108 1 DIVISION OF ADMINISTRATIVE HEARINGS DEPARTMENT OF ADMINISTRATION, STATE OF FLORIDA 2 SUGAR CANE GROWERS COOPERATIVE ) 3 OF FLORIDA; ROTH FARMS, INC.; and ) WEDGWORTH FARMS, INC., ) 4 ) Petitioners, ) 5 vs. )DOAH Case No. 92-3038 SOUTH FLORIDA WATER MANAGEMENT ) 6 DISTRICT, an agency of the State ) of Florida; et al., ) 7 Respondents. ) - - - - - - - - - - - - - - - - - x 8 FLORIDA SUGAR CANE LEAGUE, INC., ) UNITED STATES SUGAR CORPORATION; ) 9 and NEW HOPE SOUTH, INC., ) Petitioners, ) 10 vs. )DOAH Case No. 92-3039 SOUTH FLORIDA WATER MANAGEMENT ) 11 DISTRICT, an agency of the State ) of Florida; et al., ) 12 Respondents. ) - - - - - - - - - - - - - - - - - x 13 FLORIDA FRUIT AND VEGETABLE ) ASSOCIATION; LEWIS POPE FARMS; ) 14 W.E. SCHLECHTER & SONS, INC., ) and HUNDLEY FARMS, INC., ) 15 Petitioners, ) vs. )DOAH Case No. 92-3040 16 SOUTH FLORIDA WATER MANAGEMENT ) DISTRICT, an agency of the State ) 17 of Florida; et al., ) Respondents. ) 18 - - - - - - - - - - - - - - - - - x 100 Southeast 2nd Street 19 Miami, Florida March 16, 1993 20 9:15 a.m. - 6:00 p.m. 21 DEPOSITION OF ALLEN S. LEFOHN 22 VOLUME II - P.M. SESSION 23 Taken before RICHARD BURSKY, Registered 24 Professional Reporter and Notary Public in and for the State of Florida at Large, pursuant to Notice of 25 Taking Deposition filed in the above cause. 109 1 AFTERNOON SESSION 2 1:20 p.m. 3 BY MR. NETTLETON: 4 Q. Dr. Lefohn, would the addition of a 5 limiting nutrient to an oligotrophic water body cause 6 eutrophication in your opinion? 7 A. Well, my expertise is not eutrophication 8 in terms of the limiting nutrients and such. So I 9 feel like I am answering questions that are not 10 associated with my educational background. 11 Q. What in your opinion is the significance 12 of soil acid versus rainfall acid with regard to 13 ecological effects? 14 A. Could you rephrase that question? I think 15 I know what your point is but I would like to hear 16 that again in a different way. 17 Q. How do you attribute the effects of acid 18 through soil cycling or soil processes versus acid 19 from rainfall? 20 A. In some cases the high concentrations of 21 aluminum that were observed in the surface waters 22 occurred at the same time the pH depressions occurred 23 during runoff and so aluminum was acting like a 24 tracer that it was coming from the soil and not from 25 deposition. 110 1 And the acids in most cases, almost all 2 cases the soils that were we were looking at were in 3 forests and pHs were fairly low and aluminum was 4 observed in high concentrations at the same time. 5 And overland flow was playing a minor role in places 6 such as Hubbard Brook and other places that were 7 documented. 8 Q. Is there a difference though in the 9 ecological effect on the biota or the microbial 10 communities or what have you if the acid is from 11 deposition or from soil processes? 12 A. In some case is yes, in some cases no. 13 Q. Can you describe the different cases? 14 A. Yes. In some cases you have weak acids 15 associated with the organic acids coming down and 16 they have a different effect than strong acids such 17 as sulfuric acid. But sometimes in fact the strong 18 acids were coming off as Dr. Krug had found due to 19 natural processes. 20 A. They were not just, in other words, 21 sulfuric acid coming off into a water body was just 22 not attributable to sulfuric acid falling out of the 23 sky. 24 THE WITNESS: Did you get that? 25 Q. I don't know if I got it. 111 1 A. H2SO4 is the sulfuric acid. 2 Q. Right. 3 A. The paradigm was that if it is sulfuric 4 acid it is impacting your lakes, the sulfate that is 5 in your lakes and the acid that is in your lakes 6 comes from the atmosphere, comes from man-made SO2 7 reacting to form sulfuric acid. However, what Dr. 8 Krug found, it is possible to actually get strong 9 acids through natural processes going off into the 10 water body. It was just not weak inorganic acids 11 that were running off. 12 Q. Is there a difference in effect depending 13 on the source of the acid? Again, maybe I am 14 misunderstanding your answer. 15 A. My answer was that sulfuric acid will 16 probably have a different effect than the weak 17 organic acid just on its ability to titrate. So 18 effectively the sulfuric acid though is just not 19 attributable to man-made activities. In other words, 20 from man you get sulfuric acid or nitric acid. From 21 the natural processes you also get sulfuric acid plus 22 weak organic acids. Therefore there is a difference 23 potentially between the two acids but to simply say 24 one acid is coming from man and other acid is coming 25 from natural is incorrect. 112 1 Q. I understand that. I am not sure whether 2 you still answered my question. Is there a 3 difference in effect to the biota depending on the 4 source? 5 A. I am not sure I understand the question. 6 Q. Is there any difference in the effect on 7 the biota from the source of the sulfuric acid or is 8 sulfuric acid sulfuric acid? 9 A. Sulfuric acid is sulfuric acid, that's the 10 point I was trying to make. 11 Q. From what we have discussed this morning 12 and your educational background in chemistry and all, 13 I am curious as to what it is in your background and 14 your education and your experience which you consider 15 qualifies you to be an expert on the effects of the 16 various chemical compositions on, essentially the 17 ecological effects -- 18 MR. BLANK: I object to the form of the 19 question. 20 Q. Can you tell me what you consider your 21 qualifications are to provide opinion testimony in 22 that regard? 23 A. Yes. I am highly experienced in the 24 characterization of deposition data, wet deposition 25 data and some of the bulk data that I have looked at, 113 1 the linkages to biological effects is coming from the 2 fact that I ran the ecology branch in Corvallis, 3 Oregon and during that time my responsibility was 4 working with the biologists in dealing with the 5 effects component and relating physical chemistry to 6 that. 7 Since then I have worked very closely with 8 scientists from all over the world on the ecological 9 side and the vegetation effect side concerning what 10 exposure elicits an adverse effect, and consequently 11 my responsibility is not just the characterization 12 but actually taking it all the way to the exposure 13 response equation with the help of those individuals 14 who have actually done the fumigation experiments in 15 the laboratory. 16 And I have actually designed the exposure 17 regimes hand in hand with the researchers who 18 implemented the actual exposure of the crops. 19 Q. Am I correct then that your qualifications 20 with regard to ecological effects derive from what 21 you have just described as your experiences with 22 working with the biologists and so forth as opposed 23 to your educational background per se in chemistry? 24 A. The chemistry part of the characterization 25 and such comes from my training in chemistry, physics 114 1 and math. The biological experience comes from 2 working with the researchers themselves and 3 essentially designing experiments for them, as well 4 as my appreciation of running or participating in the 5 creating of national programs in Washington, D.C. 6 Q. What research if any has the Animal 7 Ecology Lab in Oregon done with regard to nutrient 8 responses? 9 A. Let me give you a quick history. 10 The animal ecology branch no longer exists 11 and the laboratory no longer exists. EPA decided to 12 abolish its theme laboratories about the time I left 13 EPA. 14 However, the Corvallis Environmental 15 Research Laboratory was heavily involved in 16 eutrophication, the facility itself was a body and 17 within that there were components of which one was 18 the national ecological research laboratory which I 19 ran and the ecology lab. The lab itself to date was 20 very involved in eutrophication activities. 21 Q. What was your role in those eutrophication 22 activities? 23 A. While I was in Corvallis? 24 Q. Yes. 25 A. None. 115 1 Q. What about before or after you were there? 2 A. Before I was in Corvallis my role, as I 3 indicated earlier, was working in Washington, D.C. to 4 put together the program that involved working with 5 the laboratory people in Corvallis while I was in 6 Washington, D.C. as well as working with the 7 laboratory people in Las Vegas who were worried about 8 the helicopters and aircraft and platform material. 9 Q. That is what you described earlier 10 concerning the EPA survey? 11 A. Yes, that's correct. 12 Q. Have you done any other work other than 13 that with regard to eutrophication research? 14 A. Since then, no. 15 Q. What about before then, had you done 16 anything regarding eutrophication research? 17 A. I don't believe so. It was when I was in 18 Washington, D.C. 19 Q. When were you first contacted about 20 providing services for this particular litigation? 21 A. I believe September or October of 1992. 22 Q. Who contacted you? 23 A. Counsel. 24 Q. Who specifically? 25 A. I think it was, Bill Earl was my first 116 1 contact. 2 Q. Are you under contract with Mr. Earl's 3 firm to provide services? 4 A. Yes. 5 Q. Is that a written contract? 6 A. Yes. 7 Q. How are you being compensated? 8 A. On an hourly basis. 9 Q. What is your hourly rate? 10 MR. BLANK: Go ahead, you can answer. 11 THE WITNESS: My hourly rate? 12 MR. BLANK: Yes. 13 A. My hourly rate is $125 an hour. 14 Q. You charge the same thing for testimony as 15 your research activities? 16 A. No. 17 Q. What is the difference? 18 A. Double during the time I am here. 19 Q. So you are charging 250 an hour for 20 testimony? 21 A. When I am in a deposition. 22 MR. NETTLETON: Off the record. 23 (Discussion off the record) 24 MR. NETTLETON: Back on the record. 25 BY MR. NETTLETON: 117 1 Q. Had you ever worked for Mr. Earl's firm 2 prior to this engagement? 3 A. No. 4 Q. Are you being paid by the firm of Peeples, 5 Earl & Blank or by their clients, the Florida Sugar 6 Cane League? 7 A. The checks are from counsel. 8 Q. What have you been asked to do in this 9 case? 10 A. I have been asked to -- there are two 11 phases. One phase is to look at what has been done 12 to date and draw some conclusions concerning what I 13 have evaluated and the second phase has to do with 14 rendering an opinion concerning what the deposition 15 loading is in Southern Florida, in South Florida. 16 Q. When you say look at what has been done, 17 what are you referring to? 18 A. Evaluate the material that has been 19 provided to me. 20 Q. Have you been asked to do anything else 21 other than the two things you have just mentioned? 22 A. Not that I can recall. 23 Q. What materials have been provided to you? 24 A. I have seen the SWIM Plan -- there was a 25 list that was provided to you recently, so I am doing 118 1 that from memory right now. 2 Q. All right. 3 A. So pardon me if I do not remember each of 4 the numerous items. 5 There was the SWIM Plan, there was the 6 draft paper by Grimshaw that I gather is under peer 7 review, at least I received it as the paper had gone 8 off to be reviewed. There is a piece by Walker that 9 was done I believe for the Justice Department that 10 was an evaluation of the Everglades National Park 11 data that the District had collected. 12 There was a memo from I believe Grosser 13 dealing with quality assurance, there was an Appendix 14 E that Mr. Grimshaw had written I believe in 1989 15 that appeared to be supporting material to the 16 information that went into the SWIM Plan. There were 17 several tables and such concerning the calculations 18 of how the average concentrations for the twelve 19 sites were calculated and the flow rated mean were 20 calculated. 21 There may be other additional things I 22 have just not remembered at this point that were in 23 the list. 24 There were a couple of publications also. 25 MR. NETTLETON: Mark that, please. 119 1 (A letter dated March 1, 1993 from Mr. 2 Blank, was marked Lefohn Deposition Exhibit 2 for 3 identification) 4 BY MR. NETTLETON: 5 Q. Showing you what has been marked as 6 Exhibit 2, is that the list of documents contained in 7 that letter that you are referring to? 8 A. Yes. 9 Q. Other than those documents set forth on 10 there, have you been provided any additional 11 materials? 12 A. Yes. Recently I received a diskette from 13 counsel that had been sent from the District to 14 counsel that contained some volume information that 15 was associated with some of the samples collected at 16 some of the sites. These were data that we had asked 17 for previously. 18 Q. Is that the information that is referred 19 to in the final paragraph of the letter that the 20 League's counsel was awaiting receipt of? 21 A. Yes, I believe so. 22 Q. When did you receive that? 23 A. Around March 5. 