STATE OF FLORIDA
DIVISION OF ADMINISTRATIVE HEARINGS


CASE NOs. 92-3038
92-3039
92-3040

SUGAR CANE GROWERS COOPERATIVE OF
FLORIDA, INC., ROTH FARMS, INC.,
and WEDGWORTH FARMS, INC.,

and

FLORIDA SUGAR CANE LEAGUE, INC.,
UNITED STATES SUGAR CORPORATION,
and NEW HOPE SOUTH, INC.,

and VOLUME I

FLORIDA FRUIT AND VEGETABLE
ASSOCIATION, LEWIS POPE FARMS,
W.E. SCHLECHTER & SONS, INC., and
HUNDLEY FARMS, INC.,

Petitioners,
vs.

SOUTH FLORIDA WATER MANAGEMENT
DISTRICT,

Respondent,
and

MICCOSUKEE TRIBE OF INDIANS OF
FLORIDA, the UNITED STATES OF
AMERICA, FLORIDA DEPARTMENT OF
ENVIRONMENTAL REGULATION, and
FLORIDA WILDLIFE FEDERATION,

Intervenors.


_____________________________________________________________

DEPOSITION OF
DR. CURTIS POLLMAN

_____________________________________________________________




ACCURATE REPORTING
A/K/A
MACDONALD COURT REPORTING SERVICE
204 West University Ave., Suite 7, Gainesville FL 32601
(904) 373-1126 ù (800) 329-1133 ù FAX (904) 375-6249

Pursuant to due notice, the deposition of the above-
named witness was taken by the Petitioners, Sugar Cane
Growers Cooperative of Florida, Roth Farms, Inc., and
Wedgworth Farms, Inc., before Mary Macdonald, Court Reporter
and Notary Public, State at Large, at 204 West University
Avenue, Gainesville, Florida, on Tuesday, February 16, 1993,
commencing at 12:00 noon and Wednesday, February 17, 1993,
commencing at 9:00 a.m.










APPEARANCES:

SUSAN HILL PONZOLI, Esquire, Assistant United States
Attorney, Southern District of Florida, 155 South Miami
Avenue, Miami, Florida 33130;

WILLIAM H. GREEN, Esquire, Post Office Box 6526,
Tallahassee, Florida 32314;

PAUL L. NETTLETON, Esquire, 400 International Place, 100
Southeast Second Street, Miami, Florida 33131, of counsel
for South Florida Water Management.


ALSO PRESENT:

Ronald D. Jones, Ph.D., Florida International University 3a
I N D E X


SWORN TESTIMONY OF: PAGE:


DR. CURTIS POLLMAN

Direct Examination by Ms. Ponzoli 4

Cross Examination by Mr. Nettleton 287


EXHIBITS:

Pollman Exhibit No. 3
First two pages of report entitled:
"Mercury Emissions to the Atmosphere
in Florida, Final Report" 211

Pollman Exhibit No. 4
Document with cover page entitled "Mercury
Contamination in the Everglades Ecosystem,
a Plan of Study for U.S. EPA, Region IV, Greer
Tidwell, Administrator, Atlanta, Georgia, by
Jerry Stover and Delbert Hicks, U.S. EPA,
Region IV, Environmental Services Division
Ecological Support Branch, Athens, Georgia" 214

Pollman Exhibit No. 5
Document with title page "Chesapeake Bay
Agreement 1992 Amendments" 216

Pollman Exhibit No. 6
Document with first page entitled "Proposed
Research on Water Quality Model for Water
Conservation Area 2A, Duke Wetland Center,
April 1992" 237

Pollman Exhibit No. 7
Document with cover letter on Duke University
letterhead, addressed to Dr. Pollman, from
Dr. Richardson, dated September 2, 1992 237

Pollman Exhibit No. 8
Document entitled "Privileged and Confidential ÄÄ
Notes on INTECOL's IV International Wetland
Conference Special Session ÄÄ An Ecological
Analysis of the Everglades", by C.D. Pollman 255

Pollman Exhibit No. 9
Document entitled "SFWMD Everglades SWIM
Plan, C. Pollman ÄÄ report, reviews,
weekly updates" 258

Pollman Exhibit No. 10
Document entitled "Privileged and Confidential
Draft Everglades SWIM Review 2/6/91" 264
3b
Pollman Exhibit No. 11
Document entitled "Everglades SWIM Plan
Challenge Review, 3 December 1992" 268

Pollman Exhibit No. 12
Document entitled "Water Quality and
the Everglades: Options for the Future" 275

Pollman Exhibit No. 13
Document entitled "South Florida Water
Management District Everglades SWIM Plan ÄÄ
Curt Pollman correspondence", with cover
letter dated May 31, 1990, addressed to
Mr. Green from Dr. Pollman 276

Pollman Exhibit No. 14
Document entitled "McClave", with fax
transmittal, dated 7/11/91 on Info Tech
letterhead, from Dr. McClave to Dr. Pollman 283



Thereupon,
CURTIS POLLMAN,
being first duly sworn, testified as follows:

DIRECT EXAMINATION
BY MS. PONZOLI:
Q Would you state your name for the record, please?
A My name is Curtis D. Pollman, P-o-l-l-m-a-n.
Q And your address, Dr. Pollman?
A It's 353 Northeast Boulevard, in Gainesville,
Florida.
Q And your phone number?
A (904) 378-2762.
Q Okay. For the record, Dr. Pollman, I'm Susan Hill
Ponzoli, I'm an Assistant United States Attorney and I
represent the Respondent/Intervenor The United States in the
SWIM Challenge proceedings.
I will be asking you a number of questions and if
you don't understand my question, please indicate that that
was so and I'll try to frame a better question for you.
A Okay.
Q Have you had your deposition taken before, Dr.
Pollman?
A Yes, indeed I have.
Q How many times?
A I don't recollect exactly, perhaps three to four
times.

MR. GREEN: Excuse me.
Ms. Ponzoli, should we stipulate that
objections, other than to the form of the question,
are reserved for the hearing?
MS. PONZOLI: Yes.
MR. NETTLETON: I thought that was automatic.
MR. GREEN: It is, but we're using different
court reporting services and I just wanted to
clarify it.
MR. NETTLETON: Fine.

BY MS. PONZOLI:
Q Have you been an expert witness in other cases, Dr.
Pollman?
A Yes, I have.
Q How many other cases?
A At least two other cases, possibly more.
Q You don't recall?
A I don't recall exactly.
Q Okay. And what type of an expert witness were you
in each of the two that you do recall?
A I testified principally to issues relating to water
chemistry and water quality and also sediment chemistry.
Q Do you recall the names of those cases?
A One case was the Homer Smith Seafood case and it
was a case that involved Homer Smith Seafood and the State
Department of Environmental Regulation ÄÄ the State of
Florida Department of Environmental Regulations.
Q Okay.
A The second case involved Jacksonville Shipyards and
Jacksonville Shipyards was being sued by another entity, I
don't recall who the other entity was, I believe it was a
homeowner.
Q Were you the expert witness for the Homer Smith
Seafood?
A Yes.
Q Okay. And also for the Jacksonville Shipyards?
A That is correct.
Q And were you affiliated with Mr. Green in those
cases?
A In the first case, in the Homer Smith Seafood case,
I was asked by Mr. Green to provide expert services.
In the second case I worked for a different law
firm.
Q Okay. How many times have you been an expert
witness, Dr. Pollman, on behalf of Mr. Green or his law firm?
A Well, just the Homer Smith Seafood case that I
recall.
Q How old is that case?
A I believe that case was heard in 1985.
Q Okay. I'm going to hand you a copy, Dr. Pollman,
of your Notice Duces Tecum for this deposition, and just ask
you to look through it and make sure that you believe that
you provided all the documents that are responsive to this
Notice Duces Tecum.

MS. PONZOLI: Off the record.
(Thereupon, a discussion was had off record;
whereupon, the following proceedings were held:)
MS. PONZOLI: Let's mark this Pollman No. 1
and Pollman No. 2.

(Document entitled Re-notice of Taking Deposition -
Duces Tecum, consisting of seven pages, so marked
as Pollman Exhibit No. 1.)
(Document with letterhead K.B.N., entitled Curtis
D. Pollman, Ph.D., consisting of eleven pages
[curriculum vitae], so marked as Pollman Exhibit
No. 2.)

BY MS. PONZOLI:
Q Dr. Pollman, Documents To Be Produced.
A Yes.
Q Is the CV, which sits in front of you, as Pollman
No. 2, is that your most recent CV or similar document?
A It looks like the most recent CV that I produced so
far.
Q Okay.
A It may be somewhat dated though.
Q You haven't put together a more recent one.
A No, I haven't.
Q Okay.
A I do notice though that there are certain aspects
of projects or certain projects that I have been working on
that aren't listed on the CV.
Q Okay. Then the second category, all technical,
professional or scientific publications, et cetera, that you
authored or co-authored, related to Everglades research, did
you provide those under No. 2?
A That's on page?
Q Beginning on five and going to page six.
A Yes, I did provide all documents.
Q Okay.

MR. GREEN: For the record, we hope to have
your privileged list tomorrow, it's coming over by
Fed Ex.
MS. PONZOLI: All right.
MR. GREEN: I just wanted to mention that.
MS. PONZOLI: All right.
MR. GREEN: There are some documents that were
withheld as privileged and you'll have a list of
those.
MS. PONZOLI: Do you believe that some of
those fit within the second category?
MR. GREEN: I don't know. I just wanted to
say that before you got into the category.

