EPA's Delisting Program for Hazardous Wastes: Current Limitations and Future Directions

Karen Florini, Richard A. Denison, and Peter F. Rathbun

Editors' Summary: "Delisting" is the process by which the Environmental Protection Agency (EPA) excludes a particular facility's waste from otherwise applicable lists that would designate it as hazardous under the Resource Conservation and Recovery Act (RCRA). Delisting is available for wastes that do not warrant management as hazardous, due to plant-specific variations in processes, raw materials, or other factors. The delisting process has been subject to some controversy, and in the 1984 RCRA amendments Congress attempted to ensure that delisting would not result in deregulating a waste that may still be hazardous. In this Article, the authors evaluate EPA's delisting program, with special emphasis on whether EPA has met Congress' objectives as set out in the 1984 amendments. The authors analyze EPA's reliance on a particular model that estimates potential for groundwater contamination and conclude that the model is extremely limited and disregards many potential routes of exposure to contamination. The authors also argue that EPA's delisting program neglects RCRA's land ban restrictions and clean closure requirements. The authors conclude that EPA already has the authority to correct many of the problems in the program, but that Congress may want to consider the issue during the upcoming RCRA reauthorization process.

Karen Florini is an attorney, Richard A. Denison is a senior scientist, and Peter F. Rathbun is a law clerk in the Environmental Defense Fund's Toxics Program.

[19 ELR 10558]

In 1976, amid growing concerns about "the effect on the population and the environment of the disposal of discarded hazardous wastes,"¹ Congress enacted the Resource Conservation and Recovery Act (RCRA).² The centerpiece of RCRA is subtitle C, which provides a comprehensive, cradle-to-grave regulatory system for hazardous wastes from initial generation to final disposal.³ All remaining wastes, which are putatively nonhazardous, are regulated under the much less stringent provisions of subtitle D.⁴ This bifurcated scheme of regulation will be effective in protecting human health and the environment only if solid wastes that are in fact hazardous are accurately classified as such, so that they will be managed according to subtitle C. Congress delegated to the Environmental Protection Agency (EPA) the tasks of identifying the specific characteristics of hazardous waste and listing particular categories of wastes as hazardous.⁵

When EPA promulgated its first group of hazardous waste listings in 1980, it also devised a process for delisting a waste from an individual facility.⁶ Delistings were to be available for wastes that, although meeting the terms of a designated hazardous waste category, did not appear to warrant management as a hazardous waste because of plant-specific variations in processes, raw materials, or other factors.⁷ Over the next few years, several hundred delisting petitions were filed, but EPA took final action on only a few.⁸ The Agency also granted over 150 temporary delistings; such a delisting, which could be granted if EPA concluded that there was a "substantial likelihood" that a final exclusion eventually would be granted, allowed the waste to be handled as nonhazardous while EPA completed full review of the petition.⁹

Eight years after the passage of RCRA, Congress enacted sweeping amendments that substantially overhauled the Act.¹⁰ The scope of the amendments reflected in part two key congressional concerns: first, that many dangerous wastes were escaping regulation simply because EPA had never designated them as hazardous,¹¹ and second, "that land disposal of hazardous waste [was] not providing, and in some cases cannot provide, protection [19 ELR 10559] against groundwater contamination and in many cases poses grave threats to public health and the environment."¹²

To cure these and other shortcomings, Congress enacted detailed, highly prescriptive amendments designed to ensure that "the RCRA regulatory and enforcement program [would] be conducted in a manner that controls and prevents present and potential endangerment to public health and the environment."¹³ Among other objectives, Congress sought to provide for a more aggressive approach to defining wastes as hazardous, as well as for stricter controls on land disposal of hazardous waste.

