations.

4  I guess, we don't really have the EPA summary of all

5  the nitric oxide concentrations, but it just seemed

6  like we had, sort of, generally, concluded that if, in

7  fact, it is a mixture, it's, it doesn't seem to be the

8  nitric oxide that's doing much of anything.  And then,

9  when you, when you, if you eliminate that, you're, sort

10  of, the next most abundant thing is NO2, and then you

11  start going way down in abundant for these other

12  species that we don't know much about in terms - -

13	DR. HENDERSON: I agree, or what I was

14  thinking of multi-pollutant was ultra fines and ozone

15  and - -

16	DR. LARSON: I agree with that, but I'm

17  just saying that, even though we can't, necessarily,

18  say much about that, I think we can say something about

19  the biological plausibility or the lack of it for

20  nitric oxide.  Because that seemed to be a point of

21  confusion early on in yesterday's discussion.

22	DR. HENDERSON: Okay, I think it's pretty

23  clear in the document.  That's where I read it, so in -

24  -

25	DR. LARSON: Yeah, I mean, we're just

Page 68

1  to is a major reorganization of the document to move a

2  lot of information from the earlier parts into chapter

3  five.  That's not what we intended, though.

4	DR. CRAWFORD-BROWN: I need to  answer

5  that, though.  I mean, I certainly would agree that the

6  entire document is the resource to which people need to

7  turn, but I just don't know what findings and

8  conclusions mean if it isn't a summary of the most

9  important and relevant points from the previous

10  chapters.  I, you know, I would think that there will

11  be people who are going to say, look, I don't have time

12  to read your whole document.

13	Tell me what I really need to know as a

14  policy maker, as somebody who's going to try to do a

15  risk assessment, and so forth.  Tell me what I really

16  need to know in order to be able to make those

17  determinations.  You're the scientist.  I'm not the

18  scientist.  So, I just, you know, that chapter needs to

19  be a chapter that does summarize everything from the

20  past as far as relevant conclusions are.

21	DR. COTE: You know, I was going to, I

22  meant to ask this yesterday, if there were specific

23  examples where there were more important conclusions in

24  the body than in the chapter five.  If somebody could

25  note those, just when you see them.  Things were



Page 67

1  re-, I mean, I was just reading it.  It seems, it still

2  seemed to be a point of confusion.

3	DR. HENDERSON: Okay, I got it.  We're

4  coming to the end here, and as again, this is the first

5  time we've tried this sort of process.  And I need to

6  know if everyone on the phone and sitting around the

7  table is co-, if we modify as we have discussed here

8  this morning extensively, if we modify the content of

9  these points, are you comfortable with these, this

10  being the substance of the letter that we send to the

11  Administrator.  Now, I'm not talking about

12  wordsmithing, et cetera.  Because, what will happen is

13  that this draft letter will go to all of you, so if,

14  you know, if you have wordsmithing problems, don't

15  worry about it.  It's the substance of what's in the

16  letter that I want to know if you're comfortable with.

17  And Mary, why are you raising - -

18	DR. ROSS: May I ask one final, about

19  question number eight.

20	DR. HENDERSON: Okay.

21	DR. ROSS: We had actually intended that

22  the entire ISA be the document that serves as support

23  for risk and exposure assessment.  If it is intended

24  that only the conclusions chapter be the resource for

25  risk and exposure assessment, what that's going to lead

Page 69

1  actually were written by the same people, and I think

2  chapter five reflected, as we were working, perhaps,

3  more refinement of thinking, and so, it's a little

4  disturbing if it was better the first, rather than the

5  second round of thinking, so if you ju- -

6		DR. CRAWFORD-BROWN: If we combine a list

7  of the things that are back in the earlier chapters

8  that - -

9	DR. COTE: That should be - -

10		DR. CRAWFORD-BROWN: But I think the main

11  issue had to do, also, with the fact that the writing

12  of chapter five does need to be the bridge to the user

13  of this document, and what that user needs, as

14  important information,  for the kinds of decisions he

15  or she is going to make.  And that's where I didn't,

16  you know, I, personally, I know Ellis felt the same,

17  didn't feel that that connection was quite there, where

18  somebody at chapter five began to ask, what are people

19  actually going to use this for in the end.  And that's

20  why I raise this issue of integrated.  I don't think

21  there's such a thing as integrated outside of the

22  context of the question that somebody is trying to

23  address.

24	DR. COTE: Was it that there weren't



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 19

Page 70

1  of information were missing, or both?

2	DR. CRAWFORD-BROWN: I, but I mean, my

3  personal opinion is that what happens with chapter five

4  is that, it is, what often happens with concluding

5  chapters in theses, for example, where it's just a

6  compilation of things from the earlier chapters.

7  Here's a thing from chapter one, and here's a thing

8  from chapter two.  Rather than somebody,

9  systematically, sorting through and saying, what do we

10  think we really learned from the earlier chapters that

11  are relevant to the kinds of applications that we

12  thought we were directing this report towards.

13	DR. CRAPO: I think a good example is the

14  issue I was talking about a lot this morning about the

15  dose metrics being annual average, and not telling you

16  what the people, the populations were exposed to,

17  actually, in terms of the more toxic elements of the

18  high level exposures.  And then, a discussion of that,

19  so that the person who tries to interpret the health

20  effects data, in relationship to the possible

21  exposures, both for what they know and don't know, is

22  not there in chapter five.  A person that would read

23  that and just think you had, it would just jump right

24  from the exposure data, think it had totally supported

25  all the findings.  And so, I think, that disconnect

Page 72

1  of these 47 statements, or however many it becomes, and

2  that those that are directly relevant to the issue of

3  making a judgment about the standard be highlighted or

4  marked in some way, and that the, so that's the first

5  point.  And the second point, in chapter two, you were

6  asking for examples of then something that was

7  mentioned in chapter two, but didn't show up in the

8  summary that was in chapter five.  And all of the nine

9  statements in chapter five are relevant to the issue of

10  monitoring alone.

11	DR. COTE: Right, I heard that, yes.

12	DR. COWLING: So, I was just thinking to

13  mention those examples.

14	DR. COTE: Thank you, and what you

15  provided on criteria for judgment, I thought was very

16  good, too.

17	DR. CRAPO: I'd like to add one more

18  thing.  I think the biggest thing that might come out

19  of this review of NO cycle is a recommendation that we

20  go to a one-hour daily average instead of an annual

21  average.  And, no matter what the level is set at, it

22  would totally change our science.  But, we ought to

23  set, I think that's what's needed more than anything

24  else, because I think that we're measuring the wrong

25  thing.  And, I would argue that our document ought to



Page 71

1  that we struggled with for two days here needs to be

2  obvious in chapter five.

3	DR. HENDERSON: And I lay a few, John

4  Samet had quite a bit to say about chapter five in his

5  comments.  And he had the, since so many people had the

6  same conclusion that it was really just a listing of

7  the, what, of items from the previous chapters,

8  excluding some, because several people said that, and

9  not an integration of, you know, all five.  So, I think

10  that it really does need attention.

11	DR. COTE: Clearly.

12	DR. HENDERSON: You don't have that many

13  people giving almost the same comments without there

14  being something that - -

15	DR. COTE: No, no, I wasn't disagreeing.

16  I was just trying to get more - -

17	DR. HENDERSON: Just get more examples

18  and - -

19	DR. COTE: Yeah.

20	DR. HENDERSON: Yes, Ellis.

21	DR. COWLING: I would offer two comments.

22  One is George's suggestion yesterday that, and it's

23  relevant to what Doug was just saying about, what is

24  relevant to the decisions that were made next.  And it

25  seems to me that he was suggesting that a scan be made

Page 73

1  set the, ought to, appropriately, set the background

2  for that type of a recommendation, 'cause that's where

3  I think we're headed.  And it's not in there now.

4	DR. HENDERSON: Has the shorter averaging

5  time been considered in the, by the Agency, because in

6  our discussion, I had the same thought, James, and I

7  thought, well, gee, maybe we're looking at the wrong

8  averaging time.

9	DR. CRAPO: Well, both daily and, sure, I

10  mean, one-hour and dailies is, those are two changes to

11  it.

12	DR. HENDERSON: I'm just curious if the

13  Agency has, that's ever come up.

14	DR. ROSS: Well, I mean, you can look in

15  the history of the rule making, and in 1993, actually,

16  you'd have to ask Karen Martin the specific history,

17  but we made an effort to try to breakdown into short-

18  term and long-term exposure discussions.  And within a

19  short term, there are a range of different levels.

20  Many of the epi studies use 24 hour, but we did try to

21  discuss the evidence at, related to averaging time.

22  Tox studies use a whole variety of different exposures.

23  And then, we try to make that available, to the extent

24  we can say something about peak exposures, we will.

25  There aren't many epi studies that look at peak



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 20

Page 74

1  exposures.  I'm not even sure there are many tox

2  studies, but we'll try to bring that up as much as we

3  can.  I mean, we're taking that ho-, as a comment that

4  we need to address.

5	DR. CRAPO: Remember if the ozone field

6  goes that direction, for shorter averaging times and

7  it's, because it's been there for decades, it has

8  really influenced the thinking of the evolution in that

9  field.  This data would suggest that NO2 has a toxic

10  profile similar to ozone.  In fact, it interacts with

11  ozone to make it this toxic product.  So, there's no

12  rationale for having a different, an annual averaging

13  time for NO2, and a short averaging time for ozone.  I

14  would just argue that you can use the science of the

15  ozone science to justify a lot more evaluation of why

16  NO2 ought to have the same type of short-term

17  evaluation on it.  And part of our problem is we set it

18  up wrong thirty years ago, and we've got a bad

19  collection of data to compare everything to.

20	DR. BALMES: I guess the other point that

21  we should emphasize, this is John Balmes, is that if

22  asthma exacerbation is one of the major endpoints that

23  we feel the epidemiology supports, and I certainly

24  think it does, then it makes no, you know, an annual

25  average does not protect asthmatics from exacerbation.

Page 76

1  me, the epi drives it.  We'd, logically, might think,

2  oh, it's a short term for that day that really drives

3  it, but we have to come up with a value, and if all the

4  epi is driven by lo-, the annual averages, we got a

5  tough task.

6	DR. HENDERSON: That's a good point.

7	DR. CRAPO: For particulates, we have

8  two, so we could keep both, then.  We could put a

9  short-term and a long-term standard in.  Well, we

10  couldn't, but the Administrator could.

11	DR. COTE: And if you were, if you're

12  thinking about two different kind of health effects,

13  like lung growth and asthma, there's no reason to think

14  it would be the same.  It might be, but I'm not sure, I

15  don't know.

16	DR. BALMES: And, there's also no reason

17  to have the same type of siting criteria for your

18  monitors if you're going to go to a short-term

19  standard.  Because, I can walk down a street canyon for

20  an hour and get a completely different exposure than I

21  will at a EPA monitoring site for an hour.

22	DR. ROSS: Just to remind people, we're a

23  little ahead of the process here, talking about the

24  standards already and the sited criteria.

25	DR. HENDERSON: No, we tend to jump over



Page 75

1	DR. HENDERSON: Okay, thank you.  I was

2  thinking of the ozone and, you know, the eight hour

3  standard make sense because it's much higher during the

4  daylight hours.

5	DR. CRAPO: And so NO2 is the same- -

6	DR. HENDERSON:  Is, is NO2 in the same -

7  -

8  (WHEREUPON, there was a discussion off the record.)

9	DR. CRAPO: NO's shorter than the ozone

10  peak, isn't it?

11	DR. RUSSELL: It's actually a very

12  different shape.

13	DR. HENDERSON: Yeah, I think we'd have

14  to be a little careful, but you know, I - -

15  (WHEREUPON, there was a discussion off the record.)

16	DR. HENDERSON: Terry, Terry has his hand

17  up, or you want to go to Dale?

18	DR. GORDON: I had a feeling that this

19  conversation was going to go this way, and I was

20  wondering if it did, we went to shorter term.  Are we

21  going to lose something.  It sounds like people are

22  leaning toward a short term, not a long term, and how

23  would that effect the true long-term studies, such as

24  the children's lung growth studies, which might be more

25  correlated epi wise with annual averages.  I mean, to

Page 77

1  to the endpoints, so let's, but, I really would like to

2  draw this together so that we complete our peer review

3  of the ISA document, and Ellis, would, I'll give you

4  the last call before I ask.

5	DR. COWLING: Well, I would just like to

6  support what Jim Crapo has suggested here, with the

7  additional suggestion, and this is where they did do

8  what Karen Martin told us yesterday.  What was in the

9  mind of the Administrator, and what are the policy

10  implications of having an annual standard, and what are

11  the policy implications of having a daily standard, or

12  any other standard.  And it seems to me that we ought

13  to know what was the rationale in 1971, when an annual

14  standard was selected.  And now, and then, we have the

15  other iterations in '93 and so on, what was in the mind

16  of those who made the decisions at that time.  And I

17  think if that is clarified, it would provide a more

18  rational basis for a decision about what is the proper

19  averaging time.

20	DR. BALMES: Ellis, I can tell you one

21  thing.  I was, way back in my youth, I was in the

22  public health service at EPA.  And the very first

23  criteria document, as you know, were very thin, and

24  committees were about five or six people, and the

25  process took a day, so.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 21

Page 78

1	DR. HENDERSON: Oh, you're making us all

2  jealous.  Anyway, I do want, we are skipping way ahead.

3  We're going to the next step.  And we, before we go

4  there, we need to complete our review of the ISA, which

5  is looking at the science.  So, I want to know if

6  everyone in the room and on the phone is comfortable

7  with the substance of what we're going to put in the

8  letter.  And you will see this, the draft come-, the

9  letter come out, and you will get to, we will seek

10  concurrences.

11	We always do, so, is there anyone who is not

12  comfortable with it?  John Samet, are you on the phone?

13  Oh, he's coming this afternoon, okay.  Well, we have a

14  quorum of the chartered members of CASAC here who are

15  all comfortable with this, so I consider that the

16  charter members have approved this, the substance of

17  this letter that's going to go out.

18	What comes out next is going to be the draft

19  letter, and with Angela's able help, I hope we can get

20  it out fairly soon.  And then, you must look at it very

21  carefully, and we will seek concurrence before it

22  actually goes in.  And any questions about that

23  process?  Well, I thank everybody for cooperating so

24  well with this new way of doing things.  We'll see if

25  it works out.  I don't know if it, it hasn't quite

Page 80

1  and I would ask, the first speaker is going to be

2  Lydia?

3	MS. WEGMAN: Yeah.

4	DR. HENDERSON: Okay, and Lydia, maybe,

5  you know, as you go through, you can introduce the

6  others from the Air office who are going to be

7  participating.  So, it's a real privilege to have Lydia

8  with us.  She always clarifies things.

9	MS. WEGMAN: Well, I don't know about

10  that, Rogene, but thank you very much.  My name's Lydia

11  Wegman, and I am the Director of the Health and

12  Environmental Impacts Division in the Office of Air

13  Quality Planning and Standards.  And we are the folks

14  who will be working on the exposure and risk

15  assessment, and ultimately, the advanced notice of

16  proposed rule making and the proposed rule and final

17  rule.  And I do want to introduce the folks who are

18  with me, or the ones who have done the real work on the

19  scope and methods plan for the exposure and risk

20  assessment.  And Dr. Karen Martin, who will speak in a

21  moment after I'm done, Dr. Scott Jenkins, Dr. Stephen

22  Graham, and Dr. Harvey Richmond.

23	MR. RICHMOND: I'm no doctor.

24	MS. WEGMAN: Oh, no doctor, you should be

25  a doctor, though. You do the work of a doctor.  So,



Page 79

1  reached it conclusion, but all of you, by participating

2  so readily, I think have helped it, and we may continue

3  to do this.

4	The next thing on our agenda was to move on

5  to the next document.  I think maybe it's time for a

6  break, and we'll take a fifteen minute break, then

7  we'll come back and we'll hear from the Air office.

8  And they are going to move us in the direction we keep

9  trying to go.

10  (WHEREUPON, a break was taken.)

11	DR. HENDERSON: If everybody could take

12  their seats.  Thank you, Doug.  Okay.  We're going to

13  be moving on, here comes Ron, if others could take

14  their seats.  We're going to be moving on to a

15  consultation now for our next document, which is the

16  exposure risk assessment methods document.  And, as

17  we've been saying, we keep jumping in this direction

18  from going from the science assessment to wanting to

19  participate in this part of the process.  And this is

20  our opportunity.  As a consultation, this is where we

21  can, early on in the development of this process,

22  provide advice to the Agency on the methods for

23  exposure and risk assessment.  So, we had quite a bit

24  of discussion on exposure assessment this morning.  It

25  is, we're going to hear first, then, from the Agency,

Page 81

1  this is our team on the exposure and risk planning for

2  the NOx review, the primary NOx review, and we'll be

3  coming to talk with some of you next week about our NOx

4  and SOx  secondary review.

5	I first want to say thank you all for taking

6  the time to review the work we've done, and to spend

7  the couple of days you're spending here in RTP, either

8  in person or by phone, to offer us your comments.  Your

9  comments are invaluable to us, and without the work of

10  CASAC, we would not be able to perform our work.

And I

11  just want to say how very important your work is to us,

12  to the Agency as a whole, and to public health.  And

13  thank you very much for all the work that you do.

14	I, also, just want to make one point, as you

15  offer us comments on the scope and methods plan.  As

16  you have seen, and I know several of you have commented

17  on, we've got a tiered assessment, both for the

18  exposure and risk assessments.  And, one of the reasons

19  we have the tiering is that, we don't know whether

20  we'll have the scientific evidence to go through all

21  tiers of these assessments, and we are very much

22  looking to you for advice on how to prioritize these

23  assessments, and what we do need to do, based on what



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 22

Page 82

1  are the resource constraints that we face in our

2  office.  We have constraints, not only in terms of

3  people and money, which are constraints that we always

4  face, and it does seem that we face them more every

5  year with declining budgets.  But, we also have a time

6  constraint in this case.  As you know, we have a court

7  order.  It's currently under review by the public, but

8  we very much anticipate having a final court order that

9  gives us firm dates by which we do have to complete the

10  proposed and final rules, as well as the ISA.  And,

11  that does limit what we can do.  And I want to alert

12  you to that, because I know that there is a desire on

13  all of our parts to do the maximum amount of assessment

14  that we possibly can do with the science that we have.

15  But we will, in fact, be facing some constraints, and I

16  want to seek your help in knowing what is the most

17  important thing to do within the time and resource

18  constraints we have.  So, as you think about these

19  issues and give us your advice today, I'd appreciate it

20  if you kept that in mind.

21	And now, I'm going to turn it over to Karen,

22  who is going to offer a few thoughts on multi-pollutant

23  assessments.

24	DR. MARTIN: Just, while this is, you all

25  have been talking some about the issue of multi-

Page 84

1  step, in deciding how to approach our exposure and risk

2  assessments; what is it we're assessing the risk of.

3  We can,  obviously, make some up-front assumptions, and

4  my comments yesterday were intended to help get your,

5  at least, preliminary thinking to help us do that.  But

6  our assumptions can, then, be further refined, as we go

7  through the process of doing a first phase and a second

8  phase of a risk assessment, so that, in the end, we

9  can, are in the best position to characterize what, in

10  fact, we think our quantitative assessments reflect the

11  risk of, in this case, only SO2, SO2 in combination

12  with other pollutants, SO2 as a surrogate for other

13  pollutants.  All those things are things we are, in the

14  end, going to have to speak very clearly to.  So the

15  more you help us, at this early stage in the game, with

16  some of your thinking at this stage, and recognizing we

17  can further refine that as we go about characterizing,

18  in the end, the results that we do produce.

19	Beyond that, the issue of multi-pollutant

20  standards and multi-pollutant strategies, obviously,

21  has much broader implications for all of our NAAQS

22  reviews, and for what the Office does in implementing

23  programs to address the NAAQS.  And I would just make

24  the observation that, some of the comments I heard, I,

25  perhaps, unintentionally, have the, sort of sounded



Page 83

1  pollutant approaches, and multi-pollutant standards,

2  and multi-pollutant interpretations of scientific

3  evidence, and the question is, the  tangentially

4  related to the subject we're here to talk about, our

5  plan for doing exposure and risk assessment, but it's,

6  obviously,  more broadly related to our ultimate review

7  of these primary and of two standards, and our review

8  of standards in general.  And I just wanted to take a

9  moment and make a few observations about the

10  discussion, and sort of, our view of it.

11	In the context of science assessment

12  documents, it's clearly extremely important that those

13  assessment documents do everything they can to tease

14  out, what do we know about any individual pollutants

15  effects, and what do we know about the interactions of

16  that pollutant with other pollutants, and to what

17  extent can we define specific effects related

18  individual pollutants versus to what extent does the

19  evidence limit us to only making more general

20  observations about associations of air pollution more

21  broadly.

22	All those issues are extremely important, and

23  your discussions are helping, I think, to sharpen the

24  science assessment document in that regard.  It becomes

25  important for us, of course, at least as an initial

Page 85

1  like, there is an inherent mismatch between setting

2  standards for individual pollutants, and crafting

3  control strategies that most efficiently and

4  effectively get at the mix of pollutants that are, in

5  fact, of concern.

6	And I would offer the observation that I

7  don't, really, perceive that to be a mismatch.  It's

8  clearly a distinction, but one can, clearly, have

9  standards for individual pollutants in conjunction with

10  air quality management programs that are very multi-

11  pollutant oriented, and seek to find the most efficient

12  strategies for addressing all the pollutants for which

13  we have standards.  And, you all, I mean, different

14  people have different views on that, but I just wanted

15  to make the observation that there isn't, necessarily,

16  an inherent mismatch, or contradiction in doing those

17  things.

18	The one pollutant that we truly have a multi-

19  pollutant standard for is, of course, particulate

20  matter, which is a collection of thousands of

21  pollutants.  And if you think about it, we have

22  established that as a multi-pollutant standard under

23  the guise of PM mass, and a great deal of the research

24  right now is focused on trying to tease out what are

25  the differing relative toxicities of the individual



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 23

Page 86

1  components within that mix.	So, when we

2  do find ourselves faced with what truly is a multiple

3  pollutant standard, what we set about to do is trying

4  to figure out how to separate it out.  And I think

5  that's sort of informative as to what utility there

6  might be in trying to aggregate all the other

7  pollutants into one standard, with interaction terms,

8  wouldn't our next step, logically, be trying to sort

9  out the relative toxicity.

10	So, I just wanted to offer those

11  observations, in terms of, the issue is really an

12  important one, but perhaps, it's not as much of a

13  mismatch or contradiction as one might originally

14  think.

15	Those were the points, observation points I

16  wanted to offer before we get into it, so if there's

17  nothing else we need to deal with, why don't we just

18  jump in to the overview presentation we wanted to make.

19	DR. HENDERSON: I think that's, that

20  would be good.  Donna has a question.

21	DR. KENSKI: Well, just a response, I

22  guess, to the idea that single pollutant standards

23  don't, necessarily, preclude multi-pollutant controls.

24  Well, I, you know, it's clear that, you know, a control

25  on one pollutant will almost always, you know, have an

Page 88

1  states, it's the Saint Louis area.  And so, we, in

2  fact, are very mindful of that issue.  As far as the

3  planning goes, I think Karen is addressing, you know,

4  the way in which we set standards right now, which does

5  not, in fact, preclude multi-pollutant planning.  And

6  we can set-, you know, at this point, we do need to

7  look at the pollutants individually, but that doesn't

8  prevent us from moving forward to multi-pollutant

9  planning.  And that's, definitely, what we are trying

10  to do.

11	DR. HENDERSON: Thank you.  And so,

12  Karen, are you going to, who is our first speaker for

13  the - -

14	DR. MARTIN: Scott's going to take the

15  lead in covering the opening, and Stephen and Harvey

16  will round out the opening presentation.

17	DR. HENDERSON: Okay.

18	DR. JENKINS: Okay, thanks.  My name is

19  Scott Jenkins, and I'm the health lead for the NO2

20  review and OAQPS.  And I'm going to be talking through,

21  probably, three or four slides on giving a little bit

22  of background on the current approach that we have

23  proposed in the scope and methods plan.  And then,

24  Stephen is going to talk through the exposure part of

25  it, and Harvey is going to talk through the risk part



Page 87

1  effect on other pollutants, it still, you know, the

2  burden on the states to comply with the single

3  pollutant standard requires that they produce a, you

4  know, state implementation plan.  And in that plan,

5  they have to provide for how they're going to control

6  that single pollutant, not multi-pollutants.  So, I

7  think the, you know, it would, while we get these, sort

8  of, indirect, you know, controls on other pollutants,

9  it would be more straightforward, I think, to, you

10  know, have a multi-pollutant approach that really, you

11  know, dictated this, you know, need to control all

12  pollutants, not just single pollutants.  So, while, you

13  know, while, yes, we do get controls, still the burden

14  on states is to demonstrate their control of a single

15  pollutant for a single standard.

16	MS. WEGMAN: Yeah, and I'll just respond

17  briefly.  We are very mindful of the air quality

18  manager port that the NRC issued, and we did, there is,

19  in fact, a subcommittee of the Clean Air Act Advisory

20  Committee that has looked at all the recommendations

21  coming out of the air quality management report,

22  including the one to develop multi-pollutant plans, and

23  we, in fact, have a project going on in our office to

24  pilot multi-pollutant planning with three states, North

25  Carolina, New York, and Illinois, Missouri, four

Page 89

1  of it.

2	Okay, so we're all aware, by now, that our

3  purpose here to solicit feedback on our proposed

4  approach to assessing risks and exposures.  I just went

5  through this, so I'm going to talk through, a little

6  ahead on the schedule a little bit, talk about the

7  previous review, and give a little bit of what, I hope,

8  is historical perspective.  And then, talk about the

9  scope of the plan.

10	Okay, first schedule, and Mary presented the

11  same slide yesterday, so I'm not going to go through

12  anything in detail, other than to point out that the

13  next time we will be soliciting, or will be meeting

14  with CASAC will be Spring of '08, where we'll be asking

15  for feedback on the second draft of the ISA, and the

16  first draft of the risk and exposure assessment, and

17  then, again, September of '08, for the second draft of

18  the risk and exposure assessment.  And then, the final

19  date here, this is our, the date that we anticipate

20  will become our court ordered date.  This is about five

21  months earlier, just to point out, five months earlier

22  than the dates we had originally proposed.

23	Okay, so a little bit of background, and

24  this, and I'm going to expand on this side a little

25  bit, just based on the conversation that we just had



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 24

Page 90

1  prior to the break, and that is, providing a little bit

2  of historical perspective for how the original standard

3  was set.

4	So, the original standard was based on

5  epidemiology studies that were conducted in

6  Chattanooga, Tennessee, basically, where the long-term

7  annual average levels of NO2 were correlated with

8  health effects.  The issue that arose later with those

9  studies was that the issue of confounding with other

10  pollutants, an issue of the measurement approach to,

11  for, from measuring NO2 in the studies.

12	So, what happened was that, the original

13  standard was set based on those long-term epi studies.

14  And then, every review since then has focused,

15  essentially, on the short-term issues.  And the crux of

16  the decisions that the Administrator has made are how

17  well does that existing long-term annual standard

18  protect against short peak exposures.

19	So, I think that'll become clear when I go

20  through the slide a little bit, but that just gives you

21  a little bit of a historical context.

22	So, and their talk-, specifically, about the

23  last review of the NO2 NAAQS, and the Administrator had

24  a, made a couple of conclusions regarding the

25  sufficiency and the necessity of the existing annual

Page 92

1  arrived at was that we conduct-, we and OAQPS

conducted

2  an air quality assessment, essentially, evaluating the

3  relationship between the annual average NO2 levels and

4  short-term one-hour average NO2 levels.  And, as part

5  of that evaluation, we looked at the number of

6  exceedance as a very short-term benchmark values, with

7  the assumption that were just meeting the current

8  standard.

9	So, those benchmark values were derived,

10  again, from these clinical studies that I just

11  mentioned, and, basically, the result was that, if you

12  assume that the existing annual standard is being

13  attained, the short-term levels of NO2 of potential

14  concern would be very unlikely in most parts of the

15  country.  I think Los Angeles had a few exceedances at

16  the .2 ppm level, but that was the only spot where

17  those exceedances were found.

18	So, that was, this was the structure of the

19  la-, the con-, of the decision framework for the last

20  review.  And now, I'm going to move to the current

21  review, and talk just for just a minute about the scope

22  of the planned risk and exposure assessment, and then

23  I'm going to turn it over to Stephen.

24	We hit on this a little bit yesterday, this,

25  using NO2 versus other oxides of nitrogen.  Obviously,



Page 91

1  standard.  The first conclusion was that the existing

2  annual standard will maintain annual NO2 concentrations

3  well below levels, long-term levels that are of

4  potential concern.  And those long-term levels of

5  potential concern were derived from the animal tox

6  literature.  And this is, basically, a fi-, based on

7  findings that if you expose animals for  relatively

8  long periods of time to relatively high levels of NO2,

9  you get emphysema-like lesions in the lung.  And then,

10  we're talking about, at least months of exposure to,

11  say, at least 5 ppm NO2 here.

12	So, it's pretty much, it's pretty easy to see

13  that, yes, the existing annual standard of .053 ppm

14  will protect against those sorts of long-term effects.

15	The other conclusion, and this is, really,

16  more of the focus of the last review, the other

17  conclusion was that the existing annual standard will

18  provide protection against the short-term peak NO2

19  levels that are of concern.  And those short-term

20  levels of concern were derived from the human clinical

21  literature.  This came from a set of studies showing

22  that, in asthmatics, if you expose asthmatics to levels

23  as low as, say, .2 to .3 ppm NO2, you can get increased

24  airway response in this.

25	So, the way that the second conclusion was

Page 93

1  NO2 is but one of the oxides of nitrogen that include

2  both gaseous and particulate species.

3	There are really two issues, I think,

4  embedded in this.  And that is, first, that is using

5  NO2 as a surrogate for the gaseous nitrogen oxide, and

6  the other issue is using, is focusing this review on

7  the gaseous nitrogen oxide.

8	So, regarding the first, the first statement,

9  you know, and I think this was borne out yesterday,

10  we're thinking-, we're planning to use NO2 as a

11  surrogate for the gaseous species, basically, because

12  the lack, relative lack of health effects data, and

13  actually, it came out yesterday, also, the relative

14  lack of monitoring data for gaseous species other than

15  NO2.

16	In the case of the particulate, the second

17  point that I made, the particulate nitrogen oxide, and

18  we made this point in our integrated review plan, and

19  we, this point, also, came up at our last consultation

20  with you last spring, that the particulate species are

21  addressed by the current NAAQS, and the rationale

22  provided right here, basically, the last review for

23  the, of the PM standard concluded that size

24  fractionated particle mass, rather than chemical

25  composition was the most appropriate way to address



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 25

Page 94

1  ambient PM.

2	This conclusion, obviously, is going to be

3  reassessed in the next review, or I should say, the

4  current review, since it's already kicked off. We had

5  the opening workshop.  But at present, it would be

6  redundant to, also, use the NO2 NAAQS to protect

7  against the health effects of particulate nitrogen

8  oxide.

9	Other than that, I just want to point out

10  that the assessment is going to evaluate, this will be

11  a recurring theme throughout Stephen and Harvey's part

12  of the talk, that the current assessment, we're

13  planning to assess both recent ambient levels of NO2,

14  ambient levels that are associated with just meeting

15  the current standard, and ambient levels that are

16  associated with just meeting the potential alternative

17  standards, which will be identified as we move forward.

18  And the assessment's going to focus on both short- and

19  long-term exposures.

20	So, that's all that I had to say in the way

21  of background and introduction.  I'm going to turn it

22  over now to Stephen, who is going to talk us through

23  the proposal, proposed plan for the exposure

24  assessment.

25	DR. GRAHAM: Thank you, Scott.  Could you

Page 96

1  fashion, that is going from, in a sense, a qualitative

2  evaluation and progressing to a quantitative evaluation

3  if, of course, data exists to support that type of an

4  evaluation.

5	So, the tier one, as I mentioned, is a air

6  quality characterization, and the purpose there is to

7  estimate the potential exposures, using the current, as

8  well as historical air quality data that we have

9  available to us, and use that as a surrogate for

10  exposure.  In addition, we are proposing to take a

11  glance at some of these near roadway exposures, using

12  the ambient data, using enhancement factors.  And then,

13  of course, any available concentration data and

14  emissions data that may be available to look at the

15  influence from sources, particular sources that may be

16  outdoors or indoors.

17	The locations that we considered are

18  outlined, based on the, those criteria that is, air

19  quality trans data availability, you know, number of

20  monitors, whether the data are quality assured and

21  comprehensive, and in addition, to some other criteria.

22  And we've selected Los Angeles, Houston, Atlanta,

23  Philadelphia, and Chicago, and possibly, aggregation,

24  based on some of the analyses that are going to be

25  performed here.



Page 95

1  work the clicker?  Sometimes I have a habit of talking

2  with my hands.  Okay, all right, thank you.  So, of

3  course, the general, broad goals of this exposure

4  assessment are to estimate both short-term and long-

5  term exposures, short-term being hourly, and that is

6  associated with these current levels of ambient NO2,

7  and assuming alternative levels of NO2.

8	Also, to develop these quantitative

9  relationships, based on the form of the current

10  standard, which is long-term, annual average and the

11  relationship between that average and the short-term

12  peak concentrations, which was done in the prior review

13  as well.  But in addition, I want to, also, consider

14  local source influences, which we saw was important in

15  the review of the ISA, and the impact on the exposure

16  estimates.

17	As far as the approach, it's already been

18  mentioned that we have three tiers.  The tier one is

19  air quality characterization.  And I'll go through each

20  of these in greater detail, I guess, or of course, it's

21  been in greater detail in the scope of methods

22  document.  Populations considered include the general

23  population, as well as the individuals identified as

24  susceptible or vulnerable.  And the assessment of

25  uncertainty is also going to be approached in a tiered

Page 97

1	The expected output is, of course,

2  descriptive statistics for NO2 in some of these

3  selected locations; relationships between the short-

4  term peak levels and the long-term average levels; and,

5  of course, identification of additional areas to be

6  modeled in the tier two and tier three, dependent on

7  the analysis outcome.

8	Uncertainty will, primarily, be qualitative

9  at this stage, and of course, these tier one exposure

10  assessment, the outcome is going to be used for

11  comparison with some of the health benchmarks, once

12  they are identified.

13	So, in tier two, we've got, the purpose is to

14  improve that relationship.  So, now, we're trying to

15  link the actual concentrations, themselves, to persons,

16  to humans.  And we are going to, of course, consider

17  both the on roadway and, as well as, near roadway,

18  using dispersion modeling and or enhancement factors as

19  well.

20	The model concentrations for other outdoor

21  sources, if there are any identified, as well as the

22  indoor sources, if they are identified as being

23  important in influencing these exposure estimates or,

24  shall I say, the relationship between, I'm sorry, the

25  contribution, the relative contribution between the



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 26

Page 98

1  two, that will be done as well.  And, of course,

2  consider influential factors, that is, time that people

3  do spend in particular microenvironments, as well as

4  the limited decay of NO2 indoors, and populations

5  residing near roadways.

6	The locations, of course, are going to be,

7  it's going to be a more focused analysis, and it's

8  going to be focused on the locations that have been

9  identified in the tier one analysis.

10	And the output is going to be broken up into

11  two different exposure metrics.  We've got short-term

12  exposure outcome, where we have, in addition to the

13  temporal and spatially resolved ambient air quality

14  concentration fields, that account for local sources,

15  like emissions from roadways and other sources that are

16  identified as important.  We've got estimates of the

17  number of individuals who may experience exposures of

18  concern.  Not to suggest that it's an individual

19  analysis.  It's more of a cohort-based analysis.

20	And then, of course, long-term exposure

21  estimates will include annual average exposure levels

22  within a given census tract, and could be considered at

23  a more finer resolution, say a block group or block.

24  And it's not just the annual, but also, I believe,

25  we'll be able to estimate the daily average.

