1 STATE OF FLORIDA

DIVISION OF ADMINISTRATIVE HEARINGS

Case Nos. 92-3038, 92-3039, 92-3040

SUGAR CANE GROWERS COOPERATIVE )

4 OF FLORIDA, a Florida Agricultural )

Cooperative Marketing Association, )

5 ROTH FARMS, INC., and )

WEDGWORTH FARMS, INC., )

6 )

and )

7 )

FLORIDA SUGAR CANE LEAGUE, INC. and)

8 UNITED STATES SUGAR CORPORATION; )

)

9 and )

)

10 FLORIDA FRUIT AND VEGETABLE )

ASSOCIATION, LEWIS POPE FARMS, )

11 W.E. SCHLECHTER & SONS, INC., )

AND HUNDLEY FARMS, INC., )

12 )

Petitioners, )

13 )

vs. )

14 )

SOUTH FLORIDA WATER MANAGEMENT )

15 DISTRICT, an Agency of the State )

of Florida. )

16 )

and )

17 )

MICCOSUKEE TRIBE OF INDIANS OF )

18 FLORIDA, the UNITED STATES OF )

AMERICA, and FLORIDA DEPARTMENT OF )

19 ENVIRONMENTAL REGULATION, and )

the FLORIDA WILDLIFE FEDERATION, )

20 )

Intervenors. )

21 )

22 99 N.E. 4th Street, 3rd Floor

Miami, Florida

23 February 15, 1994, 9:00 a.m.

25 Deposition of Dr. Steven P. Millard

JACK BESONER & ASSOCIATES, INC.

2 Taken before Lora Lee Knorr, Certified

3 Shorthand Reporter in and for the State of Florida

4 at Large, pursuant to Notice of Taking Deposition

5 filed in the above cause.

6 - - - - - - -

8 APPEARANCES:

ON BEHALF OF THE PETITIONERS:

PEEPLES, EARL & BLANK, P.A.

12 One Biscayne Tower, Suite 3636

Two South Biscayne Boulevard

13 Miami, Florida 33131

BY: Robert H. Blank, Esq.

ON BEHALF OF THE RESPONDENT/INTERVENORS:

UNITED STATES ATTORNEYS OFFICE

16 99 N.E. 4th Street, 3rd Floor

Miami, Florida 33132

17 BY: Thomas A. Fitzgerald,

Assistant United States Attorney

ON BEHALF OF SOUTH FLORIDA WATER MANAGEMENT

POPHAM HAIK

20 SCHNOBRICH & KAUFMAN, LTD.

100 S.E. Second Street

21 P.O. Box 019101

Miami, Florida 33131

22 BY: Jose A. Loredo, Esq.

25 - - - - - - -

JACK BESONER & ASSOCIATES, INC.

1 I N D E X

Witness Direct Cross Redirect Recross

2 Dr. Millard 4

JACK BESONER & ASSOCIATES, INC.

1 Thereupon:

2 DR. STEVEN P. MILLARD,

3 was called as a witness by the Intervenor, and

4 after being first duly sworn, was examined and

5 testified under oath as follows:

6 DIRECT EXAMINATION

7 BY MR. FITZGERALD:

8 Q. Good morning, Doctor. On the record,

9 Tom Fitzgerald on behalf of the

10 Respondent/Intervenors in this matter. You were

11 called for your deposition back in -- it was March

12 of 1993, that the fundamentals, as I recall, you

13 indicated you hadn't testified about.

14 If you want to take a break at any

15 time, just tell me. If you don't understand a

16 question, please advise me, and I'll try and put

17 it into some kind of format that makes sense for

18 you. And if, in the course of the questions, in

19 order to answer a question you need to assume

20 something or need additional parameters in order

21 for to you be able to formulate a reasonable

22 response, please just advise me, and we will try

23 and sort that out as we go along.

24 Doctor, in February of this year, 1994,

25 the firm with whom you maintain a consulting

JACK BESONER & ASSOCIATES, INC.

1 relationship, Peeples, Earl and Blank, filed a

2 document in this matter captioned "Disclosure of

3 Expert Fact Witnesses" on behalf of Florida Sugar

4 Cane League, Incorporated and United States Sugar

5 Corporation. In paragraph 14 they designated you

6 as a potential witness in this matter and

7 identified a subject matter of expected testimony.

8 Since July of 1993, have you had

9 occasion to speak with anyone concerning what the

10 subject matter of your expected testimony will be

11 in this case?

12 A. Yes.

13 Q. And with whom did you speak?

14 A. Mr. Blank.

15 Q. Okay. And did you discuss the areas

16 that you would be designated an expert in or

17 simply look at the areas of work that you were

18 performing on behalf of the firm?

19 A. I talked about both.

20 Q. And what was your understanding as a

21 result of those conversations of the subject

22 matter of expected areas of testimony for you?

23 A. Expected areas of testimony would

24 include my opinion on the limits that are set in

25 Appendix "E", the validity of those limits, how

JACK BESONER & ASSOCIATES, INC.

1 the entry and access data was gathered in the

2 period of '92 on to '93 validates or does not

3 validate or changes the model used in Appendix "E"

4 to set limits for the -- I'm talking about

5 Loxahatchee.

6 Also, with regard to Loxahatchee, I

7 expect I'll testify about whether I've been able

8 to find any sort of correlation between the

9 phosphorus load coming into Loxahatchee and the

10 ambient water phosphorus at the fourteen sampling

11 stations that are used to determine the limit.

12 Let's see, possibly I'll testify on

13 work that I'm expected to do that has to do with

14 looking at phosphorus loads coming out of the EAA

15 and phosphorus loads coming into the Everglades

16 National Park.

17 Q. Did I understand you just now said the

18 work you may do in those areas -- those last two

19 on phosphorus loads out of EAA and phosphorus

20 loads into Everglades National Park?

21 A. That's correct.

22 Q. So you have not commenced that work

23 yet?

24 A. No, I'm still waiting on some data for

25 that work.

JACK BESONER & ASSOCIATES, INC.

1 Q. What data are you awaiting?

2 A. I'm waiting to receive phosphorus --

3 let's see, I'm waiting to receive data from a

4 number of structures that border the EAA and also

5 data that flows into the Everglades, data from

6 structures where the water flows into Everglades

7 National Park. I'm waiting for data that will be

8 in the form of, I believe, monthly loads, because

9 my current understanding -- and I believe I

10 received data from some of those structures a

11 while back.

12 But as I recall, the form of the data

13 was phosphorus loads, say, every other week,

14 something like that, and the kind of sample was

15 not always the same. There were about ten or

16 eleven different samples codes, several of which I

17 think indicated a composite and several of which

18 indicated a grab. And, so, I'm waiting for data

19 from Environmental Services and Permitting where

20 they would be computing what the monthly load is.

21 Based on my understanding -- my

22 understanding is an algorithm -- that is what the

23 district uses.

24 Q. When have you been advised you will

25 receive that data?

JACK BESONER & ASSOCIATES, INC.

1 A. I wasn't given a specific date. I

2 expect it will probably be within a couple of

3 weeks.

4 Q. How long will it take you or how long

5 do you anticipate it will take you to conduct your

6 analysis of phosphorus loads out of the EAA and

7 into ENP once you receive the data, assuming the

8 data is in usable format?

9 A. I would anticipate anywhere from two

10 weeks to a month.

11 Q. So you expect it within, say, two

12 weeks, and then two weeks to a month -- so we're

13 looking at about six weeks. And as soon as you

14 complete that work, are your final opinions

15 manifest then in that subject area upon completion

16 of the work or do you have to sit there and

17 eliminate the scatter plots and curves and the

18 graphs that evidently you experts in this case

19 seem to enjoy producing to the baffling of the

20 attorneys?

21 MR. BLANK: I feel compelled to object

22 to the form of the question.

