Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 1 of 35 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MARYLAND WATERKEEPER ALLIANCE, INC. Plaintiff, Civil Action No. WMN-10-487 v. ALAN AND KRISTIN HUDSON FARM, et al. Defendants DEFENDANT PERDUE FARMS, INC. S MEMORANDUM IN SUPPORT OF PROPOSED FINDINGS OF FACT AND CONCLUSIONS OF LAW James L. Shea (jlshea@venable.com) Michael Schatzow (mschatzow@venable.com) Thomas M. Lingan (tmlingan@venable.com) Maria E. Rodriguez (merodriguez@venable.com) M. Rosewin Sweeney (mrsweeney@venable.com) VENABLE LLP 750 E. Pratt Street, Suite 900 Baltimore, MD 21202 (410) 244-7400 (410) 244-7742 facsimile Attorneys for Defendant Perdue Farms, Inc.
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 2 of 35 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MARYLAND WATERKEEPER ALLIANCE, INC. Plaintiff, Civil Action No. WMN-10-487 v. ALAN AND KRISTIN HUDSON FARM, et al. Defendants TABLE OF CONTENTS Page INTRODUCTION...1 I. Plaintiff Did Not Prove A Clean Water Act Violation....2 A. There is no evidence that the Hudson Farm poultry houses discharged pollutants...6 1. Plaintiff did not prove that the Hudson Farm poultry house fans emitted fecal bacteria or more than insignificant amounts of nutrients...7 2. Plaintiff did not prove that pollutants travel from the poultry houses to a farm ditch....11 3. The cows are responsible for the high pollutant counts found at HF02 and at WK A and B....15 B. Alan Hudson should be deemed to have a permit...18 C. Poultry house fan emissions are not subject to a zero discharge limit...20 D. Plaintiff did not prove that water from Hudson Farm reaches a navigable water of the U.S...22 II. Perdue Does Not Control Environmental Compliance At The Hudson Farm....22 CONCLUSION...29
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 3 of 35 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MARYLAND WATERKEEPER ALLIANCE, INC. Plaintiff, Civil Action No. WMN-10-487 v. ALAN AND KRISTIN HUDSON FARM, et al. Defendants TABLE OF AUTHORITIES CASES Assateague Coastkeeper v. Alan and Kristin Hudson Farm, 727 F. Supp. 2d 433 (D. Md. 2010)...2 Assateague Coastkeeper v. Maryland Dep t of the Env t, 200 Md. App. 665, 28 A.3d 178 (2011)...19 Browning-Ferris Indus. of Illinois v. Ter Maat, 195 F.3d 953 (7th Cir. 1999)...23 Ortiz v. Principi, 274 F.3d 1361 (Fed. Cir. 2001)...3 Sierra Club v. Cripple Cr. and Victor Gold Min. Co., 00-CV-02325- MSK-MEH, 2006 WL 2882491 (D. Colo. Apr. 13, 2006)...3 United States v. Bestfoods, 524 U.S. 51 (1998)...23 In re: Oil Spill by the Oil Rig Deepwater Horizon in the Gulf of Mexico, on April 20, 2010, 844 F. Supp. 2d 746 (E.D. La. 2012)...23 United States v. Manigan, 592 F.3d 621 (4th Cir. 2010)...3
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 4 of 35 STATUTES 33 U.S.C. 1316 (2006)...22 OTHER AUTHORITIES Mueller & Kirkpatrick, Evidence 3.3 (1995)...3
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 5 of 35 IN THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MARYLAND WATERKEEPER ALLIANCE, INC. Plaintiff, Civil Action No. WMN-10-487 v. ALAN AND KRISTIN HUDSON FARM, et al. Defendants DEFENDANT PERDUE FARMS, INC. S MEMORANDUM IN SUPPORT OF PROPOSED FINDINGS OF FACT AND CONCLUSIONS OF LAW Following a non-jury trial of this matter, Defendant Perdue Farms, Inc. ( Perdue ) submits both Proposed Findings of Fact and Conclusions of Law and the following supporting memorandum. INTRODUCTION This case is easily resolved by routine application of logic and law. Despite the airplanes, helicopters, zoom photography and multiple opportunities to sample the air, soil and water on the Hudson Farm, Plaintiff produced absolutely no evidence of a discharge from the poultry houses at any time, let alone between October 2009 and April 2010. A photograph of caked material on a poultry house fan or one of brown patches of ground underneath certain fans does not begin to establish a discharge of pollutants into waters of the United States. The only support for Plaintiff s claims came from the expert it hired. Bruce Bell is a professional, plaintiff-side witness who describes himself as an engineer with a stormwater management background, and who conducted no tests despite ample opportunity to do so. 1
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 6 of 35 Defendants, by contrast, presented the testimony of a tenured professor of microbiology who has taught at Virginia Tech since 1986, and is called upon routinely by government and environmental groups to determine the sources of water pollution, especially in rural, agricultural watersheds. He has never before served as an expert witness and has no ties to industry. Moreover, he has written hundreds of scholarly articles and books, including a textbook on microbial source tracking in 2011. (Hagedorn, Oct. 22, 2012 (AM) at 21:21-22:4.) Plaintiff s expert, conversely, lists in his CV that he most recently published an article as part of a conference held 16 years ago. (See PX 27.) Even if there were no other evidence dictating a defense verdict (which there is, see below), and this case came down to a battle of experts, the testimony of the more experienced, better credentialed, and unbiased expert witness is far more trustworthy. When cross-examining Defendants expert, Dr. Charles Hagedorn, who testified unequivocally that this was an open and shut case (Hagedorn, Oct. 22, 2012 (AM) at 95:10-12) and that the chicken operation was not the source of water pollution on or near the Hudson Farm, Plaintiff s counsel repeatedly asked Dr. Hagedorn whether he had done any testing and whether he could rule out a contribution from the poultry. (Hagedorn, Oct. 22, 2012 (PM) at 6:18-24.) Plaintiff has it all wrong. It was not up to Defendants to prove that the poultry operation on the Hudson Farm did not discharge pollutants; it was up to Plaintiff to prove that it did not that it could have or might have but that it did. Plaintiff failed. Accordingly, Defendants are entitled to judgment. I. Plaintiff Did Not Prove A Clean Water Act Violation. In order to prove a Clean Water Act violation, Plaintiff needed to show by a preponderance of the evidence that Defendants discharged a pollutant to navigable waters from a point source. Assateague Coastkeeper v. Alan and Kristin Hudson Farm, 727 F. Supp. 2d 433, 2