24 Q. When did you receive the other items set 25 forth in the letter? 120 1 A. The item A is a paper I received from 2 Bowersox about six weeks ago. 3 The Grimshaw '89 I probably had received 4 about two months ago. 5 The paper by Grimshaw, Shih and Gatz may 6 have been about a month ago. Grosser was probably 7 about a month. And the NADP/NTN data summary 8 information I have had for a long time and that was 9 just in my records, get that from the disk. 10 The disk rain information may be in 11 February, early February; the Surface Water 12 Improvement -- the SWIM Plan itself probably in 13 November and the Walker piece probably also in 14 November. 15 Q. With regard to the SWIM Plan item, I 16 believe the reference there is to the supporting 17 information document. How many volumes of the SWIM 18 Plan did you receive? 19 A. Three. 20 Q. Other than those items and the information 21 you just recently received on March 5, is there any 22 other information or materials that have been 23 provided to you? 24 A. I have been provided a lot of other -- not 25 a lot of other but other material which I am not 121 1 necessarily dependent upon for drawing conclusions. 2 Q. What other materials were provided? 3 A. Several reports from the USGS concerning 4 rain chemistry, one dated 1965, the other dated 1980 5 or 1981. 6 I have been provided -- are you asking 7 that others have provided to me associated with this 8 case? 9 Q. Right. 10 A. That people from all over the country or 11 people just associated with the case itself? 12 Q. Well, let me break it down. First what 13 were you provided by the Peeples Earl & Blank law 14 firm with regard to this case? 15 A. There were several appendices associated 16 with the report, but I can not remember the exact 17 title of the report, basic summary information of 18 data, not necessarily data but analysis and reports 19 that others had done in the past over the last 20 several years, some of them District reports, some of 21 them government reports from USGS, various reports. 22 I am sorry I can't be more specific at this time. 23 Q. What did the reports deal with? 24 A. Some dealt with water chemistry itself, 25 surface water chemistry. Some of the reports dealt 122 1 with the quality assurance program, components of the 2 quality assurance program. 3 I have received the actual data for the 4 USGS three month comparison study for colocated 5 samplers for the 1992 period. 6 The other appendices and such, it is very 7 difficult for me to remember right now all of the 8 details of all that information. But most of that I 9 did not depend on, those other things that I have not 10 cited here I have not depended on for drawing my 11 conclusions to date. 12 Q. You refer to these as appendices to a 13 report. What report are you referring to? 14 A. A report that was done by, for counsel and 15 they were appendices of xeroxes of reports and such 16 that others provided, others had done. 17 Q. Who prepared the report? 18 A. I believe it was Hydrologic Associates. 19 Q. Do you know when that report was prepared? 20 A. I'm thinking 1991. 21 Q. Can you tell me what that report was 22 about? 23 A. It was a summary of the different types of 24 monitors that the District had used to date, the 25 types of quality assurance protocols that were being 123 1 followed, some of the problems that were associated 2 with the various monitoring approaches, some of the 3 techniques. 4 Q. When you refer to types of monitors, what 5 are you referring to? 6 A. Different types of wetfall collectors or 7 dryfall collectors and different designs. 8 Q. When you refer to the QA/QC what are you 9 referring to? 10 A. Some of the quality assurance/quality 11 control protocols that were being followed in terms 12 of how the samples would be preserved. 13 Q. Do you recall what the report stated about 14 that? 15 A. No. It was general. My recollection is 16 there were no bottom lines concerning strengths, 17 weaknesses of different techniques. It was more or 18 less a documentation of what was going on. 19 Q. Were the reports that were appended to it 20 cited as the authority for whatever they were stating 21 inside the report? 22 A. Not necessarily. To me, it appeared to me 23 that some of the appendices were just what had been 24 done in certain areas. For example, the Walker 25 report was one of the appendices. 124 1 Q. The problems with monitoring, what 2 problems did they identify? 3 A. I think the data collection problems in 4 terms of the efficiency was one issue that may have 5 been raised. The preservation of the samples may 6 have been another. 7 Q. Any others you can recall? 8 A. No. I can not recall any others at this 9 time. 10 Q. Did Hydrologic Associates reach any 11 conclusions in their report? 12 A. I believe they did not. I believe it was 13 mostly a narrative of what was going on. 14 Q. Do you know what time period was covered 15 by the report? 16 A. I believe it was through 1990. And I 17 don't recall, other than I know that the District has 18 samples going back to 1974, but I don't recall how 19 far it went back. 20 Q. Were you asked to provide any comments on 21 the report? 22 A. No. 23 Q. Other than the Hydrologic Associates 24 report and the appendices thereto, were you provided 25 any additional information or documents other than -- 125 1 A. From counsel or from other sources? 2 Q. From counsel. 3 A. There was quality assurance information 4 that was provided to me, of which I passed on to an 5 associate of mine. 6 Q. Who is your associate? 7 A. Well, associates meaning a peer which is 8 Dr. Krupa. 9 Q. Why did you pass it to him? 10 A. Because Dr. Krupa was a subcontractor to 11 my company, A.S.L. & Associates, to evaluate the 12 quality assurance and the quality control. 13 Q. Can you spell Krupa? 14 A. K R U P A. 15 Q. The QA/QC that you were provided, was that 16 a specific document or program? 17 A. No, it was whatever documentation we could 18 find which included the Grosser piece, the memo that 19 was provided to us sometime the beginning of the 20 year. I think it was drafted in January 1993 but it 21 was provided to us in February. So it is whatever 22 information that had been sent to us, some of the 23 appendices that I referred to plus Grosser. 24 Q. Did this include any information 25 concerning laboratory analysis QA/QC? 126 1 A. I don't recall. For this specific case my 2 responsibility was to identify the sources and just 3 send them on to Dr. Krupa. 4 Q. What do you mean by identify the sources? 5 A. What I meant, what material was available 6 I passed on. 7 Q. Anything else that was provided to you by 8 counsel? 9 A. At this point I can not think of anything 10 else. 11 Q. What other reports or materials or 12 documents did you receive from third parties with 13 regard to your work in this case? 14 A. I have -- I have received annual reports 15 from the National Acid Deposition Program on a 16 routine basis and that material resides in my office. 17 It is public information, the documents are public 18 information distributed by the government. 19 I had asked and received from Dr. Bowersox 20 a copy of his paper which you have a copy of also. 21 I have received raw data from governmental 22 sources of the NADP sites in Florida, national acid 23 deposition sites in Florida. 24 And I have received a tape from Electric 25 Power Research Institute on some of its wet 127 1 deposition data. 2 Q. What kind of tape? 3 A. I am sorry, it is a magnetic tape of data. 4 Q. What type of information is on that tape? 5 A. Wet deposition data, it is associated with 6 the anions and cations which include orthophosphate. 7 Q. Is that included in the EPRI data base we 8 discussed earlier? 9 A. Yes, that's correct. 10 Q. Any other information you received from 11 third parties? 12 A. I have received recently as late as late 13 last week references to two peer review papers that 14 deal with the determination of wet chemistry in terms 15 of how the calculations are made for volume weight 16 concentration. 17 Q. Do those papers provide any novel ideas 18 that haven't been followed in the past concerning 19 volume weighting? 20 A. What it did was supply me with additional 21 information in the peer review literature for support 22 for some of the conclusions that I drew concerning 23 the, why one uses volume weight concentration. 24 Q. Anything else you received from third 25 parties? 128 1 A. Yes. There is one other thing I did 2 receive from counsel. There is a paper by Hendry and 3 Brezonik and Edgerton and they have, that paper 4 summarizes the deposition data for Florida for the 5 late 1970s. 6 Q. Any additional materials that you have 7 received with regard to this case? 8 A. Nothing additional I can think of at this 9 time. 10 MR. BLANK: Just for clarification 11 purposes, because I don't want there to be any 12 confusion in terms of what he has received, and I 13 don't know that you have covered it in your 14 questions, but I know you have received the 15 District's rainfall data and that was referred to in 16 the first paragraph of the letter and not one of the 17 inserts. 18 A. Yes, I am sorry. Yes, I am sorry, I did 19 receive a diskette of those data. 20 Q. I was referring to things that are not 21 stated in the letter that has been marked as Exhibit 22 2. 23 A. The item F on the list was not the 24 chemistry data. I believe it was associated more -- 25 there may have been some chemistry data on there but 129 1 the reason that diskette was sent to us was at the 2 time it was thought that was the volume data but in 3 fact it turned out to be for every day the rainfall 4 amount for just a few sites. I did receive on a 5 separate diskette a copy of the data starting from 6 1974 through I believe 1992 for some of the sites. 7 Q. And that's what is referred to in the 8 opening paragraph of the letter? 9 A. Which sentence are you referring to, this 10 sentence -- 11 Q. The District's rainfall data for South 12 Florida. 13 A. Yes. 14 Q. Any other materials you have received from 15 either counsel or third parties relating to your work 16 on this case? 17 A. Not that I can recall at this time. 18 Q. Has any information been conveyed to you 19 orally as opposed to by documents that you have 20 utilized for your work in this case? 21 A. No, I am fairly certain I have depended 22 upon written materials. 23 Q. Other than the materials we have discussed 24 that have been provided to you, have you reviewed any 25 other materials of any nature for purposes of your 130 1 work on this case? 2 A. I have reviewed my own Kriging papers. 3 Q. Which specific ones? 4 A. The White House report, the State of 5 Science Report No. 7 as well as the paper on page 17, 6 Lefohn, Knudsen and McEvoy 1988 as well as the paper 7 on page 17 by Knudsen and Lefohn, The Use of Spatial 8 Statistics to Characterize Regional Ozone Exposures. 9 Q. Why were you reviewing your Kriging papers 10 for purposes of your work on this case? 11 A. For evaluating the Kriging procedures that 12 were followed by the District. 13 Q. Did you reach any conclusions as to the 14 appropriateness of the Kriging procedures being 15 followed by the District? 16 A. Yes. 17 Q. What is your conclusion? 18 A. That the procedures were inappropriate. 19 Q. Were inappropriate? 20 A. Yes. 21 Q. What is that based upon? 22 A. It was based upon several things. 23 One, the data that were used were 24 inappropriate to be used in the Kriging. 25 Number two, it was uncertain that the 131 1 model itself was appropriate. 2 And number three, the way the data 3 themselves were characterized was inappropriate. 4 Q. I will come back to that in a second. 5 Are there any other materials that you 6 have reviewed with regard to your work that you have 7 done on this case? 8 A. I don't recall any. 9 Q. With regard to your opinion that the 10 District's Kriging procedure was inappropriate, you 11 stated first that was based upon the fact that the 12 data was inappropriate, is that correct? 13 A. That's correct. 14 Q. Why do you say the data was inappropriate? 15 A. Because the samples that were measured 16 came from different years. In some cases there were 17 three samples or two samples collected entirely for a 18 particular data point. In other cases they, the data 19 may have been gathered for longer periods of time but 20 not at the same periods of time as other data that 21 were used. For example, data may have been gathered 22 for the period 1976 through 1982 or 1983 and stopped 23 and for another site the data would have been 24 gathered from 1985 through -- into 1989. 25 And that is just an inappropriate way to 132 1 use the data to Krig. 2 Q. Why is that inappropriate? 