658 1 UNITED STATES DISTRICT COURT FOR THE 2 SOUTHERN DISTRICT OF FLORIDA 3 **************************** 4 UNITED STATES OF AMERICA, * Plaintiff * 5 * Case Number VS. * 88-1886-CIV 6 * Hoeveler SOUTH FLORIDA WATER * 7 MANAGEMENT DISTRICT, ET AL.,* Defendants * 8 ***************************** 9 10 Deposition of WILLIAM W. WALKER, JR., 11 taken on behalf of the defendants South Florida 12 Water Management District and John R. Wodraska 13 pursuant to the applicable rules of the Federal 14 Rules of Civil Procedure, before Nancy L. 15 Eaton, Registered Professional Reporter and 16 Notary Public within and for the Commonwealth 17 of Massachusetts, at the offices of Skadden, 18 Arps, Slate, Meagher & Flom, One Beacon Street, 19 Boston, Massachusetts, on Monday, February 11, 20 1991, commencing at 10:00 a.m. 21 22 23 LINDA MARIE MacDONALD, RPR-CM REGISTERED PROFESSIONAL REPORTER 24 60 LAWRENCE ROAD, PLYMOUTH, MA 02360 (508) 747-6615 659 1 A P P E A R A N C E S: 2 UNITED STATES ATTORNEY'S OFFICE By A.U.S.A. Richard Harrison 3 155 South Miami Avenue, Suite 600 Miami, FL 33130 4 for the United States of America. 5 SKADDEN, ARPS, SLATE, MEAGHER & FLOM By Attorney Laura B. Ahearn 6 1440 New York Ave., N.W. Washington, D.C. 20005 7 for South Florida Water Management District and John R. Wodraska. 8 PEEPLES, EARL & BLANK 9 By Attorney Rick J. Burgess One Biscayne Tower, Suite 3636 10 Miami, FL 33131 for the cities of Belle Glade and 11 Clewiston. 12 STATE OF FLORIDA OFFICE OF GENERAL COUNSEL By Asst. Gen. Counsel David A. Crowley 13 Department of Environmental Regulation Twin Towers Office Building 14 2600 Blair Stone Road Tallahassee, FL 332301 15 for the Florida Department of Environmental Regulation. 16 17 A L S O P R E S E N T: 18 Douglas Robson, Consultant, SFWMD John Davis, Consultant, Belle Glade 19 and Clewiston Peter Ghavami, Legal Assistant, 20 Skadden Arps 21 22 23 24 DEPOSITION OF WILLIAM W. WALKER, JR. 660 1 I N D E X 2 Witnesses Examination WILLIAM W. WALKER, JR. 3 (By Ms. Ahearn) 661 4 5 E X H I B I T S 6 Number For ID 7 DX 33 Confidential Memo 705 8 DX 34 Water Quality Trends, Executive Summary 722 9 DX 35 Declaration of 9-1990 723 DX 36 Article - Techniques of 10 Trend Analysis 750 DX 37 Article - Nonparametric 11 Trend Test for Seasonal Data with Serial Dependence 750 12 13 14 15 16 17 18 19 20 21 22 23 24 DEPOSITION OF WILLIAM W. WALKER, JR. 661 1 DIRECT EXAMINATION - CONTINUED - BY MS. AHEARN 2 Q. Good morning, Doctor Walker. 3 A. Good morning. 4 Q. Just to let you know, you are still under oath 5 from last week throughout this deposition. 6 Did you review any materials 7 regarding South Florida water quality over the 8 weekend? 9 A. No. 10 Q. Took it off. Last week before we commenced 11 your deposition, it is my understanding that 12 you met with counsel for the government on 13 Monday and Tuesday, is that correct? 14 A. Yes, I believe that's correct. 15 Q. Over that Monday and Tuesday, did you look at 16 materials regarding water quality in South 17 Florida? 18 MR. HARRISON: You may answer yes 19 or no. 20 A. Yes. 21 Q. Did you look at materials other than those that 22 you provided to be produced in conjunction with 23 your deposition? 24 A. No. DEPOSITION OF WILLIAM W. WALKER, JR. 662 1 Q. Can you describe for me the general nature of 2 the documents and materials you looked at prior 3 to the commencement of this deposition last 4 week? 5 MR. HARRISON: Objection, 6 counsel. You're going into my two-day 7 preparation session with Dr. Walker. He 8 already testified the materials we used that we 9 went over over the weekend were materials that 10 he has provided for this deposition. I think 11 it is inappropriate for you to ask further as 12 to what Dr. Walker and I went into. It is 13 clearly attorney client. 14 MS. AHEARN: I didn't ask him what 15 nature. I asked him the nature of the 16 materials that he did look at. That would 17 clearly go to the deponent's recollection, his 18 ability to seize on certain facts that might 19 relate to one type of the science you have been 20 looking at versus the other, so I think my 21 question was appropriate. 22 MR. HARRISON: Counsel, it goes 23 into whatever my mental impressions are, what 24 we were going to work for in this deposition. DEPOSITION OF WILLIAM W. WALKER, JR. 663 1 It is entirely beyond the scope of anything you 2 have a right to go into and wish you would go 3 into the science, but not asking what him and I 4 did at our preparation session. I am not going 5 to let him answer those questions. 6 Q. Dr. Walker, I'd like to get a complete overview 7 of the opinions that you have formulated that 8 you would be prepared to give at trial in this 9 case. Now, let me make clear, I'm not asking 10 you to tell me what the United States attorneys 11 have or haven't told you they'll want you to 12 testify to. I am simply trying to identify the 13 opinions in your mind which you have formulated 14 to such a degree that you as a scientist would 15 be prepared to testify as to those opinions at 16 trial. 17 MR. HARRISON: Counsel, just to 18 clarify any misinterpretation of your 19 characterization, the United States has never 20 told Dr. Walker what they want him to testify 21 to nor would we ever -- or at least I certainly 22 as an attorney would not. 23 MS. AHEARN: I was simply trying 24 to fend off your objection. DEPOSITION OF WILLIAM W. WALKER, JR. 664 1 MR. HARRISON: We have discussed 2 areas which are necessary, or we as lawyers 3 believe are necessary, for proof in this case, 4 but Doctor Walker's opinions are his own. The 5 United States has certainly never put opinions 6 in Doctor Walker's head and I just want to 7 clear up that. 8 Q. Have you formulated such opinions that you 9 would be prepared to give at trial regarding 10 the current existence of water quality problems 11 within Everglades National Park? 12 MR. HARRISON: Objection to form. 13 Counsel, just to make sure we have 14 a clear record, are you asking about stuff that 15 Doctor Walker has opinions he has formulated 16 based on his own work or opinions that he would 17 be comfortable in giving based on work of other 18 and reliance thereon? 19 MS. AHEARN: What I'd prefer to 20 do, Mr. Harrison, is identify those opinions 21 and then I will give Dr. Walker the opportunity 22 to explain his bases, identify what he relies 23 on for each opinion. 24 MR. HARRISON: Okay. DEPOSITION OF WILLIAM W. WALKER, JR. 665 1 A. I believe that I testified last week that one 2 of the tasks in which I am currently involved 3 is a study of trends at monitoring stations 4 within the Park. I have not completed that 5 task and I have not at this time formulated 6 definite opinions regarding that analysis. 7 Q. Have you formulated tentative opinions at this 8 time? 9 A. The analysis that I've completed thus far 10 indicates that increasing trends in phosphorus 11 concentrations are likely at some of the 12 stations monitored within the Park. 13 Q. Are these increasing trends in total 14 phosphorus? 15 A. That's correct. 16 Q. Any other constituents? 17 A. The preliminary analysis also indicates a 18 decreasing trend in dissolved oxygen 19 concentration at one station. 20 Q. What station is that? 21 A. I believe it was station P33. 22 Q. When you say that trends are likely based on 23 your preliminary analyses, can you quantify 24 this likelihood? DEPOSITION OF WILLIAM W. WALKER, JR. 666 1 A. The likelihood is defined in the same way that 2 I define the likelihood in the trend analysis 3 that I conducted on the inflows to the Park as 4 described in the report that I submitted to the 5 Justice Department and as described in the 6 declaration that I signed for the Justice 7 Department regarding trends in water quality at 8 inflow points to the Park. 9 Q. Are you using the identical statistical 10 procedures in your Park marsh station analysis 11 that you used in the analysis reflected in 12 Exhibit 17, your final report? 13 A. The fundamental statistical procedure that is 14 used to examine or to estimate the likelihood 15 of trend is the same, that is, the Seasonal 16 Kendall test. 17 Q. Does the work differ in some other respects, 18 for example, maybe in how you have to adjust 19 the database? 20 A. What do you mean by adjust? 21 Q. Well, I asked you if you were using the same 22 procedures and in answering you told me that 23 you are using the Seasonal Kendall test. I am 24 trying to confirm whether or not other DEPOSITION OF WILLIAM W. WALKER, JR. 667 1 procedures used in Exhibit 17 are also being 2 employed in the marsh station analysis or not. 3 A. At this point in my research, I have not 4 completed the investigation of hydrologic 5 factors as they may contribute to variations in 6 the water quality at marsh stations. 7 Q. So is it too early to tell whether you will use 8 the same procedures? 9 A. That's correct. 10 Q. And then you told me that in terms of 11 quantifying this likelihood, you're doing it in 12 the same manner as Exhibit 17. Have you 13 quantified for the marsh stations the 14 likelihood of increasing trends in total 15 phosphorus? 16 A. The statistical procedure that I'm using tests 17 the hypothesis that is called the null 18 hypothesis of no increasing trend or no change 19 in the data. 20 The likelihood is expressed in 21 terms of the probability that that hypothesis 22 is true, and when the likelihood -- when that 23 probability is less than .1, as described in my 24 report on the inflow trends, I'm using that as DEPOSITION OF WILLIAM W. WALKER, JR. 668 1 a working definition of differentiating between 2 stations with trends and stations without 3 trends. 4 Q. Have you also done analyses of the marsh 5 stations using a probability factor of less 6 than .05? 7 A. Well, the Kendall test estimates that 8 probability. The interpretation of the 9 importance of that probability in terms of 10 whether or not a trend exists is really not a 11 statistical question. The procedure estimates 12 the probability and I am just using .1 as a 13 convenient way of summarizing the results and 14 tabulating the results. I'm also reporting the 15 exact probability levels as they are derived 16 from the Seasonal Kendall test. 17 Q. At the probability level of less than .10, how 18 many of how many total marsh stations show some 19 trend? 20 A. I don't recall those details. 21 Q. How many total marsh stations have you analyzed 22 so far? 23 A. I believe there are nine stations. 24 Q. Do you recall whether using a probability level DEPOSITION OF WILLIAM W. WALKER, JR. 669 1 of less than .05 you get a certain number of 2 these nine stations showing a likely trend? 3 A. I don't recall. 4 Q. How many total water quality constituents are 5 you looking at at these marsh stations? 6 A. I don't recall the exact number. 7 Q. Is it more than twenty? 8 A. I don't recall the exact number. 9 Q. Do you recall if it is more than ten? 10 MR. HARRISON: Objection. Asked 11 and answered. 12 A. I don't recall the exact number. 13 Q. Are there documents among those you produced 14 for the deposition that would reflect the 15 number of constituents being analyzed as well 16 as those that show likely trends at given 17 probability levels? 18 A. Yes, there are. 19 Q. Would these be computer printouts? 20 A. I guess you could define them as computer 21 printouts, yes. 22 Q. What would you define them as? 23 A. Tables. 24 Q. Can you give me identifying features of these DEPOSITION OF WILLIAM W. WALKER, JR. 670 1 tables so that I could find them in your 2 documents? 3 A. I recall a series of tables that are 4 collections that are in a similar format to 5 those that are included in Appendix A of 6 Exhibit 17, that is, my report on the inflow 7 trends. And I recall some one or more tables 8 that summarize results of the Seasonal Kendall 9 test for different stations and water quality 10 components. 11 Q. Great, thanks. That will give me some 12 guidance. Do you recall approximately when 13 these tables were generated? 14 A. They would have been generated in -- within the 15 last two months. 16 Q. Is it your opinion as an environmental engineer 17 specializing in water quality that this 18 likelihood of an increasing trend in total 19 phosphorus at certain Park stations constitutes 20 an existing water quality problem in the Park? 21 MR. HARRISON: Object to form. 22 Inflow stations or marsh stations or both? 23 Q. The marsh stations that we just discussed. 24 A. As I stated earlier, I have not completed my DEPOSITION OF WILLIAM W. WALKER, JR. 671 1 analysis of these data. I have not completed 2 my analysis of the hydrologic factors as they 3 may influence these measurements, and so I 4 cannot at this point state one way or another 5 whether I believe these trends from the 6 analysis that I have completed thus far 7 constitute water quality problems. 8 Q. Is that answer also applicable if we consider 9 other work that you've done regarding water 10 quality in South Florida in addition to the 11 marsh trend analysis? 12 A. No. 13 Q. Based on that entirety of work that you've 14 done, do you have an opinion whether there are 15 existing water quality problems within the 16 Park? 17 A. The work that I've completed on the inflows to 18 the Park indicates that most of the stations 19 that are monitored at those inflow points show 20 increasing trends in phosphorus, decreasing 21 trends in nitrogen to phosphorus ratio during 22 the period of monitoring that I examined. 23 Inasmuch as the Park ecosystems 24 are downstream of those inflow points, I DEPOSITION OF WILLIAM W. WALKER, JR. 672 1 believe that those trends at the inflow points 2 constitute water quality problems for the Park. 3 Q. Is it your opinion as an environmental engineer 4 specializing in water quality that this 5 likelihood of trend which your work shows of 6 increasing phosphorus concentrations in Park 7 inflows constitutes a current as opposed to 8 threatened problem for the Park? 9 MR. HARRISON: Objection, 10 counsel. Asked and answered. 11 A. I don't understand your question. 12 Q. I asked you if you were of the opinion whether 13 a water quality problem existed within the Park 14 and your answer suggested to me that the trend 15 you see at the perimeter of the Park is 16 upstream of marsh stations and in that respect 17 constitutes a problem. From the perspective of 18 the interior marsh stations, is this an 19 existing problem or a threatened problem? 20 A. My opinion is based upon analysis of data that 21 was collected between December of 1977 and 22 September of 1989. The fact that phosphorus 23 concentrations increased at Park inflow points 24 during that period indicates to me that during DEPOSITION OF WILLIAM W. WALKER, JR. 673 1 that period the Park -- the downstream 2 locations in the Park -- were subjected to 3 those higher phosphorus concentrations and that 4 would constitute water quality problems. 5 Q. Have you an opinion whether the trends of 6 increasing phosphorus concentrations and 7 inflows which you report has caused any 8 imbalance in natural populations of aquatic 9 flora or fauna within the Park? 10 MR. HARRISON: Objection. Lack of 11 foundation to show that he's done that type of 12 work. 13 Also objection to the definition 14 of the term balance. 15 A. My work has focused on water quality measures 16 at Park inflows and within the Park and not on 17 direct measurements of flora and fauna and the 18 imbalance thereof. 19 Q. Consulting the work of others on flora and 20 fauna, have you any opinion? 21 A. No. 22 Q. Have you any opinion whether natural 23 populations of aquatic flora or fauna within 24 the Park are imbalanced? DEPOSITION OF WILLIAM W. WALKER, JR. 674 1 MR. HARRISON: Objection, 2 counsel. I don't see where it differs from 3 your last question. 4 MS. AHEARN: I think my last 5 question had a causation element in it which 6 this question does not. 7 MR. HARRISON: Object to form. 8 Imbalanced from what? 9 A. Would you rephrase the question? 10 Q. Have you any opinion whether there exists an 11 imbalance in natural populations of aquatic 12 flora or fauna within the Park? 13 MR. HARRISON: Object to form. 14 A. My work is focused on the analysis of water 15 quality data and has not been -- I have not 16 analyzed data on that topic. 17 Q. Have you any opinion as to the cause of the 18 increasing trend in total phosphorus which you 19 report in Park inflows? 20 A. I have identified categories of causes that 21 could be factors but I have not quantitatively 22 evaluated them. 23 Q. What are the categories of factors which you've 24 identified? DEPOSITION OF WILLIAM W. WALKER, JR. 675 1 A. The first category would be that the increasing 2 trends in phosphorus at Park inflows reflect 3 increases in the concentrations or loadings of 4 phosphorus discharged to the water conservation 5 areas from the adjacent watersheds. 6 The second category is related to 7 a loss in the retention capacity or the 8 assimilative capacity of the water conservation 9 areas to absorb or remove the phosphorus 10 loadings discharged from -- discharged to the 11 water conservation areas from adjacent 12 watersheds. 13 The third category has to do -- 14 pertains to changes in the distribution of flow 15 or changes in water management, the 16 distribution of flow into the Park which may 17 have influenced phosphorus concentrations at 18 certain stations during this time period; and 19 the fourth category of cause has to do with 20 climatologic variations which I believe could 21 be related or -- could be related to the 22 apparent trends. 23 Q. Before you -- strike that. 24 You said you hadn't quantified DEPOSITION OF WILLIAM W. WALKER, JR. 676 1 these categories of factors. Do you presently 2 have any preliminary opinions as to which if 3 any or all of these factors are causally 4 related to the increasing trends in total 5 phosphorus that you report? 6 MR. HARRISON: Object to 7 characterization. I think he's already said 8 that he's identified all four as being causally 9 related. That was my understanding. Feel free 10 to clear it up. 11 A. I'm sorry. I lost your question. Would you 12 repeat your question or rephrase it? 13 Q. At this point are you ready to give an opinion 14 whether or not each of these categories of 15 factors is causally related to the trends that 16 you report? 17 A. As described in Exhibit 17, Series C of the 18 analysis adjusted for effects of hydrologic 19 factors including antecedent rainfall, water 20 elevation and in the case of the analysis of 21 trends in nutrient transport, the work also 22 considered variations in flow. 23 Based upon that, I believe that 24 the fourth category that I've identified DEPOSITION OF WILLIAM W. WALKER, JR. 677 1 earlier, that is, the category relating to 2 climatologic or hydrologic variations, is a 3 relatively unimportant factor. 4 With respect to the other three 5 factors, my current opinion is based upon the 6 analysis of the inflow data; and the 7 preliminary analysis of the inflow data to the 8 water conservation areas, that is, analysis of 9 the inflows to the Park and the analysis of the 10 inflows to the water conservation areas which 11 generally showed similar patterns, and I 12 therefore believe, based upon the work that 13 I've completed thus far, that it is likely that 14 increasing loads to the water conservation 15 areas and increasing concentration at inflow 16 points to the water conservation areas are 17 causative factors contributing to the trends at 18 the Park inflow points. 19 With respect to the other factors 20 and with respect to overall quantifying each of 21 these factors and putting numbers on each of 22 them, I have not as yet completed my work so 23 that I would be able to do that. 24 Q. Now, with one, your category one, where you say DEPOSITION OF WILLIAM W. WALKER, JR. 678 1 that it is likely that it is a causative factor 2 and your category four, climatologic 3 variations, where you say it is probably 4 relatively unimportant, are you relying there 5 on the work you've done to investigate 6 statistical correlation with those factors and 7 total phosphorus concentration and loadings? 8 A. I have, to the best of my ability, attempted to 9 remove, using the statistical techniques that 10 are described in Exhibit 17, those portions of 11 the variations in phosphorus concentrations 12 that are related to hydrologic factors and I 13 have found trends remaining in the data. 14 Q. And those trends you believe are likely to have 15 some causal relationship with increased 16 concentrations or loadings of phosphorus 17 discharged to the water conservation areas, 18 that was your category one? 19 A. I believe that category one is likely to be a 20 factor contributing to those trends. 21 Q. And am I correct, your work is still too 22 preliminary to reach that type of tentative 23 opinion regarding your categories two and 24 three? Can you tell us now whether categories DEPOSITION OF WILLIAM W. WALKER, JR. 679 1 two and/or three have likely causal 2 relationships with total phosphorus? 3 MR. HARRISON: Counsel, if you're 4 asking for a preliminary opinion, I have no 5 objection. He's already stated that the work 6 has not progressed to the point that he has 7 firm opinions on that. 8 A. My work has not completed to the point of being 9 able to quantitatively rank these various 10 potential causes. 11 Q. Let me just try and ask it one more way. It 12 seems to me you told us that in the final 13 category of causative factors, it is likely 14 that your category one is in and your category 15 four is relatively unimportant. I am not 16 clear, where do you stand currently on 17 categories two and three? 18 A. I am still in the process of evaluating all of 19 these categories. All of these are preliminary 20 conclusions that we're discussing regarding 21 causation and preliminary opinions. 22 Q. I recognize that. If you've told it to me, I 23 apologize, I missed it. Do you have a 24 preliminary opinion on the causal relationships DEPOSITION OF WILLIAM W. WALKER, JR. 680 1 of your categories two and three to the 2 phosphorus inputs to the Park? 3 A. I have preliminary opinions regarding category 4 three. 5 Q. And what is that opinion? 6 A. As I stated earlier, it is possible that trends 7 at some of these stations may be related to 8 changes in the distribution of flow entering 9 the Park. The one -- one example of that, if I 10 could refer to Exhibit 17? 11 (Witness looked at document.) 