Congress also attempted to correct defects in the delisting program. In particular, Congress sought to ensure that "there would no longer be a risk that delisting a waste means releasing a waste which may still be hazardous from regulation."¹⁴ Congress' skeptism about the reliability of land disposal heightened its concerns about the adequacy of the delisting program, since most delisted wastes go to nonhazardous waste land disposal facilities that are far less carefully controlled than facilities governed by RCRA subtitle C.¹⁵

This Article examines the status of EPA's delisting program and its relationship to other aspects of the RCRA program, with emphasis on the congressional objectives articulated in the 1984 amendments. As the term suggests, delisting is the inverse of listing, so the Article first briefly summarizes the listing process. It next describes the basic components of the delisting program and then presents an analysis of a variety of shortcomings in the program as implemented to date. Finally, the Article outlines suggestions for strengtheningthe program.

Background: The Designation of Wastes as Hazardous

A given waste can be classified as hazardous either by being listed as such by EPA or by exhibiting one of the characteristics defined by EPA.¹⁶ To date, EPA has listed approximately 110 waste streams as hazardous. A waste may be listed if it may cause serious illness, if it contains one or more of the substances on a designated list of toxic constituents,¹⁷ or if it exhibits one or more of the defined characteristics.¹⁸

The Agency has created two types of listings: non-specific wastes, which are generated by a variety of industries (e.g., certain spent solvents), and specific wastes, which are produced by a particular industry (e.g., distillation bottoms from production of aniline).¹⁹ Once listed, not only the waste stream itself but also any materials derived from the waste or mixtures containing the waste — in any amount — are deemed to be listed wastes as well.²⁰ When listing a waste, EPA identifies its primary hazardous property as toxic, reactive, ignitable, corrosive, or acutely hazardous.²¹ In addition, about 600 generic chemical products are listed and become subject to regulation when discarded, spilled, burned, or applied to land.²²

A waste is also hazardous if it exhibits any of the four characteristics defined by EPA: ignitability, corrosivity, reactivity, and extraction procedure (EP) toxicity.²³ These characteristics are quite narrow; for example, the toxicity characteristic covers only eight metals and six pesticides.²⁴ Generators are obliged to determine whether a waste exhibits any of these characteristics.²⁵

As a practical matter, not all wastes that fall within a listed category are necessarily hazardous. Some facilities use different processes or raw materials than others within the same industry and thus produce wastes of differing toxicity. In addition, wastes that fall within a listed category as a result of the derived-from or mixture rule may differ substantially from the initial waste that was the subject of the listing.

To provide a way out of the hazardous waste system for such materials, if they are in fact nonhazardous, EPA created the delisting process. Any generator may petition for delisting of wastes produced at its facility. Because disposal and other management costs are considerably higher for hazardous wastes than for nonhazardous wastes, generators have a great incentive to pursue delistings.

As discussed above, a waste can be hazardous either by virtue of being listed or by virtue of exhibiting any of the designated characteristics. Once delisted, a facility's waste is not automatically exempt from regulation under RCRA subtitle C, but rather is only removed from the lists. Thus, generators of delisted wastes must continue to determine whether their wastes exhibit any of the characteristics of hazardous waste specified in the subtitle C regulations.²⁶ Any [19 ELR 10560] waste that does so must be managed as a hazardous waste, regardless of whether or not it has been excluded from the lists.

The Delisting Program

Statutory and Regulatory Context

Congress gave its implicit imprimatur to the delisting process in the 1984 RCRA amendments when it enacted certain procedural and substantive changes but left EPA's basic concept unaltered.²⁷ In particular, the 1984 amendments disallowed the Agency's practice of granting temporary delistings even before completing a full review or providing public notice.²⁸ Public notice and opportunity to comment are now required for all EPA delistings.²⁹

In addition, the amendments required EPA to reassess all exclusions granted previously, using somewhat more stringent standards; any delisting not reaffirmed by November 1986 (following notice and comment) would lapse automatically.³⁰ Most of the activity in EPA's delisting program during 1985-86 involved reassessments of the pre-1984 delistings; EPA subsequently denied many of those petitions.³¹