23 MR. FITZGERALD: Noted, but I think the

24 hearing officer will see the inherent veracity of

25 that.

JACK BESONER & ASSOCIATES, INC.

1 THE WITNESS: No, when I say my estimate

2 of two weeks to a month, that includes coming to

3 an opinion -- final opinion.

4 BY MR. FITZGERALD:

5 Q. So a minimum of six weeks from today --

6 somewhere in the range of four to six weeks from

7 today you may have final opinions on that subject

8 area?

9 A. Uh-huh.

10 Q. What phosphorus trend models have you

11 examined since our March '93 initial deposition?

12 MR. BLANK: If you don't understand the

13 question, Doctor, ask him for clarification.

14 THE WITNESS: Okay. Yes, I guess when

15 you say "phosphorus trend", I guess I need you to

16 be more specific.

17 BY MR. FITZGERALD:

18 Q. Paragraph 14 of the document I've

19 previously described, subparagraph B, says your

20 subject of expected testimony -- the first three

21 words are "phosphorus trend models". So what does

22 that mean to you in the context of the work you've

23 been doing, your consultation with other attorneys

24 -- experts and the attorneys with regard to what

25 your testimony is going to be about in this case?

JACK BESONER & ASSOCIATES, INC.

1 MR. BLANK: You're not compelled to

2 speculate, Doctor. If you don't understand, say

3 so.

4 THE WITNESS: Right. Okay. You could

5 classify the work that I'm doing for Loxahatchee,

6 analyzing the Loxahatchee model and the entry and

7 access data -- analyzing that. You could classify

8 that under phosphorus trend analysis because there

9 is a separation in time between the data that was

10 gathered between '78 and '83 and the data that's

11 -- also entry and access data of '92 to '93.

12 And there -- you could look at the

13 question of whether the phosphorus concentration

14 and the entry and access period is really

15 different from the phosphorus concentration in

16 either the base line period of '78 to '79 or the

17 post base line period of '79 to '83. Whether I'd

18 really call that looking for a trend or -- I

19 probably wouldn't call it looking for a trend.

20 I'd call it looking for a difference just between

21 those two time periods.

22 BY MR. FITZGERALD:

23 Q. Have you completed your work on that

24 particular aspect?

25 A. No, I'm still working on it.

JACK BESONER & ASSOCIATES, INC.

1 Q. When do you on anticipate you'll have

2 final opinions on whether there are differences in

3 phosphorus concentrations from the base line

4 period -- the post base line period and the entry

5 and access period?

6 A. Within a couple of weeks.

7 Q. Well, you're going to testify about

8 statistics I see, and I'm going to skip over that

9 one just as I did in March. That's just a little

10 too generic.

11 You indicate, back in March, that you

12 were not engaged at that time in any STA, storm

13 water treatment area in modelling. Have you

14 engaged in any analysis or modelling the storm

15 water treatment areas since then?

16 A. Would the work that I did for the

17 mediation plan fall under that? I wouldn't think

18 so.

19 Q. No, I don't think so.

20 A. No.

21 Q. Okay. You bring up the mediation plan.

22 When you say that, do you mean the work that's

23 reflected in some of your documents provided by

24 counsel that occurred roughly in the July through

25 October, November time frame of 1993 relating to

JACK BESONER & ASSOCIATES, INC.

1 settlement negotiations in the case?

2 A. Yes.

3 Q. Okay. Let me make one thing clear,

4 Doctor, based on a ruling of the hearing officer,

5 as we understand it, I will not be asking you

6 questions intended to elicit anything involving

7 your work related to the settlement negotiations.

8 However, if in order to answer another question

9 you have to address that, please advise me so we

10 can sort that out before you go into what you did

11 there.

12 The only, I guess, exception to that is

13 your ultimate factual opinions for the areas

14 you're not designated in are, in part, based on

15 some work did you during that period which is just

16 occidental, then you need to tell me that, too, so

17 we can sort that out.

18 I did recognize that in the documents

19 provided there was a fair amount of material

20 related to things like payment schedules and risk

21 assessments and some models which you did there,

22 and I don't intend to ask you about that. So when

23 I say STA models, I mean STA models in the sense

24 that it appears in the SWIM Plan adopted by the

25 board in 1992.

JACK BESONER & ASSOCIATES, INC.

1 A. Okay.

2 Q. So I take it then your answer,

3 essentially, is no.

4 A. That's correct.

5 Q. All right. So when it says STA models,

6 we can sort of ignore that, except incidently the

7 phosphorus concentration limits, I think, you've

8 probably already covered to some degree.

9 Have you done any work related to

10 phosphorus concentration limits other than what

11 you've just described initially about the earlier

12 testimony, in other words, Loxahatchee work, entry

13 and access, that sort of thing?

14 A. I haven't done any work addressing

15 phosphorus concentration limits, no, as stated in

16 the SWIM Plan.

17 Q. Except as previously outlined in your

18 initial remarks; is that correct?

19 A. That's right. That's right, except as

20 I had previously explained.

21 Q. Have you done any analysis or are you

22 in the process of doing anything regarding water

23 quantity trends in the Everglades protection area

24 or the Everglades agricultural area?

25 A. No.

JACK BESONER & ASSOCIATES, INC.

1 Q. Is it your understanding you would be

2 doing any water quality trend work?

3 A. No, I don't believe so.

4 Q. Okay. One area you've indicated of

5 subject matter is the QAQC data base work. What

6 QAQC data work have you done in connection with

7 this case?

8 A. With regard to the entry and access

9 data in Loxahatchee, there were field replicates

10 taken. So that gave me an opportunity to look at

11 how much variability is constructed by just small

12 scale field variability, how much of that -- how

13 much of the overall variability in the data is

14 contributed just by field replicate data. That's

15 just using the ES&P data and the Department of

16 Justice data. I understand field replicates were

17 taken. We were only given averages of those field

18 replicates.

19 Q. Did you compare the ESP data to the

20 federal average data for samples at the same

21 sites, on the same dates, etc., etc.?

22 A. Yes, I have --

23 Q. And what opinions, if any, did you

24 derive from that comparison?

25 A. They don't agree very well.

JACK BESONER & ASSOCIATES, INC.

1 Q. How much variance or what analysis did

2 you subject them to?

3 A. I've done some very simple plots. So

4 far I have looked at -- on a station by station

5 basis. So for each station I've looked at a plot

6 of the ES&P data for each date.

7 So the picture that I have in my head

8 is that I'm -- on the access you have the date

9 12/92, 1/93, 2/93, etc., etc. And you can plot

10 the three ES&P field replicates or make a box

11 based on those three, but it's a joint --

12 basically it's just you look at the three

13 obvious. So you can look at those where on the

14 one access I'm thinking of the logarithm of the

15 total phosphorus.

16 So you can look at those observations,

17 and then you can also plot on the same plot, the

18 Department of Justice observation, which is

19 apparently the average of the three field

20 replicates. And, so, you can just look at that by

21 eye to see where the Department of Justice data

22 follows relative to the ES&P data. So you can

23 make a plot like that for each station.

24 And then another simple plot that I've

25 made is simply looking at the fourteen station

JACK BESONER & ASSOCIATES, INC.

1 average for the ES&P data for each date and

2 comparing that to the fourteen station average

3 with the Department of Justice data.

4 Q. And what were the results of that

5 comparison?

6 A. Ideally, if there was fairly good

7 agreement between the two data sets, you would see

8 something that should tend to cluster along a line

9 that ideally has slope of one and an intercept of

10 zero. But if you do that -- if you make a plot,

11 it looks -- the data are scattered all over the

12 place. It doesn't look like very good fit.

13 There's not very good agreement.

14 Q. Back in March you talked about four

15 different factors that would affect your review of

16 any data, things like, you know, whether the

17 samples were taken in the same fashion, processed

18 the same way, etc. Are those the types of factors

19 that you would look at to account for the

20 variability or variance in the data set -- the two

21 data sets, the ESP and the DOJ data set?