12 Figure 2 on page 4 shows the 13 monthly flows that were released at the various 14 inflow points to Shark slough including S12A, 15 S12B, S12C, S12D and S333; and as is described 16 in the report, there have been various changes 17 in the relative magnitudes of those release 18 points, in particular, the latter years as -- 19 in an effort to provide increased flow to the 20 northeast portion of Shark River slough, there 21 was an increased flow through the structure 22 S333. 23 The S333 structure is located 24 right adjacent to the L67A canal and it is DEPOSITION OF WILLIAM W. WALKER, JR. 681 1 possible that some of the -- some portion of 2 the overall trend which I identified for the 3 combined discharge to Shark slough, that is, 4 the station S12_334, is related to that 5 increased discharge of canal water directly 6 into the Park in the later portion of the 7 record. 8 The overall magnitude of that 9 effect, if I could refer to Figure 10 on page 10 37, there is a trend line shown for what I call 11 station S12T which is the composite discharge 12 through the S12s, A, B, C and D, and that, as 13 indicated, there is a significant increasing 14 trend. There is also a trend line shown for 15 the station S12_334 which is the total 16 discharge to Shark slough. 17 And my preliminary opinion 18 regarding an effect of water management is that 19 the difference in the trend between that 20 measured at or for S12, for the S12 composite, 21 and the trend reported for the Shark River 22 slough composite could be related to the 23 increase inflow through S333. 24 Now, it is possible that trends DEPOSITION OF WILLIAM W. WALKER, JR. 682 1 also exist at S333, so we don't know whether 2 the difference between S12T and S12_334 is just 3 an effect of water management or whether there 4 are also effects of the other causal factors 5 involved here as well; but that is one example 6 of water management -- of a case where water 7 management may influence trend at a particular 8 station. 9 Q. And this preliminary opinion regarding this 10 water management change as reflected in S12T 11 and S12_334 is based on correlative analyses 12 that are described in Exhibit 17; is that 13 correct? 14 A. No, it is based upon common sense and mass 15 balance. 16 Q. Are you relying on any particular body of 17 scientific literature in deriving this opinion? 18 A. I am relying on the fundamental principle of 19 mass balance. 20 Q. What experimental work have you done concerning 21 potential causative factors for the increasing 22 trends that you report for Park inflows? 23 A. I've done no experimental work. That is not 24 part of the scope of work and is not part of DEPOSITION OF WILLIAM W. WALKER, JR. 683 1 what I normally do in the course of analyzing 2 data from a system such as this. 3 Q. Have you reviewed any experimental work on this 4 topic performed by others? 5 A. We are discussing a very large and very complex 6 system that is operated for many purposes, and 7 the feasibility of doing experimental work to 8 evaluate causes of trends, the feasibility of 9 experimenting with this whole system to 10 evaluate causes is nonexistent. 11 Q. Would you agree that it is hazardous as a 12 matter of science to accept correlation as 13 sufficient proof of causation? 14 MR. HARRISON: Object to form. 15 Are you speaking in particular context of this 16 particular study Doctor Walker has done or are 17 you just asking any field of science? 18 Q. I'm asking if he would agree with that 19 statement. 20 A. Correlation can be used as a basis for forming 21 opinions regarding causation. It is not 22 necessarily hazardous to do so. It is on a 23 case specific basis. Correlation is sufficient 24 -- is a factor that can be used in assisting DEPOSITION OF WILLIAM W. WALKER, JR. 684 1 one in formulating opinions regarding 2 causation. 3 Q. Is correlation alone a sufficient basis on 4 which to form an opinion constituting proof in 5 court? 6 MR. HARRISON: Objection, counsel. 7 Q. Let me rephrase the question. Dr. Walker, in 8 your opinion is correlation alone sufficient on 9 which to base an opinion concerning causation 10 of a water quality trend? 11 MR. HARRISON: Object to form. 12 Question is too broad. 13 A. It is valid to consider correlation as a factor 14 in formulating one's opinions regarding 15 causation. 16 Q. Would you base an opinion on this matter on 17 correlative statistics alone? 18 MR. HARRISON: Objection. Asked 19 and answered. 20 A. The correlative statistics or the trends that 21 have been reported in the work that I've done 22 so far describe the development of a problem at 23 Park inflow points over the period of record 24 that was monitored. DEPOSITION OF WILLIAM W. WALKER, JR. 685 1 The interpretation of causation 2 can be based partially upon those observed 3 trends and upon interpretation of other factors 4 and other observation and other work that have 5 been made that has occurred in the water 6 conservation areas. 7 Q. Developing your opinions concerning the water 8 quality in the Park, have you relied on 9 specific work performed in the water 10 conservation areas other than your own? 11 A. I have relied on the description of the work 12 that has been done that is documented in the 13 SWIM plan regarding the development of 14 eutrophication in certain of the -- certain 15 areas of the water conservation areas. 16 Q. Any other documents or reports concerning work 17 in the water conservation areas on which you've 18 relied? 19 A. The list of or the documents which I turned 20 over to the Justice Department contain a number 21 of references that are related to this topic 22 that are outside of the SWIM plan, so those 23 would also be included in the list of documents 24 that I have read and relied upon. DEPOSITION OF WILLIAM W. WALKER, JR. 686 1 Q. Do you have any opinion as to whether water 2 quality problems currently exist within the 3 Loxahatchee Refuge? 4 MR. HARRISON: Object to form. 5 A. I believe that the trend analysis of water 6 conservation area inflow and outflow points 7 that I described last week as one of the tasks 8 with which I am currently involved indicated a 9 likely trend or trends at one or more of the 10 outflow points from the water conservation 11 areas and those trends were increasing 12 phosphorus concentration. 13 Q. So you found those trends specifically at 14 points leading into the Refuge? 15 A. I believe they were trends at points coming out 16 of the Refuge. 17 Q. Would those be the S10 structures? 18 A. I believe that's correct. 19 Q. Any other structures or points leading out of 20 the Refuge at which you found these trends? 21 A. I don't recall. 22 Q. Have you any opinion as to whether there exists 23 an imbalance in natural populations of aquatic 24 flora or fauna in the Refuge? DEPOSITION OF WILLIAM W. WALKER, JR. 687 1 MR. HARRISON: Objection, 2 counsel. No foundation. 3 A. I have not studied the imbalance of natural 4 flora or fauna in the Refuge and I have no 5 opinion on that. 6 Q. Have you any opinion as to what factors have 7 caused the water quality problem at the Refuge 8 as you've defined it, your likely trends in 9 increasing total phosphorus at outflow points? 10 A. My work in Water Conservation Area 1 or the 11 Loxahatchee National Wildlife Refuge has been 12 very preliminary and I have not formulated any 13 such opinions in that area. 14 Q. That would include tentative opinions. 15 A. I really haven't thought about it as much as 16 the other -- the Park inflow points. I have 17 not completed my analysis of Loxahatchee. 18 Q. Doctor Walker, do you have an opinion as to 19 what numeric level as a total phosphorus 20 standard for inflows to Everglades National 21 Park should be adopted? 22 MR. HARRISON: Object to form. 23 Adopted for what purpose? And also the term 24 standards is not defined. DEPOSITION OF WILLIAM W. WALKER, JR. 688 1 Q. I'll rephrase the question. 2 MR. HARRISON: I am going to allow 3 him to answer anything that he believes 4 standards for inflows. I just want to make 5 sure that the question is couched right as to 6 what purpose and what type of standard you're 7 talking about, counsel. These questions can be 8 misinterpreted. 9 Q. Dr. Walker, in the context of ONRW standards, 10 have you an opinion as to what number should be 11 used as the limitation for acceptable mean 12 annual total phosphorus concentrations for 13 inflows to Everglades National Park? 14 MR. HARRISON: Object to form. 15 A. The ONRW methodology as we developed it in a 16 number of working sessions with the District 17 staff did not involve setting standards on a 18 mean annual timeframe, so your question makes 19 no sense to me. 20 I can't answer it. 21 Q. Do you have an opinion as to whether numeric 22 limitations for phosphorus -- total phosphorus 23 inputs to Everglades National Park should be 24 implemented? DEPOSITION OF WILLIAM W. WALKER, JR. 689 1 A. Yes. 2 Q. What is the nature of the limitations which you 3 believe should be implemented? 4 MR. HARRISON: Object to form. 5 You may answer. 6 A. My recommendations concerning phosphorus, 7 interim phosphorus standards for inflows to 8 Everglades National Park, are described in what 9 was introduced as Exhibit No. 28 last week. 10 Essentially these are limits that are based 11 upon the first five years of monitoring data in 12 each Park basin, that is, Shark River slough, 13 Taylor slough and the coastal basin, and these 14 are standards that would apply to five-year 15 running flow-weighted mean concentrations for 16 each basin. The specific values for these 17 basins are derived directly from the monitoring 18 data and are identified in Exhibit 28. 19 Q. Could you please direct me in Exhibit 28 to the 20 specific numbers which you recommend be 21 employed as these limitations? 22 A. Well, with respect to the flow-weighted mean 23 concentrations, these values are identified for 24 each basin in Table 1 of Exhibit 28, and in the DEPOSITION OF WILLIAM W. WALKER, JR. 690 1 case of Shark slough, the standard would be 2 designed to preserve a long term flow-weighted 3 mean concentration not greater than .0083 4 milligrams per liter. For any five-year period 5 the flow-weighted mean concentration would not 6 exceed .009 milligrams per liter. 7 In the case of Taylor slough as 8 shown in Table 1, the standards are designed to 9 preserve a long term flow-weighted mean of 10 .0078 milligrams per liter and the five-year 11 running flow-weighted mean concentration for 12 any five-year period would not exceed .0091 13 milligrams per liter. 14 And in the case of the coastal 15 basin, the standards would preserve a long term 16 flow-weighted mean of .0084 milligrams per 17 liter. The value for any five-year period 18 would not exceed .0096 milligrams per liter. 19 These numbers are derived directly from the 20 monitoring data for the first five years of 21 record in each basin using the statistical 22 methods that we agreed upon and developed with 23 the District staff under the ONRW discussions. 24 Q. Have you any opinion as to whether these DEPOSITION OF WILLIAM W. WALKER, JR. 691 1 numeric limitations that you've just pointed 2 out to us on Table 1 should be implemented 3 through the South Florida litigation? 4 MR. HARRISON: Objection. You may 5 answer. 6 A. These limits would insure that the inflow 7 concentrations to the Park in each basin would 8 not exceed the values that I've listed. 9 Exactly how the limits are implemented would 10 not matter. 11 Q. Have you any opinion as to whether these 12 limitations are achievable? 13 MR. HARRISON: Object. Lack of 14 foundation. 15 A. As I mentioned, these limits are based upon the 16 first five years of monitoring data for each 17 basin. They therefore reflect a condition that 18 once existed at the inflows to the Park. 19 Therefore, I believe they are achievable. 20 Q. Have you any opinion as to what reasonable and 21 reliable measures, if any, should be 22 implemented upstream of the Park to achieve 23 these standards? 24 MR. HARRISON: Objection. DEPOSITION OF WILLIAM W. WALKER, JR. 692 1 Foundation. 2 You may answer. 3 A. Yes. 4 Q. What is your opinion in that regard? 5 A. I believe that measures should be taken to 6 control or reduce the loading of phosphorus 7 entering the water conservation areas from the 8 various sources. 9 Q. What reasonable and reliable methods should be 10 implemented to control phosphorus loading to 11 the water conservation areas? 12 MR. HARRISON: Object to the form 13 of the question and to the foundation or lack 14 of. 15 A. I would generally categorize those methods as 16 methods that are directed at reducing the 17 loading and concentration of phosphorus in the 18 watersheds, in runoff from the watersheds that 19 drain into the water conservation areas using 20 any of a variety of best management practices 21 or other techniques for doing so. 22 Another category of control would 23 be use of what has been termed by the district 24 as water management areas or buffer areas for DEPOSITION OF WILLIAM W. WALKER, JR. 693 1 -- to provide interception of phosphorus 2 loadings from the adjacent land uses and 3 removal of phosphorus prior to discharge into 4 the water conservation areas. 5 Q. Is it necessary to reduce existing phosphorus 6 loads to the water conservation areas in order 7 to comply with your interim standards in Table 8 1? 9 MR. HARRISON: Object to form. I 10 think that the question requires more 11 specificity. His analysis includes several 12 combinations of structures. Are you asking 13 about everything or a particular basin? 14 A. Would you repeat the question, please? 15 Q. Is it necessary to reduce your existing 16 phosphorus loads to water conservation areas in 17 order to comply with the interim standards in 18 your Table 1 of Exhibit 28? 19 A. Well, in the case of Taylor slough and the 20 coastal basin, these systems are not directly 21 -- these systems are not directly linked to 22 the water conservation areas. There is some 23 linkage, but they are also influenced by other 24 watersheds, so it is very difficult to answer DEPOSITION OF WILLIAM W. WALKER, JR. 694 1 your question with -- as to whether reducing 2 phosphorus loads to the water conservation 3 areas is necessary to achieve these limits for 4 Taylor slough and the coastal basin. 5 With respect to Shark slough, I 6 believe that some reduction in the loads to the 7 water conservation areas would be needed in 8 order to comply with these limits. 9 Q. Do you know how much in terms of metric tons 10 per year? 11 A. My work has not progressed to the point of 12 being able to predict that or to quantitatively 13 specify that. 14 Q. Can you predict how much loading in terms of 15 metric tons per year could be reduced by the 16 variety of BMPs which you say are available? 17 A. No, I cannot. There are a range of techniques 18 and the particular performance of those 19 techniques and the amount of phosphorus that 20 would be controllable using those techniques 21 would depend upon the specific techniques and 22 the specific conditions that are present in the 23 watersheds and how effectively these are 24 applied and enforced. DEPOSITION OF WILLIAM W. WALKER, JR. 695 1 Q. Have you done any investigation of those 2 techniques, the conditions that exist in the 3 watersheds and how they can be enforced? 4 A. Not personally, no. 5 Q. Have you reviewed the work of others in this 6 regard? 7 A. I believe that my review has been limited to 8 whatever discussion is contained in the SWIM 9 plan. 10 Q. Are there any specific BMPs or similar methods 11 which, in your opinion, are more likely to lead 12 to a reduction of phosphorus loading? 13 MR. HARRISON: Object to form. 14 A. That's a very general question. If you could 15 be more specific, maybe I could answer that. 16 Q. Are there specific BMPs and similar measures 17 which, in your opinion, should be implemented 18 in order to reduce phosphorus loadings to the 19 water conservation areas? 20 A. I have not analyzed in any detail the various 21 options that are available, and one category of 22 control that seems to me to make some sense for 23 -- at least for the point of view of further 24 investigation -- would be more careful control DEPOSITION OF WILLIAM W. WALKER, JR. 696 1 of water level in the agricultural fields. 2 Q. Without further field investigation, would you 3 be able to determine what reduction in 4 phosphorus loading could be achieved through 5 more careful control of water level? 6 A. I have not done any field investigation. 7 Q. Let me put it this way. Does the data exist, 8 the information exist such that you could 9 determine the potential phosphorus reduction 10 from more careful control of water level or is 11 more field work still necessary to create that 12 basic raw data? 13 A. Well, my research has not focused on the 14 quantitative aspects of that. I believe that 15 that is a category of management option that 16 should be implemented. There may be other 17 experts and a body of research that would be 18 available to make quantitative predictions in 19 that regard but that is not my field. 20 Q. And you corrected me in that the Taylor slough 21 basin and coastal basin aren't so directly 22 impacted by the loadings to the water 23 conservation areas. What are the methods that 24 should be implemented to ensure that your DEPOSITION OF WILLIAM W. WALKER, JR. 697 1 interim standards from Table 1 in Exhibit 28 2 are met in those two basins? 3 MR. HARRISON: Object to the 4 characterization of his prior testimony. 5 A. I have not investigated in any way the control 6 possibilities for these watersheds, but I will 7 state that the numeric limits that are 8 specified in Table 1 are very close to the 9 values that were measured in the last five 10 years of data that I looked at. In other 11 words, under -- at least looking at the last 12 five years of data -- the system would already 13 be very close to being in compliance with these 14 limits. 15 Q. So that additional control methods might not be 16 necessary? 17 A. I wouldn't go so far as to say that because the 18 -- the analysis the trend analysis indicated 19 significant increasing trend especially in the 20 Taylor slough basin, so some form of control 21 may be necessary to arrest that trend in order 22 to ensure compliance with these limits. 23 Q. For purposes of establishing these interim 24 standards for phosphorus concentrations into DEPOSITION OF WILLIAM W. WALKER, JR. 698 1 the Park, is there any important difference 2 between a level of .0083 milligrams per liter 3 and .01 milligrams per liter? 4 MR. HARRISON: Objection to form. 5 A. I am not in a position to state whether or not 6 such differences are important. These are the 7 numbers that came out of the statistical 8 analysis. If someone wants to round them off 9 in one direction or another, that would be up 10 to them. That would be a policy decision. But 11 from the point of view of having a set of 12 numbers that come directly out of the baseline 13 data, these are the numbers that come out and I 14 feel uncomfortable with rounding them off and 15 using those rounded off values as standards. 16 Q. Have you any opinion as to whether a standard 17 of .01 milligrams per liter as opposed to .0083 18 milligrams per liter would have any impact on 19 the biota within the Park? 20 MR. HARRISON: Counsel, object, 21 lack of foundation. 22 A. I have seen no indication that .01, that a 23 standard of .010 milligrams as compared to the 24 .0083 milligrams per liter would not result in DEPOSITION OF WILLIAM W. WALKER, JR. 699 1 some impact. 2 Q. Have you seen any information which suggests 3 the limit of .01 milligram per liter would 4 adequately protect the Park from adverse 5 impacts? 6 A. These numbers in Table 1 of Exhibit 28 are 7 based upon the first five years of record for 8 each station. And the numeric values assume a 9 five-year averaging period. Now, a level of 10 .01 milligrams per liter applied to a shorter 11 timeframe, shorter than five years, might be 12 statistically the equivalent of this standard 13 of .0083 and provide the same level of 14 protection, but the important distinction here 15 is that .01 would apply to a shorter timeframe 16 than five years. 17 Q. So .01 would be acceptable if it applied to the 18 five-year averaging period but nothing shorter? 19 A. No, the .0083 which works, as I stated earlier, 20 in the case of Shark River slough for any five- 21 year period, the standard would be .009 for any 22 five-year period .009 should not be exceeded 23 and that would apply to a five-year period. If 24 one wanted to adopt a standard of .01 DEPOSITION OF WILLIAM W. WALKER, JR. 700 1 milligrams per liter and provide the same level 2 of protection, then one would have to use an 3 averaging period that is shorter than five 4 years. 5 Q. I get you. Thank you. 6 (Recess.) 7 Q. Doctor Walker, would a flow-weighted mean total 8 phosphorus concentration of .01 milligrams per 9 liter be consistent with historic inflow and 10 marsh data for the Park? 11 A. Again, you would have to specify what timeframe 12 you're talking about and which inflows and 13 which marsh stations you're talking about. 14 Q. You would be uncomfortable making such a 15 general statement as the one framed in my 16 question? 17 A. You would have to specify the timeframe and the 18 stations that you are referring to. 19 Q. What is the scientific rationale for the first 20 five year of sampling period of record for each 21 basin adopted in your Exhibit 28? 22 A. The concept of a five year running average was 23 something that was developed in the discussions 24 that we had with the District staff regarding DEPOSITION OF WILLIAM W. WALKER, JR. 701 1 ONRW standards. I believe it was originally 2 suggested by the District staff. 3 Q. So you just rely on the District for purposes 4 of establishing this parameter? 5 A. Well, I have analyzed the data and I'm familiar 6 with the variations that have occurred in 7 phosphorus over this timeframe and I believe 8 that a five-year basis for the standard is a 9 reasonable one. 10 Q. Have you looked at other periods of record to 11 see what kind of standards they would generate? 12 A. Yes. 13 Q. How many other periods of record have you 14 looked at? 15 A. Well, the original period of record that was 16 investigated under the ONRW was the period that 17 was specified, I believe, by some legislative 18 or regulatory body in Florida for the ONRW 19 standards. It was specified that we would use 20 data for the period from -- excuse me, data 21 through March of 1981, so the first period of 22 record that we examined used all the data that 23 we had up through that timeframe, and so that 24 would be roughly a three-year period of record DEPOSITION OF WILLIAM W. WALKER, JR. 702 1 from 1978 through 1981. 