United States v. Ottati & Goss, Inc., United States of America v. Ottati & Goss, Inc., Appeal of State of New Hampshire, Intervenor/plaintiffUnited States v. Ottati & Goss, Inc., United States of America v. Ottati & Goss, Inc., Appeal of State of New Hampshire, Intervenor/plaintiff
Nеarly ten years ago the Environmental Protection Agency (“EPA”) began this lawsuit by asking a court to require several companies to clean up a thirty-four acre hazardous waste site near Kingston, New Hampshire. The suit eventually consisted of two claims that are relevant here, both made under the Comprehensive Environmental Response, Compensation and Liability Act of 1980 (“CERCLA”). First, EPA said that “actual or threatened release[s]” of “hazardous substance[s]” from the site posed “an imminent and substantial endangerment to the public health or welfare or the environment.” It therefore asked the district court, in light of the problem, “to grant such relief as the public interest and the equities of the ease may require.”
The suit was complicated. The United States originally brought suit under § 7003 of the Resource Conservation and Recovery Act (RCRA),
While the litigation proceeded, EPA itself began to clean up part of the site, adding to the costs it intended to recover, and two of the defendants also began to help clean up. In the meantime, the court divided the trial itself into two phases. The first phase would determine which defendants the law required to help clean up and to pay costs. The second phase would determine precisely what further cleanup actions the law (
The court began the first phase of trial on December 5, 1983, concluded the trial on June 13, 1985, and found that the law required fifteen defendants either to engage in further cleanup or to pay part of the costs, or both.
See United States v. Ottati & Goss, Inc.,
EPA raises three sorts of legal issuеs on this appeal. First, its appeal raises a general question of statutory interpretation — a question with implications for other cases as well as this one. It asks what legal standard a court should use in deciding whether to grant the particular relief EPA requests when EPA brings an action under
the first sentence
of
Second, EPA argues that the record does not adequately support the relief determinations that the district court made. It believes that the record required that court to order IMC to clean up the site somewhat more thoroughly. EPA’s arguments are highly fact-specific and evidence-related. We have examined those portions of the record (compiled over ten years and amounting to morе than 40,000 pages) that the parties have cited in their briefs. Having done so, we conclude that the district court's determinations of fact and of proper relief are adequately supported, with one exception. The exception consists of the cleanup ordered in respect to volatile organic compounds (VOCs).
Third, EPA raises two “miscellaneous” arguments, one dealing with a district court statement about liability, the other dealing with that court’s refusal to award EPA certain indirect costs, a refusal apparently meant as a kind of “sanction” for improper behavior. We have rejected EPA’s “liability” related argument as not now properly before us. We have decided to remand the case for further explanation about the “sanction.”
In sum, we have affirmed the district court’s decision with two exceptions, each of which requires further district court proceedings. The first exception concerns VOC-related relief; the second concerns sanctions.
We shall now explain how we have reached these legal conclusions.
I
APA “Arbitrary/Capricious” Review and a CERCLA Injunction
The EPA’s initial argument — an argument with implications beyond the confines of this case — concerns application of the Administrative Procedure Act’s “arbitrary/capricious/abuse of discretion” standard in determining a proper injunctive remedy. The issue arises because, while the court was in the midst of the liability phase of the trial, EPA began an
administrative
proceeding to determine an appropriate cleanup remedy. This proceeding lasted about two years, it involved the creation of a seven-volume administrative record, and it led to an EPA document called the “Record of Decision” (ROD), U.S. Exh. 375 (Phase II), which said, among other things, that IMC should undertake the very kind of cleanup for which EPA later argued in court. Once it produced this document, EPA’s lawyers told the district court judge that, when he ordered a cleanup remedy, the law
required
him to order just what the ROD set forth, unless he found the ROD to be “arbitrary and capricious or otherwise not in accordance with law.”
1. EPA clean up/private party reimbursement,42 U.S.C. §§ 9604 , 9607. EPA itself can clean up a hazardous waste site, use money from a “Superfund” to pay the cost of clean up, and then obtain reimbursement from the site’s owners or operators (and various other private persons).Section 9604(a)(1) authorizes EPA “to act ... to remove or arrange for the removal of, and provide for remedial action relating to such hazardous substance_”Section 9607(a) makes relevant private parties “liable for ... all costs of removal or remedial action incurred by the United States Government....”