22 A. In order to try and investigate why

23 you're not getting the results that you expect,

24 you need to go back and look at the whole sampling

25 process.

JACK BESONER & ASSOCIATES, INC.

1 Q. Did you encounter any problems with the

2 data set produced by ES&P?

3 A. What do you mean by problems?

4 Q. Well, did you employ all the data in

5 your analysis provided by ESP for the three

6 replicates, on every occasion, from every site?

7 A. When I'm talking about the plot, the

8 plots that I was just describing, the first set of

9 plots where I looked at each station, I compared

10 the Department of Justice data with the three

11 field replicates, what I looked at -- and then

12 when I talked about the plot where I compared the

13 fourteen station average, initially I took the

14 average of the field replicates. This is all

15 after log transforming the data.

16 After taking the log of the data, I

17 took the average of the field replicates at each

18 station on each date, and then I took the fourteen

19 station average. And, so, I compared that against

20 the fourteen station average based on the

21 Department of Justice data.

22 Q. I'm not sure I heard the answer to the

23 question I was trying to ask. Maybe I didn't put

24 it too artfully.

25 Did you or anyone screen the data

JACK BESONER & ASSOCIATES, INC.

1 before those two plot analyses to account for

2 outliers or data that was suspect to -- based on

3 the best professional judgment of anybody involved

4 in the process?

5 A. I have done some -- what's called

6 outlier analysis on the ES&P data. I also have

7 gotten information from John Davis, where he has

8 gone back and looked in log books, the sampling

9 log books, and he has identified approximately

10 five particular observations where it was noted in

11 the log book that it took several passes -- at

12 least two passes through the filter before they

13 felt that the -- whatever was suspended in the

14 water they basically had gotten rid of in order to

15 get a "clean sample".

16 Whenever you see -- so I take it from

17 John Davis, in his best professional judgment, he

18 said that it would possibly be legitimate to leave

19 those particular observations out of an analysis.

20 Q. Okay. On these five -- or whatever

21 number that Dr. Davis decided in his best

22 professional judgment or suspect or whatever

23 reasons, did you then exclude those from the two

24 plots you were discussing?

25 A. I haven't done that yet. I haven't

JACK BESONER & ASSOCIATES, INC.

1 done that yet.

2 Q. Am I correct in my understanding that

3 if the samples were tainted with excess dissolve

4 particularly or they were clogged filters, they

5 had little chunks of fish or titan in that, that

6 would result after the acid digest process used

7 ESP and elevator process for phosphorus --

8 A. That's my understanding.

9 Q. And from whom do you derive that

10 understanding?

11 A. From conversations with Dennis

12 Lettenmaier and John Davis.

13 Q. In your experience doing environmental

14 modelling -- your experience was detailed in the

15 earlier depos in March. Is it unusual to do this

16 type of outlier analysis and exclude some field

17 testing data because of -- it's called obvious

18 problems in the collection or processing process.

19 A. No, it's not unusual to have problems

20 with certain samples in environmental monitoring

21 data or industrial data in any kind of process.

22 Q. When that arises, is it your practice

23 then to use the best available data that the --

24 after your outlier analysis and consultation with

25 people who would have an understanding of the

JACK BESONER & ASSOCIATES, INC.

1 processes involving the parameters you're

2 examining?

3 A. My usual procedure is to use all the

4 data first and do both parametric and

5 nonparametric analysis, if possible. And then

6 based on the conversation with the people who are

7 experts in their field -- experts in the subject

8 matter, and their opinions on what can and cannot

9 be discarded, then I would go back and do the same

10 analysis with that revised data set as well. So I

11 would end up doing several analyses.

12 Q. Would it be possible to take two sets

13 of data that are in a related area, you know, like

14 phosphorus concentrations where the two sets were

15 gathered by different techniques and, therefore,

16 can't be treated as a single base and, through

17 some statistical method, pool that data and

18 examine the data through some pooling process that

19 would give you a feel for the variables that

20 you're interested in?

21 A. Your question is a little too vague for

22 me.

23 Q. All right. Is it possible to look at a

24 specific data set and pool it with a general data

25 set, not unique to the system you're examining, to

JACK BESONER & ASSOCIATES, INC.

1 develop a trend model?

2 A. I don't understand what you mean by

3 specific and general -- a specific data set and

4 general data set.

5 Q. If you have a data base that covers

6 fifty marsh systems or wetlands systems, and

7 phosphorus data is available in that universal

8 data set -- constructive data set, ESP goes out to

9 Loxahatchee, for example, and develops the data

10 set that you've examined, is it possible to pool

11 that data and learn something or get a feel for

12 the functioning of the systems and their

13 phosphorus processes that are involved?

14 A. It's certainly possible to combine data

15 that have been gathered from several different

16 areas, using several different techniques, but you

17 have to be very careful about reaching any

18 conclusions.

19 Q. Why is that?

20 A. Well, because of all the factors that

21 are involved. If you have data gathered from

22 several different marsh systems, and you're trying

23 to pool them to try and figure out some general

24 law, it may be that "general law" operates

25 different in the different systems. And if you

JACK BESONER & ASSOCIATES, INC.

1 pool all the data together and come up with some

2 supposedly general law based on that pool data, it

3 may be that the only reason you see that is

4 because the general law holds in maybe eight out

5 of ten cases, and there are the other two cases

6 where it maybe works the opposite way. Those are

7 overshadowed by the fact that you've got eight out

8 of ten which work one way so that the data masks

9 out the effects of the other two.

10 Q. If you wanted to data to see if -- say

11 the model you were developing for your specific

12 situation, in fact, bears a relationship with the

13 general data set, could you, as Dr. Davis did with

14 the Loxahatchee data, could you prescreen the

15 other data sets -- other eight or ten systems with

16 people knowledgeable in the appropriate

17 disciplines and the processing of the variables

18 you're trying to predict and maybe eliminate some

19 of those that would mask and give you an erroneous

20 result?

21 A. Certainly you can always try to do

22 that.

23 Q. In pooling data like that or in trying

24 to construct through a pooling mechanism, would a

25 bayesian analysis be a way to do that?

JACK BESONER & ASSOCIATES, INC.

1 A. Certainly you could do it, use a

2 bayesian analysis, but I have to tell you that my

3 expertise on bayseian approach is very limited, if

4 nonexistent.

5 Q. Okay. All right. If you were going to

6 -- if you were tasked with that process, how would

7 you do it?

8 MR. BLANK: I'm going to enter a

9 standard objection as to the form and the line of

10 questioning. It seems we're getting into

11 hypothetical questions here, and I'm not certain

12 the witness is really in a position to really

13 discern what you're trying to accomplish here, but

14 go ahead, as best you can.

15 MR. FITZGERALD: Well, I'm not too

16 concerned with him trying to discern what I'm

17 trying to say, but he is designated as a

18 statistician, and I'm questioning him on the

19 statistical methods that he would employ or that

20 are available to do this sort of thing.

21 MR. BLANK: My only concern is I'm not

22 sure that he understands this sort of thing.

23 THE WITNESS: Yeah, I mean, I wasn't --

24 I'm sort of getting lost now in the general line

25 of questions. Can we step back for a minute, and

JACK BESONER & ASSOCIATES, INC.

1 you describe to me exactly what it is that you

2 want?

3 BY MR. FITZGERALD:

4 Q. If I want to take the ESP data and

5 compare it to a national data basis on wetlands

6 that has phosphorus data and pool that data, after

7 appropriate screening of the national data base,

8 how would you, as an expert statistician, have

9 experience in environmental modelling to pool that

10 type of data?