2 Q. Any other periods of record that you examined? 3 A. I believe some calculation were done early 4 using shorter time periods, one year, for 5 example, and the District had originally 6 proposed using a ten-year or using the entire 7 period of record for their monitoring program 8 as a basis for setting the standard. 9 Q. What would be the scientific rationale for 10 using the entire period of record? 11 A. Well, it wasn't our suggestion. That was the 12 suggestion by the -- made by the District. 13 Q. Do you recognize any scientific rationale for 14 using the entire period of record? 15 A. I believe their rationale was that was all the 16 data they had. It was the longest period of 17 record that was available with a consistent set 18 of laboratory techniques and procedures and so 19 forth. 20 Q. You accept the District's position on this 21 point in terms of establishing the standards? 22 A. Not in terms of using the ten-year or the 23 entire period of record for setting the 24 standards, no, I do not. DEPOSITION OF WILLIAM W. WALKER, JR. 703 1 Q. When you looked at the ONRW period, how did the 2 numbers derived from that period of record 3 compare to those in Table 1 of Exhibit 28? 4 A. They were similar with the exception of, excuse 5 me, with the exception of Taylor slough and 6 coastal basin because the monitoring periods 7 for Taylor slough and coastal basin did not 8 begin until 1983. We had no data for the ONRW 9 timeframe prior to March of 1981 for those 10 basins, so we only had data for Shark slough 11 prior to 1981. 12 Q. How did the values for Shark slough for the 13 ONRW time period compare to those reflected in 14 Exhibit 28? 15 A. Very similar. 16 Q. Were they higher or lower? 17 A. I think they were in terms of the upper 95 18 percent confidence limit for the flow-weighted 19 mean which is the value not to be exceeded in a 20 five-year period. They were essentially 21 identical. I don't recall. It was within .001 22 milligrams per liter. 23 Q. That number not to exceed that you just 24 referred to, what value is that in Exhibit 28? DEPOSITION OF WILLIAM W. WALKER, JR. 704 1 A. .0090. 2 Q. Could you please take this line in Table 1 for 3 Shark slough in Exhibit 28 and explain the 4 meaning of each entry in the categories across 5 that line? 6 A. You want me to explain each of the columns in 7 Table 1? 8 Q. Please. 9 A. The first column identifies the basin, Shark 10 slough, Taylor slough or coastal basin. The 11 second column identifies the number of samples 12 that were available during the baseline periods 13 for the first five years of monitoring for each 14 of those basins. 15 The next column identifies the 16 flow-weighted mean concentration that was 17 calculated for each of those basins during the 18 first five years of record. 19 The next column that's in 20 parentheses identifies the upper 95 percent 21 confidence limit for the flow-weighted mean 22 that was calculated for each of the basins 23 during the first five years of record. 24 Q. And that's the number you would employ as the DEPOSITION OF WILLIAM W. WALKER, JR. 705 1 standards for purposes of computing the five 2 years? 3 A. That number would be the number that would be 4 compared with the five year flow-weighted mean 5 concentration for any period in the future for 6 the purposes of determining compliance with 7 this standard. 8 Q. Please proceed. 9 A. The rest of this table has apparently been 10 obscured by someone's handwritten notes that 11 are not mine and I can't continue explaining 12 this table. 13 (Off the record discussion.) 14 (Confidential Memorandum was 15 marked Exhibit 33.) 16 Q. Doctor Walker, could you please identify the 17 document which has been marked as Exhibit 33? 18 A. This appears to be a partial copy of a 19 confidential memorandum that I prepared for the 20 Justice Department in February of 1990 21 regarding phosphorus standards for inflows to 22 Everglades National Park. 23 Q. And when you say it is partial, are you 24 referring to pages 8, 9 and 10 which are DEPOSITION OF WILLIAM W. WALKER, JR. 706 1 captioned for figures which do not appear here? 2 A. Upon reviewing this document, that's all that I 3 note at this moment as being missing. There 4 may be other things missing, but I don't see 5 anything at this moment. 6 Q. Did you author Exhibit 33 in its entirety? 7 A. Yes. 8 Q. And to whom did you direct this document? 9 A. This was directed to the US Department of 10 Justice, what was then the Land and Natural 11 Resources Division. 12 Q. Anyone else to whom you sent this document? 13 A. I don't recall whether I sent copies directly 14 to anyone else or not. 15 Q. Do you recall, did you receive comments back on 16 this document? 17 A. No, I don't recall receiving any comments. 18 Q. Do any more recent versions of this document 19 exist? 20 A. Not of this particular document, no. 21 Q. May I turn your attention to page 3 of Exhibit 22 33, the next to last paragraph and the first 23 sentence of the last paragraph read: 24 "Total phosphorus should not DEPOSITION OF WILLIAM W. WALKER, JR. 707 1 exceed .03 mg/liter in more than 2.8% (or 2.1%) 2 of the samples collected in each basin. 3 "This is just justified based on 4 inflow data for the ONRW time frame (regardless 5 of water management)." 6 Do you still agree with that 7 statement in Exhibit 33? 8 MR. HARRISON: Object to 9 characterization and foundation. Statement 10 without being read in contact with the whole 11 paragraph may be out of context. I am not 12 saying that it is, but there has been no 13 predicate laid that he agrees with that 14 statement as read into the record. 15 Q. If there is a context, the document will 16 further provide it. 17 A. This statement was offered as an alternative 18 expression of the standards as is identified in 19 Paragraph 7 on page 3 of Exhibit 33. This way 20 of expressing these standards targeted the .03 21 milligram per liter level and was a statistical 22 way of using that, expressing the standards 23 with respect to that frame of reference. 24 As is stated at the end of DEPOSITION OF WILLIAM W. WALKER, JR. 708 1 Paragraph 7, because of the underlying 2 lognormal frequency distribution of the data, 3 the alternative frequency standard, that is, 4 the frequency greater than .02 milligrams less 5 than 10 percent of the time would provide 6 essentially the same level of protection. 7 So this is just, this alternative 8 frame of reference of .03 milligrams per liter 9 is another way of expressing the standards that 10 I derived from the historic data and just 11 included in the report for consideration by the 12 Justice Department and by the Park staff. 13 Q. Was this standard as expressed in the 14 highlighted paragraph on page 3, was that 15 proposed to the District? 16 A. Not to my knowledge. 17 Q. Why not? 18 MR. HARRISON: Objection, counsel. 19 Proposed by whom? If you're talking about by 20 the Unified Federal Response to SWIM, that was 21 a Department of Justice mental processes that 22 you cannot go into even if the doctor does 23 know. Don't answer if -- 24 A. I can't answer that. DEPOSITION OF WILLIAM W. WALKER, JR. 709 1 Q. Because of Mr. Harrison's instruction? 2 MR. HARRISON: If it was expressed 3 to the District during the course of Doctor 4 Walker's ONRW meetings, which are not 5 privileged meetings that he held with the 6 District, I have no problem, counsel, but you 7 have not asked; and if in fact it was 8 expressed, whether or not it was expressed in 9 the Unified Federal Response which came from 10 the Department of Justice, that is clearly 11 attorney-client and I am not going to let him 12 go into why or why not the Department of 13 Justice chose which response to put in SWIM. 14 A. I don't recall having discussed this particular 15 way of expressing the standards with the 16 District in the context of the ONRW meetings. 17 Q. Is there any reason why this was not discussed 18 with the District in any context? 19 MR. HARRISON: Same objection, 20 counsel. 21 A. Well, from a statistical point of view, as I 22 pointed out earlier, it is because it partially 23 it is consistent and is partially redundant 24 with the frequency base standard that was DEPOSITION OF WILLIAM W. WALKER, JR. 710 1 already being discussed with the District. 2 Q. I'm just trying to find out, do you know of any 3 reason why this statement of the standard was 4 not put forth to the District? 5 MR. HARRISON: Objection, counsel. 6 Q. Do you know of any reason? 7 MR. HARRISON: Objection, counsel. 8 Q. Simply a yes-or-no question. 9 MR. HARRISON: You may answer 10 that. 11 A. No. 12 Q. I am going to address your attention up in the 13 paragraph numbered 6 starting at the end of the 14 7th line there is the sentence: 15 "The precise concentration level 16 at which biological damage occurs is unknown." 17 As you sit here today, do you believe that 18 statement to be true? 19 A. I believe that increasing phosphorus 20 concentration above an existing condition in 21 the Park or elsewhere has a potential of doing 22 biological damage. I do not know the precise 23 concentration level in terms of milligrams per 24 liter at which biological damage starts or DEPOSITION OF WILLIAM W. WALKER, JR. 711 1 stops. 2 Q. The next line reads: "Estimates range from .01 3 to .03 milligrams per liter," then referencing 4 Doctor Ron Jones, .01-.03 milligrams per 5 liter. What is the source of that information 6 concerning Doctor Jones? 7 A. That was a range that I recall Doctor Jones 8 having discussed in early meetings with Doctor 9 Jones and Park staff and the attorneys as being 10 a possible level at which impacts could be 11 quantified based upon his preliminary work. 12 Q. That was at more than one meeting? 13 A. Yes. 14 Q. Was this discussed at your February 8, 1990 15 meeting which is referenced on the first page 16 of Exhibit 33? 17 MR. HARRISON: Object to form and 18 I am going to object to going into specifics 19 about the February meeting, other than what is 20 reflected in this document. This document 21 obviously constitutes what the United States 22 could have claimed attorney-client and work 23 product privileges on. It has been turned 24 over. I will not say that the privilege is DEPOSITION OF WILLIAM W. WALKER, JR. 712 1 waived. I do not intend at this time to seek a 2 return of this document. 3 You may go into the scientific 4 bases that he knows for anything stated in this 5 document, his scientific opinion on anything in 6 this document, but I will not allow you to go 7 into the source of any discussions he may have 8 had with respect to why information was in here 9 other than his own scientific bases. 10 Strategy decision which may have 11 come up at the February 1989 meeting are 12 clearly privileged and are improper inquiry 13 even regarding this document. 14 Q. Doctor Walker, were Doctor Jones' estimates for 15 level at which biological damage occurred 16 discussed at the February 8 meeting? 17 A. I don't recall whether he addressed that topic 18 specifically at that meeting or not. 19 Q. Do you recall any discussions of this issue 20 subsequent to the time you prepared Exhibit 33? 21 A. Yes. 22 Q. What was said during those discussions? 23 MR. HARRISON: Objection, 24 counsel. I will let you go into what Doctor DEPOSITION OF WILLIAM W. WALKER, JR. 713 1 Walker knows or does not know that is in the 2 heads of the scientists. I would appreciate it 3 if you wouldn't characterize it as coming out 4 of that meeting. 5 If you want to ask Doctor Walker 6 what he knows about Ron Jones' current opinion 7 on those numbers or what he does not know, Ron 8 Jones is undergoing a deposition. I am sure 9 you already have that information. 10 MS. AHEARN: I think there has 11 clearly been a subject matter waiver as to 12 everything that's raised in this document and 13 the other documents which you've divulged. 14 MR. HARRISON: Counsel -- 15 MS. AHEARN: I am moving on. I 16 think our record is clear. Doctor Walker 17 received information from other scientists and 18 he reports it in his drafts; and when they talk 19 about it subsequently, you won't let me find 20 out what's being developed here. 21 MR. HARRISON: Counsel, I want to 22 clarify the record. You may go into the 23 scientific basis that is in Doctor Walker's 24 head, or if he knows the scientific bases in DEPOSITION OF WILLIAM W. WALKER, JR. 714 1 any of the other government experts' heads, 2 regarding what is in this document. But as 3 long as your questions do not go to why 4 something was presented, why it wasn't, why it 5 was transmitted to the District, what I 6 characterize are strategy decisions, I don't 7 think we'll have a problem and I don't think 8 you need to go into those. 9 Q. Doctor Walker, do you know if Doctor Jones 10 still estimates this precise concentration at 11 which biological damage occurs as .01 to .03 12 milligrams per liter? 13 A. To my knowledge Doctor Ron Jones has not 14 precisely quantified a value or a range at 15 which biological damage occurs. 16 Q. Do you know if Doctor Jones still estimates 17 that this range may be as high as .03 18 milligrams per liter? 19 A. To my knowledge Doctor Jones has not identified 20 a particular concentration level at which 21 biological damage begins to occur. 22 Q. Could you please turn to page 11 of Exhibit 33. 23 MR. HARRISON: Eleven, counsel? 24 Q. Eleven. Did you compute the values that appear DEPOSITION OF WILLIAM W. WALKER, JR. 715 1 on page 11? 2 A. Yes. 3 Q. The entries for S12+S333, do those entries 4 reflect the same structures that are included 5 for the basin Shark slough on Table 1 of 6 Exhibit 28? 7 A. I don't recall if they reflect precisely the 8 same structures. There is an adjustment for 9 the flow through S334 that is subtracted 10 through S333 and I don't recall whether the 11 values in Exhibit 33 are adjusted for that. 12 Okay. Upon further review on page 13 6, it is indicated that the flows to S333 are 14 adjusted for the flows to S334, so as far as I 15 can tell or I recollect, the values for 16 S12+S333 referred to on page 11 of Exhibit 33 17 are the same structures as the values referred 18 to as Shark slough in Table 1 of Exhibit 28. 19 Q. Did you use the same computational methods to 20 derive the flow-weighted means in each exhibit? 21 A. I did. 22 Q. Do the numbers in Table 1 of Exhibit 28 include 23 outliers? 24 A. The numbers in Table 1 on Exhibit 28 were DEPOSITION OF WILLIAM W. WALKER, JR. 716 1 computed using the methodology that we had 2 agreed upon with the district which -- and that 3 methodology included a systematic procedure for 4 screening for outliers, so Table 1, the numbers 5 in Table 1, do not include the effects of 6 outliers. 7 Q. How did the District express its agreement to 8 this procedure for excluding outliers? 9 A. In the course of one or more of the meetings 10 that we had with the District, there was a 11 verbal agreement and there was also a written 12 agreement, I recall, in the form of a letter 13 from Paul Trimble or excuse me, I don't recall 14 Paul's last name, but there was a letter from 15 District staff that indicated that we had 16 reached agreement regarding methodology for 17 handling outliers. 18 Q. Is that among the documents you produced, that 19 letter? 20 A. I believe so. 21 Q. Do you recall by name any other person from the 22 District who expressed agreement? 23 A. There was general agreement in the meeting, in 24 more than one meeting that we had with the DEPOSITION OF WILLIAM W. WALKER, JR. 717 1 District regarding ONRW on the methodology. 2 Q. Did anyone holding an executive office or 3 position with the District agree to this 4 methodology for outliers? 5 MR. HARRISON: Objection, 6 foundation. Not shown that Doctor Walker would 7 know what that title is and who would hold it. 8 A. Based upon my limited knowledge of the 9 District's organizational framework, I believe 10 that the highest ranking person who was present 11 at the meeting at which we agreed upon a method 12 and the concept of screening the data for 13 outliers was Thomas McVicar. 14 Q. And it is your testimony Mr. McVicar agreed 15 with this methodology? 16 A. To my recollection there was a general 17 agreement in the meeting. 18 Q. Did Mr. McVicar specifically agree? 19 A. He was present at the meeting and I don't 20 recall his voicing any objection to the 21 agreement. 22 Q. Let me make sure I understand this. With 23 reference to page 11 in Exhibit 33, do these 24 values, does this table reflect that if you DEPOSITION OF WILLIAM W. WALKER, JR. 718 1 employ the period of record for discharges to 2 Shark River slough before 1985 and include 3 outliers, the flow-weighted mean is 17.9 4 milligrams per liter -- excuse me, parts per 5 billion -- total phosphorus? 6 A. No. 7 Q. Utilizing the second period of record that is 8 reflected in this table on page 11 and 9 including outliers, what is the flow-weighted 10 mean for the S12s plus S333 structures? 11 A. The second period of record reflects the 12 measurements that were collected after 1985. 13 The flow-weighted mean concentration excluding 14 outliers was 16.9 parts per billion. The 15 flow-weighted mean concentration including 16 outliers was 17.9 parts per billion. 17 Q. Thank you for correcting me. Then when you use 18 "all" for period of recording, what period of 19 record is that? 20 A. That would be the entire period of record from 21 the District's database: December of 1977 22 through September of 1989. 23 Q. And using that period of record, what is the 24 flow-weighted mean for discharges to Shark DEPOSITION OF WILLIAM W. WALKER, JR. 719 1 River slough including outliers? 2 A. The flow-weighted mean including outliers is 3 13.1 parts per billion. 4 Q. Could I turn your attention back to the bottom 5 of page 3, Exhibit 33. 6 Did you test for lognormality of 7 the historic pre-1981 distribution of 8 flow-weighted mean total phosphorus 9 concentrations? 10 A. Yes. 11 Q. What test procedure did you employ? 12 A. I believe the test involved comparing the 13 observed and the predicted frequencies of total 14 phosphorus concentrations exceeding .01, .02 15 and .03 milligrams per liter where the observed 16 frequencies would be calculated directly from 17 -- by counting the number of measurements that 18 were above those values and the predicted 19 frequencies would be calculated from the log 20 mean and log standard deviation of the 21 measurements calculated during that same time 22 period and applying the lognormal frequency 23 distribution equations. 24 Q. Did you do that type of analysis for each of DEPOSITION OF WILLIAM W. WALKER, JR. 720 1 the structures? 2 A. I believe I applied that only to the Shark 3 River slough discharge. 4 Q. The total value or for each structure? 5 A. I don't recall whether -- I may have done 6 testing involving the individual structures as 7 well. I don't recall. 8 Q. Did you record the results of this testing 9 procedure in any way? 10 A. I recall doing it on a spreadsheet. I don't 11 recall whether I saved those results or not. 12 Q. When did you do this testing? 13 A. It would have been in January probably or early 14 winter of 1990. 15 Q. And you did this on your personal computer? 16 A. Correct. 17 Q. Let me make sure I've made this clear. Is it 18 that you can't remember if you did the testing 19 to each of the structures discharging to Shark 20 River slough? 21 MR. HARRISON: Objection. Asked 22 and answered. 23 A. I may have done it to each of the structures. 24 I don't recall. DEPOSITION OF WILLIAM W. WALKER, JR. 721 1 Q. Do you recall if you did the test to the 2 combined? 3 A. I did perform it for the combined discharge to 4 Shark slough. 5 Q. This was all done on the same occasion, the 6 same session at your computer? 7 A. Not necessarily. 8 Q. Is a flow-weighted mean of lognormal 9 concentrations distributed lognormally? 10 A. Not necessarily. 11 Q. When would the flow-weighted mean of lognormal 12 concentrations be distributed lognormally? 13 MR. HARRISON: Object to form. 14 You mean with respect to Shark 15 River slough, the same topic we're on or 16 generally? 17 A. I can't answer your question because I don't 18 know what the question refers to. You'd have 19 to be more specific. 20 Q. This isn't a statistical issue question 21 addressed on a conceptual level? 22 A. Not unless you define what you mean by the 23 distribution of the flow-weighted mean. 24 Q. The frequency distribution? DEPOSITION OF WILLIAM W. WALKER, JR. 722 1 A. I'm sorry, I still can't answer your question. 2 It is not specific enough. 3 (Water Quality Trends, Executive 4 Summary, was marked Exhibit 34.) 5 Q. Doctor Walker, can you please identify Exhibit 6 34? 7 A. This is entitled Water Quality Trends at 8 Inflows to Everglades National Park, Executive 9 Summary. 10 Q. Did you prepare this document? 11 A. No. 12 Q. Do you know who did? 13 A. I believe that Doctor Robert Harris prepared 14 this. 15 Q. Have you seen this document before today? 16 A. I believe I have seen it, yes. 17 Q. Do you know why Doctor Harris prepared this? 18 MR. HARRISON: Objection, if it 19 calls for a direction or instruction from the 20 Department of Justice. If you happen to know a 21 scientific basis on which Doctor Harris 22 provided it, you may answer. 23 A. I can't answer. 24 Q. Based on your counsel's instruction? DEPOSITION OF WILLIAM W. WALKER, JR. 723 1 A. Yes. 2 Q. Did Doctor Harris write any of what appears in 3 Exhibit 7? 4 A. Are you asking me if Doctor Harris coauthored 5 Exhibit 7? 6 Q. Did Doctor Harris coauthor Exhibit 7? 7 A. No. 8 Q. What do you mean by coauthor? 9 A. I mean whether Doctor Harris wrote any of 10 Exhibit 7. 11 Q. So Doctor Harris did not write any of Exhibit 12 7? 13 A. That's correct. 14 MR. HARRISON: Counsel, I do 15 suggest before you go into the next exhibit we 16 have lunch. 17 MS. AHEARN: Why don't we do a 18 simple identification, Mr. Harrison. 19 (Declaration of September 1990 was 20 marked Exhibit 35.) 21 Q. Doctor Walker, can you please identify Exhibit 22 35? 23 A. This is a declaration that I prepared in 24 September of 1990 that describes my research DEPOSITION OF WILLIAM W. WALKER, JR. 724 1 results regarding the South Florida problem. 2 Q. Did Doctor Harris write any of this? 3 MR. HARRISON: Object to form, 4 counsel. I think "any" in this situation has 5 to be very tightly defined. 6 A. Doctor Harris prepared an early draft of this. 7 Q. Did you prepare Exhibit 35 in its entirety? 