BY MS. PONZOLI:
Q All right. Dr. Pollman, I don't recall seeing many
actual publications written by you, provided to the United
States.
Do you believe that you provided those copies?
No. 2 actually says a list and then No. 3 is a copy
of them.
A Right, and these are documents relating to
Everglades research.
Q Yes, sir.
A I did provide all such documents.
Q Are they listed on your CV, the ones that you've
written or did you not author. . .
A My publications have dealt with systems principally
other than the Everglades.
Q Okay. So that's why there really were no
Everglades publications?
A That's correct.
Q And you have provided the reports, other than the
ones that Mr. Green may have listed on his privileged list?
A That's right.
Q All right. Then under No. 4 you provided all
documents relating to research done in the Everglades from
'75 to date?
A That's right.
Q Okay. Then I think that No. 5 is somewhat
repetitive of No. 4, all documents relating to research done
in the EPA or the EAA, as defined in the '92 SWIM Plan, those
also were provided?
A That's correct.
Q And No. 6, again, is really Everglades research and
that would have been provided?
A Yes, indeed.
Q Okay. Why don't you look through the rest of the
list, Dr. Pollman, and ensure me that you have indeed
provided all these categories, other than what Mr. Green had
put on his privileged list?
A Unless it was protected by Mr. Green, I have
provided you with everything in accordance with your request.
Q Okay.
Then Pollman No. 2, Dr. Pollman, I would like to
return to that later to discuss it, but I think you can
identify that as your most recent CV?
A It appears to be. There may be other copies that
my firm may have produced that might be more recent that I'm
not aware of, but it appears to be my most recent CV.
Q I'll return to discuss that later.
Dr. Pollman, what I'd like to do first this
afternoon is to understand what exactly your role, on behalf
of the Cooperative, has been over the last several years and
what your role will be at the time of trial.
So that's really the gist of where I'm trying to
understand what you have done.
When did you first become involved with this review
or this whole SWIM Challenge issue?
A If memory serves me correctly, and I don't remember
the actual precise date, I believe I got involved in either
February or March of 1990.
Q Okay. And you were retained at that time by Mr.
Green; is that right?
A That is correct.
Q On behalf of the Cooperative?
A Yes.
Q Okay. And what were you retained to do at that
time?
A I was retained at that point really to provide some
technical expertise in reviewing the SWIM Plan as it was
being formulated at that point in time, also to provide some
technical guidance or critiquing, if you will, on other
documents that were being produced. For example, early on,
one of the things that I performed was a review of Dr.
Richardson's ÄÄ one of his reports.
Q Okay.
A So that was one of the first things I initially
did.
I guess to summarize it in a nutshell, it was
really to provide some technical guidance and expertise.
Q Okay. Have you, over the course of time, Dr.
Pollman, become, would it be fair to say, a scientific
coordinator for the Cooperative?
A In part I think that would be a fair assessment.
It certainly doesn't characterize the entire role that I have
played, but I think that would in part describe what I have
done.
Q How would you characterize the rest of your role?
A Well, I also provided some direct technical input
and some technical analysis, if you will, on certain issues,
and I see that as beyond the simple role of a coordinator.
Q Any other role that you believe is a generic role
that you have provided in this matter?
A I really can't think of any other role at this
particular point in time.
Q Okay. By a scientific coordinator, did you assume
that role early on in 1990 or was there a time that that sort
of evolved?
A Well, I guess it should be understood that there
was no specific point in time that a decision was made that I
would serve as a scientific coordinator. It was a role that
evolved, I guess, based upon my recognition of the technical
issues involved and my feeling that we needed some better,
more high-powered expertise, if you will, in different areas
that I didn't feel comfortable, that I was fully qualified to
provide good guidance on.
Q Okay. And your direct technical input and analysis
that has gone specifically to which issues?
A Principally water chemistry, water quality related
issues.
I have also provided some input on the modeling
work that has been done.
Q At the time of trial do you anticipate testifying?
A Yes, I do.
Q Okay. And you will testify at the time of trial on
which issues?
A Well, I would imagine that I would be testifying on
issues that relate, again, primarily to water chemistry. I
may be testifying on related issues that touch upon water
chemistry, such as, for example, sediment/water interactions,
that sort of stuff; perhaps some aspects of the phosphorus
cycle.
Water chemistry is a fairly broad area, so there
are certainly a number of areas that fall within water
chemistry.
It may be that I also provide some expert testimony
on the modeling work that the Growers Co-op is performing,
which I haven't done at this point, but that remains to be
seen.
Q You don't know whether you would actually sit at
trial and use the models or a model as a piece of evidence
that you would be testifying on water chemistry or water
quality directly from that model?
A Very likely would.
Q Okay. Since you did not provide the model among
your documents ÄÄ is that accurate?
A That's correct.
Q Then I assume that your final opinions have not
been formed; is that accurate?
A That is correct.
Q Okay. What do you anticipate ÄÄ what new
additional information do you anticipate will influence your
final opinions, other than what you presently have?
A That's a difficult question to answer.
I guess there's information that practically we can
obtain and then there's information that I would like to
obtain.
As far as information that is practical, in terms
of getting that information in hand at this point in time,
certainly one set of data I'd like to see is a verified data
set for the Loxahatchee National Wildlife Refuge.
I'd like to see more data on the STA Design. I'd
like to see more information on settling velocity.
I'd like to see more information on research that
addresses the issue of nutrient thresholds.
I'd like to see more information on issues that
touch upon dose response relationships, which I guess is also
related to nutrient thresholds.
Q How would you distinguish the two?
A Well, I would distinguish the two insofar as I
believe that there needs to be framework, a conceptual
framework at the minimum, but ultimately a quantitative
framework for establishing a dose response relationship in
the Everglades area. More specifically, I think that would
fall under the framework of a mathematical model that really
quantifies relationship between say nutrient loads leaving
the EAA and ultimately some biological response in the
Everglades.
I don't know if it's practical, that sort of goal,
but that's certainly the type of information I would like to
see before I would arrive at any sort of final conclusion.
Q Has such a model been successfully built and
applied elsewhere?
A There are certainly ecosystem type models around,
nothing I believe that is available for the Everglades at
this point in time, but certainly other systems have been
modeled from the cradle to the grave, if you will, looking at
basic hydrology, looking at nutrients and water chemistry and
then taking the system on up into trophic level responses.
It's a difficult process, I will admit, but it has
been done.
Q Where has this been done, Dr. Pollman?

A I couldn't give you a specific example off the top
of my head.
I know, for example, that Dr. Odum's group, the
University of Florida, specializes in the so-called systems
type model; in fact, I know that Tom Fontaine, at the Water
Management District, has been involved in that type of
modeling at one point or another, and I believe Bob Constanza
has been engaged in such modeling efforts as well.
There's an entire journal devoted to this type of
thing.
Q Excuse me?
A Journal, scientific publication.
Q Okay. Do you have any personal knowledge of the,
for lack of a better word, the success of these models to
actually real-world responses?
Do you understand my question?
A Let me rephrase your question.
Q Okay.
A The question, as I understand it, is do I have any
understanding or feeling for the efficacy of these models, in
terms of whether or not they accurately predict system
response.
Is that your question?
Q I think that's close enough.
A I'll state at this point that I believe that
modeling is a bit of an art, some might even call it an
arcane art. And particularly when you get into ecosystem
type modeling, and I think that the success of ecosystem
models has probably been limited at this point, that's really
just a conjecture on my part.
Q Why is that?
A Because I'm not up on the full breadth of the
literature of ecological modeling but I feel like I should be
in terms of answering your question definitively.
There may be some models that could do a wonderful
job of simulating ecosystem response.
Q But you're not aware of them.
A I'm not aware of them, because I'm not that up on
the literature at this point.
Q Okay. Do you know anybody who you would say has
more expertise in assessing the success, as I have put it, or
you have put it, of some of the ecosystem models?
A Well, yeah, there are a number of people that would
be perhaps without peer in that area.
Q Okay.
A H.T. Odum at the University of Florida.
Q Okay.
A Bernard Patten, at the University of Georgia.
Q Okay.
A Those two names come to mind.
Q Okay. You said you would like to have a verified
data set for Loxahatchee Wildlife Refuge.
I have two questions on that.
A Okay.
Q One, are you looking to see a data set from the
entry that is presently going on; is that what you are
awaiting?
A That's another issue.
Yes, I believe that that data set that relates to
the entry by the League into the Loxahatchee will provide
some very useful information, but that was not what I was
referring to when I mentioned the verified data set.
Q Okay. What does verify mean in that question?
A What I'm referring to, when I mention a verified
data set for the Loxahatchee, is the following: We have
received the data set ÄÄ actually we received several
different data sets, either from the District or from Mark
Maffei, from the Fish and Wildlife Service, and there has
been a number of problems with these data sets, and if I take
these data sets at face value I'm concerned about the
integrity of the data sets, I'm not absolutely convinced that
they're as bad as they appear to be, but I would like to see
somebody who is in a position of familiarity with the systems
and is a bona fide custodian of these data sets, and I would
like to see somebody stand behind the data and say: Yes,
this is indeed the accurate data set for the Loxahatchee and
we stand behind it.
Q Okay.
A At which point I can evaluate the integrity of the
data sets.
Q Okay. You mention that they appear to be bad.
What are the apparent badness of these data sets?
A Well, there are a number of things. I think the
data are fundamentally internally inconsistent, their basic
relationships in water chemistry, which appear to be violated
and I'd like to be satisfied that those relationships are
indeed ultimately satisfied, to be redundant with the use of
that word, satisfaction.
Q Do you think that actually can be done? I mean,
you're indicating that ÄÄ I guess implicit in your answer is
that you probably will never be really satisfied with the
data.
A No, that's not an accurate statement.
I very likely could be satisfied.
Q Okay.
A But the data set we have in our hands right now
clearly appears to have some problems and maybe those
problems relate to how the data are entered into the data
base and it could be something as simple as a column shift in
the data, so that values for a particular parameter and for a
particular station are inaccurately represented in that data
set and if by shifting those columns into the proper sequence
all the data are properly aligned, we go back and recalculate
these relationships, everything may fall back into place, at
which point I would then have a higher level of comfort with
the data set.
Q Okay. Who collected the data sets?
A If memory serves me correctly, we first noticed the
problem when we got a total phosphorus data set, I believe it
was from the District via Environmental Services and
Permitting.
I may not be accurately representing the chronology
here, but we did end up with a data set ultimately from the
District that appeared to have the phosphorus values out of
phase, so to speak. There appeared to be a column shift in
the data and we were able to correct those problems, as far
as the phosphorus data was concerned, by looking at data that
was subsequently published by the District and the SWIM Plan.
We subsequently received a data set from Mark
Maffei.
Again, one of the things that we have been very
interested in receiving was a data set that included all
water quality parameters. We knew that such water quality
parameters existed, but had never seen them published before
and we finally received a data set from Maffei and I
immediately was able to see, just by looking at the printout,
that there appeared to be some fundamental problems with the
data, and then at that point we decided to do a more detailed
analysis and looked at interparameter relationships, for
example, to establish whether or not the data were internally
consistent.
Q Is this the data set for the Interior Marsh
Stations?
A That's correct.
Q And, to the best of your knowledge, who collected
this data?
A I believe the District collected the data. I
believe the data or data that was published ÄÄ I can't
recollect the date, but it was in the mid-'80s, by an
individual named Millar.
Q Okay. And are there ways that you could pursue
verifying this data that exists, but has not been pursued?
A Well, we have asked for the data from the District
and. . .
Q The same data set that you received from Dr.
Maffei?
A Yes, that's correct.
Q Okay.
A We asked for all the data that exists for the
Loxahatchee National Wildlife Refuge from the District,
including all parameters, other than phosphorus.
Q Okay.
A And we recently received a data set from George
Shee, I believe, that included nutrient values, stage values
and chloride values, but none of the other water quality
parameters.
Q Okay. Do you know what parameters were measured at
the Refuge?
A I don't remember exactly the full suite of
parameters that were measured, but I can give you an
indication of what was measured.
Q Okay.
A I believe that all the major cations and all the
major anions were measured. I believe laboratory
conductivity was measured. I believe the pH was measured. I
believe that there might even be field conductivity
measurements. There may be some other measurements as well,
and, of course, nutrients were measured as well.
Q And the internal inconsistencies were what?
A Well, as an example, since indeed all the major
cations and anions were measured you should be able to
predict the conductivity of the sample, based upon those
measurements, and compare it with the measure of
conductivity. That's one check.
Q All right.
A I would also expect that there would be a close
relationship between sodium and chloride values.
Q Going back to the conductivity, you were not able
to do that? It was not consistent when you carried it out?
You could not predict the conductivity that you asked for?
A That's right, there were some problems there.
Q What were the problems?
A Well, the relationships are not observed. . .
Q Okay.
A . . .that I expected.
Q Let me just ask you ÄÄ what I guess is puzzling me
is, is there any way to satisfy your concerns with this data
set?
A Not as it currently exists.
Q Do you think that there is any way that it exists
anywhere else that could satisfy your concern?

MR. GREEN: Object to the form.
You may answer.

BY MS. PONZOLI:
Q That's just lawyer talk.
A I can't say that if my problems were ÄÄ say the
major ion chemistry are fully resolved, then I wouldn't still
be convinced of the absolute integrity of the data, I might
still have some problems, but certainly if I could be assured
that the internal consistency of the data were there I would
feel much more comfortable with the data set.
Q I think what I'm driving at, Dr. Pollman, it seems
to me that other than the column shift, to which you
referred, which could be rectified, if there's a problem with
the data I don't see how it's fixed a decade down the road, I
mean, that's the problem as a non-chemist, as I'm sitting
here I'm thinking: I don't think it can be fixed.
Do you think that there's some way that it could be
fixed?
A I don't know. I don't know. It's hard to answer
that question without actually seeing the "corrected data
set" in hand. It may be that the corrected data set would
still have some problems with it.
Q And is it your belief that if a corrected data set,
in all probability, exists at the South Florida Water
Management District, it just simply hasn't been located, or
whatever reason, it has not been placed in your hands; is
that accurate?
A No, that's not accurate.
Q Okay.
A I'm not optimistic.
Q I wasn't either, but I wanted to like get down to
the bottom line with you. I mean. . .
A I would hope against hope that this data set
exists, but I'm not optimistic that such a data set exists.
Q What would lead you to believe that it could exist,
since it hasn't been produced to date?
A Well. . .

MR. GREEN: Object to the form.