11 MR. BLANK: When you refer to ESP data

12 --

13 MR. FITZGERALD: On Loxahatchee.

14 MR. BLANK: You're talking about the

15 entry and access data?

16 MR. FITZGERALD: Dr. Millard has been

17 using that term.

18 THE WITNESS: It would involve several

19 steps. As far as what I would do, first of all, I

20 will have to familiarize myself with the national

21 data base, understand where that data came from,

22 how it was collected, what laboratory procedures

23 were used to analyze it, etc., etc., and then I

24 would have to integrate that knowledge with the

25 knowledge I have about how the ES&P data were

JACK BESONER & ASSOCIATES, INC.

1 collected and the laboratory protocol that was

2 used to analyze that data.

3 I would have to talk with people

4 knowledgeable in the field of wetlands ecology and

5 ask them about what, if any, particular

6 differences they might expect to see since you

7 have, in the national data base -- I assume that's

8 composed of data from several different areas,

9 several different wetlands.

10 MR. FITZGERALD: Let's just, for the

11 sake of argument, assume hundreds.

12 THE WITNESS: Okay. Hundreds of

13 wetlands. And then based on those conversations

14 with experts and integrating them, all that

15 knowledge, at that point I may decide that it may

16 or may not be feasible to pool the data for

17 whatever particular purpose it is that I've been

18 called to do.

19 BY MR. FITZGERALD:

20 Q. Have you been asked to conduct that

21 type of process in this case?

22 A. What I've been asked to do is take the

23 ES&P data -- entry and access data and compare

24 that data with the data that are given in Appendix

25 "E" that were used to set the limits for the

JACK BESONER & ASSOCIATES, INC.

1 Loxahatchee National Wildlife Refuge.

2 Q. Back in March, when we were looking at

3 this, you stated that you had not examined the

4 district's phosphorus model for limit purposes in

5 Loxahatchee contained in Appendix "E", but rather

6 you had conducted your analysis tests based on Dr.

7 Walker's model which was different. Have you

8 since conducted an analysis of the Appendix "E"

9 district model?

10 A. Yes, I've done some analyses based on

11 that data.

12 Q. Okay. Now, using the ESP data you've

13 just described, have you conducted a comparison of

14 that against Dr. Walker's model contained in his

15 original model that you looked at in his '91

16 paper?

17 A. No, I've only been looking at comparing

18 the ES&P entry and access data with the modelling

19 data that's contained in Appendix "E".

20 Q. Okay. In analyzing Dr. Walker's '91

21 model, you produced a draft report that reflected

22 a bootstrapping analysis and some other

23 consideration and listed a number of areas of

24 possible further work and inquiry. Did you ever

25 finalize that report?

JACK BESONER & ASSOCIATES, INC.

1 A. No.

2 Q. Why not?

3 A. I wasn't asked to.

4 Q. Did you ever conduct a bootstrap

5 analysis with Appendix "E" and phosphorus trend

6 for Loxahatchee?

7 A. No.

8 Q. Why not?

9 A. I wasn't asked to.

10 Q. You testified in March the

11 bootstrapping method was a valid statistical tool

12 for getting a feel for the marsh and variance of

13 -- and just sort of a subject of the validity of

14 the modelling effort. Why would that not be

15 equally valid with respect to the district

16 Appendix "E" model?

17 A. Well, the main point of the draft or

18 report that I produced that commented on Walker's

19 report was to get a feel for the amount of wiggle,

20 the amount of variability in the limits, because

21 the limits are -- the way the limits were set,

22 they were really a realization of a random

23 variable. And the point is that had I gone back

24 and done the same, gathered the same sort of data

25 that I gathered in the base line period '78 to

JACK BESONER & ASSOCIATES, INC.

1 '83, had I gone back and possibly gathered on

2 different days, in slightly different locations,

3 those are two possible sources of variability.

4 Had I done that, what kind of limits would I come

5 up with then? Would they be comparable? How much

6 different would they be?

7 And, so, the point of that, of my

8 report on the bootstrap analysis -- the main point

9 was to try and get a feel for the record for that.

10 Q. Okay. In doing that, a bootstrap as I

11 understood, it, with the substation program, the

12 way you wrote it, it was limited observation

13 obviously to the existing data set. You were not

14 able to substitute data from other locations, on

15 other dates, per se, but were merely able to

16 provide an analysis on any given day by substation

17 of the fourteen by fourteen data you had.

18 A. That's correct.

19 Q. Okay. So when you say it gives you a

20 feel for what might have happened, if you sampled

21 at different places, different days, then those

22 apply to develop the model, and it's that's sort

23 of best professional judgment kind of wiggle

24 analysis, isn't it? I mean, you don't have the

25 data --

JACK BESONER & ASSOCIATES, INC.

1 A. You can't construct the data, right,

2 but that's the point of the bootstrap.

3 Q. Okay. But inherent in that, isn't the

4 weakness -- isn't the weakness of bootstrap why it

5 only gives you a feel or an approximation?

6 A. It's a limits mechanism. It kind of

7 puts a bound around the data set you do have and

8 gives you a feel for what the variance might be.

9 Yes, it probably gives you a low data on the

10 variance.

11 Q. Why does it give you a lower bound?

12 A. Because of what you just said.

13 Q. So why didn't you do that for the

14 district's model? It sounds like a pretty good

15 idea.

16 A. Well, since the ES&P data were

17 gathered, we've been using that to compare to the

18 district's model. Rather than doing bootstrap

19 analysis, we've been using the ES&P data to

20 compare the data with the district's data --

21 compare as far as whether the relationship between

22 the fourteen station average of the log TP, the

23 relationship of that variable with the average

24 station still loads, whether the slope and the

25 intercept of the line are about the same, and

JACK BESONER & ASSOCIATES, INC.

1 also, whether there's real any difference -- any

2 difference in the intercept of that line, in other

3 words, in layperson's term, whether the phosphorus

4 concentration, the entry and access period is

5 really any different than the phosphorus

6 concentration in the base line period.

7 Q. You indicated that in conducting your

8 analysis you log transformed the data first. Why

9 did you do that?

10 A. Well, the district model is based on

11 doing that.

12 Q. Okay. But you just told me you weren't

13 testing the district's model in the way you did

14 Dr. Walker's model, and you essentially had a

15 clean slate. You've got distribution period of

16 record data base. You've got the ESP data from

17 the entry and access Loxahatchee.

18 So it seems to me you probably have the

19 freedom to handle the data however you choose to

20 determine if, in fact, there is a quantifiable

21 correlation between the phosphorus concentration

22 and the station in Loxahatchee, yet you choose to

23 do a log transformation. Was there something that

24 had led you to do that independent --

25 A. No, I think you're misrepresenting what

JACK BESONER & ASSOCIATES, INC.

1 my directed task was. My directed task was to

2 take the district's model, and assume that that's

3 the model that you have to use, and then look at

4 the entry and access period data to see how that

5 compares with the district's model.

6 Q. So you had a preconceived goal when you

7 went into the analysis?

8 MR. BLANK: Object to the form of the

9 question.

10 BY MR. FITZGERALD:

11 Q. Did you have a preconceived goal when

12 you went into the analysis?

13 A. My task, as I understood it, was to

14 look at the ES&P data and entry and access data

15 and compare it with the district data in Appendix

16 "E". If you're talking about a goal, my goal was

17 to compare -- look at the entry and access data

18 and see how closely it matched the district's

19 model.

20 MR. FITZGERALD: I'd like to mark this

21 document as deposition exhibit 13. It is a four

22 page document, bate numbered 1210001 through

23 1210004.

24 (The document referred to

25 was thereupon marked as

JACK BESONER & ASSOCIATES, INC.

1 Respondent's Exhibit Number

2 13 for Identification,

3 a copy of which is attached

4 hereto.)