8 MR. HARRISON: Object to form. 9 A. The words and opinions expressed in Exhibit 35 10 are my own. 11 Q. Who actually wrote Exhibit 35? 12 A. I wrote Exhibit 35. 13 Q. How many prior draft declarations were there? 14 MR. HARRISON: Objection, 15 counsel. 16 A. I don't recall. 17 MR. HARRISON: This document was 18 written specifically for a legal purpose and 19 you've already established it was for the 20 summary judgment I think last week. I am not 21 going to go into the preparation of the 22 document. 23 MS. AHEARN: Mr. Harrison, the 24 government has made multiple entries on DEPOSITION OF WILLIAM W. WALKER, JR. 725 1 privileged lists attributing this document and 2 draft to different authors. I am simply trying 3 to sort out how many documents the government 4 has here. 5 MR. HARRISON: I object to your 6 characterization. I think that you're probably 7 misinterpreting something. 8 MS. AHEARN: If I am, this is my 9 chance to straighten it out. 10 MR. HARRISON: You can ask him how 11 many. I won't let you go into the substance or 12 who had what input to any of the drafts. You 13 have this declaration before you and you have 14 had Doctor Walker's testimony that it 15 represents his words and his opinions. You can 16 ask him how many drafts that he personally 17 wrote. 18 Q. How many drafts authored by persons other than 19 yourself of your declaration are you aware of? 20 MR. HARRISON: Objection, 21 counsel. I will not permit him to go into what 22 other assistance or technical consultants may 23 have had, whether they have had input or not 24 into this. You're looking at the final DEPOSITION OF WILLIAM W. WALKER, JR. 726 1 product. Doctor Walker already testified it 2 was his own, and if you want to go into how 3 many prior drafts Doctor Walker prepared, feel 4 free to do so. 5 MS. AHEARN: You won't permit me 6 to determine what other documents Doctor Walker 7 was able to consult incorporating this final 8 product? 9 MR. HARRISON: As far as I am 10 concerned, counsel, go ahead. I don't want to 11 fight over this particular area. I'll let you 12 ask. I won't let you go into the substance, 13 but you can go over how many drafts he has 14 seen. 15 A. I will have to ask for the question again, 16 please. 17 Q. How many drafts of the declaration of Doctor 18 William Walker, other than those drafted by 19 yourself, are you aware of? 20 A. I recall one. 21 Q. And who was the author of that one draft? 22 MR. HARRISON: Object to form. I 23 think draft also needs to be defined, counsel. 24 Talking about an entire document or an outline DEPOSITION OF WILLIAM W. WALKER, JR. 727 1 or whatever? I think it needs to be defined. 2 A. There was one document that was prepared by 3 Doctor Robert Harris that essentially 4 summarized important points in my report on 5 trend analysis and presented those points, 6 extracted those points from my report and 7 provided an example of the format and the 8 presentation that would be appropriate for a 9 declaration. 10 Q. Did you utilize this draft by Doctor Harris in 11 preparing the final declaration? 12 A. The final declaration is my own words and my 13 own opinions. 14 Q. Did you utilize Doctor Harris' draft? 15 A. I read Doctor Harris' draft to provide some 16 frame of reference, but I subsequently prepared 17 my own version that expresses my own words and 18 my own opinions. 19 MS. AHEARN: Thank you. Let's 20 have lunch. 21 (Luncheon recess.) 22 AFTERNOON SESSION 23 MR. HARRISON: Counsel, prior to 24 lunch I told you I'd try to get ahold of Lloyd DEPOSITION OF WILLIAM W. WALKER, JR. 728 1 Pike, the attorney for the Corps of Engineers 2 in Jacksonville, that I had heard that Col. 3 Malson was not going to be made available on 4 February 19th and the District does have a 5 notice for the 19th and the 20th, as I 6 understand, for his deposition based on dates 7 that he was available; but Senator Graham has 8 requested his presence and the Chief of 9 Engineers, and I believe the Division 10 Commander, General Hatch and General Sabke, I 11 think, to be in Washington,DC and it is my 12 understanding that that is for the 20th. 13 I do not know whether or not that 14 was precipitated because of the case or the new 15 governor or what precipitated it, but Colonel 16 Malson is being requested to come to Washington 17 for Senator Graham. 18 MS. AHEARN: You don't know if it 19 precipitated by the proceeding involving South 20 Florida? 21 MR. HARRISON: It would be pure 22 speculation, but I would be surprised if the 23 discussions with the Senator do not center on 24 the issues in the Everglades lawsuit. If we DEPOSITION OF WILLIAM W. WALKER, JR. 729 1 find out that has been communicated to Colonel 2 Malson or to anybody with the government as to 3 what the reason for that briefing was, I'll let 4 you know, but I am going to have to ask you to 5 withdraw that notice, or obviously I'd have to 6 file a motion for protective order. There is 7 nothing I can do about it. 8 I did get Mr. Pike to reaffirm the 9 dates that are on Colonel Malson's calendar now 10 that he has blocked off for Lloyd Pike for 11 deposition, and that's April 22nd through April 12 24th or the entire week of April 29th through 13 May third, and let me know if you want me to 14 file a motion for protective order on the 19th 15 or how we should proceed. 16 I am hoping that I won't have to 17 do that since I don't think that there is 18 anything any of us can do about this particular 19 meeting. I would like -- I think it would be 20 most judicious for the District to notice him 21 right now before his calendar fills up for one 22 of those two timeframes in April, but that's 23 simply my suggestion. 24 MS. AHEARN: I know, Rick. I DEPOSITION OF WILLIAM W. WALKER, JR. 730 1 thank you. Unfortunately, I followed the 2 advice once before and it didn't work for you, 3 and I think I thank you and I will convey the 4 information to those who need to know. 5 MR. HARRISON: Will you let me 6 know whether or not you're going to continue to 7 seek him for February 19th so I might file a 8 protective order? I've held back because the 9 reasons for that briefing are totally beyond 10 our control, and I think if you delve back into 11 your side, you may well find that some of the 12 reasons were precipitated by certain elements 13 of the state. 14 I certainly don't object to him 15 going up there, but I can't be torn between 16 Senator Graham and this case and we simply 17 can't produce him. 18 MS. AHEARN: I have just learned 19 about this new conflict for Colonel Malson and 20 I will convey the information to the people who 21 will decide what action to take and I am sure 22 they'll let you know. 23 MR. HARRISON: Thanks. 24 (Off the record discussion.) DEPOSITION OF WILLIAM W. WALKER, JR. 731 1 DIRECT EXAMINATION - CONTINUED - BY MS. AHEARN 2 Q. I'd like the record to reflect that we have 3 provided an unmarked version of Table 1, the 4 second page to Exhibit 28, and I've asked 5 Doctor Walker to insert that unmarked page into 6 the deposition copy of Exhibit 28. 7 Doctor Walker, does this Table 1 8 now reflect the table as you prepared it? 9 A. I believe so. 10 Q. I'd like to turn your attention to Exhibit 17. 11 Now, Doctor Walker, we've 12 discussed this report briefly on Friday and the 13 application of the Seasonal Kendall test to the 14 District's surface water quality data for 15 inflows to the Park for 1977 through 1989. 16 Have you applied the Seasonal 17 Kendall test to surface water quality data for 18 Park inflows that are collected by any agency 19 other than the District? 20 A. No, I have not. 21 Q. In the second paragraph of the abstract, you 22 report that trend magnitudes range from four 23 percent a year at S12D to 21 percent per year 24 at S332. Does this mean that in water year DEPOSITION OF WILLIAM W. WALKER, JR. 732 1 1990 total phosphorus levels at S12D would be 2 four percent higher than levels from water year 3 1989? 4 A. No, not necessarily. 5 Q. Does the trend analysis reflected in Exhibit 17 6 make any prediction of water quality for years 7 subsequent to the period of record actually 8 employed in the test? 9 A. No. As is stated in the conclusions of the 10 report, on what I believe is page 44, although 11 I can't tell because the page numbers have been 12 somehow missed in the copying process, but on 13 the last page of the text in my report in the 14 middle of the paragraph it states the trends 15 detected for the 1977 to 1989 period cannot be 16 extrapolated into the past or into the future. 17 Q. I believe you told us that you -- correct me if 18 I'm wrong -- that you've acquired the water 19 quality data for the year following the year of 20 record employed in Exhibit 17; is that correct? 21 A. I do not have the data in a raw form. 22 The only information which I have, 23 which I believe was entered into an exhibit 24 already, was a flow-weighted mean concentration DEPOSITION OF WILLIAM W. WALKER, JR. 733 1 calculated by the District for, I believe, it 2 was CS12's and S333 for water year 1990. 3 Q. Is that summary data adequate for you to test 4 whether any trend is evident for water year 5 1990? 6 A. No. 7 Q. Do you have any plans to perform analyses to 8 determine if trends exist subsequent to the 9 period of record analyzed in Exhibit 17? 10 A. I do not have any definite plans to do that. 11 Q. Now, you explained how the Kendall test was 12 applied to three time series. One is all 13 concentration data, the second concentration 14 data collected on days with appreciable 15 discharge; and thirdly, concentrations 16 suggested for hydrologic factors including an 17 antecedent rainfall and water surface 18 elevation. 19 Did you perform analyses for all 20 twenty water quality components for each of 21 these three series? 22 A. Yes. 23 Q. How? 24 A. Yes. DEPOSITION OF WILLIAM W. WALKER, JR. 734 1 Q. How did you select the twenty water components 2 to analyze? 3 A. These water quality components reflect those 4 that were consistently sampled by the District 5 during this period and which were provided to 6 the Park staff for the purpose of setting ONRW 7 baseline standards. 8 Q. So was it largely data availability that 9 directed the selection of components? 10 MR. HARRISON: Object to form. 11 A. It was availability of consistently collected 12 data. 13 Q. And by consistently collected, you mean in 14 terms of the field methodology employed? 15 A. I mean in terms of sampling frequency and 16 period of record. 17 Q. You also looked at nutrient transport rates. 18 Is that what is also commonly referred to as 19 loading? 20 A. Yes. 21 Q. And in your abstract you report that an 22 increasing trend in phosphorus transport rate 23 is indicated for Taylor slough. Was any trend 24 indicated for the other structures or basins? DEPOSITION OF WILLIAM W. WALKER, JR. 735 1 A. Yes. 2 Q. Were the other trends detected decreasing 3 trends? 4 A. There were no decreasing trends in total 5 phosphorus that exceeded or that had 6 probabilities less than .1. 7 Q. And by total phosphorus you mean the total 8 phosphorus transport rate? 9 A. Total phosphorus transport rate, correct. 10 Q. Any trends with probabilities using a standard 11 of .05? 12 MR. HARRISON: Of decreasing 13 trends, counsel? 14 Q. Yes. 15 A. Would you fully repeat your question? I'm 16 sorry. We started talking about trends and now 17 we're talking about decreasing trends. Would 18 you please repeat your question fully? 19 Q. I asked you, you're reporting the increasing 20 trend at Taylor slough, and I asked you if 21 there were trends at the other basins and 22 stations, and you said yes, so I'm trying to 23 clarify that those were decreasing trends, 24 inferring from reporting an increasing trend DEPOSITION OF WILLIAM W. WALKER, JR. 736 1 for Taylor slough that those other trends were 2 decreasing. 3 A. Well, that inference is incorrect. That's your 4 assumption. Those are not my results. 5 Q. Then what I am asking for is your results, if 6 you can point me to, if there were trends at 7 the other structures for phosphorus transport 8 rate, what were they? 9 A. In the case of Shark River slough, when I 10 examined all of the data including the days on 11 which the structure was sampled when there was 12 no flow, and during those days the calculation 13 of loading would be completely insensitive to 14 the concentration, that first series of tests 15 indicated there was no trend at Shark slough. 16 Q. Doctor Walker -- I'm sorry. 17 A. When I examined the data collected on days when 18 there was flow passing through the structure, 19 there was an increasing trend indicated in 20 Shark slough, and that trend magnitude was 9.9 21 percent per year. 22 Q. And Doctor Walker, you're answering in 23 reference to a table in Exhibit 17. Could you 24 please point me to the page where I'll find DEPOSITION OF WILLIAM W. WALKER, JR. 737 1 this? 2 A. The table I am referring to is Table 6 and I 3 believe, although I can't -- just from looking 4 at the adjacent pages -- because the page 5 number has been omitted from the copy and it 6 looks like it is on page 41. 7 Q. Doctor Walker, during our next break I will see 8 if I can find a copy that wasn't cropped at the 9 top if that will make it easier to see the 10 pages for you. 11 On the first page of the text of 12 Exhibit 17 in the first paragraph, 8th line, 13 you report that in retaining some 94% of the 14 total phosphorus load discharged from the 15 sources you discuss above, between 1979 and 16 1988, the WCAs had been utilized as water 17 quality buffers. 18 Could you characterize the 19 phosphorus uptake efficiency reflected by that 20 94 percent figure? 21 MR. BURGESS: Sorry, counsel, 22 which page are you on? 23 MS. AHEARN: First on 24 introduction. DEPOSITION OF WILLIAM W. WALKER, JR. 738 1 MR. HARRISON: I'm sorry. The 2 page labeled introduction, counselor? 3 MS. AHEARN: Yes. It doesn't have 4 a page number on it. 5 MR. HARRISON: When you said first 6 page of the text, I was looking at the 7 abstract. The very next page after the cover 8 sheet. 9 A. That sentence refers to the fact that based 10 upon the District's calculation of nutrient 11 loads into and out of the water conservation 12 areas, 94 percent of the total phosphorus load 13 that is discharged into the water conservation 14 areas is retained within the water conservation 15 areas. 16 Q. Have you performed any verification of those 17 calculations? 18 A. I have not recalculated the total phosphorus 19 budgets for each structure leading into and 20 leading out of the water conservation areas as 21 performed by the District. 22 Q. Do you have any information to suggest that 23 this 94 percent uptake figure is wrong? 24 A. Assuming that the District's calculation of DEPOSITION OF WILLIAM W. WALKER, JR. 739 1 total phosphorus loads from the various 2 structures is correct, and assuming that the 3 District's assumptions regarding the loading of 4 phosphorus in rainfall and atmospheric load is 5 correct, one would calculate the removal 6 efficiency of 94 percent for this entire period 7 of record, but I have not verified or tested 8 those specific assumptions myself. 9 Q. Those are assumptions which you have employed 10 in any work you've performed on the South 11 Florida matter? 12 MR. HARRISON: Objection. 13 Employed -- he obviously employed them here. 14 You mean in his reports, counsel? 15 A. Yes. 16 Q. Can you characterize a 94 percent phosphorus 17 uptake rate relative to uptake rates that have 18 been studied throughout North America? 19 MR. HARRISON: Objection, 20 foundation. 21 A. I can characterize it generally in relation to 22 uptake rates that were present in the 23 particular water resource systems that I've 24 studied. DEPOSITION OF WILLIAM W. WALKER, JR. 740 1 Q. Okay. How would you characterize it relative 2 to the systems you've studied? 3 A. It is relatively high, unusually high for -- in 4 relation to the systems that I've studied. 5 Q. And when we talk about the systems that you've 6 studied, are those the ones reflected in your 7 project summaries, for example, Exhibit 2? 8 A. That's correct. 9 Q. A little bit far farther down in the 10 introduction with respect to WC and nutrient 11 trap efficiency, you state the longevity of 12 this is unknown. 13 Do you believe that to be true 14 today, that the longevity of WC and nutrient 15 trap efficiency is unknown? 16 A. Generally, yes. 17 Q. Could it be indefinite? 18 A. Could what be indefinite? 19 Q. The nutrient trap efficiency longevity? 20 MR. HARRISON: Objection, counsel, 21 based on what assumptions, based on the 22 assumption that the nutrients continue as they 23 are now? There is a million assumptions that 24 go into that assumption. DEPOSITION OF WILLIAM W. WALKER, JR. 741 1 MS. AHEARN: I don't think I want 2 to ask him the question a million different 3 ways. 4 MR. HARRISON: No, but I think in 5 all fairness you should characterize what 6 you're asking in assumptions. 7 A. The question of whether a trap efficiency of 8 this magnitude could be maintained indefinitely 9 must be considered in relation to the 10 literature studies that the study reported in 11 the literature on wetlands where people have 12 quantified inputs and outputs over periods of 13 time and people have documented declines in 14 phosphorus trapping efficiency over time. 15 Q. Would you be able to apply that information 16 from the literature to the water conservation 17 areas of the Everglades? 18 A. One would have to consider that information in 19 evaluating the longevity of the phosphorus 20 trapping efficiency in the Everglades. 21 Q. The areas under consideration in the 22 literature, are they comparable to the water 23 conservation areas? 24 A. Comparable in what way? DEPOSITION OF WILLIAM W. WALKER, JR. 742 1 Q. In size and configuration, in climate, in 2 macrophyte, soil composition? 3 A. Each system has its own unique characteristics. 4 Q. Are you aware of any studies of the longevity 5 of nutrient uptake efficiency in any wetland 6 comparable to Water Conservation Area 3? 7 MR. HARRISON: Object to form. 8 A. Studies would be located perhaps in different 9 regions and of different sizes, but there are 10 studies that show that wetlands used for waste 11 water treatment such as these water 12 conservation areas have been used, in other 13 words, being used as water quality buffers have 14 a tendency to lose that buffer capacity over 15 time. 16 Q. Could you identify one wetland that's been 17 studied in the literature that you would point 18 to as most comparable to the Water Conservation 19 Area 3A for these purposes? 20 MR. HARRISON: Object to form. 21 Q. The best example that we could employ? 22 MR. HARRISON: My objection, 23 counsel, is just over the definition of the 24 words comparable and best. DEPOSITION OF WILLIAM W. WALKER, JR. 743 1 A. I've stated that there were probably no systems 2 to my knowledge that are comparable in terms of 3 size or location. All that I stated is that 4 there have been waste water treatment system 5 wetlands that are used for waste water 6 treatment that have been documented to lose 7 their phosphorus trap efficiency over time. 8 Such studies are referenced in an article that 9 was prepared by -- I believe it was Nichols 10 published in 1983 in the Journal of Water 11 Pollution Control Federation and in a 12 subsequent article by Richard Nichols that was 13 in a book published by Van Nostrand in 1984 or 14 1985. I don't recall the particular names of 15 the wetlands. 16 Q. A bit farther down in the introduction you make 17 reference to Belanger, et al, 1989. What was 18 that study about? 19 A. That study was a review of eutrophication 20 problems in the water conservation areas and 21 the Everglades as prepared by Belanger, Scheidt 22 and Platko which appeared in Lake and Reservoir 23 Management in 1989. 24 Q. This a review paper as opposed to original DEPOSITION OF WILLIAM W. WALKER, JR. 744 1 research? 2 A. I believe it was a mixture of both. I don't 3 recall exactly. 4 Q. Have you relied on any original data reported 5 by Belanger at all? 6 A. I reviewed that particular paper by Belanger, 7 et al. 8 Q. Do you know if there is any primary data or 9 original research in the next reference Lake 10 Okeechobee Technical Advisory Council, 1990? 11 A. I am not certain whether that contains any 12 original research. 13 Q. Do you know of any original research which 14 supports the statement: "Increases in 15 phosphorus concentration have been associated 16 with changes in plant communities and with 17 declining water quality in portions of the 18 WCAs"? 19 A. The research cited in Belanger, et al and the 20 research cited in the Lake Okeechobee Advisory 21 Council report. 22 Q. Have you reviewed that research that is cited 23 in these two sources? 24 A. I have reviewed these two sources of DEPOSITION OF WILLIAM W. WALKER, JR. 745 1 information, these two publications. 2 Q. The next to last sentence on this page reads: 3 "Tests for trends in nutrient transport rate 4 (kg/day) and ENP basin are also conducted." 5 When did you conduct those tests? 6 A. Late summer, early fall of 1990. 7 Q. Had you conducted them prior to preparing the 8 first draft of your water quality trends 9 report? 10 A. No. 11 Q. Why wasn't this included in the first set of 12 analyses you did before preparing the draft 13 report? 14 A. The first report focused on trends in 15 concentration and identified such trends, and 16 with the additional timeframe permitted between 17 the draft and the final, and I believe there 18 was also some feedback from one or more of the 19 reviewers of that first draft, that information 20 on nutrient transport rates would be useful, so 21 it was partly a matter of additional time at 22 the suggestion of the reviewer that the 23 analysis of nutrient transport was undertaken. 24 Q. Did you have a deadline for that first draft? DEPOSITION OF WILLIAM W. WALKER, JR. 746 1 MR. HARRISON: I am going to 2 object to your question, counsel. If there is 3 some sort of a deadline, if it is his own 4 deadline, fine. If it is some deadline that 5 the Department of Justice gave him for purposes 6 of the information, it is beyond the scope of 7 anything reasonably calculated to lead to the 8 discovery of admissible evidence. 9 MS. AHEARN: I merely asked if 10 there existed a deadline. I certainly think it 11 is discoverable. It'd help me whether only so 12 much research could be done by Doctor Walker 13 before his first draft or this is his first 14 submission and he considered that a completed 15 package. 16 Q. My question, again, is did you have a deadline? 17 MR. HARRISON: I will permit him 18 to answer. 