Page 100

1	So, as I said, the locations are the same as

2  those identified previously.  The expected output is

3  going to be the counts of people exposed one or more

4  times to several NO2 levels, based on, of course,

5  health information obtained from the ISA, and because

6  APEX is a time series model, and the averaging time,

7  that is of interest, can be taken out of the, or shall

8  I say developed as an exposure metric.  And we also

9  have counts of personal occurrences of a particular

10  exposure.

11	And the uncertainty can be a little bit more

12  quantitative in a sense.  We can look at, again, model

13  inputs, where data exists for describing both a

14  variability and the  uncertainty in them, and of

15  course, model formulation.  If we have estimates of

16  personal exposure that are available to compare that,

17  as well as microenvironmental concentrations.  That's

18  it.  Thank you.

19	MR. RICHMOND: Thank you, Stephen.  I'm

20  going to walk you through a few slides on the risk

21  assessment.  First of all, overview goals of the risk

22  assessment are to estimate the number of occurrences of

23  short-term air quality events and number of people

24  exposed at, or above, various potential health effect

25  benchmarks associated with alternative NO2 scenarios.



Page 99

1	And, in addition, we've got the ratios of

2  exposure to ambient, which could be useful for

3  extrapolating to other areas that we had not modeled.

4  And uncertainty would be addressed, of course, through

5  various sensitivity analyses, limited sensitivity

6  analyses, based on input distributions and other model

7  inputs, as well as measured comparisons, if there are

8  data that exist for particular microenvironments.  That

9  would be compared to model estimates.

10	In tier three, of course, it's a more refined

11  approach, and here we are focusing on addressing more

12  particulars about human physiology, including time,

13  well, that's not physiology, but time, location,

14  activity patterns, and their physiology.  Using the air

15  concentration fields developed from a tier-two

16  approach, that is where we have the on and near roadway

17  concentrations, and using the EPA's APEX model for

18  estimating exposures.

19	Locations, of course, are built upon what had

20  been identified in the tier one and used in the tier

21  two analysis.  And, oh, I forgot to mention, of course,

22  the APEX model is capable of estimating individual

23  exposure estimates.  It's a time series exposure model.

24  And we also have capabilities to estimate indoor

25  sources using that model.

Page 101

1  If a tier two assessment is conducted, we'd also

2  provide health risk estimates for NO2 health endpoints

3  associated with alternative scenarios.  And for any

4  tier, we'll, of course, identify and characterize key

5  assumptions and the variability and uncertainty

6  associated with the assessments.

7	As Scott and Stephen have said, the scenarios

8  evaluated are for both recent air quality, simulating

9  the current standard, which is a difficult challenge,

10  given the levels are much lower than the current

11  standard; and air quality levels just meeting potential

12  alternative standards, which could be short- or long-

13  term standards.

14	There's a two tiered approach here.  Proposed

15  one is the, in tier one, potential health effect

16  benchmark levels, which will be based on a review of

17  the revised ISA, would be compared to, first, air

18  quality and then, exposure estimates generated by the

19  tiers that Stephen's gone through.

20	Tier two, if it's judged feasible, and

21  they're, and of sufficient utility for decision making,

22  we involve combining concentration response, if it's

23  based on epi; or exposure response based on controlled

24  human exposure data, with exposure estimates to

25  generate population risk estimates.  It's what we'd



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 27

Page 102

1  like to have, ideally.  We may or may not be able to

2  get there.

3	And next slide, okay.  The tier one, as I've

4  said, the air quality levels from a tier one exposure

5  assessment, or estimated exposure levels from tier two

6  or three,  would be compared to potential health effect

7  benchmark levels for several example urban areas.

8  Those would be the same areas that Stephen's talked for

9  the air quality and exposure tiers.

10	We have identified, very tentatively, in a

11  benchmark of, in the .2 to .3 ppm one-hour averaging

12  time range, based on the controlled human exposure

13  studies, of effects that have been observed in

14  asthmatic, both children and adult asthmatic.  There's

15  uncertainty about those health effect benchmarks that

16  we see, we'd be using alternative benchmark levels to

17  illustrate the impact of alternative choices about the

18  lowest exposure levels of concern.

19	In terms of variability, we address that by

20  doing the analysis in different geographic areas.

21  Population variability in response but it would have to

22  be addressed qualitatively.  We don't have, I think,

23  data to distinguish that very well.  And the projected

24  outcomes would be the number of  occurrences of air

25  quality levels at or above several benchmarks, or

Page 104

1  tier two assessment would be conducted, including the

2  outcome and insights gained from the tier one

3  assessments, both with the exposure and risk

4  assessment, and availability of information and data

5  required to conduct a tier two assessment on the

6  adequacy of concentration response functions, baseline,

7  and getting baseline incidents data for things like

8  hospital admissions and emergency department visits for

9  the example urban areas.

10	Then, the utility or value added to the

11  decision process beyond the insights provided by a tier

12  one assessment, and the feasibility of conducting the

13  assessment within the time constraints that we have.

14	Next slide.  Based on our preliminary

15  analysis of the first draft ISA, the most likely

16  candidate endpoints are listed here.  I think that was,

17  generally, in agreement with what I heard yesterday in

18  the discussion on ISA.  But the strongest evidence from

19  the epi would be for respiratory related morbidity

20  endpoints, including hospital admissions, especially

21  for asthmatics; respiratory related emergency

22  department visits; and respiratory symptoms, such as

23  cough and wheeze, particular in children and

24  asthmatics.

25	Risk estimates, if we do proceed to this



Page 103

1  number of times in a given year that a population or

2  individual experiences various exposure levels of

3  concern.

4	To our next slide.  And a tier two, if

5  conducted, would estimate the number of individuals in

6  selected populations for several example urban areas

7  expected to experience specified health effects more

8  similar to the ozone and PM risk assessments that we've

9  completed in the last couple of years.

10	We judged that it would be more likely that

11  would be based on the epidemiological literature.

12  Preliminary judgment is  that controlled human exposure

13  studies don't provide enough information to identify

14  credible exposure response relationships.  There's

15  enough information to judge benchmarks for the health

16  endpoints, but it's difficult to see how to get

17  exposure response relationships across the range of

18  interest.

19	We're still evaluating.  We're, obviously,

20  listened carefully to what you've said over the last

21  day and a half.  And look forward to seeing how the

22  revised ISA responds to those, in terms of whether

23  there's sufficient epidemiological evidence adequate to

24  conduct a credible quantitative risk assessment.

25	The criteria listed here for determining if a

Page 105

1  stage, we would propose to conduct both sing-, use both

2  single and multi-pollutant models.  And uncertainty

3  would be addressed similar to how we've handled in PM,

4  the statistical or sample size uncertainty, that we

5  would provide confidence intervals around point

6  estimates of risk, and representing a range of results,

7  based on different epidemiological studies.

8	Expected outputs are listed here, in terms of

9  we would look at estimated incidents that can express

10  the results in a number of different ways.  Incidents

11  per hundred thousand and or percent of incidents.  And

12  this would address hypothetical change in incidents

13  associated with moving from just meeting the current

14  standard to just meeting potential alternative

15  standard.

16	The final part of the risk characterization

17  is several things that we've tried to either put the

18  more limited example, like, you know, urban areas

19  analysis, is one to summarize U.S. air quality

20  information, and discuss the various health effects

21  that we were not able to quantify from the ISA.  So,

22  that would certainly be part of the exposure risk

23  report to provide context for those things that we do

24  deal with quantitatively, and would include those air

25  quality statistics for all air as the U.S. based on the



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 28

Page 106

1  NO2 monitoring data.  And that we'd also provide

2  information, national scale information on the size of

3  potentially susceptible or vulnerable populations as

4  part of that.

5	DR. JENKINS: Okay and this slide just

6  has the charge questions to CASAC.  This is actually a

7  condensed version of those charge questions, because we

8  actually had too many to fit on a single slide, but I

9  think this captures them.

10	DR. HENDERSON: Thank you.  Karen, do you

11  have more to present?

12	DR. MARTIN: I don't believe so, unless

13  there are some specific questions.

14	DR. HENDERSON: Okay, now, Lianne, you're

15  on the phone?

16	DR. SHEPPARD: I am.

17	DR. HENDERSON: When are you leaving?

18  I'm just trying to - -

19	DR. SHEPPARD: I have two more hours.

20	DR. HENDERSON: Oh, okay.  I was going-,

21  that's good, Lianne.  I just wanted to allow you to ask

22  questions if you were leaving.

23	DR. BALMES: Rogene?

24	DR. HENDERSON: Yes.

25	DR. BALMES: This is John Balmes.  I'm

Page 108

1  lot of concern about this morning.  But let's start in

2  on the discussion of the air quality section, which is

3  very brief in this report, and has lots of discussants.

4  So, we may have more discussion and, than there is

5  text, here.

6	SPEAKER: Than material.

7	DR. HENDERSON: But, what I'd like to do

8  is run through, you know, there's three, the air

9  quality section, the exposure section, and the risk

10  assessment section.  We're going to have, open the

11  discussion up to anyone who has any questions.  We

12  tried to group people as to their interest, rightly or

13  wrongly, so let's start out with, on the air quality,

14  Ellis, did you have some comments you wanted to make,

15  or you know.

16	DR. COWLING: I must say that the general

17  impression I have is that the approach is being, that

18  is being proposed is reasonable, and that I have

19  confidence that it will be pursued within the limits of

20  time available that were mentioned.  The five months

21  shorter time frame, I'm sure has caused some anxiety

22  within the staff about how to get all the things done

23  that they had hoped that they can accomplish.  But, I'm

24  satisfied that the approach being proposed is

25  reasonable.



Page 107

1  going to have to leave to teach in a few minutes, and I

2  will be returning in time for the health effects

3  discussion, which is currently scheduled for, what is

4  it, 2:15 your time?

5	DR. HENDERSON: We may be getting to that

6  a little earlier, because I plan to have a working

7  lunch.

8	DR. BALMES: Well, I just, I have to

9  teach, so I won't be able to join the call till a

10  little bit after 2:00 your time.

11	DR. HENDERSON: Okay.

12	DR. BALMES: Just so you know.

13	DR. HENDERSON: Okay.

14	DR. BALMES: I don't have a specific

15  question right now, though, just letting you know.

16	DR. HENDERSON: Okay, and if you, let us

17  know when you join in.  Okay, I found it fascinating

18  our discussion just before we started this about, you

19  know, the annual average standard and how that relates

20  to peak exposures.  And this group has, obviously,

21  addressed that, and I, in a statement, if the existing

22  annual standard is obtained, short-term NO2 levels of

23  potential concern would be unlikely in most parts of

24  the country.

25	That, would, addresses some things we had a

Page 109

1	DR. HENDERSON: Thank you, Ellis.  Donna,

2  do you have any comments on this, the proposed methods

3  for air quality evaluation?

4	DR. KENSKI: Well, I guess I have a

5  question.

6	DR. HENDERSON: You need to get close to

7  your mike.

8	DR. KENSKI: Okay, sorry.  In answer to

9  the, I guess, the charge question about whether it was

10  appropriate to use historic data, I thought that was

11  the logical approach.  I, the question I had was in

12  the, how you're modeling expected exceedances, and it

13  gives an exponential model here, and I just wondered if

14  there were any discussion about that choice of model,

15  and whether you considered other models.

16	DR. GRAHAM: That's a good question,

17  thank you.  We've got, that, actually, had been used in

18  the previous review.  So, as I was looking to, I guess,

19  duplicate that effort, but in addition to that, we are

20  also going to look at an alternative model that looks

21  at, like a logistic regression, so it'd be more

22  probability based.

23	DR. KENSKI: Okay, yeah, I think that's

24  appropriate.

25	DR. GRAHAM: Thanks.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 29

Page 110

1	DR. HENDERSON: Is that all, though, or

2  did you - -

3	DR. KENSKI: Oh, well, another question

4  about your choice of, and I'm not sure if this really

5  fits in, maybe it does.  Your choice of cities, New

6  York wasn't on this list, and that seemed odd to me

7  that, 'cause it was one of the higher, definitely one

8  of the higher concentration cities.  So, if you're

9  looking for peak exposures, it seems like you'd choose

10  those urban locations, and it has lots of monitors.

11	DR. GRAHAM: Right, I think Philadelphia

12  was selected over New York, per se, because of, it was

13  representative of a northeastern region, but the key

14  feature there was the availability of additional data,

15  including very refined roadway counts, and other data

16  that had been developed previously through other

17  research.  So, I thought it would be a slam dunk, per

18  se.

19	DR. KENSKI: Okay.

20	DR. HENDERSON: Is that all?  Tim Larson,

21  are you on the phone?

22	DR. LARSON: Yes, I am.

23	DR. HENDERSON: Okay, did you have

24  comments on the methods for the air quality section?

25	DR. LARSON: Yes, a couple.  It wasn't, I

Page 112

1  I guess, the question is, how do you see that going

2  down?

3	MR. RICHMOND: Okay, this Harvey

4  Richmond.  Let me try to address that.  In, if we can,

5  the risk assessment has said, at first one tier,

6  there's, if we're looking at results of either the air

7  quality in tier one or exposure, either from a tier two

8  or tier three, either APEX or otherwise exposure

9  estimates, we're comparing, we're using all of the air

10  quality information.

11	We're using the monitoring, but also,

12  enhancing that with additional information to try to

13  estimate, either, a surrogate for exposure or getting

14  the distribution of exposures.  That's then going to be

15  compared with a health benchmark levels that are based

16  on the controlled human exposure studies.

17	So, we're trying to match exposure with an

18  exposure response or an exposure, you know, an effect

19  observed in a clinical setting.  That's one use of it.

20  And separately, if the epidemiology is deemed that it's

21  sufficiently one that is likely causal, or you know,

22  whatever we decide to go down on that continuum of

23  causality, that we're going to quantify, if we quantify

24  an effect from the epidemiological literature, those

25  studies, I agree with you, are based on the ambient



Page 111

1  guess, I have to calibrate my thinking here.  The

2  short-term standard, the health effects for that are

3  based on the analysis of existing monitoring sites for

4  short-term NO2 levels, I assume.

5	Is that the, I mean, is, it seems to me that

6  the relationship between the short-term values and the

7  long-term values, and how you do that, depends on what

8  sort of the health basis you're deriving, using to

9  derive that short-term value.  If it's the epi, then

10  that's one thing, and then, would you just, sort of,

11  use the statistics from existing sites.

12	If it's independent toxicology clinical human

13  exposures, et cetera, then it seems like, potentially,

14  there are different relationships between the long-term

15  and short-term values for parts of urban areas that may

16  be more relevant than ones at the monitoring sites.

17  So, I'm not clear, it's not clear to me which is the

18  basis for your health risk assessments.

19	To the extent that it's the epi, then I

20  suspect that the existing monitoring statistics are

21  relevant.  To the extent it's other, then I would

22  suggest that the relationship, I mean, in the extreme

23  case, for instance, the one-hour peak exposure is while

24  you're commuting, which has nothing to do with any of

25  the statistics or distributions at the monitoring.  So,

Page 113

1  monitoring fixed site monitors, usually the average

2  across several monitors.

3	And if we were to do a risk assessment, based

4  on epi, it would not be based on the exposure analysis.

5  It would be based on the ambient fixed site monitors.

6  I hope that clarifies it.

7	DR. LARSON: Yeah, that helps, thank you.

8  Well, at least to the extent that it's based on the

9  human clinical studies, I would suggest caution here

10  using your near roadway or traffic related impact as

11  models, because I think you get very different

12  relationships spatially and temporally in a flat road,

13  than you would in a built up urban area.  So, I mean,

14  the residential areas of Philadelphia or parts of

15  Chicago or Houston maybe that's, your approach is fine.

16	But we're doing studies in Chicago, and I

17  know for a fact that, is part of our cohort lives in

18  downtown Chicago, and we can't and don't have any

19  success using those kinds of approaches in that area.

20  So, if there was some way to qualitatively screen those

21  parts of the urban areas that are subject, is more

22  complicated confinement effects, so that you would

23  limit your exposure assessment in some way, based on

24  that.

25	I would think that's more defensible.  You



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 30

Page 114

1  get a qualitatively different in a distribution and

2  it's not only, then becomes determined by the geometry

3  of your urban area, but also you end up with this

4  question about the vertical distributions within those

5  confined areas.

6	And then it, then you  really up, you really,

7  you know, it's an open question as to where the,

8  whether you've got buildings with open windows, or

9  whether you've got inlets of a building in a

10  residential apartment on the roof or on the ground or

11  whatever, and it gets real complicated in a hurry.  And

12  the, I guess, the basic point is that a simple, sort

13  of, Gaussian dispersion model do not have much skill in

14  those parts of the urban area.

15	So, I mean, maybe it's a qualitative way to

16  assess, you know, there are methods for doing that.

17  You know, you can look at building footprints, and

18  overlay building heights.  We've done that in New York,

19  and we're doing it in Chicago.  And you can, sort of,

20  identify the areas that are, you can just look at

21  Google, actually, and probably do the same thing, but

22  more quantitatively, you could do it that way.  But I

23  would say your proposed approach air mod, et cetera, is

24  fine as long as it's, sort of, single family

25  residential, but otherwise, it's questionable.

Page 116

1  the exposure field as to what they think about that.

2	DR. HENDERSON: And that brings me to a

3  point that I had meant to bring up earlier.  For all of

4  you who have not yet submitted your individual comments

5  on this methods document, that's just essential in this

6  case, because the letter will not, it will be more pro

7  forma when it's not going to list the consensus.  So,

8  the advice that you want to give to the Agency in this

9  consultation will come in the form of your individual

10  written comments, so that increases their importance.

11  Tim, is that, are you, have you completed your comments

12  on the air quality section?

13	DR. LARSON: Well, I, yeah, I think

14  that's, wait a minute.  It certainly, EPA, I think, is

15  sponsoring some of the work I mentioned, so, you know,

16  we certainly can do our best to work with the, provide

17  whatever information we have.  We have quite a bit of,

18  Lianne is involved with this, too.  We have quite a bit

19  of NO2 passive monitoring data, saturation data in

20  several cities that we're talking about.

21	DR. HENDERSON: I'm sure they'd

22  appreciate having all the information you can give

23  them.  Ted, would you - -

24	MR. RICHMOND: Can I just ask one

25  clarifying question?



Page 115

1	DR. HENDERSON: Do you have a response to

2  that, I mean, to that, Harvey?

3	MR. RICHMOND: One thing is I'll note

4  historically, it's one of the reasons why there hasn't

5  been an NO2 exposure analysis in the reviews that I was

6  involved in '85 and '93, is doing this is much more

7  challenging than doing an ozone exposure assessment.

8  And we acknowledge that.

9	I think we're, you know, we're saying, we're

10  trying to push the envelope as far as we can, and we're

11  still in the learning phases as to how far that is, and

12  we'd certainly be interested in the kind of information

13  that, Tim, that you've cited that, you know, if we, for

14  some of the example cities, if you have relevant

15  information, or ideas on how best to do it, or if we,

16  simply, you think if the advice of this committee is

17  we, simply, aren't able to credibly do certain parts of

18  the analysis.

19	Obviously, from a public health standpoint,

20  you'd be interested in those levels in those places

21  where you're saying it's most difficult to conduct the

22  assessment.

23	So, that's the challenge we face, and this is

24  a general road map.  We'll see as we get into it, but

25  we look to the advice of this committee and experts in

Page 117

1	DR. HENDERSON: Sure.

2	MR. RICHMOND: Tim, that saturation data,

3  is it for short-term averaging?

4	DR. LARSON: It's, no, it's two-week

5  average - -

6	MR. RICHMOND: Two-week average.

7	DR. LARSON: Yeah, so that's a problem,

8  but, yeah, it's a problem.  But it - -

9	MR. RICHMOND: Okay, I just wanted to

10  clarify it.

11	DR. LARSON: It can identi-, I mean,

12  we've done models, though.  We, in New York,

13  specifically, we've done, we've implemented the OSPM

14  model for New York City, which is kind of an

15  interesting exercise.  And, to do that, you have to do,

16  or you have to have information on building footprints.

17  It's a massive undertaking, but so, we have, we have

18  hourly predictions compared to our measurements.  And

19  they compare pretty well.

20	I mean, it is reasonably, if you expect to

21  model that.  So, they do have skill, and they, and we

22  do have predictions on an hourly basis, both as an

23  urban background model, and superimposed on that is a,

24  is an OSPM model, and using Mobile Six and the traffic

25  - - model for New York City.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 31

Page 118

1	DR. HENDERSON: Okay, let's move on to

2  Ted.

3	DR. RUSSELL: Okay, very much like the

4  prior individuals, I was, generally, pleased with the

5  scope document.  One of the things, I think, that comes

6  out of just recognizing that you're going to be doing a

7  number of these, if you look at it more than one a

8  year, to get it down to a well-oiled, sort of,

9  approach, is one of my first recommendations.

10	And I realize you all are, may not have tons

11  of resources or whatever, but just something that could

12  be done such that these things become very automatic.

13  You know, I look at where we've come in other areas of

14  the modeling world, and I think that this could be done

15  when you're looking at exposure and risk analysis as

16  well, just to make it so it's not as, such a huge

17  effort every time.

18	A few things, one of them was, in regards to

19  what you, when you were presenting this morning, you

20  said something about locations considered, and you have

21  the five cities, and then, an aggregation of others.

22  And then you said, possibly, or maybe you said possibly

23  before you said that.  And I, it brings back what we

24  discussed in the lead panel was that, we're, I think,

25  as a whole, more interested in the U.S. than in

Page 120

1  read this, it wasn't just peak exposures.  It was all

2  exposures.  So, I'm just hoping that you're looking at

3  doing the exposure analysis at a national level, not

4  just for five cities, both peak as well as long term.

5	DR. GRAHAM: The air quality data is

6  going to be evaluated nationally.  When we look at the

7  focused exposure analyses, that's going to be on

8  individual cities that had been identified.

9	DR. RUSSELL: I think we would have some

10  interest in, maybe I'm not, maybe I'm singular here, of

11  really the interest at a national level somehow

12  extrapolating or doing something to get, to give us an

13  idea of what's happening nationally, not just at those

14  cities.  Because it came up in the lead.  It came up in

15  the ozone, as well, that that would be some important

16  information.

17	And, let's see, one, just a minor comment, in

18  your model, in terms of how to go away from roads use-

19  -

20	DR. MARTIN: Just, if I may?

21	DR. RUSSELL: Yes.

22	DR. MARTIN: Just to come back to the

23  point you just made.  I can't not make the observation

24  that, as we move to exposure modeling, what we and you

25  both are looking for is to develop enhancements to



Page 119

1  individual cities.

2	I'm just wondering how you can, how you would

3  plan not to do an aggregation of others, because I

4  think that would be somewhat more instructive to us, to

5  show us, on a national basis, the exposures of concern.

6  So, do you want to respond to that, how you're going to

7  make that decision not to do an aggregation of others,

8  or?

9	DR. GRAHAM: Delete possible.

10	DR. RUSSELL: Okay.

11	DR. GRAHAM: No, the intent was, in the

12  original analysis, they had done Los Angeles as a

13  separate area, and everything else was just lumped

14  together.  And here I was proposing, okay, we can do

15  multiple locations, and if it's of value, we can look

16  at these other locations as well.  And the criteria, in

17  a sense, for selecting the individual areas is based on

18  the fact that, we do have some information on the fact

19  that there are more peak occurrences.

20	In these other locations, there may not be.

21  So, in a sense, the model for predicting peak exposures

22  over a particular level, it just may fall apart,

23  because there are no peak exposures.  So, that's why I

24  say possible.

25	DR. RUSSELL: Okay, though, the way I

Page 121

1  those models to be, to address the location-specific

2  issues that are so central to understanding exposures

3  to NO2.

4	Once you zoom out and now say, oh, let's do

5  that on a national scale, you lose all your local

6  specificity, which is why we would approach a national

7  look on the basis of air quality, recognizing it's a

8  pretty gross approximation, and looking on the local

9  level, to try to tease out the nuances of exposures

10  around roadways and building canyons and those sorts of

11  things that we couldn't possibly do on a national

12  scale.

13	So, I mean, we're not trying to be resistant

14  to say, yes, it would be nice to know exposures on a

15  national scale, but I think the best we can reasonably

16  do is to do the more generalized air quality, look on

17  the national scale and to tease out the details

18  locally.

19	DR. RUSSELL: Dale has a comment, I

20  think, in response.

21	DR. HATTIS: Yeah, but, yes, essentially,

22  you've got three legs of a parallelogram approach here,

23  at least, in your plan.  You've got the local

24  assessment of air quality.  You've got the local

25  assessment of exposure.  You've got the national



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 32

Page 122

1  assessment of air quality.

2	I think it's not too much to hope that you

3  apply the lessons from the comparison of the local

4  assessment of air quality and the local assessment of

5  exposure to make, at least, a preliminary national

6  projection of the exposures, if the regularities you've

7  observed in your fully analyzed cases apply to some

8  portion of the national air quality data.

9	I mean, you, probably, wouldn't apply the

10  Philadelphia comparison directly to South Dakota, but

11  you, probably, want to apply it to some portion of

12  South Dakota, maybe, in a fraction of Fargo or

13  whatever, and, to some extent, get.

14	But anyway, that's the basic idea, is that

15  it's, it might not be a tremendous expense and of

16  effort to do that distributional projection for an

17  appropriate fraction of the country, or the country as

18  a whole, you know, even though you don't want to do,

19  in detail, the country as a whole.  You can get an

20  approximation from that, from the comparison.

21	DR. MARTIN: And then, of course, you're

22  left with making the judgment, does the approximation

23  so assume away all the details that are important that

24  you're left with, clearly we could create numbers, but

25  would they be meaningful.  And if they're not going to

Page 124

1  as opposed to standing by the roadside.  So, I guess

2  that I'm thinking that a lot of your, most of your key

3  data that's going to explain health effects are going

4  to come from understanding local variability.

5	And I, and it's where I'm going with this is,

6  is the, you need to associate locally, but you'd

7  probably have to have wide, a careful thought and

8  discussion about the variances even locally, 'cause

9  you're only very, barely touching that, that part of

10  the parameter.

11	Sounds to me like the, if you're doing a near

12  roadway, for example, comparison, you might have

13  various people that are experiencing that for which,

14  that have levels that are many, many times others that

15  are in the same environment because of the way they,

16  the way they were exposed to it.  And so, I'm curious

17  how you'd handle that variability as you let that

18  average out.

19	And that's just, actually, taking exactly

20  your comment and taking it the other direction, even

21  more extreme at the local level that we, actually, need

22  to understand the health effects.

23	DR. GRAHAM: Well, that is part of the

24  plan.

25	DR. CRAPO: Yeah, I like that part, that



Page 123

1  be meaningful, then we would argue, they're not worth

2  creating in the first place.

3	So, that's what we have to deal with, and we

4  can clearly look at it, but that's always the tension,

5  as you point to with lead as well.  And, I didn't want

6  to just slip by this and so.

7	DR. CRAPO: I wanted amplify what you're

8  saying, and maybe, even take it further the other

9  direction, which is that, your data on local data is

10  really still coming from your primary monitoring

11  stations, I assume.  That's correct?  I mean, your,

12  like your, near roadway monitoring stations, and things

13  like that, gives you local data that you'll use for

14  looking at some of the variation at the more local

15  level.

16	But, in fact, what we're learning, as we, or

17  the mornings is, or what we're learning is that there,

18  the local level has even tremendous variability within

19  that.  There's each gra-, I mean, gradients across the

20  roadway are falling away as a, you know, being 10  feet

21  versus 100 feet versus 1,000 feet for roadway has a

22  huge impact on the levels.  And even being in a car

23  dramatically changes the level.

24	I was told during the break that the school

25  buses are 4.6 times higher than ambient inside the bus,

Page 125

1  you had one-hour average, I just really compliment you

2  on the fact that your, you did have the one-hour

3  averaging goals in there.

4	DR. GRAHAM: Well, and to take into

5  account, specifically, roadway, on roadway and within,

6  given buffer distances of the roadway.

7	DR. CRAPO: And what about inside cars?

8	DR. GRAHAM: Sure, on road equals in

9  vehicle.

10	DR. CRAPO: 'Cause that could've, that

11  could, actually, do you have, are you going to have

12  personal monitoring measurements to give you that data?

13	DR. GRAHAM: No, based on modeling.

14	DR. CRAPO: You're just going to model

15  it, okay.  That would be key to this whole thing.

16	DR. HENDERSON: Okay, Ted, did you have

17  more to - -

18	DR. RUSSELL: I think that was,

19  primarily, it, and I'm not sure where we ended up on

20  it.

21	DR. HENDERSON: Well, I heard that we

22  need the local data to be able to relate it to health

23  effects.  I mean, that's, to me, that's, but - -

24	DR. RUSSELL: Right, I think we want the

25  local data, and, just, when we sit here and talk about



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 33

Page 126

1  things later, I think we're going to want to know what

2  are the national implications.  And maybe I'm one of

3  those who says, if you give me an approximate number

4  and say it's really approximate, that's a fine thing.

5  You know, I don't, that doesn't bother me, because I

6  know what I'm basing my discussions on.

7	That is, it's a number and there's

8  significant uncertainty, but it, at least, it's

9  something that gives me an idea of what's happening

10  nationally.  So that was the, that's my major concern

11  there.  And, let's see.  Also, one of the other issues

12  that I, or thoughts I had on this was, how do you plan

13  to provide ambient versus total exposure risks, total

14  exposures and related risks in your assessment, 'cause

15  I think having that comparison would be insightful.

16	MR. RICHMOND: Well, in the past, and

17  we've addressed this in the CO exposure model, where we

18  included passive smoking in gas stoves, we're able to,

19  since it's driven by a model, the exposure part, at

20  that tier of the assessment, to both report total, as

21  well as, just with the ambient.

22	In other words, basically, turn the indoor

23  sources off in the model, and how much is the ambient,

24  both including ambient outdoors and the ambient that

25  penetrated indoors, but in the absence of those indoor

Page 128

1	DR. GRAHAM: In a sense, it's no longer

2  supported, so it may have been used traditionally, and

3  it may look like a reasonable approach now.  I guess,

4  I'm trying to think of the future.  And AERMOD is

5  actually a little bit more advanced.  It's based on

6  boundary layer theory versus stability classes, and has

7  additional capabilities addressing turbulence and

8  meandering.

9	And, I know that the AERMOD doesn't have a

10  line source option right now, but it is something that

11  is being considered in the near future, not in time for

12  this particular review, but it will have that

13  capability.  Right now, what we are proposing is to do

14  link-based emissions, so, and I think it is being

15  applied right now in New Haven, Connecticut, and there

16  has been a paper published recently using CALPUFF,

17  which is a similar type of dispersion model, to do this

18  near roadway estimation.

19	So, it's, I think, not an unreasonable

20  approach, and there will be, I guess, portions of it

21  that, of CALINE that may be investigated.  I think you

22  had also mentioned in your comments earlier about the

23  conversion from NOx to, or shall I say NO to NO2.  So,

24  that may be an important feature there.

25	But, again, it was to look for, look towards



Page 127

1  sources.  So, we do try to provide that perspective of

2  how much is due to the ambient problem, as opposed to

3  indoor concentrations as well.

4	DR. RUSSELL: And that will be included

5  in this?

6	MR. RICHMOND: Yes.

7	DR. HENDERSON: Okay, Christian, do you

8  have?

9	DR. SEIGNEUR: Yes, I only have one point

10  I want to address.  It's, when it does a tier two

11  exposure assessment and the use of the model used to

12  calculate the air quality concentrations, in your

13  document, you mentioned you plan to use AERMOD.

You,

14  also, mentioned the model CALINE4.

15	My understanding is that AERMOD was the route

16  for stacks, dispersion of protons from stacks.  CALINE4

17  is most specific to roadways.  So, could you clarify

18  why you're planning to use AERMOD other than

CALINE4?

19	DR. GRAHAM: Absolutely.  While it had

20  been recommended to me that I use CALINE for, I'm

21  sorry, AERMOD for few reasons, CALINE, from what I

22  understand is, the developer of that model had recently

23  retired, and Air B has no initiative to continue on

24  developing that model.  So, we can consider that an

Page 129

1  the future, and, in addition, the fact that we are,

2  also, going to look at additional sources, some of

3  which are stationary type sources.  So, why not use one

4  model to head all the emission sources.

5	DR. SEIGNEUR: Okay, yeah, that will be

6  fine.  My recommendations, though, would be that if

7  you're going to use AERMOD for roadways, that it would

8  be evaluated prior to, with data graded near roadways

9  prior to its application.  'Cause EPA, typically,

10  requires people to evaluate the models before they are

11  applied.  So, in this case, since AERMOD has not been

12  formerly evaluated for roadway application, that, you

13  know, EPA would do that.

14	DR. GRAHAM: Right, yeah, and I did

15  forget to mention that, that AERMOD is the recommended

16  model, at least for dispersion.

17	DR. HENDERSON: Now, does anyone else

18  have something they want to comment on the air quality

19  section, 'cause after this, we'll move on to the

20  exposure.  Yeah, Ron.

21	DR. WYZGA: I have some questions.  Let

22  me say that, I'm very impressed by the approach, and, I

23  think, a really good understanding of the exposures

24  here, and I think you're, really, getting it.

25	I think you, it's going to be a challenging



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 34

Page 130

1  job, but my question was, when you talk about just

2  meeting the current standard scenario, I just want to

3  see if I understand it.

4	The current standard is 0.053 ppm, and let's

5  say, that you come up with a health benchmark, a new

6  one, of .05.  And that, if, let's say, one of your

7  cities, your concentration is .04.  Does this mean

8  that, in your risk assessment, you're going to assume

9  that the people are exposed to .053, as opposed to .04?

10	DR. GRAHAM: That's a tough one.

11	MR. RICHMOND: Our dilemma is, for our

12  purpose, not an impact assessment.  We're looking at,

13  in the country, we have levels, typically, down at .03

14  and below.  So, we're well below the current standard,

15  annual averages for the current standard.

16	How do we assess what the risk is for meeting

17  the current standard.  It's not the risk from recent

18  air quality, which is lower.  Is there a scenario, you

19  know, that we look at as a hypothetical scenario, that

20  matches exactly the current standard at the monitoring

21  network, in the design monitor just meets the 53, you

22  know, ppb.

23	There are two, you know, sort of, basic

24  choices, and I'd be interested in comments from the

25  committee.  The one we put forward but were, like to

Page 132

1  standard.

2	DR. WYZGA: My only concern is that, I

3  think, the risk estimate that you come up with is going

4  to be misread as to, this is the current risk, and - -

5	MR. RICHMOND: Right, and it's not, and I

6  appreciate that.

7	DR. WYZGA: - - and I think that's

8  something that, really, if it's done this way, you need

9  a very strong statement telling people what it is not.

10	MR. RICHMOND: Right, and we agree.  It's

11  a very hypothetical, and I agree, sort of, with your

12  comments that if we go down this path, whichever way we

13  do to simulate the current standard, we need to make it

14  clear how unlikely that is, given current NOx

15  stationary controls, given NOx vehicle controls, you

16  know, that's a very unlikely scenario.  But, that is,

17  sort of, the baseline if you're looking at, what are

18  the risks that would be, if you were just meeting the

19  current standard.

20	DR. HENDERSON: Important point.  Yes,

21  Kent.

22	DR. PINKERTON: Although this may be

23  somewhat of a trivial question, I noticed in figure

24  two, when it shows the NOx emissions that, where

25  they're coming from, and I understand that the focus



Page 131

1  hear views on, is to use recent, meaning recent the

2  last time places did not attain or were just in

3  attainment of standard.  In L.A., it wasn't that long

4  ago, had levels that were approximately the current

5  standard.

6	That's one approach to use historical air

7  quality in the '90's, generally, when some of these

8  locations were just meeting the current standard.  The

9  other choice is to use some kind of roll up approach.

10  And then, the question is, do you do it proportionately

11  and roll all the monitors up from current levels to

12  just meeting standard.  And, I believe, UARG had some

13  comments about that very issue, so I'd encourage you to

14  look at that.