BY MS. PONZOLI:
Q Okay.
A I presume I can still answer the question?
Q You may.
A Okay. Could you repeat the question?
Q I guess what I'm sort of asking is the same
question around and around, is that unless there is a data
set that the District just hasn't located to turn over to
you, it would appear that you have what is to be had from the
'80s, you do not find it a satisfactory data set for the
reasons that you have stated; is that accurate?
A That's right.
Q Okay. And I am asking you, doesn't that seem to be
the end of the discussion as to your satisfaction with that
data set?
Is there any possibility that there's a way of
fixing this?
A Well, I might be accused of being an optimist, but
one thing that I have noticed throughout this whole process
of being involved in this SWIM Challenge is that data sets
have come and gone, there have been a series of revisions
that have been made to data sets. As an example, I think
there's been a whole litany of changes to loading
calculations coming out of the EAA and those changes and
calculations, I know at least on one occasion, reflects a
change in the water chemistry data set that was used.
So given the history of the way data have filtered
out of the District, it may be ÄÄ I think it lies within the
bounds of possibility that there may be an accurate set that
is buried somewhere that people haven't ferreted it out,
because they haven't looked hard enough.
Q Let's assume, Dr. Pollman, that no verified data
set is forthcoming for the Refuge, what would be the
implications of that to your way of thinking?
A Well, I have serious reservations then about the
quality of the data set and its applicability for deriving
water quality standards for the Refuge.
Q In other words, if you could not, in your opinion,
obtain a better data set than you presently have been
presented with, it would be your opinion that the limits
established for the Refuge are not supported by substantial
data.
A By good data or by accurate data.
It's hard to know whether or not the data we have
is accurate, given the fact that there seem to be serious
problems with the overall integrity of the data sets.
Q Are your concerns with that Refuge data set
reflected in any single document, Dr. Pollman?
A No, they're not. I have not written anything at
this point to summarize or detail my concerns.
Q Do you have, in your opinion, a recommendation for
how one would go about establishing standards for the Refuge,
if one were to have to put aside the data sets with which you
have been presented?
A That's a ÄÄ that could be a very long answer, I
guess.
My feeling about setting standards relates to
ultimately defining some sort of dose response relationship
or defining very clearly what your goals are by establishing
such criteria and. . .
Q You would have to do either dose response
relationship or set goals, you said?
A Yes.
Q And the goals would be set by what?
A The goals relate to your management objectives. In
other words, what is it fundamentally that one is trying to
protect or achieve through the establishment of say total
phosphorus criteria, and I think that there also needs to be
a relationship between those criteria and the so-called
forcing functions or the outside influences that are being
regulated, so that these internal standards are being met.
In other words, there needs to be, I feel, a dose
response relationship or a cause and effect relationship
between ultimately discharges from the EAA and the internal
marsh concentrations and I think ultimately that relates to a
good model of framework, which will give you those predicted
capabilities.
Then to take it one step further, then I think
there needs to be a relationship between those set phosphorus
concentrations and the ecosystem response. I think we need
that understanding established as well.
Q Okay. Are the present limits set on, to the best
of your knowledge, the best available data for the Refuge?

MR. GREEN: Object to the form.

BY MS. PONZOLI:
Q Okay.
A I have a hard time believing that the data from
1978 through '81 or '83, that's embodied in the current
criteria for the Loxahatchee, are the best data available for
the Refuge, I actually find that incredulous. There must be
better data around. There may not be, but I'm amazed that
there aren't, if that's the case.
Q But, to the best of your knowledge, is that the
best available data?
A Well, I do believe there are data that have been
collected by the University of Florida.
I have never seen the water chemistry data
published, but I have a suspicion that there are water
chemistry data available for that system that are more recent
in nature.
Q Dr. Pollman, returning to my question, however, to
the best of your knowledge, the criteria for the Refuge have
been done on the best available data, despite the fact that
you have your concerns regarding that data.
A Could you repeat that?
Q Okay. To the best of your knowledge, Dr. Pollman,
have the criteria for the Refuge been made on the best
available data, despite the fact that you have explained to
me your concerns regarding that data?
A Well, I guess ÄÄ no, I don't think they have been
made on the best ÄÄ that the criteria has been based on the
best available data, because I do believe there are better
data available, I just haven't seen those data.
Q Okay. You believe that the best available data is
some other data, but you have not seen it.
A I have a suspicion, yes.
Q Okay. The best available data is something about
which you have a suspicion.
A For example, there's a Master's thesis by an
individual named Pope, who collected sediment chemistry data.
I believe in that thesis there's a mention that water quality
data has been collected as well. I have not seen that.
Q Why would you think these are the best available
data, without having seen them?
A Well, if the data set that is the currently extant
data set, that was used that involved the criteria, that
holds as the one and only data set and it is verified in its
current incarnation as the data to use, then it's one of the
worst data sets that I have ever looked at in terms of
internal consistency.
Q Okay.
A In fact, I'll go one step further. It is the worst
data set that I have seen in terms of internal consistency.
Q Okay. I think your unhappiness is clear, but you
have seen no other data sets to date that are better.
A That is correct.
Q Have you looked at the Kitchens and Richardson data
regarding the Refuge?
A I have not seen their water chemistry data.
Q Okay.
A Let me amend that statement.
Q Go ahead.
A I have seen isopleths of some of their data, but I
have not seen the raw data and I've not been in any position
to evaluate the integrity of their data set as well.
Q Is there a reason for that?
A Yes. The reason is we don't have their raw data
set.
Q Have you asked for it?
A I can't recall whether we asked for it or not and
my recollection is that Wiley Kitchens is not going to be
responsive to a direct request from us for his data.
Q Are you affiliated with the University of Florida,
Dr. Pollman?
A Yes, I have an adjunct appointment.
Q Is it the custom that an adjunct professor would
have access to another professor's raw data?
That's been the problem, from my understanding, the
University of Florida would not turn that raw data over.
A Being an adjunct professor does not confer any
special privilege in terms of getting access to another
professor's data.
Q Might even impeded it?
A Yeah, I think that's a fair statement to make.
Q Okay. So returning to the type of information that
you were interested in having: dose response, what is the
cause and effect relationship between water from the EAA and
the interior marsh or a model that is set up for ecosystem
responses; weren't those the three types of research that you
felt might be information that you would have confidence in?
A Yes, that's fundamentally correct, but I should
clarify one point.
Q Okay.
A When I refer to a model, there are a couple of
different types of models that one could look at. At the
minimum I would like to see a geochemical model that's
coupled to a hydrology model, that relates loads from the EAA
to the Loxahatchee and the potential impact those loads might
have on various points within the Loxahatchee and that,
again, would be a comminuted, sort of a hydrologic/
geochemical model.
To take it one step further, to look at ecosystem
response, then you would have to add a module to that, that
coupled model, that looks at ecosystem response.
Q Do you understand Dr. Richardson's dosing
experiment to be a dose response, in the manner in which
you're referring to the type of research you would like to
have done?
A I'm not intimately familiar with his experimental
design. I reviewed it some time ago and if memory serves me
correctly, I was not entirely convinced that his experiment
would lead to the dose response type relationship that I
would like to see resolved, but my recollection might be
incorrect; moreover, it may be that he subsequently modified
his design to establish more of a dose response relationship.
I really don't know exactly what he's doing.
Q Okay. To your knowledge, is anyone doing the kind
of research that you're talking about on dose response, cause
and effect relationship between EAA Water Interior Marsh
Stations or hydrologic/geochemical modeling, keyed into an
ecosystem response model; is anyone doing these different
types of researches?
A I'm not sure if anybody is doing the dose response
type relationship research. That, I believe, needs to be
done.
Q Okay.
A It may be that the District has set aside money or
asked for money to do this type of research; but, again, I
can't speak knowledgeably on that topic.
Q Okay.
A I believe that there are some modeling efforts
underway. I believe that the District's model, hydrologic
model for South Florida Bay, accommodates to some extent the
spatial resolution, I think, that one would need to establish
this cause and effect relationship that I would like to see
established between say external loads and internal response.
Q Okay.
A I don't remember the last part of your question.
Q It was your hydrologic/geochemical model into an
ecosystem response model.
A Oh, okay.
I believe that the District has plans underway to
develop an ecosystem response model, they have their own name
for it and I believe Bob Constanza is involved.
I believe, also, again I'm not speaking
definitively here, but I believe that that model will not be
up and running until 1997.
Q Are KBN/Tetra Tech or Duke expanding any of their
models to include these concepts?
A Not an ecosystem response model at this point, no.
Q How about the hydrologic/geochemical model?
A Yes, Tetra Tech is involved in developing a
hydrologic/geochemical model.
Q Okay.
A The geochemical modeling really is limited to the
extent, at this point in time, where we could predict
phosphorus concentrations.
I guess I should qualify this statement, as far as
ecosystem response is concerned.
Duke is interested at this point in trying to
establish a wetland phosphorus model, based upon nutrient
dynamics in Water Conservation Area 2A.
Q Okay.
A It's not what I would consider to be an ecosystem
response model, where you have different trophic levels, but
it would be more of an empirical model that looks at the
vegetation as a big black box or the ecosystem as a big black
box, as a removal mechanism for phosphorus.
Q But they're only in the conceptual stage of this.
A I would say that the Duke folks are in the
conceptual stage. They've done a little bit of empirical
data analysis and it's really hard to say where their
modeling efforts are going to take them.
Tetra Tech is in the process of having ÄÄ they have
conceptualized their hydrologic model at this stage and are
in the process of refining it and calibrating it and
ultimately verifying it, but that's not been done yet.
Q What, in your estimation, is the time frame for the
dose response research that might be necessary.
A Are you asking me how long do I think it will take
to perform that research?
Q Yes, sir.
A Certainly several years.
Q Okay. And the cause and effect relationship, the
external loading to Interior Marsh Stations concentration,
how long do you think that would take?
A That might actually proceed more rapidly, because
that is fundamentally a modeling exercise, hydrologic
modeling exercise, where you set up a system, look at the
flows in the system and the topography in the system and you
develop a model then that accounts for those flows and the
distribution of water across the Everglades. I think that
actually could proceed more rapidly and I believe a lot of
the framework is in place for doing that, coupling that
hydrologic model to the geochemistry might be ÄÄ perhaps
might have to wait to be done in parallel with the ecosystem
response type stuff and dose response type stuff, because
often you need to have algorithms or equations in that model
that account for phosphorus uptake and losses particularly.
Q Then, in other words, returning back to the
Interior Marsh concentrations, you believe that ÄÄ and the
external loads from the EAA, you believe that the data set
exists from which you could build a model to do ÄÄ or could
build a model from the data sets that exist.
A They may exist. I'm not familiar enough with the
hydrologic data to speak to its quality, in terms of do we
have adequate stage measurements, for example, at various
points in the EAA to set up such a model and I really would
have to defer that question to the hydrologic modelers.
Q Okay.
A I can't. . .
Q Assuming the data sets do not exist, how large a
data set would you need to have. . .
A Well, again, the size of the data set and the size
of the model ultimately really relate to the type of
questions that you are trying to ask and answer.
If, for example, we're strictly concerned with,
say, concentrations of phosphorus entering into the
Everglades National Park at the S-12 structures or
concentrations entering into the Loxahatchee S-5 and S-6 or
Water Conservation Area 2A, if we're just concerned about
those control points, then a spatial resolution of the model
doesn't have to be all that dense. You could get away with a
model that's more of a lumped model, spatially aggregated so
that the scale resolution is rather coarse, and you could
verify the performance of the model pretty readily through
the calibration verification exercise, but if you're more
concerned about internal dynamics, then you have to sort of
get into a more complicating model effort, and the more
complicated the model is the more data it will require and it
may be that that data just doesn't exist and that somebody
would have to set up monitoring stations, for example, stage,
and if you're going to predict phosphorus concentrations,
then you set the model for phosphorus at these various
points.
In short, it doesn't make a lot of sense to set up
a very complicated, very dense model in terms of a high
degree of spatial resolution if you don't have the data set
behind it, the hydrology and the chemistry to calibrate and
verify the model. It's really a shot in the dark.
Q I appreciate the spatial concerns, but I was really
thinking of temporal ones.
A Temporal concerns I think are going to be dictated
by the hydrology.
Q My question is very simple.
Do you need a year or two years?
A Of monitoring?
Q Yes.
A Oh, gosh. Certainly more than a year.
Q Okay.
A You would need at least several years worth of
data, I would imagine.
Q Could you get by with a couple of years worth of
data?
A Well, modelers have done that in the past and I'm
not going to, again, speak for the hydrologists here in terms
of what they feel comfortable with doing.
Q But in terms of water chemistry, are you
comfortable with just a few years of data?
A If needs must, yes.
I guess if the calibration exercise ÄÄ if the model
is well-calibrated and if we have adequate data to do some
verification on the model, I would feel fairly comfortable
with it.
Q Okay.
A But there's going to be uncertainty and I think
that uncertainty needs to be embedded in the modeling process
and you need to evaluate that uncertainty in your
predictions.
Q Okay. Are you aware of, Dr. Pollman, anywhere
where actual threshold ÄÄ nutrient thresholds have been
documented as causing particular changes?
A As a nutrification?
Q Yes.
A I would imagine there's been a number of studies
that have looked at biological response as a function of say
phosphorus additions and nitrogen additions. I think that
type of research has been done for decades.
Now, what if somebody has established a threshold
concentration, per se, for the Everglades, that's an entirely
different question.
Q I really was asking a broader question.
Are you aware of where a threshold concentration
has been established in any ecosystem?
A There's been a number that has been used that was
developed in Wisconsin, back in the early '50s, established,
I believe, ten micrograms per liter phosphorus is the
threshold between oligotrophic systems and eutrophic systems.
Q But that's really a different question, isn't it?
A I'm not sure I understand.
Q You were talking about a nutrient threshold.
A Yes.
Q Did you mean a nutrient threshold between an
oligotrophic system and a eutrophic system?
A Well, in reality it probably would be between a
mesotrophic system and a eutrophic system.
The stages in lake trophic status really go from
oligotrophic conditions to mesotrophic conditions to
eutrophic conditions, ultimately to hypereutrophic
conditions.
Q Do you think ultimately that that type of question
or threshold becomes a political question?