3 A. Because the appropriate way would have 4 been to Krig -- to calculate the deposition and to 5 Krig the deposition for each year and use all the 6 data that were available by each year, and then get 7 values for each year into the area which they were 8 interested in which was the WCAs and to come up with 9 an annual value and then average that if they were 10 looking for an average value between the periods of 11 1974 into 1989. 12 Instead, what appeared to be done, and I 13 say appeared because the documentation was not very 14 strong in the SWIM report, was that they 15 calculated -- they collapsed all the data for twelve 16 different monitors, some are, for example, Homestead 17 had only a very few number of samples, it may have 18 been three or four samples. Another site had only 19 nine samples and they treated them in an equivalent 20 way which means they weighted them in evaluating 21 saying two or three samples gave a quick snapshot of 22 what was going on at that site over a nine or ten or 23 twelve year period. 24 And for another site they may have had 199 25 samples over the same time period which perhaps 133 1 better represented what was going on over several 2 years. So they essentially Kriged data, tracing the 3 data at that time at each point as if it were 4 weighted in an equal way and that was incorrect. 5 Q. What is inappropriate about the model 6 itself? 7 A. It was unclear why universal Kriging was 8 used. Ordinary Kriging is a technique that usually 9 is applied and that if there tends to be a trend in 10 the data or something like that, then they go to a 11 different model of Kriging. 12 And we attempted to get from the District, 13 obtain from the District the information that would 14 have allowed us to get a better feel for why 15 universal Kriging was used. And to the best of our 16 knowledge at this time the documentation just is not 17 there. It may be there but we have not received it. 18 We have requested it. Hence -- 19 Q. What -- 20 A. Go ahead. 21 Q. I am sorry. 22 What is the difference between universal 23 Kriging and normal Kriging? 24 MR. BLANK: Do you mean ordinary Kriging? 25 Q. Ordinary Kriging, excuse me. 134 1 A. As such. A different model is used and 2 the person that I work with on this is Dr. Knudsen. 3 He is a coauthor and he is the Kriger. And Dr. 4 Knudsen and I have discussed the approach that was 5 used by the District and based on his conclusions I 6 am now reflecting that in what I am saying now in the 7 difference between ordinary and universal Kriging. 8 He is an actual modeler that uses the data, does the 9 Kriging physically. 10 Q. So your on conclusion is based upon your 11 discussions with Dr. Knudsen and his opinions 12 concerning the District's Kriging procedures? 13 A. No. You asked me what the difference was 14 between ordinary Kriging and universal Kriging and 15 what I am saying, what I am responding to that 16 question is that Dr. Knudsen would be more qualified 17 to tell you the difference in the models. 18 What Dr. Knudsen has indicated to me is 19 that it is unclear why the District selected 20 universal Kriging over ordinary Kriging but that one 21 point is not the basis of my conclusion concerning 22 inadequacy of the Kriging effort. 23 Q. The third item you mentioned was the way 24 the data itself was characterized was inappropriate. 25 What do you mean by that? 135 1 A. The data, to use simple arithmetic means 2 is just totally inappropriate to determine what the 3 deposition was. In my opinion, the District was 4 attempting to determine what the deposition was in 5 the WCAs. To simply use the arithmetic means is 6 incorrect. 7 Q. What should they have used? 8 A. If they wanted to do what they did, first 9 of all, I have already made the point about, it is 10 unclear whether the data would even support the use 11 of Kriging because of the fact of the different time 12 frames that the data were gathered. 13 If the data were -- if you could use the 14 data, what should have been done is either calculate 15 the deposition at each site, supposing at each of the 16 twelve sites, supposing you had good coverage for 17 each year, and then Kriged that deposition into the 18 WCAs, or what some people have done is use the volume 19 weighted concentrations and the amount of rainfall 20 from data bases that are available and that doesn't 21 necessarily mean the rain gauge values at each -- you 22 could only use the rain gauge values at each monitor 23 but you have airports and such, and you essentially 24 do Kriging of both. 25 You Krig the rainfall amount, rain gauge 136 1 value and you also Krig the volume weight 2 concentration and you multiply the two to get the 3 deposition in the area, in the WCAs. 4 Q. What precisely is the source of your 5 information as to the Kriging procedures that you are 6 assuming were followed by the District? 7 A. The SWIM Plan, the, I want to say 8 supporting document, what I referred to, supporting 9 information document. 10 Q. Other than discussing the universal versus 11 ordinary Kriging with Dr. Knudsen, did you discuss 12 any of the other aspects of your Kriging opinion with 13 him? 14 A. Yes. Dr. Knudsen and I discussed what I 15 discussed right here, which is the inadequacy, the 16 potential inadequacy of the data base because of the 17 not overlapping of the monitoring collection at the 18 same time and the fact that they were Kriging just 19 the concentration instead of the deposition. 20 Q. Were the factors -- let me back up. Did 21 Dr. Knudsen review the SWIM Plan document as well? 22 A. Yes. 23 Q. Were these opinions concerning the 24 inadequacy of the District's Kriging procedure first 25 arrived at by Dr. Knudsen or yourself? 137 1 A. I drew some preliminary conclusions for 2 background purposes and then sent the information to 3 Dr. Knudsen once I obtained it. 4 Q. Describe to me what happened after that. 5 A. He independently went through three 6 volumes of the SWIM Plan, the supporting information 7 document was the piece that he primarily drew his 8 conclusions on. But there was a small part of the 9 appendix which is the third document that had a 10 contradictory statement in it that talked about 13 11 sites used for Kriging and he noted that at the time, 12 whereas the SWIM Plan second volume referred to the 13 12 sites. 14 Q. Where is Dr. Knudsen located? 15 A. Montana Tech in Butte, Montana. 16 Q. Is he being compensated for his review of 17 this material? 18 A. Yes, he is. 19 Q. Who is paying him? 20 A. I am paying him. He is under contract to 21 me. When I say me, A.S.L. & Associates, my 22 corporation. 23 Q. Why did you contract with Dr. Knudsen? 24 A. Dr. Knudsen is a Kriging expert in that he 25 actually does the Kriging, does the mathematics, is 138 1 probably one of the experts, one of the most highly 2 thought of experts in the world and I believe he was 3 also cited in the SWIM Plan itself. Dr. Knudsen and 4 I have worked very closely together since about 5 1984-85 on Kriging air quality information. 6 As a mathematician statistician which he 7 is, I respect his opinion and have used him as an 8 independent resource. 9 Q. Has Dr. Knudsen ever been involved in 10 Kriging with regard to phosphorus concentrations? 11 A. No, to the best of my knowledge, no. 12 Q. Have you ever been involved in anything 13 like that? 14 A. The Kriging of phosphorus, no. 15 Q. Are you relying upon the opinions of Dr. 16 Knudsen to support your own opinions? 17 A. No. 18 Q. Do you consider Dr. Knudsen more qualified 19 in the area of Kriging than yourself? 20 A. He is very qualified in the area of the 21 actual use of the data in the modeling. He is more 22 qualified than I am in terms of the actual Kriging. 23 But I consider myself highly qualified in terms of 24 the interpretation. 25 Q. How do you evaluate whether one Kriging 139 1 method is superior to another? 2 A. In terms of the actual mathematics and the 3 model used, that is -- I personally depend on a 4 mathematician similar to Dr. Knudsen basically to 5 judge whether or not the assumptions built into the 6 Kriging allow for the use of ordinary Kriging versus 7 universal Kriging versus any other. 8 Q. So you are relying on Dr. Knudsen for that 9 particular aspect of your opinion? 10 A. As to whether the model itself was 11 appropriate, that's correct. 12 Q. Is it possible to validate or verify which 13 Kriging method is more appropriate? 14 A. By looking -- there are a lot of ifs on 15 this. If the data base itself were appropriate and 16 could be used, then it would be possible to validate 17 which model would be most appropriate, if the data 18 base itself were appropriate to be used as input. 19 As an alternative we attempted to locate 20 background information from the District to give us 21 some ability to judge that without having to rerun 22 the data and that data was not available, at least 23 not made available to us. 24 Q. Has such a verification ever been done 25 with regard to phosphorus in rainfall to your 140 1 knowledge? 2 MR. BLANK: I object to the form of the 3 question. 4 THE WITNESS: Am I supposed to respond? 5 MR. BLANK: You can go ahead and answer. 6 THE WITNESS: All right, I am sorry. 7 A. The actual Kriging of phosphorus data, of 8 deposition data, I am unaware of other than what was 9 attempted by the District. 10 Q. What was attempted by the District to your 11 knowledge? 12 A. To my knowledge, they were attempting to 13 estimate what the deposition in the WCAs were. 14 Q. And that is what you are criticizing, what 15 your opinion is, their procedures were inappropriate, 16 is that right? 17 A. To get to the estimate of the phosphorus 18 deposition, total phosphorus deposition in the WCAs, 19 I believe that the information that was generated can 20 not scientifically be supported. 21 Q. Are you aware of whether the District or 22 any other party, party generically, any other person 23 or entity, has attempted to validate or verify what 24 particular Kriging method is more appropriate for 25 Kriging phosphorus in rainfall? 141 1 A. I am unaware of anyone else validating the 2 various models. 3 Q. Is there a general consensus in the 4 scientific community regarding the appropriate 5 Kriging method for phosphorus in rainfall? 6 A. There is a consensus on the general 7 approach concerning the Kriging of wet deposition 8 data. 9 Q. What is that consensus? 10 A. The consensus I have described earlier is 11 using deposition information and using that to create 12 or using volume weighted concentration coupled with 13 rainfall information and Kriging both to then 14 determine deposition in the area that you are 15 interested in. 16 Q. You had indicated previously in part of 17 your testimony that you had attempted to obtain some 18 additional data from the District. What data are you 19 referring to? 20 A. Is your question regarding the Kriging or 21 regarding the wet deposition? 22 Q. I can do both. Let me back up and ask a 23 more general question. Is there any information that 24 you have requested that you have not received? 25 A. Yes. 142 1 Q. Tell me what that is. 2 A. We have requested background information 3 concerning the Kriging in terms of the models that 4 were used, how the models were fitted, what the 5 results of those fittings were in terms in the form 6 of graphics or tables. We have not received 7 anything. 8 Q. Do you know if any such things exist? 9 A. I do not know. 10 Q. Anything else you have requested but have 11 not received? 12 A. Yes. 13 Q. What? 14 A. Several times on December 18 counsel 15 requested from the District the volume information or 16 the amount of rainfall information associated with 17 each sample that was analyzed in the District's data 18 base for the 12 sites. 19 What we were sent or what counsel was sent 20 which was passed on to me was the diskette that I 21 previously described that is listed here that was a 22 1.8 megabyte file that contained every day's amount 23 of rainfall for a few sites and the data were, could 24 not be used. They were inappropriate. 25 Following that, counsel did request again 143 1 the information I have just described and 2 approximately ten days ago I received a diskette that 3 contained the volume information associated with the 4 bulk samplers as well as the event samplers for the 5 period of I believe 1974 through 1978 and then it 6 skips a year and a half, it goes from July 1978 until 7 the beginning of -- I am sorry, until 1988. Let me 8 back off from 1974 to the middle of 1988 and then it 9 skips a time period and then picks up again in 1990. 10 So therefore there is a void for what the 11 volume information was for those data that were 12 collected for that year and a half period. 