15	They were actually arguing not to use the

16  historical approach, but that it would be better

17  rolling up the monitors that were nearest the road, and

18  then, rolling up the other monitors not as much based

19  on relationships between near roadway monitors and the

20  other monitors.  So, that's an alternative approach.

21  And I don't think we're fixed, yet, on exactly which

22  approach, but we put forward as, to get reaction, at

23  least.  You know, how else are you going to do it.

24  Otherwise, we don't have any results for risk or

25  exposure or air quality that approximate the current

Page 133

1  has been, primarily, on urban areas and near roadways,

2  but I noticed that close to 20 percent of NOx emissions

3  come from off highways.  And so, I'm just curious if

4  rural areas or areas of high agricultural activity, do

5  they contribute to NOx emissions, and are we missing

6  something by only focusing on urban areas or near

7  highways?

8	MR. RICHMOND: I don't know if we're

9  prepared to say much.  The one thing I will note, I

10  think we were going to look at an air quality tier one,

11  was major power plants sometimes are sited.  I know in,

12  I think it was Charlotte Mecklenburg, I know the case

13  where it was sited, just outside the ozone non-

14  attainment area, coal-fired power plant.  And we have

15  the ability of modeling to see what kind of NOx levels,

16  NO2 levels would we expect around some of those point

17  sources.

18	They may or may not be peaks of concern.

19  Maybe they're still,  with controls that we have on,

20  don't reach those levels.  I don't know if on, have any

21  information on that, but that's the kind of thing we

22  will look at in the screen analysis, do we have any

23  potential problems outside due to some of these may be

24  that we know which sources from the emission inventory

25  are major contributors to that.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 35

Page 134

1	DR. KENSKI: Kent, maybe I could clarify

2  a little bit.  That off-highway category includes

3  sources like construction equipment and marine, you

4  know, boats, lawnmowers, all those things.  And so, you

5  could make the assumption that they, generally, follow,

6  sort of, a population distribution.  I mean, the

7  distribution of emissions in that category would be

8  highly correlated with population.  So, to that extent,

9  you could assume that it was more urban and less rural.

10  Although, certainly, you know, farm equipment is a part

11  of that category.

12	DR. PINKERTON: Okay, thanks.

13	DR. HENDERSON: Okay, I think we've,

14  actually, already moved into the second section, the,

15  our exposure section, and - -

16	DR. SHEPPARD: Rogene, before we

17  continue, I wanted to comment a little bit more on the

18  air quality modeling.

19	DR. HENDERSON: Go right ahead.

20	DR. SHEPPARD: You know, well, before I

21  start, if everybody who's on the phone could mute their

22  line.  That would be really helpful.  You can press

23  star six if you don't have a mute button.  So, the

24  complexity of the modeling, I think, is ex-, of the air

25  quality model is extremely challenging.  And Tim

Page 136

1  important again, depending upon the purpose of what,

2  what the analysis is trying to do.

3	The other comment I have, beyond that, is, if

4  it could simplify the work, or at least make it clearer

5  to those of us who are reviewing it, how much each tier

6  is completely conditional on the previous tier, and to,

7  perhaps, take out anything that's overlapping.  For

8  instance, between tier one and two, that looks like

9  there's some different overlapping efforts that are

10  going to be done, and can those, can some of that be

11  removed and done in only one tier.

12	DR. HENDERSON: Do you have a response,

13  anybody want to respond to a question about the

14  overlapping of the tiers.  Again, we've gotten into the

15  exposure area, but that's fine.  That's where we're

16  supposed to go.  No comments.

17	DR. GRAHAM: We'll take a look at that.

18	DR. HENDERSON: You'll take a look at,

19  okay.  That's all we need.  Thank you, Lianne.  Did you

20  have more to, comments to make?

21	DR. SHEPPARD: I do, but maybe I'll wait

22  until other people talk about exposure, and then chime

23  in later.

24	DR. HENDERSON: Okay, well, I know you

25  still have, you'll still be there another hour or so?



Page 135

1  touched on that a bit with the street canyon issue.

2  And there's so many assumptions that are in here.  And

3  it, it's, also, I was, it really hit home, the comment

4  about the resource limitations.  And this is a,

5  potentially, a huge effort to get it right.

6	I think simplifications are possible,

7  depending on the purpose of the analysis, and I'm, it

8  strikes me as this air quality modeling is being done

9  for many different purposes, which means that

10  simplifications, if you had only one purpose, may not

11  be as easy.

12	You know, if you just want exceedances, you

13  might be able to simplify in different ways, than if

14  you wanted predictions.  Because you're going to be

15  using the predictions, for instance, in the APEX model.

16  If you're focusing just on long-term exposure, there

17  are simplifications; but if you want the short-term,

18  one-hour, that means a lot more complex model.

19	It's not so clear that temporal and spatial

20  variation in NOx are separable in the sense that, when

21  you're really near roads, the temporal patterns are,

22  probably, really different than locations far away from

23  roads.

24	So, thinking about which monitors are

25  representative for the analysis becomes really

Page 137

1	DR. SHEPPARD: Yeah, right.

2	DR. HENDERSON: Doug, you have comments

3  on the exposure method, the tiered approach, et cetera.

4	DR. CRAWFORD-BROWN: There's no lunch

5  first, then?

6	DR. HENDERSON: No.

7		DR. CRAWFORD-BROWN: Just wanted to know

8  where we stood with lunch, that's all.

9	DR. HENDERSON: Oh, well, lunch is coming

10  in thirty minutes.  Oh, lunch is ready.  I suggest we

11  have a working lunch.  I thought, maybe, according to

12  our schedule, we would go to, we would have lunch at

13  12:30, that we might make a little more progress, but

14  if hunger pangs are striking, I don't mind.

15	DR. CRAWFORD-BROWN: I'm not taking the

16  rap for this here.  I just wanted - -

17	DR. GORDON: Let me take the rap.

18	DR. HENDERSON: When you said to know,

19  it's good to know what the plan is.  Let's work a

20  little longer, I feel like.

21	DR. CRAWFORD-BROWN: Okay, I don't care.

22	DR. HENDERSON: I'm not hungry yet.

23	DR. CRAWFORD-BROWN: Is this on?  Is this

24  one on?  I'm not, is there a reason you're holding the?

25	DR. HENDERSON: It's just because I can't



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 36

Page 138

1  reach and yeah, it's on.

2	DR. CRAWFORD-BROWN: Okay, well, first, I

3  like the exposure section, as I mentioned in my

4  comments.  It, really, is very much in the line of a

5  wide range of other kinds of assessments that the EPA

6  has done over the years.  And it, really, is, I would

7  say, you know, partially in answer to Ted's issue, the

8  exposure and the risk side is starting to get pretty

9  automated these days.

10	The models are not quite plug and chug,

11  because situations change quite dramatically.  I was

12  very comforted when you said that you would use the

13  epidemiological results with air quality information

14  and the clinical studies for the, I hope I'm getting

15  this right, for the actual inter-subject variability

16  kinds of calculation, 'cause I always worry about using

17  the epidemiological results to get your slope factor,

18  or whatever, and then, also, doing inter-subject

19  variability.

20	Because the epi results, in fact, already

21  have that convolved inside of it.  And so, I hope I'm

22  understanding that correctly.

23	MR. RICHMOND: That is correct.  And if

24  you'll look at the ozone staff paper, and risk

25  assessment, you'll see that's exactly what we say - -

Page 140

1  you could do is, you could imagine that the different

2  tiers are different levels of uncertainty, or you can

3  imagine that the different tiers address different

4  kinds of questions of one, the lower tiers having to do

5  with questions about the upper percentiles of exposure.

6  And the other ones covering the whole exposure realm.

7  And then, the only other comment I would make now, the

8  rest are all in my written comments, is, I do think on

9  the uncertainty side, you've got a significant amount

10  of work to do there.

11	You always will have that.  You've got this

12  challenge of combining the, what are going to be

13  necessarily qualitative aspects of uncertainty with

14  more quantitative aspects of uncertainty, aspects of

15  uncertainty that have to do with scenario

16  specification, and so forth, and other aspects having

17  to do with uncertainty in parameter values.

18	And I'll be interested to see how you fold

19  those things together into some, sort of, overall

20  judgment of uncertainty here.  I agree with the

21  direction you were, sort of, heading, which is to make

22  it, you know, to leave this sort of expert judgment as,

23  and sort of semi-quantitative uncertainty bounds in the

24  assessment.

25	I think that will be important, rather than



Page 139

1	DR. CRAWFORD-BROWN: Yeah, exactly the

2  same thing, yeah.

3	MR. RICHMOND: - - where we had clinical

4  data and epi data.  We made that point.

5		DR. CRAWFORD-BROWN: Yeah, good, okay.  I

6  wasn't quite clear on one thing, which has to do with,

7  as you move from tier to tier, are you moving from tier

8  to tier because of things that you see in the previous

9  tiers assessment, like a screening method, for example,

10  that say, oh, if I look at the upper 95 percentile,

11  boy, that's really large.

12	So, I, that risk is large, so I better do a

13  more detailed one.  Or I look at it and it's very

14  small, so I don't need to do the more detailed one.  Or

15  are you moving from tier to tier based on whether the

16  data are available to move to the next tier.  I'm

17  assuming, maybe, a little bit of a combination of

18  those.

19	DR. GRAHAM: Right, I'd say both.  And

20  the hope would be that the prior tier is, in a sense,

21  more conservative or, well, I don't want to say,

22  hopefully, it's more uncertain, but we want to reduce

23  the uncertainty in progression from going from, say, a

24  tier one to a tier two or tier three.

25	DR. CRAWFORD-BROWN: Okay, 'cause is

Page 141

1  thinking that everything can be reduced entirely to the

2  kind of more quantitative probability density functions

3  on uncertainty.  But, in the end, you'll just have to

4  figure out how you're going to present that as a story,

5  the overall uncertainty.  But I thought the

6  methodologies were quite good.  That's all.

7	DR. HENDERSON: Thank you.  Terry Gordon,

8  do you have some assessment, or advice to give on the

9  exposure section?

10	DR. GORDON: Well, I'd say quantitative

11  risk assessment is a weak point of mine, so I don't

12  have really much to say, except exposure.

13	DR. HENDERSON: Well, this is exposure,

14  yes.

15	DR. GORDON: Well, see my confusion of

16  the terms shows - -

17	DR. HENDERSON: Well, we're talking about

18  the exposure section, not the risk assessment.  I mean,

19  the tiered exposure approach.

20	DR. GORDON: So, not the-, we're on the

21  health effects or not?

22	DR. HENDERSON: No.

23	DR. GORDON: Oh, no.

24	DR. HENDERSON: So, are you still, are

25  you just still, okay, that's fine.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 37

Page 142

1	DR. GORDON: Well, I was going to read

2  Ellis' comment.

3	DR. GORDON: I have no experience in

4  which to basically form judgment.

5	DR. CRAWFORD-BROWN: That didn't stop me

6  from talking.

7	DR. GORDON: Well.

8	DR. HENDERSON: I got the message.  Okay,

9  Jim Ultman, are you on the phone?  Have to wait for

10  people to un-mute.  Lianne, you said you wanted to save

11  your comments.  Do you have any further comments, and

12  then I'll open it up to the whole group.

13	DR. SHEPPARD: Well, you know, I'm

14  looking over what I wrote.  And I prepared an extensive

15  set of comments, but a lot of them are fairly detailed,

16  and probably aren't worth discussing now.  But the, my

17  comment about the purpose of the tiers, and I also, it

18  resonated with me, the previous comment about, exactly,

19  what is the goal of each tier, and are they

20  representing different kinds of questions, or are they,

21  really, just progressions of better information.

22  Because, in the exposure tier, it states pretty clearly

23  that they'll be using interpolated hourly NO2

24  concentrations.

25	It says measurements, but presumably,

Page 144

1  I missed it on Terry, but anyone should have, should

2  feel free to comment on the approach they're using for

3  exposure assessment, particularly the tiered approach.

4  Do we have any more comments?

5	DR. HATTIS: Yeah, I just want to just

6  reinforce that, you know, I do think that if you stop

7  at the some of the lower tiers, you'll probably not

8  produce the kind of information that will later be

9  needed in, at least, impact assessments, if not the

10  primary decisions.

11	DR. HENDERSON: Good, and I believe Frank

12  Speizer had a similar comment that he would be

13  disappointed if you stopped at tier one.

14	DR. GORDON: Rogene, I - -

15	DR. HENDERSON: Okay, Terry.

16	DR. GORDON: I might be making myself

17  more confused, but on page 22, it has a long-term

18  exposure approach, as if it's going to be using annual

19  averages.  And then, when I get to page 31, I was going

20  to talk about some, one health comment.  It says

21  they're not going to use that in the health risk

22  assessment.  So, if that's true, why are you going to

23  do this work?  I don't agree with not doing it,

24  actually.

25	DR. GRAHAM: It was probably found, I



Page 143

1  there'll also be predictions, because they'll be over

2  space as well.  And, you know, that is just a, I mean,

3  we're, as Tim alluded to, we're struggling with that

4  here in the project we're doing.

5	That's a huge, huge undertaking to do that

6  well.  So, there has to be a number of simplifying

7  assumptions to even do it at all.  And, of course, you

8  know, then you start to question how good it is.  And,

9  again, it depends on the purpose.  I guess the only

10  other comment, with respect to the exposure modeling

11  is, it seems to me that the, well, when it moves to the

12  level of the Monte Carlo simulation with the APEX

13  model, that could be expanded to incorporate some key

14  assumptions.

15	So, not just doing sensitivity analyses,

16  looking at one or two different assumptions; but

17  incorporating explicit structure for what those

18  different assumptions or models could be.  And, then,

19  getting a more explicit estimate of the uncertainty

20  that has to do with more than just the distribution, or

21  just the underlying assumptions, but also variability

22  in what those assumptions are.

23	DR. HENDERSON: Thank you, Lianne.  Now,

24  do others, as I say, these names listed here are just,

25  trying to divide up according to interest.  Obviously,

Page 145

1  don't want to say probably.  It was founded in the

2  current form of the standard.  That is, it's a annual

3  average.

4	DR. HENDERSON: Is that the answer

5  you're, I mean, does that answer your question?

6	DR. GORDON: Tradition, yeah.

7	DR. HENDERSON: Yeah, okay.

8	MR. RICHMOND: Well, one of the, I mean,

9  this is sort of the linkage to get back and forth

10  between the exposure and risk.  One could envision

11  doing a long-term, addressing long-term air quality in

12  the tier one, or long-term exposure, doesn't mean that

13  quantitatively that we have enough information to

14  address, or we don't, you know, depending on both

15  causality and the level of information on concentration

16  response relationships from the epidemiology, do we

17  have enough to make a credible quantitative risk

18  assessment, not qualitative con-, you know, concerns

19  about the health endpoints that may be shown, but

20  enough to, basically, move to that next step

21  quantitatively.  And, so I mean, there could be very

22  much a distinction between whether we do long-term air

23  quality or exposure, as opposed to a tier two,

24  quantitative, long-term risk assessment.

25	The question on benchmark is it's difficult,



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 38

Page 146

1  given the issue that Doug and others have brought up,

2  the epi studies long-term don't tell you, like the

3  clinical studies, what's a benchmark exposure level.

4  You don't know what the exposure was.  You know what

5  the ambient monitors were.  That's all.

6	DR. HATTIS: Well, because you don't,

7  they don't directly tell you about it, they certainly

8  give you a clue, from which you can reason, given your

9  other information about what they're likely to have

10  been.

11	MR. RICHMOND: It, again, depends,

12  there's all sorts of questions about is it really the

13  cumulative long-term average, or is it, as Dr. Crapo

14  mentioned earlier, is it that someone sees a peak so

15  many times per week.  There's all sorts of

16  possibilities, from a health standpoint, in terms of

17  what the real, underlying cause of those long-term

18  effects are.

19	DR. HATTIS: Yeah, and I think you need

20  to fairly characterize those uncertainties by doing it

21  a couple of different ways, and say, okay, what are the

22  differences in expectations that we get from these

23  different possible states of the world.

24	DR. HENDERSON: Thank you, Ed?

25	DR. AVOL: I submitted some written

Page 148

1  the current standard?

2	DR. HENDERSON: Yeah, go ahead, Harvey.

3  We talked about that earlier.

4	MR. RICHMOND: Yeah, I thought we had

5  discussed that about twenty minutes ago, but - -

6	DR. HENDERSON: Yeah.

7	DR. SHEPPARD: Yeah.  Well, maybe, okay,

8  maybe I just missed it.

9	MR. RICHMOND: Oh, okay, but, well, what

10  I laid out is, it is a problematic challenge to, you

11  know, to deal with how do you assess the exposures or

12  risk just meeting the current standard.  There are

13  different approaches.  One is to use historical air

14  quality when the levels were just meeting the current

15  standard back typically in the 90's for some of these

16  example urban areas.

17	The other approach would be to rec-, you

18  know, do some statistical adjustment just like we've

19  done in ozone and PM, where it's been to ra-, you know,

20  adjust things, air quality adjustment procedures to

21  adjust the distributions downward to meet a standard,

22  but effectively, rolling up distributions to just meet

23  the standard.  And there are different ways you could

24  do that, as to whether it's proportionately all the

25  monitors, or whether we make distinctions between the



Page 147

1  comments.  I won't go through all those, but one area I

2  did want to ask about was the issue of the Tiger

3  mapping, in terms of the road designations, and in

4  terms of modeling and assessing the portion of

5  population that may be within or near roadways.

6	There are ways, I point out one way that we,

7  sort of, found out there was a problem and what we did

8  about it, but there are other ways to do it.  But I

9  just wanted to get some confirmation that, in fact,

10  you're either going to ground troop it or do some

11  sensitivity analysis or something, or move to something

12  other than that.

13	DR. GRAHAM: I'm sorry, I was looking at

14  my notes.  Yeah, we are aware of that data as well.

15  And, it's not to say that we were just going to look at

16  the one data source that I mentioned, but the Tiger

17  road, it wouldn't be exclusive.  But the Tele Atlas

18  would be used.

19	DR. HENDERSON: Are there any other

20  comments or questions people have about the exposure

21  assessment section?

22	DR. SHEPPARD: Yeah, this is Lianne

23  Sheppard.  I have another question, and what does it

24  mean to look at exposure for just meeting current

25  standard, when so many of the measurements are below

Page 149

1  near roadway and non-near roadway monitors.

2	So, we were looking for feedback from the,

3  particularly, the air quality experts on this

4  committee.

5	DR. SHEPPARD: So, I guess, my question,

6  then, is, should we even do that, since the mo-, since

7  the current data are below the standard.

8	MR. RICHMOND: Right, the other

9  alternative is, then, we, otherwise, have no exposure

10  or risk associated with the current standard.  It would

11  really be the recent air quality and what standard that

12  would be associated.  So, if a place that only maximum

13  has .03 today annual average, that they were,

14  basically, looking at standards at that level and

15  below.  That's the choices we face, and you know.

16	DR. CRAWFORD-BROWN: Rogene, may I,

17  several points have touched on that.

18	DR. HENDERSON: Yes, go ahead, Doug.

19		DR. CRAWFORD-BROWN: This worries me just

20  a little bit, because the question comes up as to what

21  do you mean by meeting the standards.  If what you mean

22  by meeting the standards is that everybody goes to

23  0.053, then that's one thing.

24	But my argument is going to be, that when

25  people meet the standards, kind of like in water, it's



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 39

Page 150

1  the same issue, they put in place mitigation strategies

2  that do drop significant areas well below the

3  regulatory limit, and that is part and parcel with

4  meeting the standard.

5	So, I don't think that a scenario in which

6  everybody has gone up to .053 is, in fact, a scenario

7  that is meeting the standard in the way that meeting

8  the standard actually plays itself out.

9	MR. RICHMOND: Yeah, let me clarify that.

10  When we say meeting the standard, standards, typically,

11  have been implemented by, for example, large regional

12  areas.  The CMSA basis, it's not just L.A. County, but

13  it's the L.A. at CMSA is, typically, a definition for a

14  non-attainment area.  When we say adjusting the air

15  quality, we mean at the highest monitor in that area.

16  We're not talking about using or adjusting air quality

17  so that every single monitor in an area is just at the

18  current standard.  It is the design monitor within that

19  urban area.  So, it's like, for the whole New York

20  area, if we did New York, or Philadelphia.

21	DR. CRAWFORD-BROWN: Okay, so in your - -

22	MR. RICHMOND: So that, so none of it,

23  nowhere have we ever adjusted air quality in any of our

24  analyses in the past so that all monitors with an area,

25  when we do just meeting standard scenarios, it means

Page 152

1  one of the things, having looked at your comments, we

2  may want to, at least, take maybe the worst case

3  situation, maybe it's Los Angeles, and look, you know,

4  look at the mobile models.  What if we were to double

5  vehicle things, would we still have a problem?

6	We still might not come up to that level

7  given control technologies and given stationary

8  technologies, that was the industry argument, that even

9  given what's in place and can't be rolled back, that

10  you can't envision the scenario that gets there, but

11  that's something we could look through modeling in a

12  more limited number of areas.

13	DR. HATTIS: Yeah, and I think that's a

14  reasonable alternative, to say, okay, what is the worst

15  possible deterioration that we can reasonably imagine

16  under the current scenario.  I mean, that can include,

17  you know, non-attainment of ideal compliance with

18  everything, but.

19	DR. LARSON: Harvey, this is Tim again.

20  When your scenario of just meeting the standard is

21  going to based on the actual location of the, the

22  worst, the highest EPA monitor in that area?

23	I mean, what if you did your modeling

24  exercise and found that there were a whole bunch of

25  places that currently don't meet that standard within



Page 151

1  only at the design monitor, and what are the

2  relationships at the other monitors as they flow up and

3  down.

4	DR. CRAWFORD-BROWN: Oh, okay, okay.

5  Okay, there's something you said earlier that made me

6  think it was different from that.  Thank you.

7	MR. RICHMOND: All right, okay.

8	DR. HENDERSON: Well, and, I think Ron

9  brought up the important point.  You wouldn't want to

10  say that the risk associated with meeting the current

11  standard, or the current risk when it's, actually,

12  lower. So, I, that's, you could misinterpret that.

13  Dale, did you have something?

14	DR. HATTIS: Yeah, I think the concern is

15  to have a realistic scenario.  And in the, the

16  realistic scenario that could get you to back up to the

17  roll up, type methodology that you want to think about,

18  is imagine a future of possible growth in traffic or

19  other things that you could reasonably imagine, where

20  you could be deteriorating the air quality enough to

21  get you to near the, near compliance with the current

22  thing.  So, I think it may well be that your roll up

23  scenarios are the easiest thing to do along those

24  lines.

25	MR. RICHMOND: Right, and in that vein,

Page 153

1  the urban area.

2	MR. RICHMOND: Well, what you, are you

3  talking about the exposure modeling, air qu-, I'm a

4  little lost as to which point - -

5	DR. LARSON: Yeah, I mean, your, yeah.  I

6  mean, is it based on the current, the - -

7	MR. RICHMOND: Whether you meet the

8  standard, by definition, is at the monitors.  It, you

9  know, the mo-, we're supposed to be, have taken that

10  into account.

11	We realize in setting the CO standard and the

12  ozone standard that, no, the highest level may not, you

13  know, depending on the pollutant, may not be at the

14  monitor, but designing scenarios for alternative

15  standards is based on the monitoring network, not,

16  we'll then look at the implications through exposure

17  analysis in modeling to see what's the distribution of

18  exposures in the population, no matter where they are,

19  but it is based on simulating standards that are met at

20  the monitoring network and by definition of the

21  standard at the design monitor.

22	DR. LARSON: Okay.

23	DR. HENDERSON: Okay, does anyone who is

24  on the phone have any comments?  I don't want to ignore

25  you.  I've already heard from Lianne and Tim.  Okay, I



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 40

Page 154

1  suggest that there seems to be a strong move towards

2  lunch.  We have lunch next door.  I would like for us

3  to use this as sort of a working lunch.  Maybe be

4  prepared to start our discussions again at 1:00 in

5  here.  You can bring your lunch in here and eat, or you

6  can eat there, or do whatever you want to do.  But I'd

7  like to start the discussions at 1:00 and Vanessa has

8  something.

9	DR. VU: Since we are short of time, I'd

10  also like to invite OAR representatives to join, grab a

11  lunch here so you can quickly go back here.

12	DR. CRAWFORD-BROWN: With us?

13	DR. VU: No, no, I mean, get the lunch

14  and come back here at 1:00, since it's a, you know,

15  sandwich buffet, whatever.

16	DR. CRAWFORD-BROWN: They get in line

17  first, then.

18	DR. HENDERSON: Well, I think that's

19  great, so we will begin our discussions here at 1:00,

20  and you can use your time now to eat in there or eat

21  in, to bring it in here,  whatever you want.

22  (WHEREUPON, the morning session was concluded.)

23	DR. HENDERSON: We want to, very good.

24  Doug is my bell ringer.  We want to get started

25  discussing this last section of the methods document.

Page 156

1  is data from several studies looking at children's lung

2  function, for example, which is a long-term, sort of,

3  exposure.  And so, it seems like there needs to be some

4  reconciliation.

5	Yesterday, we had some discussion about,

6  well, perhaps maybe there is a need for a short-term

7  NO2 standard.  But, so, I just want to offer that up as

8  a comment for discussion, or at least for

9  consideration.  In any case, here, it just, sort of,

10  says, well, so, we're just not going to do it.  But I

11  think it needs to be, at the very least, it needs to

12  be, sort of, just supported or substantiated or

13  something.

14	The other comment, which is a small comment.

15  I, also, have some small written comments.  But,

16  there's a, in section 4.3, there's a discussion about

17  health responses.  And again, it, sort of, focuses more

18  on short-term effects.  And, sort of, disregards long-

19  term losses in lung function.  And yet, based on what

20  we talked about yesterday and today, one of the

21  conclusions in the ISA is going to be that long-term

22  lung function is an important issue.  So, it seems like

23  the two documents are not going to be quite consistent.

24	MR. RICHMOND: In response, we said

25  preliminary, based on what we saw in the first draft



Page 155

1  That is the one on the health risk assessment.  And I

2  think there'll be a lot of comments on that.  A lot of

3  people interested in it.  John Samet is calling in a

4  1:45, so we'll have his comments later.

5	But, we have all of our Air Office crew here.

6  Well, that was a great lunch, and I thank Vanessa and,

7  who has already left for arranging it for us.  That was

8  a nice way of handling things.

9	So, now, we're going to open our discussion

10  on the health risk assessment approach, and what, and

11  offer it, our advice to the Agency as to whether we

12  think they're using the right approach, or if it could

13  be improved.  And Ed, you are one of the first

14  discussants.

15	DR. AVOL: Okay, thank you.  I have, I

16  guess, two comments.  One has to do with the risk

17  assessment scope overview itself.  The, what's laid out

18  on page 31 talks about how the draft ISA leads to a

19  suggestion that the strongest health findings are for

20  one-hour and twenty-four-hour averaging times, so

21  there's not going to be any risk assessment for longer

22  term exposures.  And that's disquieting, I guess.

23	The current standard is an annual, long-term

24  standard.  The document says, or this document says,

25  we're only going to look at short-term exposure.  There

Page 157

1  ISA, which we're waiting, and we said we will make the

2  assessments based on the second draft ISA that will be

3  coming out before the risk assessment.  So, we will

4  look at and work closely with and see it to see how

5  those issues are addressed in the second draft ISA.

6	DR. HENDERSON: Okay, now, is, John

7  Balmes, are you on the phone?

8	DR. SHEPPARD: Rogene, I think he said he

9  was coming back at 2:15.

10	DR. HENDERSON: Oh, that's right.  I knew

11  that.  I'm sorry.

12	DR. SHEPPARD: Can I take his place?  I'm

13  leaving kind of earl-, soon.

14	DR. HENDERSON: So, would you like to

15  make your comments now?

16	DR. SHEPPARD: Yeah, and I'd like to

17  follow up on Ed's comment, because one of my main

18  concerns was for each of the tiers of the risk

19  assessment is that, I think we need to have very

20  clearly stated criteria for what particular outcomes

21  and populations and so on will be used.  And those

22  should be specified in advance.  And presumably, they

23  will come from the results of the ISA, as has been

24  stated.  But, the criteria for choosing them, for

25  instance, would it be only the out-, the outcomes that



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 41

Page 158

1  are measure-, assessed as likely causal, as an example.

2  Those, then, get brought forwarded to the risk

3  assessment.

4	My other, fairly major comment, with respect

5  to the risk assessment is, I think it needs to be

6  expanded to have three tiers.  And the first tier

7  should be the qualitative risk assessment.  And, so,

8  that it doesn't, it becomes as important as the other

9  tiers, and it also becomes the foundation for the

10  quantitative risk assessment.

11	So, all the different outcomes are reviewed,

12  but some of them, presumably, can't be easily

13  quantified in a quantitative risk assessment, but

14  they're still important, and they get discussed in the,

15  in what I would suggest would be the first tier, which

16  is the qualitative assessment.  And then, some of them

17  meet the criteria for being brought forward for

18  quantitative assessment, and they, therefore, go up to

19  the next levels.  So, I think that's a fairly  major

20  change in the organization that I recommend.

21	I think the criteria for even discussing the

22  quantitative or even conducting the quantitative risk

23  assessment need to be specified in advance as well in

24  this document.  And I, actually, recommend that all

25  tiers of all, both the exposure and the risk

Page 160

1	I guess, looking back at the ozone criteria

2  document, when we had a fair amount of discussion about

3  the linear versus logistic function, you ended up

4  using, doing a new analysis, which had different

5  weights for the linear and the logistic function.

6	That is an example of moving in that

7  direction.  It was presented more as a sensitivity

8  analysis, but the, you know, that's the, that's what

9  I'm thinking, that a lot of, a number of different

10  assumptions and uncertainties of those would be

11  incorporated into the estimates that are produced.

12	MR. RICHMOND: But again, that was a

13  sensitivity analysis.  There wasn't an assignment of

14  how much weight to put on the different choices.  We

15  put forth a base case assumption, and we looked at the

16  impact of alternative assumptions.  I'm hearing you say

17  that you want us to do more than that.  I'm still left

18  puzzled as to what you would be recommending us to do

19  differently.

20	DR. SHEPPARD: Okay, I'll try to

21  articulate that more clearly in writing.

22	DR. HENDERSON: Okay, is that all you had

23  to say right now, Lianne?

24	DR. SHEPPARD: Yes.

25	DR. HENDERSON: Okay, thank you.  And



Page 159

1  assessment, be discussed, even if the discussion is

2  that these are the reasons why we can't do this, if

3  that is the ultimate decision.

4	From what I've seen, I think that all tiers

5  should be done, at least so far, but that remains to be

6  seen.

7	And then my last, fairly major comment is the

8  quantitative risk assessment needs to, also,

9  incorporate some integrated uncertainty assessment that

10  goes beyond the  sensitivity analysis.

11	DR. HENDERSON: Okay.

12	MR. RICHMOND: Just on the last point,

13  this is Harvey Richmond.  Could you clarify what

14  approaches, either in written comments or today, when

15  you use the word integrated uncertainty assessments,

16  what you would envision that, what approaches being

17  used to carry out such integrated assessments?

18	DR. SHEPPARD: Yeah, and the devil of

19  those, that kind of thing is in the details, of course.

20  I'm sure that's why you asked the question.  And I

21  guess, the idea here is that the, and I'll try to

22  expand a little bit more than I have already.  The idea

23  is that the, we go beyond sensitivity where we assume a

24  different set of fixed assumptions to allowing for

25  multiple different assumptions.

Page 161

1  then, James Crapo, do you have something?

2	DR. CRAPO: Yeah, I think I've made most

3  of the points, so to make the, I like the risk

4  assessment model and the health endpoints.  I thought

5  you were choosing the appropriate ones, and I like,

6  particularly, that you were, included a focus on short-

7  term exposures and short averaging times.

8	And, I assume, if I read it correctly, you're

9  going to continue that as you do the tier two

10  epidemiology.  You're also going to look at the

11  possibility of using correlations with something other

12  than the national average, annual average, but rather

13  the short-term exposure peaks, which I'd really

14  encourage that.

15	Because I think that one of the most

16  important outcomes that can come out of this analysis

17  to help us develop data that would convince us it's

18  important to change the form of the standard or not.

19  Because I think that, I think form of the standard is,

20  probably, one of the more important questions to

21  address at this point in time.  And I'd like to see the

22  risk assessment provide us better information to make

23  an informed decision on that.  But I think that's

24  already part of your goal as I read it, so I'm very

25  pleased with what I see.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 42

Page 162

1	DR. HENDERSON: Thank you, James.  Is

2  Steve Kleeberger still here?  I don't see him.  He's

3  not, he wasn't going to be on the phone, was he?

4	DR. NUGENT: That's correct, and he did

5  send an email this morning saying he may be unlikely to

6  be here.

7	DR. HENDERSON: Oh, okay.  Then we'll go

8  on to Kent Pinkerton.

9	DR. PINKERTON: Okay, again, I think this

10  is a really well-executed document.  I think the

11  concern that I have is one that was also expressed by

12  Frank Speizer.  And that's just the concern that with

13  the tiered approach, that one may be tempted to stop at

14  tier one and not go beyond.  And I think that becomes

15  problematic, because there's such rare occurrences of

16  excursions of above the set standard, that one could

17  argue, well, there is no need to go to tier two and

18  look at these potential health effects.

19	But since we see so many instances of

20  significant health effects associated with ambient

21  concentrations of NO2 that are well below the

22  established standard, that I think it's just important

23  that that be really emphasized, that many of these

24  studies really need to go beyond just tier one and go

25  on to tier two and, occasionally, to tier three.

Page 164

1  national priority - -

2	DR. HATTIS: - - but I do think that

3  getting an idea of the quantitative significance of the

4  health effects that you think are likely is important

5  from a national priority setting standpoint.  You know,

6  one, of course, your main job is to inform on decisions

7  about the revision of the criteria, this particular

8  criteria standard.

9	But, also, it seems to me, that you are also

10  feeding into a national discussion about how we should

11  devote our resources to changing the mix of air

12  pollutants that we are exposed to, and, as well as

13  other problems.  And so, trying to be as thorough as

14  you can about allowing people to project national

15  impacts is an important function, okay.

16	Because this isn't going to stop at the, with

17  your meeting of the deadlines that you have in front of

18  you.  And, you know, people are going to continue to

19  try to understand how, you know, how they should be

20  devoting resources to this problem.  Because of that,

21  in part, I think you ought to really, seriously,

22  consider stretching a bit to include the kind of

23  effects that are based upon the chronic observations,

24  particularly the children's lung function growth.

25	Let me say that that's important because it



Page 163

1	But, again, I think the way it is written,

2  the health endpoints seem to be very appropriate.

3  Again, keep in mind, as you look at susceptible

4  populations, again, to keep in mind children, those who

5  have asthmatic conditions.  Also, I don't know if we've

6  really reached a point at this point in time, but are

7  there potential differences based on gender, with

8  regard to the health effects associated with nitrogen

9  oxide exposures.

10	Again, I think, some of my questions about

11  why only cities and not areas that are not city or

12  population based are not being included, but I think,

13  Donna, you helped me understand that a little bit

14  better.  So, and I think those are, pretty much, the

15  extent of my comments.

16	DR. HENDERSON: Thank you, Kent.  Dale

17  Hattis observed that I'm not too sharp here this

18  morning.  I skipped him.  That was not intentional,

19  Dale.  So, we'll hear from Dale right now.

20	DR. HATTIS: That's all right.  Anyway, I

21  want to second the thing that some of the, many of the

22  comments that, in fact, Kent Pinkerton has just made.

23  But just to say, a little more strongly.  I'm going to

24  be really disappointed if you stop at tier one.  Of

25  course, avoiding disappointing me is not a huge

Page 165

1  sets, if, in fact, it's true that the NOx or NO2

2  changes that, that sets the baseline for lung function

3  over a lifetime, which deteriorates, which grows

4  through childhood and early adulthood, and then starts

5  to deteriorate over time until you get to less and less

6  function as you get to our age.  And so, that's, and

7  that is, in fact, directly related to mortality as

8  well.