MR. GREEN: Object to the form.

BY MS. PONZOLI:
Q Okay.
A I don't ÄÄ I don't know how to answer that
question. I have never considered it in that context. I
think if politics is driving a threshold number, then I think
that is a very na‹ve approach and the system is not best
served by such a number.
Q Then do you think that that type of a question, the
nutrient threshold, is a scientific question?
A It should be a scientific question, yes.
Q And not a management question?
A Ultimately it needs to be a management question, as
based upon good science.
Q But it is ultimately a scientific question, in your
opinion?
A The science has to be there to establish the
question.
Now, how that threshold ÄÄ and I think I should say
for the record also that I don't believe that there is a
single number, that there's a magic number that when you
cross that threshold, so to speak, that you now go from a
mesotrophic system or an oligotrophic system to a mesotrophic
system.
Q You may have a moving number.
A It's in all likelihood a continuum and through that
continuum you're likely to have a range of different effects.
Q Okay.
A And also I think it's a naive approach to focus in
on nutrients when other factors, geochemical factors and
hydrologic factors, also have an effect on the ecosystem
response.
So I think that in part that helps translate to a
continuum ÄÄ response continuum.
As an example, I think ÄÄ this is something not
really related to the Everglades issue ÄÄ but there has been
considerable interest in lake acidification to the response
of fish to aluminum, as a toxicant, and aluminum effects on
fish are mitigated by the presence of calcium. So you can't
establish just a single number for aluminum concentrations as
giving a toxic response, you have to consider other factors
as well. For example, what are the calcium concentrations,
and I believe the same type of phenomena might also be
present in a system such as the Everglades, where ecosystem
response or system response is dictated by a number of
different factors, not any one factor in isolation.
Q So I guess implicitly you're uncomfortable with
what? A single number for phosphorus entering the water
concentration; is that accurate?
A Yes, I'm a bit uncomfortable with that.
Q And you would prefer what? In your opinion that
there be a more sophisticated approach?
A Well, I'll go back to the ecosystem modeling
approach.
I think I would like to see what the objectives
are, what the management objectives are for the system and
then let the model dictate then what the effluent
limitations, if you will, should be.
Q So you don't start with science, you start with
management objectives.

MR. GREEN: Object to the form.

BY MS. PONZOLI:
Q You can answer.
A Pragmatically speaking, when you look at a system,
I see a host of competing objectives, a number of different
entities that need and demand water and, of course, then
there's the ecosystem itself and if we were to look at just
the ecosystem and were not to consider other needs for water
in the region, then I think it becomes a fairly
straightforward exercise for establishing what the criteria
are.
I have a concern though, just a gut-level concern
at this point, that these competing interests for water have
to be factored into whatever the quote "solution for the
Everglades" is.
I'm afraid that the whole system down there,
meaning the urban system, the agricultural system and the
natural system can't have its cake and eat it too, that if
you rob Peter to pay Paul, Peter will be robbed, and I'm not
sure how much insult Peter is willing to put up with before
Peter starts crying uncle.

(Thereupon, a short recess was had; whereupon, the
following proceedings were held:)

BY MS. PONZOLI:
Q I'd like to return, Dr. Pollman, really to one of
the areas where we began and that's in your role as a
technical advisor on the SWIM Plan.
A Okay.
Q And I'd like to ask you your early reviews, initial
reviews, on the SWIM Plan in 1990.
A Okay.
Q What were your most fundamental concerns regarding
the SWIM Plan at that time?
A If memory serves me correctly, I believe that my
biggest concern was that I felt the cart was being placed
before the horse, that there was what I felt to be undue
emphasis on controlling nutrients almost exclusively without
any real emphasis on looking at hydrologic effects to the
system as a whole.
I guess to be more specific, my feeling is that
hydrology is important to this system, hydroperiod and what
not, in that the SWIM Plan needs to be able to account for
hydrologic impacts of whatever is being mandated by the SWIM
Plan, and I felt that although it was stated in the SWIM Plan
that hydrology was of concern, that it was really being paid
lip service and there was not a clear framework for
evaluating what those hydrologic impacts would be.
Q You mean the hydrologic impact of the SWIM Plan or
something else?
A Principally the SWIM Plan.
Q And the SWIM Plan at that time, the basic remedial
measures were basically what?
A I don't recall what the basic remedial measures
were, except to reduce loads out of the EAA by some
percentage and I don't recall what that percentage was.
Q Do you remember what the target concentration was
at that time into the Water Conservation Areas?
A I believe the numbers started at thirty parts per
billion and went up to fifty parts per billion.
Q As it presently stands; is that accurate?
A That's what I recollect. I don't know if that's
accurate or not.
Q Okay. Do you know the sizing of what was once
called the Water Management Areas?
A No, I don't recall the precise sizing.
Q Okay. And it was your belief that in order to
reduce the effluent concentrations to thirty parts per
billion into the Water Conservation Areas that there would be
significant hydrologic impacts?
A That was really one of my concerns.
I guess I had a problem with the SWIM Plan insofar
as I felt that, again, the dose response relationship had not
been formulated and I also felt that the goals of the SWIM
Plan were rather nebulous and it was not clear in my mind how
the goals of the SWIM Plan would be achieved by the SWIM Plan
itself.
Q And you understood those goals to be what? Do you
recall?
A Well, there was certainly, I think, a number of
goals. I think one of the goals was ÄÄ I may be wrong, I
can't remember back that far, it might have been to restore
hydroperiod, for example; reduce eutrophication-related
impacts, that would have been another goal, but I don't
recollect exactly what those goals were.
Q Okay. If a dose response research had begun in
1990, is it your opinion that it could have been completed by
now?
A I think we would be a good bit further along the
way in terms of resolving some of these basic questions.
Those questions may not be fully resolved, but certainly we
would have been a lot further on.
Q Did you suggest that dose response research be done
to anyone?
A I don't recollect, but I may have.
Q You don't recall if you suggested it to your
clients?
A I have mentioned it to my client that this was the
type of work needed to be done. I certainly don't recollect
advising my client to fund that type of research.
Q Did, to your knowledge, your client or you, on
behalf of your client, approach the Water Management District
and suggest that this would be the appropriate way to go
about a SWIM Plan?
A I have very little direct contact with the
District. In fact, I have had very little direct contact
with the District throughout this whole affair. So I don't
imagine that it really would have been in my capacity to
contact the District and say this is what you folks should be
doing. Now, whether the attorneys involved representing the
Growers Co-op talked to the District about this is the type
of stuff that should be done, I can't answer that question.
Q Your role has been primarily in support of
litigation.
A I don't know, actually. My role really has been to
advise the client on technical issues and if it relates to
litigation principally, I guess so, but I try to keep out of
those aspects of the Challenge.
Q So you only answer those questions that are put to
you?
A Principally, yes.
Q Okay.
A I may pursue some ideas that I think are
scientifically important, you know, somewhat on my own, but
I'm generally in contact with Mr. Green as to the types of
things that I'm evaluating.
Q So you respond to the technical issues that Mr.
Green requests you to respond to?
A That's correct.
Q All right. In your review of that early SWIM Plan
in 1990, were you prepared to offer testimony on water
chemistry/water quality?
A I think it would have been very difficult at that
point in time to offer much testimony in terms of water
chemistry and water quality, because I had not seen much at
that point of water chemistry and water quality.
Q You were just evaluating the Plan along the lines
you've indicated?
A That's right.
Q And pretty soon another SWIM Plan came out?
A Yes, a plethora of plans came out.
Q All right. I would like to know when, to the best
of your recollection, you began to interact with the other
various experts who are providing technical input to the
Cooperative.
Do you understand my question?
A Could you be a bit more specific?
Q I believe ÄÄ is it Mr. McClave or Dr. McClave?
A Dr. McClave.
Q Dr. McClave and then Dr. Horvath and there's Dr.
Richardson and Mr. Gherini and so forth.
A Right.
Q Okay. Can you give me some evolution of how this
came to be?
A Yes, I can. I can't give you precise dates.
Q That's all right. Years would be nice, if you can,
but if you can't. . .
A Well, even years might be difficult.
Q All right.
A But I think I can hone it down a little bit.
Q Okay.
A I can't recall whether I got Tetra Tech involved
first or whether I got Info Tech involved first. Info Tech
is the firm that Dr. McClave is the president of. I believe
I probably contacted Jim McClave first to help me review
William Walker's affidavit and that affidavit certainly you
will recollect was prepared in response to the federal
lawsuit.
Q Okay.
A Or in support of the federal lawsuit.
I believe probably at about the same time I asked
Steven Gherini to assist me. I felt at that point in time,
it became very clear to me, that one of the things that was
missing in this whole SWIM process was a conceptual framework
for evaluating management alternatives. I believe that's
very key, in terms of managing a system like this you have to
have a good idea of what the response is going to be and have
a fairly good idea of what the likelihood of success is going
to be before you embark upon such a venture and, as far as I
could tell, based upon my reading of the SWIM Plan and the
limited knowledge I had of the inner-workings of the
District, it seemed apparent that no such model framework
existed and that one would need to be developed so that,
again, management options could be evaluated and assessed.
So I contacted Steve Gherini and I believe I talked to my
client as well and suggested that this is something that we
should contemplate doing or at least evaluate the
possibilities of doing, and then I recommended that Tetra
Tech be the firm to do that work.
Q This would have been in 1990?
A I believe it was probably in December of 1990, the
end of 1990.
Q Okay.
A The only other expert that I believe that the Co-op
has retained, that I have worked with directly, is Lloyd
Horvath and he was brought into the fold principally to look
at seepage-related issues in the Everglades agricultural
area. At that time we were evaluating the BMP Rule, and one
of the issues that we were concerned about was what was the
effect on seepage on hydrology in the EAA and how much
control did farmers have over hydrology ÄÄ controlling the
release of water from their farms if seepage was a major
contributor to the overall water budget for those systems.
So that really summarized Mr. Horvath's
involvement.
Q Okay. And Mr. Horvath did some seepage studies of
farms, cooperative farms, that abut the Loxahatchee Wildlife
Refuge; is that accurate?
A That's correct.
Q And that required constant pumping?
A That is correct.
Q Okay. And did you also participate in a great deal
of the analyses for the BMP Rule?
A Yes, I did assist quite a bit.
Q Okay. Now, my understanding is that the
Cooperative, at least factually, is satisfied with the BMP
Rule; is that accurate?
A I'm not. . .
Q I'm not asking for a legal conclusion, just
factually as you understand it.

MR. GREEN: Object to the form.