19 Q. Thank you. 20 A. I had committed to completing a draft of the 21 trend analysis report by June or July, I don't 22 recall exactly the date, in 1990. I had 23 indicated to the Justice Department that the 24 draft would be available by that time. There DEPOSITION OF WILLIAM W. WALKER, JR. 747 1 was not a specific hard line deadline, but 2 there was a commitment on my part to complete 3 the report by early summer 1990. 4 Q. In your study plan or outline for that draft, 5 was there included the test for trends and 6 nutrient transport rates? 7 A. I don't recall. 8 Q. On page 3 under the paragraph data set, the 9 last sentence there, you state the District's 10 monitoring program provides a consistent water 11 quality database which is not subject to 12 various difficulties. 13 We talked about some of these 14 before generically. We haven't talked about 15 varying interval samples. How invariable are 16 the sampling intervals reflected in the 17 District monitoring database? 18 A. The District's monitoring program during this 19 period was generally biweekly, that is, a 20 sample was collected from each structure every 21 two weeks. There may have at certain times 22 been deviations from that general overall 23 program, but I would characterize the overall 24 program as being a biweekly sampling program. DEPOSITION OF WILLIAM W. WALKER, JR. 748 1 Q. Did you check the database to determine on how 2 many occasions and to what extent the biweekly 3 schedule was not followed. 4 MR. HARRISON: I will object to 5 form, counsel, in that the predicate of your 6 sentence assumes that it was not followed at 7 some point in time, so you're assuming a fact 8 that is not in evidence. 9 A. As I stated, overall monitoring frequency was 10 biweekly. There may have been periods of 11 deviations from that. The one period that I 12 recall was in 1988 at the S12s when the 13 sampling frequency may have been somewhat more 14 frequently than biweekly. The Seasonal Kendall 15 test that I've employed here is applicable to 16 programs that have modest levels of variations 17 in sampling frequency. 18 In other words it is not 19 absolutely necessary that a sample be collected 20 every two weeks in order to perform a valid 21 trend analysis. 22 Q. Wasn't the Seasonal Kendall test designed for 23 monthly water quality samples? 24 A. Not specifically for monthly. DEPOSITION OF WILLIAM W. WALKER, JR. 749 1 Q. In its original creation, wasn't the Seasonal 2 Kendall test developed and tested using monthly 3 data? 4 A. I believe the first reference to the Seasonal 5 Kendall test contains reference to monthly data 6 sets, but there is nothing special about the 7 Seasonal Kendall test that requires monthly 8 data. 9 Q. Do you know if the developers of the Seasonal 10 Kendall test, Hirsch, Slack and Smith, have 11 explored whether use of their test is 12 appropriate when you have multiple observations 13 in each season as defined in the test? 14 A. Yes, I believe that subject area is addressed 15 in their papers. 16 Q. Could you point me to which paper it is where 17 they address this? 18 A. It may be in both papers or it may be in one. 19 I don't recall. 20 Q. But there are two papers that you have in mind; 21 is that correct? 22 A. There were two papers that were published in 23 Water Resources Research that describe Seasonal 24 Kendall test as developed by Hirsch, et al. DEPOSITION OF WILLIAM W. WALKER, JR. 750 1 Q. And did you produce copies of those two 2 articles with the materials produced for this 3 deposition? 4 A. Yes. 5 (Articles were marked Exhibits 36 6 and 37.) 7 Q. Doctor Walker, could I please ask you to 8 identify first Exhibit 36 and secondly Exhibit 9 37. 10 A. Exhibit 36 is a copy of a paper entitled 11 Techniques of Trend Analysis for Monthly Water 12 Quality Data by Hirsch, Slack and Smith from 13 Water Resources Research, February 1982. 14 Exhibit 37 is a copy of a paper 15 entitled a Nonparametric Test -- sorry, 16 Nonparametric Trend Test for Seasonal Data with 17 Serial Dependence by Hirsch and Slack published 18 in Water Resources Research in June of 1984. 19 Q. Are these the two articles to which you just 20 referred? 21 A. Yes. 22 Q. Could I ask you to turn to page 110. It is the 23 pagination of the original article in Exhibit 24 36. Could you please read the sentence that DEPOSITION OF WILLIAM W. WALKER, JR. 751 1 starts at the bottom of that page. 2 Could you read it aloud for the 3 record, please? 4 Actually, Doctor Walker, if I 5 could have you read the first sentence that 6 starts in that paragraph and then the sentence 7 that starts the bottom of the page. 8 MR. HARRISON: You mean the very 9 bottom like over in the right-hand column, 10 counsel? There is one that starts at the 11 bottom of the page on the left-hand side. 12 MS. AHEARN: On the right-hand 13 side of the page. 14 MR. HARRISON: I'd also like the 15 record to reflect that this document is in 16 excess of 10 to 12 pages and Doctor Walker has 17 not sat here and recently reviewed it. 18 A. If you would kindly repeat your request, what 19 is it that you want me to read? 20 Q. On page 110, the last paragraph on the 21 right-hand side, the first sentence in that 22 paragraph. 23 A. "A possible modification of the Seasonal 24 Kendall test would involve using multiple DEPOSITION OF WILLIAM W. WALKER, JR. 752 1 observations for each month rather than 2 limiting the time series to one observation a 3 month." 4 Q. Then the sentence that begins at the very 5 bottom of the page. 6 A. "We have not explored the use of this modified 7 test and it is not clear whether it would be 8 preferable to use all of the available data or 9 to take, say, the medians of the multiple 10 observations in each of the months and use them 11 in the former version of the test." 12 Q. Using the District's biweekly data for Park 13 inflows, am I correct that for the various 14 seasons you look at in the Seasonal Kendall 15 test, you may have seasons, I guess, with no 16 data value for some reason, with one value, 17 with two values or with as many as three? 18 MR. HARRISON: Object to form. 19 Compound. 20 A. Depending upon the station and the water 21 quality component that we're analyzing, there 22 would be some variation in the number of 23 observations from -- for any given season and 24 year. DEPOSITION OF WILLIAM W. WALKER, JR. 753 1 Q. Am I correct that before running the Seasonal 2 Kendall test you need to adjust your database 3 that you're operating with one value per 4 season? 5 A. Within each year and season, that's correct,, 6 there could be a maximum of one observation 7 that is actually used in the test. 8 Q. When you have two values for any given season, 9 how do you select the one value to use in the 10 test? 11 MR. HARRISON: Object to form. 12 Are we speaking generally now or of the 13 specific problem that Doctor Walker has worked 14 on? 15 Q. Is there a general convention on that, Dr. 16 Walker? 17 A. As stated in Exhibit 17, when applying the test 18 on a monthly basis, I have taken the median 19 concentration measured within each month and 20 year. 21 Q. You followed that approach when you've had 22 three data values in a given season? 23 A. That's correct. 24 Q. Is that a generally accepted approach for DEPOSITION OF WILLIAM W. WALKER, JR. 754 1 producing the one per season value needed in 2 the Seasonal Kendall test? 3 MR. HARRISON: Object to form. 4 A. That is the approach that was used in the 5 computer program developed by Crawford, et al 6 that is referenced in Exhibit No. 17, that is, 7 the approach that was used by the USGS in the 8 computer program that they developed for 9 applying the Seasonal Kendall test. 10 Q. Does Crawford, et al discuss the rationale for 11 using the median of multiple seasonal values? 12 A. I recall a statement that that is in fact what 13 they -- that is the algorithm they have 14 incorporated into the program. I don't recall 15 whether they discussed the rationale. 16 Q. Are you aware of any discussions in the 17 scientific literature of the rationale for 18 using a median when you have multiple values in 19 a season? 20 A. I believe this topic is also discussed in an 21 article by Van Belle and Hughes. 22 Q. Is that among the papers that you produced? 23 A. Yes. 24 Q. What is the rationale they give for using the DEPOSITION OF WILLIAM W. WALKER, JR. 755 1 median among multiple seasonal values? 2 A. They discuss in general the options that are 3 available for summarizing -- for dealing with 4 multiple observations within a given season. 5 Q. What options are available? 6 A. I don't recall exactly what their article 7 encompassed. 8 Q. Did you explore any options in your Seasonal 9 Kendall test other than the median approach? 10 A. The alternative that I recall investigating was 11 essentially adjusting the seasonal length and 12 that is reported in Exhibit 17. 13 Q. So you didn't try taking the mean of all 14 multiple values, correct? 15 MR. HARRISON: Mean as opposed to 16 median, counsel? 17 Q. Mean as opposed to median. 18 A. No, I believe that that option is coded into 19 the software that I developed for performing 20 the test, but I don't recall ever actually 21 applying it because I believe that the median 22 is more appropriate statistic than the mean 23 when we're dealing with a nonparametric 24 procedure such as this. DEPOSITION OF WILLIAM W. WALKER, JR. 756 1 Q. If you had three data values in a season of 2 one, two and ten, what is the value you would 3 employ for the Seasonal Kendall test? 4 A. Having adopted the algorithm of taking the 5 median, I would employ, I would use the middle 6 value. 7 Q. Two? 8 A. If the values were one, two and ten, I believe 9 you stated, then I would have used two. 10 Q. Does that adjustment of the database produce a 11 sample of independent and identically 12 distributed random variables? 13 MR. HARRISON: Object to form. 14 A. When there is more than one observation per 15 season, the summary value that you pick for a 16 particular season, the more observations you 17 have, the less variability would be associated 18 with that particular seasonal summary. 19 Q. If that answered my question, Doctor Walker, I 20 apologize, but I have been studying this a 21 lot. If that answered my question, I am afraid 22 it went right by me. 23 By adjusting the raw data to 24 select one value per season using the median of DEPOSITION OF WILLIAM W. WALKER, JR. 757 1 multiple seasonal values, do you create a 2 sample of independent and identically 3 distributed random variables? 4 MR. HARRISON: Object to form. 5 A. That procedure may introduce some variations 6 from one season and year to another in the 7 variance and the tradeoff between taking 8 medians within each season or using other 9 procedures is discussed, as I mentioned, I 10 believe in the article Van Belle and Hughes. 11 Q. Do they render an opinion on whether the 12 resulting sample -- 13 A. They discuss the alternatives. They discuss 14 the various pros and cons of the different ways 15 of handling this. 16 Q. And what way of handling this are they most pro 17 on? 18 A. I don't believe that they come out pro or con 19 on a particular method. They discuss the 20 options. I have used the option that was 21 employed in the computer program that was 22 written by the US Geologic Survey. 23 Q. Is the null hypothesis of the Seasonal Kendall 24 test predicated to assume that the data are a DEPOSITION OF WILLIAM W. WALKER, JR. 758 1 sample of independent and identically 2 distributed random variables? 3 A. That's an assumption of the test, that's 4 correct. 5 Q. We talked about using the mean or the median 6 when we have multiple seasonal values. Is it 7 also possible to use flow-adjusted 8 concentrations? 9 A. What do you mean by flow-adjusted 10 concentration? 11 Q. Flow-weighted concentrations. Excuse me, if I 12 used the wrong terminology. 13 A. As I mentioned earlier, the Seasonal Kendall 14 test is a nonparametric procedure and 15 calculating a flow-weighted mean within each 16 season would be possible, but it would 17 generally not be consistent with a 18 nonparametric approach. 19 Q. When you composite discharges through the S12 20 structures, you flow weight across those 21 structures, do you not? 22 A. To calculate a composite value for each 23 sampling date, correct. 24 Q. Was that calculation consistent with the DEPOSITION OF WILLIAM W. WALKER, JR. 759 1 nonparametric method? 2 A. There was no alternative in that case if one 3 wants to base the test on the time series of 4 flow-weighted mean concentrations across the 5 total S12 cross-section, there was no 6 alternative but to use the flow-weighted mean 7 across the individual stations on each date. 8 Q. So there is no option but to use a particular 9 step that's not consistent with nonparametric 10 methodology? 11 MR. HARRISON: Object to form. 12 Characterization. 13 A. I cannot think of a nonparametric way of 14 representing the flow-weighted cross-section 15 for the S12s or Shark slough other than just 16 calculating the flow-weighted means. 17 Q. The data from the District, what field 18 methodology was used to collect the nutrient 19 samples? 20 A. Could you be more specific? What do you mean 21 by field methodology? 22 Q. Was it a grab sample, cross-sectionally 23 integrated or continuous sample, whatever EPA 24 methods they have with all the bottles and how DEPOSITION OF WILLIAM W. WALKER, JR. 760 1 you use it? What happened in the field with 2 what instrumentation to collect those samples? 3 A. At the Park inflow points these samples are 4 grab samples to my knowledge. 5 Q. Are they cross-sectionally integrated samples? 6 MR. HARRISON: Object to the form 7 of the question. 8 A. If they are, I am unaware of that fact. These 9 are the data that were presented to the Park 10 for the purpose of setting water quality 11 standards at the inflows to the Park, and this 12 is the information that was available for that 13 purpose. 14 Q. Was the Seasonal Kendall test designed for use 15 on cross-sectionally integrated samples? 16 MR. HARRISON: Again, object to 17 form. You'd need a lot better definition. 18 A. In streams when subject to high velocities and 19 scouring conditions, when there is, for 20 example, erosion occurring in the watershed and 21 so forth, there are certain advantages to 22 collecting cross-sectionally integrated samples 23 from the point of view of calculating loads and 24 calculating flow-weighted concentrations and so DEPOSITION OF WILLIAM W. WALKER, JR. 761 1 forth. 2 The important thing is that the 3 sampling methodology, whether it is integrated 4 or a simple grab sample, is consistent over 5 time for the purpose of trend analysis. There 6 is no requirement that the samples be 7 cross-sectionally integrated in order to apply 8 the trend detection technique. 9 Q. Have you considered whether it would be 10 preferable to use cross-sectionally integrated 11 samples? 12 A. If one were designing the program, it would 13 generally be preferable to sample with a cross- 14 sectional integration if one had that degree of 15 freedom. 16 Q. Let me back you up to the computer program 17 which you said Crawford, et al, USGS use to 18 select the median among multiple seasonal 19 values. Did Crawford, et al verify validity of 20 their method using simulations? 21 A. Crawford, et al included Hirsch -- excuse me, 22 the authors to that article included Slack and 23 Hirsch who were the original coauthors of the 24 Seasonal Kendall test. They performed DEPOSITION OF WILLIAM W. WALKER, JR. 762 1 extensive simulations of the Seasonal Kendall 2 test that are described in Exhibit 36 and there 3 may be some simulations in Exhibit 37, and it 4 was -- I do not know whether they explicitly 5 simulated the effects of using, of taking the 6 medians within each season, but I know that the 7 inclusion of that particular algorithm taking 8 medians within each season in the computer 9 program by Crawford, et al was done under their 10 -- with their judgment, application of their 11 judgment. 12 Q. You talked about whether the grab sample or 13 integrated sample, the important thing being 14 the consistency of the sampling technique. We 15 talked a couple of days ago about review of the 16 actual technique methodologies and QA/QCs used 17 by the South Florida Management District in 18 collecting this water quality data. I want to 19 make sure at this point the record is clear and 20 I've asked you a question that is broad enough 21 to make sure you've covered the point you've 22 made in relation to Exhibit 17. 23 How for purposes of preparing 24 Exhibit 17, and preparing the analyses DEPOSITION OF WILLIAM W. WALKER, JR. 763 1 reflected in it, did you ascertain whether the 2 District database met that criteria of 3 consistency? 4 MR. HARRISON: Counsel, I think 5 your question is still too broad. This is 6 still directed at all the water quality 7 components or to phosphorus or though 8 phosphorus? 9 Q. If you can maybe address the whole water 10 quality database, and if we need to divide it 11 up, then I'd be more than happy to do that. 12 A. The surface water quality monitoring program 13 operated by the District is described in 14 District Publication 88-3 by Germain and Shaw 15 in 1988. That's a reference in Exhibit 17. 16 I have read that document and do 17 not recall seeing in that document any 18 discussion of changes in sampling methodology 19 or sampling procedure or analytical procedures 20 with reference to the Park inflow points. 21 Q. Do you know how many different field 22 technicians were responsible for collecting the 23 samples that generated the database utilized in 24 Exhibit 17? DEPOSITION OF WILLIAM W. WALKER, JR. 764 1 A. Not specifically, no. 2 Q. Do you have a general idea? 3 A. No, I have no idea. 4 Q. Have you done any review of the training of the 5 District personnel involved in this water 6 quality monitoring program? 7 A. No, I have not. 8 Q. In the second paragraph on page 3 of Exhibit 9 17, about the middle of that paragraph, you say 10 that a spatially-averaged rainfall time series 11 for WCA-3A has been constructed by averaging 12 data in nine gauges in and around the 13 reservoir. What data is that? 14 A. Excuse me, would you give me the page number 15 again for that reference? 16 Q. Page 3. 17 MR. HARRISON: Center paragraph, 18 counsel? 19 MS. AHEARN: Yes. 20 A. Those -- that spatially-averaged rainfall time 21 series was constructed by averaging data from 22 nine daily rainfall gauges that are located in 23 and around Water Conservation Area 3A. 24 Q. Is that data from the District? DEPOSITION OF WILLIAM W. WALKER, JR. 765 1 A. That's data that would be managed and stored on 2 the District's computer system. I am not 3 certain whether the District actually collects 4 all of those measurements or whether there 5 might have been some other agencies involved. 6 Q. Did you look to any source for this rainfall 7 data other than what was available through the 8 District? 9 A. Yes, I did. 10 Q. What other agencies are those? 11 A. The Corps of Engineers and the United States 12 Geological Survey. 13 Q. Did you incorporate data from all three sources 14 in your analysis? 15 A. I incorporated data from the nine stations that 16 had a continuous record during most of this 17 monitoring period. There may have been some 18 gaps in some of the stations, but these are the 19 routinely -- routinely monitored rainfall 20 stations in and around Water Conservation Area 21 3A. 22 Q. Did any corrections to this database need to be 23 made before you performed your analysis? 24 A. What do you mean by corrections? DEPOSITION OF WILLIAM W. WALKER, JR. 766 1 Q. We just talked about the water quality database 2 and it is my understanding that you took what 3 was available from the District and after 4 reviewing a publication that described that 5 monitoring program, relied on that data as 6 presented. 7 Friday we talked about the flow 8 data. As I recall, you explained to us you 9 took the District's flow data, but then had 10 USGS technicians make corrections to that 11 database. I'm just trying to figure out what, 12 if either one of those two processes or 13 something different happened here with the 14 rainfall data. 15 Did you adjust the District data 16 and accept that as presented or were other 17 avenues pursued in order to compile whatever 18 database you ultimately used? 19 MR. HARRISON: Object to form. 20 A. Well, that was a very long statement that you 21 just made. I'd like to correct or amend one 22 thing, and that is regarding your premise on 23 the water quality data. I did, I mentioned the 24 fact that I had reviewed the article, the DEPOSITION OF WILLIAM W. WALKER, JR. 767 1 report by Germain and Shaw that describes the 2 monitoring program, but the other basis for my 3 consideration of the water quality data as a 4 frame of reference for doing this trend 5 analysis was that it was presented to us by the 6 district staff and it was represented to us by 7 the district staff as being adequate for the 8 purpose of setting inflow standards to the 9 Park. 10 That was another reason why I 11 believe that the data, that if there were 12 methodology problems relating to the water 13 quality data, that the District staff would 14 have indicated that to us under the ONRW 15 process, so that was an additional reason, 16 aside from the report by Germain and Shaw, that 17 I relied upon in accepting this water quality 18 data for the purpose of a trend analysis. 19 With respect to the rainfall data, 20 the data were derived from the District 21 computer with the assistance of the US 22 Geological Survey. 23 Q. I'm sorry, the USGS assisted in getting the 24 data off of the District's computer? DEPOSITION OF WILLIAM W. WALKER, JR. 768 1 A. That's correct. 2 Q. At that point were any changes or modifications 3 to the data extracted made? 4 A. Well, in order to calculate an average value 5 for these nine stations, there were -- we had 6 to account for fact that at individual stations 7 there were periods of time, for example, when 8 no measurements were taken and there were also 9 periods of time when daily values were not 10 recorded. 