Opinion
CIVIL ACTION NO. CV105-159.
February 25, 2008.
F. Edwin Hallman, Jr., Richard A. Wingate, Decker, Hallman, Barber Briggs, Atlanta, GA, James S. Murray, Wm. Byrd Warlick, Warlick, Tritt, Stebbins Hall, LLP, Augusta, GA, for Plaintiffs.
Delora L. Kennebrew, U.S. Attorney's Office, Savannah, GA, for Defendant.
ORDER
Plaintiffs, R. A. McElmurray, III, R. A. McElmurray, Jr., Richard P. McElmurray, and Earl D. McElmurray (collectively, the "McElmurrays"), filed the above-captioned case against the United States Department of Agriculture ("USDA"), seeking judicial review of an administrative decision, which denied the McElmurrays' application for a "prevented planting" federal farm subsidy.
Presently before the Court are the parties' cross-motions for judgment on the administrative record. Because the agency's decision was arbitrary and capricious, Plaintiffs' motion will be GRANTED and Defendant's motion will be DENIED.
BACKGROUND
The City of Augusta operates the Messerly/Butler Creek Wastewater Treatment Plant, which treats industrial and household wastewater. Administrative Record ("AR") 1862.
Before Congress passed the Clean Water Act in 1972, industrial wastewater effluent was dumped into the nation's rivers, oceans, and other waterways, not subject to much, if any, oversight or regulation. See Rapanos v. United States, 165 L. Ed. 2d 159, 168 (2006). One infamous result of this pollution was that the Cuyahoga River, near Lake Erie in Cleveland, Ohio, caught on fire in the 1960s.
After unregulated dumping of industrial pollutants into the nation's rivers was prohibited, effluent from industries began being routed through the municipal wastewater treatment plants across the country, along with household sewage. At municipal treatment plants, wastewater is treated to remove chemicals, pathogens, and toxic metals from the effluent. These materials are concentrated in the byproduct remaining after treatment, sewage sludge. This byproduct also contains beneficial materials like those found in commercial fertilizer. AR 1233-34. Municipalities were left with a considerable amount of sewage sludge to dispose of in some manner. See Peter Scalamandre Sons, Inc. v. Kaufman, 113 F.3d 556, 559 (5th Cir. 1997). In the late 1970s, the treated sewage sludge was re-christened "biosolids" and a "land application/recycling" program was started.
The Clean Water Act recognizes that municipal sewage sludge contains toxic pollutants, and it requires that the United States Environmental Protection Agency ("EPA") establish numerical limitations for each such pollutant. 33 U.S.C. § 1345(d)(2)(A)(i) (2001). In 1979, the EPA enacted rules governing the land application of sludge to farmland where crops are grown. 40 C.F.R. § 257.4 (2007). In 1993, the EPA enacted the "Part 503 Sludge Rule," which further regulates the amounts of heavy metals that may be contained in biosolids applications, and reinforced the agency's view that such municipal waste is safe for spreading on farms where crops are grown. 40 C.F.R. Part 503 (2007).
Because the sludge applications that took place in this case ended before Part 503 was enacted, the Part 503 Rules do not supercede the Part 257 regulations in the instant dispute. "Retroactivity is not favored in the law. Thus, congressional enactments and administrative rules will not be construed to have retroactive effect unless their language requires this result." Bowen v. Georgetown Univ. Hosp., 488 U.S. 204, 208 (1988). The McElmurrays insist that Part 257 governs, and the USDA has never advanced any argument explaining why Part 503 should apply retroactively.
The EPA's Inspector General has criticized the EPA's biosolids program sharply, finding in a 2002 report that the "EPA does not have an effective program for ensuring compliance with land application requirements of Part 503. Accordingly, while EPA promotes land application, EPA cannot assure the public that current land application practices are protective of human health and the environment." AR 1485, 1518.
Likewise, the Fifth Circuit has noted that the experts have yet to reach a consensus regarding the safety of land application of sewage sludge generally. Scalamandre Sons, 113 F.3d at 561-62.
Since 1938, the McElmurrays have owned and operated a family dairy farm near Hephzibah, Georgia. In the 1970s, Augusta developed a land application program, whereby treated sewage sludge from the Messerly plant was recycled as fertilizer and applied to private farmland, at no cost to the farmers. In 1979, the McElmurrays and Augusta entered into a series of agreements, and the City began applying its sewage sludge at the McElmurrays' farm. Plaintiffs contend that they were told the fertilizer was safe, and the applications continued on their land through 1990.
According to R. A. McElmurray, III, in November 1990, he was having trouble with his crops. McElmurray described the problem to his brother-in-law, who had a degree in agriculture from the University of Georgia. McElmurray related that his brother-in-law opined that the problem was probably aluminum toxicity. Thereafter, McElmurray asked Augusta's land application supervisor to test for aluminum in the sludge. When the result was high, McElmurray ceased allowing sludge applications on his family's farmland. AR 1743.
McElmurray conceded that he did not quit planting the land involved in this dispute until 1998. The land produced a full crop that year, but planting was ceased due to "[1]iability, and what it was doing to our dairy cows[.]" AR 1777. According to Plaintiffs, only years after the sludge applications took place did they learn the full extent of the damage that the sewage sludge had wrought on their land. The McElmurrays accused the City of withholding pertinent information about the particular locations on their land where the sludge was applied, the volume applied, and the presence and amount of toxic metals contained in the sludge. The McElmurrays contend that the sludge poisoned plants grown on the land, which were fed to their dairy cattle, causing the cows to become seriously ill and die.
As part of the Farm Bill of 2002, Congress provided certain farmers with subsidies, which were based on historical acreage and yields, not current production choices. Direct and Counter-Cyclical Program, 67 Fed. Reg. 64,748 (Oct. 21, 2002). A farmer could establish his base acres and payment acres by including "any acreage on the farm that the producers were prevented from planting during the 1998 through 2001 crop years to covered commodities because of drought, flood, or other natural disaster, or other conditions beyond the control of the producers. . . ." 7 U.S.C. § 7911(a)(1)(A)(ii) (2007 Supp.) (emphasis added).