9	So, that has, sort of, long term implications

10  for lifetime function and survival that might not be

11  apparent from just saying, okay, well, we're going to

12  lose X percent of FEV1 for kids who have more than they

13  need to begin with, you know.

14	So, I think that's, so I think that's a

15  reason to take that possibility seriously, and to, it's

16  worth a little bit of a stretch, if you have to admit

17  that you have three or four or even tenfold uncertainty

18  in that, well, okay, it might still be important.

19	DR. HENDERSON: Thank you, Dale.  I'm

20  going to ask Ronald Wyzga to make his comments, and

21  then we'll open it up, and everybody can make comments.

22	DR. WYZGA: Thank you.  I think the plan

23  as written is a very good plan.  I think it's very

24  thorough, very thoughtful.  I think the difficulties

25  are going to be in the implementation.  It's a



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 43

Page 166

1  formidable task.  You have a tremendous challenge, and

2  I think that to come up with something that's going to

3  be, with your resources available and with the data

4  available, is going to be, accepted by a wide community

5  is going to be a challenge.

6	But, I applaud the approach you've taken.  I

7  think, when I think, you know, particularly, tier

8  three, and you look at the current epidemiology

9  studies, they tend to look, they tend to use linear

10  models, which suggest that there's no threshold.  And,

11  that's one area, what worries me particularly, where

12  you're looking at the just meeting current standard, in

13  the sense that if an area is still well below any

14  standard or proposed standard, because you're using a

15  linear model, you're going to overestimate the risks

16  for that area.

17	And I really worry that that could be

18  misinterpreted, and I urge you to, sort of, think about

19  both how you present that, and is there some way to get

20  around that problem.  I don't have an obvious solution

21  to it.  I think you've done some thinking about it, but

22  I urge you to think further about it, and to the extent

23  that you can't resolve it, it's going to be very

24  important how you present the results so that they

25  don't mislead the public.  But, otherwise, thank you, I

Page 168

1  point in time.

2	DR. HENDERSON: I think that's the

3  problem of double counting, are we double counting.  I

4  mean, how many deaths, how many times can a man die.  I

5  mean, I'm just joking, of course, but what I mean is -

6  -

7	DR. HENDERSON: - - is there any double

8  counting.  I think that's a logical question.

9	MR. RICHMOND: No, Rogene, no, we didn't,

10  really, indicate mor-, we didn't include mortality in

11  that preliminary list.  It's morbidity endpoints, and

12  the hospital admission studies, that have NO2 may, or

13  may not, be some of the same studies that pointed to

14  ozone or PM where they were using - -

15		DR. CRAWFORD-BROWN: Yeah, I don't know.

16  I just get a sense that I keep seeing the same kinds of

17  studies appear in documents, and, you know.

18	DR. HENDERSON: Yeah, you're right, it's

19  the morbidity we're concerned about, but, okay.  Is

20  there anybody on the phone who wants to make a comment

21  and has not?  We are - -

22	DR. POSTLETHWAIT: I think Ted wants to

23  comment.



Page 167

1  think you've done a great job.

2	DR. HENDERSON: Are there other people

3  who want to make comments on the health effects.  Doug,

4  I think does.

5	DR. CRAWFORD-BROWN: Just one small one,

6  and it's more of a medit issue.  As we've been having

7  this discussion about the epidemiological studies, and

8  the contribution of NOx and ozone and PM and so forth,

9  this,  probably, is a time for you to start thinking

10  about looking at, what I would sort of call, the mass

11  balance of the various risk studies that you're doing

12  to see if they add up to something more than the total

13  decrement that's seen in the epidemiological studies.

14  I just wonder, if you add it up, what you calculate for

15  NOx, and what you calculate for ozone, and what you

16  calculate for PM.

17	They're all based on, sort of, the same kind

18  of epidemiological results from which, we hope, we're

19  tearing apart the various relevant contributions, but I

20  just don't know.  I don't know.  If you added them up,

21  would this be something like TRIM, for example, where

22  TRIM had problems with more stuff coming out of a

23  compartment than ever went into the compartment, you

24  know.  It's sort of a mass balance, kind of, thing

25  there, that I thought I would find interesting at some

Page 169

1  quick.  We've got all the time in the world.

2	DR. RUSSELL: No, I prefer making it

3  quick.  Just, carrying on something from what James

4  said is that, if we're going to be looking at a,

5  possibly, a new standard, when you look at your table

6  two, or any of the other analyses, just keep in mind

7  that, maybe look at various alternative standards and

8  forms of standards for our assessment.

9	DR. HENDERSON: Yes, I think that would

10  be very good for the analysis to see.  Well, it's been

11  a very productive day and a half.  I'm, I really want

12  to thank everybody for working so hard, and for

13  staying, most of you staying to the end of the meeting.

14  And, I hate that we're going to miss John Samet,

15  apparently, but he has written, has he sent in his

16  comments, his written comments?

17	DR. NUGENT: I don't think we have

18  comments on the methods document, but his assistant

19  said he'd be on the line at 1:45.  This was scheduled

20  for 2:15 on the agenda.

21	DR. HENDERSON: Oh, I know, I know the

22  problem, but I'm just sitting here.  I don't think

23  people want to sit fifteen minutes to wait, I mean.

24	DR. HATTIS: He can sign on and say his

25  piece and that's fine, but we won't hear it.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 44

Page 170

1	DR. HENDERSON: But you all will be a the

2  airport.

3	DR. CRAWFORD-BROWN: I think it's $14.38

4  we'll earn during that time, so keep at it.

5	DR. HENDERSON: Okay, well, it, I do

6  value John and, the two Johns comments, but I think

7  that they have, since we're not trying to reach a

8  consensus, I mean, we're trying to get all of

9  everybody's comments, that we can just get their

10  written comments, which is what is needed.  And again,

11  if you haven't turned in your written comments, well,

12  be sure you do that.  Ed is wanting to say something.

13	DR. AVOL: Yeah, I would just ask that if

14  the Agency staff have any questions, based on what

15  they've heard, that they would like to get

16  clarification on.

17	MR. RICHMOND: I have one.

18	DR. HENDERSON: Okay.

19	MR. RICHMOND: We put forth, I know it's

20  preliminary for putting aside the long-term children's

21  health study, but for short-term, from the clinical, we

22  identified a preliminary range of .2 to .3.  Are we in

23  the right ballpark, or do the people who are familiar

24  with the clinical evidence think it's something other

25  than that range for one hour, based on the controlled

Page 172

1  we're - -

2	DR. CRAPO: - - and homes and things like

3  that.  So, you're going to extrapolate that that might

4  be there and use that.  I like that idea.

5	MR. RICHMOND: But I'm saying, on the

6  health side, are we in the right range - -

7	DR. CRAPO: On the health side, but the,

8  so you're going to, actually, look to see if there's

9  health effects, and you're going to model with that,

10  perhaps, might be that will be - -

11	MR. RICHMOND: Well, we're going to see

12  if there are exposures of, what we call, our term is

13  exposures of concern, which doesn't mean that everyone

14  who sees that exposure will, necessarily, be affected,

15  and we've explained - -

16	DR. CRAPO: I know, and I understand

17  that.  But this going to be all extrapolation data,

18  based on modeling from - -

19	MR. RICHMOND: Model data that's a

20  combination of both ambient and modeled inputs through

21  the exposure model.

22	DR. CRAPO: I think that's a very good

23  idea.  I'd love to see the data, 'cause it addresses

24  exactly what I've been talking about.

25	DR. HENDERSON: And I think he was,



Page 171

1  studies that we have, based on the evaluation in the

2  ISA.

3	DR. CRAPO: I have a question, what do

4  you mean by .2 or .3.  Are you looking for, are you

5  going to model the - -

6	MR. RICHMOND: A level of concern, if

7  you're going to compare either air quality or

8  exposures.  We did this in ozone, as you remember.

9	DR. CRAPO: Right.

10	MR. RICHMOND: But we had .06, .07, .08,

11  so we don't have to settle on a single level, but is

12  that lev-, a range at which we, at least, are

13  interpreting our evaluations, because - -

14	DR. CRAPO: All right, well, let me be

15  sure I unders- -

16	MR. RICHMOND: - - I think that that's

17  where the clinical studies start to, kind of, you know,

18  the lowest level at which effects in asthmatics are

19  being observed.

20	DR. CRAPO: I like this, except I'm not

21  sure how you're going to model it.  Because, in fact,

22  we don't have any documented exposures at that level

23  from the air monitoring stations.  You have to

24  extrapolate into cars and buses and - -

25	MR. RICHMOND: Well, that's what

Page 173

1  Harvey was probably asking Ed clinically that was what

2  you would consider appropriate levels.  Haven't you

3  done studies with children?

4	DR. AVOL: Not children in chambers at .2

5  or .3, no.  I mean, the work we did in chambers was a

6  higher, was with adults.  So, I don't know that it's, I

7  can directly relate to this.

8	MR. RICHMOND: You talking about, there

9  were a number, there's a table summarizing a number of

10  asthmatic studies that go down as low as .2.

11	DR. AVOL: And so, I mean, in that sense,

12  again, as James said, I mean, I think, what you lay out

13  is fine.  My, not withstanding my previous comment

14  about long-term studies, long-term is more - -

15	DR. HATTIS: Let me make a somewhat

16  modest further comment on that, and that is that, you

17  should, when you measure a statistically significant

18  decrement at lung function, or a change in the

19  responsiveness, you're talking about, not necessarily,

20  a uniform response or a uniform susceptibility within

21  that group that's measured.

22	And we have prior information about how

23  variable people are in their susceptibility in a large

24  number of other context.  We have a database of that

25  that's on our website.



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 45

Page 174

1	But, anyhow, so, in pharmacodynamic

2  variability, in that kind of local responsiveness,

3  happens to be, tend to be more variable than lots of

4  other stuff, basically, of the order of geometric

5  standard deviation of three or a bit more for that kind

6  of variability, if I remember correctly.

7	But, you know, so, you can, in fact, by

8  imposing a, kind of a log-probit function, with that

9  amount of spread, you can make a, just as you can make

10  distributional characterizations of the exposures, you

11  can make distributional characterizations of the likely

12  variability in susceptibility, and get something more

13  that may, in fact, you know, say, you know, for your

14  first percentile population, you might have sensitivity

15  that's outside of the range that you've measured for

16  the average concentration that's capable of changing

17  this group, right.

18	And so, any how, so, it is possible to do a

19  slightly more involved analysis that, maybe, take you

20  half hour rather than fifteen minutes, that you might

21  do for the una, with single variable analysis.

22	You also asked the question about how do we

23  do an integrated, you know, characterization of

24  uncertainty.  And there's a, this is, perhaps, a longer

25  answer, but basically, you characterize the

Page 176

1  there are ways of dealing with this, although, they do

2  require a bit of creativity and maybe creativity is a

3  bad word.

4	MR. RICHMOND: A lot of creativity on

5  work.

6	DR. HATTIS: Yeah, you know, but,

7  nevertheless, you know, it's possible.  And I think

8  it's possible without inordinate resources.  I mean, I

9  think of the, you know, the poor guy who's over in the

10  other part of ORD dealing with trichloroethylene and

11  he's having to deal with Markov chain Monte Carlo

12  simulations of to do uncertainties and variability for

13  trichloroethylene and project from animal data to

14  people.  And, you know, and he's large-, mostly one

15  guy, you know.  So, lots of people have, you know,

16  resource constraints, but you know, it's a hard problem

17  to do quantitative assessments.  But it can, you know,

18  it's not impossible.

19	DR. HENDERSON: Thank you, Dale.  You

20  almost took up the fifteen minutes, but not quite.

21	DR. HATTIS: I'm sorry.

22	DR. HENDERSON: No, it's okay.

23	Yeah, I can just change it.  Kent, go ahead.

24	DR. PINKERTON: This is a, just a

25  question about, under the risk assessment overview, you



Page 175

1  uncertainties in the exposure, and it's faster to say

2  that than to do it, and the uncertainties in the

3  susceptibility and concentration response slopes and

4  other things of that sort, and you, basically, convo-,

5  you know, basically, convolute those two with a Monte

6  Carlo simulation.

7	But you do have to, you know, sometimes these

8  are better done than other times, and it is a matter of

9  an evolving art as to how to choose the distributions

10  that you use to characterize each of the uncertainties.

11	MR. RICHMOND: And in this area, it is

12  not straightforward.  The clinical data, even when I'm

13  suggesting .2 to .3, some asthmatic studies, controlled

14  studies, have found effects.  Some have them have been

15  repeated, and haven't found the same level under the

16  same kind of conditions.  So, this is no easy matter to

17  assign probability or simply pick distributions out.

18	DR. HATTIS: Right, and so, you might

19  want to, you know, do some combined analysis that says,

20  well, there's some chance that the population

21  distribution of susceptibilities is in this range, and

22  some with this kind of mean in standard deviation, and

23  some chance that it's in some other range that would be

24  compatible with the observation that, you know, was not

25  found in a particular population.  So, anyhow, so,

Page 177

1  had mentioned there that the EPA would not develop risk

2  estimates for NO2 related effects associated with long-

3  term NO2 exposures.  And, I think you stated that you

4  wouldn't do that, based on the fact that the findings

5  are inconclusive, or at best, suggestive.  And I'm just

6  wondering, does that mean that you think that in doing

7  short-term exposure assessments, you might be able to

8  address issues that may have, with regard to NO2

9  exposures, that may have long-term effects?

10	MR. RICHMOND: No, I don't think that's

11  what we're saying.  One, as Lianne mentioned, there, we

12  have applied in the past and was envisioning here,

13  it's, selecting which health endpoint is first looking

14  at causality, and in the past, we have and proposed

15  here to do things that were likely causal, not to

16  quantitate risk, or develop risk, quantitative risk

17  estimates for things that were only suggestive or

18  limited whatever final terms the ISA ends up.

19	So, we are, that is, part of the screening

20  criteria in terms of determining how far we go, and

21  then, looking at what kind of information we have as

22  well, in terms of even once you get past.  There were

23  endpoints for ozone like inflammation, which were

24  clearly likely causal, or causal.

25	But we didn't have sufficient information to



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 46

Page 178

1  generate, we felt, a credible exposure response

2  relationship, and we still considered that in the

3  review, that endpoint, in the discussion and

4  evaluation, but we didn't do a quantitative, let's

5  produce how many people have different degrees of

6  inflammation.

7	So, just because we don't quantitate

8  something in terms of producing some number of people

9  have this many health effects, doesn't mean we're

10  ignoring the other health endpoints.

11	DR. HENDERSON: I have a question that

12  has occurred to me as I sit here.  I seem to remember

13  that some of the toxicology studies suggested

14  development of a tolerance to NO2, am I right, does

15  that happen, or am I getting it confused with ozone

16  or - -

17	MR. RICHMOND: I'll defer to Ed.

18	DR. HENDERSON: Are there development,

19  are there animal tox studies showing development of

20  tolerance?

21	DR. PINKERTON: I think there are, yes.

22  Oftentimes, NO2, as people are likely to be aware,

23  behaves in a similar manner to ozone.  It's just that

24  you have to have  much higher concentrations to get the

25  equivalent response.  But I do believe that there is a

Page 180

1  first draft of the actual document and the second

2  draft.  So, we should be seeing this document two more

3  times, and - -

4	MR. RICHMOND: Just to clarify, no, you

5  won't be seeing this document.  You'll be seeing a

6  draft exposure risk assessment report.

7	DR. HENDERSON: Oh, okay.

8	MR. RICHMOND: I mean, there's a huge

9  difference.  This was the road map, this plan.  We

10  don't, we plan to take into account your comments and

11  the comments of the public in figuring out what we

12  ultimately do.  But the revised methods and what we

13  actually do will be in, along with the results, in the

14  first draft risk assessment, that's targeted for March.

15	DR. HENDERSON: Your report.  Okay,

16  that's good, and I'm glad you - -

17	MR. RICHMOND: I just want to make sure

18  they do, and on the scheduling and we don't, under the

19  new process, we don't produce a final of this plan.

20	DR. HENDERSON: Oh, no.

21	MR. RICHMOND: The plan is a living - -

22	DR. HENDERSON: No, no, no, no.  I'm

23  just, but this is leading to a document that will be

24  reviewed two more times.

25	MR. RICHMOND: Right.



Page 179

1  tolerance that is developed with persistent exposure to

2  NO2.

3	DR. HENDERSON: And that's very hard to

4  take into account at setting any standard, I'm sure,

5  but.  Well, okay, I think we, now.  I will do as Ed did

6  earlier.  Anything, any advice or that you were

7  expecting to get that you haven't gotten and would like

8  for us to comment while you have this great group of

9  investigators here?

10	DR. GRAHAM: After your other comment, I

11  did have a question specifics about, we had selected, I

12  think, five locations, and I briefly said there were

13  criteria in their selection.  And, I think it was Ed

14  had commented, why not Phoenix and Denver, and I was

15  just wondering, why those might want to be included?

16	DR. AVOL: I picked those two in looking

17  at previous annual standards in violations of the

18  standards, and just thinking about distribution and

19  representation of the national picture that Phoenix and

20  Denver offered other sorts of geography and exposure

21  issues than just the urbanized cities like New York and

22  Philadelphia, sort of thing.

23	DR. HENDERSON: Any more advice for the

24  Agency.  We will be seeing this again, of course, but

25  the consultation comes first, and then there's the

Page 181

1	DR. HENDERSON: But thank you for

2  correcting.  This has gotten so complicated with the

3  new process, but it's good to be precise, so I'm glad

4  you corrected that.  And, perhaps, we can - -

5	DR. MARTIN: If I might, perhaps, it

6  might be worth saying, just a little bit of

7  clarification about what you can expect to see in the

8  first draft - -

9	DR. HENDERSON: Good.

10	DR. MARTIN: - - of the risk assessment

11  report versus what you can expect to see in the second

12  draft of the risk assessment report.  And I'll just

13  layout an initial major distinction, and you folks can

14  add to it as you will.  We talked about estimating

15  exposures and risks associated with various alternative

16  standards.

17	And first of all, looking at just air

18  quality,  current levels of air quality, and then, just

19  attaining the current standard.  Those are the

20  scenarios that we anticipate putting into the first

21  draft of the risk report, and that you will be seeing

22  those results and estimates associated with those two

23  scenarios in the Spring, at the same time you see the

24  second draft integrated science assessment.

25	Subsequent to that, in the second draft, we



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 47

Page 182

1  would then, at that point, decide what alternative

2  standards we would then, additionally, do exposure and

3  risk estimates for.  Because at that point, we would

4  have the benefit of the second draft science

5  assessment, and the benefit of your review of that

6  document, to help inform an appropriate range of

7  alternative standards that would reasonably be applied

8  to finish out the exposure and risk assessment.

9	So, those results you'll see in the second

10  draft assessment.

11	DR. HENDERSON: Okay, that's helpful.

12	DR. MARTIN: And I wanted to make that

13  point, because in the past, when we've come out with

14  the first draft assessment, what everyone's looking for

15  is, what is the risk associated with the range of

16  alternative standards, and that, that's what we'll do

17  in the second draft.

18	DR. HENDERSON: Second draft, in the

19  first draft, you'll have the risks associated with the

20  current exposures and the, if you reach the higher

21  levels of the - -

22	DR. MARTIN: And it relates to the

23  comment I made yesterday.  We really don't want to get

24  ahead of ourselves.  We don't want to start projecting

25  to what alternative standards may be appropriate to

Page 184

1  until I guess receive sort of the next step.  So, as a

2  starting point, I think it was fine.  And then, I think

3  we'll have to see what comes.

4	DR. HENDERSON: Okay, John.  Yeah, and

5  we've had a very good clarification of what the next

6  steps will be.  Did you get to hear that?

7	DR. BALMES: Yeah, I did hear that.  I

8  heard that, yeah.

9	DR. HENDERSON: Okay, so that was very

10  informative.  And, okay, John.  Well, we hope you

11  didn't rush over to, we were just waiting for you to,

12  in order to adjourn, to tell you the truth.

13	DR. BALMES: Oh, okay, well, then.  I'm

14  sorry to hold anybody up from adjourning.

15	DR. HENDERSON: Well, we've had

16  interesting discussions.  Really, this last discussion

17  was most helpful, and we wouldn't have had it if we

18  hadn't have been kind of waiting for you.

19	DR. BALMES: Oh, okay, okay.

20	DR. HENDERSON: So you contributed.

21	DR. BALMES: I'll be in person next time.

22	DR. HENDERSON: Okay.  Thanks a lot,

23  John, for calling in.

24	DR. BALMES: Okay, bye.

25	DR. HENDERSON: Okay, I think we are



Page 183

1  consider until we've had the benefit of your review of

2  the second draft of the science assessment, where the

3  inferences and conclusions are more sharply defined.

4	DR. HENDERSON: Good advice, Karen.

5	DR. MARTIN: Did you offer anything.

6	SPEAKER: No, that's okay, well stated.

7	DR. HENDERSON: Good.

8	SPEAKER: Rogene, somebody may have

9  just - -

10	DR. HENDERSON: That's what I thought.

11  Is there someone that had come on the phone?

12	DR. BALMES: ;Yeah, this is John, hi.

13	DR. HENDERSON: Oh, you have no idea how

14  happy we are to hear from you, John.

15	DR. BALMES: Yeah, no, I just listened.

16	DR. HENDERSON: Okay, well, we would like

17  your comments on the risk assessment part of this

18  methods document.  And, I'll tell you that the response

19  before you has been generally positive.  You probably

20  didn't get to hear all that, but - -

21	DR. BALMES: No, I, you know, I'll tell

22  you, Rogene.  I don't, I didn't provide written

23  comments on it at this point.  I didn't have, I guess,

24  very much to say, because, in a sense, it was such a

25  general template that I didn't see too much to say

Page 185

1  finished in our, it is the job of our DFO to adjourn

2  us.

3	DR. NUGENT: Well, thank you all for

4  being here, and then the next steps will be for me to

5  send around a draft of the document we spoke about this

6  morning, and send a draft of the minutes around for

7  your comments.  And I guess, even before I do that,

8  I'll be contacting you about scheduling the May meeting

9  to get your availability, so, I look forward to seeing

10  you again and thank you.  Meeting's adjourned.

11  (WHEREUPON, the PUBLIC MEETING was adjourned at 1:45

12  p.m.)

13

14

15

16

17

18

19

20

21

22

23

24

25



US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page 48

Page 186

Page 188



1	CAPTION	1

2	The foregoing matter was taken on the date,	2

3  and at the time and place set out on the Title page	3

4  hereof.	4

5	It was requested that the matter be taken by	5

6  the reporter and that the same be reduced to	6

7  typewritten form.	7

8

8	Further, as relates to depositions, it was	9

9  agreed by and between counsel and the parties that	10

10  the reading and signing of the transcript, be and	11

11  the same is hereby waived.	12

12	13

13	14

14	15

15	16

16	17

17	18

18	19

19	20

20	21

21	22

22	23

23	24

24	25

25

Page 187

1	CERTIFICATE OF REPORTER

2  COMMONWEALTH OF VIRGINIA

3  AT LARGE:

4	I do hereby certify that the witness in the

5  foregoing transcript was taken on the date, and at

6  the time and place set out on the Title page hereof

7  by me after first being duly sworn to testify the

8  truth, the whole truth, and nothing but the truth;

9  and that the said matter was recorded

10  stenographically and mechanically by me and then

11  reduced to typewritten form under my direction, and

12  constitutes a true record of the transcript as

13  taken, all to the best of my skill and ability.

14	I further certify that the inspection, reading

15  and signing of said deposition were waived by

16  counsel for the respective parties and by the

17  witness.

18	I certify that I am not a relative or employee

19  of either counsel, and that I am in no way

20  interested financially, directly or indirectly, in

21  this action.

22

23

24  MARK REIF, COURT REPORTER / NOTARY

25  SUBMITTED ON OCTOBER 25, 2007



0	 		41 41:1 141:1

0.053 130:1 149:23

00 100:1

0002 2:1

0003 3:1

0004 4:1

0005 5:1

0006 6:1

0007 7:1

0008 8:1

0009 9:1

01 101:1

02 102:1

03 103:1 130:13

149:13

04 104:1 130:1, 1

05 105:1 130:1

053 91:13 130:1

150:1

06 106:1 171:10

07 107:1 171:10

08 89:14, 17 108:1

171:10

09 109:1

1

1,000 123:21

10 10:1 53:13, 14

110:1 123:20

100 123:21

11 11:1 111:1

12 12:1 112:1

12:00 31:1

12:30 137:13

13 13:1 113:1

14 14:1 114:1

15 15:1 29:1

46:16, 24 47:11,

16 48:1, 23 49:1

115:1

16 16:1 116:1

17 17:1 117:1

18 18:1 118:1

19 19:1 119:1

1971 77:13

1993 73:15

1:00 154:1, 1, 14,

19

1:45 155:1 169:19

185:11

 	2

2 91:23 92:16 102:11

170:22 171:1

173:1, 10 175:13

20 20:1 120:1 133:1

200 28:1 29:1 46:15,

21, 22, 24 47:10

50:1 54:1, 1

2005 37:24

2007 2:1

201 26:11

21 21:1 121:1

22 22:1 122:1 144:17

23 23:1 123:1

24 24:1 48:1, 1, 14,

16 54:25 73:20

124:1

24-hour 23:1

25 2:1 10:20 25:1

125:1

26 26:1 126:1

27 27:1 127:1

28 28:1 128:1

288,000 54:1

29 29:1 129:1

2:00 107:10

2:15 107:1 157:1

169:20

3

3 91:23 102:11

170:22 171:1 173:1

175:13

30 30:1 130:1

31 31:1 131:1 144:19

155:18

31st 5:18

32 32:1 132:1

33 33:1 54:11 133:1

34 34:1 134:1

35 35:1 135:1

36 36:1 136:1

37 37:1 137:1

38 38:1 138:1

39 39:1 139:1

4

4.3 156:16

4.6 123:25

40 40:1 140:1

42 42:1 142:1

43 43:1 143:1

44 44:1 144:1

45 45:1 145:1

46 46:1 146:1

47 47:1 72:1 147:1

48 48:1 148:1

49 49:1 149:1

5

5 91:11

5-5 54:24

50 50:1 150:1

500 26:20 35:12

51 51:1 151:1

52 52:1 152:1

53 53:1 130:21 153:1

54 54:1 154:1

55 55:1 155:1

56 56:1 156:1

57 57:1 157:1

58 58:1 158:1

59 59:1 159:1

6

60 60:1 160:1

61 61:1 161:1

62 62:1 162:1

63 63:1 163:1

64 64:1 164:1

65 65:1 165:1

66 66:1 166:1

67 67:1 167:1

68 68:1 168:1

69 69:1 169:1

7

70 70:1 170:1

71 71:1 171:1

72 54:1, 1 72:1

172:1

73 73:1 173:1

74 74:1 174:1

75 75:1 175:1

76 76:1 176:1

77 77:1 177:1

78 78:1 178:1

79 79:1 179:1



 		abundant 66:10, 11

30:1 32:1, 1 38:12

 	8

80 80:1 180:1

81 81:1 181:1

82 82:1 182:1

83 83:1 183:1

84 84:1 184:1

85 85:1 115:1 185:1

86 86:1

87 87:1

88 88:1

89 89:1

8:00 31:1

9

90 90:1

90's 131:1 148:15

91 91:1

92 92:1

93 77:15 93:1 115:1

94 94:1

95 10:19 95:1 139:10

96 96:1

97 97:1

98 98:1

99 99:1

99th 54:1, 1, 10

9:00 39:14

acknowledge 58:1

115:1

add 5:21 9:25

38:1, 20, 23

40:20, 21 41:1, 14

58:1 59:10 72:17

167:12, 14 181:14

added 7:13 9:1

104:10 167:20

addition 21:23 95:13

96:10, 21 98:12

99:1 109:19 129:1

additional 54:20

77:1 97:1 110:14

112:12 128:1 129:1

additionally 182:1

address 12:1 61:1,

19 65:19 69:23

74:1 84:23 93:25

102:19 105:12

112:1 121:1 127:10

140:1 145:14

161:21 177:1

addressed 11:21

16:12 17:1 93:21

99:1 102:22

105:1 107:21

126:17 157:1

42:14 45:19 52:1

57:1 59:22 62:10

64:14 65:1 67:21

69:1, 19 70:17

73:15 75:11

78:22 93:13 106:1,

1 109:17 114:21

124:19, 21

125:11 128:1

131:15 134:14

144:24 150:1

151:11 158:24

172:1 180:13

acute 47:12, 14, 15

adjourn 184:12 185:1 adjourned 185:10, 11 adjourning 184:14 adjust
148:20, 21 adjusted 150:23 adjusting 150:14, 16 adjustment 148:18,

20

administrator 3:1, 1

6:1, 17 67:11

76:10 77:1

90:16, 23

admission 168:12

admissions 104:1, 20

 		addresses 107:25

admit 165:16

A

a.m 31:1

ability 133:15

able 27:15 68:16

78:19 81:10

98:25 102:1 105:21

107:1 115:17

125:22 126:18

135:13 177:1

accept 58:1

accepted 166:1 accomplish 108:23 according 137:11

143:25

account 98:14

125:1 153:10 179:1

180:10

absence 63:1 126:25

absolute 20:24

absolutely 33:1

57:23 127:19

172:23

addressing 41:25

85:12 88:1 99:11

128:1 145:11

adequacy 104:1

adequate 103:23

across 8:10 16:21

38:23, 25 50:14

53:1 103:17

113:1 123:19

act 87:19

action 61:25

activity 99:14 133:1

actual 21:12 25:1

39:1 97:15

138:15 152:21

180:1

actually 8:20

10:13 11:19

12:14 16:1, 14

17:24 20:19

21:14 23:21 26:1

adult 64:1 102:14 adulthood 165:1 adults 173:1 advance 157:22

158:23

advanced 80:15 128:1

adverse 41:1 42:19

44:1 50:1 51:1, 22

52:1, 21

advice 79:22 81:22

82:19 115:16, 25

116:1 141:1 155:11

179:1, 23 183:1

advise 19:15

advisory 2:1 87:19 affected 172:14 aermod 127:13, 15,

18, 21 128:1, 1

129:1, 11, 15

against 90:18 91:14,

18 94:1



age 165:1

agency 2:1 35:23

45:20 65:21

73:1, 13 79:22, 25

81:12 116:1 155:11

170:14 179:24

agenda 5:1 79:1

169:20

agg 13:15

aggregate 13:15 86:1

aggregation 96:23

118:21 119:1, 1

afternoon 2:20 32:23

78:13

ago 74:18 131:1

148:1

ahead 3:17 20:1

26:24 36:12

45:20 76:23 78:1

89:1 134:19

148:1 149:18

176:23 182:24

agreed 51:24

agreeing 16:18 48:21

agreement 6:18 61:22

104:17

agricultural 133:1

air 2:1 42:24 44:1

59:1 79:1 80:1, 12

83:20 85:10 87:17,

19, 21 92:1

95:19 96:1, 1,

18 98:13 99:14

100:23 101:1,

11, 17 102:1, 1,

24 105:19, 24,

25 108:1, 1, 13

109:1 110:24

112:1, 1 114:23

116:12 120:1

121:1, 16, 24

122:1, 1, 1

127:12, 23

129:18 130:18

131:1, 25 133:10

134:18, 24 135:1

138:13 145:11,

22 148:13, 20

149:1, 11

150:14, 16, 23

151:20 153:1 155:1

164:11 171:1, 23

181:17, 18

airborne 65:18

airport 170:1 airway 62:13 91:24 albuquerque 14:1 alert 81:25 82:11
allow 106:21 allowing 159:24