2. EPA seeks court injunction,42 U.S.C. § 9606(a) (first sentence). Where there is an “imminent and substantial endangerment to the public health or welfare or the environment,” the EPA may ask the Departmеnt of Justice “to secure such relief as may be necessary” and the “district court ... shall have jurisdiction to grant such relief as the public interest and the equities of the case shall require.” (Emphasis added.)
3. EPA order/court enforcement,42 U.S.C. § 9606(a) (second sentence) and§ 9606(b) . The second sentence of42 U.S.C. § 9606(a) gives EPA the power to “take other action ... including ... issuing such orders as may be necessary to protect public health and welfare and the environment.” Subsection (b) provides that EPA may bring an action in “district court to enforce such order” (and to obtain a fine for its violation).
4. EPA order/EPA clean up/private party reimbursement/treble damages.42 U.S.C. § 9606(a) (second sentence);§ 9604(a) ;§ 9607(a) , (c)(3). The EPA could issue an “order” under§ 9606(a) (second sentence), but, instead of asking a court to enforce the order, the EPA could conduct a cleanup itself under§ 9604(a) . It could then obtain reimbursement from relevant private parties under§ 9607 ; but, this time the private parties would be liable for treble damages, for§ 9607(c)(3) says, when a person “fails without sufficient cause” to follow a remedial “order,” he “may be liable ... for punitive damages in an amount ... not more than three times ... any costs incurred by the Fund_”
See Developments in the Law
— Toxic
Waste Litigation,
99 Harv.L.Rev. 1458, 1485-96 (1986).
See also In re Acushnet River & New Bedford Harbor,
The second set of background circumstances, relevant only to this casе, is that EPA has here pursued a combination of the first and second courses of action. Its claim for money rests upon its own cleanup expenditures and the reimbursement provided by
In this case, the United States has not issued an administrative order to any defendant and is not seeking treble damages or civil penalties_ The defendants in this case will have an opportunity to challenge EPA’s selection of a remedy without seeking additional penalties or treble damages.
Memorandum of the United States in Support of its Motion for Pretrial Ruling (district court record, no. 1064) at 26. Moreover, the “Record of Decision” refers only to future action that EPA wants the defen
These background circumstances make clear that the specific legal issue before us is
not
whether the courts, in some general or abstract sense, can overturn EPA decisions where they are not arbitrary, capricious, an abuse of discretion, or otherwise unlawful; obviously, the answer to this general question is “generally no.” When the EPA asks a court, for example, to enforce a lawful (nonarbitrary) EPA order, the court must enforce it.
See Acushnet,
First, nothing in the language of the relevant sentence of the statute suggests that the court is required to accept EPA’s decision about proper relief. The sentence simply says that the court “shall have jurisdiction to grant such relief as the publiс interest and the equities of the case may require.” The words “shall have jurisdiction” imply a discretionary legal power; the word “equities” and the words “public interest” imply court-selected, not agency-selected standards.
Cf. Amoco Production Co. v. Village of Gambell,
480 U.S 531, 541-46,
Second, when the Administrative Procedure Act uses the words “arbitrary, capricious, abuse of discretion,” it does so in speaking of the powers of a
“reviewing court”
to “set aside agency action, findings, and conclusions found to be” unlawful.
Third, to read the statute in this way does not significantly handicap EPA. In the circumstance where it needs immediate judicial relief, it can ask the court for an injunction under the first sentence of
Fourth, it is not unusual for a regulatory statute to provide
both
for an agency to enter an order, after fairly lengthy proceedings, and
also
for the agency to ask a court for preliminary relief (say, an injunction) where circumstances require.