BY MS. PONZOLI:
Q Okay.
A What I recollect is that the Co-op and the District
reached some sort of agreement, I guess that both parties
found tractable, that's what I recollect.
Q Okay. Let me ask you, because I think there may be
a whole area of testimony that I may not spend much time on,
but I need to make sure.
Is it your opinion that the Cooperative can achieve
the requirements of the BMP Rule?
A I have no opinion on that matter.
Q You have no opinion on that?
A That's right.
Q Do you know anyone who, on behalf of the Co-op,
would have such an opinion?
A I don't know anybody that is formulating an opinion
as far as that is concerned.
Q Would it be accurate that you have no opinion as to
whether a higher load reduction could be achieved by the
Cooperative?
A I have no ÄÄ that's correct, I have no opinion on
that matter at this point.
Q Do you anticipate formulating one?
A If the Co-op asks me to look into it, then
certainly we'll look into it, but I have no indications that
this is an issue that they're interested in at this point.
Q You have no knowledge that you would be asked to
form such an opinion?
A That is correct, at this point.
Q All right. You brought Mr. Gherini in, you were
looking for a conceptual framework that would bring together
what management options?
A We were looking for a conceptual framework that
would allow us to evaluate what the effects of various
options would be on the hydrology of the Everglades and
phosphorus dynamics.
Q Okay. Before I go like into what I understand to
be the Tetra Tech model and the evolution of that model, let
me get straight in my own mind how many models are really out
there with which you're affiliated on behalf of the
Cooperative.
A Okay.
Q All right. There's the Tetra Tech ÄÄ is it a
hydrology phosphorus dynamic model?
A The title of the model, I believe, is The
Everglades Phosphorus and Hydrology Model.
Q Okay. Now, is there one model or two models at
Duke with which you have participated in some role?
A Well, there's one modeling effort, I should say.
Q Okay.
A The Duke model I think is really, as far as I'm
concerned, doesn't exist at this point in time because
there's nothing, as far as I can tell, nothing concrete that
has been developed that we can actually use and evaluate, but
there is a model that the Duke researchers are working on
developing at this time and it's been an evolutionary
process.
Q Okay.
A The model approach has gone from, as far as the
Duke model is concerned, from more of a conceptual framework
that was based on a conceptual model that Richardson had
proposed in one of his publications, to more of an empirical
type model, I guess it's probably a hybrid conceptual
empirical model, but the details of which you would have to
probably get from Dr. Reckhow.
Q Is Dr. Reckhow the one who is principally
developing this model?
A Actually a graduate student is doing the principal
work, under his direction.
Q Song-Qian?
A That's correct, but Dr. Reckhow is principally
responsible for that model.
Q Okay. I apologize, but I don't understand your
explanation of the difference between conceptual model and
empirical model.
A Okay. A conceptual model is basically what the
name implies. It's not a mathematical formulation, it's a
conceptual formulation of how a system might work. In its
simplest form a conceptual model might take the shape of a
simple diagram that shows flows moving through a system or
interactions in the system between various so-called
compartments, but it's decidedly not quantitative or non-
rigorous in a mathematical sense in its presentation. It
doesn't have mathematical equations, it hasn't gone to that
stage yet where you start taking the conceptual formulation
that you know that there's a relationship between
phytoplankton and zooplankton. We know that such a
relationship exists and that's a conceptual model, so to
speak, but taking that conceptual framework and using it to
predict something is another step and that involves taking
data and developing relationships or mathematical algorithms
to describe the interactions between those compartments.
Now, an empirical model is an entirely different
beast. An empirical model, in its purest form, is simply a
relationship between two variables. As an example, we know
that typically there's a relationship between hydrology and
phosphorus concentrations and that they appear to be
inversely related, that under high flow you tend to have
lower phosphorus concentrations in many parts of the
Everglades. So just a simple empirical model just might
establish quantitatively what that relationship would be, but
it's not really based upon any underlying principles.
I don't know if I'm making this very clear.
Q That's fine. I'm following you just fine.
Please continue, I'm with you so far.
A Okay. There's another type of model, there's a
mechanistic type of model and it's a product of the
conceptual model and that's where you start defining a model
in terms of ÄÄ well, we know that there's a relationship
between say phosphorus concentrations and algal uptake of
phosphorus and that might be described by a first order
differential equation or there might be some other kinetic
process that's used to describe that process. That's when
you start getting into a mechanistic type of model.
Q Now, I think you told me, but I don't. . .
Is the one that Dr. Reckhow and the graduate
student, Song-Qian, is that a conceptual model?
A Well, it's a product of the conceptual model. It
was born out of a conceptual model that Curtis Richardson
developed, but it's taken on a life of its own, so to speak.
It's now moved in somewhat of a different direction,
principally because of the type of data that was available.
Conceptual models are, in my line of thinking, well
and good provided you have the type of data that you need to
take that conceptual model and develop the algorithms to use
that model for any capability, but if the data doesn't exist
then you have to start becoming creative, as it were, and
start lumping different compartments together, because you
don't have a complete understanding of the processes, but you
might be able to establish a good empirical relationship.
So this is what I think the Duke researchers have
done and I think they're working with the constraints of
their data set and decided that a statistical approach is
probably going to yield the most fruit.
Q Is that because of the limited spatial and
temporal. . .
A Well, I think it's probably a reflection of the
types of data that has been collected, that you really don't
have the process-oriented type data that you need to take
Curtis Richardson's conceptual model and bring it to
fruition.
Q So they're going to have to use what they have to
statistically predict what they think will happen; is that
accurate?
A I believe that's accurate.
Q Okay.
A I think you're going to have to talk to those folks
to get the nitty-gritty of what they're doing.
Q All right. Dr. Richardson has described his as a
landscape approach.
Do you know what that means?
A Could you repeat that, please?
Q Dr. Richardson has described his model as using a
landscape approach.
Do you know what that means? Does that have any
particular meaning to you?
A Well, I find that a curious use of words, because I
know that the District is developing their own landscape
model, so I believe that's the name that they're applying to
their ecosystem type model.
What Dr. Richardson means by a landscape model, in
this particular case, I really couldn't answer.
Q I may have misunderstood him. I think that's what
he said, but I haven't read his deposition yet.
So what data do you understand Dr. Reckhow and the
graduate student to be using to build their model?
A Well, I would say principally they're using flow
data for the entry points into the Water Conservation Area
2A. They're probably ÄÄ again, I can't speak definitively,
but they're probably using core data and they're certainly
using water quality data from various sampling points within
Water Conservation Area 2A to develop their model.
Q Are the core data and the water quality data their
own data?
A No, I believe a good bit of it that they're using
is the District's.
Q Dr. Reddy and Water Management District?
A Yes.
Q Okay.
A I should qualify the statement, as far as core data
is concerned, because I really don't know if they're using
core data or not.
Q Okay. I understand the limits of the answers that
you're giving me.
So if they're using flow data that would be Water
Management District data also.
A That's right.
Q All right. And if they're using core data, which
you're not completely sure if it's the District's or their own.
A Or they may not even be using it.
Q Or they may not even be using it.
A Right.
Q And then if they're using water quality, which
again may be their own and/or the District's.
A My suspicion is, again this is a hunch, is that
they would be using both types, both their own data and what
the District has developed.
Q Okay.
A But it may be that they are restricting their
efforts strictly to District data, because the District has
also the ancillary flow data that goes with it.
Q Anything else?
A Not that I can speak of knowledgeably.
Q Do they have any vegetation data that they would be
feeding into their ecosystem response?
A Well, this is not ÄÄ I don't see this as an
ecosystem response model, this is more of a phosphorus uptake
model.
Q Okay. I'm sorry. I think you did say that.
So that would be the limit of what you believe they
have included as their data.
A Yes.
Q And is this just essentially, quite simply, a model
showing the phosphorus uptake within Water Conservation Area
2A?
A I believe that's a fair way to describe it.
Q All right.
A There may be one other source of data, I would
imagine, that they would need.
Q Okay.
A And that would be deposition data, atmospheric
deposition data, but they're probably just making the same
assumption for deposition that is published in the SWIM Plan.
Q Okay.
A There's certainly no empirical data for the Water
Conservation Area 2A.
Q Are they using this in a predictive capacity for
other Water Conservation Areas or for STAs?
A Well, I can't answer that question, because, first
of all, the model isn't developed yet.
Q Okay.
A I guess it remains to be seen, once the model is
developed, whether it's going to be applicable to any other
areas in the Everglades.
Q Do you understand what the goals are of the model?
A Well, we had certainly hoped that we would be able
to use this model to predict, help predict, STA performance,
but it may prove that that is a na‹ve expectation or hope.
Q Why na‹ve?
A Well, because the systems are likely going to be
fundamentally different. You're talking about a managed
versus an unmanaged system. Vegetation, I believe, are going
to be different.
I guess it really remains to be seen what the STA
design is ultimately going to encompass. I think
fundamentally we're talking about two different systems.
So, if the model proves to be a statistical model,
I guess I'm going to have to have some reservations about
taking the statistical model out of the bounds of Water
Conservation Area 2A and applying it to the STAs.
Q It sounds as though this model is dying as we sit
here.

Is that your understanding?

MR. GREEN: Object to the form.
BY MS. PONZOLI:
Q This model is losing vitality as we sit here asking
questions and I'm wondering why.

MR. GREEN: Object to the form again.

BY MS. PONZOLI:
Q Okay.
A I guess we're not putting all our eggs in one
basket. In other words, we're not looking for the Duke model
to be the end-all of all models to predict phosphorus
retention in any portion of the Everglades. We have limited
expectations.
Q Okay. Let me ask you this.
Did your expectations become limited at some point
in time? Had they been greater at a prior point in time?
A They were greater at the very outset, at the
conceptual model stage.
Q What caused them to diminish?
A I think cognition, the fact that we really don't
have the data to take a conceptual model like what Curtis
Richardson has proposed and flesh it out to become a good
predictive tool.
Q Okay.
A And that what we're going to end up with instead is
really an empirical model.
Q Has someone explained this to Dr. Richardson?
A I have not.
Q Do you think Dr. Richardson is of the same opinion
that you are, regarding this model?
A I have no idea.
Q Do you interact with Dr. Reckhow on this model, on
any sporadic basis?
A On a sporadic basis, yes.
Q Does Dr. Reckhow share your concept of the
limitations of this model?
A We have never sat down and discussed the
limitations of this model.
My own belief is that he would be rather
conservative in its application.
Q And if he were conservative, assuming that you
are ÄÄ are you also conservative? Would that be a fair term?
A I'm a social/liberal and a fiscal conservative.
Yes, I consider myself to be a conservative.
Q Okay. Then what would be a conservative use of
this model?

MR. GREEN: Object to the form.
DR. POLLMAN: A strict application of the
model, if it's purely an empirical model, would
be to restrict its use to Water Conservation Area
2A.

BY MS. PONZOLI:
Q For phosphorus control in 2A?
A For phosphorus uptake.
Q I'm sorry, phosphorus uptake.
A Yes.
Q Okay. When is this model, in its present form,
anticipated to be completed?
A Well, I can't answer that question exactly.
Duke has really unfettered, as far as any sort of
time frame in concerned ÄÄ well, the money that was given to
Duke was with no strings attached and it was given for them
to do research with and they are doing this research on this
model development and we certainly would like it sooner than
later, but I'm not sure when they're going to be able to
finish their efforts.
Q All right.
Is that the twenty thousand that was to fund the
graduate student?
A That's correct.
Q Okay. That was funded directly by the Cooperative,
was it not?
A I believe so, yes.
Q Did you have some hand in requesting that funding
in ÄÄ I don't recall. . .
A Yes, I did. I talked to, I believe, Mr. Green and
Mr. Ward, Jeff Ward with the Sugar Growers Cooperative, about
that I thought it would be a good idea for Duke to perform
this exercise.
Q Okay. Had you thought it would have more
application than you presently believe it has, at the time
that you proposed that funding?

MR. GREEN: Object to the form. I don't think
he said he proposed funding.

BY MS. PONZOLI:
Q Well, at the time that you had, you know, suggested
the exercise ÄÄ is that a fair. . .
A Yes, I guess so.
Q Had you anticipated it would have greater
application?
A I had anticipated that it would or it might have
applicability beyond Water Conservation Area 2A, that is
correct.
Q Okay.
A I do not want to give a premature death and burial
at this point. It may be that the model does prove to be
applicable beyond Water Conservation Area 2A. I haven't seen
the model. The model hasn't been fully developed yet. We'll
have to wait and see what happens with the model.
Q Is it proceeding as rapidly as you thought it
would?
A No, it has not.
Q Do you know why?
A I can hazard a guess and that is that. . .

MR. GREEN: Well, you're not required to
speculate.
DR. POLLMAN: No, I don't know.

BY MS. PONZOLI:
Q Would you like to hazard a guess? I'd be
interested in your guess.
It was your lawyer who interposed. . .

MR. GREEN: That's all right. I don't mind if
you answer.