Prevented plant[ing] means, for the purpose of establishing base acres under § 1412.201, the inability to plant a crop with proper equipment during the established planting period for the crop or commodity. A producer must prove that the producer intended to plant the crop and that such crop could not be planted due to a natural disaster rather than managerial decisions. The natural disaster that caused the prevented planting must have occurred during the established planting period for the crop.
While it is not very material, in light of the stipulation made by Deputy Administrator Johnson, discussed below, the Court takes notice of the language used in the statute. The law does not appear to support government counsel's suggestion at oral argument that the Court should view the McElmurrays' claim skeptically because they did not qualify under the law for the credit, but were only able to apply because a special exception was made for them.
7 C.F.R. § 1412.103 (2007).
On January 15, 2003, Plaintiffs submitted a request for acreage/disaster credit to the USDA, listing environmental contamination of the land on their application as the reason for the "prevented planting." The McElmurrays listed the intended crops as 907.1 acres of cotton and 204.8 acres of corn for the years 1999 to 2001. The following day, the McElmurrays submitted additional forms, stating that their request included an additional 559.1 acres of cotton and 59.5 acres of corn for the years 1999 to 2001. The total request was for a prevented planting credit of 1466.2 acres of cotton and 264.3 acres of corn. AR 2134.
While it may seem odd at first blush, the parties agree that cotton is a food-chain crop. It is common for cows to be fed cotton hulls after the cotton lint is removed from the plant (and people consume beef and dairy products), and cottonseed oil is a common ingredient in many snack foods that people eat, like potato chips. AR 1262. Moreover, there is substantial evidence that cotton is not an economically viable crop without considering the marginal value of cottonseed. AR 1049-50 1055-56.
At first, Plaintiffs' applications were reviewed by the USDA's Farm Service Agency ("FSA") County Committee. After a preliminary review by the County Committee, the McElmurrays' application was denied because the damage was not caused by a natural disaster, as the County Committee believed was required for relief. Yet, a superior FSA official in Washington, D.C., John A. Johnson, reversed the basis for that determination. Johnson, the FSA Deputy Administrator for Farm Programs, stipulated that the McElmurrays could receive the subsidy if their land was contaminated, and the contamination caused the McElmurrays to refrain from planting the intended acreage. On April 22, 2003, the FSA County Committee again denied Plaintiffs' application for payments.
The McElmurrays appealed to the FSA State Committee. This five-member committee of farmers oversees USDA farm programs in Georgia, sets local policies, and settles agriculture-related disputes that involve farmers and public policy. After reviewing the record and conducting multiple hearings, the FSA State Committee voted in favor of Plaintiffs' application, by a vote of three to two. In finding for the McElmurrays, the State Committee discounted the advice of its attorney, Donald Kronenberger, who had opined that the State Committee was bound by certain documents provided to the Committee by the EPA, and had to deny the McElmurrays' application. AR 1988 2745.
However, the State Committee's decision was stayed, pending a review by the FSA's Deputy Administrator for Farm Programs, pursuant to 7 C.F.R. § 1412.102(d). Although the entire agency record was forwarded to Johnson, there is no indication that the Deputy Administrator reviewed the file. AR 2134 2433. On March 18, 2004, the Deputy Administrator overruled the State Committee and denied Plaintiffs' application. AR 2256-57. In part, Johnson's determination was based on a decision of the Richmond County Superior Court, which had granted summary judgment in favor of Augusta, against the McElmurrays in a related civil lawsuit. AR 2000-01. At the time, that decision was on appeal before the Georgia Court of Appeals. AR 2066. Johnson's decision was made over the State Committee's continuing objection. AR 0002 2259-60.
On April 22, 2004, Plaintiffs filed another appeal, this time with the USDA's National Appeals Division ("NAD"). On September 2 and 3, 2004, a final hearing was held before NAD hearing officer James Mark Jones. On December 3, 2004, Jones upheld the denial of the farm credit, finding no error in the FSA's decision to deny the McElmurrays' application, which was based on certain opinions provided by the EPA.
During the NAD appeal process, Jones opined that he did not have the authority to determine whether the land was contaminated, and suggested that the EPA had decided that the land was not polluted. To the contrary, Plaintiffs' counsel, F. Edwin Hallman, Jr., indicated that the EPA had not resolved the issue properly, and argued that the question of contamination was appropriately before Jones. AR 2633-34. Jones also stated that, as far as his review was concerned, "anybody's that's been untruthful, is not going to make any difference." AR 2682 2694. Based on these statements, it appears that Jones' view of his authority in deciding the case was unduly narrow, which preordained his conclusion in favor of the agency. To the extent that Jones found the EPA's position questionable or unreliable, either because of the underlying data it was based on, or because the sister agency failed to consider the actual applications presented by the McElmurrays, then Jones should not have relied on, or deferred to, such findings. AR 1495.
On January 3, 2005, Plaintiffs brought this action for judicial review of the NAD's final administrative determination in the United States District Court for the Northern District of Georgia, pursuant to 7 U.S.C. § 6999 (1999). On September 12, 2005, the case was transferred to the Southern District of Georgia.
On December 27, 2005, Plaintiffs amended their complaint, and on February 2, 2007, they moved to supplement the administrative record. On March 5, 2007, the USDA moved for judgment on the administrative record. On September 28, 2007, Chief Judge William T. Moore, Jr., denied Plaintiffs' motion to supplement the administrative record. On October 4, 2007, Chief Judge Moore reassigned the case to the undersigned for plenary disposition.
STANDARD OF REVIEW
Judicial review of the USDA's final determination to deny a prevented planting credit is governed by the Administrative Procedures Act ("APA"). 7 U.S.C. § 6999 (1999); 5 U.S.C. § 701-706 (2007). An agency's decision, including its actions, findings, and conclusions, should not be overturned unless the decision is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law" or unless it is "unsupported by substantial evidence." 5 U.S.C. § 706(2)(A) (E) (2007).