See, e.g.,
Federal Trade Commission Act,
Fifth, the authority cited by EPA to the contrary is not persuasive. The legislative history it cites does not specifically discuss the
first
sentence of
A number of district courts have considered this question and have reached differing conсlusions. We agree with the reasoning of the courts in
United States v. Hardage,
We are not persuaded by other cases that have held that an “arbitrary and capricious” standard applies in a
For these reasons, we do not believe that the district court, in a
II
EPA has raised a host of evidence-related issues. Although the district court adopted most of EPA’s suggestions for relief, EPA believes the record required it to order yet more stringent relief in respect to contamination by 1) metals, 2) volatile organic compounds (VOCs), and 3) chemicals called “PCBs.” Obviously, reviewing a 40,-000 page record to determine whether a specific fact-based finding has adequate support poses difficult practical problems for a busy court of appeals. Recognizing that the parties have had a full opportunity to argue these matters at length in the district court, that the district court is better able than we are to resolve such issues, and that the factual issues before us concern not whether the site will be cleaned up, but rather likely concern who should pay the added cost of making it extremely clean, we have held both EPA and IMC to the specific language of their briefs. That is to say, we have reviewed the particular record pages they have cited in their briefs, and we have asked ourselves whether, in light of the district court’s greater familiarity with this case, the appellant, EPA, has convinced us that the district court acted outside its fairly broad legal powers. With one exception — concerning VOCs — we are convinced that its determinations have adequate record support.
We shall review these claims briefly here, writing only enough to explain to the parties the basis for our conclusions.
A
Metals
EPA argues that the district court shоuld have held IMC responsible for, and ordered it to clean up, deposits of four metals (iron, manganese, nickel, and arsenic) in the groundwater. The district court did not do so for two independent reasons. First, the district court recognized that EPA, in its complaint, had listed five specific hazardous chemicals that contributed to ground and water contamination, but had not listed metals. The district court also believed that EPA had not adequately in
Second, the district court nonetheless allowed EPA to introduce its “metals contamination” evidence conditionally, so that it could make a finding in case this appeals court reversed its “proof limitation” order. After the partiеs introduced evidence, the district court concluded that EPA had not proved any “metals contamination” on the site. Rather, it concluded that the site, even before any contaminating activity took place, contained natural deposits of these metals. It found that EPA had failed to show any concentration of metals higher than the natural metal deposits that already existed.
We believe the record adequately supports both these conclusions.
1. The proof-limiting order. As we have just said, the court granted IMC’s proof-limiting motion because it believed that to permit EPA to read its complaint to include a charge of “metals contamination” would unfairly surprise IMC. EPA attacks the lawfulness of the court’s “proof-limiting” order by pointing to: 1) language in its complaint (other than in paragraph 42) which states in general terms that IMC (and others for whom it is responsible) released “hazardous” materials; 2) an instance, prior to Day 56 of trial, in which it introduced evidence of metal contamination; and 3) an effort by IMC’s counsel, prior to Day 56 of trial, to cross-examine a witness abоut metal contamination. The first of these matters, the presence of general “contamination” language in the complaint, is beside the point. IMC’s “proof-limitation” motion did not rest upon the literal language of the complaint. Rather, it rested upon the fact that a) the specific list of five chemicals in the complaint, plus b) almost four years of pre-trial activity and fifty-five trial days without significant mention of its potential liability for metal contamination, plus c) the absence of any reason why the government failed to mention metals earlier, taken together, made it unfair to read the complaint to include metals or to permit a “metal-related” complaint amendment.
The second and third matters show little. The earlier instance in which the EPA told IMC about metals consists of its effort on day 43 of trial to introduce evidence that a different defendant, Geochem (for whose actions IMC is not responsible), contaminated a different site, the 0 & G site (for which IMC is not responsible), with metal deposits — a claim that the court eventually rejected when it dismissed Geochem from the suit on the ground that no one (at that other site) had released metals. The earlier instance of cross examination by IMC consisted of its cross examination on day 55 of trial, the day before it made its “surprise-based” motion.
The district court has adequate legal power to interpret general language in a complaint in light of what four years worth of legal activity indicates the parties mean by that language; and it can refuse to permit an amendment where such an amendment would unfairly surprise a party.
See, e.g., Isaac v. Harvard University,
2.
No metal contamination.