164:14

alluded 143:1

am 30:1 80:11 106:16

110:22 178:14, 15

ambient 8:23 21:17

22:12, 19 29:1

33:1 35:13 51:1

52:19 94:1, 13,

14, 15 95:1

96:12 98:13 99:1

112:25 113:1

123:25 126:13, 21,

23, 24, 24 127:1

146:1 162:20

172:20

alone 72:10

alongside 33:16 61:1 alpha 22:18, 22 already 33:1 76:24

94:1 95:17

134:14 138:20

153:25 155:1

159:22 161:24

alternative 94:16

95:1 100:25 101:1,

12 102:16, 17

105:14 109:20

131:20 149:1

152:14 153:14

160:16 169:1

181:15 182:1, 1,

16, 25

analyses 10:17 96:24

99:1, 1 120:1

143:15 150:24

169:1

analysis 12:12,

13, 18 14:1

15:17 19:18, 23

97:1 98:1, 1,

19, 19 99:21

102:20 104:15

105:19 111:1 113:1

115:1, 18 118:15

119:12 120:1

133:22 135:1, 25

136:1 147:11

153:17 159:10

160:1, 1, 13

161:16 169:10

174:19, 21 175:19

analyze 8:16

analyzed 21:24

50:1 122:1

among 49:12

amount 33:10 82:13

140:1 160:1 174:1

amplify 123:1

angela 2:1, 1 3:1,

12 4:16 6:1, 22

39:1 64:10

angela's 39:14 78:19

angeles 28:20 30:1

92:15 96:22 119:12

152:1

animal 91:1 176:13

178:19

animals 91:1

anne 4:1

annex 9:14 10:1,

1, 23 23:23

annexes 9:24 10:1

annual 8:15, 16

23:1, 15 26:1 27:1

29:1 46:24 47:1, 1

49:1, 15, 20, 24

53:25 70:15

72:20 74:12, 24

75:25 76:1

77:10, 13 90:1,

17, 25 91:1, 1,

13, 17 92:1, 12

95:10 98:21, 24

107:19, 22

130:15 144:18

145:1 149:13

155:23 161:12

179:17

annually 23:1, 1

answer 10:12 26:10

68:1 109:1 138:1

145:1, 1 174:25

answering 63:14

answers 40:1

anticipate 82:1

89:19 181:20



apart 119:22 167:19

apartment 114:10

anxiety 108:21

apex 99:17, 22 100:1

112:1 135:15

143:12

anybody 9:1 34:19

56:13 136:13

168:20 184:14

anyhow 174:1 175:25

anyone 78:11

108:11 129:17

144:1 153:23

anything 2:22 7:12

12:13 32:19 38:1

53:1, 1 59:10

64:18 66:1 72:23

89:12 136:1

179:1 183:1

anyway 63:21 65:22

78:1 122:14 163:20

anywhere 22:14

apparent 165:11 apparently 169:15 appear 16:1 49:23

168:17

appears 43:24

applaud 166:1

application 129:1,

12

applications 70:11

applied 128:15

129:11 177:12

182:1

apply 122:1, 1, 1,

11

applying 20:12, 14

appreciate 45:23

82:19 116:22 132:1

approach 65:21

84:1 87:10 88:22

89:1 90:10 95:17

99:11, 16 101:14

108:17, 24

109:11 113:15

114:23 118:1

121:1, 22 128:1,

20 129:22 131:1,

1, 16, 20, 22

137:1 141:19

144:1, 1, 18

148:17 155:10,

12 162:13 166:1

approached 95:25

approaches 83:1

113:19 148:13

159:14, 16

approaching 7:10

appropriate 8:1

93:25 109:10, 24

122:17 161:1 163:1

173:1 182:1, 25

appropriately 73:1

approved 78:16

approximate 126:1, 1

131:25

approximately 131:1

approximation

121:1 122:20, 22

area 8:1 13:10 46:21

88:1 113:13, 19

114:1, 14 119:13

133:14 136:15

147:1 150:14,

15, 17, 19, 20, 24

152:22 153:1

166:11, 13, 16

175:11

areas 9:18, 19 15:1,

1, 13, 14, 21

16:24 17:1, 22

31:1 32:11 49:16

97:1 99:1 102:1,

1, 20 103:1

104:1 105:18

111:15 113:14,

21 114:1, 20

118:13 119:17

133:1, 1, 1, 1

148:16 150:1, 12

152:12 163:11

aren't 28:1 73:25

115:17 142:16

argue 72:25 74:14

123:1 162:17

arguing 29:22 131:15

argument 29:24

149:24 152:1

arose 90:1

arranging 155:1 arrived 92:1 aside 170:20

art 175:1

articulate 52:24

160:21

aspect 64:16 65:15

aspects 59:1, 1

140:13, 14, 14, 16

assess 9:1 23:16

59:1 94:13

114:16 130:16

148:11

assessed 158:1

assessing 84:1

89:1 147:1

assessment 8:1

67:23, 25 68:15

79:16, 18, 23,

24 80:15, 20 81:17

82:13 83:1, 11,

13, 24 84:1 89:16,

18 92:1, 22 94:10,

12, 24 95:1, 24

97:10 100:21, 22

101:1 102:1 103:24

104:1, 1, 1, 12,

13 108:10 112:1

113:1, 23 115:1,

22 121:24, 25

122:1, 1, 1

126:14, 20

127:11 130:1, 12

138:25 139:1

140:24 141:1,

11, 18 144:1, 22

145:18, 24

147:21 155:1,

10, 17, 21

157:1, 19 158:1,

1, 1, 10, 13,

16, 18, 23

159:1, 1, 1 161:1,

22 169:1 176:25

180:1, 14

181:10, 12, 24

182:1, 1, 10, 14

183:1, 17

assessment's 94:18

assessments 81:18,

21, 23 82:23 84:1,

10 101:1 103:1

104:1 111:18 138:1

144:1 157:1

159:15, 17

176:17 177:1



assign 175:17

assignment 160:13 assistant 169:18 associate 124:1

71:10

attribute 60:1 61:21

64:18

availability 96:19

64:12, 14 146:25

155:15 170:13

173:1, 11 179:16

avol's 61:1

associated 20:1

104:1 110:14 185:1	 	

21:13 46:1 51:1

52:20, 21 94:14,

16 95:1 100:25

101:1, 1 105:13

149:10, 12

151:10 162:20

163:1 177:1

181:15, 22 182:15,

19

association 36:20

44:23

associations 36:24

43:1 45:11 51:21

52:21 83:20

assume 92:12 111:1

122:23 123:11

130:1 134:1 159:23

161:1

assuming 12:15 50:20

95:1 139:17

assumption 92:1

134:1 160:15

assumptions 84:1,

1 101:1 135:1

143:1, 14, 16, 18,

21, 22 159:24,

25 160:10, 16

assured 96:20

asthma 74:22 76:13

asthmatic 62:1, 12

102:14, 14 163:1

173:10 175:13

asthmatics 29:1

74:25 91:22, 22

104:21, 24 171:18

atlanta 32:10 96:22

atlas 147:17

atmosphere 47:24

65:21

audio's 5:12

attain 131:1 attained 92:13 attaining 181:19 attainment 131:1

133:14

attention 51:19

available 8:1

10:16 11:1, 1

73:23 96:1, 13, 14

100:16 108:20

139:16 166:1, 1

average 8:15, 15

23:1, 1, 15, 22

26:1 27:1 29:1

45:1 46:24 47:1,

1, 1 48:1, 15

49:1, 20, 25 70:15

72:20, 21 74:25

90:1 92:1, 1

95:10, 11 97:1

98:21, 25 107:19

113:1 117:1, 1

124:18 125:1 145:1

146:13 149:13

161:12, 12 174:16

averaged 23:1, 1, 1

averages 25:19, 20

27:1, 15 48:1

49:15 53:25

54:11 75:25 76:1

130:15 144:19

averaging 23:11,

22 25:1, 10, 11

27:1 45:1, 1 73:1,

1, 21 74:1, 12, 13

77:19 100:1 102:11

117:1 125:1 155:20

161:1

australian 12:24

16:10, 11

automated 138:1

automatic 118:12

aware 18:21 37:14

89:1 147:14 178:22

away 13:1 45:1

120:18 122:23

123:20 135:22

avoided 56:17

avoiding 163:25

avol 33:13 40:10,

11, 24 41:17

42:1 43:17 49:1, 1

59:16 61:1

B

background 8:1

73:1 88:22 89:23

94:21 117:23

bad 29:12 40:18

74:18 176:1

balance 9:24 17:20

167:11, 24

ballpark 170:23

balmes 3:25, 25

31:1, 1, 14, 17,

22 35:25 36:1,

15 37:1, 21, 22

45:13, 14 46:17,

20 50:17, 17 53:23

59:11 74:20, 21

76:16 77:20

106:23, 25, 25

107:1, 12, 14

157:1 183:12,

15, 21 184:1,

13, 19, 21, 24

barely 124:1

base 19:21 160:15

based 13:1, 14 14:15

16:1 39:1 45:21

48:15 57:1, 17

59:13 65:23

81:23 89:25

90:1, 13 91:1 95:1

96:18, 24 99:1

100:1 101:16,

23, 23 102:12

103:11 104:14

105:1, 25 109:22

111:1 112:15, 25

113:1, 1, 1, 1, 23

119:17 125:13

128:1 131:18

139:15 152:21

153:1, 15, 19

156:19, 25 157:1

163:1, 12 164:23

167:17 170:14,

25 171:1 172:18

177:1

baseline 104:1, 1



132:17 165:1

basic 114:12

122:14 130:23

basically 24:1 25:17

36:19 55:1 56:20

90:1 91:1 92:11

93:11, 22 126:22

142:1 145:20

149:14 174:1, 25

175:1, 1

basing 126:1

basis 19:25 77:18

111:1, 18 117:22

119:1 121:1 150:12

battle 55:16

beach 33:16

become 89:20 90:19

118:12

becomes 29:25 42:1

72:1 83:24 114:1

135:25 158:1, 1

162:14

begin 154:19 165:13

beginning 38:19 behaves 178:23 behind 35:10 46:1 believe 19:10

40:16 41:11

48:16 98:24 106:12

131:12 144:11

178:25

bell 154:24

benchmark 92:1, 1

101:16 102:1,

11, 16 112:15

130:1 145:25 146:1

benchmarks 97:11

100:25 102:15,

25 103:15

beneficial 35:22

benefit 182:1, 1

183:1

best 16:1 84:1

115:15 116:16

121:15 177:1

better 8:1 10:15

20:1 39:23 43:25

51:1, 1 52:23

53:1, 19 55:23

57:1, 14 58:19

61:23 69:1

131:16 139:12

142:21 161:22

163:14 175:1

beyond 34:18 42:25

84:19 104:11 136:1

159:10, 23 162:14,

24

bias 35:1

biased 11:25 13:18

16:1

biases 12:14 15:1, 1

34:1

bigger 13:1, 23

biggest 17:23

44:24 72:18

billion 26:1, 1

49:18 54:10, 11

biologic 32:1 61:19,

22

biological 29:20, 25

60:23 61:1 62:23

66:19

biologically 49:24

biology 62:1

bit 3:20 6:24 9:25

18:16 34:11

43:23 54:1 56:1

71:1 79:23 88:21

89:1, 1, 23, 25

90:1, 20, 21 92:24

100:11 107:10

116:17, 18 128:1

134:1, 17 135:1

139:17 149:20

159:22 163:13

164:22 165:16

174:1 176:1 181:1

block 98:23, 23

blown 13:1 boats 134:1 body 68:24 borne 93:1 bother 126:1

bothering 55:14 boundary 128:1 bounds 140:23 boy 139:11

break 79:1, 1, 10

90:1 123:24

breakdown 73:17

breaker 63:19

breaking 7:1

breathing 14:12,

13 27:23 28:1

65:11

bridge 69:12

brief 5:1 14:21

108:1

briefly 87:17 179:12

bring 21:20 22:15

65:16 74:1 116:1

154:1, 21

bringing 2:1 29:1

35:19

brings 116:1 118:23

broad 53:1 95:1 broader 37:1 84:21 broadly 83:1, 21 broken 98:10 brought
16:23 21:1

22:1 40:1 49:10

146:1 151:1 158:1,

17

budgets 82:1

buffer 125:1 buffet 154:15 building 114:1,

17, 18 117:16

121:10

buildings 114:1

built 99:19 113:13

bullet 11:16

22:17, 20 36:1

59:14 60:11

61:18 63:1

bunch 152:24

burden 87:1, 13

bus 123:25

buses 123:25 171:24

busy 32:11 button 134:23 bye 184:24

C

calculate 127:12

167:14, 15, 16

calculation 138:16

calibrate 111:1

california 28:20

33:15 64:24

caline 127:20, 21

128:21

caline4 127:14,



16, 18

calpuff 128:16

candidate 104:16

canyon 30:13 76:19

135:1

canyons 121:10

capabilities 99:24

128:1

capability 128:13

capable 99:22 174:16

capture 15:15, 19,

23 27:1 39:22

captured 6:21

11:12 18:1

captures 65:14 106:1

capturing 15:1 30:11

32:1

car 123:22

care 39:1 137:21

careful 37:16

75:14 124:1

carefully 7:1

78:21 103:20

carlo 143:12 175:1

176:11

carolina 87:25

carry 159:17 carrying 169:1 cars 125:1 171:24 casac 3:22, 23

5:19 10:1 35:20

51:20 52:18

78:14 81:10

89:14 106:1

case 24:21 27:21

62:11 82:1 84:11

93:16 111:23 116:1

129:11 133:12

152:1 156:1 160:15

cases 15:1 32:1

48:24 122:1

categories 61:13

category 134:1, 1,

11

causal 42:22 61:21

62:1, 19, 21,

22, 25 112:21

158:1 177:15,

24, 24

causality 112:23

145:15 177:14

cause 16:19 23:14

44:1 46:18 47:1

50:1 52:1 65:10

73:1 110:1 124:1

125:10 126:14

129:1, 19 138:16

139:25 146:17

172:23

caused 108:21

causes 46:1

causing 30:15 caution 9:11 113:1 cd 35:11, 17

census 98:22

central 20:14, 25

21:1, 14, 22 22:13

27:1 49:16 121:1

certain 47:1, 24

115:17

certainly 10:21

34:17 36:25 44:1

46:13 48:1 54:1

68:1 74:23

105:22 115:12

116:14, 16

134:10 146:1

cetera 10:25 24:16

42:19 47:1 67:12

111:13 114:23

137:1

chain 176:11

challenge 101:1

115:23 140:12

148:10 166:1, 1

challenged 43:18

challenging 115:1

129:25 134:25

chambers 173:1, 1

chance 6:1, 25 7:1

175:20, 23

change 25:1 28:20

29:21, 24 36:17

48:22, 25 55:10

58:13 72:22 105:12

138:11 158:20

161:18 173:18

176:23

changed 45:1 46:18

60:12

changes 49:1, 13, 20

50:1 59:1 73:10

123:23 165:1

changing 164:11

174:16

chapter 16:20

35:16 41:25

61:19 67:24

68:1, 18, 19, 24

69:1, 12, 18, 25

70:1, 1, 1, 22

71:1, 1 72:1, 1,

1, 1

chapters 68:10

69:1 70:1, 1, 10

71:1

characterization

95:19 96:1

105:16 174:23

characterizations

174:10, 11

characterize 34:1

84:1 101:1

146:20 174:25

175:10

characterizing 84:17

charge 6:23, 25 7:18

19:14 34:18, 20,

21, 24 39:1

61:15 64:10 106:1,

1 109:1

charlotte 133:12

charter 78:16 chartered 78:14 chattanooga 90:1 chemical 44:1 93:24
chicago 96:23

113:15, 16, 18

114:19

childhood 165:1

children 49:1, 13,

14 63:24 64:25

65:1 102:14 104:23

163:1 173:1, 1

children's 46:10

52:1 64:24 75:24

156:1 164:24

170:20

chime 136:22

choice 109:14 110:1,

1 131:1

choices 102:17

130:24 149:15

160:14

choose 35:14, 23



110:1 175:1

chooses 35:21

choosing 157:24

161:1

christian 127:1

chronic 47:11 164:23 chug 138:10 circumstances 56:16 cite 17:11 44:14
cited 115:13

cities 13:16, 19

14:10 110:1, 1

115:14 116:20

118:21 119:1

120:1, 1, 14 130:1

163:11 179:21

city 15:23 117:14,

25 163:11

clarification 9:21

10:1 23:1 31:1

60:1 170:16

181:1 184:1

clarified 77:17

clarifies 80:1 113:1

clarify 38:13 117:10

127:17 134:1 150:1

159:13 180:1

clarifying 116:25

classes 128:1 clean 2:1 87:19 clear 20:12 36:1

65:14 66:23

86:24 90:19

111:17, 17

132:14 135:19

139:1

clearer 136:1

clearly 22:1, 10

36:1 58:23 71:11

83:12 84:14

85:1, 1 122:24

123:1 142:22

157:20 160:21

177:24

clicker 95:1

clinical 21:1, 11

22:1 33:1 57:20

91:20 92:10 111:12

112:19 113:1

138:14 139:1 146:1

170:21, 24

171:17 175:12

clinically 173:1

close 13:1 49:17

54:18 57:23 63:1

109:1 133:1

closely 157:1

closer 54:13

clue 146:1

cmsa 150:12, 13

co 33:1 67:1

126:17 153:11

co-pollutants 39:24,

25

coal-fired 133:14

coefficients 29:10

coherence 39:1 45:17

50:18, 20 51:10

53:10

coherencies 53:22

coherent 43:19 53:1

cohort 113:17 cohort-based 98:19 collection 74:19

85:20

com 12:18

combination 84:11

139:17 172:20

combine 69:1

combined 175:19

combining 101:22

140:12

combustion 44:21

45:11 51:21 52:1

comes 42:20, 23

78:18 79:13

118:1 149:20

179:25 184:1

comfortable 3:1

50:19 67:1, 16

78:1, 12, 15

comforted 138:12

coming 39:13 47:19

67:1 78:13 81:1

87:21 123:10

132:25 137:1

157:1, 1 167:22

comma 52:22

comment 3:21 4:22

22:1 40:12, 13

74:1 120:17 121:19

124:20 129:18

134:17 135:1 136:1

140:1 142:1, 17,

18 143:10 144:1,

12, 20 156:1,

14, 14 157:17

158:1 159:1

168:20, 23 173:13,

16 179:1, 10

182:23

commented 81:16

179:14

comments 4:1, 10, 15

5:1, 1, 14, 17, 20

24:1 33:19 37:23

38:1, 1 39:1

40:1 71:1, 13,

21 81:1, 1, 15

84:1, 24 108:14

109:1 110:24

116:1, 10, 11

128:22 130:24

131:13 132:12

136:16, 20 137:1

138:1 140:1

142:11, 11, 15

144:1 147:1, 20

152:1 153:24

155:1, 1, 16

156:15 157:15

159:14 163:15,

22 165:20, 21

167:1 169:16,

16, 18 170:1, 1,

10, 11 180:10,

11 183:17, 23

185:1

committee 2:1 3:1

65:15 87:20

115:16, 25

130:25 149:1

committees 77:24

communities 49:14 community 166:1 commuting 111:24 comparability 27:19
comparable 16:1 21:1

27:14

compare 2:21 11:22

12:19 20:1 27:1,

15 33:23 48:17

74:19 100:16

117:19 171:1



compared 2:18

15:25 99:1

101:17 102:1

112:15 117:18

comparing 22:1 112:1

comparison 48:12, 14

97:11 122:1, 10,

20 124:12 126:15

comparisons 99:1

compartment

167:23, 23

compatible 175:24

compilation 2:16, 17

3:1 70:1

compile 33:22

complete 3:16 5:1

77:1 78:1 82:1

completed 103:1

116:11

completely 29:21

76:20 136:1

complex 135:18

complexity 134:24

compliance 151:21

152:17

complicated 113:22

114:11 181:1

compliment 125:1

comply 87:1 component 12:12 components 58:13

61:1 86:1

composition 93:25

comprehensive 96:21 con 92:19 145:18 concentration 12:22,

23 96:13 98:14

99:15 101:22 104:1

110:1 130:1 145:15

174:16 175:1

concentrations

8:24 14:15 16:1, 1

17:1 21:1, 1, 13

49:19 66:1, 1 91:1

95:12 97:15, 20

99:17 100:17

127:1, 12 142:24

162:21 178:24

concept 65:1

concern 65:17 85:1

91:1, 1, 19, 20

92:14 98:18 102:18

103:1 107:23 108:1

119:1 126:10 132:1

133:18 151:14

162:11, 12 171:1

172:13

concerned 44:24 58:1

168:19

concerns 65:14

145:18 157:18

concluded 66:1 93:23

154:22

concluding 70:1

conclusion 44:12

45:1 48:1 51:23

56:18, 21 71:1

79:1 91:1, 15, 17,

25 94:1

conclusions 67:24

68:1, 20, 23 90:24

156:21 183:1

conclusive 12:1

concur 42:18 51:1

concurrence 39:1

78:21

concurrences 78:10

concurs 51:1 52:18

condensed 9:14 16:24

35:16 106:1

condition 21:22

conditional 136:1

conditions 163:1

175:16

conduct 92:1

103:24 104:1 105:1

115:21

conducted 90:1

92:1 101:1 103:1

104:1

conducting 104:12

158:22

confidence 45:1

105:1 108:19

confined 15:1

17:25 18:1 30:13

114:1

confinement 113:22

confirmation 147:1 confirmed 18:21 confounded 30:21 confounder 66:1

confounding 90:1

confused 144:17

178:15

confusion 66:21 67:1

141:15

conjunction 85:1

connecticut 128:15 connection 69:17 conscious 40:15, 17 consensus 65:25

116:1 170:1

conservative 139:21

consider 10:21 78:15

95:13 97:16 98:1

127:24 164:22

173:1 183:1

consideration 2:24

156:1

considered 73:1

95:22 96:17

98:22 109:15

118:20 128:11

178:1

considering 43:12

consistency 51:10

53:1, 1

consistent 43:19

53:1 57:14

58:15, 16, 17

156:23

consistently 19:1

45:10

constraint 82:1

constraints 82:1, 1,

1, 15, 18 104:13

176:16

construction 134:1

consultation

79:15, 20 93:19

116:1 179:25

contacting 185:1

contain 3:1 contains 36:1 content 67:1 context 24:1, 1

69:22 83:11

90:21 105:23

173:24

continue 10:24

79:1 127:23 134:17

161:1 164:18



continuum 112:22

contradiction

85:16 86:13

contribute 133:1

contributed 59:1

184:20

contribution

97:25, 25 167:1

contributions 167:19

contributors 133:25

control 85:1 86:24

87:1, 11, 14 152:1

controlled 101:23

102:12 103:12

112:16 170:25

175:13

controls 86:23 87:1,

13 132:15, 15

133:19

convenience 5:20

conversation 7:1

23:1 75:19 89:25

conversion 128:23

converted 45:1 convey 6:17 convince 161:17 convo 175:1 convolute 175:1
convolved 138:21 cooking 27:25 cooperating 78:23 correct 19:19 33:1

38:16 123:11

138:23 162:1

corrected 181:1

correcting 181:1

correctly 29:13

36:22 138:22 161:1

174:1

correlate 29:12

correlated 27:22

29:1 75:25 90:1

134:1

correlates 53:14

correlation 29:1,

1 44:25

correlations 52:1

161:11

corresponded 24:18

cote 37:12 38:1 60:1

61:16 62:1 63:1,

18, 21 64:1, 1

65:1 68:21 69:1,

24 71:11, 15, 19

72:11, 14 76:11

cough 104:23

could've 125:10 counting 168:1, 1, 1 countries 45:10 country 8:12 92:15

107:24 122:17, 17,

19 130:13

counts 100:1, 1

110:15

county 150:12

couple 36:1 54:12

81:1 90:24 103:1

110:25 146:21

course 49:11 83:25

85:19 95:1, 20

96:1, 13 97:1,

1, 1, 16 98:1,

1, 20 99:1, 10,

19, 21 100:1, 15

101:1 122:21 143:1

159:19 163:25

164:1 168:1 179:24

court 82:1, 1 89:20

cover 32:14

covered 32:20 37:1 covering 88:15 140:1 cowling 71:21

72:12 77:1 108:16

crafting 85:1

crapo 8:1, 22

22:25 23:1 25:24

26:10, 13, 18,

22 28:1, 13

29:1, 19 30:1,

14 40:17, 20 41:1,

11 44:1 46:14,

18 47:1, 13

48:1, 11, 18 49:21

50:10 52:1 53:1

55:17 56:1 57:1,

25 62:12 70:13

72:17 73:1 74:1

75:1, 1 76:1

77:1 123:1

124:25 125:1,

10, 14 146:13

161:1, 1 171:1, 1,

14, 20 172:1, 1,

16, 22

crawford-brown 56:15

59:21 68:1 69:1,

10 70:1 137:1,

1, 15, 21, 23

138:1 139:1, 1, 25

142:1 149:16, 19

150:21 151:1

154:12, 16 167:1

168:15 170:1

create 122:24

creates 19:23 41:1 creating 123:1 creativity 176:1, 1,

1

credible 103:14,

24 145:17 178:1

credibly 115:17

crew 155:1

criteria 9:1 72:15

76:17, 24 77:23

96:18, 21 103:25

119:16 157:20,

24 158:17, 21

160:1 164:1, 1

177:20 179:13

critical 12:12 58:1

criticize 46:1

cross 10:1

crux 90:15

cumulative 32:1

146:13

curious 42:14

73:12 124:16 133:1

current 8:13, 14

51:1 52:19 88:22

92:1, 20 93:21

94:1, 12, 15 95:1,

1 96:1 101:1, 10

105:13 130:1, 1,

14, 15, 17, 20

131:1, 1, 11, 25

132:1, 13, 14,

19 145:1 147:24

148:1, 12, 14

149:1, 10 150:18

151:10, 11, 21

152:16 153:1

155:23 166:1, 12

181:18, 19 182:20

currently 11:17 82:1

107:1 152:25



cut 63:16

cycle 72:19

D

dailies 73:10

daily 48:1, 14 72:20

73:1 77:11 98:25

dakota 122:10, 12

dale 12:1 17:1

21:1 22:1 75:17

121:19 151:13

163:16, 19, 19

165:19 176:19

data 8:1, 18 10:16

12:23 15:20, 24,

25 16:20 23:23

24:13 25:1, 12

26:1 27:1 28:19

44:22 45:1, 1, 15,

18, 22 53:1

57:16 70:20, 24

74:1, 19 93:12, 14

96:1, 1, 12, 13,

14, 19, 20 99:1

100:13 101:24

102:23 104:1, 1

106:1 109:10

110:14, 15 116:19,

19 117:1 120:1

122:1 123:1, 1, 13

124:1 125:12,

22, 25 129:1

139:1, 1, 16

147:14, 16 149:1

156:1 161:17 166:1

172:17, 19, 23

175:12 176:13

database 33:21

173:24

date 24:18 89:19,

19, 20

dates 82:1 89:22

day 3:16 6:15

23:1, 1, 1, 11

26:1 33:17 44:13

48:22 55:1 76:1

77:25 103:21

169:11

daylight 75:1

days 71:1 81:1 138:1 deadline 5:17 deadlines 164:17

deal 19:1 63:19

85:23 86:17 105:24

123:1 148:11

176:11

dealing 30:19

62:18 176:1, 10

deals 39:24

dealt 44:23 deaths 168:1 decades 74:1 decay 98:1

decide 9:1 44:1 57:1

112:22 182:1

decided 57:1 63:1

deciding 84:1

decision 77:18 92:19

101:21 104:11

119:1 159:1 161:23

decisions 53:1 69:14

71:24 77:16

90:16 144:10 164:1

declining 82:1

decrement 167:13

173:18

decrements 46:10

49:12

deemed 112:20

defensible 113:25 defer 178:17 define 58:1 83:17

defined 58:1 183:1

defining 42:1

59:13 60:12, 13

definitely 88:1

110:1

definition 58:19

150:13 153:1, 20

definitive 56:18

degree 12:21 14:18 degrees 178:1 delete 119:1 deliberately 44:1
delineate 22:11 demonstrate 87:14 density 141:1 denver 179:14, 20

department 104:1, 22 depend 60:13 dependent 97:1 depending 27:19

135:1 136:1 145:14

153:13

depends 27:1 111:1

143:1 146:11

derive 111:1

derived 91:1, 20

92:1

deriving 111:1

describing 36:22

100:13

descriptive 97:1

descriptor 43:22

design 130:21 150:18

151:1 153:21

designations 147:1

designing 153:14 desire 82:12 detail 89:12

95:20, 21 122:19

detailed 62:24

63:1 139:13, 14

142:15

details 69:25 121:17

122:23 159:19

deteriorate 165:1

deteriorates 165:1 deteriorating 151:20 deterioration 152:15
determinations 68:17 determined 114:1 determining 32:22

103:25 177:20

develop 87:22 95:1

120:25 161:17

177:1, 16

developed 99:15

100:1 110:16 179:1

developer 127:22

developing 127:24

development 63:25

64:1 65:1 79:21

178:14, 18, 19

deviation 24:25

174:1 175:22

deviations 24:11

devil 159:18 devote 164:11 devoting 164:20 dfo 185:1 dictated 87:11 die
168:1

diego 14:1



diff 62:21

difference 14:1, 18,

18, 22 17:1

19:24 22:11

31:10 47:17 55:1

57:13 180:1

differences 13:1

15:13 17:1 18:1, 1

20:24 21:16

27:22 33:1 47:10

146:22 163:1

different 7:24 11:23

13:22 14:1, 23

23:21, 22 27:14,

18, 18 45:10

53:1 54:23 59:1

73:19, 22 74:12

75:12 76:12, 20

85:13, 14 98:11

102:20 105:1, 10

111:14 113:11

114:1 135:1, 13,

22 136:1 140:1, 1,

1, 1 142:20

143:16, 18 146:21,

23 148:13, 23

151:1 158:11

159:24, 25

160:1, 1, 14 178:1

differently 160:19

differing 85:25

difficult 61:20

101:1 103:16

115:21 145:25

difficulties 165:24

dilemma 130:11 dimension 45:1 dioxide 37:24 direct 20:1 54:1

61:21

directing 70:12

direction 43:1

74:1 79:1, 17

123:1 124:20

140:21 160:1

directions 29:21

directly 8:21 12:18,

19 16:1 21:1 42:19

43:1 52:19 72:1

122:10 146:1 165:1

173:1

director 80:11

disagree 12:11

disagreeing 48:1

71:15

disappointed

144:13 163:24

disappointing 163:25

disconnect 70:25 discretionary 56:1 discuss 32:18

73:21 105:20

discussants 108:1

155:14

discussed 16:25

18:16, 18 32:23

35:1, 20, 20

67:1 118:24

148:1 158:14 159:1

discussing 32:14

142:16 154:25

158:21

discussion 3:16,

17 4:10, 14 5:1

7:23 11:13 14:21

16:18 26:12 28:1

29:20 30:21

34:16 43:12, 18

49:1 54:24 59:17

62:1 66:21 70:18

73:1 75:1, 15

79:24 83:10 104:18

107:1, 18 108:1,

1, 11 109:14 124:1

155:1 156:1, 1, 16

159:1 160:1 164:10

167:1 178:1 184:16

discussions 59:1

65:24 73:18

83:23 126:1 154:1,

1, 19 184:16

disease 30:16

dispersed 25:18

dispersion 97:18

114:13 127:16

128:17 129:16

disproportionately

61:1

disquieting 155:22

disregards 156:18 distance 18:1 33:22 distances 125:1 distinction 85:1

145:22 181:13

distinctions 148:25

distinctive 64:1, 10 distinguish 102:23 distortions 54:20 distributed
2:1 25:1 distribution 12:15