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 7 of 35 444 (D. Md. 2010); Sierra Club v. Cripple Cr. and Victor Gold Min. Co., 00-CV-02325-MSK- MEH, 2006 WL 2882491 at *11 (D. Colo. Apr. 13, 2006) ( [Plaintiff] has the burden of proof by a preponderance of the evidence ). To prove a fact by a preponderance of the evidence requires the trier of fact to believe that the existence of a fact is more probable than its nonexistence. Ortiz v. Principi, 274 F.3d 1361, 1365 (Fed. Cir. 2001) (quoting In re Winship, 397 U.S. 358, 371-72 (1970) (Harlan, J., concurring) (emphasis added)); United States v. Manigan, 592 F.3d 621, 631 (4th Cir. 2010) (quoting Concrete Pipe & Prods. of Cal., Inc. v. Constr. Laborers Pension Trust for S. Cal., 508 U.S. 602, 622 (1993)). It is not enough for the plaintiff to prove that its proffered facts are possible or even that they are equally as plausible as an alternative set of facts; instead the plaintiff must persuade[] the factfinder[] that the points in question are more probably so than not. Mueller & Kirkpatrick, Evidence 3.3 (1995) (emphasis added). See also Ortiz, 274 F.3d at 1365 (the factfinder must decide what he believes most probably happened ) (citing In re Winship, 397 U.S. at 371 n. 3) (emphasis added)). In other words, if the defendant s version of events is just as likely to be true as the plaintiff s, the defendant wins. In this case, Plaintiff did not prove that a discharge more likely than not occurred between October 2009 and April 2010. Plaintiff pointed to the topography of the Hudson Farm (Bell, Oct. 16, 2012 (AM) at 41:21-42:10, 45:14-46:10), architecture of the CAFO (which, like every poultry CAFO, was designed to keep water away from the poultry houses) (id. at 37:4-16), and to the fact that the Hudson Farm poultry houses (again, like all others) have fans (id. at 38:5-41:2). 1 None of that proves anything. 1 Alan Hudson testified that the builder of the poultry CAFO at his farm used a standard cookiecutter design with a swale and a swale pipe. (Hudson, Oct. 17, 2012 (PM) at 91:8-92:5.) Both Gary Kelman (Kelman, Oct. 23, 2012 (AM) at 14:3-6) and David Mister (Mister, Oct. 23, 2012 3
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 8 of 35 Plaintiff s case depended entirely on the testimony of its expert, Dr. Bruce Bell. Dr. Bell is a consultant who earns about half of his income from being a professional witness, and testifies on behalf of plaintiffs approximately 90% of the time. (Bell, Oct. 16, 2012 (AM) at 11:24-12:9.) The Court qualified Dr. Bell as an expert in environmental engineering, fate and transport of pollutants, as well as environmental sampling [ironic, given that he did none in this case], stormwater management, environmental microbiology and chemistry. (Bell, Oct. 16, 2012 (AM) at 11:16-20, 14:21-25.) Dr. Bell testified that because the water on and near the Hudson Farm contained pollutants, and because poultry house pollutants theoretically could reach the tested water via his alleged pathways, 2 the pollutants found in the water must have come, at least in part, from poultry litter. (Id. at 58:12-59:11.) Plaintiff focused particularly on the Maryland Department of the Environment s ( MDE s ) test results at sampling point HF02 on January 26, 2010, because the poultry houses on the Hudson Farm lie between sampling points HF02 and HF03, where one water test showed a lower pollutant count on the same day. The fact that the results were obtained on only one day and that only one sample was tested at each location is by itself enough to disqualify the test results from being used to identify a pollution source. Because demonstrable and significant variation exists among results of tests of the same water source taken at the same time, one set of results is of very limited use. Dr. Hagedorn testified that he participated in an experiment with nine of his students, in which each stood one meter away from a neighboring tester and sampled (AM) at 81:14-82:3) testified that the Hudson Farm looked typical. Mr. Mister stated that a swale and a swale pipe between poultry houses were characteristic of poultry CAFOs on the Eastern Shore. (Mister, Oct. 23, 2012 (AM) at 81:14-82:3.) 2 Dr. Bell defined a pathway as a route by which pollutants can come in contact with whatever you don t want it [sic] to come in contact with... (Bell, Oct. 16, 2012 (AM) at 22:1-2 (emphasis added).) 4
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 9 of 35 water from the same stream at ten different locations across the width of the stream. The results were widely and alarmingly inconsistent, ranging from 10 to 110,000 parts per million ( ppm ). (Hagedorn, Oct. 22, 2012 (AM) at 86:2-19.) William Beatty of MDE also testified that there can be enormous variability in surface water sample results and confirmed that sample results taken minutes apart in the same location can vary greatly. (Beatty, Oct. 10, 2012 (PM) at 180:9-23.) Samples taken at different locations or on different days will vary even more. (Id. at 180:24-182:11.) In short, one test result taken at one site on one day cannot be used to draw conclusions about water outside the sampling bottle and it certainly cannot be used to determine the source of any pollution allegedly revealed by the result. Nonetheless, Dr. Bell testified that he needed to do no further testing to come to his conclusion. (Bell, Oct. 16, 2012 (AM) at 78:4-11). Plaintiff visited the farm twice and took photographs including a photograph of a fan featuring what Plaintiff would have the Court believe was caked manure and feathers, but that every witness who was asked, except Dr. Bell, identified as something else 3 and could have petitioned the Court at any time for access to the farm. (See, e.g., photos at PX 167 and PX 169, showing visits in October 2010 and November 2010.) 4 Despite plenty of chances, however, Plaintiff and its expert did none of the testing that 3 See Testimony of Tammie Seyfert (Oct. 15, 2012 (PM) at 59:14-60:2 ( [I]t just looks like dust to me. )); Testimony of Charles Hagedorn, (Oct. 22, 2012 (PM) at 20:25 ( accumulated dust )); Testimony of Jeff Smith (Oct. 11, 2012 (AM) at 69:18 ( dust residue )); Testimony of Alan Hudson (Oct. 17, 2012 (PM) at 101:11-16 ( It s not as dark as litter. And it s just not chicken litter. )). 4 That close-up photograph of a fan served as a stark reminder that all Dr. Bell had to do was extend his hand a few inches to grab a sample of the caked material visible in the photograph. He did not preferring to refer to the material as manure without a basis. 5