Unfortunately, the 1984 amendments were silent as to the status of delistings previously granted by the 25 states that had been authorized to implement RCRA in lieu of EPA.³² At least 14 of those states had active delisting programs and some of them had granted temporary and final delistings using EPA's former criteria.³³ Although the temporary state delistings expired along with the federal ones in November 1986, EPA has interpreted RCRA not to require that states reevaluate the final pre-1984 delistings under the more stringent standards of the 1984 amendments.³⁴ Recently, some states have received authorization to carry out the more stringent delisting provisions under the 1984 amendments, and at least one state has recently granted a delisting petition.³⁵

As revised following the 1984 amendments, EPA's delisting regulations set forth two requirements that must be satisfied before a waste may be delisted:

(1) The petitioner must demonstrate … that the waste produced by a particular generating facility does not meet any of the criteria under which the waste was listed as a hazardous … waste; and

(2) … the Administrator must determine, where he has a reasonable basis to believe that factors (including additional constituents) other than those for which the waste was listed could cause the waste to be a hazardous waste, that such factors do not warrant retaining the waste as a hazardous waste.³⁶

The latter requirement was added by the 1984 amendments, closing a loophole that "historically [had] resulted in some wastes which are still hazardous being exempted from the hazardous waste lists" because those wastes, although still containing hazardous constituents, no longer met the criteria for which they were originally listed.³⁷ Depending on the waste's hazard type as assigned by EPA when initially listing the waste (i.e., toxic, ignitable, reactive, corrosive, or acutely hazardous), additional requirements also apply. Most delistings to date have involved wastes designated as toxic. For such wastes, a petitioner must demonstrate either that the waste does not contain the constituent(s) that caused it to be listed, or that the waste "is not capable of posing a substantial present or potential hazard to human health or the environment when improperly treated, stored, transported or disposed of, or otherwise managed."³⁸

Agency Practice

The steps in the delisting process are relatively simple:³⁹ A generator submits a petition containing information on its manufacturing processes, the amount and characteristics of the waste stream, and analytic data on the presence of various constituents. EPA reviews the petition and may request supplemental information.⁴⁰ Once the Agency finishes reviewing the completed petition, it publishes a proposed grant or denial of the petition and allows a public comment period, generally 45 days.⁴¹ EPA then publishes a final grant or denial. A grant can be made conditional on certain steps being taken, such as ongoing testing of the waste. All excluded wastes are listed, along with any conditions limiting the exclusion, in the Code of Federal Regulations.⁴²

[19 ELR 10561]

As already noted, most of EPA's attention through late 1986 was devoted to reevaluation of the pre-1984 petitions. In 1987, EPA published only a handful of delisting notices, but the program gained momentum beginning in early 1988. Between early 1988 and mid-1989, EPA published over 50 notices on delisting actions (i.e., final grants, final denials, proposed grants, or proposed denials).⁴³

Unfortunately, the Agency's substantive review of delisting petitions has some serious deficiencies. Specifically, EPA has chosen to base its delisting decisions almost exclusively on a particular groundwater model that disregards other exposure routes and suffers from a variety of methodological flaws. In addition, and perhaps of greater long-term significance, EPA is applying its delisting methodology in a manner that undercuts other critical components of RCRA. Both problems are explored below.

The Vertical and Horizontal Spread (VHS) Model

For each of the delisting proposals published during 1988 and the first half of 1989,⁴⁴ EPA relied heavily on the results of a particular environmental fate and transport model. That model, known as the vertical and horizontal spread (VHS) model,⁴⁵ attempts to estimate the potential for groundwater contamination following the leaching of certain toxic constituents from a waste disposed of at an unlined landfill.⁴⁶ The model predicts concentrations of contaminants at a hypothetical drinking water well located 500 feet downgradient from the disposal site.⁴⁷ If the predicted concentrations exceed levels of regulatory concern (e.g., drinking water standards), the delisting petition is denied.⁴⁸