15:1, 21 24:12, 19

32:1 112:14

114:1 134:1, 1

143:20 153:17

175:21 179:18

distributional

122:16 174:10, 11

distributions

23:21 25:1 99:1

111:25 114:1

148:21, 22

175:1, 17

distributive 25:10

disturbing 69:1 divide 20:21 143:25 division 80:12 doctor 80:23, 24,

25, 25

document 3:18 5:1, 1

7:1 9:1 14:1 18:15

44:12 45:1 46:1,

12 51:1, 1 52:22

53:1 55:12

56:19, 24 62:1

66:23 67:22

68:1, 1, 12

69:13 72:25

77:1, 23 79:1, 15,

16 83:24 95:22

116:1 118:1 127:13

154:25 155:24,

24 158:24 160:1

162:10 169:18

180:1, 1, 1, 23

182:1 183:18 185:1

documented 171:22

documents 83:12,

13 156:23 168:17

done 15:22 34:1,

10 37:16 51:13,

14, 15 54:1 80:18,

21 81:1 95:12 98:1

108:22 114:18

117:12, 13 118:12,

14 119:12 132:1

135:1 136:10, 11



138:1 148:19 159:1

166:21 167:1 173:1

175:1

donna 86:20 109:1

163:13

door 154:1

dose 22:1 31:1

53:11, 12, 15

65:10 70:15

double 20:20 61:11

152:1 168:1, 1, 1

doug 58:24, 25 65:13

71:23 79:12

137:1 146:1 149:18

154:24 167:1

downtown 28:20

30:1 113:18

downward 148:21

dr 2:1, 10, 13, 15

3:15, 25 4:1, 1,

1, 1, 1, 1, 1, 17,

20, 23 5:1, 1, 11,

23, 24 6:1, 12

7:1, 17, 21 8:1,

1, 19, 22 9:1,

1, 10, 11, 12, 18,

19, 20 10:1, 1,

10, 13 11:1, 1, 1,

11 12:1, 10

13:1, 1, 12, 14

14:1, 1, 11, 13,

17, 20 15:1

16:1, 13, 17 17:1,

10, 13, 19, 20

18:10, 17, 20,

24 19:1, 1, 1,

1, 14, 17 20:1,

1 21:1, 11, 21,

22, 25 22:1, 17,

20, 22, 23, 24, 25

23:1, 17, 20 24:1,

1 25:15, 17, 21,

24 26:1, 10, 13,

16, 18, 22, 24, 25

27:1, 10, 13 28:1,

1, 1, 10, 13,

15, 17 29:1, 15,

19 30:1, 1, 1,

14 31:1, 12, 14,

16, 17, 19, 22, 23

32:13 33:13, 19,

25 34:1, 1, 1, 15,

24 35:1, 25

36:1, 1, 11, 13,

25 37:1, 1, 1,

1, 12, 21 38:1, 1,

1, 10, 11 39:1,

10, 11, 12, 13,

15, 16, 17, 19, 20

40:1, 1, 1, 1, 10,

17, 20, 22, 24

41:1, 1, 11, 14,

17, 21, 23 42:1,

1, 10, 13, 16,

18 43:11, 15,

17, 20 44:1

45:13 46:14, 17,

18, 20, 23 47:1,

1, 13 48:1, 1, 10,

11, 13, 18, 21

49:1, 21 50:1,

1, 10, 16, 17, 23,

24, 25 51:15, 18

52:1, 11, 13,

17, 25 53:1, 18,

23 54:1, 1, 14,

15, 17, 21, 22

55:1, 1, 11, 12,

13, 17, 25 56:1,

1, 12, 13, 15,

17 57:1, 1, 1, 16,

18, 19, 21, 25

58:1, 11, 18, 20

59:11, 15, 16, 19,

21, 23 60:1, 1,

11, 16, 21 61:1,

1, 14, 16 62:1, 1,

12, 16, 18, 20, 21

63:1, 1, 1, 11,

13, 15, 17, 18,

20, 21, 22, 23

64:1, 1, 1, 1,

1, 1, 12, 13,

14, 21, 23 65:1,

1, 1, 1, 12, 23

66:13, 16, 22,

25 67:1, 18, 20,

21 68:1, 21

69:1, 1, 10, 24

70:1, 13 71:1, 11,

12, 15, 17, 19,

20, 21 72:11,

12, 14, 17 73:1,

1, 12, 14 74:1, 20

75:1, 1, 1, 1, 11,

13, 16, 18 76:1,

1, 11, 16, 22,

25 77:1, 20 78:1

79:11 80:1, 20,

21, 21, 22 82:24

86:19, 21 88:11,

14, 17, 18 94:25

106:1, 10, 12, 14,

16, 17, 19, 20,

23, 24, 25

107:1, 1, 11,

12, 13, 14, 16

108:1, 16 109:1,

1, 1, 1, 16, 23,

25 110:1, 1, 11,

19, 20, 22, 23, 25

113:1 115:1 116:1,

13, 21 117:1, 1,

1, 11 118:1, 1

119:1, 10, 11,

25 120:1, 1, 20,

21, 22 121:19,

21 122:21 123:1

124:23, 25

125:1, 1, 1, 10,

13, 14, 16, 18,

21, 24 127:1, 1,

1, 19 128:1 129:1,

14, 17, 21

130:10 132:1, 1,

20, 22 134:1,

12, 13, 16, 19, 20

136:12, 17, 18,

21, 24 137:1, 1,

1, 1, 1, 1, 15,

17, 18, 21, 22,

23, 25 138:1

139:1, 1, 19, 25

141:1, 10, 13, 15,

17, 20, 22, 23, 24

142:1, 1, 1, 1, 1,

13 143:23 144:1,

11, 14, 15, 16, 25

145:1, 1, 1 146:1,

13, 19, 24, 25

147:13, 19, 22

148:1, 1, 1 149:1,

16, 18, 19

150:21 151:1, 1,

14 152:13, 19

153:1, 22, 23

154:1, 12, 13, 16,

18, 23 155:15



157:1, 1, 10,

12, 14, 16 159:11,

18 160:20, 22, 24,

25 161:1 162:1, 1,

1, 1 163:16, 20

164:1 165:19, 22

167:1, 1 168:1, 1,

15, 18, 22, 24, 25

169:1, 1, 17,

21, 24 170:1, 1,

1, 13, 18 171:1,

1, 14, 20 172:1,

1, 16, 22, 25

173:1, 11, 15

175:18 176:1,

19, 21, 22, 24

178:11, 18, 21

179:1, 10, 16,

23 180:1, 15,

20, 22 181:1, 1,

1, 10 182:11,

12, 18, 22

183:1, 1, 1, 10,

12, 13, 15, 16, 21

184:1, 1, 1, 13,

15, 19, 20, 21,

22, 24, 25 185:1

draft 3:10 5:22

34:13 67:13

78:1, 18 89:15,

16, 17 104:15

155:18 156:25

157:1, 1 180:1, 1,

1, 14 181:1, 12,

21, 24, 25

182:1, 10, 14, 17,

18, 19 183:1

185:1, 1

drafting 39:1

dragging 6:1

dramatically

123:23 138:11

draw 51:19 77:1

dreams 37:11

driven 48:1, 1, 1

76:1 126:19

drives 76:1, 1

driving 50:1

drop 150:1

due 127:1 133:23 dunk 110:17 duplicate 109:19

during 23:1, 11

59:20 75:1

123:24 170:1

E

earl 157:13

earlier 51:20 68:1

69:1 70:1, 10

89:21, 21 107:1

116:1 128:22

146:14 148:1 151:1

179:1

early 66:21 79:21

84:15 165:1

earn 170:1

easier 6:1 easiest 151:23 easily 33:23

62:14, 15 158:12

easy 62:16, 17 91:12

135:11 175:16

eat 154:1, 1, 20, 20

ed 26:24 33:18 34:23

36:1 40:11 42:23

49:1 64:15

146:24 155:13

170:12 173:1

178:17 179:1, 13

ed's 157:17

effect 13:24 16:1

18:18 21:23 52:1

75:23 87:1

100:24 101:15

102:1, 15

112:18, 24

effectively 85:1

148:22

effects 10:25 14:1

18:1 20:1 25:13

29:1, 1, 14

34:21 41:1

44:19, 20 47:20,

23 48:1, 19

49:1, 23 50:10, 12

51:1, 22 66:1, 1

70:20 76:12 83:15,

17 90:1 91:14

93:12 94:1

102:13 103:1

105:20 107:1 111:1

113:22 124:1, 22

125:23 141:21

146:18 156:18

162:18, 20 163:1

164:1, 23 167:1

171:18 172:1

175:14 177:1, 1

178:1

efficient 85:11

efficiently 85:1

effort 34:10 73:17

109:19 118:17

122:16 135:1

efforts 136:1

eight 45:1 58:21, 22

59:1 64:11 65:13

67:19 75:1

either 36:21 38:19

52:14 56:1 58:1

63:19 66:1 81:1

105:17 112:1, 1,

1, 13 147:10

159:14 171:1

elements 70:17

elevated 19:11 eliminate 66:1 ellis 59:1 69:16

71:20 77:1, 20

108:14 109:1 142:1

email 162:1

embedded 93:1 emergency 104:1, 21 emission 129:1

133:24

emissions 7:24 96:14

98:15 128:14

132:24 133:1, 1

134:1

else 9:1 34:19, 19

49:1 72:24 86:17

119:13 129:17

131:23

encourage 131:13

161:14

endpoint 177:13

178:1

endpoints 74:22 77:1

101:1 103:16

104:16, 20

145:19 161:1 163:1

168:11 177:23

178:10

emphasize 29:16 33:1

74:21



emphasized 32:18

46:12 162:23

emphysema-like 91:1

enhancement 96:12

97:18

enhancements 120:25

enhancing 112:12 entire 67:22 68:1 entirely 141:1

epa 5:17 15:1, 18

16:14 17:22 43:1

66:1 76:21 77:22

116:14 129:1, 13

138:1 152:22 177:1

epa's 99:17

envelope 115:10 environment 124:15 environmental 2:1

61:1 80:12

envision 145:10

152:10 159:16

envisioning 177:12

epi 14:14 18:10,

23 19:25 33:11

43:1 46:1 48:15

50:19 57:1

73:20, 25 75:25

76:1, 1 90:13

101:23 104:19

111:1, 19 113:1

138:20 139:1 146:1

epidemiologic

42:23 45:15, 22,

25 51:1, 1 52:18

55:1

epidemiological

12:23 16:1 17:17

36:19 37:1 103:11,

23 105:1 112:24

138:13, 17

167:1, 13, 18

epidemiology

20:11, 17 22:1

38:22 45:22

47:18 48:1 52:20

74:23 90:1

112:20 145:16

161:10 166:1

equal 44:16

equals 125:1

equation 24:21

equipment 134:1, 10

equivalent 178:25

error 13:10 19:1,

1 21:23

et 10:25 24:16 42:19

47:1 67:12

111:13 114:23

137:1

especially 36:22

104:20

essential 116:1

essentially 13:17

23:25 24:1, 10, 11

25:1, 1, 1 27:13

41:25 54:1 90:15

92:1 121:21

established 85:22

162:22

establishing 20:1

estimate 16:1 24:21,

22, 24 95:1 96:1

98:25 99:24 100:22

103:1 112:13 132:1

143:19

estimated 17:16

102:1 105:1

estimates 27:1 95:16

97:23 98:16, 21

99:1, 23 100:15

101:1, 18, 24,

25 104:25 105:1

112:1 160:11

177:1, 17 181:22

182:1

estimating 17:14

99:18, 22 181:14

estimation 128:18

evaluate 54:22

55:1 94:10 129:10

evaluated 101:1

120:1 129:1, 12

evaluating 92:1

103:19

evaluation 74:15, 17

92:1 96:1, 1, 1

109:1 171:1 178:1

evaluations 171:13

events 100:23

everybody 78:23

79:11 134:21

149:22 150:1

165:21 169:12

everybody's 170:1

everyone 2:1 3:19

67:1 78:1 172:13

everyone's 182:14

everything 2:17 3:1,

1 6:19 7:19

23:1, 1 35:12 49:1

68:19 74:19

83:13 119:13 141:1

152:18

european 15:24

evidence 40:25 42:23

43:1 53:1, 21 54:1

58:15 73:21

81:20 83:1, 19

103:23 104:18

170:24

evolution 74:1

evolving 175:1 ex 134:24 exacerbation

74:22, 25

exact 3:1

exactly 27:1 44:1

55:19 124:19

130:20 131:21

138:25 139:1

142:18 172:24

examine 36:20

examining 38:22

example 18:13

24:13 27:23 62:1

70:1, 13 102:1

103:1 104:1 105:18

115:14 124:12

139:1 148:16

150:11 156:1 158:1

160:1 167:21

examples 68:23 71:17

72:1, 13

exceedance 92:1

exceedances 92:15,

17 109:12 135:12

exceeding 8:13, 13

except 30:13 48:18

55:25 141:12

171:20

excluding 71:1

exclusion 40:15 exclusive 147:17 excursion 26:18, 19,



21, 22, 23

excursions 162:16

exercise 117:15

152:24

exist 28:11, 14 99:1

existing 12:16

14:1 16:1 23:23

90:17, 25 91:1,

13, 17 92:12

107:21 111:1,

11, 20

exists 96:1 100:13

expand 9:16, 25

60:21 89:24 159:22

expanded 16:24

143:13 158:1

expect 17:18

28:22, 23 34:1

117:20 133:16

181:1, 11

expectation 24:19

expectations 146:22

expected 97:1

100:1 103:1

105:1 109:12

expecting 179:1

expense 122:15

experience 21:1

98:17 103:1 142:1

experiences 103:1

experiencing 124:13

expert 140:22

experts 115:25 149:1

explain 21:15 24:1

33:1 35:12 44:25

47:11, 12 54:23

59:18, 18 60:1, 1,

1, 15, 18 61:1

63:25 67:23, 25

70:24 73:18

76:20 79:16, 23,

24 80:14, 19 81:1,

18 83:1 84:1 88:24

89:16, 18 91:10

92:22 94:23

95:1, 15 96:10

97:1, 23 98:11,

12, 20, 21 99:1,

23, 23 100:1,

10, 16 101:18, 23,

24, 24 102:1, 1,

1, 12, 18 103:1,

12, 14, 17 104:1

105:22 108:1

111:23 112:1, 1,

13, 16, 17, 18, 18

113:1, 23 115:1, 1

116:1 118:15

120:1, 1, 24

121:25 122:1

126:13, 17, 19

127:11 129:20

131:25 134:15

135:16 136:15,

22 137:1 138:1,

1 140:1, 1

141:1, 12, 13, 18,

19 142:22 143:10

144:1, 18

11 98:17 99:18

107:20 110:1

111:13 112:14

119:1, 21, 23

120:1, 1 121:1, 1,

14 122:1 126:14

129:23 148:11

153:18 155:22

161:1 163:1 171:1,

22 172:12, 13

174:10 177:1, 1

181:15 182:20

express 105:1

expressed 23:1

162:11

extend 32:19

extended 12:13 extensive 142:14 extensively 67:1 extent 73:23

83:17, 18

111:19, 21 113:1

122:13 134:1

163:15 166:22

extrapolate 171:24

172:1

extrapolating 99:1

120:12

extrapolation 172:17

extreme 31:25 111:22

124:21

extremely 83:12,

22 134:25

extrinsically 60:24

49:24 124:1

145:10, 12, 23	 	



explained 172:15 explicit 143:17, 19 explicitly 5:1 exploits 11:1
exponential 109:13 expose 91:1, 22 exposed 13:20

20:19 26:20

30:23 70:16 100:1,

24 124:16 130:1

164:12

exposing 27:25 29:22

exposure 8:1, 10

11:1 13:15 15:15

18:11, 12 21:16

29:23 30:22 32:1

146:1, 1 147:20,

24 149:1 153:1, 16

155:25 156:1

158:25 161:13

172:14, 21 175:1

177:1 178:1 179:1,

20 180:1 182:1, 1

exposures 8:17 10:16

13:15 17:15, 16

18:22, 23 21:13

22:11, 12 27:1, 1,

17, 22 29:17

32:22, 24 33:12

34:1 51:1 53:25

70:18, 21 73:22,

24 74:1 89:1 90:18

94:19 95:1 96:1,

F

face 82:1, 1, 1

115:23 149:15

faced 86:1

facing 82:15

fact 12:11 13:19

14:1, 25 16:1

18:21 20:17, 23

24:18 25:1 29:1

30:24, 25 43:17

44:15 47:18

49:19 52:1 55:21

66:1 69:11 74:10

82:15 84:10 85:1

87:19, 23 88:1,

1 113:17 119:18,

18 123:16 125:1



129:1 138:20 147:1

150:1 163:22

165:1, 1 171:21

174:1, 13 177:1

factor 14:25 49:22

55:1 56:11, 25

138:17

factors 18:1, 1 64:1

96:12 97:18 98:1

factual 69:25

fair 160:1

fairly 10:19 11:16

16:1 23:12 33:23

46:1 48:19 49:1

78:20 142:15

146:20 158:1, 19

159:1

fall 24:20 25:1

41:23 61:1 119:22

falling 123:20

falls 8:21 familiar 170:23 family 114:24 fancy 37:25 fargo 122:12 farm
134:10

fascinating 107:17 fashion 12:1 96:1 faster 175:1

fat 25:1 fault 63:16 favor 44:16

feasibility 104:12 feasible 101:20 feature 110:14

128:24

feedback 89:1, 15

149:1

feeding 164:10

feel 3:1 10:18, 24

43:1, 10 69:17

74:23 137:20 144:1

feeling 9:1 75:18

feels 60:14

feet 28:1 47:24

123:20, 21, 21

felt 35:23 69:16

178:1

fence 43:1

fev1 165:12

fi 91:1

field 36:18 74:1,

1 116:1

fields 98:14 99:15

fifteen 26:1 28:1

47:23 79:1

169:23 174:20

176:20

fifth 65:10

fifty 20:18

figure 43:13 86:1

132:23 141:1

figured 38:14

figuring 180:11

final 67:18 80:16

82:1, 10 89:18

105:16 177:18

180:19

finally 38:22

finding 45:10 46:11

findings 42:19 45:25

51:1, 1 52:18 53:1

68:1 70:25 91:1

155:19 177:1

fine 6:1 19:13 24:1,

1 42:1 58:1 113:15

114:24 126:1 129:1

136:15 141:25

169:25 173:13

184:1

finer 98:23

fines 66:14

finish 39:1 182:1 finished 185:1 firm 82:1

first 3:10 7:12

10:15 22:1

24:13, 14 28:22

34:18, 20 36:13

59:14 60:11

64:23 67:1 69:1

72:1 77:22 79:25

80:1 81:1 84:1

88:12 89:10, 16

91:1 93:1, 1, 1

100:21 101:17

104:15 112:1 118:1

123:1 137:1

138:1 154:17

155:13 156:25

158:1, 15 174:14

177:13 179:25

180:1, 14 181:1,

17, 20 182:14, 19

fit 106:1

fits 110:1

five 13:1 18:1, 1

24:15 36:10, 15

38:1, 10 39:10,

11, 11, 21 40:1

61:17 68:1, 24

69:1, 12, 18, 25

70:1, 22 71:1,

1, 1 72:1, 1 77:24

89:20, 21 108:20

118:21 120:1

179:12

fix 22:24

fixed 113:1, 1

131:21 159:24

flat 113:12

flattered 36:14 flaw 37:18 floating 27:1 flow 49:1 151:1 focus 8:15
91:16

94:18 132:25 161:1

focused 85:24

90:14 98:1, 1

120:1

focuses 156:17

focusing 29:1

51:12 93:1 99:11

133:1 135:16

fold 15:1 54:12

140:18

folks 5:19 80:13, 17

181:13

footprints 114:17

117:16

forget 129:15

forgot 99:21

form 30:19 95:1

116:1 142:1

145:1 161:18, 19

forma 116:1

formal 6:21 former 61:1 formerly 129:12 formidable 166:1 forms 169:1

formulation 100:15

forth 68:15 140:16

145:1 160:15 167:1



170:19

forward 88:1 94:17

103:21 130:25

131:22 158:17

185:1

forwarded 158:1

foundation 158:1 founded 145:1 fraction 8:12

122:12, 17

fractionated 93:24

frame 108:21 framework 92:19 frank 144:11 162:12 free 144:1

freeway 33:16 50:1

freeways 30:1

31:1, 10, 21

frog 35:13

front 24:1 31:1 58:1

164:17

fuel 52:1

fuller 14:1 fully 122:1 function 32:1

38:24 46:1 49:1,

11, 12, 20

156:1, 19, 22

160:1, 1 164:15,

24 165:1, 1, 10

173:18 174:1

functions 104:1

141:1

fundamental 44:12

furnace 52:1

future 65:19

128:1, 11 129:1

151:18

G

gained 104:1

game 40:18, 19

59:20, 22 84:15

gary's 7:1

gas 27:25 126:18

gaseous 93:1, 1,

1, 11, 14

gather 7:1

gaussian 114:13

gee 73:1

gender 163:1

gene 41:1

general 9:1, 21

32:1, 12 48:1

65:1, 17, 24 83:1,

19 95:1, 22 108:16

115:24 183:25

generalized 121:16

generally 24:14

60:24 66:1

104:17 118:1 131:1

134:1 183:19

generate 101:25

178:1

generated 101:18

genetic 40:13

41:1, 15

genetically 41:22

genetics 58:21

geographic 60:24

102:20

geography 179:20

geometr 24:22

geometric 24:23,

24 174:1

geometry 114:1

george 20:1 22:1,

1 50:24 52:15

george's 71:22

gets 27:1 33:16 49:1

54:12 114:11

152:10

getting 10:15

23:12 29:1 104:1

107:1 112:13

129:24 138:14

143:19 164:1

178:15

given 9:15 55:1, 1

98:22 101:10 103:1

125:1 132:14, 15

146:1, 1 152:1, 1,

1

gives 11:25 51:13

82:1 90:20

109:13 123:13

126:1

giving 7:1 37:19

71:13 88:21

glad 180:16 181:1

glance 96:11

goal 142:19 161:24

goals 95:1 100:21

125:1

gold 26:16

gone 101:19 150:1 goodies 44:1 google 114:21

gordon 9:1, 11, 18

16:17 75:18 137:17

141:1, 10, 15, 20,

23 142:1, 1, 1

144:14, 16 145:1

gotten 136:14

179:1 181:1

gra 123:19

grab 154:10

graded 129:1

gradient 15:10 28:16

gradients 15:15

17:23 28:10, 13

123:19

graham 80:22 94:25

109:16, 25

110:11 119:1, 11

120:1 124:23

125:1, 1, 13

127:19 128:1

129:14 130:10

136:17 139:19

144:25 147:13

179:10

granted 57:22

graph 23:25

great 5:1 8:22 43:21

85:23 154:19 155:1

167:1 179:1

greater 48:1, 23

62:13 95:20, 21

greyer 46:21

gross 121:1

ground 14:24 28:1, 1

114:10 147:10

group 59:13 98:23

107:20 108:12

142:12 173:21

174:17 179:1

groups 5:18 8:10

grows 165:1

growth 49:12 50:1

63:25 64:1, 15, 16

65:1, 1 75:24

76:13 151:18



164:24

growths 49:12

gst 41:18

guess 7:24 22:1,

18 26:13 36:13

37:1 55:1 66:1

74:20 86:22

95:20 109:1, 1, 18

111:1 112:1 114:12

124:1 128:1, 20

143:1 149:1

155:16, 22

159:21 160:1

183:23 184:1 185:1

guessed 26:13

guise 85:23

guy 176:1, 15

H

habit 95:1

half 44:16, 17

103:21 169:11

174:20

hand 6:1 75:16

handle 10:15 124:17 handled 105:1 handling 155:1

hands 95:1

happen 67:12 178:15 happened 90:12 happens 70:1, 1

174:1

happy 18:12 42:1

58:1 183:14

hard 8:16 41:1

169:12 176:16

179:1

harder 65:11

harvey 80:22

88:15, 25 112:1

115:1 148:1 152:19

159:13 173:1

harvey's 94:11

hate 169:14

hattis 12:10 13:1,

14 14:1, 13, 20

16:1 17:13 18:24

19:1, 1, 17

21:22 23:20 24:1

25:17 26:1 27:1,

13 33:25 54:1,

17 64:1 121:21

144:1 146:1, 19

151:14 152:13

163:17, 20 164:1

169:24 173:15

175:18 176:1, 21

haven 128:15

haven't 44:23 50:1

170:11 173:1

175:15 179:1

having 30:24 47:22

53:20 74:12 77:10,

11 82:1 116:22

126:15 140:1, 16

152:1 167:1 176:11

he'd 169:19

he's 78:13 162:1

176:11, 14

head 7:1 129:1

headed 73:1 heading 140:21 heads 29:1 healing 11:1

health 10:24 13:14

15:1 29:1, 1, 13

34:19, 21 36:20,

23 41:1 44:1 46:10

48:1, 19 49:1,

1, 23 50:10, 11

51:1, 22 52:1,

22 61:21 64:24, 25

70:19 76:12

77:22 80:11

81:12 88:19 90:1

93:12 94:1 97:11

100:1, 24 101:1,

1, 15 102:1, 15

103:1, 15 105:20

107:1 111:1, 1, 18

112:15 115:19

124:1, 22 125:22

130:1 141:21

144:20, 21

145:19 146:16

155:1, 10, 19

156:17 161:1

162:18, 20

163:1, 1 164:1

167:1 170:21

172:1, 1, 1 177:13

178:1, 10

hear 5:13, 25 7:11

17:1, 1, 1 32:19

39:25 79:1, 25

131:1 163:19

169:25 183:14,

20 184:1, 1

heard 31:1, 1, 1

59:11 72:11

84:24 104:17

125:21 153:25

170:15 184:1

hearing 5:1 160:16

heart 44:1

heavily 31:24

height 11:13, 18, 23

12:1 15:1 18:1, 17

28:24

heights 14:23 15:13,

18, 21 114:18

hello 2:1, 10

help 9:1 34:1

37:19 78:19

82:16 84:1, 1,

15 161:17 182:1

helped 79:1 163:13

helpful 10:1 51:25

134:22 182:11

184:17

helping 9:1 83:23

helps 12:20 113:1

henderson 2:1, 10,

13, 15 4:1 5:1, 24

6:1, 12 7:1, 17

8:1, 19 9:1, 10,

12, 19 10:1, 10

11:1, 1 12:1 14:1,

11, 17 17:1, 19

18:10, 20 19:1, 1,

14 20:1 21:21,

25 23:17 24:1

25:15, 21 26:16,

24 29:15 32:13

34:1, 15 35:1

36:1, 11, 25 37:1,

1 38:1, 10 39:1,

11, 13, 16 40:1,

22 41:1, 14, 21

42:10, 16 43:11

50:1, 16, 24 51:15

52:13 54:1, 14, 21

55:1, 11, 12 56:1,

13 57:1, 16, 19

58:20 59:15



60:1, 16 61:14

63:1, 13, 17,

20, 22 64:1, 13,

21 65:1, 1, 12

66:13, 22 67:1, 20

71:1, 12, 17, 20

73:1, 12 75:1,

1, 13, 16 76:1, 25

78:1 79:11 80:1

86:19 88:11, 17

106:10, 14, 17,

20, 24 107:1,

11, 13, 16 108:1

109:1, 1 110:1,

20, 23 115:1

116:1, 21 117:1

118:1 125:16, 21

127:1 129:17

132:20 134:13,

19 136:12, 18,

24 137:1, 1, 1,

18, 22, 25

141:1, 13, 17, 22,

24 142:1 143:23

144:11, 15

145:1, 1 146:24

1, 15, 17 53:12

54:19 59:17 60:14,

18 61:1 70:18 91:1

133:1

higher 8:14 19:1

21:15 62:22 75:1

110:1, 1 123:25

173:1 178:24

182:20

highest 26:1 30:12

45:1 54:1, 1

150:15 152:22

153:12

highlighted 72:1

highly 134:1 highways 133:1, 1 historic 109:10 historical 89:1

90:1, 21 96:1

131:1, 16 148:13

historically 115:1

history 35:10 73:15,

16

hit 50:1 92:24 135:1

ho 74:1

hourly 48:14 95:1

117:18, 22 142:23

hours 23:12 50:1

75:1 106:19

houston 96:22 113:15

huge 118:16 123:22

135:1 143:1, 1

163:25 180:1

human 36:18 91:20

99:12 101:24

102:12 103:12

111:12 112:16

113:1

humans 97:16

hundred 13:1 18:1

26:1 33:17 47:21

48:24 49:18

58:10 105:11

hunger 137:14

hungry 137:22

hurry 114:11 hypothesizing 49:22 hypothetical

105:12 130:19

132:11

147:19 148:1, 1

hold 5:25 184:14	 	

149:18 151:1

153:23 154:18,

23 157:1, 10, 14

159:11 160:22,

25 162:1, 1 163:16

165:19 167:1

168:1, 1, 18, 25

169:1, 21 170:1,

1, 18 172:25

176:19, 22 178:11,

18 179:1, 23

180:1, 15, 20,

22 181:1, 1

182:11, 18

183:1, 1, 10,

13, 16 184:1, 1,

15, 20, 22, 25

here's 70:1, 1

herring 11:20

hey 63:11

hi 183:12

hice 4:16, 18, 18,

21, 25 5:1, 15

high 14:10 23:1,

10 26:18 30:1,

holding 137:24 home 52:1 135:1 homes 172:1

hope 22:15 59:1

65:11 78:19 89:1

113:1 122:1

138:14, 21

139:20 167:18

184:10

hoped 108:23

hopefully 139:22 hoping 120:1 horizontal 17:23 horizontally 13:1
hospital 104:1, 20

168:12

hospitalization

42:24

hot 26:1 28:22

hour 26:1 45:1 48:1,

1, 14, 16, 25

54:11, 25 73:20

75:1 76:20, 21

136:25 170:25

174:20

I

i'd 4:1, 21 7:1,

11 19:17 42:1

57:25 72:17

82:19 108:1 130:24

131:13 139:19

141:10 154:1, 1

157:16 161:13,

21 172:23

i'll 4:12, 23 11:1

37:22 51:16 52:1

77:1 87:16 95:19

115:1 136:21

140:18 142:12

159:21 160:20

168:24 178:17

181:12 183:18,

21 184:21 185:1

i'm 3:11 5:18 6:20

10:14 16:13, 18

17:20 18:18 22:1

27:24, 24 28:1

29:1, 11 30:1, 14,

20 32:10 34:11

36:14 38:21 39:11,

19 40:25 42:14, 24



43:1, 23 44:13, 24

46:21 48:1, 1,

1, 1, 16, 21, 22

49:22, 22 50:1,

19, 22 55:13

58:1 59:1 62:10

63:1 64:14 66:16

67:11 68:17

73:12 74:1 76:14

80:21, 23 82:21

88:19, 20 89:1,

11, 24 92:20, 23

94:21 97:24 100:19

106:18, 25 108:21,

23 110:1 111:17

116:21 119:1

120:1, 10, 10

124:1, 1, 16

125:19 126:1, 1

127:20 128:1

129:22 133:1 135:1

137:15, 22, 24

138:14, 21

139:16 142:13

147:13 153:1

157:11, 12

159:20 160:1,

16, 17 161:24

163:17, 23

165:19 168:1

169:11, 22

171:20 172:1

175:12 176:21

177:1 179:1

180:16, 22 181:1

184:13

i've 2:17 23:20

46:18 47:16

102:1 153:25 159:1

161:1 172:24

idea 6:24 8:11 38:24

59:12 86:22 120:13

122:14 126:1

159:21, 22 164:1

172:1, 23 183:13

ideal 152:17

ideally 102:1 ideas 115:15 identi 117:11

identification 97:1

identified 94:17

95:23 97:12, 21,

22 98:1, 16

99:20 100:1 102:10

120:1 170:22

identify 101:1

103:13 114:20

ignore 153:24

ignored 38:15 ignoring 178:10 illinois 87:25 illustrate 102:17 imagine
140:1, 1

151:18, 19 152:15

inadvertently 35:1

incidents 104:1

105:1, 10, 11, 12

include 2:25 7:19

9:22 37:21, 22

93:1 95:22 98:21

105:24 152:16

164:22 168:10

included 2:20 6:19

7:20 10:12, 18

22:16 35:1 36:1,

17, 21 126:18

127:1 161:1 163:12

179:15

includes 13:16

35:12, 17 60:22

134:1

including 19:20

87:22 99:12 104:1,

20 110:15 126:24

inconclusive 177:1

incorporate 60:12

143:13 159:1

incorporated 160:11

incorporating 143:17

increased 59:25

60:1, 1, 1, 10

91:23

increases 116:10

independent 111:12

indicate 51:1, 1

52:18 168:10

indices 54:24

indirect 87:1

individual 24:1

38:1, 1 83:14,

18 85:1, 1, 25

98:18 99:22

103:1 116:1, 1

119:1, 17 120:1

individually 38:25

88:1

individuals 60:1

95:23 98:17

103:1 118:1

indoor 12:25, 25

16:1 20:1 21:1, 17

22:1 27:16, 22

32:24 33:1 97:22

99:24 126:22, 25

127:1

indoors 8:12 96:16

98:1 126:25

induces 62:1

industry 152:1

infection 46:1

infer 12:22, 24 15:1 inferences 183:1 inferred 12:20 inflammation 62:13

177:23 178:1

influence 41:15

96:15

influenced 74:1

influences 7:25

13:22 95:14

influencing 97:23

influential 98:1 inform 164:1 182:1 information 8:1 9:15

11:22 16:1 17:1

19:21 23:14

25:22 33:14, 21

36:1 41:18 61:25

66:1 68:1 69:14

70:1 100:1 103:13,

15 104:1 105:20

106:1, 1 112:10,

12 115:12, 15

116:17, 22

117:16 119:18

120:16 133:21

138:13 142:21

144:1 145:13, 15

146:1 161:22

173:22 177:21, 25

informative 86:1

184:10

informed 161:23

impact 11:18 12:1

15:1 95:15



102:17 113:10

123:22 130:12

144:1 160:16

impacts 13:1 36:23

42:20 52:21

80:12 164:15

implementation

87:1 165:25

implemented 117:13

150:11

implementing 84:22

implications 51:23

77:10, 11 84:21

126:1 153:16 165:1

importance 11:13

21:1, 18 29:16

32:15, 20 116:10

important 7:1 12:1

13:10 15:16

18:19 21:19

31:15 33:1 34:17

40:1 42:15 46:11

68:1, 23 69:14

81:11 82:17 83:12,

22, 25 86:12 95:14

97:23 98:16 120:15

122:23 128:24

132:20 136:1

140:25 151:1

156:22 158:1, 14

161:16, 18, 20

162:22 164:1,

15, 25 165:18

166:24

imposing 174:1

impossible 176:18

impressed 44:13

129:22

impression 51:13

108:17

impressive 45:16

improve 19:15 97:14 improved 36:1 155:13 inhaling 27:1 inherent 85:1, 16
inherently 60:1 initial 83:25 181:13 initiative 127:23 injury 64:1 65:1,
1 inlet 15:1, 17 18:17

28:23

inlets 15:14 19:1

114:1

innate 60:1

inordinate 176:1

input 99:1

inputs 99:1 100:13

172:20

inside 123:25

125:1 138:21

insightful 126:15

insights 104:1, 11

instance 111:23

135:15 136:1

157:25

instances 162:19

instead 3:1 26:1

36:18 38:17

46:15 72:20

instructive 119:1

integrated 69:20, 21

93:18 159:1, 15,

17 174:23 181:24

integration 36:1

71:1

intended 61:24

63:1 67:21, 23

68:1 84:1

intent 62:1 119:11

intentional 163:18

inter-subject

138:15, 18

interaction 61:1

86:1

interactions 83:15

interacts 74:10 intercept 24:22 interchangeably

55:17 56:1

interconversion 58:1

interconvert 56:1

57:10

interest 100:1

103:18 108:12

120:10, 11 143:25

interested 115:12,

20 118:25 130:24

140:18 155:1

interesting 117:15

167:25 184:16

internal 27:16

interpolated 142:23

interpret 35:11

39:25 70:19

interpretations 83:1

interpreting 20:11

21:19 171:13

intervals 105:1

intrinsic 60:1

introduce 35:1 80:1,

17

introduction 94:21

invaluable 81:1 inventory 133:24 investigated 128:21 investigators 179:1
invite 154:10 inviting 4:10 involve 101:22 involved 115:1