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 10 of 35 Dr. Hagedorn testified they certainly could and should have done. (See, e.g., Hagedorn, Oct. 22, 2012 (AM) at 45:9-12, 47:7-8.) 5 A. There is no evidence that the Hudson Farm poultry houses discharged pollutants. Not a single person aside from Plaintiff s professional expert witness has concluded that the Hudson Farm poultry operation discharged pollutants. David Mister, who is both an experienced Concentrated Animal Feeding Operation ( CAFO ) inspector who has conducted hundreds of inspections on behalf of the State (Mister, Oct. 23, 2012 (AM) at 67:22-68:15) and the Regional Coordinator of the Maryland Department of Agriculture (id. at 64:24-65:1), inspected the Hudson Farm CAFO the day after Plaintiff sent its Notice of Intent to sue ( NOI ). Mr. Mister concluded that there was no visual evidence of discharge and recorded that he saw no issues with the poultry operation with regard to the CAFO permit. (DX 136; Mister, Oct. 23, 2012 (AM) at 82:8-85:18.) David Bramble, who inspected the Hudson Farm CAFO on January 26, 2010, along with almost a dozen other MDE employees (Bramble, Oct. 23, 2012 (AM) at 39:4-40:11), found no evidence of anything being transported to and from the farm within the production area. (Bramble, Oct. 23, 2012 (AM) at 40:13-20.) Gary Kelman, head of the MDE s CAFO section visited the farm on July 26, 2012. (Kelman, Oct. 23, 2012 (AM) at 2:22-25, 13:1.) He determined that the Hudson Farm seemed like a typical farm that wasn t 5 Dr. Hagedorn provided explicit examples of the type of testing that Plaintiff needed to do, testifying, for instance, that one of the many things Dr. Bell could have done was test for bound bacteria in poultry house air by using an Andersen sampler, which is a column with filters containing different pore sizes. Air pulled through the filters collects the different sized particles and when the filters are removed and washed, the collected material can be measured. If bacteria are present on the material collected, then they can be separated out and grown on an agar plate so that they can be counted. (Hagedorn, Oct. 22, 2012 (AM) at 45:13-21.) The Andersen Sampler is just one specific example of the testing that Plaintiff could have done and shows how simple it would have been for Plaintiff to collect evidence, had it wanted to do so. 6
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 11 of 35 necessarily an outlier. (Id. at 14:3-6.) No one from MDE (or any other agency) elected to bring an enforcement action against the Hudson Farm for discharges from the poultry operation. (Hudson, Oct. 18, 2012 (AM) at 6:21-24.) Following repeated visits, inspections and testing by various state agencies, Hudson Farm was charged with only one violation: it was fined $4000 for illegal storage of biosolids. (Hudson, Oct. 18, 2012 (AM) at 6:13-14.) Even that fine was later stricken. (Id. at 6:15-20.) 1. Plaintiff did not prove that the Hudson Farm poultry house fans emitted fecal bacteria or more than insignificant amounts of nutrients. Plaintiff s case, which began with a biosludge pile that it originally insisted was chicken manure, and at one point seemed to be about excessive land application of manure, continued to evolve at trial. While Plaintiff most recently proclaimed that trace amounts of litter on heavy use area ( HUA ) pads led to illegal discharges of pollutants, it seems to have retreated from that theory at trial. Plaintiff s case included only infrequent and vague references to the tracking of litter out of the houses on feet and equipment. Plaintiff did not identify any specific uncovered poultry litter outside of a house, much less make any attempt to trace the location or the timing of the alleged litter s existence to the high bacterial and nutrient counts detected off the farm. Instead, Plaintiff shifted its entire focus to emissions of dust by poultry house fans onto the swale that lies between the two houses. The bulk of Plaintiff s case on that point was the testimony of Dr. Bell. He articulated the simplistic reasoning which sums up the entirety of Plaintiff s case: poultry house fans emit dust; dust is equivalent to litter; litter contains pollutants; therefore, by emitting dust, poultry house fans discharge pollutants. (Bell Oct. 16, 2012 (AM) at 22:1-4, 33:17-19, 20-24, 39:18-23, 40:24-41:2, 72:11-13.) Dr. Bell based the conclusion that poultry house fans emit pollutants on literature. (Id. at 70:8-10.) When asked on cross-examination to discuss the literature that served as the basis 7
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 12 of 35 for his supposition, Dr. Bell was unable to name even one article. (Id. at 70:11-71:18.) Upon redirect examination, when prompted by Plaintiff s counsel, Dr. Bell stated that he was familiar with the two studies specifically identified by Plaintiff s counsel: one by Just ( An Aerobiological Perspective of Dust in Cage-Housed and Floor-Housed Poultry Operations ) and another by Davis and Morishita ( Relative Ammonia Concentrations, Dust Concentrations, and Presence of Salmonella Species and Escherichia coli Inside and Outside Commercial Layer Facilities ). 6 (Id. (PM) at 13:7-22, 14:11-15; PX 916; DX 185.) Neither supports Plaintiff s theory. 7 6 Curiously, when Plaintiff s counsel cross-examined Dr, Hagedorn about a study conducted in Germany, and Dr. Hagedorn responded that they should discuss chicken houses a little closer to home in Ohio by focusing on the Davis and Morishita study, counsel responded by saying We re going to talk about that, I assure you. (Hagedorn, Oct. 22, 2012 (PM) at 17:17.) She never raised it again. 