The scope of review under the "arbitrary and capricious" standard is narrow and a court is not to substitute its judgment for that of the agency. Nevertheless, the agency must examine the relevant data and articulate a satisfactory explanation for its action including a "rational connection between the facts found and the choice made." . . . In reviewing that explanation, we must "consider whether the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment." . . . Normally, an agency rule would be arbitrary and capricious if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. The reviewing court should not attempt itself to make up for such deficiencies; we may not supply a reasoned basis for the agency's action that the agency itself has not given.Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (internal cited and quoted sources omitted) (emphasis added).
Substantial evidence is more than a scintilla, and must do more than create a suspicion of the existence of the fact to be established. "It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion," . . . and it must be enough to justify, if the trial were to a jury, a refusal to direct a verdict when the conclusion sought to be drawn from it is one of fact for the jury.NLRB v. Columbian Enameling Stamping Co., 306 U.S. 292, 300 (1939) (internal case citation omitted) (emphasis added).
The Eleventh Circuit has explained that "[t]he substantial evidence test is no more than a recitation of the application of the `arbitrary and capricious' standard to factual findings."Fields v. United States, 173 F.3d 811, 813 (11th Cir. 1999). The agency must give reasons for its findings. When the evidence is in conflict, the agency must explain why it credited some probative evidence but not the conflicting evidence. Vemco, Inc. v. NLRB, 79 F.3d 526, 529 (6th Cir. 1996). The substantial evidence standard does not excuse the agency from its duty to engage in reasoned decision-making. Haebe v. Dep't of Justice, 288 F.3d 1288, 1301 (Fed. Cir. 2002).
"Except as otherwise provided by statute, the proponent of a rule or order has the burden of proof." 5 U.S.C. § 556(d) (2007);Am. Trucking Ass'ns, Inc. v. United States, 344 U.S. 298, 319-20 (1953); Dir., Office of Workers' Comp. v. Greenwich Collieries, 512 U.S. 267, 272-81 (1994). In this case, the McElmurrays bear the burden of proof because they sought the federal subsidy. AR 2440.
While Daubert does not apply to agency decisions in any formal respect, the principles underlying that decision do apply. Pasha v. Gonzalez, 433 F.3d 530, 535 (7th Cir. 2005). Significantly, the APA demands that agency decisions not be based on unreliable evidence, and an agency must provide a coherent reason for refusing to consider the testimony of expert witnesses. Chao v. Gunite Corp., 442 F.3d 550, 559 (7th Cir. 2006). In other words, "deference has its limits." Id.
Nonetheless, contrary to Plaintiffs' repeated contentions throughout the administrative proceedings, agencies may rely on hearsay in making their determinations. Richardson v. Perales, 402 U.S. 389, 402-04 (1971); AR 1427. The APA provides that any oral or documentary evidence may be considered, so long as the agency excludes irrelevant and immaterial evidence. 5 U.S.C. § 556(d) (2007).
The Court's consideration of the case is limited to the administrative record before the agency when the USDA made its determination to deny Plaintiff's application for prevented planting credits. Dkt. No. 61; see Alabama-Tombigbee Rivers Coal. v. Kempthorne, 477 F.3d 1250, 1262 (11th Cir. 2007) (court should consider evidence outside the administrative record "only where there is initially `a strong showing of bad faith or improper behavior' by the agency").
DISCUSSION
The issue presented in this case concerns whether the McElmurrays' land was contaminated by sludge applications such that the soil was unsafe for growing food-chain crops. The only dispute presented in this case concerns whether the McElmurrays' land was too polluted to use. The agency has never disputed the question of causation, and the evidence of record supports a finding that causation was established. AR 1777.
To determine whether Plaintiffs have met their burden of proof, the Court will examine the sludge data provided by Augusta, the views of the experts as to contamination, and the EPA's contributions, in turn. Along the way, the Court will examine the proof of contamination, and consider the appropriate remedy in light of the evidence submitted.
I. Augusta's Data
Much of the evidence that was considered by the federal agencies in this case, and by Plaintiffs' experts, is based on data collected by the City of Augusta, with respect to its sludge application program from 1979 to 1990. Although there is a broad consensus that Augusta's reports were unreliable, incomplete, and in some cases, fudged, the City's information is an integral part of this case.
According to the deposition testimony of Hugh Avery, Augusta's sewage sludge land application supervisor beginning in 1984, the City's sludge application data going back to 1979 were inaccurate, and the records predating his tenure were "in shambles." AR 2604-05. Specifically, Avery testified that the records were incomplete and missing critical information about which fields received sludge applications. AR 2604.
Jeff Larson, an official with the Georgia Environmental Protection Division ("EPD"), conducted an audit of the Messerly plant in 1998, and reported in an internal memorandum that problems with the sludge application program persisted, even after the program had been delegated in part to a reputable contractor, AMSCO, Inc. Larson stated that two hundred truckloads of sludge were leaving the facility for land application each day, "much of which may not be meeting requirements[.]" AR 0985 1669.
Larson found fault with the City's digestion system and its inappropriate sludge sampling techniques. Larson asserted that the City ignored certain results to make the program look better than it was in fact. AR 1668 1670. The plant was in "very poor condition," with major units rusted and out of service. Larson also reported that management at the facility was "literally a joke[,]" and that the "staff is the most demoralized bunch of people I have ever witnessed[.]" AR 0986.
The final EPD report based on Larson's investigation found that "[t]he sludge regulations are based on a well run pretreatment program which is not the case in Augusta. The sludge is highly corrosive. . . ." AR 1670. The report recommended that the plant be shut down immediately. AR 1671. Neither the USDA nor the EPA asserted that conditions at the Messerly plant had deteriorated since 1990. Indeed, Larson indicated that the plant had "been grossly neglected for years." AR 0986.
Dr. Lewis Goodroad, Plaintiff's expert soil scientist, reported that Augusta manipulated its data by averaging lab results over several months in an attempt to reduce the levels of metals present in the sludge. AR 0681. A former Supervisor of the Messerly Wastewater Treatment Plant, Allen Saxon, confirmed that this was the case. AR 0808. An employee of the USDA, Tommy Weldon, agreed that it "would be hard to come to a conclusion based on [Augusta's] data[,]" because of the City's "sloppy record-keeping and inaccurate data." AR 2758.