13 Q. The '89-'90 time period somewhere? 14 A. Half of '88, all of '89. 15 Q. Do you know if that data exists anywhere? 16 A. I do not. Counsel has requested that 17 information. 18 Q. How would you use that information if you 19 receive it, if it exists? 20 A. If it were to exist, I would calculate the 21 volume weight concentrations. 22 Q. For what purpose? 23 A. To basically get an idea of what the 24 concentrations would have looked like on an annual 25 basis in a bucket. That could then be turned over, 144 1 turned into deposition data for those sites. 2 Q. Has the accuracy of the Kriging approach 3 for phosphorus ever been validated? 4 MR. BLANK: I object to the form of the 5 question. 6 You can answer it. 7 THE WITNESS: I am thinking of the answer. 8 (Pause) 9 A. Assuming an appropriate technique had been 10 used, no, not that I am aware of. 11 Q. Other than the three items that you 12 mentioned that you are basing your opinion on 13 concerning the inappropriateness of the District's 14 Kriging procedure, is there anything else that you 15 are relying on for that opinion? 16 A. I am sorry, there are several opinions 17 that I have talked about. 18 Q. As I understand it, you testified that in 19 your opinion the District's Kriging procedure was 20 inappropriate and that opinion was based upon 21 essentially I guess three sub-opinions, that the data 22 was inappropriate, the model itself was inappropriate 23 and the way the data itself was characterized was 24 inappropriate. Is there anything other than those 25 three items that support your opinion that the 145 1 Kriging procedure used by the District was 2 inappropriate? 3 A. No. That is pretty much what I based my 4 opinion on, but I would like to state that I did not 5 say that the model that was used was wrong, I said it 6 was impossible to determine whether it was even 7 correct, whether you should use ordinary Kriging 8 versus universal Kriging. 9 Q. What would you need to know to determine 10 which model is more appropriate? 11 A. You would have to actually have done 12 deposition calculations and doing that you then 13 looked at the period of record for each of the 14 monitoring sites and then determine whether or not 15 your data are sufficient to even Krig, whether you 16 even have enough data sites to even want to Krig to 17 determine the deposition in the area that you are 18 looking for. 19 Q. Are you saying you would volume weight the 20 rainfall phosphorus concentrations and then Krig 21 them? 22 A. There are two ways to do that. You 23 calculate the deposition at each site and use the 24 volume weight concentration with rain gauge reads 25 from other sites as well as the monitoring sites and 146 1 Krig both of them to get your deposition. 2 Q. And then multiply them? 3 A. Concentration with rain gauge amount. 4 Q. By using the approach you just described, 5 wouldn't that be multiplied errors, then? 6 A. Yes. But you have errors in everything 7 you do. I mean, it is a matter of doing it both 8 directions. 9 The concentration times amount carries an 10 error in the estimation for the concentration as well 11 as the estimation for the rain gauge reading and then 12 there is an error that you can carry over with 13 Kriging just using the deposition itself. And using 14 both ways you get an idea of what the relative errors 15 are, comparing one to the other. 16 Q. Why is it that you use volume weighted 17 concentrations? 18 A. Well, my opinion of why you need volume 19 weight concentration goes all the way back to 20 Walker's piece where he used the Everglades National 21 Park Research Center data that he obtained from the 22 District for the period late 1987 into early April 23 1989. 24 What he found was there was an inverse 25 relationship between the amount of rainfall and the 147 1 concentration of total phosphate. What I mean by 2 that is when you had heavy rains or there was heavy 3 rains the concentration was less than when you had 4 trace samples and the concentration was high. 5 What that says is that if you use an 6 arithmetic mean which the District did, you would 7 essentially end up with a different value than one 8 would get if you just wanted to calculate which is 9 what they were trying to do, find out what the 10 concentration was in the bucket over the period of 11 time they were interested in, from 1987 to 1989. 12 The only way you can do that is with the 13 volume weighted type concentration, because what it 14 does is give weighting to this problem associated 15 with the inverse relationship. 16 Q. Didn't the District provide information in 17 the SWIM Plan of both arithmetic as well as volume 18 weighted concentrations? 19 A. No. 20 Q. I assume what we just discussed, your 21 opinion concerning the Kriging procedure, is at least 22 one of the items covered under what you were asked to 23 do with regard to what has been done to date and your 24 conclusions, is that correct? 25 A. Could you, would you please? There are a 148 1 lot of ifs and buts in that. 2 Q. I will rephrase. 3 When I asked what you had been asked to do 4 by counsel you provided me with two things, one was 5 to look at what has been done to date and to draw 6 conclusions from that and secondly was opinions on 7 deposition rate. 8 A. Deposition amount. 9 Q. Other than the Kriging procedure opinion 10 that you have just expressed, what other opinions 11 have you reached with regard to what has been done to 12 date? 13 A. The table that is in the supporting 14 information document is incorrect. 15 Q. What table are you referring to? 16 A. The table that summarizes the flow 17 weighted mean column is that, that column does not 18 relate to whatever flow weighted means was supposed 19 to have meant. 20 Q. What you do you mean, it doesn't relate? 21 A. In the Grimshaw calculation or piece that 22 I referred to as an Appendix E in one of the reports, 23 there were several tables in there that were labeled 24 volume weighted concentration. They were in fact not 25 volume weighted concentration in those tables. 149 1 And that information was used in that 2 piece to carry over to form the table that appeared 3 in the supporting document. 4 Q. Are you saying that the table in the 5 supporting information document of the SWIM Plan 6 contains numbers which purport to be flow weighted 7 concentrations which in your opinion are not actually 8 flow weighted concentrations? 9 A. I had inferred that flow weighted mean was 10 the same as volume weighted concentration. I then 11 went back to the Appendix E and -- I am sorry. Let 12 me be careful, not the Appendix E in the SWIM Plan 13 but the appendix in the Grimshaw piece. And those 14 tables are labeled volume weighted but in fact they 15 are not the volume weighted values and that 16 information was then transferred into the information 17 in the SWIM Plan. 18 Q. Why do you say they were not, what are you 19 basing that on that they were not volume weighted 20 numbers? 21 A. Because volume had nothing to do with 22 those samples. 23 Q. How do you know that? 24 A. By looking at the data that were collected 25 from the District by Walker in his piece, he actually 150 1 listed in one of the tables, it may have been Table 1 2 or Table 2, the events data, meaning the two week 3 samples or the four week samples. 4 So I knew what the concentrations were of 5 total phosphorus in wet and dryfall. 6 By taking that information it is possible 7 then to compare that to the tables that were 8 constructed by Grimshaw in his Appendix E. And in 9 fact what he had, what Grimshaw had in his left 10 column was the average concentration, arithmetic 11 average concentration for individual months. 12 For example, he took all the January 13 numbers, all the February numbers, March numbers, no 14 matter what year it was, and he then had a table that 15 said for January the arithmetic average was value X 16 for two samples, that is an example, for February it 17 may have been three samples across several years, et 18 cetera. 19 What Grimshaw then did based on that is 20 reconstructed the numbers in the SWIM plan he sample 21 weighted. There were three samples in January, 22 multiplied it by the arithmetic mean and added to 23 that the two samples that may have been in February, 24 added it, and then divided by the total number of 25 samples at that site. 151 1 That is what is in the column that is 2 labeled flow weighted mean which is not volume 3 weighted concentration. 4 Q. Do you know whether the samples were in 5 effect the, consisted of similar volumes or not? 6 A. They did not consist of similar volumes at 7 least for the Everglades National Park site which 8 listed the volumes. 9 I will answer that in two ways. 10 What Walker used was rain gauge readings 11 from Homestead Air Force Base. That is different 12 than the volume, because he could not obtain from the 13 District the volume information. 14 Second, I have since received the volume 15 information recently from counsel that came from the 16 District through counsel and those volumes are 17 different. 18 Q. Different from what? 19 A. I think your question was do they vary 20 from sample to sample. They are different. 21 Q. Let me see if I have this right now. 22 You expressed the opinion that the table 23 in the SWIM Plan supporting information document 24 contains flow weighted mean numbers or purports to 25 which are taken from the Grimshaw document which in 152 1 your opinion do not actually reflect flow weighted or 2 volume weighted concentrations, is that correct? 3 A. My assumption -- I will answer that 4 question -- but my assumption is that flow weighted 5 means were the same as volume weighted means, given 6 that they were not volume weighted means. 7 Q. Just so I am clear, you are basing that on 8 the fact that you have obtained what the actual 9 volumes were in the samples? 10 A. No. 11 Q. Explain to me again how you know that they 12 were not volume weighted. 13 A. Because by using the data in the table 14 that Walker had, I was able to go to the table that 15 Grimshaw had for the Everglades National Park in that 16 particular Appendix E. 17 Given the data from November 1987 through 18 April, beginning of April of 1989 I was able to 19 reconstruct the Grimshaw table for the Everglades 20 National Park. 21 And what that means is the left column in 22 the Grimshaw table had something called phosphorus 23 concentration. That concentration that was listed 24 under the months of January, February, March, et 25 cetera, through December was the arithmetic mean of 153 1 the number of samples for each of those months across 2 several years. What Grimshaw then did was take the 3 number of samples that were made up in that 4 arithmetic mean and simply multiplied the two 5 together and then add it and divide by the total 6 number of samples. The volume had no role in any of 7 that calculation. 8 Q. Do you know whether the Walker data and 9 the Grimshaw data came from the same source, same 10 sampling? 11 A. Yes. 12 Q. Same data base? 13 A. Yes, because I was able to reconstruct the 14 Grimshaw table from using the Walker data. 15 I am sorry, when I say Walker data I mean 16 the data that Walker summarized that he had gotten 17 from the District, the exact number of samples, the 18 exact number of samples per month, everything matched 19 what was in the Grimshaw table. 20 Q. And Walker also reported volume in his? 21 A. No. He used the Homestead Air Force Base 22 rain gauge reading. 23 Q. Are you likewise then using the volume 24 that you determined from the Homestead rainfall gauge 25 to determine what the volume is? 154 1 A. I haven't done anything. 2 Q. You have not reanalyzed volume weighting 3 the numbers then, is that correct? 4 A. That is correct because we are waiting for 5 the data to come from the District for the incomplete 6 year and a half. 7 Q. You have not done that with regard to the 8 rest of the data, though? 9 A. You mean other sites? 10 Q. No, with regard to the park for other than 11 the year and a half that you are missing? 12 A. Could you rephrase that, please? 13 Q. Let's move on. I may come back to that. 14 What other opinions have you reached with 15 regard to what has been done to date? 16 A. It may be getting late and I may be 17 getting tired, but I am thinking, I am thinking I may 18 have covered all my conclusions concerning what has 19 been done to date. I would like to reserve the 20 possibility over the next period a light bulb goes 21 off and I am able to say yes, I forgot something. 22 Q. Okay. 23 With regard to the opinion we just 24 discussed concerning the information in the 25 supporting information document, are you relying on 155 1 anything other than what you just described, 2 comparison of the Walker numbers and the calculation 3 of the numbers in the Grimshaw report? 