In any event the record adequately supports the district court’s substantive conclusion that there was no metals contamination. That conclusion rested upon its finding that a particular well dug on the land, well W-ll, is a “background well,” that is to say, a well that has not been contaminated by the waste site and thus reflects the natural state of the water. Well W-ll had water with high concentrations of several
EPA challenges only the first step of this reasoning. It denies that the district court could find that well W-ll was a “background” well. It points to what several documents introduced at trial show as very high metal concentrations, particularly of iron and arsenic, in well W-ll. It says that these concentrations are much higher than those present at wells outside the dump sites but they are about the same as wells on the dump sites. See U.S.Exh. 634A (Ph. II). EPA also points to testimony of its expert, Dr. Sanborn, to the effect that well W-ll was not a background well, but rather was drilled for a special “bedrock test” purpose: although the slope of the ground is such that one might expect water to flow from the spot where well W-ll was drilled toward the dump site (thus suggesting that well W-ll’s site is “background”), if bedrock below well W-ll is lower than bedrock at the dump site, water may flow backwards along the bedrock, bringing contamination to well W-11. Dr. Sanborn also said he thought that well W-ll’s concentrations do in fact reflect such contamination, not background. See Tr. Day 22 (Ph. II) at 80-84.
The problem for EPA is that IMC points tо other evidence that suggests the opposite. For example, U.S.Exhibit 448H.3 apparently compares levels of iron, nickel, arsenic and manganese at well W-ll with levels in wells W-16, W-17, and W-23 (all background wells), and also with levels in wells W-13 and W-14 (all dump site wells) and with various other wells. Some wells seem to have high concentrations of the metals; others have low concentrations— all without any obvious pattern. IMC also quotes a study done by an EPA contractor, which says that the “arsenic is more likely indicative of natural conditions in the area.” U.S.Exh. 659 (Ph. II) at 5-12. The very report that speaks of well W-ll as having been dug to test bedrock levels concludes that “subsequent analysis ... does not indicate” any flow of water from the dump site towards well W-ll, U.S.Exh. 659 at 4-2; and maps of bedrock levels do not show the level differential for which EPA argues. Indeed, the very expert upon whom EPA relies testified earlier, and more vigorously, that well W-ll was indeed a background well.
We have read the hundreds of pages of tables of numbers to which EPA and IMC have referred us, with such guidance as the parties have provided in their briefs, kеeping in mind the legal fact that EPA has the burden of persuading us that the district court’s conclusion is “clearly erroneous.” It has not done so. Thus, we find the district court’s conclusions about metals to be lawful for this second, substantive, as well as for the first, procedural, reason.
B
Volatile Organic Compounds (VOCs)
The EPA claims that the district court should have ordered IMC to clean up the soil on the 5.88 acre GLCC site further in order to reduce the concentration of several volatile organic compounds (VOCs). The district court concluded that IMC’s 1984 cleanup efforts reduced VOC concentration sufficiently, and that EPA itself should pay for any further VOC reduction efforts. EPA says 1) that the court should have measured IMC’s efforts against EPA’s “one part per million” VOC soil concentration standard, and 2) that, in any event, IMC failed to meet even the more generous “five to ten parts per million” standard that the court approved.
1.
The standard.
After reviewing the briefs and the record pages cited, we cannot say that the district court
had
to conclude that “the public interest and the equities of the case,”
On the other hand, IMC points out that the very report and expert study on which EPA relies says that a VOC level of 7 parts per million in the soil would reduce lifetime exposure cancer risks to one in a hundred thousand, a level that, in various contexts (including EPA’s basic studies), EPA asserts to be its “goal.” U.S.Exh. 375 at 21. Moreover, EPA monitored IMC’s cleanup efforts, EPA was fully aware that IMC’s method would reduce VOC soil concentrations to levels of five to ten parts per million, yet, at that time, EPA did not urge further cleanup. (See Tr. Day 59 (Ph. I) at 55, and EPA contractor note saying that the “clean excavated soil was effectively aerated,” Tr. Day 63 (Ph. II) at 48.) IMC meets EPA’s argument that lower VOC soil concentrations are now needed to protect against higher water concentrations by noting that the district court separately ordered run-off water cleanup to appropriately low levels. And it also says that the the special additional cleanup method (digging up, heating in a special machine, and then replacing about 14,000 cubic yards of soil) will cost an added several million dollars (IMC has already spent about $2.6 million), all for very little purpose (since one part per million is not significantly safer than five or ten).