There is also evidence that the City fabricated data from its computer records in an attempt to distort its past sewage sludge applications. AR 502-03. In January 1999, the City rehired Saxon to create a record of sludge applications that did not exist previously. Saxon prepared sludge spreadsheets in 1999, which showed cumulative loading calculations for the first time in the twenty-year history of the City's land application program. AR 0798-818, 844-52, 659-685.
In other instances, there is evidence that Augusta altered its records to show that the sludge was applied to different, incorrect fields. Handwritten notes on the City's records contradict the number of acres involved, and the volume of sludge applied, as those figures are represented in the 1999 spreadsheet developed by Saxon. AR 2598. Other evidence indicates that City officials altered the spreadsheets in 1999 in an attempt to remove any record of the application of hundreds of thousands of gallons of sludge to hundreds of acres on the McElmurrays' farm. AR 0643-47. Goodroad reported that 18.9 million gallons of sludge had been applied to Plaintiffs' fields but was not recorded by Augusta. AR 0650.
Notwithstanding these facts, USDA employee Ronald Carey testified that evidence that Augusta changed its records years after applications were made, to reflect that the sludge was applied to larger plots of land than was actually the case, would not concern him. AR 2590.
The McElmurrays contend that Augusta's records, under-representative though they are, show that Augusta violated federal law in placing the sludge onto their land, citing, inter alia, 40 C.F.R. § 257.3-5 (2007). This federal regulation governs allowable cadmium and polychlorinated biphenyl ("PCB") limits. Plaintiffs contend that this violation is plain evidence of contamination of Plaintiffs' land and the unsuitability of the property for the production of food-chain crops. AR 658-685. The Court will explore that evidence and regulation below.
II. The Experts' Responses: Hall and Haaland Describe the Evidence of Contamination
During the administrative proceeding, Plaintiffs presented credible evidence from qualified experts that supported their contention that their farmland was contaminated. That evidence was not considered by the EPA or the USDA, but the McElmurrays' applications were denied anyway.
William L. Hall is a professional engineer and the CEO of NewFields, Inc., an environmental consulting firm based in Atlanta, Georgia. Plaintiffs retained Hall and NewFields as experts in separate litigation against the City of Augusta relating to the sludge applications to their land. On April 1, 2003, Hall signed an affidavit that was submitted to the FSA and included in the administrative record. AR 0329-0336. Hall has extensive experience with respect to the impact of heavy metals on the environment, and has been the project manager on seven Superfund sites that reached final closure. AR 0329, 0361-68, 0691-92.
Hall made extensive findings about Augusta's sludge data and the specific instances of contamination on the McElmurrays' farm. Hall opined that about 2,234 acres of the McElmurrays' farm was unusable, due to contamination from the heavy metals contained in the sewage sludge. AR 0330. Hall noted that high contaminant concentrations were based on the limited sampling that had been completed, and opined that there was a correlation between cow mortality and the consumption of silage, which is animal feed made from forage plants, grown on contaminated fields. AR 0331.
Hall reported that Augusta allowed companies to dump industrial waste into an open pit at the Messerly plant, and that the City failed to monitor the amount and type of waste being dumped into the pit while the McElmurrays were receiving sludge. Hall also faulted the plant's managers for failing to keep records showing when and where dangerous contaminants were placed on the McElmurray land. AR 0332 0782. Hall recounted that the sludge applications were unpredictable and variable in terms of the kinds and amounts of contaminants contained in the sludge. This resulted in "hot zones" of extremely high contaminant ratings on random parts of the McElmurray farm. AR 0333.
Dr. Goodroad reported that former county agent Bill Craven had agreed that sludge applications on the McElmurrays' land were not uniform. AR 0372.
Of particular concern, Hall noted that over ten percent of samples showed highly elevated cadmium concentrations, at levels up to seven times the limits that had been established at some Superfund sites, which were being remediated under the Comprehensive Environmental Response, Compensation, and Liability Act ("CERCLA"), 42 U.S.C. § 6901- 6992k (2003).
Further, Hall criticized the City's sampling practices, explaining that Augusta took less than five cubic feet of dirt per million cubic feet of soil, and only within the top eight inches of the soil column. According to Hall, this part of the soil is the least likely to retain contaminants over time, due to leaching. Hall points out that the City's data shows that the sludge contaminant concentrations became highly erratic, with extreme metal concentration spikes, beginning in 1986. Hall opined that this time frame coincided with a significant increase in mortality in the McElmurrays' dairy herd, when compared with the state average. AR 0335.
In an expert report, Hall reported specific shortcomings in Augusta's field update report data, which purport to record "year to date" ("YTD") and "lifetime total" ("LTT") applications of particular heavy metals on the McElmurrays' land. The reports are inconsistent in that they show YTD figures that match LTT figures and, relatedly, subsequent application data that does not account for prior applications in reckoning the LTT data.
In other instances, the field update report data show cumulative LTT figures that decrease from one application to the next. AR 0342 0350. Still, Augusta's data indicated that cadmium and molybdenum levels on the McElmurray farm were above regulatory limits in certain instances, in amounts ranging from 37% to 1400%. AR 0352-53. Hall opined that the high concentration of molybdenum in the McElmurrays' silage was particularly serious, given the time that had elapsed since the sludge was placed on the land. The McElmurray samples were taken in 1998, eight years after Plaintiffs halted the land application program. AR 0356.
Additionally, Hall faulted Augusta's data for lacking information. Complete months and years were missing from the field update reports, which meant that Augusta's sludge application estimates were under-reporting the toxicity of the soil by a wide margin. Hall also called attention to the City's failure to monitor molybdenum, despite evidence of its presence, given that it is a known hazard on land used by dairy cattle. AR 0343. Hall reported that after the City learned about high concentrations of molybdenum in its sludge, it failed to notify researchers at the University of Georgia about the presence of this heavy metal. Because the University scientists failed to test for molybdenum, the researchers' advice to apply lime to raise the soil's pH level, and thereby limit crop toxicity, was faulty or incomplete. AR 0348.