4 MR. BLANK: I object to the form of that 5 question. 6 Q. Other than what you just discussed, are 7 you relying upon anything else with regard to your 8 opinion that the information in the supporting 9 information document is incorrect concerning flow 10 weighted or volume weighted, assuming those are the 11 same? 12 A. There were also errors in that table, 13 there were transcription errors going from the 14 Grimshaw piece into the table in the SWIM Plan, there 15 was at least one site where the flow weighted mean 16 was incorrectly transcribed. 17 The data had been calculated in the 18 Grimshaw piece but somehow was incorrectly placed in 19 the table. 20 The other thing is that the Kriging used 21 arithmetic means, not the flow weighted means. 22 Q. I am sorry, say that again. 23 A. The Kriging used arithmetic means 24 according to the best information we had available in 25 the SWIM Plan, according to the SWIM Plan. 156 1 Q. Does that relate back to your first 2 opinion on the District using inappropriate Kriging 3 procedures? 4 A. Yes. 5 MR. NETTLETON: Please mark this as the 6 next exhibit. 7 (A document entitled The South Florida 8 Water Management District's Methods for the 9 Collection and Interpretation of the Quality and 10 Quantity of Rainfall, 27 September 1989, was marked 11 Lefohn Deposition Exhibit 3 for identification) 12 BY MR. NETTLETON: 13 Q. I have handed you what has been marked as 14 Exhibit No. 3. Is this the Grimshaw paper you have 15 been making reference to? 16 A. It is. 17 MR. NETTLETON: Mark that as No. 4. 18 (A document entitled Surface Water 19 Improvement and Management Plan for the Everglades, 20 pages 158 through 162, was marked Lefohn Deposition 21 Exhibit 4 for identification) 22 BY MR. NETTLETON: 23 Q. I am also showing you what has been marked 24 as Exhibit 4 which I will state for the record is an 25 excerpt from the supporting information document of 157 1 the Everglades SWIM Plan dated March 13, 1992. 2 Specifically it contains pages 158 through 162. 3 Is the information or the table that you 4 have been referring to with regard to your second 5 opinion included in this Exhibit No. 4? 6 A. 4 is, yes, one of the tables that I 7 referred to. 8 Q. Dr. Lefohn, with regard to Exhibit 4, can 9 you point out to me precisely which table it is you 10 are saying has the incorrect information concerning 11 flow weighted or volume weighted mean? 12 A. On Exhibit 4, the column that has been 13 labeled flow weighted mean -- 14 Q. Which page are you on, sir? 15 A. Sorry, page 160, Figure 28. The flow 16 weighted mean column is not volume weighted 17 concentration. 18 Q. What table are you referring to in Exhibit 19 3 where this information was purportedly transferred 20 from? 21 A. I am sorry, in Exhibit 3 there were 22 several tables from which the table was purportedly 23 transferred from. Table 4, there is no page number, 24 is the table that I had used to reconstruct from the 25 Walker data how that, how Table 4 was created. 158 1 Q. What other tables from Exhibit 3 were 2 involved in the transposition over to Exhibit 4? 3 A. Well, Table 1 is the summary table of many 4 of the tables in Exhibit 3. You will notice on page 5 Exhibit 4, Figure 28, that the CLE site has listed 6 flow weighted mean of .033. 7 Q. Right. 8 A. If you look at Table 1 on the figure -- I 9 am sorry, on Exhibit 3, under CLE it is .186. 10 Q. That is the transcription error you were 11 referring to? 12 A. That is correct. 13 Q. Do you know what CLE station is? 14 A. Clewiston site. 15 Q. Have you reviewed the data to determine 16 which is the correct number? 17 A. Neither of them are correct. If you are 18 asking me if that was an error that went from the 19 Table 1 on Exhibit 3 over to Table 28, Table 1 is the 20 value that I -- the value in Table 1 was a value I 21 was able to reconstruct from the tables in the back 22 of Exhibit 3. 23 Q. I know we are all speculating at this 24 point but it obviously appears that the CLE site from 25 Table 1 on Exhibit 3, assuming it was transferred 159 1 over, the number that was included in Exhibit 4 is 2 actually for the one above it, B 50. 3 A. B 50. 4 Q. Is that correct? Is that what it looks 5 like? 6 A. You are speculating. It appears to be a 7 transcription error. 8 MR. NETTLETON: Please mark this as the 9 next exhibit. 10 (A document entitled Rainfall Total 11 Phosphorus Concentrations and Loadings in Everglades 12 National Park, prepared for Environ Corporation and 13 US Department of Justice, by William W. Walker, Jr., 14 August 1989, was marked Lefohn Deposition Exhibit 5 15 for identification) 16 MR. NETTLETON: Off the record. 17 (Thereupon, a brief recess was taken, 18 after which the following proceedings 19 were had) 20 MR. NETTLETON: Back on the record. 21 BY MR. NETTLETON: 22 Q. Dr. Lefohn, have we handed you what has 23 been marked as Exhibit No. 5? 24 A. Yes. 25 Q. Is that the Walker paper you were 160 1 referring to previously in your testimony? 2 A. Yes. 3 Q. Can you tell me what you utilized from the 4 Walker paper or you relied upon to reach your 5 conclusion? 6 A. Page 6, Table 1. 7 Q. Can you tell me how this table was used by 8 you to determine that the information contained in 9 the SWIM Plan was incorrect? 10 A. What I did was I looked at Table 4 of what 11 I am calling the Grimshaw piece. 12 Q. That is Exhibit No. 3? 13 A. That is correct. And I segregated the 14 data by month independent of the year. 15 For example, if you look at June on Table 16 4 and you then refer to Table 1 of Exhibit 5, under 17 June there were only, on Table 1, there are only two 18 entries, June 8, 1988 and June 29, 1988. 19 In the fourth column there is a total that 20 is listed on that table for June 8, 1988. It is .004 21 ppm. And on June 29, 1988 it is .10. 22 I added the two values together and took 23 the simple arithmetic mean of the two values. The 24 addition of the two is .014 parts per million. I 25 divided that by 2 which is .007 parts per million. 161 1 On Table 4 the entry in the first column 2 under total phosphorus concentration is .007 which is 3 the arithmetic mean of those two values. 4 Likewise, if you segregate the samples by 5 month you can get the left column, the information in 6 the left column in Table 4 from Table 1. 7 Q. Okay. 8 A. Now given that information, if you then, 9 now referring to Table 4 in Exhibit 3, if you 10 multiply, for example, looking at the month of 11 January, .054 which is the arithmetic mean of three 12 samples, times 3, and you add to that .025 times 1, 13 and you add to that .034 times 3, and you keep doing 14 that all the way through December, you are 15 essentially adding together 12 products of each one 16 representing sample size times the average 17 concentration. 18 I then get a sum from that total and I 19 divide that by the sample size. 20 That value for the Everglades National 21 Park -- 22 Q. Can I stop you there for a second? 23 A. You may. 24 Q. When you say you divide by the sample 25 size, what do you mean? 162 1 A. I mean if you go to the third column it 2 says sample size and add. 3 Q. To the number of samples? 4 A. Yes. 5 You then -- there are 23 samples and if 6 you then go to Table 1 of Exhibit 3 you will see the 7 entry under sampling station ENP of .034 in the right 8 column under volume weight. That is not a volume 9 weighted concentration, that is a sample weight 10 concentration. 11 Q. Am I correct then that if you take the 12 numbers on Table 4, multiply the first column by the 13 third column, add up the totals of each of those and 14 divide that sum by the total number of samples you 15 come up with .034? 16 A. Yes. That's a sample weighted mean. 17 I did that same calculation for each of 18 the other sites, not using the Walker paper but using 19 just the tables in the Grimshaw piece. 20 Q. I am sorry, without using the Walker? 21 A. There is no reason to use the Walker paper 22 because the Walker paper only dealt with Everglades 23 National Park. 24 So there are additional tables there that 25 I think are before Table 4 and after Table 4 that you 163 1 could then reconstruct the information that was 2 summarized in Table 1 under the column volume weight 3 which is not volume weight. 4 Q. What else did you use the Walker paper 5 for, if anything? 6 A. What I used the Walker paper for also for 7 Table 1 was to convince myself that the District data 8 that it had given me for rain gauge values for every 9 day was useless information, i.e., I couldn't use it 10 for my purposes which was to calculate the 11 precipitation weighted mean. And that was because I 12 did not know when the start date was for each sample. 13 Table 1 has an item A, footnote A -- on 14 the first line, 11/17/87 -- and that footnote states, 15 11/01/87 starting date assumed. 16 I believe Walker had the same problem I 17 had which was he did not know when the sample started 18 and so he assumed a starting date. 19 Based on his assumption and based on the 20 fact that I had already reached that conclusion that 21 assumptions would have to be made, I decided that the 22 data that had been given to me by the District could 23 not be used to calculate precipitation weighted means 24 based on the daily rain gauge values. 25 Q. Going back to Exhibit 3 and Table No. 1 -- 164 1 A. Yes. 2 Q. -- have you done any analysis to determine 3 how the numbers reached under the column labeled 4 Simple have been reached or calculated? 5 A. Yes. Simple -- 6 Q. Before you answer that, let me strike the 7 question and ask you a preliminary question. 8 Is what you just described as occurred is 9 it reflected under the volume weighted column under 10 Table 1, would you consider that a simple arithmetic 11 mean? 12 A. No. 13 Q. Okay, then I will repeat my original 14 question which is have you done any analysis to 15 determine how the number under the sample column has 16 been calculated. 17 A. Yes. The sample column or -- some 18 preliminary analysis using the data that had been 19 given to me from counsel that had come from the 20 District's data set appears to be simple arithmetic 21 mean. And that is in fact what shows up in Exhibit 22 4, Figure 28 under the column, simple arithmetic 23 mean, which is the third column of Table 4. 24 Q. Describe for me the difference between a 25 simple arithmetic mean and a sample weighted mean. 165 1 A. A simple arithmetic mean simply takes all 2 the numbers and adds them up independent of trying to 3 weight a particular month more than another, it 4 simply treats all the data together, it simply, you 5 go from start to finish, whereas if you had an 6 average that occurred in the month of January, you 7 had 15 samples in that month, you would weight that 8 month of January more than some other month if 15 9 were the highest number of samples that were measured 10 in that month. 11 So you are giving more weight to those 12 months that add more samples. That has nothing to do 13 with volume. 14 Q. Now that it has been a little bit longer, 15 let me go back to a question I asked you before. 16 Other than the two areas we have 17 discussed, have you reached any other opinions 18 concerning what has been done to date by the 19 District? 20 A. Nothing more than I have already discussed 21 today. 22 Q. Are you relying upon anything in the 23 Walker document, Exhibit No. 5, other than Table 1 24 that we have talked about? 25 A. To reach the conclusions that I have 166 1 already talked about? 2 Q. Yes, sir. 3 A. No. 4 Q. In Exhibit No. 4 which is the excerpt from 5 the SWIM Plan supporting information document, is the 6 information concerning Kriging that you referred to 7 with regard to your first opinion contained in this 8 excerpt? 9 A. Yes, most of it. 10 Q. Can you tell me where that is? 11 A. I believe page 162. There is no page 12 number on the page previous to 162 which I assume is 13 page 161. In the last paragraph it starts out with 14 saying a statistical procedure was employed. 15 Q. And that through the first full paragraph 16 on 162 is the information you recall referring to? 17 A. Yes. And additional -- perhaps the next 18 paragraph also, Table 16. 19 Q. Is it your understanding that the 20 information contained in Table 16 is the result of 21 the Kriging exercise employed by the District? 22 A. I believe so. 23 Q. With regard to Exhibit 3, the Grimshaw 24 1989 paper, other than the sections or tables we have 25 discussed, are you relying upon anything in that 167 1 document for the conclusions you have expressed other 2 than that which we discussed? 