7 Instead, both articles make clear that litter particles make up only a small part of the dust found inside a poultry house (see, e.g., PX 916 at 2 ( The sources of dust from a poultry facility include dried fecal matter and urine, skin flakes, ammonia, carbon dioxide, pollens, feed and litter particles, feathers (which produce allergen dandruff), grain mites, fungi, spores, bacteria, viruses and their constitutes, peptidoglycan, β-glucan, mycotoxin and endotoxin )) and neither concludes that the dust emitted by poultry house fans contains fecal bacteria. Although Davis and Morishita isolated bacteria outside the poultry houses that they studied, they did not determine that the isolated bacteria came from inside the poultry houses and specifically stated that quantitative analysis would need to be performed to determine [] if these bacteria are identical to those isolated from inside the facilities (DX 185 at 34.) In addition, the Davis and Morishita study is inapplicable to the Hudson Farm because it involved caged layer houses with much larger bird capacities ranging from 50,000 to 180,000 caged birds suspended over manure pits. Because the manure pit houses did not use litter, there was nothing to absorb ammonia. (Id. at 31.) Moreover, the air sampling towers were placed in front of ventilation fans that were ventilated from the level of the manure pits and ran constantly. (Id.) By contrast, the fans on the Hudson Farm do not run constantly. Two sidewall, or ventilation fans, on each house, typically run for only one minute of every five minutes, which amounts to 12 minutes per hour. (LeKites, Oct. 15, 2012 (AM) at 26:7-13, 30:14-32:25.) Only one of those faces the swale, while the other faces the HUA pads on the east ends of the houses. (See PX 18 showing the locations of the sidewall fans, including one fan at each end of the poultry houses.) The others are triggered only by elevated temperatures and turn off once the desired temperature is reached. (LeKites, Oct. 15, 2012 (AM) at 26:7-13, 30:14-21, 32:23-25.) Tunnel fans are usually used only for cooling the birds down in warm weather 8
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 13 of 35 Dr. Hagedorn provided a much more detailed, deliberate and thorough analysis, explaining that very few fecal bacteria live in the air inside a poultry house. As soon as fecal bacteria are excreted by a chicken, those bacteria begin to die off. (Hagedorn, Oct. 22, 2012 (AM) at 44:14-18.) Dr. Hagedorn concluded that whether and how much fecal coliform and E. coli are in poultry litter will depend upon how the poultry house in which the birds live is managed. (Id. at 44:11-18; PX 916 at 2 ( bacteria depend on management processes that control relative humidity, temperature, type and age of the litter ); PX 916 at 4 ( the types of feed and litter may alter the types and levels of bacteria ).) Next, Dr. Hagedorn testified that once fecal bacteria leave the animal s gut and are exposed to air inside a poultry house, their die-off rate will increase. (Hagedorn, Oct, 22, 2012 (AM) at 40:19-23.) Those that entered the air free floating will die off rapidly. (Dr. Bell agrees.) (Id. at 44:17-18.) Even those fecal bacteria attached to minute particles of organic matter would have little protection against the oxygen in the air and would not survive much longer. (Id. at 45:2-6.) 8 Dr. Hagedorn also addressed nutrients inside a poultry house. He noted that although the dust inside a chicken house will contain nutrients such as nitrogen, phosphorous, and ammonia (id. at 45:25-46:8), the concentration of any airborne nutrients will be much less than any nutrients in the poultry litter on the floors of the poultry houses. (Id. at 47:1-3, 47:18-48:2, Hagedorn, Oct. 22, 2012 (PM) at 18:2-4.) Again, he concluded that nutrient concentrations could be significantly reduced by house management. (Hagedorn, Oct. 22, 2012 (PM) at 10:14- months. (Id. at 30:4-10, 91:22-92:8.) During the winter months, two tunnel fans are set to go on only in emergencies; Alan Hudson covers the inside of the remaining four with plastic. (Hudson, Oct. 18, 2012 (AM) at 37:19-38:17.) 8 Bacteria that attach themselves to large particles would fall to the ground and therefore would not be susceptible to being blown out by a fan. (Hagedorn, Oct. 22, 2012 (AM) at 44:24-45:6.) 9
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 14 of 35 19, 13:14-20, 100:18-25.) For instance, the Hudson Farm poultry CAFO uses a poultry litter amendment (PLT), which is an acid that binds with nutrients and reduces the amount of ammonia that would otherwise be released into the air. (LeKites, Oct. 15, 2012 (AM) at 20:17-22, 94:24-95:22.) It is applied year round at every flock placement. (Id. at 20:19-25, 95:20-22.) 9 Most importantly, Dr. Hagedorn testified that the composition of air, dust or fan emissions can be readily determined by simply measuring them. Despite the fact that the premise of Plaintiff s claim that poultry house fans emit significant amounts of nutrients or that they emit fecal bacteria at all is improbable, Dr. Bell did not analyze any of the litter on the Hudson Farm, take any air samples or test any dust from inside the chicken houses. (Bell, Oct. 16, 2012 (PM) at 5:5-16.) 10 9 Todd LeKites estimated that PLT reduces ammonia 30 to 40 parts per million, or probably more than 50 percent. (LeKites, Oct. 15, 2012 (AM) at 95:13-19.) PLT was used during the sampling period between October 2009 and April 2010. (Seyfert, Oct. 15, 2012 (PM) at 94:1-14.) The Farm Visitor Log indicates that PLT was delivered and spread on December 10th and 13th of 2009. (PX 417 at PER028052.) Tammie Seyfert testified that PLT was in use at the Hudson Farm poultry houses throughout the relevant period and that she was sure of that because she could see the PLT on the litter. (Seyfert, Oct. 15, 2012 (PM) at 102:25.) Plaintiff did not take into account any use of PLT on the Hudson Farm. 10 The fans produced few emissions during the relevant period to begin with. All fans are turned off when birds are removed and none are turned on again until the day before a new flock arrives. (Hudson, Oct. 17, 2012 (PM) at 98:17-21 (day before birds arrive he set[s his] fans up so they come on as needed ).) When the MDE tested on January 26, 2010, the fans had been turned off for the prior 12 and 15 days, respectively for each house. (DX 116-PER028247 showing flock removal dates of January 11 and 14, 2010.) A new flock was not delivered until February 1, 2010 (PX418-PER028072), well after the sampling date. Further, the tunnel fans on the Hudson Farm poultry houses rarely operate in the winter. (LeKites, Oct. 15, 2012 (AM) at 91:25-92:8 (tunnel fans are typically not used during winter and during winter all but 1-2 tunnel fans are covered over with plastic); Seyfert, Oct. 15, 2012 (PM) at 100:4-14 (tunnel fans were set up for back up purpose only during winter of 2009-2010); Hudson, Oct. 18, 2012 (AM) at 37:21-38:17 (tunnel fans used mainly during warm weather months, and operate in [s]ummertime most of the time, during the winter he covers over with plastic all but two tunnel fans.).) October through April is the coolest part of the year, so the tunnel fans operate the least during those months. January, when the MDE tested, is usually the coldest month of all. 10