BY MS. PONZOLI:
Q I understand the constraints of the answer.
A My suspicion is that the model progresses somewhat
slower than anticipated, because the work is being down by a
graduate student in concert with professors who have other
sets of priorities and this is not their sole responsibility
of work.
Q Dr. Reckhow, other than supervising this graduate
student, does he have other work that he does on behalf of
the Cooperative?
A There's only one other thing that I'm aware of that
Dr. Reckhow is involved in and I'm not sure if he's going to
follow through with this or not, but I know that the
Cooperative is interested in having him evaluate Dr. Walker's
reanalysis of settling rates in Water Conservation Area 2A.
Q So rather than Dr. Richardson, it will be Dr.
Reckhow who does the analysis of the settling rates?
A Well, it may be both of them, actually. I don't
know.
Q And that's the only other task that you're aware of
that Dr. Reckhow is performing on behalf of the. . .
A As far as I know.
Q Do you know any other tasks that he's performing on
behalf of the League?
A Well, maybe the League has interest in the analysis
also.
Q So you have no idea whether it would be a split
funding or anything of that nature?
A I have no idea.
Q And those monies are not run through KBN?
A That is correct.
Q Okay. Has anyone asked you or, to your knowledge,
Dr. Reckhow to look at Dr. Walker's chloride data when that
becomes available?
A Has anybody asked Dr. Reckhow?
Q Has anybody asked you or, to your knowledge, asked
Dr. Reckhow to do that task?
A I don't know if anybody asked Dr. Reckhow to look
at it.
Q All right.
A We're interested in looking at it.
Q So you anticipate that you will be looking at it
when it becomes available?
A Either I or Steven Gherini, or both of us together.
Q All right.
A And possibly Dr. McClave as well.
Q Are you aware of a meeting where Dr. McClave and
Dr. Walker met regarding the BMP Rule and did some work
together reconciling, I guess, some differences between their
analyses?
A I recollect when that occurred, yes.
Q Okay. Do you have any memory of what the
reconciliation was between the two of them?
A No, I don't. I was not privy to that meeting.
Q Okay.
All right. I'd like to talk then about the
Everglades phosphorus and hydrology model.
Dr. Pollman, what I would like to do is have you
describe for me the Everglades phosphorus and hydrology model
as it presently exists and then I'll do the history of its
evolution afterwards. Rather than having you build it for me
as it came through time, I would like to know right now what
that model consists of.
Why don't you give me a general overview and then
I'll have some specific questions.
A Well, the model domain is the Everglades
Agricultural Area, the various Water Conservation Areas and
the Everglades National Park.
The model is designed really to predict flows at
major control structures and phosphorus concentrations at
major control structures.
Q Bottom line, that's pretty much it, right?
A Yeah.
Q It does not include the lake.
A The lake may be added to the model. My belief is,
and I don't know this for a fact, the lake may be included as
part of the model, but there will not be a module in there,
per se, that would predict stage dynamics in the lake. The
lake may be added as a boundary condition of the model.

Q But you don't know for sure.
A That's right.
Q When will that decision be made?
A Well, actually that is ÄÄ I don't know if a
decision has to be made or not. It's something that Steve
Gherini and his staff would be deciding on, as far as their
modeling efforts are concerned.
Q And Florida Bay is not included?
A That's correct.
Q Okay. What data have been included, if you could be
more specific, if you know? If you know dates, time frame?
A The data that have been used are the data that have
been basically summarized in the SWIM Plan. So I believe the
time frame would be 1978 through ÄÄ 1979 through 1988, and I
believe that we subsequently obtained ÄÄ that we have
acquired three more years worth of data from the District in
the past few months. So that would be 1989, 1990 and 1991,
if memory serves me correctly.
Q And these have been fed into the model?
A They will be used in the model: verification
exercises.
Q And that exercise is not going on yet?
A It may be ongoing.
Q But you don't know.
A Certainly it's not completed at this point.
Q Do you think it's ongoing?
A It may be ongoing at this point.
Q Okay.
A It's certainly on tap to be done.
Q Will you participate in the model verification
exercise?
A Not directly.
Q What does that mean?
A It means I will certainly review it.
Q What does the model verification exercise consist
of?
A Well, it probably would be a dynamic verification,
where we look at a time series of flows and various control
points or various major structures in the Everglades versus
predicted flows.
Q Just so I'm clear.
You would go back in time and pick a certain time
and place and make sure that it did, in fact, happen the way
your model said it happened?
That sounds wrong to me.
A When you verify a model, what you're trying to do
is take a data set that was not used in the actual model
development and see how the model performs. In other words,
look at the predicted response versus the observed response.
Q Where would you obtain a separate data set for flow
and water quality?
A Well, as an example, what one could do is take a
data set and divide it in half or divide it three-quarter the
data set to be used to calibrate the model, so that you try
and optimize the match between simulated values and observed
values and then take a look at another data set, the
remaining portion of the original data set that was not used
in the calibration process, and then you look at the
predicted values versus the observed values and see if they
match up very well.
Q Okay. Just so I'm crystal clear.
The hydrology data, specifically from the SWIM Plan
that you have included, includes what?
A Well, it would include all the flow data we can get
from the District at the major control structure and it would
also include stage values of various points within the
Everglades as well.
What we would like to do is try and match both
predicted stage values and predicted flows, to the extent
that we can.
Q Do you have the major canals included in your
model?
A Yes.
Q You have the marsh flow-throughs included.
A What do you mean marsh flow-throughs?
Q I mean, is all the water that is moving in a system
included in your model?
A The model is intended to be a complete, as
possible, representation of the hydrologic balance of the
whole Everglades.
Q So that includes surface water and groundwater.
A Groundwater is not a specific component of the
model, per se. It may be that that proves to be an estimated
parameter of the model, because there's very little
information of groundwater flow in the system down there.
Q Are there other models that include groundwater or
not?
A Models for the Everglades?
Q Yes.
A I'm not aware of any. There may be, but I'm not
aware of any.
Q Have you compared the hydrology of your model to
other models?
A Yes, we have.
Q Okay. Which other models have you compared it to?
A There is a model that the University of Florida has
developed that I believe it's really a subset of South
Florida Water Management District's hydrologic model and we
have compared. We've done some preliminary exercises where
we have compared their output versus our output to see if we
could get a match.
Q Did you get a match with the University of Florida
AEA model?
A Well, before I answer that question I should caveat
my answer by saying that everything that's been done at this
point has really been a preliminary exercise.
Our preliminary indication is that we do not get a
very good match.
Q Okay. Have you matched it against the South
Florida Water Management District's model?
A We don't have the South Florida Water Management
District model up and running.
Q Are you attempting to get it?
A I believe we have made requests to get it, but I
can't answer that question definitively.
Q You're not sure if you requested it or not?
A I believe we have requested it in the past, I don't
believe we have it in an executable form.
Q Are you continuing to request it?
A I believe we requested it recently.
Q Do you believe that that's something that your
model needs to be tested against?
A We certainly would like to evaluate our model
against their model and understand any differences if there
are any.
Q Do you have a theory as to why you did not have a
good match against the University of Florida AEA model?
A Well. . .

MR. GREEN: Excuse me. Just for the record, I
don't know what AEA means. I don't think it's been
defined and I want to make sure we're talking about
the same thing.
MS. PONZOLI: Is it the Adapted Environmental
Assessment?
DR. POLLMAN: That's correct.
MS. PONZOLI: I'm sorry, I was just. . .
MR. GREEN: That's all right.

BY MS. PONZOLI:
Q Okay.
A Yes, it is the AEA model.
I really hesitate to speculate as to why our model
might differ from the AEA model.
We have really only done a preliminary analysis at
this point and I don't want to derive any conclusions that
might result in comparing apples to oranges. Suffice it to
say that we are concerned about the hydrologic balance in the
AEA model.
Q How did it differ from the AEA model?
A I don't recall exactly.
Q Do you recall the margin of difference?
A I can't express it in quantitative terms. I can
state it was substantial.
Q Okay. Do you have an opinion of the AEA model or
the accuracy of the AEA model? If that's a fair word for it.
Is accuracy a fair word to use in regard to a
model ÄÄ is efficacy a better word?
A Yeah, but probably both are appropriate words to
use.
I can't comment on that model without spending more
time looking at it.
Q So you don't care to offer an opinion? You have no
opinion?
A I have no opinion.
Q You have no opinion on the AEA model.
A That's correct.
Q Okay. Do you have an opinion on the South Florida
Water Management District hydrology model?
A No, I do not.
Q Okay. Did you include seepage losses in the
system, in your hydrology model?
A I don't recall.
Q Did you include losses to evapotranspiration?
A Yes, we did.
Q Okay. I asked you, I think it was a crude
question, which is what caused the problem, I'm sure, but as
I understand models you will have grid cells; is that right?
A No, not necessarily.
Q Okay. Yours does not contain grid cells?
A Well, the model is set up a little differently than
the AEA model, which has spacing of maybe one kilometer by
one kilometer grid.
The model is set up in what we call a series of
CSTRs or continuous stirred tank reactors.
Q How does that differ from a grid?
A Well, fundamentally, there's not a whole lot of
difference. It's just really how you spatially set up a
model.
Q Is it just a larger ÄÄ like a great big vat, in a
matter of speaking?
A Exactly.
Q Okay. How big is your vat?
A Well, the vats vary in sizes. One vat would be
Loxahatchee Wildlife Refuge.
Q Okay.
A Again, I don't recall how many CSTRs are in the
model, but there are probably six or so at the minimum.
The EAA would be a CSTR.
Water Conservation Area 2A would be a CSTR.
And so on.
Q Okay. So each of the Water Conservation Areas, the
EAA and the Everglades National Park would each be one of
these large vats.
A That's right.
Q Okay.
A And then we're looking at enhancing the spatial
resolution within the Loxahatchee and also Water Conservation
Area 2A.
So actually a Water Conservation Area might be a
CSTR. I couldn't tell you how many it would be, but it could
be three, you know, it could be four.
Q Okay.
A The same for the Loxahatchee. The Loxahatchee
could be two CSTRs, it could be more.
Q How do you decide how to break a CSTR? I mean,
obviously the Water Conservation Areas have some rationale to
them.
A Right.
Q But how would you decide within one how to break it
into ÄÄ how to further divide it?
A Well, one reason why you might subdivide it into
different CSTRs would be because you want more information on
what is happening within a particular area.
Now, if we are only concerned with what comes in
and what comes out of Water Conservation Area 2A, then
arguably a single CSTR would be an adequate approach, but we
are interested in perhaps understanding more of the dynamics
within Water Conservation Area 2A, which leads us to an
enhanced spatial resolution.
Q Okay. Does 2A break hydrologically into two
rationally different units?
A Into two units?
Q Well, two or more. I mean, how would you decide?
A Well, I think I'm going to leave that up to the
hydrologists to decide, because it's fundamentally a
hydrologic question.
Q From Mr. Gherini?
A Yes.
Q Does that make sense to you somehow though?
A What? To break it into different CSTRs?
Q Yes.
A Yes, it makes sense to me.
Q Why?
A Why?
Q Yes.
A Because I'm interested in understanding the
dynamics of phosphorus uptake in Water Conservation
Area 2A and I think in terms of looking at the settling
velocity, I think we need to look at an understanding of what
the settling velocity may be or may not be in Water
Conservation Area 2A. I think we need to break Water
Conservation Area 2A into a different number of CSTRs as
opposed to one CSTR.
Q And you're using only Water Management District
data for analyzing 2A or are you also incorporating Dr.
Richardson's data?
A We likely will be using Dr. Richardson's water
quality transect data.
Q Have you done that yet?
A No.
Q When will that effort begin?
A Well, it may be starting as we speak. If it hasn't
started then we'll start it in the near future.
Q When is it projected to be completed?
A I really can't speak for Tetra Tech. Hopefully
soon.
Q How soon is soon? Are we talking about a month, a
year. . .
A Several months, perhaps.
Q Now, which of Dr. Richardson's water quality
transect data are we referring to?
A Are you asking specifically for a transect number?
Q No, no, I'm not using the transect number. I'm
assuming you're using ÄÄ are you using all the transects?
A I couldn't tell you exactly what the dates are. We
received some data earlier this fall from Dr. Richardson, I
believe it's from his first year of monitoring. I don't
believe we have all his data at this point.
Q Okay.
A If we decide to use his data we would be using at
least the first year's worth of monitoring.
Q And, in your experience, how many years does it
take to get a good data set?
A You can construct a model, a hydrologic/geochemical
model, in one year's worth of data.
I'm doing that on a different project right now.
We're modeling the hydrologic and major ion dynamics in some
seepage lakes in Florida, based upon eighteen months' worth
of data, and a number of modeling studies have been
developed, using that, say, one year's worth of data.
So one year may prove to be useful, but you always
want more data than you have, that almost goes without
saying, scientists and engineers want more than they are ever
given to work with, but we probably could fly with one year's
worth of data, at least get some good calibration results.
Q You say get good calibration results, what do you
mean?
A Well, when you set up your model we're really
trying to do two things. We're trying to predict hydrology
and phosphorus losses from the system, and having transect
data for phosphorus concentrations will help us to pick
appropriate rate loss coefficient to describe phosphorus
uptake in the model.
Q I guess my thought was that one year seemed rather
thin for a data set.
A It's not that unusual to use one year's worth of
data.
Q Okay. And I think I had asked you, but I didn't
nail it down, what you were going to be using of Dr.
Richardson's. Was it his core data or. . .
A No. Our model really doesn't ÄÄ it would be
principally surface water data.
Q And in your model you're doing phosphorus ÄÄ going
back to the name of it ÄÄ are you talking about total
phosphorus or. . .
A Total phosphorus.
Q And how are you going to use Dr. Richardson's? I
don't think his is in total phosphorus, is it?
A I don't believe he has total phosphorus values.
Q All right. So you'll use only his total phosphorus
surface water data; is that accurate?
A I don't want to restrict myself to using total
phosphorus data alone, but. . .
Q Sure.
A . . .the model is a total phosphorus model.
Q Do you see any problems in using Dr. Richardson's
water quality data?