5 BY MR. FITZGERALD:

6 Q. Is that your handwriting?

7 A. Yes, it is.

8 Q. Okay. So these are your notes from

9 some point in time. In fact, the numbering may

10 not match, and that's one thing I wanted to ask

11 you, whether the four pages are from the same

12 Orlando meeting on Thursday, 12/9/93, which is

13 what's at the top of the second page, and down at

14 the bottom it says 1210002.

15 A. Uh-huh.

16 Q. And then there's a page number 2

17 handwritten behind it, which is 0003, page number

18 3, which is 0004 at the top page, which is 0001.

19 Is that part of the same notes taken or is that --

20 they came stapled together.

21 MR. BLANK: What's the question?

22 MR. FITZGERALD: I'm trying to find out

23 if the top page of notes belong as part of the set

24 of three pages thereafter, which are sequentially

25 numbered in the doctor's handwriting.

JACK BESONER & ASSOCIATES, INC.

1 THE WITNESS: They were probably written

2 on the same pad of paper. Whether they were

3 separated by separate pages or not, I can't tell

4 you.

5 BY MR. FITZGERALD:

6 Q. All right. So you can't tell me if it

7 was part of the same meeting, the first page

8 0001.

9 A. These notes were written during this

10 meeting in Orlando.

11 Q. Okay. So you were sort of going side

12 by side with a list of things that occurred to you

13 to do as a result the discussion in Orlando. For

14 example, number 5 on that first page says: "Redo

15 the percent cattail analysis use DOJ data."

16 A. Uh-huh.

17 Q. Had you done a cattail analysis before

18 that?

19 A. Yes. Yes, I was given data from

20 Breedlove, Dennis and Associates on percent

21 cattail that was taken along approximately --

22 let's see, A, B, C, D, E, F -- about six transects

23 in the Loxahatchee, and along the transect they

24 took measurements of percent cattail at a

25 particular place, what the ambient water total

JACK BESONER & ASSOCIATES, INC.

1 phosphorus concentration was, also what the soil

2 phosphorus concentration was and what the water

3 depth was.

4 Q. Did you attempt to correlate the data

5 to determine if there was any predictor variable?

6 A. I looked at several plots of percent

7 cattail or some transformation of percent cattail

8 versus the other variables, water TP, soil TP,

9 water depth.

10 Q. And what opinion, if any, did you form

11 based on your analysis of those plots and

12 comparison of the cattail percentage against other

13 variables?

14 A. It turned out that the percent cattail

15 appeared to be correlated with water depth in --

16 for water depth less than four. It was a positive

17 correlation. There were about four observations

18 where water depth was six feet or higher, and in

19 those instances there was no cattail.

20 So it could very well have been some

21 sort of parabolic relationship, but the data only

22 went up to a certain water depth. And then there

23 weren't -- I don't think -- I don't recall there

24 were any observations, say, between four feet and

25 six feet.

JACK BESONER & ASSOCIATES, INC.

1 So, except for the observations that

2 were six feet or over, there were about four

3 observations where the water depth was six feet or

4 over. So there was correlation between percent

5 cattail and water depth. There was no correlation

6 between percent cattail and ambient water TP or

7 soil TP.

8 Q. You say that the correlation is

9 positive. How positive? By what statistical

10 measure was it positive?

11 A. If you're asking me to recall, say,

12 something like R-squared value, I can't -- I don't

13 recall that, but the picture clearly showed a

14 positive correlation.

15 MR. BLANK: I think you may be

16 misconstruing the way he's using the term

17 positive.

18 MR. FITZGERALD: Well, that's what I'm

19 trying to explore. If he can give us an R-2 value

20 or if he could give the "P" value, I would know

21 exactly what he meant, and that's how we went with

22 the -- because the analysis of Dr. Walker's work

23 was discussed at great length back in March.

24 BY MR. FITZGERALD:

25 Q. So when you say positive, what do you

JACK BESONER & ASSOCIATES, INC.

1 mean?

2 A. I mean, if you look at the scatter

3 plot, the plot shows that as water depth

4 increased, percent cattail cover increases.

5 There's a general relationship there.

6 Q. Did you seek to model that

7 relationship?

8 A. Yes, I did.

9 Q. And what --

10 A. It was statistically significant. I

11 don't remember the "P" value. I think, in fact, I

12 probably modelled it initially as a quadratic

13 effect so that in increase with water depth and

14 then at a certain point it started to decrease

15 with water depth.

16 Q. So you modelled it with a parabolic

17 function then.

18 A. That's correct.

19 Q. Did you make any effort to determine,

20 by discussions with field specialists, whether, in

21 fact, it was in parabolic but acidotic at the end?

22 A. I talked with some people at Breedlove,

23 Dennis and Associates, George Carlson, for one,

24 about the behavior of the cattail species and

25 whether it made sense that the cattail would

JACK BESONER & ASSOCIATES, INC.

1 behave that way according to water depth.

2 Q. And what did he tell you?

3 A. He told me that, based on his

4 knowledge, for the species of cattail, that was

5 the predominant cattail in that area. If I recall

6 correctly, it was his understanding that its

7 favored water depth was somewhere between two and

8 four feet, something like that.

9 Q. Is the plot you're referring to

10 included in the materials that you provided to us?

11 A. I don't believe so.

12 Q. Why not?

13 A. It's part of a draft report that's

14 preliminary.

15 Q. All right. In March you provided us

16 your draft report on the bootstrapping effort; did

17 you not?

18 A. Yes.

19 MR. FITZGERALD: Counsel, I have a

20 little problem with that not being provided. The

21 witness is supposed to have his final opinions now

22 anyway. The hearing officer made that very

23 clear. And he's telling me that his final opinion

24 as to one of the most significant areas he's going

25 to testify in won't be provided for six weeks at

JACK BESONER & ASSOCIATES, INC.

1 the soonest which, of course, is beyond the end of

2 discovery in this case. He's not providing a

3 draft report on which he has stated here he is

4 going to be relying.

5 BY MR. FITZGERALD:

6 Q. What remains to be done on the draft

7 report?

8 A. As far as what remains on the report,

9 one thing is this item number 5 because, as I

10 said, in the report I initially did the analysis

11 with a parabolic fit. Another way to do the

12 analysis would be to just use only water depths

13 that were four feet or less and then model that.

14 Q. Assuming you did that, it seems

15 intuitive to me that it gives you a higher

16 positive correlation; is that correct?

17 A. No, not necessarily. In fact, it might

18 make it a little bit lower because of the way the

19 data is spread out. As I said, we had water

20 depths between about zero and four feet, and then

21 it might have been a little higher, and then not

22 really anything between four and six, and then at

23 six feet or more there were about four

24 observations.

25 The problem with fitting a parabolic

JACK BESONER & ASSOCIATES, INC.

1 curve like that is that these observations over --

2 at six feet are more -- are basically the ones are

3 determining a lot about the parabolic curve. So

4 if you cut out those observations and just use the

5 depths at four feet or less and just stay with a

6 straight line, then you may not necessarily get as

7 high an R-squared. It's not necessarily going to

8 be thorough. You'll get a higher R-squared if

9 that's what you're saying.

10 Q. This effort or possible effort to

11 finalize your draft report sounds like, to me --

12 and correct me if I'm wrong -- a matter of a

13 couple of minutes work with your software

14 package. It does not seem to be a particularly

15 involved operation; is that correct?

16 A. Well, it would take not too much time

17 probably to refit the model, but it would take a

18 while to then compare that model to the original

19 fit and then draw conclusions from that.

20 Q. Did you calculate R-square for the

21 original fit?

22 A. For the quadratic fit?

23 Q. Yes.

24 A. Yes. I don't recall what it is.

25 Q. And you don't have that report with

JACK BESONER & ASSOCIATES, INC.

1 you?

2 A. No.

3 Q. And it's not in any of the materials

4 you provided to us?