116:18 174:19

isa 3:10, 17 4:11,

12 5:1 9:1, 25

10:1, 1, 1, 22

11:18, 21 12:1

14:22 19:1, 10,

15, 19 24:1 26:1

32:18 36:1 42:1

45:1 53:1, 1

54:1 67:22 77:1

78:1 82:10 89:15

95:15 100:1 101:17

103:22 104:15,

18 105:21 155:18

156:21 157:1, 1,

1, 23 171:1 177:18

isa's 9:22

irrelevant 30:25 irritant 62:14 66:1 it'd 109:21

it's 2:17 3:1, 1,

1 6:1 8:1, 24

9:1 10:19, 20

12:21 13:12, 25

15:1 16:1 17:11

19:12, 18, 22

20:10, 23, 23

24:1, 22 30:1, 1

32:16 37:1, 1,

1, 10, 18 40:16,

24 41:11 42:1

44:14, 15, 16,

20 45:1, 23, 25

47:1, 19 49:1



50:18, 19 51:25

54:1 55:15, 24

57:14, 22, 23

58:15, 17, 24

61:20 63:16, 18,

18 64:18 66:1,

22 67:15 69:1 70:1

71:22 73:1 74:1, 1

75:1, 11 76:1 79:1

80:1 82:1 83:1, 12

85:1 86:12, 24

88:1 91:12, 12

94:1 95:17, 20

98:1, 1, 18, 19,

24 99:10, 23

101:20, 22, 25

103:16 111:1,

12, 17, 19, 21

112:20 113:1

114:1, 1, 15,

24, 25 115:1, 21

116:1 117:1, 1, 1,

17 118:16 119:15

121:1 122:1, 15

124:1 126:1, 1, 1,

19 127:10 128:1,

1, 19 129:25

130:17 132:1, 1,

10 135:1, 19

137:19, 25 138:1

139:13, 22

144:18 145:1, 25

147:15 148:19,

24 149:25

150:12, 13, 19

151:11 152:1

154:14 161:17

162:22 165:1,

15, 23, 25

166:23 167:1, 24

168:11, 18

169:10 170:1,

19, 24 173:1

175:1, 23 176:1,

1, 16, 18, 22

177:13 178:23

181:1

isn't 15:18 25:16

47:1 68:1 75:10

85:15 164:16

items 71:1

iteration 51:12

iterations 38:18

77:15

issue 10:15 11:1

27:1, 1 34:16,

25 42:1 43:19

53:10, 11, 15,

15 57:1 69:11,

20 70:14 72:1, 1

82:25 84:19

86:11 88:1 90:1,

1, 10 93:1

131:13 135:1 138:1

146:1 147:1

150:1 156:22 167:1

issued 87:18

issues 6:16 8:1

10:21 18:13 53:1

59:17 61:1 82:19

83:22 90:15 93:1

121:1 126:11 157:1

177:1 179:21

J

james 10:11, 14

25:21 43:20

45:14 46:23 73:1

161:1 162:1

169:1 173:12

jealous 78:1

jenkins 80:21 88:18,

19 106:1

jill 30:1

jim 4:1, 1, 1

45:14 63:11, 13

77:1 142:1

job 130:1 164:1

167:1 185:1

joe 33:14

john 3:25 4:18 5:1

31:1 35:25

36:15, 15 37:22

38:1 43:18 45:13

50:17 71:1 74:21

78:12 106:25 155:1

157:1 169:14 170:1

183:12, 14

184:1, 10, 23

johns 170:1

join 107:1, 17

154:10

joking 168:1

jotted 6:14

joyce 36:12

ju 69:1

judge 103:15

judged 101:20 103:10 judging 17:17 judgment 72:1, 15

103:12 122:22

140:20, 22 142:1

jump 58:17 70:23

76:25 86:18

jumping 79:17

justice 61:1

justify 74:15

K

karen 73:16 77:1

80:20 82:21

88:1, 12 106:10

183:1

kenski 57:1, 18,

21 58:1, 18

86:21 109:1, 1, 23

110:1, 19 134:1

kent 132:21 134:1

162:1 163:16, 22

176:23

key 21:1 35:1, 14,

23 36:1 43:24

101:1 110:13 124:1

125:15 143:13

keywords 37:18, 25

kicked 94:1 kid 50:1 kids 165:12

kinds 14:1 15:15

28:10 53:1 69:14

70:11 113:19

138:1, 16 140:1

142:20 168:16

kleeberger 162:1

knew 38:13 56:1

157:10

kudos 58:25

L

l.a 131:1 150:12, 13

la 92:19

lab 41:18

lack 66:19 93:12,

12, 14

laid 36:1 53:21

148:10 155:17



lakewood 28:20

30:1 33:15

large 11:16, 17

30:20 33:24 49:1

139:11, 12

150:11 173:23

176:14

larger 11:19 13:1

14:25 19:21

larson 4:1, 1

15:1, 1 16:13

17:10, 20 18:17

22:17, 22, 24

28:1, 10, 15 31:1,

1, 12, 16 33:19

46:23, 23 47:1

48:1, 10, 13, 21

50:1 62:18, 21

64:23 65:23 66:16,

25 110:20, 22,

25 113:1 116:13

117:1, 1, 11

152:19 153:1, 22

last 11:14 20:1

23:15 36:1 38:21

40:25 42:14 44:1

51:1 52:15 56:18

61:18 77:1 90:23

91:16 92:19 93:19,

20, 22 103:1, 20

131:1 154:25

159:1, 12 184:16

late 40:18

later 21:18 23:24

55:10 90:1 126:1

136:23 144:1 155:1

latest 36:18

lawnmowers 134:1

lay 51:1 71:1 173:12

layer 128:1 layout 181:13 lead 33:10 67:25

88:15, 19 118:24

120:14 123:1

leading 180:23

leads 155:18 leaning 75:22 learned 70:10 learning 115:11

123:16, 17

least 29:11 30:18

83:25 84:1

91:10, 11 113:1

121:23 122:1 126:1

129:16 131:23

136:1 144:1

152:1 156:1, 11

159:1 171:12

leave 60:14 107:1

140:22

leaving 106:17, 22

157:13

led 16:1

lee 4:1

legs 121:22 lengths 27:15 lesions 91:1

less 25:1, 1, 18

35:17 134:1 165:1,

1

lessons 122:1

let's 7:1, 16 20:17,

18 34:15 40:20, 20

59:1 64:10, 22

77:1 108:1, 13

118:1 120:17 121:1

126:11 130:1, 1

137:19 178:1

letter 3:1 6:1

17:1 18:14, 20

29:16 33:1 38:1

39:1 59:1 65:17

67:10, 13, 16

78:1, 1, 17, 19

116:1

letting 107:15

lev 171:12

level 14:24 15:1

16:1 19:1 21:1

23:1, 13 24:15, 16

28:1, 1, 1 29:23

30:17 45:16

70:18 72:21

92:16 119:22

120:1, 11 121:1

123:15, 18, 23

124:21 143:12

145:15 146:1

149:14 152:1

153:12 171:1,

11, 18, 22 175:15

levels 12:15, 17, 19

17:18 19:12 20:18,

25 23:10, 12, 22

29:1 30:12, 15, 22

35:14 47:23, 25

73:19 90:1 91:1,

1, 1, 1, 19, 20,

22 92:1, 1, 13

94:13, 14, 15

95:1, 1 97:1, 1

98:21 100:1

101:10, 11, 16

102:1, 1, 1, 16,

18, 25 103:1

107:22 111:1

112:15 115:20

123:22 124:14

130:13 131:1, 11

133:15, 16, 20

140:1 148:14

158:19 173:1

181:18 182:21

lianne 39:12, 13,

17, 20 106:14,

21 116:18 136:19

142:10 143:23

147:22 153:25

160:23 177:11

lifetime 165:1, 10

likely 27:21 28:1

30:1 34:1 42:22

44:20 49:25

61:1, 1 62:19, 22,

24 103:10 104:15

112:21 146:1 158:1

164:1 174:11

177:15, 24 178:22

limit 82:11 83:19

113:23 150:1

limitations 135:1

limited 98:1 99:1

105:18 152:12

177:18

limits 47:22 108:19

line 24:20 25:1 35:1

40:22, 23, 23

128:10 134:22

138:1 154:16

169:19

linear 160:1, 1

166:1, 15

lines 40:11, 24

151:24

link 52:1 57:15

97:15



link-based 128:14

linkage 44:1 145:1

list 2:18, 19, 21

5:1 6:1, 10, 13,

13, 19 69:1

110:1 116:1 168:11

listed 2:19 32:21

103:25 104:16

105:1 143:24

listened 23:1 103:20

183:15

listing 6:21 71:1

lit 37:16

literature 17:10, 14

19:21 46:1 91:1,

21 103:11 112:24

little 3:20 4:13

9:25 10:15 35:10

42:21 54:1, 12, 19

20 131:1 135:22

179:12

log 24:1, 18, 22,

23, 24 25:1, 1, 1

log-probit 174:1

logical 109:11 168:1 logically 76:1 86:1 logistic 109:21

160:1, 1

long 30:25 33:12, 16

58:1, 1 75:22 91:1

95:1 101:12 114:24

120:1 131:1 156:18

165:1 177:1

long-term 49:13

73:18 75:23 76:1

90:1, 13, 17 91:1,

1, 14 94:19

95:10 97:1 98:20

173:10

lower 47:22 61:1

101:10 130:18

140:1 144:1 151:12

lowest 47:19

102:18 171:18

lumped 119:13

lunch 107:1 137:1,

1, 1, 10, 11, 12

154:1, 1, 1, 1,

11, 13 155:1

lung 46:1 49:1,

11, 12, 20 50:1

75:24 76:13 91:1

156:1, 19, 22

164:24 165:1

173:18

lydia 80:1, 1, 1, 10

lynwood 33:14, 15

56:1 69:1 75:14

111:1, 14 135:16	 	

76:23 88:21

89:1, 1, 1, 23, 24

90:1, 20, 21 92:24

100:11 107:1, 10

128:1 134:1, 17

137:13, 20

139:17 149:20

153:1 159:22

163:13, 23

165:16 181:1

live 32:1 59:14

lives 113:17

living 29:1 180:21 lo 24:1 25:18 76:1 local 95:14 98:14

121:1, 1, 23, 24

122:1, 1 123:1,

13, 14, 18

124:1, 21

125:22, 25 174:1

locally 121:18

124:1, 1

located 19:11

location 18:1

99:13 152:21

location-specific

121:1

locations 18:1 47:24

59:18 96:17 97:1

98:1, 1 99:19

100:1 110:10

118:20 119:15, 16,

144:17 145:11, 11,

12, 22, 24

146:1, 13, 17

155:23 156:1, 21

170:20 173:14,

14 177:1

longer 9:1, 23

25:1 34:11, 12

128:1 137:20

155:21 174:24

los 28:20 30:1 92:15

96:22 119:12 152:1

lose 75:21 121:1

165:12

losses 156:19

lost 153:1

lot 11:22 15:20

17:12 23:13

31:1, 12, 13, 20

32:1 45:1 53:1, 20

55:1 68:1 70:14

74:15 108:1

124:1 135:18

142:15 155:1, 1

160:1 176:1 184:22

lots 45:10 53:1

108:1 110:10 174:1

176:15

louis 88:1

love 5:13 172:23

low 14:12, 14

23:1, 12, 13 91:23

M

magnitude 47:17, 17

main 69:10 157:17

164:1

maintain 91:1

major 33:22 59:1

68:1 74:22

126:10 133:11,

25 158:1, 19 159:1

181:13

maker 68:14

man 168:1

management 85:10

87:21

manager 87:18

manner 178:23

map 31:1 115:24

180:1

mapping 147:1

maps 9:22 march 180:14 marine 134:1 marked 72:1 markov 176:11

martin 73:16 77:1

80:20 82:24

88:14 106:12

120:20, 22

122:21 181:1, 10

182:12, 22 183:1

mary 9:19 10:1



34:1 51:16

54:14, 21 67:17

89:10

mass 85:23 93:24

167:10, 24

massive 117:17

match 112:17 matches 130:20 material 6:1 9:22

108:1

matter 14:1 16:1

72:21 85:20 153:18

175:1, 16

max 46:25 47:1, 1

48:1, 14, 24 54:25

55:1

maximum 54:1 82:13

149:12

may 3:21 11:1 13:11,

22 17:1, 1 22:1

27:23 34:25

41:15 43:1 44:1

49:17 51:1 52:17

54:18, 20, 20

58:13 61:1, 1

62:17 67:18 79:1

96:14, 15 98:17

102:1, 1 107:1

108:1 111:15

118:10 119:20,

22 120:20 128:1,

1, 21, 24 132:22

133:18, 18, 23

135:10 145:19

147:1 149:16

151:22 152:1

153:12, 13

162:1, 13

168:12, 13

174:13 177:1, 1

182:25 183:1 185:1

maybe 8:24 11:20

12:1 16:23

19:12, 20 30:1

32:14 34:11 50:1

65:1, 18 73:1 79:1

80:1 110:1

113:15 114:15

118:22 120:10,

10 122:12 123:1

126:1 133:19 134:1

136:21 137:11

139:17 148:1, 1

152:1, 1 154:1

156:1 169:1 174:19

176:1

mean 9:1, 14 10:1,

1, 18 12:17

14:1, 1, 18

16:25 17:11

18:11 20:21

24:23 25:13, 15,

23 27:21 28:15, 18

30:1 33:1 34:1, 1,

10 35:1 37:1, 1, 1

38:11 41:1, 11,

17, 24 42:1, 1, 15

43:25 44:1 46:1

47:1, 1 48:19

49:1, 1, 22

50:20 51:1

53:13, 16 54:1

55:15 56:1, 1, 10,

14 57:11, 17, 19

59:24 61:24 62:19,

23 63:23 64:14

66:25 67:1 68:1, 1

70:1 73:10, 14

74:1 75:25 85:13

111:1, 22 113:13

114:15 115:1

117:11, 20

121:13 122:1

123:11, 19

125:23 127:25, 25,

25 130:1 134:1

141:18 143:1

145:1, 1, 12, 21

147:24 149:21,

21 150:15

152:16, 23

153:1, 1 154:13

168:1, 1, 1 169:23

170:1 171:1 172:13

173:1, 11, 12

175:22 176:1 177:1

178:1 180:1

meandering 128:1

meaning 35:15, 19

131:1

meaningful 122:25

123:1

means 10:1 14:14

22:21 25:1

135:1, 18 150:25

meant 6:12 33:14

35:1 68:22 116:1

measure 21:14 57:12,

12 58:1 158:1

173:17

measured 14:16 15:25

16:1 99:1 173:21

174:15

measurement 33:1

90:10

measurements 16:1

19:24 20:1, 1,

10 54:1 57:1, 1

117:18 125:12

142:25 147:25

measuring 18:22

55:21, 21 72:24

90:11

mechanism 61:24

62:24

mechanisms 63:1, 1

mechanistic 41:19 mecklenburg 133:12 medical 30:18

42:25 53:1, 1

medit 167:1

meet 148:21, 22

149:25 152:25

153:1 158:17

meeting 2:1 89:13

92:1 94:14, 16

101:11 105:13,

14 130:1, 16

131:1, 12 132:18

147:24 148:12,

14 149:21, 22

150:1, 1, 1, 10,

25 151:10 152:20

164:17 166:12

169:13 185:1, 11

meeting's 185:10

meets 130:21 member 4:11 members 3:22, 23

4:14 5:1 7:1 38:16

78:14, 16

ment 37:1

mention 5:1 36:15

37:11 72:13

99:21 129:15

mentioned 4:1

18:14 36:14, 16



37:1, 10 41:1

42:11 72:1 92:11

95:18 96:1

108:20 116:15

127:13, 14

128:22 138:1

146:14 147:16

177:1, 11

message 142:1

met 29:25 153:19

meters 13:1 28:1, 1,

16, 25

method 137:1 139:1

methodologies 141:1 methodology 151:17 methods 3:18 5:1,

1 79:16, 22

80:19 81:15

88:23 95:21

109:1 110:24

114:16 116:1

154:25 169:18

180:12 183:18

metric 29:12 44:25

45:11 100:1

metrics 29:23

70:15 98:11

microenvironmental

100:17

microenvironments

98:1 99:1

microphone 2:12

middle 40:18

midnights 49:1

mike 5:25 6:1 109:1

mind 25:25 30:10

46:18 77:1, 15

82:20 137:14

163:1, 1 169:1

mindful 87:17 88:1

mine 141:11 minimize 12:1 minor 120:17 minute 79:1 92:21

116:14

minutes 7:1 107:1

137:10 148:1

169:23 174:20

176:20 185:1

misinterpret 151:12

misinterpreted

166:18

mislead 166:25

mismatch 85:1, 1, 16

86:13

misread 132:1

miss 169:14

missed 144:1 148:1 missing 70:1 133:1 missouri 87:25 misspoke 31:19
mitigation 150:1

mix 85:1 86:1 164:11

mixture 66:1

mo 149:1 153:1 mobile 117:24 152:1 mod 114:23

model 97:20 99:1, 1,

17, 22, 23, 25

100:1, 12, 15

109:13, 14, 20

114:13 117:14, 21,

23, 24, 25

119:21 120:18

125:14 126:17, 19,

23 127:11, 14, 22,

24 128:17 129:1,

16 134:25

135:15, 18

143:13 161:1

166:15 171:1, 21

172:1, 19, 21

modeled 97:1 99:1

172:20

modeling 97:18

109:12 118:14

120:24 125:13

133:15 134:18,

24 135:1 143:10

147:1 152:11, 23

153:1, 17 172:18

models 105:1

109:15 113:11

117:12 121:1

129:10 138:10

143:18 152:1

166:10

modest 173:16

modification 52:15 modified 19:20 51:17 modify 67:1, 1

moment 4:24 80:21

83:1

mon 8:23

money 26:14 82:1

monitor 13:1 20:25

30:1, 1 47:1

130:21 150:15, 17,

18 151:1 152:22

153:14, 21

monitoring 11:18

12:1, 16 15:11

31:1 32:22 43:1

72:10 76:21

93:14 106:1 111:1,

16, 20, 25

112:11 113:1

116:19 123:10,

12 125:12 130:20

153:15, 20 171:23

monitors 11:14, 23

12:1, 20 14:1,

23 15:1, 18

16:14 17:1, 22

18:18 19:11

20:15 21:14

22:14 28:1, 21

30:1, 11, 11 31:10

32:11 33:1 34:1

47:1 49:16 54:18

76:18 96:20 110:10

113:1, 1, 1

131:11, 17, 18,

19, 20 135:24

146:1 148:25 149:1

150:24 151:1 153:1

monte 143:12 175:1

176:11

months 89:21, 21

91:10 108:20

mor 168:10

morbidity 104:19

168:11, 19

morning 4:1, 1, 12

26:17 31:1 39:1

67:1 70:14 79:24

108:1 118:19

154:22 162:1

163:18 185:1

mornings 123:17

mortality 165:1

168:10

mostly 176:14

move 3:17 7:11



68:1 79:1, 1 92:20

94:17 118:1 120:24

129:19 139:1, 16

145:20 147:11

154:1

moved 134:14

moves 143:11

moving 45:20 49:14

59:17 79:13, 14

88:1 105:13 139:1,

15 160:1

multi 82:25 85:10,

18

multi-pollutant

59:1, 1 65:15

66:14 82:22

83:1, 1 84:19,

20 85:22 86:23

87:10, 22, 24

88:1, 1 105:1

multi-pollutants

65:20 87:1

multi-word 43:22

multiple 86:1 119:15

159:25

mute 134:21, 23

myself 27:25 144:16

N

naaqs 35:1 84:21, 23

90:23 93:21 94:1

name's 80:10

national 106:1 119:1

120:1, 11 121:1,

1, 11, 15, 17,

25 122:1, 1

126:1 161:12

164:1, 1, 10, 14

179:19

nationally 120:1, 13

126:10

nature 28:14

nearest 131:17

necessarily 31:1

45:17 66:17

85:15 86:23 140:13

172:14 173:19

necessity 90:25

negative 35:1 44:15,

17

neglected 3:12

network 130:21

153:15, 20

nevertheless 176:1

nice 121:14 155:1

night 23:10 30:25

40:18 44:1

nine 72:1

nitric 65:25 66:1,

1, 20

nitrogen 37:23 92:25

93:1, 1, 1, 17

94:1 163:1

no's 48:1 75:1

no2 10:19 15:11,

18 16:14 17:1,

22 20:1 21:13

23:22 29:22

30:24 32:1

36:21, 23 41:1

43:24 44:1, 1 46:1

51:1 55:15, 19,

20, 24 56:1, 1

57:1, 1, 1, 12,

15, 17, 23 58:1,

1, 12, 13 62:1, 1,

14 66:10 74:1, 13,

16 75:1, 1 88:19

90:1, 11, 23 91:1,

1, 11, 18, 23

92:1, 1, 13, 25

93:1, 1, 10, 15

94:1, 13 95:1, 1

97:1 98:1 100:1,

25 101:1 106:1

107:22 111:1 115:1

116:19 121:1

128:23 133:16

142:23 156:1

162:21 165:1

168:12 177:1, 1, 1

178:14, 22 179:1

no2's 23:1

nobody's 36:25

nodding 7:1

non 133:13

non-attainment

150:14 152:17

non-near 149:1

none 5:1 18:1 150:22

noon 31:1

nor 36:21

normal 24:1, 18

25:1, 1 62:1

normally 25:1

north 87:24 northeastern 110:13 note 4:1 68:25 115:1

133:1

noted 42:10

notes 6:14, 18

147:14

nothing 86:17 111:24

notice 80:15

noticed 132:23 133:1

nowhere 150:23

nox 7:25 10:17

11:1 22:12, 12

29:17 32:16

36:18 52:1 55:1,

14, 15, 22, 23

56:10 57:1, 1, 12,

13 58:13, 17 81:1,

1, 1 128:23

132:14, 15, 24

133:1, 1, 15

135:20 165:1

167:1, 15

nrc 87:18

nuances 121:1

nugent 3:15 4:1,

1, 1, 1, 17, 20,

23 5:1, 11, 23

39:15, 17 162:1

169:17 185:1

O

oaqps 88:20 92:1 oar 154:10 objection 56:14

occasionally 162:25 occur 23:10 64:1 occurred 178:12 occurrences 100:1,

22 102:24 119:19

162:15

occurs 46:13 49:25

observation 84:24

85:1, 15 86:15

120:23 175:24

observations 16:10

83:1, 20 86:11

164:23

obtained 100:1



107:22

obvious 52:10

53:22 71:1 166:20

obviously 83:1 84:1,

20 92:25 94:1

103:19 107:20

115:19 143:25

odd 110:1

october 2:1 5:18

off-highway 134:1

offer 71:21 81:1, 15

82:22 85:1

86:10, 16 155:11

156:1 183:1

offered 179:20

offhand 27:16

office 79:1 80:1, 12

82:1 84:22 87:23

155:1

offices 34:1

oftentimes 178:22

oh 37:1 42:18

58:24 60:1 63:17

64:13 76:1 78:1,

13 80:24 99:21

106:20 110:1 121:1

137:1, 10 139:10

141:23 148:1 151:1

157:10 162:1

169:21 180:1, 20

183:13 184:13, 19

okay 2:13 4:1, 23

5:1 6:1, 1 7:1, 1,

11, 18 8:1, 19 9:1

10:1 11:1 12:1

13:16, 20 21:25

22:24 24:25 26:23,

24 28:19 30:14

34:15 36:11, 13

37:1, 1, 1 38:1,

1, 10 39:1, 20

40:1, 1 42:10 55:1

57:1 60:1, 16

61:14 63:17, 19,

22 64:1, 1 65:12

66:22 67:1, 20

75:1 78:13 79:12

80:1 88:17, 18

89:1, 10, 23

95:1 102:1

106:1, 14, 20

107:11, 13, 16, 17

109:1, 23

110:19, 23 112:1

117:1 118:1, 1

119:10, 14, 25

125:15, 16 127:1

129:1 134:12, 13

136:19, 24

137:21 138:1

139:1, 25 141:25

142:1 144:15 145:1

146:21 148:1, 1

150:21 151:1, 1,

1, 1 152:14

153:22, 23, 25

155:15 157:1

159:11 160:20, 22,

25 162:1, 1 164:15

165:11, 18

168:19 170:1, 18

176:22 179:1

180:1, 15 182:11

183:1, 16 184:1,

1, 10, 13, 19, 19,

22, 24, 25

one's 47:11 62:16

one-hour 25:19 45:1,

1 46:25 47:1, 1, 1

48:24 53:25

54:1, 25 55:1

72:20 73:10 92:1

102:11 111:23

125:1, 1 135:18

155:20

ones 31:21, 23, 24

32:1 37:22 41:1, 1

80:18 111:16 140:1

161:1

open 15:13, 14

31:1 52:1 62:1

108:10 114:1, 1

142:12 155:1

165:21

opening 88:15, 16

94:1

opinion 51:1 70:1

opportunity 37:12

79:20

opposed 11:24 27:1

33:1 124:1 127:1

130:1 145:23

opposite 12:1

option 128:10

ord 176:10

order 28:25 47:16

50:1 68:16 82:1, 1

174:1 184:12

ordered 89:20

orders 47:17 organization 158:20 organize 56:20 organized 39:24 oriented
85:11 original 90:1, 1, 12

119:12

originally 86:13

89:22

orphan 127:25

ospm 117:13, 24

others 41:18 43:1,

10 62:17 79:13

80:1 118:21 119:1,

1 124:14 143:24

146:1

otherwise 38:14 46:1

112:1 114:25

131:24 149:1

166:25

ought 10:22 21:1

51:12 72:22, 25

73:1 74:16 77:12

164:21

over-emphasized

12:11

overall 13:23 140:19

141:1

overemphasized 13:25

overestimate 166:15 overestimates 32:1 overlap 6:24 overlapping 136:1,

1, 14

overlay 114:18

overstate 43:1

overview 86:18

100:21 155:17

176:25

overwhelmingly 44:22

ourselves 86:1

182:24

outcome 45:21 62:1

97:1, 10 98:12

104:1

outcomes 36:21



38:23, 25 44:1

61:21 102:24

157:20, 25

158:11 161:16

outdoor 12:17 22:13,

19 27:17 32:24

33:1 97:20

outdoors 8:13

96:16 126:24

outlined 96:18

output 97:1 98:10

100:1

outputs 105:1

outside 12:16

69:21 133:13, 23

174:15

oxidant 59:1 62:15

oxidative 41:19

oxide 65:25 66:1, 1,

20 93:1, 1, 17

94:1 163:1

oxides 92:25 93:1

ozone 45:1 66:14

74:1, 10, 11,

13, 15 75:1, 1

103:1 115:1 120:15

133:13 138:24

148:19 153:12

160:1 167:1, 15

168:14 171:1

177:23 178:15, 23

P

page 54:24 59:15

61:17 144:17, 19

155:18

palliative 66:1, 1

p.m 185:12

panel 3:22, 23 60:14

118:24

pangs 137:14

paper 37:24 44:10

128:16 138:24

papers 37:14, 17

paradigms 11:1

paragraph 32:14 33:1

44:21

paragraphs 2:1

parallelogram 121:22

parameter 18:19

124:10 140:17

parcel 150:1

park 13:1 parochial 32:10 partial 61:18 partially 138:1 participate
79:19

participating 79:1

80:1

particle 93:24

particular 9:23

28:19 96:15 98:1

99:1 100:1

104:23 119:22

128:12 157:20

164:1 175:25

particularly 59:1,

10 144:1 149:1

161:1 164:24

166:1, 11

particulars 99:12

particulate 85:19

93:1, 16, 17, 20

94:1

particulates 76:1

passed 2:15

passive 116:19

126:18

past 68:20 126:16

150:24 177:12, 14,

22 182:13

pasted 63:16

path 132:12 pathways 41:20 pattern 23:19 patterns 99:14

135:21

peak 23:14 47:23

48:1, 12 53:25

57:11, 11 73:24,

25 75:10 90:18

91:18 95:12 97:1

107:20 110:1

111:23 119:19, 21,

23 120:1, 1 146:14

peaks 8:1 23:1 47:14

49:17 133:18

161:13

peer 3:10 77:1

penetrated 126:25

people 3:18 5:25

6:1, 22 7:1 8:12

14:11, 13 16:1

17:1, 1 20:19 21:1

27:1 29:1, 1

32:1 37:13 38:1

46:1 50:19 59:12

60:1, 17 68:1,

11 69:1, 18

70:16 71:1, 1,

13 75:21 76:22

77:24 82:1 85:14

98:1 100:1, 23

108:12 124:13

129:10 130:1 132:1

136:22 142:10

147:20 149:25

155:1 164:14, 18

167:1 169:23

170:23 173:23

176:14, 15

178:1, 1, 22

per 17:24 24:14

26:1, 1 44:1 49:18

54:10, 11 105:11

110:12, 17 146:15

perceive 85:1

percent 8:11

10:19, 20 18:1,

1 20:18 58:10

105:11 133:1

165:12

percentile 24:15, 16

54:1, 1, 10 139:10

174:14

percentiles 140:1

perception 41:22 perfect 25:1, 1 63:1 perform 81:10 performed 96:25
perhaps 34:1 69:1

84:25 86:12

136:1 156:1 172:10

174:24 181:1, 1

period 3:21 64:16

periods 23:1 91:1 persistent 179:1 person 3:1 42:24

70:19, 22 81:1

184:21

personal 11:1

18:23 22:11, 12,

19 27:1, 1 47:1

70:1 100:1, 16



125:12

personally 45:19

69:16

persons 97:15

perspective 49:10

89:1 90:1 127:1

pharmacodynamic

174:1

phase 84:1, 1

phases 115:11

philadelphia 96:23

110:11 113:14

122:10 150:20

179:22

pick 12:1 175:17

picked 179:16 picture 179:19 phoenix 179:14, 19 phone 3:13, 19,

20, 22, 23 4:1, 19

5:13 6:1 31:1

39:12, 17, 18 67:1

78:1, 12 81:1

106:15 110:21

134:21 142:1

153:24 157:1 162:1

168:20 183:11

piece 169:25

pieces 69:25 pilot 87:24 pinkerton 132:22

134:12 162:1, 1

163:22 176:24

178:21

pinto 28:17 34:1

physiology 99:12,

13, 14

places 47:1 50:1

115:20 131:1

152:25

placing 13:1

plan 36:1 80:19

81:15 83:1 87:1, 1

88:23 89:1 93:18

94:23 107:1

119:1 121:23

124:24 126:12

127:13 137:19

165:22, 23

180:1, 10, 19, 21

planned 92:22

planning 80:13

81:1 87:24 88:1,

1, 1 93:10 94:13

127:18

plans 87:22

plant 133:14 plants 133:11 plaus 62:23 plausibility

29:20, 25 32:1

51:10 53:10 61:19,

23 66:19

plausible 53:1

player 43:25

plays 150:1

please 3:22 5:12, 14

37:15, 18

pleased 118:1 161:25

plenty 53:12

pm 85:23 93:23

94:1 103:1 105:1

148:19 167:1, 16

168:14

plot 26:1 27:11

plots 23:20 24:1 plotted 24:10, 17 plotting 24:1

plug 138:10

point 9:20 10:1,

14 20:1 21:20 22:1

24:13 28:19

33:13 36:1 39:1

40:1 49:10 50:1

54:1 59:1 60:1

66:20 67:1 72:1, 1

74:20 76:1 81:14

88:1 89:12, 21

93:17, 18, 19 94:1

105:1 114:12 116:1

120:23 123:1 127:1

132:20 133:16

139:1 141:11 147:1

151:1 153:1 159:12

161:21 163:1, 1

168:1 182:1, 1, 13

183:23 184:1

pointed 28:1 38:16

42:1 168:13

points 2:19 11:1

24:20 67:1 68:1

86:15, 15 149:17

161:1

policy 9:15 51:23

68:14 77:1, 11

pollutant 65:20

83:1, 16 85:11,

18, 19 86:1, 22,

25 87:1, 1, 15

153:13

pollutants 32:16

36:23 59:1, 1

65:18 83:14, 16,

18 84:12, 13 85:1,

1, 1, 12, 21

86:1 87:1, 1,

12, 12 88:1

90:10 164:12

pollution 83:20

polymorphisms 41:15

ppb 28:1 29:1, 1

46:15, 16, 21, 22,

24, 24 47:16, 21

48:1, 23 49:1 50:1

53:13 54:1 130:22

poor 176:1

ppd 26:20

popped 37:17

population 17:15, 16

26:20 31:20, 25

32:1 33:24, 24

60:22 61:10 62:1

64:18, 19, 20

95:23 101:25

102:21 103:1

134:1, 1 147:1

153:18 163:12

174:14 175:20, 25

populations 40:12

41:1 42:1 60:1

61:1, 20 70:16

95:22 98:1 103:1

106:1 157:21 163:1

port 87:18

portion 122:1, 11

147:1

portions 10:25

128:20

position 84:1

positive 35:1 44:16,

17 183:19

possibilities 146:16

possibility 161:11

165:15

possible 12:21 27:11



70:20 119:1, 24

135:1 146:23

151:18 152:15

174:18 176:1, 1

possibly 82:14 96:23

118:22, 22

121:11 169:1

postlethwait 10:13

11:1 26:25 27:10

34:24 41:23 42:1

43:20 59:19, 23

60:11 62:1, 16, 20

63:1, 11, 15, 23

64:1, 1 65:1

168:22

potential 17:1

32:1 91:1, 1 92:13

94:16 96:1

100:24 101:11,

15 102:1 105:14

107:23 133:23

162:18 163:1

potentially 106:1

111:13 135:1

power 133:11, 14

powerful 52:1

ppm 35:12 53:14

91:11, 13, 23

92:16 102:11 130:1

practical 25:25

precise 33:1 48:1

181:1

precisely 33:10

preclude 86:23 88:1 predicting 119:21 predictions

117:18, 22 135:14,

15 143:1

prefer 57:1 169:1

preliminary 84:1

103:12 104:14

122:1 156:25

168:11 170:20, 22

prepared 133:1

142:14 154:1

present 4:15 5:1

10:13 11:1 94:1

106:11 141:1

166:19, 24

presentation 86:18

88:16

presented 89:10

160:1

presenting 118:19

press 134:22

presumably 142:25

157:22 158:12

pretty 11:17 43:16

45:16 66:22 91:12,

12 117:19 121:1

138:1 142:22

163:14

prevent 88:1

previous 55:1 68:1

71:1 89:1 109:18

136:1 139:1 142:18

173:13 179:17

previously 100:1

110:16

primarily 15:1 44:20

48:16 97:1

125:19 133:1

primary 51:20 52:1

81:1 83:1 123:10

144:10

prior 90:1 95:12

118:1 129:1, 1

139:20 173:22

prioritize 81:22

priority 164:1, 1

privilege 80:1

pro 116:1

probability 24:1, 17

109:22 141:1

175:17

probably 6:1 23:14

28:1 33:23 40:1

48:1, 1 53:19

55:22 59:1 63:1

88:21 114:21

122:1, 11 124:1

135:22 142:16

144:1, 25 145:1

161:20 167:1 173:1

183:19

problem 12:19 16:11,

19, 22 17:1, 1

18:11, 22, 25,

25 20:1, 10, 24

21:1, 24 32:17

33:11 35:20

44:24 45:16, 20

50:1 59:1 74:17

117:1, 1 127:1

147:1 152:1 164:20

166:20 168:1

169:22 176:16

problematic 148:10

162:15

problems 38:1 43:1

52:12, 13 67:14

133:23 164:13

167:22

procedures 148:20

proceed 5:1 104:25 proceeds 4:14 process 36:1 67:1

76:23 77:25

78:23 79:19, 21

84:1 104:11 180:19

181:1

produce 84:18 87:1

144:1 178:1 180:19

produced 160:11

producing 178:1 product 74:11 productive 169:11 products 44:21 45:11

51:21

profile 74:10

profound 48:19

program 34:1

programs 84:23 85:10 progress 137:13 progressing 96:1 progression 139:23
progressions 142:21 project 87:23

143:1 164:14

176:13

projected 102:23

projecting 182:24 projection 122:1, 16 proper 77:18 proportion 13:1
proportionally 28:1 proportionately

131:10 148:24

proposal 94:23

propose 5:1 105:1

proposed 80:16, 16

82:10 88:23

89:1, 22 94:23

101:14 108:18,

24 109:1 114:23



166:14 177:14

proposing 24:1 96:10

119:14 128:13

protect 74:25

90:18 91:14 94:1

protection 2:1 91:18

protective 15:1 protons 127:16 prove 41:13 provide 38:1, 1

51:25 77:17

79:22 87:1 91:18

101:1 103:13

105:1, 23 106:1

116:16 126:13

127:1 161:22

183:22

provided 72:15 93:22

104:11

providing 5:16 90:1

proximity 13:21

17:24, 25 47:1

public 2:1 3:21 4:1,

10, 11, 15 5:1,

1 15:1 51:1

52:22 77:22

81:12 82:1

115:19 166:25

180:11 185:11

published 37:24

41:18 128:16

pulled 6:22

158:1, 16

qualitatively 102:22

113:20 114:1

quality 42:24 60:1

80:13 85:10 87:17,

21 92:1 95:19

96:1, 1, 19, 20

98:13 100:23

101:1, 11, 18

102:1, 1, 25

105:19, 25

108:1, 1, 13 109:1

110:24 112:1, 10

116:12 120:1

121:1, 16, 24

122:1, 1, 1 127:12

129:18 130:18

131:1, 25 133:10

134:18, 25 135:1

138:13 145:11,

23 148:14, 20

149:1, 11

150:15, 16, 23

151:20 171:1

181:18, 18

quantified 158:13

quantifies 25:18

quantify 105:21

112:23, 23

quantitate 177:16

178:1

quantitative 7:23

64:10 67:19

69:22 83:1 86:20

107:15 109:1, 1,

11, 16 110:1 112:1

114:1, 1 116:25

130:1 131:10

132:23 136:13

143:1 145:1, 25

147:23 149:1, 20

159:20 168:1 171:1

174:22 176:25

178:11 179:11

questionable 114:25

questions 7:15

9:24 34:18, 20, 21

61:1 78:22

106:1, 1, 13, 22

108:11 129:21

140:1, 1 142:20

146:12 147:20

161:20 163:10

170:14

quick 22:1 61:17

168:24 169:1, 1

quickly 2:21 154:11

quite 6:24 18:16

24:1 30:22 69:17

71:1 78:25 79:23

116:17, 18 138:10,

11 139:1 141:1

156:23 176:20

quorum 78:14

pure 57:23

19:23 84:10 95:1	 	



purpose 89:1 96:1

97:13 130:12

135:1, 10 136:1

142:17 143:1

purposes 135:1

pursued 108:19 push 115:10 pushing 47:25 puts 29:21 putting 170:20

181:20

puzzled 160:18

Q

qu 153:1

qualitative 27:12

96:1 97:1 114:15

140:13 145:18

96:1 100:12 103:24

140:14 141:1, 10

145:17, 24 158:10,

13, 18, 22, 22

159:1 164:1 176:17

177:16 178:1

quantitatively 8:1

105:24 114:22

145:13, 21

question 6:23, 25

7:15, 16, 16,

17, 18, 18 8:1, 21

10:10, 12 11:12

22:25 25:25 32:1

38:1, 1 39:1, 10

40:1, 10, 11, 14

41:24 42:16 53:13,

14 59:25 60:19

61:15, 17 63:1

R

ra 148:19 raise 69:20 raising 67:17

random 19:1 21:23

range 49:1 73:19

102:12 103:17

105:1 138:1

170:22, 25

171:12 172:1

174:15 175:21,

23 182:1, 15

rap 137:16, 17

rare 162:15 rate 27:23 rates 7:23 8:1

rather 37:19 42:15

65:20 69:1 70:1

93:24 140:25



161:12 174:20

ratio 13:23 22:19

rational 77:18

rationale 74:12

77:13 93:21

ratios 99:1

re 67:1

reach 133:20 138:1

170:1 182:20

reached 79:1 163:1

reaction 131:22 reader 10:23 36:1 readily 79:1 reading 3:11, 14

7:10 18:15 67:1

ready 7:11 137:10

real 25:24, 25 49:23

50:11, 12 80:1, 18

114:11 146:17

realistic 151:15, 16

realize 118:10

153:11

really 8:16, 17

10:1, 1 12:18

13:24 15:19

27:1, 12 29:1 30:1

41:1, 1 43:21

44:11, 23 45:24

50:21, 25 55:23

56:1 58:14 66:1, 1

68:13, 15 70:10

71:1, 10 74:1 76:1

77:1 85:1 86:11

87:10 91:15 93:1

110:1 114:1, 1

120:11 123:10

125:1 126:1

129:23, 24 132:1

134:22 135:1,

21, 22, 25

138:1, 1 139:11

141:12 142:21

146:12 149:11

161:13 162:10, 23,

24 163:1, 24

164:21 166:17

168:10 169:11

182:23 184:16

realm 140:1

reason 29:1 37:17

38:12 76:13, 16

137:24 146:1

165:15

reasonable 108:18,

25 128:1 152:14

reasonably 117:20

121:15 151:19

152:15 182:1

reasons 81:18

115:1 127:21 159:1

reassessed 94:1

rec 148:17

recall 18:15 receive 16:1 184:1 receiving 39:1 recent 36:19 94:13

101:1 130:17

131:1, 1 149:11

recently 127:22

128:16

recognize 43:1 46:1

recognizes 51:20

recognizing 43:1

84:16 118:1 121:1

recommend 158:20, 24

recommendation

12:1 72:19 73:1

recommendations

87:20 118:1 129:1

recommended 127:20

129:15

recommending 160:18

reconcile 12:21 reconciliation 156:1 record 5:17 26:12

28:1 75:1, 15

recurring 94:11

red 11:20 redraft 56:20 reduce 139:22 reduced 141:1

redundant 42:1, 1

94:1

refer 38:1

references 10:1

37:15, 19

referred 37:22

refers 64:24, 25 refine 84:17 refined 84:1 99:10

110:15

refinement 69:1

reflect 84:10

reflected 22:1 69:1

regard 20:11 46:1

61:20 83:24

163:1 177:1

regarding 90:24 93:1

regards 118:18 region 110:13 regional 150:11 regression 24:20

25:13 109:21

regular 31:1

regularities 122:1 regulations 63:1 regulatory 150:1 reinforce 144:1
reiterate 5:16 relate 34:21

125:22 173:1

related 8:1 33:10

73:21 83:1, 1,

17 104:19, 21

113:10 126:14

165:1 177:1

relates 107:19

182:22

relating 4:11

relation 34:1

relationship 53:24

58:12, 16 70:20

92:1 95:11

97:14, 24 111:1,

22 178:1

relationships

12:22 22:13

49:19 95:1 97:1

103:14, 17

111:14 113:12

131:19 145:16

151:1

relative 10:17 17:22

18:1 25:19 59:13

85:25 86:1

93:12, 13 97:25

relatively 91:1, 1

relevant 9:15 17:14,

17 35:18, 23 48:13

68:1, 20 70:11

71:23, 24 72:1,

1 111:16, 21

115:14 167:19

remains 159:1



remarks 5:1

remember 14:22

21:1 23:19 43:23

44:1 74:1 171:1

174:1 178:12

remind 39:1 76:22

removed 136:11 reorganization 68:1 repair 64:1 65:1, 1 repeat 52:15
repeated 175:15 repor 57:21

report 70:12 87:21

105:23 108:1

126:20 180:1, 15

181:11, 12, 21

reported 57:22

reporting 35:1 represent 60:1 representation

179:19

representative 12:16

31:20 110:13

135:25

representatives

154:10

representing 46:1

105:1 142:20

represents 24:12

require 176:1 required 104:1 requires 87:1 129:10 research 85:23

110:17

residential 113:14

114:10, 25

residing 98:1

resistant 121:13 resolution 98:23 resolve 166:23 resolved 98:13
resonated 142:18 resource 67:24