To the extent it has any relevance, Hall noted that even though 40 C.F.R. Part 503 limits concentrations of molybdenum to 75 parts per million ("ppm"), the sample concentrations on the McElmurrays' land ranged from 25 ppm to almost 140 ppm. AR 0344. Hall drew attention to the fact that the USDA expressed concern about the molybdenum levels permitted in the EPA's Part 503 Rules. The USDA recommended that the EPA reduce the ceiling concentration limit for molybdenum in biosolids to 54 ppm. Even under the EPA's more relaxed limit, 75 ppm, Hall pointed out that Augusta's sludge was applied at about twice that level in some cases. AR 0756. Nonetheless, it is not apparent that this particular test result shows contamination of the soil, in light of the McElmurrays' protestations that Part 503 does not apply in the instant case.
Dr. Ron Haaland, an Auburn University professor in the School of Agriculture, was hired by Augusta's attorney as an expert witness in the Superior Court litigation. AR 0423. Haaland performed testing at the McElmurrays' farm, and concluded that the soil was not unsafe for growing food-chain crops. Haaland blamed any ill effects from the sludge on the McElmurrays' failure to pay attention to detail and oversee the sludge application program properly. AR 0420 1374.
The McElmurrays took issue with Haaland's soil-testing methodology before the State Committee. Plaintiffs asserted that Haaland attempted to find a way to discredit the McElmurrays' samples and show no contamination on their property. The McElmurrays claimed that Haaland set up their property using a nine acre grid system, and pulled one sample from each acre in the nine acre grid. Plaintiffs submit that Haaland then combined the samples together to dilute any results showing the presence of contaminants. AR 1868.
Although Haaland is the only expert that the parties have disclosed that tested the McElmurrays' soil and disagreed with Plaintiffs' experts' conclusions of contamination, the agency never responded to this criticism of Dr. Haaland's methodology. At oral argument, the government's lawyer declined to address this point, leaving lingering doubt about there being any evidence that supports the government's determination that the land was not contaminated.
Evidence related to the pH level of the soil also supports Plaintiffs' position that the land was too polluted to grow crops for human consumption. Food-chain crops may not be grown when the pH of the sludge and soil mixture is less than 6.5 and the cadmium level therein exceeds 2 ppm. 40 C.F.R. § 257.3-5(a)(1)(i) (2007). Nor may such crops be grown where the annual application of cadmium from solid waste exceeds 0.5 kilograms per hectacre, or .45 pounds per acre. 40 C.F.R. § 257.3-5(a)(1)(ii) (2007).
Plaintiffs' evidence shows that sewage sludge with cadmium concentrations of between 4.2 ppm (January 1980) and 1200 ppm (February 1990) were deposited on Plaintiffs' farmland for ten years. Many fields contained annual cadmium deposits that were two or three times the federal limit. AR 1132-1157. According to the information supplied by Augusta, the pH level of the sludge and soil mixture at the McElmurrays' farm was below the 6.5 minimum consistently. These figures were accepted as credible by Plaintiffs and their experts, and the EPA, which relied on Augusta's data only in reaching its conclusions in this case. AR 892-913.
Another factor supporting Plaintiffs' argument that the land was contaminated is that certain metals react to the soil's pH level differently. Augusta advised the McElmurrays to keep the pH level of their soil elevated, to attenuate the effect that certain heavy metals would have on their crops. Generally, most metals will accumulate from the soil into the plants grown thereon when the soil has a low pH level. However, molybdenum and arsenic are the exception to this rule. AR 1783. According to experts retained by both parties, molybdenum accumulates in plants more easily and directly when soil pH levels are high. AR 0345 0411. As a result, Augusta's suggestion that applying lime to raise the pH level would mollify any contamination concerns was misleading or incomplete. AR 0348.
Other specific evidence showed that heavy metals were found at levels that were above the regulatory limits on the McElmurrays' farm, making the land unfit for food grown for human consumption. On one piece of property alone, antimony levels registered at 96.8 ppm, while the regulatory limit was 4 ppm. Arsenic registered at 44.2 ppm, more than twice the amount allowed by law. Cadmium was found at a level of 6.41 ppm, which was more than three times the level deemed safe under the law. Selinium registered at 5.4 ppm, although the cleanup standard provided under the law was set at 2 ppm. Thallium was found at 51.6 ppm on that particular piece of property, although the regulatory limit is 2 ppm. AR 1801-03. The levels were similar on other parcels farmed by the McElmurrays. AR 1803-06.
According to the evidence contained in the administrative record, Thallium is quite dangerous to dairy herds. AR 0916. Plaintiffs maintain that Thallium was used as a catalyst by NutraSweet in making its sweetener, and NutraSweet was the largest user of the Augusta sewer system during the 1980s. AR 1808. The McElmurrays contend that the City did nothing to limit large or illegal dumping, like that by NutraSweet. A 1998 EPD audit provided some support for this contention, finding that "[t]here are no local limits for conventional pollutants" at the Messerly plant. AR 1669.
This portion of the administrative record discusses the limits allowed under Georgia law. At oral argument, Plaintiffs' attorney conceded that federal law controlled, but reported that Georgia law was coextensive with federal requirements in this respect. Although counsel for Defendant expressed no opinion about the applicability or the relevance of state law, the Government's lawyer did not disagree that the relevant state and federal standards were the same.
At oral argument, the McElmurrays noted that the administrative record showed that Augusta's lab reports demonstrated that PCBs were placed on their land at a level in excess of 5,000 ppm, even though the allowable limit under EPA standards was 2 ppm. See 40 C.F.R. § 257.3-5 (2007). The government has not disputed that characterization of the evidence, and it is supported by the administrative record. AR 0535.