3 A. Yes. There are additional tables which I 4 referred to but did not name that were before and 5 after the Everglades National Park summary table. 6 Q. And that is with regard to the, I think 7 you mentioned that if you went through each of the 8 tables for each of the stations you could add them up 9 and do the same calculations -- 10 A. That's correct. 11 Q. Anything other than the tables in the 12 Grimshaw 1989 document that you relied on? 13 A. I would like to take a second. 14 Q. Sure. 15 (Pause) 16 A. I believe that is it. 17 Q. The second area that you indicated counsel 18 had asked you to express an opinion on was opinions 19 on, I originally had written down deposition rate in 20 South Florida but then I think you said that should 21 be deposition amount. 22 A. Yes. 23 Q. What is the distinction you are making 24 between rate and amount? 25 A. Amount would be the total value that one 168 1 would get over the year. A rate is a function of 2 time of some sort. And I am just saying the average 3 of a year or something like that would be the amount 4 of deposition. 5 Q. When you say the amount of deposition, are 6 you talking essentially volume of water or -- 7 A. No. The kilograms per hectare of the 8 total phosphorus that occurred in the area. 9 And then from that you get a loading value 10 which is your -- which is in Table 16 in Exhibit 4. 11 Q. In deposition are you talking about both 12 wetfall and dryfall? 13 A. Yes, total. 14 Q. Have you formulated an opinion concerning 15 the deposition amount of total phosphorus in South 16 Florida? 17 A. Not at this time. 18 Q. Can you tell me what you have done toward 19 that goal? 20 A. It is a matter of identifying the or 21 attempting to obtain the volume weighted 22 information -- the volume information from the 23 District which as I indicated previously we have some 24 of those data. 25 In addition, I have attempted to locate 169 1 other data sets that may be some help in the future. 2 Q. What other data sets are you attempting to 3 locate? 4 A. Well, I have located the National Acid 5 Deposition data for phosphorus for the six sites in 6 Florida. 7 Q. That is the NADP? 8 A. NADP. 9 Q. Any other data sets? 10 A. Just a copy of an article that appeared by 11 Hendry, Brezonik and Edgerton that I already referred 12 to. 13 I do have some other data that relates to 14 the EPRI data base and some of the UAPSP data and I 15 have not looked at that yet. 16 Q. Have you formulated any preliminary 17 opinions at this point concerning the deposition 18 amount of total phosphorus? 19 A. No. 20 Q. The volume information you are referring 21 to is that year and a half of data, the '88 and '89 22 that you discussed earlier? 23 A. Yes, that is correct. 24 I would like to clarify something very 25 quickly. It was the other data set also from the 170 1 USGS that I have, the three months that I previously 2 referred to earlier today. 3 Q. Do you have any other data from the USGS 4 other than the 1992 three months of data? 5 A. I don't believe so. I don't think so at 6 this time. 7 Q. Do you intend to look at any earlier USGS 8 data? 9 A. If I can be convinced based on 10 recommendations from others concerning that the 11 quality of the data are adequate I would certainly 12 want to look at it if volume data were available. 13 Q. What have you heard concerning the quality 14 of the USGS data? 15 A. Well, we are talking about a lot of USGS 16 data, some of it going back to 1965 and earlier than 17 that. The summaries were for '65. 18 It seems to have similar problems as other 19 data bases have suffered from concerning the early 20 type of monitoring that went on. There was no 21 indication in the '65 report there was any volume 22 information so even if the quality were okay, were 23 appropriate, I am not sure it would be useful. 24 Q. Is the three months that you have from 25 1992, does that include volume data? 171 1 A. Yes. 2 Q. Do you know what the quality of that data 3 is? 4 A. No. I did not work on those data because 5 we have no documentation associated with receiving 6 those data. We are waiting for that. 7 Q. Do you know when that is going to occur? 8 A. No. 9 Q. What is your opinion as to the quality of 10 the NADP data? 11 A. For what it was intended to do, I think 12 the data are adequate. The quality assurance 13 protocol was very rigorous. 14 Q. With regard to the NADP data, am I correct 15 that you will be looking solely at the 16 orthophosphorus parameters? 17 A. No. I may look at other things but it is 18 premature at this time. I have a full set of data 19 that came from the 12 sites -- six sites in Florida. 20 Q. What other aspects of the NADP data will 21 you be looking at? 22 A. I don't know. That's what I am saying, I 23 have the ions, cations and anions data and it is a 24 matter of looking at patterns of type of exposures 25 that occurred during those weekly samples. 172 1 Q. How would that assist you in formulating 2 your opinion on total deposition of total phosphorus 3 amounts? 4 A. It may not. Once again, that is only 5 orthophosphate that they measured so therefore to get 6 to the total phosphorus just may not be appropriate. 7 Q. Do you have an opinion as to quality of 8 the NADP data concerning orthophosphate? 9 A. There may be some question concerning the 10 degradation of the samples. 11 Q. What is the question? 12 A. In terms of the integrity of the data, was 13 the sample adequately preserved so that you did not 14 have problems with biological activities occurring 15 within the sample that resulted in less sample being 16 measured than was actually there at the beginning. 17 Q. Do you have any opinion on that question 18 at this point? 19 A. Not at this point. 20 Q. Do you intend to formulate an opinion? 21 A. Yes. 22 Q. What type of activities, I think you 23 described them as biological activities, would affect 24 the integrity of the data on orthophosphorus? 25 A. The actual biological activity I can not 173 1 define for you. However, there may be indications of 2 degradation of the sample and what I want to do is 3 talk with the investigators in Illinois who were 4 responsible for doing the analysis as well as the 5 quality control process to find out if they have 6 looked at the issue and what their conclusions are. 7 Q. Have you talked to anybody at, is it the 8 University of Illinois? 9 A. Yes, it is the University of Illinois. 10 Q. Have you talked to anybody concerning 11 this? 12 A. Concerning phosphorus analysis, yes. 13 Q. Who have you talked to? 14 A. Dr. Van Bowersox who sent me a paper. 15 Q. He is the one who authored the acid rain 16 paper? 17 A. Yes. 18 Q. What is his role with regard to the 19 analysis of the NADP data? 20 A. I don't know his official title but he's 21 very prominent in terms of how the data are analyzed, 22 how they are stored in the computer data base and how 23 adequate the data are for various types of analysis. 24 Q. How often have you spoken with him 25 concerning this matter? 174 1 A. Twice. 2 Q. When were those conversations? 3 A. Probably in November, in the early part of 4 my analysis and probably about four weeks ago or five 5 weeks ago when I asked for a copy of his paper. 6 Q. Have you had any correspondence with him 7 concerning this? 8 A. Just when he sent me the original set of 9 information concerning how they analyzed for 10 phosphorus. 11 Q. I am sorry, what did he send you? 12 A. The information on how they analyzed for 13 phosphorus from the EPA regulations. 14 Q. What form did that take? Was that an 15 EPA -- 16 A. Xerox copies of EPA material on the 17 protocols. 18 Q. Did he send you any other material? 19 A. Not that I can recall. 20 Q. Did he send a cover letter with the 21 material? 22 A. Yes. 23 Q. Did he state anything other than, 24 enclosed? 25 A. No, typical, enclosed please find. 175 1 Q. In any of your discussions or 2 correspondence did he ever express an opinion as to 3 the quality of the data concerning the 4 orthophosphorus? 5 A. Just that in many samples across the 6 United States they were measuring many times minimum 7 detectable levels at many sites across the United 8 States. 9 Q. Can you explain to me, what does that 10 mean? 11 A. It meant low values at many sites across 12 the United States. We didn't talk about any specific 13 location but just that the values were low in 14 general. 15 Q. I guess just so I am clear, are you saying 16 he stated that the levels of orthophosphorus shown in 17 the data across the United States are at around the 18 detection limit, is that -- 19 A. Many of the samples that they had gathered 20 at many of the sites across the United States were at 21 or below minimum detectable levels which means it 22 could be there but they weren't sure how much. 23 Q. Do you know what minimum detection levels 24 he was referring to? 25 A. Well, the absolute values seem to vary 176 1 over the years. It may be somewhere in the area of 2 two parts per billion. 3 Q. Is this data with regard to wetfall or 4 bulk? 5 A. Wetfall. What I was discussing with him 6 was wetfall. 7 Q. Are you aware of whether NADP has any data 8 concerning orthophosphorus or total phosphorus in 9 dryfall? 10 A. I believe there are data, some data. I do 11 not know what the data of record looks like. 12 Q. Have you discussed any of that with, what 13 is his name? 14 A. Van Bowersox, Dr. Van Bowersox. 15 Q. With Dr. Van Bowersox? 16 A. No, I did not. 17 Q. Have you spoken with anyone else at the 18 University of Illinois or anyone who is involved in 19 the analysis of the NADP data? 20 A. No. 21 Q. Have you sent any correspondence to the 22 University of Illinois or anyone involved in the 23 analysis of the NADP data? 24 A. Pertaining to this case, no. 25 Q. Did Dr. Van Bowersox discuss any other 177 1 matters with you that might be related to this case 2 other than the data we just discussed? 3 A. No. 4 Q. Are you able to form an opinion concerning 5 the deposition amount of total phosphorus with the 6 data you currently have for any particular time 7 period? 8 MR. BLANK: For what area, counsel, South 9 Florida? 10 Q. I am sorry, for South Florida, the 11 Everglades area. 12 A. I can not at this time calculate a number 13 that I would believe in. 14 Q. If the year and a half of data concerning 15 volume from half of '88 and '89 does not exist would 16 you be able to formulate an opinion concerning the 17 deposition amount of total phosphorus in South 18 Florida? 19 A. I could certainly do the calculations for 20 the time period of record where volume data were 21 there for various sites. I would feel more 22 comfortable comparing those data with an independent 23 set of data to see if similar patterns were observed. 24 And then based on that I could begin to draw 25 conclusions. 178 1 Q. What independent set are you referring to? 2 A. Data that may exist through the published 3 information from the Hendry data from the late 4 seventies. I do not know if it continued into the 5 eighties but at least the late 1970s time period, 6 taking a look at some of the wet orthophosphorus data 7 for NADP and then attempting to correlate that with 8 some of the absolute values that have been measured 9 at some of the six sites of data or some of the 12 10 sites of data, six wet-dry sites. 11 Q. Have you attempted to do that to date? 12 A. No. I was waiting for more data, more 13 volume data from the District. 14 Q. What other scientists besides yourself 15 have you discussed the issue of total phosphorus 16 amounts in deposition in South Florida with? 17 A. In general, with Dr. Krupa, Dr. Sagar 18 Krupa, in general. 19 Q. Anyone else? 20 A. Not deposition amounts. 21 Q. Can you tell me what your discussions with 22 Dr. Krupa have consisted of? 23 A. Namely the quality of the data. 24 Q. And Dr. Krupa, I believe you previously 25 testified, is someone you subcontracted to look at 179 1 the QA/QC of the data, is that right? 2 A. That's right. 3 Q. Has Dr. Krupa expressed any opinions to 4 you concerning the deposition amount of total 5 phosphorus in South Florida? 6 A. No. 7 Q. Have you expressed any opinions to Dr. 8 Krupa concerning that? 9 A. Only the problem of not being able to have 10 the volume information available. 11 Q. Have you discussed with him any 12 preliminary opinions based on the data that you do 13 have on hand? 14 A. No. Well, not the phase one that we 15 talked about, what we talked about previously. 16 Q. Phase one meaning your opinions concerning 17 the work that has been done so far by the District? 18 A. Yes. 19 Q. Has Dr. Krupa expressed any opinions to 20 you concerning that phase one area? 21 A. Of the questions that I was asking, no. 22 Q. Have the opinions that you expressed to 23 Dr. Krupa concerning that phase one area been 24 consistent with what you have told us here today? 