There is simply no way for us to say, on the basis of this kind of general evidence, or on the basis of other highly general, tangentially related record material to which EPA refers, that the district court was wrong in its choice of standard. EPA must convince us of fact-related error, and it has failed to do so.
2. Compliance with the “five to ten parts per million” VOC standard. The district court concluded that “there is evidence in some areas that ppms total of VOC levels exceeded substantially the norm,” but “overаll ... the IMC cleanup of the soil contamination substantially cleared up the VOCs.” We take as the “norm” the “five to ten parts per million” that IMC says was its “goal.” IMC’s brief adds that “it is evident that the court accepted the propriety of IMC’s target range of 5-10 ppm when it found that TMC ... substantially cleaned up the VOCs.’ ” And we shall hold IMC to its stated arguments, just as we have done with EPA. We therefore understand EPA to argue that the record does not permit the district court to find that IMC reduced VOC levels, overall, to “five to ten parts per million,” even allowing for an occasional “hot spot” departure. We agree.
EPA points to two studies in the record, one conducted by an EPA contractor, the other representing an IMC contractor’s “check” of the EPA contractor’s sampling. After IMC finished cleaning up its site, the EPA contractor tested the results by digging a series of “test pits.” It randomly dug 62 test pits, each about eight feet long, three feet wide, and eight feet deep, evenly spaced throughout an area somewhat larger than a football field. The contractor then took soil samples from different levels within each pit. It found that soil in over thirty of the sixty-two pits contained VOCs concentrated at levels higher than ten parts per million. Many samples were much higher than ten parts per million, and the high concentration samples appeared at different levels in many pits.
IMC makes two arguments in its effort to support the district court’s conclusion. First, it says that EPA’s contractor did not test the soil in a fair way. Rather, the contractor would use a meter to detect just where in the pit concentration was the highest; then, the contractor would take soil from that place. IMC’s contractor testified that such a “hot spot” might be the size of “a softball.” Tr. Day 46 (Ph. II) at 47. Since other evidence in the record suggests that a lifetime of exposure to VOC concentration of, say, seventy parts per million would produce a cancer risk of one in ten thousand, one might ask just how a scattering of such “softballs” throughout
The difficulty with this argument is that we cannot read the record as showing no more than the occasional “softball.” IMC’s contractor’s own report indicates that thirty-three of the sixty-two wells have VOC concentrations that exceed ten parts per million (and, again, many of them exceed it by ten times or more). Although that report indicates, in respect to a few pits (pit 117, for example) that the “hot” area does not comprise the whole pit, the very fact that the contractor made notes stating, in respect to some pits, that the high concentration is found in only part of the pit, suggests that, where no such note exists, the high concentration area is fairly widespread within the pit. (The contractor testified that he would indicate in his notes whenever the “hot” area was significantly limited within a single pit. Tr. Day 46 (Ph. II) at 58-59.) The report suggests that high concentrations were limited to very small areas in only a few instances; indeed, nowhere does it indicate “softball-sized” areas; and in some instances (pit 121, for example) the report specifically notes that high concentrations exist throughout the pit (e.g., 350 ppm at two feet deep, 340 ppm also at two feet, 400 ppm at four feet in one part of the pit and 606 ppm at four feet in another part). The contractor also testified that “in other cases [the “hot spot”] ... could be half a test pit,” Tr. Day 46 (Ph. II) at 47; and he seems to have accepted a characterization of an “average” level of 87 parts per million as reasonable, id. at 112. Having read through both the “IMC” and “EPA” reports, and finding little conflict in this respect between them, we have a firm conviction that one cannot reasonably characterize the “over ten parts per million” VOC concentrations as “few” or “far between.” Rather, the studies and related testimony indicate that such “over-standard” concentrations are widespread, and in significant amounts, within the total test area.