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 15 of 35 2. Plaintiff did not prove that pollutants travel from the poultry houses to a farm ditch. Again, Dr. Bell s say-so was the only evidence that pollutants that allegedly left the poultry houses then travelled to a farm ditch. Dr. Bell provided no analysis of how that might happen or how long it might take. Having done no testing or modeling, he simply made an ipsi dixit statement that they did. Assuming that air or dust blown out by the poultry house fans might have contained pollutants, Dr. Hagedorn explained what would happen next. First, some of the air or lighterthan-air dust emitted by fans simply disperses into the atmosphere. (Hagedorn, Oct. 22, 2012 (AM) at 49:14-19 (since gravity inside and outside poultry houses is the same, dust that exists in the air inside will not settle, but will remain suspended in air when it reaches the outside).) Any bacteria or nutrients that did not remain in the atmosphere but instead landed in the swale between the poultry houses, which itself acts as a large filter, would then begin their long journey to a farm ditch, a journey inevitably blocked by many, potentially lethal obstacles. 11 Those obstacles include: Exposure to oxygen and solar radiation. (Id. at 55:8-14.) Absorption by grass or other vegetation in the swale. (Id. at 55:16-23.) Consumption by grass microbes. (Id. at 55:25-56:15.) 11 Plaintiff would have the Court believe that the swale between the chicken houses did not act as a vegetative buffer for fan emissions because tractor and truck traffic had compacted the swale s soil surface and rendered it impermeable. Other than Plaintiff s say-so, there is no proof of that. As the many photographs of the swale show (approximately 40 were admitted into evidence), the swale sometimes appears to feature tire tracks (especially when covered with snow) but at other times looks heavily vegetated and shows no discernible evidence of vehicle traffic, let alone compaction. Compare PX 149 and PX 106; PX 160; PX 163. The only way to determine the swale s ability to inhibit the movement of nutrients or bacteria (again, assuming any escaped the poultry houses) was to measure the swale s uptake. Again, Plaintiff failed to do so. 11
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 16 of 35 Absorption by the soil. Nutrient and dust molecules that fall down in between the blades of grass and end up on the soil come into contact with a biological mat of microorganisms in the upper part of the soil and in the thatch at the bottom of the grass. The biological mat contains trillions of microorganisms per cubic inch, which would consume the nutrients and degrade the dust. (Id. at 56:17-57:6, 57:22-23.) Consumption by microbes in the soil. (Id. at 56:24-57:23.) Immobilization by soil organic matter. (Id. at 57:9-15.) With rainfall, the swale would continue to act as a filter, (id. at 59:8-15), and each of the processes described above would operate anew every time the microbe or nutrient molecule was exposed to fresh surfaces, i.e., surfaces other than those on which the molecules first landed. (Id. at 59:17-22.) Any molecules that survived, traversed the swale and found themselves at the entrance to the swale pipe would there encounter more impediments. 12 Although the vegetation at the pipe s opening would not block water from entering the pipe altogether, it certainly would slow water flow down. The very substantial vegetated cover at the pipe entrance would further absorb and immobilize any remaining particles. (Id. at 60:7-9; DX 86a (photograph showing both green and dormant plants at the swale-side entrance of the pipe).) Any remaining nutrient molecules would be exposed to an active biofilm of microorganisms that exists wherever plants 12 In order to traverse the swale, the nutrients would need flowing water. There is absolutely no evidence of such a flow occurring in the swale at any time, including between October 2009 and April 1, 2010. Only one photograph of the many introduced at trial (DX 3) even shows water in the swale and that one reveals a small puddle some distance from the end of the swale. 12
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 17 of 35 are being decomposed and would themselves be used in the decomposition process. (Hagedorn, Oct. 22, 2012 (AM) at 60:11-16.) 13 Given the minimal amounts of fecal bacteria and nutrients that could even have reached the swale from the poultry houses, and the significant obstacles to transport those particles would have encountered, Dr. Hagedorn was firmly of the view that none would have reached Ditch 1. (Id. at 61:8-62:15.) Assuming that these hypothetical pollutants existed and did reach Ditch 1, however, all of the mechanisms that operated as obstacles in the swale would impede their progress to a water of the United States. In addition, they would have to deal with new obstacles posed by the ditch: up-take and degradation by algae and soil, (id. at 67:10-13), predators and denitrification (id. at 67:17-68:19). A further obstacle would be the lack of flow in the ditches. The only way nutrients could travel from the swale to areas outside the farm is to travel via water flowing downstream. The ditches are, however, intermittent, meaning that they do not always contain water. (Id. at 70:5-6, 71:1-4.) Even when water is present in Ditch 1, an additional fate and transport hindrance would be presented by the 90 degree turn south of HF02 but north of HF01. (Id. at 71:11-20.) At that elbow, water flow slows, which would cause particles to fall out of suspension and be deposited on the sides of the ditches and streams. (Id. at 71:11-16.) Moreover, nutrients can be stranded on the sides or bottom of a ditch when the water table falls. When rain ceases, water begins to drain out of the ditch both downward and laterally. At that point, any nutrients suspended in the water would be carried by the sinking water level. 13 Plaintiff repeatedly established from witness after witness that they did not see the swale pipe, even though these witnesses were mere feet from the swale end of the pipe. The obvious explanation is that the opening to the swale pipe was covered by vegetation. 13