EPA has expressed the rationale underlying its use of the VHS model for evaluating delisting petitions as follows:

EPA believes that this fate and transport model represents a reasonable worst-case waste disposal scenario for the petitioned waste, and that a reasonable worst-case scenario is appropriate when evaluating whether a waste should be relieved of the protective management constraints of RCRA subtitle C. Because a delisted waste is no longer subject to hazardous waste control, the Agency is generally unable to predict and does not control how a waste will be managed after delisting. Therefore, EPA current believes that it is inappropriate to consider extensive site-specific factors. For example, a generator may petition the Agency for delisting of a metal hydroxide sludge which is currently being managed in an on-site landfill and provide data on the nearest drinking water well, permeability of the aquifer, dispersivities, etc. If the Agency were to base its evaluation solely on these site-specific factors, the Agency might conclude that the waste, at that specific location, cannot affect the closest well, and the Agency might grant the petition. Upon promulgation of the exclusion, however, the generator is under no obligation to continue to manage the waste at the on-site landfill. In fact, it is likely that the generator will either choose to send the delisted waste off site immediately, or will eventually reach the capacity of the on-site facility and subsequently send the waste off site to a facility which may have very different hydrogeological and exposure conditions.⁴⁹

Unfortunately, the comprehensiveness of EPA's words has not been matched by its deeds in evaluating individual delisting petitions. Rather, EPA appears to rely exclusively on the VHS model in evaluating delisting petitions: all of the delistings that EPA has recently proposed to grant "pass" the model's test, while all of the petitions that EPA has proposed to deny (for reasons other than incompleteness) fail that test.

Because the VHS model is designed to predict only one attribute — namely a waste's potential for contaminating a drinking water well located 500 feet downgradient — EPA implicitly disregards the waste's potential to injure human health through any route of exposure other than consumption of contaminated drinking water.⁵⁰ The model also disregards all environmental effects, such as harm to wildlife. Further, the VHS model suffers from a number of shortcomings even in its ability to predict groundwater contamination. These flaws are discussed at length below.

Problems With EPA's Delisting Program

Failure to Consider Original Listing Criteria

EPA's own regulations require every delisting petitioner to "demonstrate … that the waste … does not meet any of the criteria under which the waste was listed."⁵¹ Most of the delisting petitions recently considered by EPA involved wastes listed because of the presence of "toxic" constituents. According to the original background documents supporting these listings, many of these wastes were listed because of concerns about total toxic constituent levels, not merely because of their potential to leach those toxins into groundwater.

[19 ELR 10562]

None of EPA's delisting proposals have expressly evaluated the hazard from total concentrations of toxic constituents in a delisting determination, even though total concentrations were often identified as a factor in listing each of the wastes. For example, when EPA listed wastewater treatment sludges from electroplating operations as hazardous waste, the Agency's first consideration in support of the listing was that these wastes "contain significant concentrations of the toxic heavy metals chromium, cadmium, and nickel, and highly toxic cyanide."⁵² To dispel any doubt that this listing was based on more than leachability, the Agency also said that:

concentrations for chromium and cadmium in the EP extract from this waste from individual sites might be less than 100 times the national interim primary drinking water standard [i.e., the level that defines the waste as hazardous under the EP toxicity characteristic]…. Nevertheless, the Agency believes that there are factors in addition to metal concentrations in leachates which justify the T listing … [including] the presence of … cyanide … in high concentrations … [and] the nondegradability of the three heavy metals and the high concentrations of cadmium and chromium in actual waste streams.⁵³

Although the Agency did not specify a precise level of concern, it characterized levels exceeding 1,000 parts per million (ppm) as being "significant concentrations" of cadmium, chromium, and nickel.⁵⁴ In an analogous listing, cyanide levels exceeding 1,000 ppm were "deemed to be of regulatory concern."⁵⁵

Despite this unambiguous record demonstrating that total constituent concentration was one of the primary listing criteria for these electroplating wastes, EPA recently proposed delisting these wastes at seven sites without explicitly considering the total concentrations of the constituents of concern.⁵⁶ At one site, cyanide concentrations in the waste were measured at 6,300 ppm, more than six times above the level identified by EPA as being of concern when the Agency originally listed the waste.⁵⁷ Another petitioner was not even required to provide total constituent data.⁵⁸ Delisting proposals for other wastes have followed a similar pattern.