11 In place of daily values there 12 would be a cumulative rainfall for a period of 13 a week, say there might be a value reported for 14 a week as opposed to the values for individual 15 dates, so those kinds of variations and factors 16 were considered in calculating the spatial- 17 average concentration or, excuse me, the 18 spatial-average rainfall for Water Conservation 19 Area 3A. 20 Q. What is the rationale for using spatially- 21 averaged rainfall time series for WCA-3A? Am I 22 correct this averages spatially rainfall data 23 throughout the entire area, the entire Water 24 Conservation Area 3A? DEPOSITION OF WILLIAM W. WALKER, JR. 769 1 A. The nine stations that I used are shown on 2 figure one. I also used a four-station average 3 that consisted only of the stations that were 4 located within the water conservation area. 5 The rationale for using the spatially-averaged 6 rainfall data was that the source of the flow 7 discharged into the Park from Water 8 Conservation Area 3A was Water Conservation 9 Area 3A and with more stations included in the 10 average, we would have a higher chance of 11 getting a representative value for the rainfall 12 on the entire system. 13 Q. Is it your understanding that all of the 14 rainfall that falls over WCA-3A ultimately 15 enters the Park? 16 A. No, not necessarily. 17 Q. Under what circumstances would that not be the 18 case? 19 A. Well, there are release points from WCA-3A that 20 do not discharge into the Park, plus there are 21 other processes aside from release as surface 22 flows. There are other fates for the rainfall. 23 Q. What percentage of the rainfall that falls 24 northeast of the Miami Canal at any time passes DEPOSITION OF WILLIAM W. WALKER, JR. 770 1 to the south of the Miami Canal? 2 MR. HARRISON: Objection. 3 Foundation. 4 A. I don't know the answer to that question. 5 Q. How much rainfall that falls over the western 6 basins, the areas to the west of WCA-3A, enter 7 Water Conservation Area 3A? 8 MR. HARRISON: Objection, counsel, 9 lack of foundation, and also I can't understand 10 the relevancy of talking rainfall. Are you 11 talking rainfall intact goes from point A to 12 point B with all its constituents and water 13 quality components? 14 A. I don't understand your question. 15 Q. Does any of the water that passes through the 16 S12 structures originate as rainfall over the 17 basins to the west of WCA-3A? 18 MR. HARRISON: Objection. 19 Foundation. 20 A. A portion of the watershed of Water 21 Conservation Area 3A is located to its west. 22 Therefore, rainfall on the western basin as a 23 result of rainfall and runoff processes may 24 contribute to Water Conservation Area 3A. DEPOSITION OF WILLIAM W. WALKER, JR. 771 1 Q. Did you try using any Theissen weighting of 2 rainfall method throughout Water Conservation 3 Area 3A relative to the S12 structures? 4 A. What was that word, counsel? 5 Q. Theissen. 6 A. Could you spell it? 7 MS. AHEARN: I doubt it. 8 MR. BURGESS: T-H-E-I-S-S-E-N. 9 A. No, I did not. It is my understanding that the 10 inflows to the Park, the schedules used to 11 determine the release of flow volume into the 12 Park under the so-called rain-driven plan are 13 also based upon rainfall over Water 14 Conservation Area 3A alone and they are also 15 based upon calculation of a spatially-averaged 16 rainfall that does not employ this Theissen 17 technique. It is a sample average of the 18 monitoring stations, so the technique that I am 19 using here is similar. 20 Q. When I asked you before about the rationale for 21 using this spatially-averaged rainfall, you 22 didn't mention the operating procedures of the 23 rainfall-driven plan. Am I correct now to 24 assume that was an additional rationale for DEPOSITION OF WILLIAM W. WALKER, JR. 772 1 using the spatial average among nine stations? 2 A. I did not consider the rain-driven plan 3 directly as a basis for using the average 4 rainfall. I just used spatially-averaged 5 rainfall to reflect the amount of rainfall I 6 think using some measure of the amount of 7 rainfall on Water Conservation Area 3A as a 8 hydrologic factor that might be correlated with 9 water quality at the S12s and S333. 10 Q. Are you aware of any documented pattern in 11 rainfall concentrations of phosphorus spatially 12 across WCA-3A? 13 A. To my knowledge, there has been no monitoring 14 of rainfall quality within WCA 3A and I have 15 seen no document that quantifies any spatial 16 gradings across water conservation areas. 17 Q. Are you aware of any information to indicate 18 that the concentration of phosphorus does 19 change as a gradient from the north to the 20 south of WCA-3A? 21 MR. HARRISON: Object on your 22 predicate. Assumes a fact not in evidence. 23 A. The information that I reviewed on atmospheric 24 quality on concentrations of phosphorus in DEPOSITION OF WILLIAM W. WALKER, JR. 773 1 rainfall does indicate that phosphorus 2 concentrations and loadings are higher in the 3 region around Lake Okeechobee and they are 4 lower in the extreme south in the vicinity of 5 the Everglades National Park Research Center. 6 Q. But that data wasn't particularly WCA-3A. 7 A. There were no stations within WCA-3A. 8 Q. Are you aware of any document or trends 9 concerning volume of rainfall over WCA-3A? 10 A. What type of trend are you referring to? 11 Q. Are you aware of any documented trend in terms 12 of the volume of rainfall? 13 A. I believe that there has been some work done by 14 the District and/or others regarding long term 15 trends in rainfall quality in South Florida. 16 Q. Are you familiar with that work? 17 A. Not in great detail, no. 18 Q. Are you familiar with any data concerning 19 changes in annual rainfall patterns in Water 20 Conservation Area 3A? 21 MR. HARRISON: Counsel, I am going 22 to object just because of the nature of it. 23 Are you talking patterns in quantity amount of 24 rainfall or in phosphorus? It seemed like DEPOSITION OF WILLIAM W. WALKER, JR. 774 1 before you asked a question on quantity or 2 volume and I heard a quality come back and I am 3 not sure. I want to make sure the witness 4 knows what he is answering. 5 Q. Are you aware of any information showing trends 6 in terms of the pattern, how much and when rain 7 comes down over southern Florida? 8 A. Well, I stated earlier I am aware of some work 9 that has investigated the possibility of long 10 term trends in the volume of rainfall. 11 Associated with that work there may have been 12 some discussion of patterns. I don't recall 13 specifically what they were. 14 Q. Would you be aware of any information 15 concerning changes in the beginning or the 16 intensity of the wet season versus the dry 17 season? 18 A. There may have been some work done on that 19 topic. 20 Q. If there was a trend in the timing and volume 21 of rainfall over South Florida during the 22 period of record of data utilized in Exhibit 23 17, is that a factor that should be accounted 24 for in the methodologies you have used? DEPOSITION OF WILLIAM W. WALKER, JR. 775 1 A. The reason that I included rainfall as a 2 hydrologic factor was to remove that portion 3 statistically, remove that portion of the 4 variability and concentrations that correlated 5 with variations in rainfall. Whether those 6 variations in rainfall reflect so-called trends 7 or whether they reflect random variations, wet 8 period, dry period, would be irrelevant. 9 Q. You mentioned that the 1980 S12 sampling was 10 perhaps more frequent than biweekly. Would 11 this more frequent sampling bias the Seasonal 12 Kendall test? 13 MR. HARRISON: Object only to your 14 characterization of the testimony. I am not 15 sure he testified that the entire year was 16 sampled that way. 17 A. To my recollection there may have been two 18 months when the sampling frequency was perhaps 19 weekly instead of biweekly. I do not believe 20 that that level of variation in sampling 21 frequency would have had a serious effect, 22 significant effect on the Seasonal Kendall 23 results. 24 Q. The types of water quality samples we are DEPOSITION OF WILLIAM W. WALKER, JR. 776 1 talking about here, they are positively skewed, 2 aren't they? 3 A. If you could be more specific, I could answer 4 your question. 5 Q. For this data is the frequency distribution 6 positively skewed for the long rate tail -- oh, 7 right tail. The long right tail? 8 MR. HARRISON: Object to form. 9 Q. I know what I am saying now. 10 MR. HARRISON: I don't think 11 counsel understands her own question and I am 12 not pretending to, but I will object to form 13 out of an abundance. Not scientific in terms 14 of art that every scientist would employ the 15 same definition. 16 A. This of course would depend upon which water 17 quality component you're talking about. 18 Q. Doctor Helsel has an article. I think you're 19 familiar with it: Less than obvious. Are you 20 familiar with an article by that title by 21 Doctor Helsel? 22 A. No, I am not. 23 Q. Have you ever talked with Doctor Helsel what 24 his definition of positively skewed today is? DEPOSITION OF WILLIAM W. WALKER, JR. 777 1 A. I know what a positively screwed distribution 2 is, if that's what you're asking, and I believe 3 I have had discussions with Doctor Helsel on 4 the topic. 5 Q. Do you know if your view of a positively skewed 6 distribution of water quality data accords with 7 that of Doctor Helsel? 8 MR. HARRISON: Again, counsel, I 9 object to your characterization. 10 Q. That's okay. Let me see if I can straighten 11 this out. 12 For the total phosphorus water 13 quality data, is the concentration frequency 14 distribution positively skewed? 15 MR. HARRISON: Caution the witness 16 to make sure you understand what the question 17 is before you answer it. 18 A. The degree of skewness would vary from one 19 station to another, but yes, there are stations 20 that I have examined here that have positively 21 screwed distributions in phosphorus 22 concentration. 23 Q. Those stations with positively skewed 24 distributions, for a month in 1988 when DEPOSITION OF WILLIAM W. WALKER, JR. 778 1 sampling was more frequency than bimonthly, 2 wouldn't the probability that the median value 3 for that month would be greater than the 4 corresponding month of the earlier, wouldn't 5 that probability be greater than one half? 6 A. I don't understand your question. 7 MR. HARRISON: I am going to 8 object. I think that's going to require a 9 definition of how many sampling points were 10 taken and I don't think you've laid an adequate 11 predicate to establish any trends. 12 Q. Okay. Doctor Walker, at one of these stations 13 where you say that the frequency distribution 14 is positively skewed, for season X in 1987, 15 where we have a biweekly sample and then in 16 1988 for season X where we have more frequent 17 than biweekly sampling, isn't the probability 18 that the median value for the 1988 season will 19 be greater than the 1987 season, isn't that 20 probability greater than one half? 21 MR. HARRISON: Counsel, is part of 22 your assumption that there were more than two 23 samples during the month in 1988 to which 24 you're referring or the timeframe? I do not DEPOSITION OF WILLIAM W. WALKER, JR. 779 1 think the question can be answered. 2 A. It would be impossible to answer that question 3 without having more information on just what it 4 is you're asking and being more specific about 5 the distributions. 6 Q. All right. Let's take year A for season X. We 7 have one sampling value and for year A +1, 8 season X, we have two sampling values. Isn't 9 the probability that the season X for year A +1 10 will be greater than the corresponding season 11 in year A, isn't that probability greater than 12 one half? 13 A. Not necessarily. 14 Q. What causes the concentration frequency 15 distribution, at those stations where you said 16 you did examine and find it, what causes those 17 distributions to be positively skewed? 18 A. That skewness reflects underlying phenomenon in 19 the system. I cannot ascribe it to a 20 particular cause. 21 Q. Are their causes inherent in the way the 22 samples are analyzed in the lab? 23 MR. HARRISON: Objection. Asked 24 and answered. DEPOSITION OF WILLIAM W. WALKER, JR. 780 1 A. The frequency distribution of the data reflects 2 the combined effect of whatever is happening in 3 the real system plus whatever happens in the 4 lab in the process of analysis. 5 Q. Do you have any documentation that shows how 6 you calculated the spatial averaging of the 7 rainfall time series? I'm looking for a 8 shortcut way so you don't have to explain it to 9 me. I have a very vague idea of what this 10 means. Is there something you can point me to 11 that would explain exactly how you conducted 12 this spatial averaging? 13 A. It is just a simple arithmetic average on each 14 day of the record across each station that has 15 a measurement. 16 Q. Weighing each station equally? 17 A. As I said earlier, there was no spatial 18 weighting. It was just a simple arithmetic 19 average. It is the same way as is used in the 20 rain-driven plan. 21 Q. There was neither weighting in terms of 22 distance from the S12 structures nor was there 23 weighting to correlate to some particular 24 spatial area associated with a given -- DEPOSITION OF WILLIAM W. WALKER, JR. 781 1 A. There was no spatial weighting. 2 Q. The last type of hydrologic variable that you 3 discuss in your data set is water elevation. 4 What is the source of the data you used there? 5 A. There is a gauge, operated, I believe, by the 6 US Geological Survey that is upstream of S12C 7 that is a consistently monitored gauge over 8 this timeframe. 9 Q. So that it came directly from USGS in Miami? 10 A. I believe it came out of the District 11 computer. The District computer would also 12 store that information. 13 Q. When structure D is open -- structure 12D is 14 open and the other exposures are closed, does 15 the actual upstream elevation -- 16 A. I did not use the gauge at S12C to necessarily 17 represent the elevation that was present at 18 every location throughout Water Conservation 19 Area 3A at any particular time. I just use it 20 as a general gauge of water elevation at that 21 particular location which, as is stated in my 22 report, is highly correlated statistically with 23 water elevations measured elsewhere in the 24 system. DEPOSITION OF WILLIAM W. WALKER, JR. 782 1 Q. Did you look at how representative across all 2 the S12 structures the gauge at S12C was? 3 A. Yes, I have the data in the data files that I 4 transferred to the District. There were some 5 elevation measurements that were collected at 6 the other locations that I do not believe were 7 as frequent; but I did look at that and compare 8 them with the water elevation at S12C. 9 Q. From that process, looking at the data that you 10 did look at, can you tell me how representative 11 the gauge upstream from S12C is of, for 12 example, stage at S12A when structure S12D is 13 open? 14 A. As I stated earlier, I am not using the water 15 elevation at S12C as a measure of water 16 elevation everywhere in the system. I am just 17 saying it is correlated with other measures of 18 water elevation in the water conservation. 19 Q. I do understand that. I'm just asking you if 20 you can provide me this other information. 21 A. There would be periods when S12D is open, for 22 example, that the water elevation upstream of 23 S12D will drop because water is moving. 24 Q. Will that impact readings at S12C gauge? DEPOSITION OF WILLIAM W. WALKER, JR. 783 1 A. It may. 2 Q. Do you know by what measure? Can we quantify 3 that impact? 4 A. I can't quantify it, no. 5 Q. Can you quantify how accurate the stage 6 readings at S12C are? 7 A. The stages are recorded to the nearest one 8 hundredth of a foot by the USGS and that would 9 in turn, since the USGS has the responsibility 10 for making the measurements and reporting the 11 data, the accuracy of the measurements would, 12 that their impressions of the accuracy of the 13 data would be reflected in the number of 14 significant digits -- digits that they report. 15 Q. So we presume the number of digits reported by 16 the USGS in their data is accurate to that 17 order of magnitude? 18 A. I am not certain, but if the accuracy was on 19 the order of a foot, say, then the USGS would 20 presumably report the measurements to that 21 resolution. They report them to a much higher 22 resolution. I don't have the precise value for 23 the accuracy of the USGS water elevation 24 measurements. DEPOSITION OF WILLIAM W. WALKER, JR. 784 1 Q. Can you quantify the accuracy of the rainfall 2 data utilized in your analysis? 3 A. No, I cannot. 4 Q. If I could turn your attention to page 4 of 5 Exhibit 17. 6 In the very lower, lower 7 right-hand corner you have a graph for rainfall 8 at Tamiani Trail. To the left on the left 9 upper corner of that is a graph for rainfall at 10 WCA-3A. Do you see the one I'm referring to? 11 It would be the fourth down in the middle 12 column and the bottom in the right-hand 13 column. 14 A. Yes. 15 Q. Did you do any comparisons of those two 16 databases? 17 A. The calculation of rainfall for WCA-3A would 18 have included the data from the Tamiani Trail 19 station as one of the stations that were used 20 in calculating that average for WCA-3A. Beyond 21 that, I did no comparisons. 22 Q. So we don't have any correlation data on how 23 closely the two data sets here track one 24 another? DEPOSITION OF WILLIAM W. WALKER, JR. 785 1 A. They are measuring only part of the same 2 thing. There is no reason to suspect that they 3 would be correlated in a very strong way 4 because we're averaging over a large area. 5 Q. What is the total area over which the Water 6 Conservation Area 3A averages apply? 7 A. I don't recall the exact acreage. Whatever the 8 size of Water Conservation Area 3A is. 9 (Recess.) 10 Q. Doctor Walker, I want to jump forward in the 11 context of the report a little bit, but it is 12 kind of related to this whole data collection 13 database generation issue I have been trying to 14 focus on this last session. 15 In Series C when you perform from 16 regression on hydrologic variables, in that 17 regression do you use all actual data points 18 per season or do you first calculate a mean or 19 a median, excuse me, for the season to use in 20 those regressions? 21 MR. HARRISON: Object to form 22 because the question doesn't specify which 23 variables you're talking about. If it makes 24 any difference, I do not know. DEPOSITION OF WILLIAM W. WALKER, JR. 786 1 A. The regression is performed using all the data 2 in the original form without reference to 3 season. 4 Q. And then from those residuals do you derive one 5 value per season? 6 A. Correct. 7 Q. That database which you developed to use in 8 your Series C, does that constitute a sample of 9 independent and identically distributed random 10 variables? 11 A. I don't see the distinction between that and 12 our previous discussion regarding use of the 13 original concentration data as compared with 14 this Series C residuals. 15 Q. The data that you subject to the Seasonal 16 Kendall test are residuals, in other words the 17 difference between certain data values and a 18 regression slope, correct? 19 A. Regression equation, correct. 20 Q. Okay. And doesn't each one of those residuals 21 differ in terms of its -- I think the term is 22 noise? 23 A. The residuals are defined as population of 24 values that is the difference between the DEPOSITION OF WILLIAM W. WALKER, JR. 787 1 measurements and the predictions. There is 2 nothing systematic about the variation in that 3 noise. 4 Q. Well, you have a number of discrete residuals 5 which are calculated from this regression 6 procedure, correct? 7 A. Correct. 8 Q. And isn't each residual value unique in terms 9 of its variance? 10 A. Residuals are from a population that has a 11 certain variance. Each residual has its own 12 value. 13 Q. The residual values are not freely 14 exchangeable? 15 A. I don't understand that question. 16 Q. They couldn't be randomly reordered? 17 A. The residuals have a certain time sequence, 18 correct, otherwise it wouldn't make any sense 19 to do a trend analysis on them. 20 Q. This may be quite simplistic, but from a lay 21 perspective, and this will all have to be 22 reduced to a lay perspective if a judge is 23 going to listen to it, a regression analysis 24 can be depicted as a number of points through DEPOSITION OF WILLIAM W. WALKER, JR. 788 1 which you would fit a regression slope, 2 correct? 3 A. Well, in this case it is a multivariate 4 regression. There may be more than one slope 5 involved. 6 Q. Maybe I misused the terminology, but a 7 regression line has a given slope, correct? 8 A. Again, if there is more than one variable, then 9 there is more than one slope. If there is more 10 than one independent variable, such as rainfall 11 and water elevation, then there is more than 12 one slope involved. 13 Q. Let's talk in terms of unit area regression. 14 If the slope of the regression line were 15 changed, we title regression line at its median 16 point, doesn't that have a different impact on 17 the variance of values all along that line? 18 A. Why would one title slope of the regression 19 line? 20 Q. I'm trying to look for the relationship among 21 all these values. Isn't it true that they are 22 not all random identically distributed, 23 randomly and independent in the sense that they 24 all have unique variants as exemplified by the DEPOSITION OF WILLIAM W. WALKER, JR. 789 1 consequences of changing the slope of the 2 regression line? 3 MR. HARRISON: Object to form. 4 A. As I stated, there is no reason to change the 5 slope of the regression line. The regression 6 line is the least square fit of the data and 7 there is no reason to change that slope. I 8 don't understand. 9 Q. If you have, if you were to use a different 10 database, then you would change the slope, 11 wouldn't you? 12 A. Whatever the database dictated, that would 13 dictate the slope. 14 Q. If you were to cast out outliers, would that 15 dictate a change in the slope? 16 A. It may. 17 Q. If for whatever reason after you have fit this 18 unit variate regression you change the valve 19 one residual, won't you necessarily change the 20 values of all the other residuals? 21 A. I don't understand why we're changing anything. 22 Q. I am asking you to address this as a 23 hypothetical then. 24 MR. HARRISON: Objection. DEPOSITION OF WILLIAM W. WALKER, JR. 790 1 A. There is no reason to get involved in changing 2 anything. There is regression equations that 3 developed from the data and they produce a time 4 series of residuals that are independent of the 5 hydrologic factors such as rainfall and water 6 elevation. I don't see any reason why anyone 7 would want to change the slope of the 8 regression line. 9 Q. Isn't it important to fully understand the 10 nature of your database to ensure that it 11 satisfies the assumptions of the methodologies 12 you're going to apply to that database? 