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 18 of 35 The suspended nutrients would percolate into the soil, accumulate on the bottom of the ditch, and be removed laterally as the water seeps into the sides of the ditches. (Id. at 73:12-23.) Given the fate and transport obstacles present on the Hudson Farm, Dr. Hagedorn concluded that the scant nutrient or bacteria originating in the poultry houses (assuming any would be emitted in the fan exhaust) could not have made their way from the swale to a farm ditch, to HF01, to either a WK or HF testing site or, certainly, to a territorial water. (Id. at 73:24-74:8.) In order for Dr. Bell to conclude as he did that fan emissions were the cause of the elevated fecal bacteria and nutrient levels found at HF02, he should have established what was in the farm ditches, groundwater, air and soil before the poultry house fan emissions could have reached them. (Id. at 62:18-64:20.) This is particularly true with respect to nitrogen and phosphorous. Those two elements are found in great quantities and concentrations on farms on Maryland s Eastern Shore because they are significant constituents of fertilizer. The soil at the Hudson Farm has for many years received large loads of those nutrients. Therefore, only by comparing samples to the background could one detect an abnormal level. Dr. Hagedorn testified that for little cost and minimum effort, Plaintiff could have taken many samples and thereby determined background levels for the nutrients and bacteria (if any) throughout the Hudson Farm. (Id. at 96:8-18.) Plaintiff did no such sampling for background levels. Similarly, if Dr. Bell had really wanted to test his hypothesis, (which was really his preordained conclusion), that the poultry houses were the source of the high pollutant counts at HF02 or at the WK testing sites, he should have performed targeted sampling, a simple process whereby one proceeds up and down different ditches to identify where the high concentrations of pollutants are located. That now you re getting hot, now you re getting cold sampling plan 14
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 19 of 35 allows one to come closer and closer to the pollution source until one identifies it. (Id. at 38:12-19.) Based on his experience in 17 states and five foreign countries tracking the sources of water pollution (Hagedorn, Oct. 22, 2012 (AM) at 28:17-23), Dr. Hagedorn concluded that the targeted sampling method, which he has successfully employed in dozens of cases, would have isolated the source of the pollutants found in the water at HF02 and at the WK sampling sites and would have confirmed that the poultry operation was not the source of the water pollution found at the Hudson Farm. (Id. at 87:12-88:21.) 14 Plaintiff did not do that testing, however, but instead relied on Dr. Bell s speculative and unsubstantiated opinion. Throughout the trial, Plaintiff remarked that the pollutants at issue are invisible (See, e.g., Opening Statement, Oct 9, 2012 (AM) at 84:19-24.) Dr. Hagedorn pointed out exactly how those pollutants could be seen, however, but Plaintiff chose not to look for them. 15 3. The cows are responsible for the high pollutant counts found at HF02 and at WK A and B. There were approximately 85 to 90 cows and calves on the Hudson Farm in 2009 and 2010 (Hudson, Oct. 17, 2012 (PM) at 81:18-25), half of which roamed freely on farm fields. 14 Dr. Hagedorn testified that he was well aware of the possibility of airborne contamination out of those [poultry] houses but that he hadn t found it or seen it on any of the other sites that he has been asked to review. (Hagedorn, Oct. 22, 2012 (PM) at 10:23-11:1.) 15 In addition, Plaintiff failed to provide any evidence regarding the volume of dust that would be necessary to produce the results obtained at HF02. During opening statement, Plaintiff s counsel stated that There s a lot of stuff being blown out of these fans. [Y]ou re going to see that there s plenty that goes on the ground. And it doesn t take much for there to be pollutants at the level of the bacteria and fecal coliform and nutrients that were seen in the endpoint sample results. (Oct. 9, 2012 (AM) at 88:12-17.) It doesn t take much (by counsel) was the last Plaintiff said on the issue. 15
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 20 of 35 (Hudson, Oct. 18, 2012 (PM) at 9:23-10:1.) The free-roaming cows produced approximately 600 tons of manure and 153,000 gallons of urine every year. (Hagedorn, Oct. 22, 2012 (AM) at 75:14-19, 76:13-20, 79:15-24.) Unlike chicken manure, which is enclosed in the poultry houses, cow waste sat out in open fields. (Hudson, Oct. 17, 2012 (PM) at 86:24-87:1.) The presence of the uncovered cow manure was well documented. MDE took photographs of numerous cow pies near Ditch 3. (See DX 4, 8.) MDE inspector David Bramble also noted its presence in an e-mail to his supervisor on January 27, 2010, the day after he inspected the farm. (DX 128.) These massive amounts of waste left in areas draining to the farm s ditches contained very high concentrations of fecal bacteria and nutrients. (Hagedorn, Oct. 22, 2012 (AM) at 76:25-77:3.) 16 Water tested in Ditch 1 at HF02 on January 26, 2010 and at WK A and B from October, 2009 to April, 2010 was impacted by Ditch 3, which flows into Ditch 1. Ditch 3 originates in the cow pasture and flows to the west, where it joins Ditch 1. 17 As vividly shown on DX58, a photograph of a well-worn cattle path, and as depicted by Alan Hudson on DX55a, every one of the cows on the Hudson Farm travelled to and from the portion of the cow pasture that drains into Ditch 3 two or three times a day in order to reach the cows sole source of drinking water. (Hudson, Oct. 17, 2012 (PM) at 83:17-84:2, 85:21-86:3.) Alan Hudson testified that he put hay bales at the entrance to Ditch 3 because he knew the pollutants allegedly discovered could not have been coming from a pile of biosolids and that s the only other place I thought anything 16 Cow urine is especially rich in nitrogen. (Hagedorn, Oct. 22, 2012 (AM) at 79:25-80:4.) 17 Plaintiff s counsel wanted to call the portion of Ditch 3 that lies between the farm road and the hay bales a drainage area instead of a ditch (Oct. 23, 2012 (AM) at 61:17-19), but as the Court correctly observed, that is only a matter of semantics. Id. The evidence shows that there is an incline to the west in Ditch 3, so that water that enters Ditch 3 at the cow pasture will flow downhill to Ditch 1. (Hudson, Oct. 18, 2012 (PM) at 16:6-18:24; Hagedorn, Oct. 22, 2012 (AM) at 80:22-81:17 (the depth of the head of Ditch 3 is approximately one and a half to two feet deep, while the depth of Ditch 3 once it crosses the road to the west is approximately three to four feet).) 16