An especially egregious example of EPA's refusal to heed its own listing criteria is found in three proposals to delist mercury-bearing wastes.⁵⁹ When these wastes were listed, the Agency said that EP toxicity test results were "not determinative in making a listing determination."⁶⁰ Nevertheless, when the Agency proposed to delist these wastes, the proposals were based solely on an evaluation of leaching potential, using EP toxicity test data as the input to the VHS model.

Neglect of Exposure Pathways Other Than Groundwater Contamination

By disregarding total constituent levels of contaminants and instead focusing exclusively on leachability, EPA has neglected a host of exposure pathways such as inhalation of airborne toxicants, direct dermal contact, and consumption of contaminated surface waters or fish taken from such waters. These pathways can be important during handling, storage, and transport, as well as following disposal of a delisted waste.⁶¹ Exposure through these routes can be significant, particularly since such exposures can occur in addition to consumption of contaminated drinking water, thereby multiplying the adverse effect on human health.

For instance, it is well established that a major pathway by which humans, especially children, are exposed to lead is by getting contaminated dusts or soils on their hands and subsequently into their mouths.⁶² Similarly, the primary route of human exposure to persistant organic chemicals is seldom groundwater contamination; rather, it is typically through consumption of contaminated foodstuffs, such as fish that have absorbed the contaminants downstream from a discharge source.⁶³

Further, airborne and waterborne exposures can represent a substantial risk, independent of a waste's potential to cause groundwater contamination. Moreover, EPA implicitly disregards the potential for exposure by multiple pathways such as airborne exposures in addition to groundwater consumption.

The preceding points have been presented to EPA during the public comment periods for several delisting proposals. In its final determinations on those proposals EPA has offered various responses.⁶⁴

[19 ELR 10563]

In addressing criticisms regarding failure to consider total constituent levels, EPA asserted that the metal constituents are tightly bound within the waste matrix and thus pose no threat.⁶⁵ EPA also asserted that the potential for airborne exposure was insignificant because some of the wastes, once treated, would be in a "monolithic and nonfriable" cement-like matrix,⁶⁶ while others would have a water content above 65 percent.⁶⁷ In addition, EPA dismissed the potential for surface-water contamination by asserting that hazardous constituent levels "will tend to be lower than the levels in EP leachate analyses," and would be further diluted in the surface-water body.⁶⁸

Unfortunately, there is no assurance that these wastes will remain in a nonfriable or wet condition over the long term. This is particularly a concern if wastes are mismanaged during storage, handling, or transit, or eventually discarded somewhere other than in a well-designed, well-operated landfill. Since EPA loses virtually all regulatory control over a waste once delisted, such mismanagement is distinctly possible. As the continuing exposures resulting from use of lead paints decades ago illustrate, metal-containing matrices may degrade over time, releasing toxic metals into the environment long after disposal. Thus, just as EPA's initial listing decisions examined both total and leachable constituent levels, so should its delisting determinations in order to evaluate long-term risks.

Finally, the Agency's response on surface-water issues again assumes that the constituents of concern must leach from the waste in order to contaminate surface water. This response ignores the fact that waste particles themselves, with toxic constituents attached, can directly enter surface waters through runoff or erosion.

In sum, EPA's narrow focus on a waste's potential to cause groundwater contamination provides little assurance that delisted wastes are in fact nonhazardous. Because controls over delisted wastes both prior to and subsequent to the time of disposal are minimal at best, wastes delisted under the current methology may continue to pose significant threats to health and the environment.