MR. GREEN: Object to the form.

BY MS. PONZOLI:
Q Okay.
A I have not evaluated the quality of his data set
yet.
Q Whose idea was it to use Dr. Richardson's data?
A I can't recall specifically. It might have come
out of a joint conversation I had with Tetra Tech, when I
became aware that there was transect data available.
Q Is there other transect data available for 2A, to
your knowledge?
A The District has some transect quality data as
well.
Q Okay. Are you going to use that or are you not
going to use that?
A I haven't discussed it, that I recollect, with
Steven Gherini. It may be that we will use it.
Q In lieu of the Duke data or in conjunction with the
Duke data?
A Either.
Q So you think that you can use them together or
exclude one and use only the other?
A We could use either/or.
The District transect data covers a different time
period. The District data that we have really covers a
different time period. I don't believe it extends through
1991. It may. It certainly wasn't taken during the same
days and so we have the opportunity to perhaps augment one
data set with another.
Q Do you know the length of time of the District data
set?
A Well, I know that there's District data that
extends back to 1976. I'm not sure what stations were
covered by that time frame.
If that's the case we likely would not go back that
far in time, but we would like to use data that starts around
1979.
Q When I was asking you, in regard to the hydrology
that was included, I believe your answer was it was as
complete as possible; is that accurate?
A Yes.
Q Okay. What, to your knowledge, is really not
included?
A We don't have specific independent data on seepage.
I think that's the biggest unknown in the model.
I believe that we have all the flow data at the
major control points that we need. We have rainfall data and
we have evapotranspiration estimates.
Q Okay. Returning to the AEA model. Which one do
you think is more accurate, the AEA or the Tetra Tech model?
A I would have to reserve judgment until we spend
more time analyzing the differences between the two.
Q Are you looking to modify your model or are you
looking for flaws in the AEA model to determine the
difference?
A We're looking to understand the differences between
the two models at this point and if there are irreconcilable
differences, and we understand what those differences are,
then we will evaluate whether our model needs to be modified
accordingly or whether we think that the AEA model is
seriously flawed.
Q What are you looking for to understand the
differences?
A Well, principally ÄÄ I can only state one thing at
this point and that would be looking at the hydrologic
balance. I think that would be the first step, looking at
how the model simulates the hydrologic balance and whether or
not you get good closure.
Q Help me here. I'm not clear.
You mean what is going in and what's coming out,
either up in the air or out the other end of the system, and
does it match in the way you thought it should match?
A I think that's a fair assessment.
Basically what you do is ÄÄ one of the things that
you would do is to see if you get a balance. In other words,
these models are all mass balance accounting of what comes in
and what goes out, and if what goes out the system is matched
by what comes in, at a very cursory level, that gives you
some confidence that you're accounting for major processes.
Q Now, the next step would be looking at the dynamics
of the response and that's really a calibration verification
exercise.
I think we were going to look at flows, predicted
flows, one model versus the other, and see how they match up,
both the models, vis … vis, with the actual observed output.
Q Okay. I think I interrupted you.
You said you were looking to understand the
differences and you said you were looking at hydrologic
balance.
A That's right.
Q And then you were getting ready to say something,
or were you?
A I don't recall.
Q Okay. That's what you're really looking at, the
hydrologic balance.
A That's right.
Q Okay. When do you think that analysis will be
completed?
A Within the coming weeks.
Q Okay. Is this one of the things that you were ÄÄ
additional things that you were looking at in your model?
A I'm not sure I understand your question.
Additional in what respect?
Q Did you come to some point of closure with your
model and then you went back and looked at additional issues
and this was one of the additional issues that you went back
to look at?
A When we started developing the model this summer I
became aware of the AEA model's existence in late summer,
sometime in August. When I became aware of that model I felt
that it was important to compare our model with the AEA model
and then we sent the model code and a copy of the executable
model to Tetra Tech for their analysis and evaluation.
So it was not something we anticipated up front.
Q Right.
A Because we were not aware of the existence of the
AEA model.
Q Okay. Is someone actively looking at the AEA model
to determine how accurate you believe it is?
A We are ÄÄ Tetra Tech is involved in doing that
evaluation. I guess they're doing some ad hoc type analysis,
as-needed type analyses, but it's not a major focal point of
our analysis, but we are using it as a guide, I guess. We
want to make sure that our model is as robust as possible.
Q That's the word I was looking for before, robust.
What else have you looked at in going back to look
at your model?
A Could you be more specific?
Q Well, there seemed to come a time at which the model
came to some point of closure and then you went back and looked
at a few more things and obviously hydrologic balance was one
of the things, and what are some of the other things?
Q Well, the model hasn't come to closure yet. The
model has been in a state of development for some months and
it was ÄÄ when we decided to look at the AEA model, the model
wasn't ÄÄ the EPH wasn't developed and I would hesitate to
say that the EPH is developed at this point, and there are
things that we feel that need to be done to the model to
achieve our objectives.
But, I guess, trying to answer your question, yes,
we have looked at the AEA model as a point of comparison and
we are concerned about getting the model properly calibrated
and verified at this point.
Q All right. Let me ask you your objectives. Let me
make sure I understand what your objectives are.
Are they to predict flows and phosphorus
concentrations at the major control structures? Are those
the objectives?
A Yes.
Q Are you looking to predict phosphorus
concentrations and in the Marsh Interior Stations?
A Are you talking about the Loxahatchee?
Q Yes. I'm sorry, I'm talking about the Loxahatchee.
A To a limited extent, yes.
Q What does that mean?
A It means I don't believe we'll have spatial
resolution to predict interior marsh concentrations at all
sixteen stations that were identified.
Q How many do you think you can get?
A I think they're going to be aggregated and I
haven't seen what the structures are going to be.
So what we'll probably have is a model that is
capable of predicting phosphorus concentrations at different
zones as opposed to different stations. That's really a
spatial resolution problem.
Q What other things are you doing to achieve the
objectives of predicting flows and phosphorus concentrations
at the major control structures? What other things, in
addition to what you have told me?
A I don't think there's anything else, other than the
fact that we're enhancing the spatial resolution in the
Loxahatchee and Water Conservation Area 2A.
It may be that we are going to enhance the
resolution in the Park itself, but I don't recall.
Q All right. I assume that the model is set up, we
now have an acronym, the EPH Model.
A I'm sure we can come up with something with more
pizzazz.
Q That's your job.

MR. GREEN: How about the "Truth" model?

BY MS. PONZOLI:
Q Okay. I assume that you have set up the "Truth"
model to tell us the truth about various scenarios; is that
accurate?
A I hate to put something in as definitive as the
"Truth" as to what it is we are going to do with the model.
It's going to be our best estimate as to what we
believe is going to happen, under various management
scenarios, and there's always going to be some element of
uncertainty and we're going to try to minimize that
uncertainty as far as those predictions are concerned.
Q Are you able to quantify the amount of uncertainty?
A We haven't done that yet.
Q Do you intend to?
A We would like to.
Q Do you think you would be able to?
A I would say that there's ÄÄ we can certainly come
up with some estimates of uncertainty. I really don't know
what those estimates would be at this point.
Q Let me ask you what would be an acceptable range of
uncertainty, in your opinion, in a model upon which you would
base the kind of decisions that might arguably be made, based
on this model, assuming it were used to make SWIM Plan type
decisions?

MR. GREEN: Object to the form.
ANSWER BY DR. POLLMAN: I don't feel qualified
to answer that question.

BY MS. PONZOLI:
Q Okay.
A And I'll tell you why.
Q Okay.
A An element of certainly a large uncertainty is
going to be STA sizing, and STA sizing evolves on a single
number at this point, the settling velocity, and it's not my
position to say what is a reasonable bound of uncertainty,
because it's ultimately an economic question and if STAs of
thirty thousand acres or thirty-five thousand acres cost,
say, three hundred to five hundred million dollars and we're
looking at a doubling ÄÄ so the settling velocity might be
off by a factor of two, and the price tag goes up to six
hundred or a thousand or a billion dollars, that's not for me
as a scientist to say whether or not that is acceptable.
So I really can't answer that question of
uncertainty. I think that's ultimately a management
question.
Q Okay. So uncertainty is a management or economic
question?
A I think so.
Q It's not a natural resource question.

MR. GREEN: Object to the form.
ANSWER BY DR. POLLMAN: I guess it all depends
on what your objectives are.
I do think that you may be able to set a
priori what the acceptable limits of uncertainty
are and if that is the case then you would have to
design a study set up that meets those limits of
uncertainty.

BY MS. PONZOLI:
Q Okay.
A But, again, I think it all depends on the questions
that you're trying to answer and what your objectives are and
if uncertainty of a factor of two is unacceptable for, say,
concentrations entering the Everglades National Park, then I
think that needs to be identified in that your studies that
are designed to address those type of questions have to
account for that.
Q Scientifically, how is uncertainty used?
A Could you be more specific?
Q Well, if you were building a model and there's
uncertainty in the model, as there will be in any model;
isn't that true?
A That's correct.
Q Okay. Some models are cruder than other models;
isn't that also correct?
A Yes.
Q So in a cruder model I would assume that the
uncertainty increases; is that accurate?
A No, that's not accurate. That's not necessarily
true.
Q Okay.
A Let's say you have the most sophisticated model
there can be.
Q Okay.
A It's very detailed, in terms of its
compartmentalization, conceptualization of the processes that
one believes are involved in describing, say, phosphorus
dynamics, every component that you put into that model has an
element of uncertainty and you may not have the information
to justify that level of description or formulation of that
model and when you account for that error and run that error
through the model you may end up with very inflated values in
terms of the uncertainty. I mean, the model output may be
highly uncertain when you account for all that embedded
uncertainty.
Some cases, in more simplified models, that perhaps
uses more of what we called lumped approach, may give you a
higher degree of confidence, because we may know those
relationships for those lumped processes with a little bit
more certainty and there may be less uncertainty about those
estimates.
This is where perhaps a combined statistical
approach or an empirical approach, a mechanistic approach,
may yield some real fruit.
So, to make a statement that a lumped or an
aggregate or a crude model is more uncertain than a more
sophisticated process-oriented model, carried out to its nth
degree in its formulation, to make that sort of blanket
statement is not accurate or correct.
Q Okay. Let me ask you this.
What is the range of uncertainty then, in your
opinion, that would be acceptable for a more simplified
model, such as the EPH Model?

MR. GREEN: Object to the form.
MS. PONZOLI: What is the source of your
objection, Mr. Green?
MR. GREEN: For what purpose? I mean, you've
talked about three or four kinds of uncertainty and
I don't think there's any context to the question.
MS. PONZOLI: All right.
MR. GREEN: You may answer the question.
MS. PONZOLI: No, I want to nail down the
uncertainties. I think that's an important
concept.