Moreover, Plaintiffs submitted evidence that the sludge contained hazardous levels of chlordane, and that it was applied to their land from 1980 to 1985, even though it was banned in 1978. AR 843-883 1109-57; Velsicol Chemical Co., et al.: Consolidated Heptachlor/Chlordane Cancellation Proceedings, 43 Fed. Reg. 12,372, 12,373 (March 24, 1978). Plaintiffs cite the following additional sources as evidence that the sludge was applied to their land in violation of federal law: AR 0329-85, 0623-837, 1064-1073; see 40 C.F.R. Part 257, 40 C.F.R. Part 261, 40 C.F.R. Part 258, Appendix I and II.
The evidence in the administrative record shows that the McElmurrays' land is contaminated and unfit for growing food-chain crops. Plaintiffs maintain that they would have violated the law by planting crops, putting human health and welfare at risk. The McElmurrays submit that the high mortality level experienced by their dairy herd is proof of the dangers associated with planting food crops on their land.
The Court concludes that the evidence of contamination on the McElmurrays' land was substantial, and the data provided by Augusta was flawed.
III. The EPA's Contributions: Mehan, Brobst, Kaufman, and Breen
The USDA submits that applications for prevented planting subsidies, like the one submitted by Plaintiffs, are usually based on the effects of natural disaster to land and crops. Because Plaintiffs' claim had a more unusual basis, alleged contamination of the land, the USDA had to consider the alleged biological effects of sewage sludge on Plaintiffs' land.
Therefore, in evaluating Plaintiffs' application, the USDA sought the opinions of officials at the EPA. The USDA recognized that it possessed limited knowledge regarding the biological effects of sewage sludge on soil, and it sought the advice of the EPA. An FSA handbook allowed it to do so, in certain instances where it lacked the expertise to make proper findings:
If a reviewing authority receives a request for review involving a technical determination by a Federal Agency other than FSA and NRCS, the reviewing authority shall . . . contact a representative of the applicable Agency to discuss and clarify the technical findings, as needed[,] . . . [and] accept as binding, written factual findings or technical determinations of the other Agency.
AR 1495.
The USDA received varying responses from EPA officials about the safety of the sewage sludge land application program and the McElmurrays' applications. Finally, the EPA declared that its official position as to the McElmurrays' petition was set out in a letter written by EPA's Assistant Administrator, G. Tracy Mehan, III. Consequently, the Court will focus on Mehan's letter first.
On December 24, 2003, Mehan wrote a letter responding to a petition from the Center for Food Safety seeking a nationwide moratorium on the land application of sewage sludge. Mehan's letter was broad in scope and only mentioned the McElmurrays' situation in a brief aside. Instead, Mehan considered a number of other issues in rejecting the proposed moratorium, concluding that "[p]etitioners do not present scientifically-based evidence or documentation that links the land application of sewage sludge or chemical pollutants allegedly contained in sewage sludge to human health and environmental impacts that are described in the petition." AR 1521.
Mehan did address Augusta's sludge application program, but all of his specific remarks focus on the Boyce dairy farm's litigation against Augusta, which was a companion case to the Superior Court lawsuit that the McElmurrays had filed against the municipality. For any opinions that Mehan does express about the Messerly treatment plant, Mehan relies on Augusta's sludge data only, which has been called into question by representatives of both parties in this case, as well as disinterested third parties, and Augusta's own representatives. AR 0023, 0332-35, 0342-43, 0350-56, 502-03, 0643-47, 0650, 0681, 0782, 0798-818, 0844-52, 0985-86, 1512-15, 1668-71, 2604-05, 2758, 2598.
Specifically, Mehan recounts the Center for Food Safety's assertion that, "On June 24, 2003, a court in Georgia ruled that the land application of sewage sludge was the legal cause of the damage to the farmland and the deaths of the farm's prize-winning cattle[.]" AR 1512. Mehan commented that the "EPA understands that the jury awarded $550,000 of the $12.5 million in damages sought by the plaintiffs without any findings of fact." AR 1512.
Mehan quoted from a letter written by Augusta's lawyer, James Ellison, to the EPA about the verdict. According to Ellison,
[o]ne of the breaches contended by the Boyces was an alleged failure to keep and maintain good records. Unfortunately and regrettably in the early days of Augusta's land application program, record-keeping was a problem, mostly due to programming problems with the biosolids application software used by Augusta. The verdict may well have represented the jury's dissatisfaction with the records maintained by Augusta.
Not surprisingly, Hallman, who also represented the Boyce family in the Superior Court case, takes issue with Ellison's characterization of the verdict. Hallman asserts that, under Georgia law, a general verdict ratifies the claims contained in the operative complaint. AR 1556 (citing Ga. Code Ann. § 9-12-1). What motivates any particular jury verdict (and the amount of damages awarded) is subject to endless speculation, and what happened in the Boyce case is not particularly germane to whether the McElmurrays' land was contaminated. Still, the information is material to the extent that it shows the basis for the EPA's opinion.
Plaintiffs argue that Defendant is wrong to rely on Mehan's letter as a factual finding or a technical determination by the EPA that Plaintiffs' land was not contaminated because Mehan's letter was not written in response to Plaintiff's applications. Mehan's letter contains no factual findings regarding Plaintiffs' land, and is not addressed to the USDA. Rather, Mehan wrote in response to a petition from a public interest group seeking a moratorium on the land application of sewage sludge in the United States.
The procedure described in the FSA Handbook for obtaining a technical determination from another agency requires a representative of an agency to "contact a representative of the applicable Agency to discuss and clarify the technical findings, as needed. . . ." AR 1495. Such was not done by the USDA's representatives with Mehan. In addition, Mehan makes clear that the petition relates only to the application of sludge under Part 503. AR 1504. As has been discussed, this law does not apply to the McElmurrays, whose land applications of sludge ceased before the enactment of the regulation in 1993. In short, Mehan's letter is largely irrelevant to the McElmurrays' applications before the USDA.
USDA employees Ronald Carey and Tommy Weldon also asked Robert Brobst, a member of the EPA's Biosolids Incident Response Team ("BIRT"), about the contamination averments made by the McElmurrays. AR 1227-1229. In response, Brobst opined in a letter that the McElmurrays' land was not contaminated. AR 1230-1240.