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 21 of 35 could be coming from. (Hudson, Oct. 18, 2012 (AM) at 3:23-24.) When Mr. Hudson was asked why he wanted to slow down the flow of water into Ditch 3 by putting hay bales at the mouth of the Ditch, he said: because the cows walk through there every day and I m sure there s manure in it. (Id. at 5:2-5.) One of the MDE officials who took the water samples on January 26, Kathleen Bassett, testified that she saw cow manure near Ditch 3: I remember it specifically as we came off the lane into the pasture. (Bassett, Oct. 11, 2012 at 139:7-8.) Photographs DX 4 and DX 8 clearly show cow patties and broken-up clumps of cow manure at the entrance to Ditch 3. (Bramble, Oct. 23, 2012 (AM) at 45:18-20; Hudson, Oct. 17, 2012 (PM) at 86:4-13.) Even Dr. Bell had to agree that cow manure washed into Ditch 3 and affected the results of the water tests conducted by MDE and the Waterkeepers. (Bell, Oct. 16, 2012 (AM) at 50:22-51:5; Hagedorn, Oct. 22, 2012 (AM) at 86:20-87:1.) Dr. Hagedorn, who specializes in determining the source of water pollution, thought the cows presented an open and shut case. (Hagedorn, Oct. 22, 2012 (AM) at 95:11.) He testified that given the high pollutant counts detected at HF02 and at WK A and B, the huge amounts of cow manure and urine deposited on the Hudson Farm fields every year and the opportunity for the cows to produce the high pollutant counts, cows were the obvious and clear culprit. (Hagedorn, Oct. 22, 2012 at 95:3-12.) 18 18 Dr. Hagedorn also testified that because the cows are large animals weighing about a half ton each, they compact the earth where they congregate, thereby reducing its permeability and enhancing runoff over that surface. (Hagedorn, Oct. 22, 2012 (AM) at 84:5-10.) Unlike the tire tracks in the swale, referred to above, every photograph in evidence that depicts the relevant area shows the path followed by the cows as they walk to the head of the cow pasture and Ditch 3. See, e.g., DX 55a; DX 58. In addition, because they are such heavy animals, the cows easily break up their own cow pies by walking on them. (Hagedorn, Oct. 22, 2012 (AM) at 90:25-91:4.) 17
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 22 of 35 The counts at HF02 for fecal bacteria were very high 300,000 ppm for fecal coliform and greater than 241,920 ppm for E. coli. (PX 16.) It is completely implausible that the trace levels of fecal bacteria that might have been emitted from the poultry house via fans could have made their way through all the obstacles and filters they faced, in anything approaching the quantities recorded at HF02. The only source which could have produced those numbers was the cows. (Hagedorn, Oct.22, 2012 (AM) at 89:10-90:22.) And because these pollutants inevitably travel in packages, the concomitant high nutrient readings at HF02 also came from the cows. (Id. at 91:12-16 (In his experience, when you solve the fecal problem, you have solved the nutrient [problem]. They come from the same source. ).) Without appropriate test results telling a different story, there simply is no basis to reach any other conclusion. 19 B. Alan Hudson should be deemed to have a permit. If the Hudson Farm CAFO had a National Pollution Discharge Elimination System ( NPDES ) permit, it would not be in violation of the CWA even if poultry house fan emissions did contain pollutants and those pollutants did make it to a farm ditch and, eventually, the Pocomoke River. As shown below, the EPA does not impose a zero discharge requirement on fan emissions 20 and it requires Best Management Practices ( BMPs ) to ameliorate fan emissions only on a case by case basis. (DX 207, NPDES Permit Writer s Manual, 4-18.) No one (aside from Plaintiff) has ever expressed uneasiness about the fans on the Hudson Farm poultry houses or raised the slightest issue about how they operate or where they are placed. In fact, the Hudson Farm poultry houses, as is standard, were designed with a swale between them. This swale acts 19 Note that Plaintiff had the opportunity to rebut everything that Dr. Hagedorn said by presenting further testimony from Dr. Bell, and that it had only one, inconsequential point to make with Dr. Bell on rebuttal. (Bell, Oct. 23, 2012 (PM) at 73:20-74:11.) 20 Gary Kelman of MDE testified that the area between the poultry houses is not subject to the zero discharge limitation. (Kelman, Oct. 23, 2012 (AM) at 20:15-17.) 18
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 23 of 35 as a vegetative buffer in the (hypothetical) scenario that any pollutants escaped from the poultry houses. The United States Environmental Protection Agency ( EPA ) adopted NPDES regulations applicable to CAFOs, including some poultry operations with wet manure handling systems, in 1976. See State Program Elements Necessary for Participation in the National Pollutant Discharge Elimination System, 41 Fed. Reg. 11458 (Mar. 18, 1976). MDE proposed a general permit for animal feeding operations and draft regulations for animal feeding operations in September 2008. (Kelman, Oct. 23, 2012 (AM) at 4:9-12; Assateague Coastkeeper v. Maryland Dep t of the Env t, 200 Md. App. 665, 678, 28 A.3d 178 (2011); 35:19 Md. R. 1737 (Sept. 12, 2008.) Alan Hudson signed a NOI to be covered by the Maryland CAFO general permit on February 26, 2009. (DX195; Kelman, Oct. 23, 2012 (AM) at 7:3-14.) Over 400 concentrated animal feeding operations applied for coverage under the general permit at around the same time. (Kelman, Oct. 23, 2012 (AM) at 8:13-21.) Because there were a limited number of individuals in Maryland qualified to prepare the required comprehensive nutrient management plan ( CNMP ), MDE offered CAFO operators an opportunity to enter into a compliance schedule to legally bridge the gap between the effective date of the permit and the earliest date that [the operator could] obtain [a] certified CNMP. (Kelman, Oct. 23, 2012 (AM) at 8:12-9:18; DX 196 (emphasis added). 21 ) 21 On November 5, 2009, MDE wrote those Maryland CAFO Operators who had submitted a notice of intent to be covered by the CAFO general permit advising them that the permit would become effective on December 1, 2009 and stating: Because of delays that will be caused by so many farms seeking development of the required Comprehensive Nutrient Management Plan (CNMP) and the limited number of certified CNMP preparers, MDE is sending you the enclosed 19