5 A. No.

6 Q. Okay. Now, in addition --

7 MR. BLANK: I'm not all together certain

8 of that. You might want to have him look at the

9 materials you have.

10 MR. FITZGERALD: We can certainly do

11 that, but I have gone through it a number of

12 times, and I see no draft or final reports that I

13 know of that discuss that, but I could be

14 mistaken. And it will certainly be gone through

15 in terms of giving Dr. Millard an opportunity to

16 find it if it's there, and we will reserve to move

17 our right to strike the witness if it's not

18 there.

19 BY MR. FITZGERALD:

20 Q. Number 7 on the first page of exhibit

21 13 says: "Redo percent cattail with DOJ data."

22 Did you do that as part of this draft report?

23 A. At this point I'm not sure what I'm

24 referring to in point number 7.

25 Q. How about point number 8? Did you look

JACK BESONER & ASSOCIATES, INC.

1 at soil phosphorus, water depth, percent cattail

2 in WCA-1 and WCA-2 using ready soil phosphorus,

3 water phosphorus interpolated --

4 A. I have started to analyze some data,

5 soil phosphorus, water depth and percent cattail

6 using ready soil for WCA-1 and WCA-2 and also

7 using percent cattail that was derived by taking a

8 map that was produced by Breedlove, Dennis and

9 Associates for WCA-2, a map of percent cattail

10 cover in that area. Someone at Environmental

11 Services got the ready station coordinates from

12 the Breedlove, Dennis and Associates cattail map

13 and attempted to determine what the percent

14 cattail cover was at that station.

15 Now, in some instances the station

16 appeared to be on the border line of two different

17 levels of percent cattail cover. So I've just

18 started to do some analyses looking at percent

19 cattail cover and the soil phosphorus at the ready

20 stations and then some data on a monthly average

21 water depth that I also obtained from

22 Environmental Services & Permitting. I haven't

23 done very much with that right now.

24 Q. When do you anticipate that will be

25 complete?

JACK BESONER & ASSOCIATES, INC.

1 A. Well, actually what we're going to do

2 is, as far as using the data that I just described

3 to you, where I'm only looking at values, either

4 actual values of the station or values

5 interpolated the percent cattail and the water at

6 the station, that analysis should be done within a

7 couple of weeks.

8 We're also going to do an analysis

9 where we take soil phosphorus and WCA-2 based on

10 creeding procedure and percent cattail based on

11 some sort of grid interpolation procedure and

12 water depth also based on a grid interpolation

13 procedure and do an analysis using all that data.

14 Q. Who is doing that?

15 A. Environmental Services & Permitting.

16 Q. Okay. That's a company -- in my

17 experience, companies rarely do anything people

18 tend to do. Who are those people you keep

19 referring to at Breedlove and ESP with whom you're

20 dealing with?

21 A. Well, Breedlove, Dennis and Associates

22 are the people that produced the percent cattail

23 map, but I'm not -- for this analysis in WCA-2,

24 I'm not talking with them directly. I'm getting

25 all my data from Environmental Services.

JACK BESONER & ASSOCIATES, INC.

1 Q. Do you have that percent cattail map?

2 A. No.

3 Q. Then how are you going to employ the

4 data depicted on it or the data used to develop

5 it?

6 A. Environmental Services & Permitting is

7 doing that for me. They will give to me the data

8 in the form of what they have given -- what they

9 have given me so far is they've given me estimated

10 percent cattail as to each of the ready stations

11 and also average monthly water depth for certain a

12 time period -- I don't recall exactly what it is

13 right now -- at each of the ready stations.

14 Q. Okay. And what's the purpose of your

15 analysis utilizing that data?

16 A. The purpose was to try and determine if

17 there's a relationship between the percent cattail

18 cover and the level of the soil phosphorus and/or

19 the average monthly water depth.

20 Q. So you're not examining water

21 phosphorus or relationship between water to the

22 phosphorus as a variable.

23 A. No.

24 Q. You're aware from your review of the

25 SWIM Plan and Appendix "E" that the district

JACK BESONER & ASSOCIATES, INC.

1 believes there's correlation between water

2 phosphorus and cattail; are you not?

3 A. Yes, I was aware of that.

4 Q. Why then are you choosing to not

5 consider water phosphorus or total phosphorus

6 concentration as a potential predictor variable

7 when that is exactly what you did with regard to

8 the percent cattail analysis you did for

9 Loxahatchee?

10 A. Well, the data were available for the

11 Loxahatchee because the measurements were taken on

12 those transects.

13 Q. Dr. Reddy also produced soil or -- I'm

14 sorry -- water phosphorus data, did he not, for

15 his tests results?

16 A. I don't know for his analysis.

17 Q. What did you mean in Exhibit 13 when

18 you said that the goal of the program is to show

19 that historically the center of Loxahatchee is not

20 impacted?

21 MR. BLANK: What page are you looking

22 at?

23 MR. FITZGERALD: The top of the page,

24 number 2, which is 1210003, the second entry

25 down.

JACK BESONER & ASSOCIATES, INC.

1 THE WITNESS: Oh, goal of -- okay.

2 Goal of program show that historically the center

3 of Loxahatchee is not impacted. I would have to

4 speculate what that refers to at this point.

5 BY MR. FITZGERALD:

6 Q. Well, from whom did you derive the

7 notion that that was a preset goal of your

8 analysis program?

9 A. It's not -- I don't believe it's

10 referring to any analysis that I'm doing. I

11 believe that's referring to -- it's my guess that

12 I'm referring to --

13 MR. BLANK: No, Doctor, don't guess.

14 THE WITNESS: Okay. I don't know.

15 BY MR. FITZGERALD:

16 Q. Is it customary to, in conducting an

17 analysis or doing the type of work you have in

18 this matter, to start out with a preconceived goal

19 as opposed to testing a hypothesis?

20 A. In my line of work as a consultant, I'm

21 obviously aware that my clients will have goals,

22 but as a statistician, it's my duty, if you want

23 to call it, or it's my purpose to analyze the data

24 and find what I find.

25 Q. The draft report that you've been

JACK BESONER & ASSOCIATES, INC.

1 referring to on Loxahatchee and the analysis

2 you've conducted, to whom have you shown that?

3 A. Which draft report are you referring

4 to?

5 Q. Do you have more than one that we

6 haven't seen?

7 A. What draft report are you talking

8 about?

9 Q. Your correlation with the positive

10 value regarding the percent cattail. With whom

11 have you shared that, the draft report?

12 A. Let's see, I sent a copy to Mr. Blank.

13 I sent a copy to John Davis. I sent a copy to

14 Mike Dennis at Breedlove, Dennis and Associates.

15 I don't recall whether I gave a copy of that to

16 Dennis Lettenmaier or not.

17 Q. When did you produce that draft?

18 A. Approximately August or September of

19 1993.

20 Q. Have you received comments back from

21 Dr. Davis or Dr. Dennis on your draft report?

22 A. We discussed it a little bit at the

23 meeting in Orlando, but these -- at which time

24 these notes were taken.

25 Q. Did they give you any written comments

JACK BESONER & ASSOCIATES, INC.

1 as to your draft report?

2 A. No.

3 Q. Back in June of '93, you sent a draft

4 report to Dennis Lettenmaier before sending

5 anything to Bob, and in that context you were

6 speaking about Mr. Blank. Was that the draft

7 report that you were referring to? Do you recall

8 that?

9 A. Yes. I sent a draft of that report to

10 Dennis, and then I was waiting for his comments on

11 the report before I sent it to Bob, and then I

12 went ahead and sent it to Bob and John and Mike

13 before I got any comments back from Dr.

14 Lettenmaier. In fact, I never got any comments

15 back.

16 Q. But it's your recollection now that, in

17 fact, you sent the original draft report to Dr.