Because Brobst concluded that Augusta's data sets were the most "complete and reliable," he used its information, and did not consider (or find any particular fault with) the information provided by the McElmurrays. Brobst's letter focused on cadmium levels at the farm, and at least in his letter, he found that cadmium levels there were within normal national background ranges. Notably, the data, which Brobst claims was obtained in 1999, puts cadmium concentrations on the Plaintiffs' land at .41 mg/kg, which is twice the national average cited by Brobst, .175 mg/kg. AR 1281-1283. Brobst also stated that other metals found in the sludge, or on the land, were within normal background ranges. AR 1238.
On December 11, 2003, Brobst further explained his results to the FSA State Committee. AR 1876-1899. Plaintiffs emphasize that on that day, Brobst made an important qualification to his earlier representation, when he conceded that his original conclusions, which were based on national background concentrations, should not, or need not, be used because those levels are dissimilar to the characteristics present in soil located in Burke County, Georgia. AR 1888, 1477, 1567-68. Perhaps more importantly, Brobst admitted that one of the McElmurrays' fields contained about forty to fifty times the allowable lifetime loading level of cadmium. AR 2652.
Brobst provides scant support for his determination that the land was not contaminated. Although his letter cites to some data in support of that conclusion, he never explains where such data were found, or how he arrived at such figures. AR 1237-38. It is difficult, if not impossible, to evaluate the trustworthiness of such a conclusion without this information.
As Plaintiffs note, Brobst's letter does not address information contained in Plaintiffs' applications, but exclusively addressed data obtained from the City of Augusta in 1999. Brobst admitted that he did not evaluate the data presented in support of Plaintiffs' applications for prevented planting credit. Because Brobst concedes that his conclusion is based on Augusta's unreliable, and to some extent, invented, data, Brobst's finding has little merit on its own.
On December 31, 2003, Plaintiffs submitted an affidavit from Hugh Kaufman, a senior policy analyst at the EPA, to the State Committee in an effort to rebut Brobst's position. Kaufman explained that he had been involved with testing and evaluating the McElmurrays' land, and opined that the McElmurrays' land was contaminated, and unfit for growing food-chain crops. AR 1478, 1487-1489, 1548.
On January 28, 2004, Barry Breen, the EPA's Principal Deputy Administrator, wrote a letter to the FSA explaining that Kaufman's affidavit was not the official view of the EPA, and that Mehan's letter was the agency's position. AR 1545. Indeed, the FSA relied on Mehan's letter as the official position of the EPA. AR 2600. Yet, there is no evidence that Mehan ever reviewed the Plaintiffs' applications, other data in the administrative record, or any of the reports detailing the sewage sludge applications on Plaintiffs' land from 1979 to 1990. AR 2663. USDA employee Carey allowed that Mehan made no specific finding that the McElmurrays' land was not contaminated. AR 2664-66.
The EPA's unexplained rejection of Kaufman's position, in favor of the largely irrelevant Mehan letter shows that the decision was not based on substantial evidence. It was arbitrary and capricious for the USDA to defer to Mehan's letter as a technical determination or a written factual finding. Sierra Club v. Martin, 168 F.3d 1, 4-7 (11th Cir. 1999). To the extent that the USDA relied on Brobst's opinions, that was arbitrary and capricious because Brobst did not consider all the relevant data.Motor Vehicle Mfrs. Ass'n, 463 U.S. at 43.
An agency may discredit the uncontradicted witness testimony based on credibility grounds, but only if the agency provides reasons for its credibility determination. Tieniber v. Heckler, 720 F.2d 1251, 1254-55 (11th Cir. 1983); NLRB v. Walton Mfg. Co., 369 U.S. 404, 406-07 (1962). Breen failed to justify why the EPA accepted Mehan's letter over Kaufman's affidavit, or even attempt to explain how Mehan's letter could qualify as a written factual finding or technical determination of the McElmurray matter. Moreover, no one at the EPA ever took the time to evaluate Plaintiffs' applications or their experts' conclusions.
Likewise, Breen failed to investigate the findings made by Kaufman. Carey asked Breen what the basis was for Kaufman's statement that the McElmurrays' land had received sludge applications making the land unsuitable for growing food-chain crops. Breen replied "I do no have information with which to answer this question." AR 1545.
As the Supreme Court has stated, "[t]he substantiality of evidence must take into account whatever in the record fairly detracts from its weight. This is clearly the significance of the requirement . . . that courts consider the whole record."Universal Camera Corp. v. NLRB, 340 U.S. 474, 488 (1951).
Other evidence of record calls into question the fairness and objectivity of the EPA's opinions with respect to the sludge land application program. The administrative record contains evidence that senior EPA officials took extraordinary steps to quash scientific dissent, and any questioning of the EPA's biosolids program.
On February 4, 2004, Dr. David Lewis, a former EPA employee, testified before the House of Representatives' Subcommittee on Energy and Mineral Resources about improper use of the scientific peer review process by senior EPA officials, with respect to a University of Georgia study relating to the Messerly plant, and the deficiencies in the agency's position in support of land application of sewage sludge. AR 1610 1616. Lewis criticized the EPA for its handling of the allegations involving the Messerly plant in Augusta, especially its reliance on the dubious data provided by the City. AR 1622-24.
Lewis' work as a microbiologist first drew national and international attention in the early 1990s when six dental patients of the same dentist in Florida contracted HIV. Lewis published a series of articles in the leading British medical journal The Lancet, showing that blood trapped in lubricants inside dental devices can escape disinfection and spread HIV, the virus that causes AIDS. This research prompted new heat sterilization guidelines worldwide. AR 1625.
Lewis explained that he had worked at the EPA for thirty-one years, and was forced to resign in May 2003 because his biosolids research was at odds with official EPA policy. AR 1619. Lewis testified before Congress that the EPA had politicized scientific research at the agency, and utilized unreliable and fraudulent data to support the continuation of the sludge land application program. AR 1619. Lewis recounted to the Committee that he researched adverse health consequences of sewage sludge from 1996 to 2003. Specifically, Lewis wrote a research paper with University of Georgia scientists that' linked chemical irritants from airborne dusts, as a result of sewage sludge applications, to children's illnesses. AR 1620.