Lack of Consistency with Other RCRA Provisions

Even beyond EPA's failure to consider alternative exposure routes, the Agency's heavy reliance on the VHS model in evaluating delisting petitions creates substantial anomalies in implementing the RCRA program as a whole. The current delisting program has the potential to significantly undercut two key aspects of RCRA: the land disposal restrictions and the clean-closure program. Each problem is discussed in turn.

Land disposal restrictions: In the 1984 RCRA amendments, Congress concluded that "to avoid substantial risk to human health and the environment, reliance on land disposal should be minimized or eliminated."⁶⁹ To effect this objective, Congress directed EPA to promulgate regulations specifying waste treatment standards that "substantially diminish the toxicity of the waste or substantially reduce the likelihood of migration of hazardous constituents from the waste so that short-term and long-term threats to human health and the environment are minimized."⁷⁰ Hazardous wastes that have been so treated are then — and only then — eligible for disposal on land. Even after treatment, the wastes must be placed in a hazardous waste landfill that holds a subtitle C permit.⁷¹

In response to this congressional mandate, EPA proposed two alternative methodologies for setting treatment standards — one risk-based and the other technology-based. Under the risk-based approach, EPA proposed to develop a "screening level" for each hazardous constituent and to "identify the maximum concentration below which the Agency believes there is no regulatory concern for the land disposal program and which is protective of human health and the environment."⁷² By contrast, under the technology-based approach, EPA would set standards reflecting the constituent concentrations achievable by applying the best demonstrated available treatment technology (BDAT), with the proviso that at least "a minimum performance standard, in the form of 'substantial' reductions in toxicity and/or mobility, must be achieved."⁷³ The BDAT standard may be set on the basis of total constituent concentrations, leachable constituent concentrations, or both.

Responding to these proposals, several commentators (including 11 members of Congress who had played leading roles in enacting the 1984 amendments) insisted that "because of the scientific uncertainty inherent in risk-based decisions, Congress had intended that treatment standards should be based on the capabilities of existing technology."⁷⁴ As a result of these comments, EPA adopted the BDAT approach.⁷⁵

According to these regulations, restricted wastes may be land disposed only if the constituent concentrations in the waste (or, for some wastes, an extract of the waste) do not exceed specified standards reflecting the levels attainable by applying BDAT. When such treated wastes are land disposed, they must be disposed in a permitted hazardous waste facility, with safeguards against migration of remaining hazardous constituents from the landfill.

Unfortunately, in reviewing delisting petitions, EPA seemingly disregards the entire land-disposal restrictions program. EPA has repeatedly proposed to delist wastes that contain constituents above the BDAT levels established for those wastes. In doing so, EPA creates the [19 ELR 10564] following noteworthy anomaly: a given waste, once delisted, can lawfully be placed in any landfill in the nation, except perhaps a hazardous waste landfill.⁷⁶

This approach does not effectuate the congressional intent that prompted enactment of the land-disposal restrictions program, to avoid reliance on inherently uncertain risk assessments when evaluating the suitability of wastes for land disposal. Given Congress' view that such techniques should be avoided in setting treatment standards for wastes to be disposed in relatively secure hazardous waste landfills, risk-based techniques plainly have even less place in evaluating disposal hazards from virtually unregulated solid waste landfills.

These concerns are more than hypothetical. As mentioned above, EPA has recently granted or proposed to grant several delistings that did not meet BDAT standards:

BDAT Actual

Constit- Std Conc. % Above

Petitioner⁷⁷ uents (ppm) (ppm) Std

Roanoke Electric Steel Lead 0.24 0.27 12.5

CF&I Steel Lead 0.24 0.251 4.6

Marquette Electronics Nickel 0.32 3.3 931.3

Clay Equipment Cadmium 0.066 0.120 81.8

Perox Chromium 0.094 0.16 70.2

By adopting a risk-based approach in evaluating petitions for delisting wastes that are subject to technology-based standards, EPA has created a delisting standard that in many cases will be easier to satisfy than the BDAT standard governing land disposal under RCRA subtitle C. This is an open invitation to hazardous waste generators to use the delisting process as a means of escaping RCRA's land-disposal restrictions — hardly the result envisioned by Congress when it created the land-disposal program.