68:1 82:1, 17

135:1 176:16

resources 118:11

164:11, 20 166:1

176:1

respect 8:1 21:20

143:10 158:1

respective 22:13

respiratory 46:1

104:19, 21, 22

respond 38:13

87:16 119:1 136:13

responds 103:22

response 7:19 11:12,

15 12:22, 24

22:1 32:1 33:1

41:16 86:21

91:24 101:22, 23

102:21 103:14,

17 104:1 112:18

115:1 121:20

136:12 145:16

156:24 173:20

175:1 178:1, 25

183:18

responses 156:17

responsive 62:13, 13

responsiveness

173:19 174:1

rest 10:19 140:1

result 42:19 43:1

92:11

results 21:17, 19

38:22, 23 84:18

105:1, 10 112:1

131:24 138:13, 17,

20 157:23 166:24

167:18 180:13

181:22 182:1

retired 127:23

returning 107:1

review 3:10 5:22

22:1 35:22 39:1

72:19 77:1 78:1

81:1, 1, 1, 1 82:1

83:1, 1 88:20 89:1

90:14, 23 91:16

92:20, 21 93:1,

18, 22 94:1, 1

95:12, 15 101:16

109:18 128:12

178:1 182:1 183:1

reviewed 55:12

158:11 180:24

reviewing 136:1

reviews 84:22 115:1

revised 101:17

103:22 180:12

revision 164:1

ri 33:1

richmond 80:22, 23

100:19 112:1, 1

115:1 116:24

117:1, 1, 1 126:16

127:1 130:11

132:1, 10 133:1

138:23 139:1 145:1

146:11 148:1, 1

149:1 150:1, 22

151:1, 25 153:1, 1

156:24 159:12,

13 160:12 168:1

170:17, 19

171:1, 10, 16,

25 172:1, 11, 19

173:1 175:11 176:1

177:10 178:17

180:1, 1, 17,

21, 25

rightly 108:12

ringer 154:24

risk 46:1 59:1 60:1,

1 67:23, 25

68:15 79:16, 23

80:14, 19 81:1, 18

83:1 84:1, 1, 1,

11 88:25 89:16, 18

92:22 100:20, 21

101:1, 25 103:1,

24 104:1, 25

105:1, 16, 22

108:1 111:18 112:1

113:1 118:15

130:1, 16, 17

131:24 132:1, 1

138:1, 24 139:12

141:11, 18

144:21 145:10, 17,

24 148:12 149:10

151:10, 11

155:1, 10, 16,

21 157:1, 18

158:1, 1, 1, 10,

13, 22, 25 159:1

161:1, 22 167:11

176:25 177:1,

16, 16, 16

180:1, 14

181:10, 12, 21

182:1, 1, 15

183:17

risks 89:1 126:13,

14 132:18 166:15



181:15 182:19

road 13:1, 1 18:1

113:12 115:24

125:1 131:17

147:1, 17 180:1

roads 13:21 31:24

33:22 47:1

120:18 135:21, 23

roadside 28:21

30:1 124:1

roadsides 29:1, 1

roadway 13:23, 24

17:24 52:1 59:18

60:17 96:11 97:17,

17 99:16 110:15

113:10 123:12, 20,

21 124:12 125:1,

1, 1 128:18 129:12

131:19 149:1, 1

roadways 13:17, 17

17:25 49:17

54:19 61:1 98:1,

15 121:10 127:17

129:1, 1 133:1

147:1

robust 62:1

rockies 40:18 59:20,

22

rogene 2:11 3:15 5:1

6:1, 1 34:14

36:1 39:15 42:13

55:1 80:10

106:23 134:16

144:14 149:16

157:1 168:1 183:1,

22

roll 3:12 131:1,

11 151:17, 22

rolled 152:1

rolling 131:17, 18

148:22

ron 7:20 28:1

36:11 37:10 38:1

39:20 55:1 56:10

79:13 129:20 151:1

ron's 38:12

ronald 165:20 roof 114:10 room 3:19 78:1

ross 9:20 34:1 51:18

54:15, 22 55:13

60:21 67:18, 21

73:14 76:22

round 56:23 69:1

88:16

route 127:15

rtp 81:1

rule 73:15 80:16,

16, 17

rules 82:10

run 108:1

running 65:11 rural 133:1 134:1 rush 184:11 russell 11:11

13:1, 12 28:1

30:1, 1 31:19,

23 42:13, 18 52:11

75:11 118:1

119:10, 25

120:1, 21 121:19

125:18, 24 127:1

168:24 169:1

 	S

saint 88:1

samet 43:18 71:1

78:12 155:1 169:14

sample 105:1

san 14:1

sandwich 154:15 satisfied 108:24 saturation 116:19

117:1

scale 24:17 34:12

54:1 106:1

121:1, 12, 15, 17

scan 71:25

save 142:10

saw 47:21 95:14

156:25

scenario 130:1,

18, 19 132:16

140:15 150:1, 1

151:15, 16 152:10,

16, 20

scenarios 100:25

101:1, 1 150:25

151:23 153:14

181:20, 23

schedule 89:1, 10

137:12

scheduled 107:1

169:19

scheduling 180:18

185:1

school 123:24

science 72:22 74:14,

15 78:1 79:18

81:24 82:14 83:11,

24 181:24 182:1

183:1

scientific 2:1 65:25

81:20 83:1

scientist 68:17, 18

scope 80:19 81:15

88:23 89:1 92:21

95:21 118:1 155:17

score 24:10

scott 80:21 88:19

94:25 101:1

scott's 88:14

screen 113:20 133:22

screening 139:1

177:19

se 17:24 44:1

110:12, 18

search 37:16

seasonally 49:1 seasons 49:1 seats 79:12, 14 second 3:16 63:1

64:25 69:1 72:1

84:1 89:15, 17

91:25 93:16 134:14

157:1, 1 163:21

180:1 181:11,

24, 25 182:1, 1,

17, 18 183:1

secondary 81:1

section 11:15, 17

22:18 108:1, 1, 1,

10 110:24 116:12

129:19 134:14,

15 138:1 141:1, 18

147:21 154:25

156:16

seeing 29:1 30:16

103:21 168:16

179:24 180:1, 1, 1

181:21 185:1

seek 78:1, 21

82:16 85:11

seem 66:1 82:1 163:1



178:12

seemed 42:1 52:1

65:13 66:1, 20

67:1 110:1

seems 9:1 16:17

19:1, 10 49:1

55:19 56:21 67:1

71:25 77:12

110:1 111:1, 13

143:11 154:1

156:1, 22 164:1

seen 81:16, 24

159:1, 1 167:13

sees 146:14 172:14

seigneur 127:1 129:1

selected 77:14 96:22

97:1 103:1

110:12 179:11

selecting 55:16

119:17 177:13

selection 35:1

36:1 179:13

semi-quantitative

140:23

send 67:10 162:1

185:1, 1

sense 57:1 60:18

61:1, 1 75:1

96:1 100:12

119:17, 21

127:25 128:1

135:20 139:20

166:13 168:16

173:11 183:24

sensitive 40:12

41:1, 19 60:24

64:17

sensitivity 99:1,

1 143:15 147:11

159:10, 23

160:1, 13 174:14

sent 6:1 63:15

169:15

sentence 36:1 38:14,

21 41:1, 14 42:14,

15 44:1, 10

50:11 51:1, 17, 19

55:1 56:16, 18

58:1 61:18 64:24

sentences 51:20

separable 135:20

separate 22:1 86:1

119:13

separately 112:20

september 89:17 sequence 58:1 series 23:20 48:1

99:23 100:1

serious 19:24 50:15

seriously 164:21

165:15

serves 67:22

service 77:22 ses 61:1 session 154:22 sets 165:1, 1 setting 20:13

34:13 35:18 85:1

112:19 153:11

164:1 179:1

settings 53:1

settle 171:11

seven 40:11, 24

41:24, 25 42:1, 12

58:20, 21 59:1, 15

61:15

several 5:18 15:1

33:16 36:18, 19

48:24 49:18 52:1

71:1 81:16 100:1

102:1, 25 103:1

105:17 113:1

116:20 149:17

156:1

shape 75:12

sharp 163:17 sharpen 83:23 sharply 183:1

sheppard 39:19 40:1,

1 43:15 106:16, 19

134:16, 20

136:21 137:1

142:13 147:22,

23 148:1 149:1

157:1, 12, 16

159:18 160:20, 24

short 4:22 6:10

73:17, 19 74:13

75:22 76:1 90:18

94:18 97:1

101:12 154:1

161:1, 1

short-term 54:23

74:16 76:1, 18

90:15 91:18, 19

92:1, 1, 13

95:1, 1, 11

98:11 100:23

107:22 111:1, 1,

1, 1, 15 117:1

135:17 155:25

156:1, 18 161:13

170:21 177:1

shorter 25:10, 11

73:1 74:1 75:1, 20

108:21

showing 44:19, 23

91:21 178:19

shown 145:19

shows 11:14 132:24

141:16

sign 169:24

significance 164:1

significant 15:1

26:19 55:1

56:11, 25 126:1

140:1 150:1 162:20

173:17

similar 27:1 74:10

103:1 105:1 128:17

144:12 178:23

similarly 18:1

simple 25:25 114:12

simplifications

135:1, 10, 17

simplify 135:13

136:1

simplifying 143:1

simply 6:17 34:25

37:1 115:16, 17

175:17

simulate 132:13

simulating 101:1

153:19

simulation 143:12

175:1

simulations 176:12

sing 105:1

single 86:22 87:1,

1, 12, 14, 15

105:1 106:1 114:24

150:17 171:11

174:21

singular 120:10



sit 125:25 169:23

178:12

site 16:22 20:14, 25

21:1, 14 22:13

31:1 49:16 76:21

113:1, 1

sited 15:1, 12 16:14

76:24 133:11, 13

sites 12:16 15:11

31:1 33:21

111:1, 11, 16

siting 17:1 32:25

76:17

sitting 29:1 30:1

42:20 43:1, 1

47:15 67:1 169:22

situation 152:1

situations 138:11

six 38:1 39:1

40:1, 10 41:24

42:11, 18 77:24

117:24 134:23

size 93:23 105:1

106:1

skill 114:13 117:21

skipped 163:18 skipping 78:1 slam 110:17

slide 89:11 90:20

102:1 103:1 104:14

106:1, 1

slides 88:21 100:20

slightly 12:10

174:19

slip 123:1

small 2:19 6:13

33:13 39:1 54:24

139:14 156:14,

15 167:1

slope 24:23 25:1

138:17

slopes 175:1

smoking 126:18

smoothed 3:1

so2 84:11, 11, 12 socio-economic 60:23 soften 43:22

solicit 89:1 soliciting 89:13 solid 52:14 solution 166:20

space 143:1

sparse 16:1

spatial 29:17 32:15,

21, 25 43:1 47:1

135:19

spatially 13:1 58:14

98:13 113:12

somebody 58:24

68:14, 24 69:18,

22 70:1 183:1

somehow 120:11

someone 146:14

183:11

someplace 58:1

somewhat 42:25 119:1

132:23 173:15

speak 2:12 4:12 5:12

80:20 84:14

speaker 2:1, 11,

14 6:1, 1, 11

7:14, 16 8:20,

23 23:1 26:1, 1,

11, 15, 21 28:12

33:18 40:19

52:24 53:16 80:1

88:12 108:1 183:1,

1

speaking 46:15

special 11:24 32:16 speciate 11:1 speciation 10:17 species 7:25 44:1

55:19 56:1 58:1

66:12 93:1, 11,

14, 20

specific 37:14, 19

61:20 62:1 68:22

73:16 83:17 106:13

107:14 127:17

specifically 90:22

117:13 125:1

specification 140:16

specificity 121:1 specifics 179:11 specified 103:1

157:22 158:23

speizer 144:12

162:12

spend 81:1 98:1

spending 81:1

sorry 40:25 55:14

97:24 109:1 127:21

147:13 157:11

176:21 184:14

sort 10:23 11:14

19:10 32:10, 12

34:10 43:1 46:1

61:11 62:1 66:1, 1

67:1 83:10 84:25

86:1, 1 87:1

111:1, 10

114:12, 19, 24

118:1 130:23

132:11, 17 134:1

140:19, 21, 22, 23

145:1 147:1

154:1 156:1, 1,

12, 17, 18 165:1

166:18 167:10, 17,

24 175:1 179:22

184:1

sorting 70:1

sorts 91:14 121:10

146:12, 15 179:20

spirit 55:1

sound 3:1

sounded 84:25

sounds 35:1 75:21

124:11

source 52:1 95:14

128:10 147:16

sources 16:1

96:15, 15 97:21,

22 98:14, 15 99:25

126:23 127:1

129:1, 1, 1

133:17, 24 134:1

south 122:10, 12

split 60:25 sox 81:1 spoke 185:1

sponsoring 116:15 spot 26:1 92:16 spread 25:12 174:1 spring 89:14 93:20

181:23

st 23:16

stability 128:1 stacks 127:16, 16 staff 108:22

138:24 170:14

stage 84:15, 16 97:1



105:1

standard 8:14

20:13 24:11, 24

28:24, 24 35:18

45:21 62:22 72:1

75:1 76:1, 19

77:10, 11, 12,

14 85:19, 22 86:1,

1 87:1, 15 90:1,

1, 13, 17 91:1, 1,

13, 17 92:1, 12

93:23 94:15

95:10 101:1, 11

105:14, 15 107:19,

22 111:1 130:1, 1,

14, 15, 17, 20

131:1, 1, 1, 12

132:1, 13, 19

145:1 147:25

148:1, 12, 15, 21,

23 149:1, 10, 11

150:1, 1, 1, 10,

18, 25 151:11

152:20, 25

153:1, 11, 12,

21 155:23, 24

156:1 161:18, 19

162:16, 22 164:1

166:12, 14, 14

169:1 174:1 175:22

179:1 181:19

standards 20:14,

21 76:24 80:13

83:1, 1, 1 84:20

85:1, 1, 13

86:22 88:1 94:17

101:12, 13 149:14,

21, 22, 25

150:10 153:15,

19 169:1, 1

179:17, 18

181:16 182:1, 1,

16, 25

standing 124:1

standpoint 27:12

115:19 146:16

164:1

stands 33:11

star 26:16 134:23

start 3:16 66:11

108:1, 13 134:21

143:1 154:1, 1

167:1 171:17

182:24

started 107:18

154:24

starting 7:1 138:1

184:1

starts 165:1

state 87:1

stated 22:1

157:20, 24 177:1

183:1

statement 35:15

43:12 50:20

52:14 93:1

107:21 132:1

statements 19:19

72:1, 1

states 87:1, 14,

24 88:1 142:22

146:23

station 33:15

stationary 129:1

132:15 152:1

stations 123:11,

12 171:23

statistical 105:1

148:18

statistically 173:17

statistics 97:1

105:25 111:11, 20,

25

staying 169:13, 13

step 78:1 84:1

86:1 145:20 184:1

stephen 80:21 88:15,

24 92:23 94:11, 22

100:19 101:1

stephen's 101:19

102:1

steps 184:1 185:1

steve 162:1

sub 61:12

sub-categories 60:22

sub-populations

13:19 30:23

subcommittee 87:19

subject 15:24

51:18 83:1 113:21

submitted 2:1, 16,

22 116:1 146:25

subsequent 181:25

subset 62:1

substance 2:25 3:1

23:16 39:1 40:1

67:10, 15 78:1, 16

substantial 13:24

substantially 8:14 substantiated 156:12 substantive 6:1, 16 suburbs
32:1, 12 success 113:19 successful 62:11 stood 137:1

stop 142:1 144:1

162:13 163:24

164:16

stopped 144:13

story 141:1 stove 27:25 stoves 126:18

sufficiency 90:25

sufficient 40:16

101:21 103:23

177:25

sufficiently 112:21

straight 24:20

straightforward 87:1

175:12

strategies 84:20

85:1, 12 150:1

stratospheric 45:16

street 15:25 21:1

30:13 76:19 135:1

strength 45:1

stress 41:19 stretch 165:16 stretching 164:22 strikes 135:1 striking
137:14 striving 65:19 strong 28:16

43:11, 16 44:1,

19, 22 45:1

51:22 132:1 154:1

stronger 62:10

strongest 42:22 43:1

104:18 155:19

strongly 46:12

163:23

struck 43:1

structure 92:18

143:17



structured 4:13

strug 9:10 struggle 9:12 struggled 71:1 struggling 143:1

suggest 65:18 74:1

98:18 111:22 113:1

137:10 154:1

158:15 166:10

suggested 77:1

178:13

suggesting 56:10

71:25 175:13

suggestion 71:22

77:1 155:19

suggestive 177:1, 17

studied 36:22 41:1

studies 12:25

13:14 14:15 16:1

17:17 18:10

19:25 20:1 21:1,

1, 12, 16, 17

33:1, 1, 12

35:1, 1, 15, 18,

24 36:1, 16, 19,

20 37:1 39:25

42:23 43:1, 1

44:14, 18 45:1

48:1, 20 50:15

54:23, 25, 25

55:1, 25 64:25

73:20, 22, 25 74:1

75:23, 24 90:1, 1,

11, 13 91:21 92:10

102:13 103:13

105:1 112:16, 25

113:1, 16 138:14

146:1, 1 156:1

162:24 166:1

167:1, 11, 13

168:12, 13, 17

171:1, 17 173:1,

10, 14 175:13,

14 178:13, 19

stuff 13:16 27:1

41:24 57:1, 1

167:22 174:1

summarize 33:1 68:19

105:19

summarized 6:23 45:1

summarizing 173:1

summary 66:1 68:1

72:1

superimposed 117:23

superimposition

63:24

support 35:1, 1, 1

45:24 46:15

67:22 77:1 96:1

supported 70:24

128:1 156:12

supports 46:22

52:1 56:24 74:23

supposed 9:1

136:16 153:1

sure 3:19 5:18

11:1 14:1 17:19

18:18 22:1 29:11

35:1 37:10 38:15

46:21 51:24

62:10 63:1, 1, 1

64:1, 13 73:1 74:1

76:14 108:21 110:1

116:21 117:1

125:1, 19 159:20

170:12 171:15,

21 179:1 180:17

surrogate 50:13

55:20, 22 56:1

84:12 93:1, 11

96:1 112:13

survival 165:10

susceptibilities

175:21

susceptibility 40:13

41:1 60:1 62:1

173:20, 23

174:12 175:1

susceptible 42:1

59:13 60:1, 1, 13,

19 61:1, 10, 12

62:1 63:10 95:24

106:1 163:1

susceptibles 64:19

suspect 111:20 symptoms 104:22 systematic 13:10, 13

15:1 19:1, 1

systematically 14:14

15:22 39:23 70:1

T

table 8:18 26:1 54:1

67:1 169:1 173:1

tables 23:23

tail 32:1

tails 25:1

taking 14:19 23:10

74:1 81:1

124:19, 20 137:15

talk 3:20 23:24 24:1

44:1 51:16 54:15

60:1 81:1 83:1

88:24, 25 89:1, 1,

1 90:22 92:21

94:12, 22 125:25

130:1 136:22

144:20

talked 43:1 45:18

59:1 102:1 148:1

156:20 181:14

talking 6:1, 10,

15 14:1 16:15

18:1, 1 23:1

28:1 30:10 43:21

49:1 53:13 67:11

70:14 76:23

82:25 88:20

91:10 95:1

116:20 141:17

142:1 150:16 153:1

172:24 173:1, 19

talks 155:18

tangentially 83:1 targeted 180:14 task 76:1 166:1 teach 107:1, 1 team
81:1

tearing 167:19

tease 83:13 85:24

121:1, 17

technologies

152:1, 1

ted 11:10 16:18

17:21 28:1

116:23 118:1

125:16 168:22

ted's 16:23 17:21

138:1

tele 147:17

template 183:25

temporal 11:1

27:11 29:17 32:15,

20, 25 98:13

135:19, 21



temporally 58:14

113:12

tempted 162:13

ten 7:1 44:15

tend 15:12 34:1 60:1

76:25 166:1, 1

174:1

tended 56:1

tends 13:18 tenfold 165:17 tennessee 90:1 tension 123:1

tentatively 102:10

term 55:16, 23

56:1 57:1 58:1

62:19 73:18, 19

75:20, 22, 22 76:1

95:1 97:1 101:13

120:1 155:22

156:19 161:1 165:1

172:12 177:1

terms 3:10 11:20

34:12 35:1 47:1

48:12 66:12

70:17 82:1 86:1,

11 102:19 103:22

105:1 120:18

141:16 146:16

147:1, 1 177:18,

20, 22 178:1

terry 9:1 75:16,

16 141:1 144:1, 15

text 108:1

thank 2:1, 1, 14

3:15 4:23, 25

5:15, 23, 24

38:1 63:21 64:1

72:14 75:1 78:23

79:12 80:10

81:1, 13 88:11

94:25 95:1 100:18,

19 106:10 109:1,

17 113:1 136:19

141:1 143:23

146:24 151:1

155:1, 15 160:25

162:1 163:16

165:19, 22

166:25 169:12

176:19 181:1

185:1, 10

thanks 5:11 33:18

88:18 109:25

134:12 184:22

that'd 33:25

that'll 90:19

that's 5:22 6:1,

23 8:23 9:10,

12, 13 10:1, 1

12:11, 12 13:1, 1,

1 14:23 15:25

16:11, 15, 23

17:1, 14, 17 20:16

21:1, 12, 18, 24

22:15 23:13, 17,

24, 25 25:15, 21

27:10, 11 30:14

31:14 32:19

33:11 35:10, 19

36:1 39:1 42:1, 1,

15 43:11, 16 45:1,

21 46:11, 20

47:1 48:1 49:1

50:1, 11, 22

51:11, 22, 24

52:14, 20 54:1, 17

56:1, 22 57:1 58:1

59:14, 23, 24

60:19 63:18, 19

66:1, 23 67:25

68:1 69:15, 19

72:1, 23 73:1,

13 76:1 78:17

86:1, 19 88:1

94:20 99:13 100:17

106:21 109:16,

23 111:10

112:14, 19 113:15,

25 115:23 116:1,

14 117:1 119:23

120:1 122:14

123:1, 1, 11

124:1, 19

125:23, 23

126:1, 10 130:10

131:1, 20 132:1,

16 133:21 136:1,

15, 15, 19 137:1

138:25 139:11

141:1, 25 143:1

144:22 146:1

149:15, 23

151:12 152:11,

13 154:18 155:22

157:10 158:19

159:20 160:1, 1

161:23 162:1, 12

163:20 164:25

165:1, 14, 14

166:1, 11 167:13

168:1, 1 169:25

171:16, 25 172:19,

22 173:21, 25

174:15, 16

177:10 179:1

180:14, 16 182:11,

16 183:1, 10

them's 47:11

theme 94:11 themselves 97:15 theoretic 41:1 theory 128:1

there'll 143:1 155:1

there's 10:21 11:12,

15, 17, 21 15:20

16:21 18:11, 22

27:21 32:23

34:19 37:1

40:12, 13, 25 41:1

43:1 44:1 47:1

52:1 53:12, 12, 14

54:24 55:1 59:1

61:25 69:21

74:11 76:13, 16

86:16 101:14

102:14 103:14,

23 108:1 112:1

123:19 126:1 135:1

136:1 137:1

146:12, 15 151:1

155:21 156:16,

16 162:15 166:10

172:1 173:1 174:24

175:20 179:25

180:1

therefore 16:1

158:18

theses 70:1

they'd 116:21

they'll 142:23 143:1

they're 12:1 13:12

15:12 17:23, 24

26:1, 1 31:1 47:19

50:14 54:19 87:1

101:21 122:25

123:1 132:25

133:19 144:1, 21



146:1 155:12

158:14 167:17

they've 170:15

thin 77:23

third 35:1

thirty 74:18 137:10

thorough 164:13

165:24

thoughtful 165:24

thoughts 5:21

82:22 126:12

thousand 33:17

105:11

thousands 85:20

tier 95:18 96:1

97:1, 1, 1, 13

98:1 99:10, 20, 20

101:1, 1, 15, 20

102:1, 1, 1

103:1 104:1, 1, 1,

11 112:1, 1, 1,

1 126:20 127:10

133:10 136:1, 1,

1, 11 139:1, 1, 1,

1, 15, 15, 16, 20,

24, 24, 24 142:19,

22 144:13

145:12, 23

158:1, 15 161:1

162:14, 17, 24,

25, 25 163:24

166:1

tier-two 99:15

tiered 81:17 95:25

101:14 137:1

141:19 144:1

162:13

tiering 81:19

tiers 81:21 95:18

101:19 102:1

136:14 139:1

140:1, 1, 1 142:17

144:1 157:18

158:1, 1, 25 159:1

three-fold 14:23

threefold 15:10

threshold 47:20

166:10

throughout 94:11

throw 62:1 65:10 tiger 147:1, 16 tightened 51:11

tightly 25:10

thurston 20:1

21:11 22:1, 20, 23

38:1, 11 50:23, 25

52:17, 25 53:18

thus 61:20

till 39:14 107:1

tim 4:1 15:1 31:1, 1

46:23 110:20

115:13 116:11

117:1 134:25 143:1

152:19 153:25

tim's 59:16

timely 2:1 tissues 64:16 toad 35:13

today 3:23 82:19

149:13 156:20

159:14

tolerance 178:14, 20

179:1

tons 118:10

top 16:14 61:18

63:25

total 126:13, 13, 20

167:12

totally 70:24 72:22

touched 135:1 149:17

touching 124:1 tough 76:1 130:10 toward 75:22 towards 33:10

70:12 128:25 154:1

tox 37:1 57:1

73:22 74:1 91:1

178:19

toxic 70:17 74:1, 11

toxicities 85:25 toxicity 57:16 86:1 toxicologic 45:18 toxicological

36:16 37:1

toxicology 45:24

46:1, 1, 14, 22

47:18, 22 50:21

53:1, 11, 12 54:13

111:12 178:13

toxicology's 47:14

tracking 44:20 tract 98:22 tradition 145:1

traditionally 128:1

traffic 18:1 23:1

31:24 49:17 113:10

117:24 151:18

trans 96:19

transformation 7:24 transformed 25:1 tremendous 122:15

123:18 166:1

tremendously 13:1

trichloroethylene

176:10, 13

tried 54:22 67:1

105:17 108:12

tries 70:19

trim 167:21, 22 trivial 10:20 132:23 troop 147:10

true 16:16 17:1,

15 19:12 21:12

32:18 41:11

49:17 57:23

58:10 61:11

75:23 144:22 165:1

truly 2:17 85:18

86:1

truth 184:12

try 34:1 55:1 59:1

68:14 73:17, 20,

23 74:1 112:1,

12 121:1 127:1

159:21 160:20

164:19

trying 15:23 22:18

28:18 43:13, 23

50:22 65:22

69:22 71:16 79:1

85:24 86:1, 1, 1

88:1 97:14

106:18 112:17

115:10 121:13

128:1 136:1 143:25

164:13 170:1, 1

turbulence 128:1

turn 3:13 4:24

68:1 82:21 92:23

94:21 126:22

turned 170:11

twelve 23:11

twenty 18:1 23:12

148:1



twenty-four-hour

46:1, 13 49:21

usually 113:1

155:20

124:22 127:22	 	

twice 28:1

two-week 117:1, 1

type 48:15 73:1

74:16 76:17 96:1

128:17 129:1

151:17

types 19:24 20:1

typically 129:1

130:13 148:15

150:10, 13

U

uarg 131:12

u.s 2:1 33:24

105:19, 25 118:25

uh 46:17 59:14 64:11

ultimate 45:21

83:1 159:1

ultimately 20:13

80:15 180:12

ultman 4:1, 1, 1,

1 63:11 142:1

ultra 66:14

un-mute 142:10 una 174:21 unaware 16:13 uncertain 139:22

uncertainties 20:1

146:20 160:10

175:1, 1, 10

176:12

uncertainty 95:25

97:1 99:1

100:11, 14 101:1

102:15 105:1, 1

126:1 139:23

140:1, 1, 13,

14, 15, 17, 20, 23

141:1, 1 143:19

159:1, 15 165:17

174:24

uncomfortable 59:12

uncomfortably 42:21 unconfined 18:1 underestimated 14:16 underlying
143:21

146:17

unders 171:15

understand 14:1 19:1

130:1 132:25

163:13 164:19

172:16

understanding 30:16,

22 31:18 62:22

121:1 124:1 127:15

129:23 138:22

understood 57:1

undertaking 117:17

143:1

unfortunately 15:19

uniform 50:14

173:20, 20

unintentionally

84:25

unique 63:1, 1 64:1,

15, 17, 18, 19

65:1

uniqueness 63:24

unless 34:19 106:12

unlikely 92:14

107:23 132:14,

16 162:1

unreasonable 128:19

up-front 84:1

upon 13:15 14:15

16:1 27:1, 19

60:14 99:19

136:1 164:23

upper 32:1 139:10

140:1

urban 15:1, 1, 21

17:1, 22 102:1

103:1 104:1 105:18

110:10 111:15

113:13, 21

114:1, 14 117:23

133:1, 1 134:1

148:16 150:19

153:1

urbanized 179:21

urge 166:18, 22

useful 7:22 27:12

53:24 65:1 99:1

user 69:12, 13

utility 86:1

101:21 104:10

usual 18:25, 25

24:19

V

value 24:12 45:23

76:1 104:10

111:1 119:15 170:1

values 11:23 25:1

92:1, 1 111:1,

1, 15 140:17

vanessa 154:1 155:1

variability 29:17

32:15, 21, 25

34:1, 1 43:1

100:14 101:1

102:19, 21

123:18 124:1, 17

138:15, 19

143:21 174:1, 1,

12 176:12

variable 16:22 27:17

65:10 173:23

174:1, 21

variances 124:1

variation 123:14

135:20

variations 8:1, 10

variety 73:22

various 50:1 99:1

100:24 103:1

105:20 124:13

167:11, 19 169:1

181:15

vary 29:18

vehicle 125:1 132:15

152:1

vehicles 33:17

vein 151:25 verbiage 35:1 44:1 version 106:1 versus 13:1 17:15

22:12 27:1 35:1

47:1, 11 48:14

54:25 83:18

92:25 123:21, 21

126:13 128:1 160:1

181:11

vertical 15:1, 20

16:22 17:1 114:1

verticality 13:18

16:19 18:1

vertically 13:1 14:1



view 12:1 83:10

views 85:14 131:1 violations 179:17 virtually 53:1 visceral 10:24 visits
42:23

104:1, 22

vote 44:16

vu 154:1, 13

vulnerability 60:10,

25

vulnerable 60:1,

1, 13, 17, 20,

22 61:1, 12

95:24 106:1

W

wait 28:1 116:14

136:21 142:1

169:23

waiting 56:19, 23

157:1 184:11, 18

walk 45:1 76:19

100:20

washed 49:1

wasn't 30:21 35:1

39:13 40:17 63:1

71:15 110:1, 25

120:1 131:1

139:1 160:13 162:1

waste 20:22

water 149:25

ways 32:1 105:10

135:13 146:21

147:1, 1 148:23

176:1

we'd 5:13, 13

34:13 75:13 76:1

101:1, 25 102:16

106:1 115:12

we'll 4:24 5:16 7:18

10:10 22:24

32:19 52:19 55:1

58:24 65:16, 18

74:1 78:24 79:1,

1, 1 81:1, 20

89:14 98:25

101:1 115:24

129:19 136:17

153:16 155:1 162:1

163:19 165:21

170:1 182:16 184:1

we're 2:24 7:1, 1

9:1, 16 15:22

16:15 18:11 21:1

29:1, 1, 10, 13

30:16, 16, 19

43:13 46:15

48:18 49:1, 13, 15

50:21 51:1, 1, 13,

14, 15 56:18,

19, 22 58:18 61:22

63:1 66:25 67:1

72:24 73:1, 1 74:1

76:22 78:1, 1

79:12, 14, 25 83:1

84:1 89:1 91:10

93:10, 10 94:12

97:14 103:19, 19

108:10 112:1, 1,

1, 11, 17, 23

113:16 114:19

115:1, 1, 1, 10

116:20 118:24

121:13 123:16,

17 126:1, 18

130:12, 14

131:21 133:1

136:15 141:17,

20 143:1, 1, 1

150:16 153:1

155:1, 25 156:10

157:1 165:11

167:18 168:19

169:1, 14 170:1, 1

172:1, 11 177:11

178:1

we've 29:11 32:11

33:1, 1 34:16

37:16 43:1 47:23

56:17 59:1 67:1

74:18 79:17

81:1, 17 96:22

97:13 98:11, 16

99:1 103:1

105:1, 17 109:17

114:18 117:12, 13,

13 118:13 126:17

134:13 136:14

148:18 163:1 167:1

169:1 172:15

182:13 183:1

184:1, 15

weak 141:11

website 173:25

week 50:1, 1 81:1

146:15

wegman 80:1, 1,

11, 24 87:16

weight 160:14

weights 160:1

welcome 3:18

well-executed 162:10 well-oiled 118:1 whammy 61:11

whatever 16:1

17:13 31:1

37:16, 18 112:22

114:11 116:17

118:11 122:13

138:18 154:1,

15, 21 177:18

whatever's 11:1

wheeze 104:23 whereas 54:10 whereupon 7:1

26:12 28:1 55:12

75:1, 15 79:10

154:22 185:11

wherever 59:24

whether 5:1 27:16

44:1, 11 56:23

59:25 60:1 81:19

96:20 103:22

109:1, 15 114:1, 1

139:15 145:22

148:24, 25 153:1

155:11

whichever 132:12

who's 3:13, 20 30:17

68:14 134:21 176:1

whoever 40:14

whole 23:1 29:24

38:24 45:15

53:15 68:12

73:22 81:12 118:25

122:18, 19

125:15 140:1

142:12 150:19

152:24

wide 124:1 138:1

166:1

wildest 37:11

windows 114:1

wise 75:25

withstanding 173:13



wonder 55:1 167:14

wondered 109:13 wonderful 63:1 wondering 10:14

30:20 43:10 56:1

75:20 119:1

177:1 179:15

wording 36:17

wordsmithing 39:1, 1

67:12, 14

work 51:11 64:21

80:18, 25 81:1, 1,

10, 11, 13 95:1

116:15, 16 136:1

137:19 140:10

144:23 157:1 173:1

176:1

worked 59:1

working 5:12 69:1

80:14 107:1 137:11

154:1 169:12

works 78:25

workshop 94:1

world 63:1 118:14

146:23 169:1

worries 149:19

166:11

worry 67:15 138:16

166:17

worst 152:1, 14, 22

worth 19:22, 22

123:1 142:16

165:16 181:1

write 4:23 38:16

51:17

writing 69:11 160:21

written 3:1, 1

5:16 7:19 22:14

23:18 37:23 38:1

51:1 58:23, 23, 25

69:1 116:10

140:1 146:25

156:15 159:14

163:1 165:23

169:15, 16 170:10,

11 183:22

wrong 72:24 73:1

74:18

wrongly 108:13

wrote 23:18 40:14

44:1 52:1 58:22,

24 65:13 142:14

wyzga 7:21 10:1 21:1

36:1, 13 37:1, 1

39:10, 12, 20,

21 40:1 55:1, 25

56:12, 17 58:11

129:21 132:1, 1

165:20, 22

Y

yearly 25:19 54:11

yesterday 2:20 4:1

6:15, 18 22:10

29:20, 24 30:21

32:23 33:20 35:1

37:23 39:21 40:1

42:21 43:18, 20

45:19 47:15, 21

52:1 55:18 59:11

65:24 68:22

71:22 77:1 84:1

89:11 92:24

93:1, 13 104:17

156:1, 20 182:23

yesterday's 4:10

66:21

yet 50:12, 22

51:1, 13 63:1

116:1 131:21

137:22 156:19

york 15:23 87:25

110:1, 12 114:18

117:12, 14, 25

150:19, 20 179:21

you'll 123:13

136:18, 25 138:24,

25 141:1 144:1

180:1 182:1, 19

you've 64:1 81:24

103:20 114:1, 1

115:13 121:22, 23,

24, 25 122:1

140:1, 11 166:1,

21 167:1 174:15

youth 77:21

Z

zoom 121:1

 

 

 

US EPA CASAC PUBLIC MEETING 10/25/07 CCR# 15676-2	Page   PAGE  