18 Lettenmaier?

19 A. Yes, I did send a draft to him.

20 Q. Okay. And if your computer mail was

21 dated sometime in June of '93, then the report

22 must have existed at least since then.

23 A. That's correct, some form of it.

24 Q. Well, that -- and you anticipated the

25 next question. Dr. Lettenmaier gave you no

JACK BESONER & ASSOCIATES, INC.

1 comments on the report?

2 A. That's correct.

3 Q. Dr. Davis and Dr. Dennis discussed it

4 with you in Orlando in December of 1993?

5 A. That's correct.

6 Q. As a result of your conversations with

7 them, did you alter the report?

8 A. No.

9 Q. Have you altered the report since 17

10 June, 1993?

11 A. I think I may have made some changes, a

12 few changes.

13 Q. What caused you to make changes in the

14 report?

15 A. I think I just made some minor changes.

16 I think there's probably minor differences between

17 whatever draft I gave Dennis Lettenmaier and the

18 draft that I sent to Mr. Blank, Dr. Davis and Dr.

19 Dennis.

20 Q. Have you sent that out for a peer

21 review of any sort?

22 A. No.

23 Q. Do you know Jim Loftis?

24 A. I know of him.

25 Q. Have you ever had a conversation with

JACK BESONER & ASSOCIATES, INC.

1 him regarding this case?

2 A. No.

3 Q. Have you ever met him?

4 A. No.

5 Q. In the materials you provided there are

6 notes that purport to reflect statements by Mr.

7 Loftis regarding Dr. Walker's work. Do you recall

8 where you came you up with that information?

9 A. It would probably -- it would have to

10 have been either in a conversation with Mr. Blank

11 or Dr. Lettenmaier.

12 Q. Has Dr. Lettenmaier had conversation

13 Mr. Loftis to your knowledge?

14 MR. BLANK: Dr. Loftis you mean.

15 MR. FITZGERALD: Dr. Loftis, correct.

16 THE WITNESS: I think they've talked,

17 yes.

18 BY MR. FITZGERALD:

19 Q. Do you know for certain?

20 A. No, I can't tell you for certain. It's

21 my belief that they have.

22 Q. When do you anticipate receiving from

23 ESP the data derived from the Breedlove, Dennis

24 Associates cattail map 2-A?

25 A. Either this week or next week.

JACK BESONER & ASSOCIATES, INC.

1 Q. How long will that analysis effort

2 take?

3 A. Once I have the data it should probably

4 take a good -- about -- oh, anywhere between three

5 and four days.

6 Q. So that will not affect or be impacted

7 by the work on "P" loads out of EAA and into

8 Everglades National Park that you're going to do

9 during that same period.

10 A. No.

11 Q. Okay. And I guess I'm still a little

12 unclear, and maybe you can put it in a nutshell

13 for me. What remains to be done on your draft

14 report to finalize it?

15 A. Go back and redo the analysis by

16 looking only at water depths between zero and four

17 feet, fit the model and incorporate any comments

18 that I get from Dr. Lettenmaier.

19 Q. Okay. You had said you sent that to

20 him in June, June 17th, and that you've gotten no

21 comments back. Do you anticipate comments now?

22 A. Well, if you recall, in June -- I'm not

23 sure exactly when the mediation plans started --

24 when the mediation process started, but I know it

25 was around that time. And, so, once that started,

JACK BESONER & ASSOCIATES, INC.

1 everything that we had been working on up to that

2 point basically got dropped, and we concentrated

3 on tasks that were asked of us during the

4 mediation at one point.

5 Q. What makes you believe that Dr.

6 Lettenmaier has comments? Has he told you he has

7 comments on the draft report?

8 A. He had requested me to send him a draft

9 of the report so that he could give me his

10 comments.

11 Q. There is no additional data for

12 Loxahatchee that you're awaiting, is there?

13 A. No.

14 Q. So all that data was from the initial

15 transecting work and vegetative work done in

16 Loxahatchee by ESP?

17 MR. BLANK: Counsel, you're referring to

18 the data related to this draft report?

19 MR. FITZGERALD: Yes.

20 THE WITNESS: I think the data --

21 MR. BLANK: Can we go off the record for

22 just a minute?

23 MR. FITZGERALD: Sure.

24 (Discussion off the record.)

25 THE WITNESS: Could you repeat the

JACK BESONER & ASSOCIATES, INC.

1 question?

2 (Thereupon the referred to question was

3 read back by the reporter as above

4 recorded.)

5 THE WITNESS: Okay. And I was about to

6 respond it was Breedlove, Dennis and Associates

7 who took that data.

8 BY MR. FITZGERALD:

9 Q. Okay. The vegetative analysis was done

10 by Breedlove, Dennis and Associates -- the

11 transecting in Loxahatchee? That's BDA, for the

12 record.

13 A. Yes.

14 Q. Yes?

15 A. Yes.

16 Q. And ESP did the analysis and sampling

17 and --

18 A. Actually, I'm not sure whether BDA did

19 that or ES&P did that, the transecting.

20 Q. Didn't you tell me that to do this kind

21 of analysis you needed to talk to the people who

22 know about the sampling mechanisms and lab

23 techniques and everything else to assess the data

24 set?

25 MR. BLANK: I object to the form of the

JACK BESONER & ASSOCIATES, INC.

1 question. I think you were referring to, counsel,

2 at that time with your concept with regard to

3 pooling the data base.

4 BY MR. FITZGERALD:

5 Q. Okay. Well, back in March we talked

6 about what a statistician needs to know about this

7 data set in order to evaluate, assist and draw

8 conclusions. Is that true with respect to the

9 data from WCA-1 that you employed in your draft

10 report?

11 A. The general things as far as talking

12 with people, how they collect it in the field and

13 how they did the analysis in the lab, yes.

14 Q. Okay. Did you do that with respect to

15 the vegetative assessments conducted by Breedlove

16 -- did you say it was BDA?

17 A. Breedlove, Dennis and Associates.

18 Q. Did you talk to the people who did

19 that?

20 A. Yes, I talked with Mike Dennis and

21 George Carlson who was the project manager.

22 Q. Okay. Did he go out and do the work

23 himself or did others do it?

24 A. I believe a number of -- I'm not sure

25 how many people were involved in that.

JACK BESONER & ASSOCIATES, INC.

1 Q. Did you ever speak to Dr. Joe Birch?

2 A. Yes, I've had one telephone

3 conversation with him.

4 Q. Was it regarding the vegetative

5 community data from Loxahatchee?

6 A. No, it was regarding the percent

7 cattail map and WCA-2.

8 Q. Have you ever spoken with Courtney

9 Hackney?

10 A. No.

11 Q. Have you ever spoken with Miles M. Bud

12 Smart?

13 A. No.

14 Q. How about W. Michael Dennis?

15 A. Yes.

16 Q. Have you spoken with anyone regarding

17 the vegetative community data in Loxahatchee

18 collected February 16th of '93?

19 A. Yes.

20 Q. Okay. When did that occur?

21 A. Sometime between when the data were

22 collected and when I sent my draft report to Dr.

23 Lettenmaier in June.

24 Q. Okay. On how many occasions did you

25 speak with Dr. Dennis regarding that subject?

JACK BESONER & ASSOCIATES, INC.

1 A. I probably spoke with Dr. Dennis about

2 twice and with George Carlson another two or three

3 times at least.

4 Q. Did Mr. Carlson ever tell you that he

5 went out and did any of this work himself?

6 A. No, as far as I know he was the project

7 manager.

8 Q. Since the March deposition did you

9 supply your bootstrap data to anyone other than

10 Dr. Lettenmaier, the bootstrap that was done, Dr.

11 Walker's work?

12 A. You mean did I supply it in the form of

13 a report?

14 Q. Uh-huh.

15 A. Besides Mr. Blank and Dr. Lettenmaier,

16 as far as I know, no, I haven't sent it to anybody

17 else as far as I can recall.