Lewis reported that a senior EPA official, Dr. John Walker, acted unethically in protecting the Part 503 sludge Rule, which Walker had helped to create. Lewis claimed that Walker had stated that he was qualified to review Lewis' microbiological research, although Walker was untrained in the field. Lewis stated that Walker approached a friend who was a corporate executive at a company involved in the sewage sludge business to help come up with criticisms of Lewis' paper. In addition, according to Lewis' testimony, Walker asked a USDA microbiologist for help with a technical review, and then plagiarized the USDA official's work as his own. Thereafter, Lewis stated that Walker distributed the critique widely, within the EPA, to trade associations, and among regulated businesses in the industry. AR 1621.
Walker also distributed an anonymous twenty-eight page critique of Lewis' research, which had not been peer reviewed, and contained false scientific arguments aimed at discrediting Lewis. Lewis told the Congressional panel that a colleague at the National Academy of Sciences, Ellen Harrison, testified in a separate proceeding that the paper damaged Lewis' reputation. AR 1621-22. Thereafter, Walker's associates attempted to pressure EPA Administrator Christine Todd Whitman to end Lewis' research immediately. AR 1627. Walker faced no discipline for his actions by the EPA. AR 1620-21.
On May 28, 2003, the EPA forced Lewis to resign for publishing articles in the leading scientific journal Nature, which were critical of the EPA's biosolids policies. During his Congressional testimony, Lewis detailed how EPA administrators attempted to force him out after his article, "EPA Science: Casualty of Election Politics," was published in Nature in 1996. Lewis described how further retaliation in 1999 by senior EPA officials, against him and his supervisor, Dr. Rosemarie Russo, prompted a separate hearing before Congress and helped spur enactment of the "No Fear" Act, a law protecting federal employees against retaliation. AR 1625-27.
The distribution of the false scientific reports by Walker caused University of Georgia officials to scrap their plans to hire Lewis after he left the EPA. Even letters from United States Senators James Inhofe and Charles Grassley, in an attempt to save Lewis' job at the EPA, were ineffective. AR 1627-28. Lewis reported that he had been blacklisted by Walker, and that he remained unemployed since he left the EPA. Lewis indicated that he had taken up an unrelated area of research without compensation because of the EPA's actions, stating that he was directing research on hepatitis C infections in Egypt. AR 1628.
IV. Summary Findings and the Appropriate Remedy
Any data that was considered by Mehan and Brobst that related to the McElmurrays' farm was that collected as of 1999. Neither official considered Goodroad's 2001 analysis detailing the deficiencies in the data collected as of 1999. The men did not discuss or acknowledge the serious limitations and deficiencies of Augusta's data. Neither official considered Plaintiffs' applications or the reports of their experts contained therein. AR 1235.
Neither Mehan nor Brobst made either a written factual finding or a technical determination about Plaintiffs' applications. Mehan, who represented the EPA's official position, did not find any material facts as to the application, and his letter was not a technical determination, but a statement of policy. Brobst may have attempted to produce a technical determination, but he did not consider the McElmurrays' applications, just old data, and he failed to consider anything the McElmurrays or their experts had to say to the contrary. Breen's conclusory rejection of the specific findings contained in Kaufman's affidavit was not binding on the USDA.
The administrative record indicates that the members of the FSA State Committee reviewed the Plaintiffs' applications thoroughly. The members of the State Committee were familiar with Plaintiffs' expert reports, and the import of that evidence. That committee voted in favor of the applications for credit. Likewise, EPA employee Kaufman was familiar with the McElmurrays' applications, expert reports, and the testing on their land. He had conducted an investigation by visiting Augusta and looking into the problems at the Messerly treatment plant. Kaufman's affidavit indicates that the land is unfit for growing crops for human consumption. AR 1487-1489. Hearing Officer Jones also considered the evidence in the case, but his comments indicated that he felt he was bound by EPA opinions to which he ought not have deferred. AR 2144. See infra, note 4.
In short, it appears that the only persons to consider Plaintiffs' applications ended up ruling in their favor, or did not believe they had the authority to rule in the McElmurrays' favor. The USDA's decision to accept a contrary decision, based on no review of the applications by the EPA, was arbitrary and capricious. The conclusions of the EPA were not based on substantial evidence, and the USDA was not compelled by their handbook to rely on the letters presented in this case.
Contrary to the McElmurrays' suggestion, that is not to say that the USDA could not defer to a sister agency if that agency made appropriate findings.
An administrative determination cannot be upheld without an articulated, rational connection between the facts before the agency and the agency's decision. Zahnd v. Sec'y of Dep't of Agric., 479 F.3d 767, 773 (11th Cir. 2007).
Because the record supports Plaintiffs' petition for farm subsidies, the Court reinstates the original decision of the FSA State Committee, and directs the USDA to grant the McElmurrays' application for prevented planting credits. Remand is inappropriate because the record was unevaluated or ignored by agency officials at the USDA and the EPA. In other words, while the record was inadequate to support the agency's decision, it is adequate to support Plaintiffs' applications.
The Court has the obligation under the APA to conduct judicial review of administrative decisions. That statute requires the Court to "hold unlawful and set aside agency action, findings, and conclusions found to be arbitrary and capricious." 5 U.S.C. § 706(2)(A). The agency "is not entitled to a second bite of the apple just because it made a poor decision — if that were the case, administrative law would be a never ending loop from which aggrieved parties would never receive justice." McDonnell Douglas Corp. v. NASA, 895 F. Supp. 316, 319 (D.D.C. 1995); Nelson v. United States, 64 F. Supp. 2d 1318, 1326 (N.D. Ga. 1999); Florida Power Light Co. v. Lorion, 470 U.S. 729, 744 (1985).
CONCLUSION
For the reasons explained above, the USDA's motion for judgment on the administrative record is DENIED, and the McElmurrays' cross-motion is GRANTED. Dkt. Nos. 54 57, respectively. The Court hereby DIRECTS the USDA to grant the McElmurrays' application for prevented planting credits.