13 MR. HARRISON: Object to form. 14 A. Yes, in general, that is true and that's 15 exactly what I've done. 16 Q. Are the regression values employed in Series C 17 independent and identically distributed random 18 variables? 19 MR. HARRISON: Object to form. 20 Not established, counsel, that those are terms 21 of art that have any meaning or whether there 22 is a degree of randomness you're looking at. 23 You call it random. He has answered the 24 question. The answer to the question may be DEPOSITION OF WILLIAM W. WALKER, JR. 791 1 nonresponsive to your question and 2 unintentionally so. 3 (Question was reread.) 4 A. Are we still waiting for an answer from me? 5 Q. Yes, excuse me. 6 A. The regression residuals generated in Series 3 7 analysis to the flow-adjusted concentrations 8 that were generated by Smith, Hirsch and Slack 9 in their study nationwide of trends in 10 phosphorus concentrations in river basins in 11 the United States, essentially I am using the 12 same methodology that they have used, and they 13 were the developers of the Seasonal Kendall 14 test; and as far as I know, the assumptions of 15 the Seasonal Kendall test are met with this 16 hydrologic adjustment procedure. 17 Q. Well, Hirsch, et al in developing this 18 technique for trend analysis employed this 19 flow-adjusted concentration to determine 20 whether there had been a change in the 21 relationship between flow and concentration, 22 correct? 23 A. Yes. 24 Q. And you're not using the hydrologic variable of DEPOSITION OF WILLIAM W. WALKER, JR. 792 1 flow in Exhibit 17 at this step of the 2 procedure, are you? 3 A. As I used flow in the trend analysis for 4 nutrient transport, but it really doesn't 5 matter whether you are using flow or rainfall 6 or water elevation. The concept is the same, 7 the concept of using a regression equation 8 related to some hydrologic factor to remove the 9 portion of variable that correlated with that 10 hydrologic factor so that the resulting trend 11 test would more closely reflect underlying 12 trend data, if it were present, with an effect 13 that was just related to variations in 14 hydrology. 15 Q. In developing the Seasonal Kendall test with 16 this method of using a time series of flow 17 adjusted concentrations, did Hirsch, et al 18 perform Monte Carlo experiments to test the 19 flow-adjusted concentration method? 20 A. I don't know. 21 Q. Do you know if the flow-adjusted concentration 22 method has elsewhere been subjected to the same 23 type of simulations and tests as the 24 straightforward, the usual form of the Seasonal DEPOSITION OF WILLIAM W. WALKER, JR. 793 1 Kendall test? 2 MR. HARRISON: Object to form. 3 A. I don't know if anyone has done Monte Carlo 4 simulations to investigate that effect 5 specifically, but I have no reason to suspect 6 that the technique of flow-adjusted 7 concentration would in any way invalidate the 8 application of the Seasonal Kendall test for 9 trends. 10 Q. Do you know of anyplace in the scientific 11 literature where it has been determined that 12 the flow-adjusted concentration method 13 satisfies the assumptions underlying the 14 seasonal Kendall test? 15 A. There is no reason to suspect on an a priori 16 basis that the flow-adjusted method does not 17 satisfy the assumptions of the Seasonal Kendall 18 test. 19 Q. You said that, correct me if I misphrase you, I 20 don't mean to mischaracterize any testimony, 21 but that it didn't matter which hydrologic 22 variable you selected, that the concept would 23 be the same. Why don't you use all hydrologic 24 variables in the flow-adjusted concentration DEPOSITION OF WILLIAM W. WALKER, JR. 794 1 procedure? 2 MR. HARRISON: Object to form. 3 You mean generally or what he did in this 4 analysis? 5 A. What do you mean by all hydrologic variables? 6 Q. Well, first you used flow, am I correct, in the 7 July 3, 1990 version of your paper? 8 A. I believe that I used flow and water elevation. 9 Q. And then in the final version of the report, 10 Exhibit 17, we see the flow variable replaced 11 with water elevation, correct? 12 A. No, that's not correct. 13 Q. That's not correct. Do you continue your 14 treatment of flow in your analysis of the 15 Series C database from the July 3, 1990 draft? 16 You seem to treat that variable in Series C the 17 same way in Exhibit 17. 18 A. No. 19 Q. Which hydrologic variables do you address in 20 Exhibit 17? 21 A. I guess the question was do you, as is shown on 22 page 8, figure 3, the variables, hydrologic 23 variables are elevation and rainfall. 24 Q. Wouldn't it also be preferable to employ the DEPOSITION OF WILLIAM W. WALKER, JR. 795 1 variable of flow? 2 A. Because of the feedback that I received from my 3 reviewers, there was a comment as to the 4 appropriateness of using flow for calculating 5 flow adjusted -- for calculating hydrologic 6 adjustment, adjusted concentrations. The 7 analysis in the final report does not emphasize 8 use of flow in the sense of the bulk of the 9 work. 10 However, flow is used as part of 11 the sensitivity analysis for S12 structures as 12 an additional explanatory variable aside from 13 rainfall and water elevation. 14 Q. Could you state the rationale for not including 15 it in the Series C analysis? 16 A. The rationale, as I mentioned earlier, is 17 described in one of the comments made by the 18 reviewers and it is also a rationale that was 19 expressed in the report by Mattraw, et al, 20 namely that flow in this system over time is 21 influenced at eight given -- at particular 22 structures by water management decisions, 23 decisions to release more water on one side of 24 the system as compared to another over DEPOSITION OF WILLIAM W. WALKER, JR. 796 1 different time periods and it was because of 2 those changes in water management that would 3 reflect not so much variations in the natural 4 system, but decisions on the part of the water 5 managers to alter the flow or release patterns, 6 that it was not appropriate to use flow as a 7 hydrologic factor for adjusting the 8 concentrations. 9 Q. That is if your objective is only to correct 10 for the influence of natural factors and no 11 management factors. 12 A. When the flow from one structure is much lower 13 during one part of the record as compared with 14 another part of the record because of the 15 intentional decision to change the release 16 pattern, it would be risky to use that flow as 17 a factor for explaining concentration 18 variations prior to a trend analysis. 19 Q. The water management changes you just 20 described, those are at least in part due to 21 directions from the United States Congress? 22 MR. HARRISON: Object. No 23 foundation. Calls for a legal conclusion. 24 Q. Okay. Dr. Walker, you said you talked in terms DEPOSITION OF WILLIAM W. WALKER, JR. 797 1 of water management decisions. Who do you 2 understand to have made decisions regarding 3 these changes in operations to the Park? 4 A. These decisions were made in conjunction with 5 the District, the Corps of Engineers and the 6 Park. 7 Q. Are you aware of any role of the United States 8 Congress in this? 9 A. Well, the United States Congress initially 10 mandated what was called the minimum delivery 11 schedule for the S12s, but beyond that, I am 12 unaware of particular involvement of the 13 Congress. 14 Q. What do you understand was the role of the 15 Corps in these water management changes? 16 A. The decisions on changes in the flow release 17 patterns for Shark slough were made by people 18 from the Park, from the Corps and from the 19 District. I cannot separate out or I do not 20 know exactly who advocated what, but this was a 21 -- these were decisions that were made in 22 concert with these three agencies and they 23 reflect how the system operated. 24 Q. When you set out to do your trend analysis, you DEPOSITION OF WILLIAM W. WALKER, JR. 798 1 wanted to determine whether or not there was a 2 trend in the phosphorus concentrations going 3 into the Park, correct? 4 A. Whether or not there was a trend in any of the 5 water quality variables that I examined. 6 Q. But you didn't want to see -- you didn't -- you 7 wanted to exclude any trend that was related to 8 certain factors? Is that correct? 9 A. I didn't want to exclude anything. I just 10 didn't want to confuse variations that were a 11 direct result of changes in water management 12 with variations that were -- that reflect the 13 underlying concentration distribution. 14 Q. But the two are overlapping, aren't they? 15 A. In the sense that the concentration data are 16 taken from the system and there were changes in 17 water management during this period, you cannot 18 totally separate out the effects statistically. 19 Q. Why did you select flow as a Series C variable 20 in your initial draft? 21 A. Because flow is a fundamental hydrologic 22 variable that is important, because if you have 23 a trend at a given flow, that's another way of 24 saying that you have a trend in the loading DEPOSITION OF WILLIAM W. WALKER, JR. 799 1 that occurs under a given condition. 2 Q. If phosphorus inputs to Everglades National 3 Park are to be reduced, could that reduction be 4 obtained by altering the flow? 5 A. Altering the flow in what regard? 6 Q. In the regard that used it in Series C, the 7 volume flow? 8 MR. HARRISON: Counsel, I'm going 9 to object because the broad nature of your 10 question to the Everglades National Park in 11 general. If that is what you mean, then I wish 12 you would so state. If you talk about one 13 particular side or basin or structure or series 14 of structures, I think the question needs to be 15 more specific. 16 Q. What I am trying to get at here is defining, 17 you talk about looking at trends, but we're not 18 looking at the entirety of phenomenon, 19 correct? I want to understand by what 20 rationale you limited the universe of changes 21 in the water quality database to determine: 22 Here is the range in which I will look for a 23 trend. 24 MR. HARRISON: Object to your DEPOSITION OF WILLIAM W. WALKER, JR. 800 1 characterization of Doctor Walker's work. 2 A. I did not limit my examination in the way that 3 you are implying. 4 Q. Maybe I just misstated my predicate wrong. I 5 don't mean to imply anything. I'm just trying 6 to get this line of questioning over. Let me 7 do it this way. 8 If we wanted to determine whether 9 there was a trend in phosphorus concentrations 10 in Everglades National Park, for purposes of 11 determining whether that trend was related to 12 activities conducted in the Everglades 13 agricultural area, would you include flow along 14 with rainfall and elevation as hydrologic 15 variables in Series C analysis? 16 A. The question of whether or not trends are 17 related to activity in the Everglades 18 agricultural area would have nothing whatsoever 19 to do with which hydrologic factors I would 20 select in doing my Series C analysis. 21 Q. If your objective was as I just stated, how 22 would you design your trend analysis? 23 A. My trend analysis is designed to document 24 whether or not increases in phosphorus DEPOSITION OF WILLIAM W. WALKER, JR. 801 1 concentration and phosphorus transport occurred 2 during that monitoring period and whether those 3 increases, whether those trends are related to 4 variations in rainfall or water elevation or 5 inflow that was the objective of the analysis. 6 Q. Isn't water elevation controlled and changed by 7 water management decision? 8 A. Yes, it is, to some extent. 9 Q. As I understood the rationale that you've 10 explained to me for changing from flow to 11 rainfall in terms of the Series C variables, 12 that was part of the rationale, correct? 13 A. Which was part of the rationale? 14 Q. That the flow is altered by water management 15 decisions? 16 A. The flow was in some cases drastically altered 17 by water management decisions, correct. 18 Q. Is that also true of water elevation? 19 A. Not to the same extent. 20 Q. To what extent is it true? 21 A. Not to the same extent as, for example, looking 22 at flows through S12A over time. If you refer 23 to figure two on page 4, you will see that, for 24 example, the flow through S12A was relatively DEPOSITION OF WILLIAM W. WALKER, JR. 802 1 low in the early part of the record and 2 increased in the later part of the record. The 3 same thing, for example, would be true for the 4 flow through S333. And those are the -- those 5 variations that we're looking at in terms of 6 flow partially reflect differences in the 7 operation or the water management strategy as 8 well as differences in climate as driven by 9 rainfall, water availability. 10 The effects of the changes on -- 11 of this change in water management on the water 12 elevation regime were not nearly as drastic or 13 as large as the effects on flow. 14 Q. Have you done some type of statistical measure 15 of say the changes in flow at S12A as reflected 16 in the upper left-hand graph and the changes in 17 the stage at S12 in particular S12C in the 18 lower middle graph? Have you done some 19 statistical measure of how drastic the changes 20 in those two variables are? 21 A. Well, we can see just by looking at the graph 22 of the water elevation upstream of S12C what is 23 referred to as upstream stage and the bottom of 24 figure two in the center that the range is on DEPOSITION OF WILLIAM W. WALKER, JR. 803 1 the order of six to maybe ten feet or so and 2 that range occurs at various times throughout 3 the period of record. 4 Q. What would be the total population in terms of 5 what's possible for those stage measurements? 6 A. Well, these are monthly values that we're 7 looking at. I don't recall exactly what the 8 minimum elevation upstream of S12C is 9 physically. It is -- I believe it is somewhere 10 -- it would be somewhat less than what is 11 shown in this figure, but I don't recall the 12 exact numbers. 13 Q. Do you know, are we looking at a variability 14 that spans ten or 100 percent of the possible 15 values? 16 In terms of the upstream stage 17 measurements? 18 A. We're looking at the range that occurred over 19 that particular timeframe between 1977 and 20 1990. I suppose theoretically higher stages 21 would be possible under extreme floods during 22 certain days, and I am not sure, as I said 23 earlier, exactly what the minimum stage would 24 be, but the minimum stage would reflect DEPOSITION OF WILLIAM W. WALKER, JR. 804 1 essentially completely drawn-down situation 2 with no water in Water Conservation Area 3A. 3 Q. If the range of possibilities are from say five 4 feet to ten feet, the world does not physically 5 hold that other options, wouldn't this graph of 6 the upstage screen show measurements are highly 7 variable? 8 MR. HARRISON: Object to the 9 hypothetical. 10 A. I don't know what you mean by highly variable. 11 The graph of the upstream stage at S12C is 12 shown in figure 2 and those are the ranges of 13 measurements. 14 Q. To a statistician's eye this graph is not 15 highly variable? 16 MR. HARRISON: The one at the 17 bottom of the center column, Figure 2, 18 counsel? 19 Q. Yes. 20 A. I am not saying that it is highly variable or 21 that it is weakly variable. It is what 22 happened. It is what was observed in the 23 system and what occurred in the system during 24 this timeframe. DEPOSITION OF WILLIAM W. WALKER, JR. 805 1 I am not characterizing it with 2 respect to degree of variability at all. 3 Q. I'm trying to get back to the rationale for 4 taking flow out of Series C, at least the part 5 on the rationale that it is subject to drastic 6 changes for water management purposes but 7 leaving the upstream water elevation which is 8 similarly subject to change because of water 9 management decision, and you directed us to 10 Figure 2 and in particular to these two graphs, 11 one of the flow at S12A and the second of the 12 upstream stage. 13 I am trying to understand, you 14 characterize the flow as highly variable and 15 the upstream stage as apparently not. 16 MR. HARRISON: I will object to 17 part of the characterization that he has not 18 said that flow and elevation are similarly 19 correlated or whatever the word you used was. 20 Other than that, go. 21 Q. So to ask you the question -- 22 A. I guess, let me try this. 23 Q. Okay. 24 A. My point is that in general the flow through DEPOSITION OF WILLIAM W. WALKER, JR. 806 1 S12A and the flow through S333 were greater in 2 the later part of the record, were higher, 3 there was more flow through those structures 4 and that, those differences from the early part 5 of the record to the later part of the record 6 reflected changes in the water management 7 decisions. 8 The elevation record upstream of 9 S12, it is hard for me to see conditions -- we 10 do not have the same situation there. We have, 11 in the early part of the record, we have 12 periods of lower elevation and we have periods 13 of high elevation, and in the latter part of 14 the record we have periods of low elevation and 15 we have periods of high elevation. So that's 16 why I'm less concerned about the effects of 17 water management on the upstream stage as 18 compared with the effects on the flow. 19 Q. Have you investigated what volume of canal or 20 levy seepage enters Everglades National Park? 21 MR. HARRISON: Object to form. 22 Compound. When you have -- is that really 23 compound or are you talking about seepage 24 through a levy out of a canal? If it is not DEPOSITION OF WILLIAM W. WALKER, JR. 807 1 compound, fine. 2 Q. Doctor Walker, have you investigated what 3 volume of seepage, whether it be directly from 4 a canal or from a canal through a levy -- I 5 just didn't want to split my question in half 6 -- enters Everglades National Park? 7 A. No, I have not. 8 Q. Do you have any opinion as to whether that 9 volume of seepage is significant in terms of 10 the flows in Everglades National Park? 11 A. I have used the daily flow data as reported by 12 the USGS in representing the flows entering the 13 Park through individual structures. Just as 14 the district has used the same information or 15 information from the same gauges in 16 constructing water balances and nutrient 17 balances for the water conservation areas. 18 Q. Do you have any estimate of what additional 19 input results from canal/levy seepage? 20 MR. HARRISON: I am not sure the 21 witness was finished answering the prior 22 question but. 23 Q. I apologize if I cut you off. 24 A. I do not have a quantitative assessment of the DEPOSITION OF WILLIAM W. WALKER, JR. 808 1 magnitude of seepage into the Park. 2 Q. Have you looked at information concerning 3 changes in canal levels into canals adjoining 4 or within the Park? 5 MR. HARRISON: Object. Compound. 6 Lack of foundation. 7 A. I have looked at information on water 8 elevations upstream of S12C in Water 9 Conservation Area 3A as is shown in Figure 2 on 10 page 4, and I have information on water levels 11 at the marsh stations monitored in the Park, 12 and I believe that the database that I have 13 assembled also contains some water elevations 14 measured at various locations in Water 15 Conservation Area 3A; but I haven't done an 16 intensive analysis of that information. 17 Q. Have you considered what degree of 18 remineralization for phosphorus recurs from 19 soils exposed or reflooded when elevations in 20 and around the Park occur? 21 MR. HARRISON: Object. Assumes 22 facts not in evidence in this deposition. 23 A. I have analyzed data from the Park inflow 24 points for trends and I have not considered DEPOSITION OF WILLIAM W. WALKER, JR. 809 1 remineralization in any of this analysis. It 2 is not part of my analysis or part of my scope 3 of work. 4 Q. Do you know of anyone who has looked at what 5 degree of phosphorus entering the Park may 6 reflect remineralization within or along canals 7 and levies at or adjoining the Park? 8 A. To my knowledge that has not been quantified. 9 Q. Could you please walk me through the procedure 10 you used to calculate the composite 11 concentrations of the flow-weighted average 12 values across the S12 structures? 13 A. We would start from the database that would 14 contain for each sampling date sample 15 concentration and the daily flow for each 16 structure: A, B, C and D; and any given 17 sampling date we would take the values from 18 each of those structures A, B, C and D and 19 calculate the flow-weighted concentration for 20 the S12s. 21 Q. Per structure? 22 A. For the entire cross-section, if that's the 23 question you're asking. 24 Q. Yes. DEPOSITION OF WILLIAM W. WALKER, JR. 810 1 A. To calculate the flow-weighted concentration 2 for the entire cross-section, I would sum the 3 products of flow and concentration for each of 4 the structures and divide by the sum of the 5 flows. That procedure would be applied to each 6 sampling date and that would produce a series 7 of flow-weighted concentrations for each water 8 quality component and each sampling date that 9 would reflect the composite discharge or total 10 discharge to Shark River slough. 11 Q. In the composites when you include the adjusted 12 flow from 334, you follow the same basic 13 calculation approach? 14 A. The same procedure except the flow through S334 15 would be subtracted from the flow through S333. 16 In that way the net value would more closely 17 reflect the net discharge through or to 18 northeast Shark River slough; and if that net 19 value after the subtraction is positive, I 20 would include it with the values for the S12 21 structures in calculating the flow-weighted 22 composite -- for the Shark slough. 23 Q. So in that case you would take the sum of the 24 products for five structures and divide by the DEPOSITION OF WILLIAM W. WALKER, JR. 811 1 sum of the flows through five structures? 2 MR. HARRISON: Objection, 3 counsel. Sum or product? I think we're mixing 4 apples and oranges. 5 It may be my mistake. 6 A. I don't recall if in the procedure that I did 7 that calculation for the combined discharge to 8 Shark slough whether I first calculated the 9 composite discharge through S12's and then 10 combined that with S333 or whether I did them 11 all together at the same time; that is, used 12 all five stations at once. There are two 13 different ways of doing that. 14 Q. Do we have any records we could look at to 15 determine which way the calculations were 16 performed? 