18 Q. Have you reviewed any subsequent work

19 by Dr. Walker since the May '91 report that was

20 the subject of your bootstrap analysis?

21 MR. BLANK: With the exclusion of the

22 mediation materials?

23 MR. FITZGERALD: Yes.

24 THE WITNESS: Yes, I've read a couple of

25 different reports by him. I've read the report

JACK BESONER & ASSOCIATES, INC.

1 that had to do with trends in Everglades National

2 Park. I don't remember the exact title. I've

3 also reviewed his report on the model that he used

4 to size the STA's.

5 I have several of his reports in my

6 possession, but I believe those are the only two

7 that I really looked at.

8 BY MR. FITZGERALD:

9 Q. Okay. With regard to the trend in

10 Everglades National Park, the first report you

11 see, have you formed any opinions based on your

12 review of that report?

13 A. I don't recall. I recall having a

14 problem with one of the things he did, as far as

15 the variables that he looked at. As I recall, his

16 response variable and his predictor variable were

17 inherently correlated because the response

18 variable were, in fact, function of the predictor

19 variable, which produces some that you call

20 spurious correlation.

21 That was -- I believe I started to

22 review that report, and then I don't think I

23 actually finished reviewing the whole report, and

24 then the mediation process came up, and so I never

25 went back to that report.

JACK BESONER & ASSOCIATES, INC.

1 Q. Do you intend to go back and reanalyze

2 that report?

3 A. No, I don't believe so.

4 Q. Do you expect to offer any expert

5 testimony or opinions at hearing in this matter

6 regarding Dr. Walker's analysis of trends at

7 Everglades National Park?

8 A. The only way that it might possibly

9 come up is, as I stated earlier, I have been asked

10 to look at loads coming out of the EAA and loads

11 coming into the Everglades National Park and

12 trying to see if there's a relationship between

13 those two sets of data. Whether that would touch

14 on Dr. Walker's analysis or not -- I don't think

15 it would, but that's the only thing.

16 Q. With regard to the model developed by

17 Dr. Walker, besides STA's, what have you done with

18 respect to that report?

19 A. I reviewed it in preparation for a

20 meeting that I had with Dennis Lettenmaier and

21 Carlos Marin.

22 Q. What was the purpose of that meeting?

23 A. The purpose of that meeting was to

24 discuss how to evaluate Walker's model and how to

25 perform sensitivity analyses on the model.

JACK BESONER & ASSOCIATES, INC.

1 Q. Did you conduct a sensitivity analysis?

2 A. No, I believe Dr. Marin is performing

3 those analyses. I was sitting in on the meeting

4 mainly just for the possibility of giving my input

5 to whatever would be done.

6 Q. I had understood from our earlier

7 discussion of your designation of testimony that

8 you don't intent to offer any expert opinions or

9 testimony at the hearing in this matter regarding

10 STA models. Is that still true or did I

11 understand your earlier testimony correctly?

12 A. Yes, I don't anticipate discussing that

13 topic.

14 Q. You haven't been involved in Dr.

15 Marin's sensitivity analysis?

16 A. No.

17 Q. You haven't reviewed it?

18 A. No.

19 Q. You haven't seen it?

20 A. I think I saw some -- I know I saw some

21 preliminary results at the meeting in Orlando.

22 Q. And what were those results?

23 A. I don't recall the details. I believe

24 the general result that I recall was that the

25 STA's would have to be sized considerably larger

JACK BESONER & ASSOCIATES, INC.

1 -- would possibly have to be sized considerably

2 larger if you change certain assumptions about the

3 model.

4 Q. Do you recall that in conducting that

5 analysis what settling rate or settling velocity

6 was used by Dr. Marin in his analysis?

7 A. I believe he used a wide range.

8 Q. So he sort of did a mini/max solution

9 on the sizing?

10 A. Yes, if I recall correctly.

11 Q. Okay. And you indicate you have

12 several other of Walker reports in your

13 possession, and you don't think you've read them.

14 May I deduce from that with keen insight that you

15 do not anticipate offering expert testimony in

16 this matter at hearing regarding those records and

17 the subject included therein?

18 A. To the best of my knowledge, yes.

19 Q. Approaching the subject from the other

20 end, in case I wasn't smart enough to ask the

21 question the right way, what expert opinions, if

22 any, do you intend to offer based on your review

23 of Dr. Walker's work at the hearing in this case?

24 A. Does Dr. Walker's work include Appendix

25 "E" in the SWIM Plan?

JACK BESONER & ASSOCIATES, INC.

1 Q. All right. Now, you understood, back

2 in March, Dr. Walker's '91 model was different

3 from Appendix "E"; did you not?

4 A. That's correct.

5 Q. Okay. And you understood that Appendix

6 "E" is the SWIM Plan.

7 A. That's correct.

8 Q. Okay. And what your clients have

9 challenged is the SWIM Plan.

10 A. Okay. Then I don't believe I'll be

11 testifying on anything that Dr. Walker has done.

12 Q. May I take it from your hesitation and

13 in cross examination of me, that you do plan to

14 offer some expert opinions about the model in

15 Appendix "E"?

16 A. Yes.

17 Q. Okay. What opinions, if any, have you

18 formed regarding Appendix "E", and are these final

19 opinions?

20 A. Preliminary.

21 Q. When do you plan to have a final

22 opinion on that?

23 A. Within the next couple of weeks.

24 Q. What is going to lead you from your

25 current preliminary opinion status to final

JACK BESONER & ASSOCIATES, INC.

1 opinions in a couple of weeks?

2 A. Mostly my analysis of the Department of

3 Justice Data.

4 Q. Which DOJ data?

5 A. The Loxahatchee data during the entry

6 and access period from '92 to '93.

7 Q. What analysis of those data are you

8 conducting that will finalize your opinions with

9 regard to Appendix "E"?

10 A. How they compare with the '78 to '83 in

11 Appendix "E".

12 Q. What statistical mechanism are you

13 using to conduct that comparison or evaluate the

14 manner in that they may or may not compare?

15 A. Well, the model in Appendix "E" is

16 based on a regression model, and assuming that

17 there is a negative correlation between the

18 fourteen station average of log TP and the stage,

19 and the limits are then derived based on this

20 regression model.

21 So one question you can ask is whether

22 the data gathered in the entry and access periods

23 still exhibit that same relationship between the

24 fourteen station average log TP and average

25 station because those limits are based on the

JACK BESONER & ASSOCIATES, INC.

1 assumptions that that relationship holds --

2 Q. No. With the correlation, I

3 understand. Have you analyzed the ESP data to see

4 if there exists the same correlation?

5 A. Yes, I've looked at ESP data, and I've

6 started to look at DOJ data.

7 Q. Okay. Now, extracting from the ESP

8 data, does the correlation appear to exist or hold

9 true with the approach taken in Appendix "E" to

10 that data?

11 A. Yes. That's a good question.

12 Q. Thank you. We sort of understood

13 that. I mean, that's sort of getting to the hub

14 of the thing.

15 A. There are some interesting features

16 about the ESP data, and there are also very

17 interesting features about the DOJ data, features

18 which are different. It's a different issue. In

19 the ES&P data, the measurements for April of '93

20 are extremely low. In fact, I think they are

21 almost all at the detection limit.

22 Q. What was the detection limit for ESP?

23 A. The detection limit was between two and

24 four parts per billion is my understanding.

25 Q. Okay. Did any of the subject matter

JACK BESONER & ASSOCIATES, INC.

1 experts from ESP or BDA or any of the other

2 consultant groups suggest to you an explanation

3 for the extremely low values in April of '93 in

4 their data?

5 A. I asked John Davis about that, and he

6 was not aware of any particular mechanism as far

7 as the sampling procedure or the laboratory

8 procedure that was any different from any of the

9 other dates.