25 A. Yes. 180 1 Q. If there are no data for volume of 2 rainfall in the same area as the sampling site for 3 total phosphorus, are you able to determine the total 4 phosphorus amount in deposition? 5 A. For that particular site? 6 Q. For that particular site. 7 A. If the volume data did not exist and the 8 rain gauge data did not exist then the volume 9 weighted or precipitated weighted values could not be 10 determined. 11 Q. How close in proximity must the data on 12 volume exist in order for you to be able to draw a 13 scientifically valid conclusion, in your opinion? 14 A. NADP focused on the precipitation as its 15 prime calculation to calculate a weighted value. It 16 then, if those data were not available, precipitation 17 rain gauge, it went to a volume value. If those data 18 were not available they did not bother to make the 19 calculation, include the sample. 20 Q. Does the NADP sampling sites all have 21 gauges at the same site the samples are taken from? 22 A. Yes. 23 Q. Is it your opinion that in order to 24 formulate a conclusion or opinion that you would need 25 data on, from gauges at the same site as the sampling 181 1 site? 2 A. Concerning calculating volume weighted or 3 precipitation weighted, that would be my opinion. 4 Q. If the District sampling sites, meaning 5 the South Florida Water Management District sampling 6 sites which produced the data concerning total 7 phosphorus concentrations do not also have 8 contemporaneous gauge readings or volume readings, is 9 there any means in your opinion to calculate volume 10 weighted total phosphorus in deposition in South 11 Florida? 12 A. That's a hypothetical question in that 13 they do have volume for some years in some samples so 14 it is possible for specific locations to calculate a 15 volume weighted concentration for a specific amount 16 of time. 17 The concern that I would have is are there 18 sufficient data covered in a particular year, set of 19 years to be able to use those data to estimate what 20 the deposition would be in the WCA area or any other 21 area of interest where you would interpolate using 22 the information. So you can calculate, I can 23 calculate, others can calculate the volume weighted 24 concentration for specific sites for specific time 25 periods at this time. 182 1 Q. But you haven't done that to date, is that 2 right? 3 A. That is correct, I have not done that. 4 Q. Do you agree with the NADP approach of 5 using volume if gauge measurements are not available? 6 A. Yes. 7 Q. Absent contemporaneous, what I mean by 8 that, in the same location, gauge or volume data, is 9 it possible in your opinion to use data from gauges 10 further away from the sampling site and how far away, 11 if possible? 12 A. It is risky at best because talking with 13 some of the people that live in this area, they talk 14 in terms of a large degree of variation in the amount 15 of rainfall over short distances. Therefore, to come 16 up with a formula that says X number of miles away is 17 acceptable and anything greater than that is not 18 acceptable is very risky. 19 Q. Have you ever done that with regard to 20 this or any other analysis you have conducted? 21 A. No. 22 Q. How does the taking of the volume versus 23 the gauge data bias the result, if it does? 24 A. The estimate from NADP is that the rain 25 gauge value represents a more complete data capture 183 1 than the volume. So therefore, assuming that 2 assumption or statement or given that NADP uses, if 3 you use volume you may be underestimating the amount, 4 the deposition that occurred during that event. 5 Q. The amount of the phosphorus? 6 A. Or any other chemical that you are 7 interested in. 8 Q. Do you agree with that assessment by NADP? 9 A. I think based on the analysis that they 10 have done, I do believe that. 11 Q. Do you have a time frame in mind of when 12 you intend to reach opinions concerning the second 13 phase of your work? 14 A. I would like to be finished sometime 15 during the summer, early summer. A lot of that is 16 contingent upon being just to being able to obtain 17 the information we have asked for in a timely 18 fashion. 19 Q. Have you been instructed not to formulate 20 any opinions based upon the data that is available 21 until you receive this additional year and a half of 22 data on volume? 23 A. No, that was my decision. 24 Q. In your opinion, how many stations would 25 be necessary to calculate a real phosphorus 184 1 deposition amount? 2 MR. BLANK: Could you repeat the question? 3 Are you saying areal? 4 MR. NETTLETON: Areal, sorry. 5 A. I don't have an opinion on that because 6 what that would depend upon would be how closely the 7 summarized data of each site as you compare site to 8 site fit the assumptions that are built into Kriging 9 and that from that you then do the variogram, go 10 through all the motions and from that you could 11 basically say how many sites could I take out and 12 still have something within a degree of confidence of 13 some defined level. So I don't feel comfortable in 14 giving you a certain answer at this point, it will be 15 variable according to the data. 16 Q. What assumptions are you talking about? 17 A. The assumptions have to do with the 18 continuity of the samples meaning you don't have big 19 jumps in values from station to station. 20 Kriging is based, was originally used for 21 mining purposes for purposes of core samples and 22 estimating what the concentrations were like at a 23 particular location. And consequently there were 24 certain assumptions built into the mathematics that 25 says this is a smooth transition, as you go from site 185 1 to site you don't have a low value, high value, low 2 value. 3 One of the assumptions is that it is 4 continuity, that it is continuous. 5 Another is that there is not necessarily a 6 trend in the data in terms, trend in the data has to 7 do with drift in the data and that is a mathematical 8 term the Krigers use and that's why they would go to 9 something like universal Kriging versus ordinary 10 Kriging. They look at the data points, they then 11 take a look at the pattern and then say do these 12 assumptions fit. 13 Q. Do you have an opinion of whether or not 14 using the data that may be available from the NADP 15 stations in Florida would be a sufficient number of 16 data points or sampling sites in order to reach an 17 opinion concerning the amount of total phosphorus in 18 deposition in Florida? 19 A. The amount of total phosphorus was not 20 measured by NADP so there has to be jumps or leaps of 21 faith to get there. 22 Q. Assuming they had measured total 23 phosphorus or change it to orthophosphorus. My 24 question is more directed to in your opinion are the 25 number of sampling sites sufficient. 186 1 A. No, those sites were spread out all over 2 the state. 3 Q. So in your opinion they would not be a 4 sufficient number? 5 A. By themselves. 6 Q. Can you tell me what you base that opinion 7 on? 8 A. That's based on the fact that you are 9 attempting -- it is similar to what we did out in the 10 United States when we were asked to estimate the 11 ozone levels across the United States. 12 We had samples in Utah and in Denver that 13 we were using to estimate what the values were in 14 Montana and Wyoming. 15 Obviously there was an air bound that was 16 equal to the magnitude of the value we were trying to 17 estimate. 18 The same situation would exist here, is 19 that you are talking about three WCAs that, I assume 20 that you are trying to determine what the area 21 loading is to those three areas. And that you have 22 sites up at Bradford and Austin Cary which is the 23 northern parts of the state, you have Cape Canaveral 24 and Quincy and Everglades National Park. 25 That is spread out throughout the entire 187 1 state and what you are doing is taking points way up 2 to the north and saying I am going to use those to 3 estimate down in the southern part of Florida. If 4 you have a density enough through the entire state 5 beyond the six sites perhaps those six sites coupled 6 with others could give you the information. By 7 themselves they would not be appropriate. 8 Q. How many sites would you need? 9 A. I think I answered that already in terms 10 of it depends on the data and the variability as you 11 look at the assumptions. There is not a quick 12 formula that says X number is what you need to do. 13 Q. If the data you collect from the existing 14 NADP sampling sites establish that the assumptions 15 corresponded or the numbers were relatively even, 16 would you then be of the opinion that you could use 17 those sampling sites, they would be sufficient to 18 determine total phosphorus deposition? 19 A. I wouldn't feel comfortable because you 20 would only be able to draw upon a couple of sites 21 that are in the middle and southern part of the state 22 to be able to draw conclusions because the further 23 you are away the less they are, those sites are going 24 to participate in your decision of what the 25 deposition is. 188 1 Q. Then it would seem to me that you must 2 have some opinion of how far apart they can be then 3 before you would feel comfortable using them. 4 A. No, because you are asking for a real 5 number. I can't give you a real number at this 6 point. 7 Q. We know the number that exists is too 8 much, the distance. 9 A. We have six sites. What I am saying is to 10 Krig those six sites to tell you what the world is 11 like in Southern Florida I believe would be 12 inappropriate, if those sites, those data are taken 13 by themselves. 14 Q. Do you believe in your opinion are the 15 sampling sites that exist within the District's 16 network sufficient, assuming you have the data that 17 is necessary, to form an opinion concerning the total 18 phosphorus deposition in Southern Florida? 19 A. Taken alone by themselves, I would feel 20 uncomfortable in depending just on those data. Even 21 if the data coverage, meaning the time period were 22 enough overlap that I could have seven or eight sites 23 taken over the same year, I could then -- and volume 24 was available, but the Grimshaw paper questioned the 25 volume from the District. That doesn't mean the data 189 1 are horrible and can't be used but it means you would 2 look to other data and compare the pattern from the 3 other data with the data from the District. 4 Q. I am not sure you answered my question 5 though. Are the number of sampling sites sufficient 6 in your opinion, assuming the quality of the data, in 7 order to reach a conclusion? 8 A. I believe the conclusion I reached and Dr. 9 Knudsen reached, given the lack of overlap, that we 10 have a great question of whether or not there are 11 enough data out there at any given moment to be able 12 to Krig. 13 In other words there was data starting in 14 '74, '75 for some sites that disappeared right then, 15 the Homestead base had three samples, I believe. 16 That certainly is not representative of that 17 location. 18 Q. Are you saying that even assuming you 19 receive this year and a half of data on volume, that 20 you do not believe you will be able to form an 21 opinion in any event concerning the total phosphorus 22 deposition amount in South Florida? 23 A. No, that is not what I am saying. What I 24 am saying is given the volume information, I can at 25 least calculate the volume weighted concentration for 190 1 sites for particular year. Based on that I can then 2 take a look and see where those sites are in 3 comparison to WCA and then make that decision whether 4 or not there are sufficient sites near the WCAs to 5 Krig into there. 6 Q. If you were to do that and then determine 7 that you could not Krig, how would you calculate the 8 areal total phosphorus deposition amounts? 9 A. It may or may not be possible to do it. 10 That's why I am looking at the other data sets to see 11 if there are other data available to be able to 12 complement the data that have been generated by the 13 District. 14 Q. When you say it may or may not, how might 15 it be possible? 16 A. How might it be possible? If in fact you 17 have patterns of exposure for specific years, given 18 different data sets that are similar to one another, 19 not necessarily in the absolute value but the 20 patterns of exposure, meaning certain areas are 21 higher than other areas and the absolute values may 22 differ but at least the indication is they are 23 higher, you then begin to say, well, maybe there is a 24 pattern here, being able to use the data, NADP data 25 for example, for orthophosphorus. 