IMC also points to testimony suggesting that it would be wasteful to require IMC to dig up the several acre site again to process 14,000 or so cubic yards of soil. An IMC expert testified that, even if nothing further is done, by the year 2000 probably more than half of the VOCs in the soil would have diffused into the atmosphere. Tr. Day 48 (Ph. II) at 43-44. Moreover, VOCs released into the groundwater will be taken care of by the groundwater treatment ordered by the court. Since IMC’s $2.6 million dollar expenditure has brought us so near absolute safety, why force expenditure of millions of dollars more to buy so little?
The problem with this second argument is that the court expressed no view about its validity or direct relevance. Rather, the court’s opinion assumes (as IMC and EPA both read it) that “five to ten parts per million” is the appropriate standard, and it goes on to say that IMC has substantially complied. We, too, therefore, express no view about this second kind of argument, except to note that the district court did not adopt it. Thus, we remand this aspect of the case to the district court so that it can devise a further VOC cleanup remedy, which, in light of its findings about danger to the public health, will adequately satisfy “the public interest.”
C
PCBs
EPA also asked the district court to order a cleanup of soil and sediments at the waste site to reduce to safe levels the concentration of a carcinogenic chemical, PCB. The court applied a standard of 50 parts per million for soil concentration, which it determined IMC’s 1984 cleanup had met, and a standard of 20 parts per million for sediment concentration, which it ordered IMC to meet. EPA now argues that the
1. The soil standard. The evidence that EPA offered to show that the soil PCB cleanup should reduce concentrations to at least fifty parts per million is strong. EPA pointed to a Toxic Substances Control Act regulation that insists (for TSCA purposes) upon incineration (or storage in special dumps) of soil with PCB concentrations greater than that. 40 C.F.R. 761.60. Of course the TSCA standard does not say that soil with less than 50 PCB parts per million is safe. Nonetheless, the evidence that EPA offered in support of its lower, 20 ppm, standard is weak. The 20 ppm evidence consists almost entirely of a two page calculation prepared by an EPA expert, Dr. Tsai (and Dr. Tsai’s explanatory testimony). In her calculation, Dr. Tsai makes clear that her conclusion — that only 20 ppm (or lower) PCB concentrations will achieve EPA’s cancer goal (a lifetime cancer risk of less than one in a hundred thousand) — rests upon her assumptions that a) developers will build residential housing on the site, b) small children, playing in the backyard, will eat dirt containing PCBs, and c) the children will eat a little bit of dirt each day for 245 days per year for three and a half years (about six ounces altogether, as we calculate it).
At the same time, IMC produced witnesses that suggested that rеsidential development was unlikely, Tr. Day 45 (Ph. II) at 73, 92; it noted that Dr. Tsai’s calculation also indicated that concentration levels as high as 70 ppm might be safe (if the children ate the dirt on only 70, rather than on 245, days per year); and it obtained testimony from EPA project manager Hohman that, in context, makes it seem as if EPA was uncertain just where to draw the line between a 70 ppm standard and a 20 ppm standard (though TSCA constrained EPA to urge a standard of 50 ppm or less). Tr. Day 41 (Ph. II) at 64-65. IMC also noted that EPA was fully aware that IMC’s own cleanup plan called for PCB reduction to 50 ppm, that EPA, at one point, called the plan “extremely commendable,” that EPA employees, contractors or officials were present throughout IMC’s cleanup, and that EPA (while not binding itself to accept IMC’s cleanup as sufficient) nonetheless said nothing at the time about reducing PCB levels further. Finally, IMC’s brief, and EPA’s studies, make clear that IMC’s cleanup, reducing PCB and VOC soil levels to 50 ppm and 10 ppm respectively, cost about $2.6 million; to reduce those levels further, to 20 ppm PCBs and 1 ppm VOCs, will cost an additional $9.3 million. One might conclude from the cited portions of the record that this amounts to a very high cost for very little extra safety.
On balance, after reading the cited portions of the record, we have no “firm conviction,”
United States v. United States Gypsum Co.,
2.
Implementation of the soil standard.