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 24 of 35 The fans on the Hudson Farm poultry houses blow out to a grassy swale, which Dr. Hagedorn explained would act as a buffer and pose many hindrances to any pollutants that managed to escape the poultry houses. (See infra.) In addition, the recently-prepared CNMP for Hudson Farm (DX 205), requires no additional BMPs related to fan emissions. That CNMP was reviewed and approved by the U.S. Department of Agriculture s Natural Resource Conservation Service ( NRCS ) and the Worcester Soil Conservation Service, the agencies with technical expertise in these plans. (Kelman, Oct. 23, 2012 (AM) 11:19-12:6.) On July 26, 2012, MDE s most experienced permit writer and the head of MDE s CAFO unit visited the Hudson Farm in the company of the CAFO expert from the Maryland Department of Agriculture, David Mister. (Id. at 12:20-13:5.) After seeing the farm and confirming that it was consistent with the description in the CNMP, MDE sent the CNMP out for public comment. (Id. at 14:7-19.) In short, no regulatory expert has ever identified problems with the poultry house fans or recommended that Alan Hudson take any additional steps to address fan emissions. Accordingly, if Alan Hudson had a permit, he would be in compliance with it. C. Poultry house fan emissions are not subject to a zero discharge limit. Plaintiff cannot base its claims on air emissions from the Hudson Farm poultry house fans because the CWA does not regulate discharges of pollutants into the air. When EPA Compliance Schedule to legally bridge the gap between the effective date of the permit and the earliest date that you are able to obtain your certified CNMP. [B]y signing and complying with the requirements in the Compliance Schedule you are able to continue your operations and you will be protected from an enforcement action by MDE related to the requirement to have a permit. The Department will use its enforcement discretion and allow the Compliance Schedule to act in place of the permit until you are able to complete all the requirements of the discharge permit. (DX 196; Kelman, Oct. 23, 2012 (AM) at 8:8-9:18.) 20
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 25 of 35 adopted the current CAFO Rule on February 12, 2003, it included no provisions addressing air pollutants from CAFOs. (DX 197, Response to Comments, A-5, A-6, 21-10, 21-38, 21-40.) Instead, the EPA explicitly stated as follows: EPA considered whether CAFO air pollutant emissions should be regulated under the CWA. However, EPA determined that the [Clean Air Act] has the authority necessary to control such emissions and they are most appropriately addressed by [that] statute. DX 197, Response to Comments on the NPDES Permitting Requirements and Effluent Limitations Guidelines for Concentrated Animal Feeding Operations, at A-5 & A-6 (emphasis added). For that reason, neither the EPA nor the State of Maryland has ever formulated a policy or identified a best management practice or other standard that relates to the operation of poultry house fans or the control of exhaust from such fans at dry poultry CAFOs. (Kelman, Oct. 23, 2012 (AM) at 19:9-17; DX 194.) Further, as of June 2, 2011, eighteen months after Plaintiff released its NOI letter to the media, the State of Maryland had not approved any CNMPs for dry poultry CAFOs that contained BMPs or any other standards or requirements for the operation of chicken house fans and their exhaust. (Id.) If the law was different, and poultry CAFOs could be held liable for a CWA violation merely because they use fans and those fans emit dust, the poultry production business would come to an end. All poultry CAFOs are constructed and operate in the same way. (Hudson, Oct. 17, 2012 (PM) at 91:8-92:14 (the chicken houses at Hudson Farm were built using a standard, cookie cutter design ); Kelman, Oct. 23, 2012 (AM) at 14:5-6 (the Hudson Farm is a typical poultry farm, not an outlier ).) Plaintiff makes much of the fact that there is a swale and a pipe between the chicken houses at Hudson Farm. But, as the evidence showed, it is a common 21
Case 1:10-cv-00487-WMN Document 202-1 Filed 11/14/12 Page 26 of 35 feature on the Eastern Shore to have a swale between poultry houses and a pipe at the end of that swale. (Mister, Oct. 23, 2012 (AM) at 81:16-82:3.) D. Plaintiff did not prove that water from Hudson Farm reaches a navigable water of the U.S. Perdue understands from the Court s earlier comments that it does not believe this issue presents a dispute. Accordingly, Perdue will not argue the point, but will submit Proposed Findings of Fact and Conclusions of Law relating to this issue in order to preserve it for appeal, if necessary. II. Perdue Does Not Control Environmental Compliance At The Hudson Farm. Plaintiff would have this Court deem Perdue an operator of the Hudson Farm CAFO solely by virtue of its involvement with the chicken-growing enterprise on the Hudson Farm. 22 As the evidence presented in this case made clear, Perdue is involved with its growers (also referred to as producers ) only to the extent that its birds might be affected. Perdue s actions were entirely consistent with its legitimate contractual interest in optimal production from its contract growers. Plaintiff s attempt to twist those sensible business procedures into proof of operational control, with the accompanying staggering liabilities, must fail. The law is clear that in order to be deemed an operator of a polluting facility, a person or entity must manage, direct, or conduct operations specifically related to pollution, that is, operations having to do with the leakage or disposal of hazardous waste, or decisions about 22 Plaintiff argued that Perdue is an operator of Hudson Farm under the definition contained in 33 U.S.C. 1316 (2006), which defines an owner or operator [f]or the purposes of this section to include any person who operates, controls, or supervises a source. Id. at 1316(a)(4) (emphasis added). The referenced section is limited solely to new sources (e.g., sources built in 2001 or later). Id. at 1316(a)(2)). It is not relevant here, therefore, since the Hudson Farm poultry houses were built in 1994. (Hudson, Oct. 17, 2012 (PM) at 89:7-22.) 22