191 1 We do have some idea of what some of the 2 orthophosphate data look like within the District and 3 you begin to compare some of that information and you 4 begin to sort out more reliable data from less 5 reliable data. 6 It is a whole building up campaign, you 7 start from the bottom and work up until you finally 8 reach a conclusion whether or not you have sufficient 9 data to Krig. If you don't, you just don't do it. 10 Q. Is there any alternative to Kriging? 11 A. As I indicated at the very beginning, 12 there are other mathematical techniques but they 13 don't give you the air analysis associated with them, 14 the air uncertainty, associated with them. 15 Any interpolation technique you use is 16 only as good as the quality of the data you have. So 17 that is the key, the quality of the data both in 18 terms of believability of data as well as its spatial 19 extent. 20 Q. If after you receive whatever data you 21 believe you need you determine that you can not Krig, 22 do you intend to use any alternative method to 23 determine the total phosphorus deposition in South 24 Florida? 25 A. I don't know at this point. I am more on 192 1 trying to find what the data sets look like before 2 judging the direction I want to go. It is just too 3 early. 4 I am going to take a rest for a second. 5 Q. Sure. 6 (Thereupon, a brief recess was taken, 7 after which the following proceedings 8 were had) 9 MR. NETTLETON: Back on the record. 10 BY MR. NETTLETON: 11 Q. Dr. Lefohn, are you aware of any unique 12 aspects of Florida's rain processes that affect the 13 chemical analysis of its phosphorus content? 14 A. I am aware of the possibility of 15 phosphorus coming into the area from, possibility of 16 coming in from long-range transport. 17 Q. Is that unique to Florida? 18 A. It's perhaps unique to this part of the 19 United States. 20 Q. From where is phosphorus coming in? 21 A. One of the possibilities is that there is 22 a contribution from the Sahara. 23 Q. Anyplace else? 24 A. Most of what I have read deals with the 25 Sahara coming into this area, the possibility of 193 1 transport from some of the other parts of Florida 2 also. 3 Q. This area, I guess we are talking -- 4 A. Southern -- 5 Q. Southern Florida? 6 A. Yes. 7 Q. Are you aware or do you understand that 8 there is transfer of phosphorus through the air from, 9 say, the EAA to the EPA, if you will? 10 A. I believe I understand your initials and 11 where it is coming from and going to. Yes, I guess 12 you are asking me if I am aware of the possibility of 13 that. And, yes, I'm aware. 14 Q. Are you aware of more than a possibility, 15 that in fact that that is occurring? 16 A. Well, all that I can say is that in 17 regards to what I have read in terms of the 18 Everglades National Park statement from Grimshaw, et 19 al., it is very difficult to attribute what the 20 amounts are that are coming from local generation, 21 attributable to local generation versus transport. 22 Q. With regard to the transport from the 23 Sahara, is such long-term transport, you think, a 24 significant amount of the phosphorus? 25 A. I don't know, because I have not seen 194 1 values and I am not sure they exist for phosphorus 2 concentrations or deposition that are associated with 3 Sahara dust for Southern Florida. 4 Q. Is that something that can be determined 5 through research? 6 A. Yes. 7 Q. Is the Sahara dust sufficiently 8 identifiable that you can attribute -- 9 A. Yes. 10 Q. Are there any unique aspects of Florida, 11 Florida's rain processes that affect the methods of 12 sampling for phosphorus in rainfall? 13 A. I am not able to answer that question. I 14 think you are asking me about processes of some sort 15 in terms of type of events or things like that and I 16 don't have an opinion on that. 17 Q. How do you monitor total phosphorus in dry 18 deposition? 19 A. A bucket, sampler of some kind is made 20 available, during times of precipitation it is 21 covered up to collect, the other side collects the 22 wet sample, during the dry period the wet side is 23 closed and the dry bucket is open to then allow the 24 deposition to settle out. 25 Q. Is any one particular type of bucket or 195 1 receptacle considered better than another in the 2 scientific community? 3 A. Once again, I think Dr. Krupa would 4 probably be better able to answer that than I can. 5 Earlier at least I discussed I was not able to versus 6 one technique over another. 7 Q. In your opinion is bulk sampling a valid 8 method for measuring total phosphorus deposition? 9 A. I believe -- I don't have an opinion 10 concerning quantitatively but based on what I have 11 read, there seems to be a lot of doubt with the 12 usefulness of some of the bulk information in terms 13 of being able to compare wet and dry samples that are 14 now taken to reconstruct those into getting a bulk 15 sample, the problem being that a lot of the bulk 16 sampling was done integrating samples over long 17 periods of time and things could have happened to 18 those samples during that time period, for example, 19 frogs being in there, other live creatures being in 20 there that contaminated the samples, a lot of things 21 that just make it very difficult to interpret those 22 data. 23 Q. With regard to these contingencies that 24 could contaminate the sample, am I correct that they 25 would also bias the sample to increase the total 196 1 phosphorus? 2 A. At this time I can think of no process 3 that would cause an under estimation of the total 4 phosphorus, yes. 5 Q. Are you done? 6 A. Yes. 7 Q. Assuming hypothetically you were asked to 8 start from scratch to determine the variability of 9 rainfall and total phosphorus deposition within 10 Florida, how would you go about doing that? 11 A. If I were knowledgeable on all the 12 techniques that were out there for wet and dryfall, I 13 would settle on those techniques that, which were the 14 focus of most researchers in today's way of 15 monitoring. 16 I would use the best available techniques, 17 mathematical techniques that would tell me where I 18 should be monitoring, based on the objective of 19 wanting to predict phosphorus loading in WCA area, 20 and then coupled with that type of methodology then 21 go ahead and develop the monitoring system based on 22 that. 23 EPA has done some of that with other 24 species. 25 Q. Species meaning what? 197 1 A. Sulfate. 2 Q. In your opinion how many monitoring 3 stations do you think you would need? 4 A. I don't have an opinion. 5 Q. In your opinion how long a period of 6 record would you need? 7 A. Several years, and I have no idea whether 8 we are talking about five years, ten years, four 9 years, three years. I would have to look at the 10 variability of the deposition as you begin to build 11 up your data set. 12 Once again, you may have some of those 13 data already available from some of your networks 14 now, from some of the sites in the network. 15 Q. If the variability you found was low, do 16 you have an opinion of how many years of monitoring 17 you would need? 18 A. No, because I am not a network designer. 19 Some of my friends are, colleagues are. And they 20 have -- they are statisticians and mathematicians and 21 they have formulas that answer that question. 22 Q. Who would you go talk to about that or 23 determine that? 24 A. EPA research Triangle Park. 25 Q. Do you have a particular person in mind 198 1 that you would talk to? 2 A. Just the people in AREAL. That's the 3 acronym. AREAL is a part of EPA research in Triangle 4 Park. 5 Q. And you would defer to whatever opinion 6 they expressed? 7 A. Concerning network design, yes. 8 Q. Can you explain how you would calculate, 9 and I don't want to be repetitive because I know we 10 went through the exercise with the exhibits, but how 11 you would calculate the mean phosphorus concentration 12 in a particular area using phosphorus concentrations 13 in rainfall data from a particular sampling location? 14 How would you calculate the mean concentration? 15 A. Volume weighted mean? 16 Q. Volume weighting. 17 A. At a particular site? 18 Q. Right. 19 A. You could use the volume of the sample 20 which appears to be what has been recorded at several 21 of the monitoring sites. The volume for that 22 particular sample times the concentration, and you 23 add to that the next concentration and volume for the 24 next sample and you continue to do that through the 25 time you are interested in, and then you divide that 199 1 product that you have summed by the total volume over 2 the time period you are interested in and that's the 3 volume weighted number. 4 Q. And am I correct that you have not done 5 that with regard to any sampling site in South 6 Florida? 7 A. Just to verify that the calculation that 8 Walker did for the '87 to '89 time period was 9 accurate, and it was. 10 Q. And it was? I am sorry. 11 A. And it was. But that was precipitation 12 amount using the Homestead data which is not what I 13 am endorsing. It was just to reproduce his numbers. 14 Q. Do you have an opinion of whether it is 15 valid to use the Homestead gauge data in combination 16 with the sampling data from the ENP to reach a 17 conclusion in that regard? 18 A. It certainly is not valid when the volume 19 data were available and they appear to be available. 20 MR. NETTLETON: Mark this as the next 21 exhibit. 22 (A document entitled Acid Rain, A 23 World-Wide Phenomenon: Perspective from the United 24 States, was marked Lefohn Deposition Exhibit 6 for 25 identification) 200 1 BY MR. NETTLETON: 2 Q. I show you what has been marked as Exhibit 3 6, an article by Van Bowersox entitled Acid Rain, A 4 World-Wide Phenomenon: A Perspective from the United 5 States. Am I correct this is the article that is 6 referenced in Exhibit No. 2 as one of the articles 7 you have reviewed and may be relying upon? 8 A. Yes. 9 Q. Can you tell me what in this article you 10 are relying on for purposes of the opinions you are 11 offering in this case? 12 A. Page 89. 13 Q. What on page 89? 14 A. The equation that showed how Dr. Bowersox 15 and his colleagues determined the index they were 16 using which was the volume weighted average 17 concentration. 18 Q. So the equation number one on that page, 19 is that right? 20 A. That's correct. 21 Q. Anything else in this paper you are 22 relying upon? 23 A. No. 24 Q. Am I correct that is the equation you will 25 be using to complete your phase two aspect of your 201 1 assignment with regard to this case, assuming the 2 data is available? 3 A. In the case of the District data, that is 4 correct, because the information is provided in 5 volume. If other data sets that I have are provided 6 with precipitation amount in inches or millimeters, I 7 will use that. 8 Q. Is there a different formula for gauge 9 readings? 10 A. Yes. In the place of volume on page 89 11 one would substitute the amount of precipitation in 12 terms of inches or millimeters so that the equation 13 would read, C sub P equals the sum of V sub I P sub I 14 divided by the summation of P sub I. 15 Q. Am I correct that in your opinion the 16 second formula would be the more appropriate, 17 assuming the data is available? 18 A. That is the direction that NADP works with 19 and it appears that would be the most appropriate. 20 MR. NETTLETON: Mark this as the next 21 exhibit, please. 22 (A document entitled Phosphorus 23 Concentrations in Rain and Atmospheric Deposition in 24 Florida, USA, was marked Lefohn Deposition Exhibit 7 25 for identification) 202 1 BY MR. NETTLETON: 2 Q. I show you what has been marked as Exhibit 3 No. 7. Am I correct that this is the article 4 referred to under capital letter C in Exhibit 2? 5 A. Yes. 6 Q. Can you tell me what if anything you are 7 relying on in this document to support your opinions? 8 A. This is a paper that raised some red 9 flags -- yellow flags to me, meaning caution, about 10 the District data. 11 Page 14, second real paragraph, which is 12 the last paragraph on that page, to paraphrase, 13 basically there was some concern that the procedures 14 were not being followed in an adequate way, therefore 15 the usefulness of the data could be called into 16 question in terms of the absolute values. 17 This was the opinion of the authors. 18 Q. Are you specifically relying on that for 19 any of the opinions you have stated here today? 20 A. No. I am relying on it in terms of, as a 21 possibility, as I move into phase two. 22 Also, it was also demonstrated that volume 23 weighted concentrations were used by the authors 24 using NADP data for wetfall information, meaning to 25 demonstrate that volume weighted concentration was 203 1 used. 2 Q. I am sorry, I lost you on that. How have 3 you relied upon that? 4 A. There was some question on whether the 5 arithmetic mean was the appropriate thing to use in 6 the pha