Clean closure: At the end of the active life of a hazardous waste facility, the facility must undergo closure pursuant to EPA requirements. There are two distinct closure tracks: closure by removal of the wastes (known as clean closure) and closure leaving waste in place. In the latter case, the facility owner or operator must comply with a variety of closure and post-closure requirements, including conducting 30 years of post-closure care at the site.⁷⁸ Such care includes groundwater monitoring and facility maintenance, activities that entail significant expense. By contrast, once a site has been clean closed, owners bear no further obligations. Unlike landfills, which are generally intended as permanent disposal sites, facilities such as surface impoundments and waste piles may — indeed, must where feasible — undergo clean closure.⁷⁹ In order to achieve clean closure, the owner or operator must remove or decontaminate not only all waste residues, but also all contaminated containment system components, subsoils, structures, and equipment, and must manage them as hazardous waste.⁸⁰

EPA has interpreted "remove and decontaminate" in the context of surface impoundments;⁸¹ this interpretation would appear to be equally applicable to the parallel language found in the standards for closure of waste piles.⁸² Rather than mandating removal of all detectable contamination at closure, the Agency allows "limited quantities of hazardous constituents [to] remain … [if they] present only insignificant risks to human health and the environment."⁸³ To ensure that these risks are in fact insignificant, closure demonstrations must "document that the [remaining] contaminants … will not impact any environmental media including ground water, surface water, or the atmosphere in excess of Agency-recommended limits or factors, and that direct contact through dermal exposure, inhalation, or ingestion will not result in a threat to human health or the environment."⁸⁴ Moreover, the demonstration must assume that potential exposures will occur "at or within the unit boundary."⁸⁵

These criteria contrast sharply with the current delisting methodology. As noted above, the VHS model evaluates only the drinking water exposure pathway; it also assumes that the nearest groundwater well is 500 feet downgradient from the site, rather than analyzing the impact at the unit boundary. The delisting demonstration thus is far less rigorous than the multi-pathway demonstration required to qualify for clean closure.

According to EPA, this disparity is warranted because closure by removal completely removes the site from regulatory control, whereas delisted solid waste will continue to be regulated under RCRA subtitle D.⁸⁶ As a practical matter, however, municipal and industrial solid waste disposal sites are currently devoid of significant federal regulatory oversight, and are subject only to minimal regulation in most states.⁸⁷

Thus, an operator who wishes to close a hazardous waste surface impoundment or waste pile could perform minimal cleanup — just enough to allow the remaining waste to pass muster under the relatively lenient delisting methodology — and [19 ELR 10565] escape further meaningful regulation. EPA has already proposed granting at least one delisting petition that would, in effect, enable the petitioner to circumvent closure.⁸⁸ This undercutting of the clean closure standards contravenes RCRA's mandate of ensuring protection of human health and the environment.

Shortcomings of the VHS Model in Predicting Drinking Water Contamination

The VHS model has notable shortcomings even in predicting groundwater contamination. Two problems arise from basic assumptions used in the model, while another results from the model's inability to process certain data. These three factors in combination reduce, and may even eliminate, the model's "conservatism" often cited by EPA.⁸⁹ The model thus could actually underestimate potential contamination, even though the model does incorporate certain conservative aspects.⁹⁰

The distance-to-well parameter: As constructed, the VHS model evaluates the potential for groundwater contamination at a hypothetical well 500 feet downgradient from the disposal site, which is presumed to be an unlined municipal solid waste landfill regulated under EPA's subtitle D regulations.⁹¹ The model implicitly assumes that no downgradient wells are located less than 500 feet away from the disposal site. Yet there is no federal requirement for a 500-foot buffer zone, either in the current subtitle D criteria or in revisions proposed in 1988.⁹² While some states require buffer zones of a variety of sizes, such requirements are not universal.