EPA argues that IMC’s cleanup did not meet even the 50 ppm PCB standard. Its evidence that this is so consists of analyses of soil samples taken from soil several feet below the surface in five of the 63 test pits. Three samples exceeded 50 ppm (56, 134, and 143 ppm, respectively). U.S. Exh. 445A (Ph. II) at table 45. The explanatory testimony to which EPA also points, however, indicates that these samples were not taken randomly. Rather the five samples were taken from areas where the experts found “discolored soils.” That being so, one might reasonably conclude that EPA could find evidence in оnly three of the 62 test pits of PCB concentrations that exceed the standard. (Although EPA’s expert also said that there were
3. PCB “sediment"standard. The district court required IMC to clean the marsh and the south brook to the point where PCB concentration in the sediments would not exceed 20 parts per million. EPA argues that the standard is too lenient: It says that the court should have imposed a standard of one part per million PCBs.
The EPA supports its argument with citations to only a handful of statements. Its own administrative report simply says in highly general language that there should be a standard of one part per million in order to protect birds, fish, and mammals, as well as human beings. U.S. Exh. 375 (Ph. II) at 25. The study upon which the EPA based its report offered several alternative remedies, only one of which recommended that same standard for that same reason. U.S. Exh. 450A (Ph. II) at 22. And Dr. Tsai testified that she chose that same standard, again, for the same reason.
IMC correctly points out, however, that the study upon which EPA based its report also, in another alternative remedy, recommended a sediment PCB standard of 50 ppm. U.S. Exh. 450A at 63. It points to a discussion in the record of a federal fish and wildlife study that finds no “significant health risk” at the pond, despite the potential 20 ppm PCB sediment concentrations. And it adds that EPA has nowhere explained how, or why, animal and human protection requires a more stringent standard. Indeed, out of the entire record, EPA mentions only a handful of highly general, conclusory sentences about the PCB sediment standard. (Kg., the EPA’s basic study simply says without elaboration, that the “action level of 1 ppm for PCB’s in sediments was selected after discussions with State and federal officials because of the potential for bioaccumulation of these PCB’s by wetland biota.” U.S. Exh. 450A at 22.) The only more concrete explanation given is that of Dr. Tsai, who, after repeating the same general conclusory sentence, added that EPA considered a “two parts per million” standard for concentration in fish. But Dr. Tsai did not explain the relation between PCB concentration in the sediment at the bottom of a pond and concentration within a fish.
In our view, the district court was entitled to hold EPA to somewhat higher standards of detailed proof than those met in the pages we have cited. And, in the absence of that proof, given the other (also skimpy) indications in the record that less stringent “alternatives” would be adequate, we believe the court could reasonably choose a 20 parts per million standard. That is to say, we cannot find that such a standard is “clearly erroneous” or its choice unreasonable.
Ill
Other Issues
1.
EPA “joint and several” liability.
The district court specifically ordered IMC to carry out “EPA’s selected groundwater remedy,”
i.e.,
the work necessary to clean up the groundwater at the GLCC site.
GLCC and IMC are also jointly and severally liable [with respect to dirty groundwater under the GLCC site] with ... the EPA.
EPA concedes that the statute permits a court to find the government itself liable for a release of hazardous waste.
See
We shall not resolve this argument on this appeal. As far as we can .tell, the court’s “liability finding” has no practical effect in respect to the judgment of the district court that is here before us under review. EPA has not pointed out any way in which this “liability” finding has made any difference, either to the speсific relief that the court ordered in its judgment or to the money or costs that it awarded. As far as we can tell, EPA simply fears that this “finding” might collaterally estop it someday from defending itself should IMC bring a different legal action seeking an EPA “contribution” to its groundwater cleanup costs.
See
EPA also wants to appeal a statement in the district court’s liability opinion that “IMC is not responsible for or liable for any releases into the environment or effects upon the public health or environment from 1955 until May, 1973” (when it bought the GLCC site from a company called Kingston Steel Drum).
2.
Liability for EPA’s “indirect costs.
” The district court awarded, to EPA, to the State of New Hampshire, and to others, costs that they incurred in helping to clean up the sites — costs for which IMC, in part, was responsible.