BY MS. PONZOLI:
Q Uncertainty as related to the EPH model, Dr.
Pollman, what would be the various forms of uncertainty that
would exist for the EPH model?
A Oh, there are a number of different sources.
Rainfall estimates, for example, are going to be a source of
uncertainty.
Q Okay. Do you have any range of uncertainty that
would apply to them?
A I can't give you what those numbers are
specifically. You're going to have to talk to, again, the
hydrologists in terms of what they think are reasonable
estimates, but perhaps five to ten percent uncertainty, maybe
upwards to fifteen percent uncertainty of rainfall estimates
are likely to accrue.
Q Okay.
A Evapotranspiration is a bigger area of uncertainty.
I think we're going to be less certain about our estimates of
ET than we are going to be about rainfall.
Estimates of flow. . .
Q How large do you think the uncertainty for ET
should be?
A Well, it depends on the temporal resolution of the
ET numbers. As you start going to finer and finer scale of
temporal resolution, the ET numbers, the estimates of error
can become inflated upwards to say fifty percent.
Q And then if you have less ÄÄ a simpler resolution,
then it deflates?
A Yes. Actually. . .
Q It seems counter-intuitive to me, but. . .
A We have a better idea of what the ET is going to be
on an annual basis than we do have on a monthly basis.
Q Okay. So up to fifty percent if you were to do a
fine temporal resolution, but let's say for an annual, what
would the ÄÄ what would the EPH Model have ET estimated on?
A You're going to have to talk to Tetra Tech for
that.
Q Okay. What range of uncertainty would you assume?
A Well, I'm not an expert as far as uncertainty in
evapotranspiration estimates, but let's say at least fifteen
percent.
Q Okay. So what other forms of uncertainty might be
in the EPH Model: rainfall, ET. . .
A Estimates of flow.
Q What would be the range of uncertainty you would
expect?
A Again, I'm not an expert in this area.
Q Right.
A But certainly at the minimum five percent, but I
would expect more likely fifteen percent errors is likely.
Q And you consider these fifteen percent errors
within an acceptable range; is that accurate?
A Well, I guess ÄÄ I think it's inappropriate to talk
about ÄÄ or perhaps na‹ve to talk about what is acceptable
and what is not acceptable. What we have to do is work with
the data that are available and then account for that
uncertainty in our estimates and make sure that we propagate
that error through our modeling so that our predictions have
the appropriate error bounds associated with them when we
finish our simulations.
Q What are appropriate error bounds? I mean, how
would one judge an appropriate error bound? Are they
numerical, can you quantify it?
A I'm not sure I know how to answer that question.
Q Well, you know where I'm going.
A No, I don't.
Q I'm looking for how much uncertainty in your model
is acceptable to you, as the modeler? As the proponent of
the model, a certain amount of uncertainty, in your own mind,
is certainly attached to that model.
A Right.
Q What do you consider acceptable in your mind,
whether or not some other person agrees that that is
acceptable?
A Okay. I think. . .
Q Now, do you understand?
A Yes, I think so.
Q Okay.
A And, again, I think the question is a bit na‹ve
because the answer is really constrained by the data. . .

MS. PONZOLI: What is it about your witnesses,
Mr. Green, they all say that I'm na‹ve.
MR. GREEN: That's not a derogatory comment.
MS. PONZOLI: I am not a statistician or a
modeler, so I'm trying, Dr. Pollman.
MR. POLLMAN: That's all right.

BY MS. PONZOLI:
Q All right.
A We are constrained by the data that we have to work
with and we certainly want to minimize the error and in an
ideal world we would like to get down to within five percent
error.
Q Okay.
A And that's typically what you can achieve from an
analytical perspective.
Q Okay.
A It's unrealistic that we will ever get to that
point, because not only are we looking at analytical error,
but we're also looking at other sources of error in the
modeling process.
We may end up with a hundred percent error. I
wouldn't be happy with that, but it may be that we can't do
any better than that, because of the data sets that we're
working with.
Q What is the usefulness of a model that has a
hundred percent error?
A I think, first of all, the notion of uncertainty is
an important one to embrace in this whole process, in that,
at least in my mind, if we're talking about a three hundred
to five hundred million dollar solution, get uncertain about
what the results are going to generate, I think that needs to
be considered in the process.
Q But my question is, what is the usefulness of a
model that has a hundred percent error?
A Well, one of the things that you can do is
identify, through error analysis or sensitivity analysis,
what elements of the model seem to be driving that
uncertainty, what is the model most sensitive to, as far as
parameter uncertainty, and that can direct us to areas of
research that should give us more bang for the buck, should
be much more cost effective, in terms of helping us get to
the bottom line and help us identify: this is the area we
need to spend our research dollars on, these are the areas of
great uncertainty and these are the things that are driving
our uncertainty, and then if we answer these questions then
we can come back and give you a more certain predictive
response.
Q Do you think that that will be the outcome of your
model?
A That may be one of the outcomes of our model, yes.
Q I understand what you're telling me about your
model's uncertainty, at least I think I understand.
Your model's uncertainty is bounded by the
uncertainty of what you have to put into it and you had no
control of what you had to put into it; is that right?
A That's correct.
Q That was accumulated by other entities and that's
what you built your model from?
A Right.
Q Do you have a good feel for the level of
uncertainty that this model will have when you're finished?
A No, that's premature to state at this point.
Q Do you think the fifteen percent is where you're
headed or much higher?
A Well, I think it's going to be more than fifteen
percent.
Q Do you add the fifteen and the fifteen and the
fifteen, is that the way it works or. . .
A The error is essentially additive. It's not going
to be fifteen plus fifteen, plus fifteen, per se.
Q Right.
A But the error of these different parameters ÄÄ
errors in these different parameters are going to be
additive.
Q Okay. What is your best guess at this time as to
where that is headed?

MR. GREEN: Asked and answered.

BY MS. PONZOLI:
Q Okay.
A I really hate to say. I don't know what it's going
to be.
Q Do you have a goal in your mind, that it won't go
beyond a certain amount?
A No, I can't set a goal, because I think the error
is going to be constrained by the quality of the data.
Q I think I'm understanding what you're saying. Your
point is not being missed, if that's any consolation to you.
A Okay.
Q I was asking you originally, before we got off on
this, what were the areas of uncertainty and you said
rainfall, ET flow.
A Right.
Q Is there anything else?
A Certainly the phosphorus uptake grade coefficient,
precipitation inputs of phosphorus also is going to be an
uncertainty.
Q Anything else?
A Oh, I'm sure there are some others, but I think
those are probably the major areas.
Q And, again, if it's five to fifteen that would be
the range of uncertainty that you would expect for each of
those?
A For phosphorus input that really ÄÄ that's a
question that almost knows no answer at this point, because
there's very limited data on phosphorus chemistry and
atmospheric deposition that I consider to be reliable.
Q So that one is totally unknown.
And the phosphorus uptake grade?
A The phosphorus uptake grade likely will be a
calibrated parameter.
Q I'm sorry, I don't know what you mean.
A Well, what we'll do is estimate the phosphorus
uptake grade based upon phosphorus concentrations that we
observe at different points and then, in essence, manipulate
that number or try different phosphorus uptake rates until we
get what we think is the best fit between the observed and
predicted phosphorus concentration data at the various
control points.
Q So you won't be able to state the uncertainty that
surrounds that?
A I'm sure we'll be able to state some uncertainty
about it and that will be an exercise that the folks at Tetra
Tech should be able to do.
Q All right. You have run some management scenarios,
is that accurate, on this model?
A We have done some preliminary exercises with the
model just to show the types of output or the type of
response that we can evaluate.
Q Okay. Can you describe some of those scenarios for
me?
A We have, with the model, looked at various ÄÄ the
effects at downstream structures at various STA
configurations.
Q If you were to build the full ÄÄ oh, what is it?
Whatever, in this one plane, the exact figure of the STA and
have the twenty-five percent BMP Rule in place, what are the
downstream effects?
Do you understand my question?
A Yes, and I don't think it's appropriate for me to
answer at this point, because the model is not ÄÄ the
development is not complete at this point in time.
Q Well, have you run that scenario?
A Yeah, but we've run it with an incomplete model. I
mean, the model is not mentally calibrated and verified.
Q What were the results on an uncalibrated and
incomplete model?
A I don't have the outputs in front of me, but I
recollect that at the structures leading into the Everglades
National Park that the effects on the concentration were
fairly nominal.
Q In other words, for the efforts upstream the result
at the Park was, as you say, nominal.
A Yeah. We didn't see much of a change, I think, in
predicted concentrations.
We did see a change in predicted flux, but that was
really related to a hydrologic change and not related to a
concentration change.
Q Okay. What was the hydrologic change?
A I don't recall. It could have been twenty percent
reduction in flow, it could have been more, it could have
been upwards to fifty percent. I don't recollect exactly.
Q That would be a fairly significant change in flow,
would it not?
A Yes.
Q Do you believe that that is going to be the result
of this model when it's finally finished?
A I don't know what the results are going to be until
we have a model that is fully calibrated and verified.
Q Do you believe you will see significant reductions
of inflow at the Park at the S-12s if the BMPs and STAs, as
envisioned in the present SWIM Plan, are fully implemented?
A It all depends on what South Florida Water
Management District decides to do, in terms of water
management.
If ÄÄ there will be flow reduction coming out of
the EAA, with the implementation of the BMPs and STAs. If
that incremental drop in flow translates to a comparable
magnitude in decline of flow at the S-12s, then you're going
to have significant decrease in flow at the S-12s, but if the
District compensates for that flow at other points, then it
may be that the District is going to be able to mitigate any
sort of flow effects at the S-12s.
Q In regard to the BMPs, if there were indeed the
twenty percent reduction in flow, is it enhanced by the time
it gets to the Park? Has it increased from twenty percent to
some greater percentage by the time it is at the Park?
A I think it's not appropriate to look at a twenty
percent reduction in flow coming out of the EAA and
translating to a twenty percent reduction in flow at the
Park. What is important to look at is the absolute magnitude
in the decrease in flow coming out of the EAA and then look
at where those cubic feet or cubic meters or acre feet of
water, whatever units you want to use, where is that decline
in water flow actually perceived in the system.
Q Where do you believe it will be perceived in the
system?
A I really don't know.
Again, I think that's a management question for the
South Florida Water Management District.
Q Possibly will this just be a reverse interim action
plan?
A Do you mean in terms of pulling water out of Lake
Okeechobee? Is that what you mean?
Q Just water simply staying there in the first place.
A I'm really not sure I understand what you mean.
Q The water would ÄÄ you believe ÄÄ where does the
water go if it's held in the EAA? Is it simply held in
place?
A I'm not familiar enough with the water balance for
the farms to answer that question.
Q Okay. Is this a Mr. Gherini question?
A You can try Mr. Gherini.
Q Okay. What has your model shown in regard to the
STAs? What will be their impact? Will there be a reduction
in flows as a result of the STAs?
A The model does show some flow impacts.
Q What does it show?
A I don't recall what the precise number is.
Q Is it three percent?
A I don't know.
Q Is it ÄÄ I mean, is it three or is it thirty, do
you have any idea if it's somewhere along. . .
A I believe it would be closer to three than to
thirty.
Q Okay. Are you familiar with Dr. Millar's estimates
of ET loss in STAs?
A No, I'm not.
Q So your objectives of the model were to predict
flow and phosphorus concentration at the major control
structures and you told me that if you fully implemented the
BMP and STA program in the SWIM Plan it only had a nominal
effect in the present runs of the model at the S-12
structures to the Everglades National Park.
A That's correct.

MR. GREEN: Object to the form. Move to
strike.
I think you were talking about phosphorus
concentrations before.
MR. POLLMAN: Would you repeat the question,
please?
MR. GREEN: I want to make sure I didn't miss
your question.

BY MS. PONZOLI:
Q You were accurate, I was talking about phosphorus
concentration, but now I want to know, you know, it will not
have a negligible effect on flow; is that accurate?

MR. GREEN: Why don't you just read back the
last question?
Do you want it read back or do you want to re-
ask it?
MS. PONZOLI: I'll re-ask it.

BY MS. PONZOLI:
Q The objective of your model was to predict flows at
the major control structures and if you fully implemented the
BMPs and STAs, your model shows that you will have reduction
of flow at the S-12 structures. . .
A That's correct.
Q . . .BMPs?
A That's correct.
Q Do you know the percentage of reduction of flow?
A No, I do not.
Q Do you have an idea of the range of the reduction
of flow, assuming no other mitigating or compensating water
manipulations were instituted by the South Florida Water
Management District?
A It could be thirty to forty percent, something like
that; but, again, I don't recall precisely.
If I had the output in front of me, I proba