17 A. It should not make any difference. There may 18 be in the files that I transferred, that I 19 supplied, there may be some file in there that 20 would indicate exactly how that was done, but I 21 can't be sure. 22 Q. I'm looking on page 5 and in the first full 23 paragraph under the descriptions of the S12T 24 and S12_334 databases, in that paragraph, you DEPOSITION OF WILLIAM W. WALKER, JR. 812 1 talk about compositing these flow-weighted 2 average values and then you write: On 3 sampling dates when no discharge occurred 4 through any structure, simple arithmetic 5 averages have been used. 6 First, I want -- I need to ask 7 you, when you say no discharge occurred through 8 any structure, does that mean no discharge 9 occurred through any one or more of the 10 structures being composited or that there was 11 no discharge through any of them, all of them? 12 Do you get my distinction? Last time you 13 did -- 14 A. If in calculating these composite discharges 15 for each date, marching through the time 16 series, I encountered a date in which there was 17 no flow through any of the sample structures, 18 but there were concentrations reported, I 19 simply averaged the concentrations. 20 Q. How did you calculate the flow-weighted average 21 value where there was flow at at least one of 22 the structures? One but not all? 23 A. If there was flow at one of the sample 24 structures, then by definition the flow- DEPOSITION OF WILLIAM W. WALKER, JR. 813 1 weighted concentration for the entire 2 cross-section would have been the concentration 3 that was measured at that structure through 4 which there was flow and that would come out of 5 the calculation procedure that I described 6 since flow was zero at the other structures, 7 they would not-- the concentration measured at 8 the other structures would not factor into the 9 calculation of the flow-weighted composite. 10 Q. So if there is flow through at least one 11 structure, do you not use simple arithmetic 12 averages? 13 A. Correct. 14 Q. Did you consider any alternative methods to 15 perform the flow-weighted averaging 16 calculations here when you had no flow? 17 A. When there is no flow, there is no flow- 18 weighted concentration. It has no definition 19 and no meaning. 20 Q. So you didn't consider any other approach under 21 that scenario? 22 A. There is no way of calculating a flow-weighted 23 concentration when there is no flow. 24 Q. When you had flow at at least some but not all DEPOSITION OF WILLIAM W. WALKER, JR. 814 1 of the structures, did you consider 2 alternatives for calculating flow-weighted 3 average values? 4 A. This is a fairly straightforward calculation. 5 I didn't consider that there were any 6 significant alternatives. 7 Q. And then you talk about doing this compositing 8 to minimize the effects of shifts and flow 9 distribution across individual outlet 10 structures during the monitoring period, did 11 you perform any analysis to quantify those 12 effects? 13 A. The purpose of the flow-weighted compositing 14 was to come up with representative values for 15 discharges to Shark slough through the S12s, 16 that is, the station that I refer to as station 17 S12T, and for representative value for the 18 total discharge to Shark slough and that is 19 what I refer to as S12_334. 20 And I subsequently tested those 21 composite time series for trend are as being 22 representative of the total flow entering those 23 portions of the slough. 24 Q. When you talk about minimizing the effects of DEPOSITION OF WILLIAM W. WALKER, JR. 815 1 shift in flow distribution, are those effects 2 -- what do you mean by the word effects there? 3 A. What I meant was I wanted to use, to provide 4 time series that were reflective of the total 5 input put to the different portions of the 6 slough and that would be less sensitive to 7 shifts in flow distribution than would the 8 values for the individual release points. 9 Q. You're looking at effects just in terms of the 10 flow volumes as between the structures and not 11 some effect on the water quality that those 12 shifts would produce? 13 A. That is correct. I was not in the process of 14 calculating these flow-weighted mean 15 concentrations for the cross-section attempting 16 to remove effects of the shifts in flow 17 distribution from the concentrations. 18 Q. When we talked about preparing the rainfall 19 database, you told us that in some instances 20 you find instead of a discrete measurement for 21 a given day a cumulative measure that might 22 span, for example a two-week period that you 23 would have to adjust to provide an appropriate 24 database, is that correct? DEPOSITION OF WILLIAM W. WALKER, JR. 816 1 A. I would have to account for that, correct. 2 Q. In calculating the flow-weighted average values 3 across the structures discharging into Shark 4 River slough, where you have flow at at least 5 one but not all of the structures, could you 6 look at flow data that is available for use in 7 those calculations? 8 A. I did look at all the flow data that were 9 available. 10 Q. Let me make sure I am clear here. 11 In calculating these flow-weighted 12 averages, did you distinguish between dates on 13 which you simply had no flow measurement versus 14 dates on which the flow was recorded as being 15 zero? 16 A. To my recollection in the case of Shark slough, 17 I did not have to make that distinction since 18 for this period of record we had measured daily 19 flows for each structure. 20 Q. So you don't have missing data points for this 21 purpose? 22 A. Correct. 23 Q. If the goal is to reflect total releases to 24 Shark River slough, couldn't you calculate the DEPOSITION OF WILLIAM W. WALKER, JR. 817 1 flow-weighted average values in instances where 2 -- couldn't you calculate the flow-weighted 3 average values by using the flow data available 4 for the two-week period centered at the 5 sampling date? 6 A. That would be an alternative way of doing the 7 calculation, correct. 8 Q. Did you consider doing it that way? 9 A. I considered doing it that way but to be 10 consistent with the methodologies that we 11 agreed upon with District staff in the ONRW 12 discussions, we agreed that assigning the daily 13 average flow to the sample concentration would 14 be an acceptable technique for calculating a 15 flow-weighted concentration at the Park inflow 16 points. 17 Q. It appears on a number of points your decision 18 was at least guided in terms of methodology by 19 agreements with District staff in the context 20 of ONRW meetings. When you selected the 21 methods you would use in your trend analysis of 22 inflows to the Park, did you know that your 23 work would be used for purposes of a lawsuit 24 against the District? DEPOSITION OF WILLIAM W. WALKER, JR. 818 1 MR. HARRISON: Object to the 2 form. Object also on any grounds of legal 3 relevancy. 4 A. I don't consider the task of setting water 5 quality standards for Park inflows any more or 6 less important or requiring any higher degree 7 or lower degree of regulation or as compared 8 with the task of analyzing the data for trends. 9 Q. But the method does change from one task to the 10 other, does it not? 11 A. The method? 12 Q. Yes. 13 A. The method for calculating flow-weighted 14 concentrations that we just discussed does not 15 change. 16 Q. Well, for example, in the calculations for the 17 ONRW process, we just went over and that 18 doesn't apply in the analysis employing the 19 Seasonal Kendall test, right? 20 A. The analysis or the objectives of the ONRW 21 process was setting water quality standards. 22 That has a different objective as compared with 23 the trend analysis. 24 Q. I know you had done some earlier preliminary DEPOSITION OF WILLIAM W. WALKER, JR. 819 1 analysis of various data. When did you 2 actually start work on the trend analysis which 3 is reflected in Exhibit 7 of your July 3, 1990 4 draft? 5 MR. HARRISON: Object, counsel. 6 If you mean by start work writing this report 7 or first looking at the raw data that 8 ultimately ended up in this report, I think 9 that needs to be specified. 10 Q. Can you identify a date when this project, 11 the trend analysis reflected in the July 3 12 draft, commenced? 13 A. I started analyzing the data in early 1990. To 14 separate a trend analysis out from my becoming 15 initially familiar with the data, graphing it, 16 displaying it, looking at the variations, 17 looking at the sampling regime and so forth, it 18 would be very difficult to determine a date 19 when I started the trend analysis. 20 Q. Could you identify when, for example, you 21 selected the methods that, the basic methods 22 you would use in your trend analysis, for 23 example, I'm using simple arithmetic averages 24 on dates when no discharge occurred through any DEPOSITION OF WILLIAM W. WALKER, JR. 820 1 structure? 2 A. I don't recall what date I made that decision. 3 Q. How many ONRW meetings did you attend before 4 commencing work on the discrete task you can 5 identify as the trend analysis leading to the 6 July 3 draft? 7 A. As I indicated in my last answer, I started 8 analyzing the data in early 1990 and it is 9 difficult for me to come up with a particular 10 date at which I started this task, this 11 particular task of trend analysis, and it is 12 therefore difficult for me to come up with a 13 number of ONRW meetings that I had attended 14 prior to that date. 15 Q. Did you discuss with district personnel which 16 method of calculating flow-weighted average 17 values would better reflect total releases to 18 Shark River slough? 19 A. I believe that we had a discussion on that 20 topic in the first or the second ONRW meeting, 21 and I also believe that if you calculate it 22 using either method for the entire period of 23 record, that the values will be similar whether 24 you use the daily flows only or whether you use DEPOSITION OF WILLIAM W. WALKER, JR. 821 1 the centered two-week flows for calculating the 2 flow-weighted mean. 3 Q. Did you discuss this issue with District 4 personnel in the context of performing a 5 Seasonal Kendall test on the data? 6 A. No. It was in the context of setting standards 7 for Park inflow points. 8 (Off the record discussion.) 9 Q. I think I wandered off a previous topic we were 10 discussing. 11 We talked about the inclusion of 12 hydrologic variables in the Series C analysis. 13 Did you consider variables other than the three 14 that we can see you used at certain points in 15 your work: Rainfall, elevation and flow? 16 MR. HARRISON: Asked and answered 17 yesterday or Friday. 18 Q. I only recall that we discussed one other, but 19 when you were outlining the work you were doing 20 in the trend analysis, did you consider 21 inclusion of other hydrologic variables? 22 A. Not to my recollection. 23 Q. Doctor Walker, in -- well, first I'd ask you to 24 -- you have Exhibit 17 before you. Could you DEPOSITION OF WILLIAM W. WALKER, JR. 822 1 also put Exhibit 16 before you? 2 On pages subsequent to the first 3 page of Exhibit 16 on the top left-hand corner 4 we see draft 9-6-90. Is that the date on which 5 this draft was generated? 6 A. I don't know for sure. It could have been. 7 Q. Do you believe to it be approximately accurate? 8 MR. HARRISON: Objection. This 9 has been asked and answered continuously. 10 Q. As opposed to believing it was really in 1920 11 or something? 12 A. I believe it to be correct within a month. 13 Q. Within a month. Could it be off as much as a 14 month? 15 A. Perhaps -- I believe the date to be 16 approximately accurate to within two weeks. 17 Q. To whom did you distribute this draft of your 18 report, the version that is Exhibit 16? I do 19 recall I asked you that question about other 20 drafts, but I don't believe I asked on this 21 one. 22 A. I don't recall exactly but I believe that one 23 recipient was Dennis Helsel. I believe I also 24 sent a copy of to Environ Corporation and to DEPOSITION OF WILLIAM W. WALKER, JR. 823 1 the Justice Department. There may have been 2 others but I don't recall. 3 Q. Did you receive any technical comments back 4 from Helsel? 5 A. Yes, I did. 6 Q. Were those oral or in writing? 7 A. Those were oral. 8 Q. Do you recall what they were? 9 A. I seem to recall that having received a message 10 on my phone answering machine from Doctor 11 Helsel that he had no problems or no comments 12 on this final version of the report. 13 Q. To your knowledge, did Doctor Helsel know that 14 this report was going to be finalized and 15 submitted to the court? 16 A. I do not know whether Doctor Helsel was aware 17 of that. I was not even aware of that. 18 Q. When did you first become aware that your trend 19 analysis report would be submitted to the 20 court? 21 MR. HARRISON: Objection, counsel. 22 Awareness obviously would have had to come from 23 trial counsel. I don't see why it is relevant. 24 MS. AHEARN: I think it is very DEPOSITION OF WILLIAM W. WALKER, JR. 824 1 relevant that a scientist knows who is going to 2 be the judge of his work. 3 MR. HARRISON: Counsel, it is 4 attorney-client privilege. It is not that 5 relevant. 6 Q. When did you become aware that your work was 7 going to be submitted to the court? 8 MR. HARRISON: Objection. If he 9 can answer without disclosing conversations 10 from litigation counsel, you may do so. 11 Q. I'm not even asking who told you or what they 12 said, just the date. 13 MR. HARRISON: Fine, and if he can 14 answer your question without disclosing 15 confidences relayed by litigation counsel, then 16 he may do so, counsel. 17 MS. AHEARN: Even if confidential, 18 the date of the confidential communication is 19 clearly discoverable and that's all I am 20 seeking here. 21 MR. HARRISON: I suppose you're 22 right. Go ahead. 23 A. I don't recall knowing that this document was 24 going to be submitted to the court until in DEPOSITION OF WILLIAM W. WALKER, JR. 825 1 fact it already was submitted in November of 2 1990. 3 Q. How did that fact come to your attention? 4 MR. HARRISON: Go ahead. If you 5 recall. I would ask you not to speculate. 6 A. Are you asking me how did I know that this 7 document had been submitted to the court? 8 Q. Yes. 9 A. I read it in the Orlando Sentinel. 10 Q. When you performed your trend analysis, 11 analyses, and prepared your report, what did 12 you understand to be the intended use of your 13 work? 14 MR. HARRISON: Objection, 15 counsel. Obviously he was hired by the 16 Department of Justice. For you to get into 17 specifically his understanding of a use cannot 18 help but disclose attorney-client 19 communications. 20 Q. Doctor Walker, when you prepared Exhibit 17, to 21 what audience did you believe you were writing? 22 MR. HARRISON: You may answer. 23 A. I was preparing a report for my client in much 24 the same way that I'd prepare a report for any DEPOSITION OF WILLIAM W. WALKER, JR. 826 1 of my other clients documenting the results of 2 research that had been done using a certain 3 database. 4 Q. But you did not know that this would be 5 submitted to a Federal Court at that time? 6 MR. HARRISON: Objection. It has 7 been asked and answered. 8 MS. AHEARN: Well, I suggest it is 9 so obvious, since he was retained by the 10 Department of Justice, it is silly to ask. I 11 am trying to find out which is which here. 12 MR. HARRISON: But you're trying 13 to figure out more than that, counsel. The 14 doctor has no actual understanding, nor should 15 he be expected to, whether portions or all or 16 any of a report go to a court. 17 I think counsel is well aware that 18 expert reports generally do not go to the court 19 because they are hearsay in and of themselves 20 and to try and ask Doctor Walker his 21 understanding of what he knows about a report 22 was going to go to a judge or justice or 23 anybody else is way beyond the doctor's 24 expectations and clearly you're getting into DEPOSITION OF WILLIAM W. WALKER, JR. 827 1 matters of attorney-client privilege that goes 2 so far beyond anything calculating to lead to 3 the discovery of admissible evidence that 4 you're wasting time. 5 MS. AHEARN: I find it remarkable 6 that you admit that the United States submitted 7 inadmissible hearsay in support of a motion for 8 summary judgment. 9 Q. If I could direct this next series of 10 questions, we can finish in the five minutes we 11 have left today. If I could direct your 12 attention to page 3 of Exhibit 16, I'd also 13 like you to refer to page 5 of Exhibit 17. 14 MR. HARRISON: Three of 16 and 15 five of 17, is that it? 16 Q. Yes. First on Exhibit 17, page 5, if you look 17 at the full paragraph about two thirds of the 18 way down on the page, the last sentence of that 19 paragraph and the first sentence of the bottom 20 paragraph, the dates reported in those two 21 sentences differ from the comparable text in 22 Exhibit 16, and I would refer you to lines 17 23 through 23 looking at the numbers down the 24 left-hand column. Could you please explain why DEPOSITION OF WILLIAM W. WALKER, JR. 828 1 those figures were changed from Exhibit 16 to 2 those that appear in Exhibit 17? 3 A. To my recollection these changes may have 4 reflected comments that I received from Bob 5 Johnson of the Everglades National Park 6 Research Center regarding changes in flow 7 distribution during this period. 8 Q. Did you do any additional sensitivity analyses 9 after the draft in Exhibit 16 was prepared? 10 A. I don't recall. 11 Q. Do you recall if you did any new sensitivity 12 analyses to change the new dates in terms of 13 operating changes? 14 A. There is nothing in my analysis that assumes 15 anything about operating changes or dates. 16 I don't recall whether I did any 17 analysis or any additional sensitivity analysis 18 between the draft in Exhibit 16 and the final, 19 Exhibit 17. I may have. 20 Q. Let me ask you just one more question and I can 21 finish this page five here on Exhibit 17. The 22 second sentence of the last paragraph refers to 23 releases made on a regular schedule. What do 24 you mean here by regular schedule, and just to DEPOSITION OF WILLIAM W. WALKER, JR. 829 1 tell where you I am coming from, can you relate 2 this to either the sampling schedule and/or 3 with regard to the seasons employed in the 4 Seasonal Kendall test? 5 MR. HARRISON: Objection. 6 Compound I believe. 7 Q. Sneaking two questions in for one. 8 MR. HARRISON: Go ahead. You may 9 answer if you can. 10 A. I believe in using the term regular schedule 11 for the early period I was referring to the 12 fact that releases were made as guided by water 13 elevations and season except under extreme 14 flood conditions and that would be as distinct 15 from later periods when releases were either 16 governed according to the flow-through plan in 17 which gates A, B, and C were left opened and as 18 distinct from releases that were made under the 19 rainfall-based plan after June of 1985 in which 20 releases were based upon antecedent rainfall 21 and evaporation from Water Conservation Area 22 3A. 23 Q. So by regular schedule, we're not talking about 24 something where you have a release on the third DEPOSITION OF WILLIAM W. WALKER, JR. 830 1 Wednesday of every month? 2 A. No. 3 Q. Is there any correlation between the sampling 4 schedule and the schedule of releases? 5 A. The sampling schedule was biweekly and release 6 is, according to my understanding, the gates 7 are generally operated or changed on a weekly 8 basis except when there is a flood risk. 9 Q. Is there any correlation between the time of 10 the sampling and the time of releases when 11 releases are made? 12 MR. HARRISON: Objection. You 13 mean statistical correlation or does the 14 District operate, have some correlation? 15 I think that needs to be 16 clarified. I'll object to form. 17 A. What do you mean by correlation? 18 Q. Is there any relationship between the sampling 19 schedule and the timing when releases are made? 20 MR. HARRISON: I still object to 21 form. 22 A. The sampling schedule was periodic; that is, 23 every two weeks. To my knowledge there was no 24 attempt in the sampling schedule to target DEPOSITION OF WILLIAM W. WALKER, JR. 831 1 specific release periods. It was operated on a 2 biweekly basis. 3 MS. AHEARN: Okay. 4 (Deposition recessed for the day 5 at 5:15 PM.) 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 DEPOSITION OF WILLIAM W. WALKER, JR. 832 1 2 Excerpt from Rule 30 (e): 3 Submission to Witness; Changes; Signing. When 4 the testimony is fully transcribed, the deposition shall be submitted to the witness 5 for examination and shall be read to or by him, unless such examination and reading are waived 6 by the witness and by the parties. Any changes in form or substance which the witness desires 7 to make shall be entered upon the deposition by the officer with a statement of the reasons 8 given by the witness for making them. 9 * * * * * * * * * * * * * * * * * * * * * * * * 10 11 I, WILLIAM W. WALKER, JR., have examined the above transcript of my testimony and it is 12 true and correct to the best of my knowledge, information and belief. Any corrections are 13 listed on the Errata Sheet. 14 ___________________________ Deponent's Signature 15 That on _______________, 1991, the 16 foregoing deposition was submitted to WILLIAM W. WALKER, the witness, for examination and was 17 read by the witness, at which time any changes desired were entered upon the deposition errata 18 sheet and that thereafter the deposition was signed by the witness before me. 19 20 _____________________________ 21 Notary Public in and for the Commonwealth of Massachusetts. 22 My Commission expires 23 24 DEPOSITION OF WILLIAM W. WALKER, JR. 833 1 COMMONWEALTH OF MASSACHUSETTS) ) 2 COUNTY OF SUFFOLK ) 3 I, Nancy L. Eaton, a Notary Public 4 within and for the Commonwealth of Massachusetts, duly commissioned, qualified and 5 authorized to administer oaths and to take and certify depositions, do hereby certify that 6 heretofore, on the date cited above, the witness personally appeared before me at the 7 above location and testified in the above captioned case; that the said witness was by me 8 duly sworn to testify to the truth, the whole truth and nothing but the truth, that thereupon 9 and while said witness was under oath, the deposition was taken down by me in machine 10 shorthand at the time and place therein named and was reduced to typewriting thereafter. 11 I further certify that the said 12 deposition constitutes a true record of the testimony given by the said witness. 13 I further certify that I am not 14 interested in the event of this action. 15 IN WITNESS WHEREOF, I have hereunto subscribed my hand and affixed my seal of 16 office this 21st day of February, 1991. 17 18 __________________________ Notary Public in and for the 19 Commonwealth of Massachusetts. 20 My Commission expires December 28, 1995. 21 22 23 24 DEPOSITION OF WILLIAM W. WALKER, JR.