$336,922.00 are indirect costs which include expenses for rent, utilities, supplies, clerical staff and other overhead expenses. These indirect costs necessary to operate the Superfund program cannot be attributed directly to the O & G/GLCC sites, and are therefore disallowed.
There is ample evidence of record why this court allowed the State of New Hampshire’s indirect costs and disallowed indirect costs of the United States. To add further fuel to the fire, the United States has the temerity to state there is no reason for its ustulation [i.e., being“burned”] by the court. Without reitеrating what is self-evident, the cavalier and hubristic actions of EPA in this litigation warrant the use of punitive measures by the court. This litigation appears to be interminable and this ruling may be considered a harbinger if future court hearings are necessitated relative to monitoring.
The United States has on many occasions since it commenced litigation on May 15, 1980 done the following: ignored court orders, delayed the progress of this case, resulting in sanctions being imposed in the liability phase, attempted to recoup the amount of the sanctions in this phase of the trial, led the court to assume it had authority to settle when it had none, at the time making a mockery of a settlement conference, when former defendant, Geochem’s motion to dismiss was granted, attempted to recover sums expended in that segment of the litigation against the remaining defendants and a general insouciance about judicial time with general oblivion to the demands of time on the defendants.
Id. at 998.
EPA, pointing to cases involving comparable cost awards, argues that the law requires the district court to award it “administrative overhead,”
i.e.,
a proportionate share of rent, utilities, administrative staff costs, etc., that are not readily allocable to one specific, rather than some other specific, cleanup site.
See, e.g., United States v. Northernaire Plating Co.,
We agree with EPA thаt ordinarily courts should allow recovery of these indirect costs. We also agree with IMC that the district court denied these costs as a sanction, that district courts ordinarily have the legal authority to impose such sanctions,
see Roadway Express, supra,
and that the record raises serious questions related to the effective administration of environmental protection laws. Why, for example, has this case taken ten years to litigate? The issues are complex, but not unfathomable. Why has the government not found a way to express its technical problems in English
{e.g.,
“small children will eat tiny amounts of dirt when they play in a yard”), instead of relying upon mazelike patterns of cross-references among regulations, statutes, and “expert jargon”? Has the government itself caused a significant amount of contamination through negligent cleanup efforts? Has the government, in fact, spent enormous administrative (and judicial) resources in an effort to force improvement from “quite clean” (10 ppm VOC, and 50 ppm PCB) to “extremely clean,” at three to four times the “quite clean” costs? Indeed, hаs the government taken account of the fact that preparing briefs and then asking an appeals court to work through such a maze, in the face of a forty or fifty thousand page record, takes considerable
To find a case worrying, however, to believe that “there must be a better way,” to wonder about the government’s priorities in the face of other, apparently more serious, environmental demands for “cleanup” time and effort — is not to find the type of governmental activity that may call for sanctions. And the parties hotly contest the existence and the unreasonableness of each specific instance of misconduct that arguably calls for sanctions. (The government, for example, says that IMC’s claim that it “was barred from discussing settlement with the governmеnt” is “false;” while IMC replies that its claim “is true,” and “it is wrong for EPA to label as false a statement that some of its own lawyers know to be true.”)
We simply cannot determine from this vast record just what it is that EPA or its lawyers may have done that is wrong and warrants the sanction of denying indirect costs; and, we do not want to guess, in so serious a matter, just what is in the district court’s mind. We therefore remand the case on this issue for the court to redetermine the “sanctions” matter, after which, if it assesses sanctions, it will state the specific factual, and legal, basis upon which they rest, including the reason to equate the amount of the sanction with the amount of the indirect costs.
Conclusion
For the reasons stated, we order the district court to do the following: 1) The court will reconsider its order in respect to VOC cleanup; it will amend that order to require IMC to clean up VOCs in the soil at the GLCC site to a level that it determines “public health” and the “public interest” require. 2) The court will reconsider the matter of “indirect costs,” explaining, as we have set forth above, any denial of those costs as a sanction. In all other respects the judgment of the district court is affirmed.
Affirmed in part, vacated in part, and remanded for further proceedings indicated in this opinion.