EPA derived the 500-foot assumption from an informal survey of 38 state agencies that estimated the proximity of drinking water wells to municipal landfills. As EPA expressly acknowledged in introducing the VHS model, however, 24 percent of the states surveyed believe that they have drinking water wells within 500 feet of landfills; indeed, 11 percent believe they have wells less than 200 feet away.⁹³

EPA rather cavalierly characterized its use of the 76th percentile as "a reasonable worst case." In effect, however, it means that the VHS model may underestimate potential groundwater contamination at up to 1,500 sites nationwide.⁹⁴

Indeed, in revising the federal criteria for municipal solid waste landfills, EPA collected data on existing facilities and found that the median distance from landfill to the nearest drinking water well was approximately 1,370 feet. Moreover, the median distances for Virginia and Ohio are both 1,000 feet, while that for Massachusetts is 797 feet. In fact, the maximum distance to wells in the Massachusetts database is only 1,295 feet.⁹⁵

The "single source" assumption: In evaluating a delisting petition, EPA directly compares the contaminant levels at the hypothetical well predicted by the VHS model, which reflects the volume of the initial waste and its leach-ability, to the regulatory levels of concern for these contaminants. The regulatory levels of concern are drinking water standards and similar health-based EPA guidelines; each level reflects the maximum "safe" amount of that contaminant in drinking water. If the predicted concentrations do not exceed those standards, the waste is eligible for delisting.

EPA's approach implicitly assumes that the petitioner's waste is the only source of contaminants that ever reaches the hypothetical drinking water well. But if the waste is disposed of in a municipal solid waste landfill (as the VHS model presumes and as can legally occur for any delisted waste), it is unlikely that the delisted waste will be the only source of leachate from the landfill. MoreoveR, other sources in addition to the landfill may also contribute contaminants to the hypothetical well.

Thus, by comparing the concentrations predicted by the VHS model for this one source of contamination directly to the regulatory levels of concern for drinking water — which reflect the "safe" level of a contaminant in drinking water from all sources — EPA again relies on an assumption that cannot fairly be characterized as a reasonable worst case. EPA thus may delist a waste when disposal of that waste in a nonhazardous waste landfill may contribute to groundwater contamination.

EPA's actual implementation of the delisting program bears out these concerns. For example, EPA has proposed to grant a delisting petition where the predicted groundwater contamination at the hypothetical 500-foot downgradient well caused by petitioner's waste alone would be 90 percent of the regulatory level of concern.⁹⁶

In response to public comments raising this issue, EPA has merely stated that it lacks "any technical basis to support a determination of an appropriate percentage reduction [below the level of concern] and believes that, without a technical basis, any resulting percentage would be arbitrary."⁹⁷ The Agency further stated that "in light of the conservative nature of the VHS model, EPA will continue [19 ELR 10566] to allow wastes to exhibit [compliance-point concentrations] up to 100 percent of the health-based standard."⁹⁸

As demonstrated in the preceding and following sections, however, the VHS model has several aspects that are far from conservative. Moreover, while choosing a percentage-reduction figure may have an element of arbitrariness, so does EPA's current approach. Indeed, because a particular landfill will almost inevitably contain multiple contaminant sources, and because solid waste landfills are frequently located in areas containing other potential or actual sources of groundwater contamination, EPA's use of a zero-percent reduction is perhaps the least supportable approach.

The model's volumetric limitations: In addition to the distance-to-well and single source assumptions, the VHS model is incapable of predicting the behavior of even moderate volumes of waste. EPA further exacerbates this problem by using annual waste volumes even for wastes generated on a multi-year basis.

As the volume of any given waste increases, the quantity or degree of contamination of leachate from that waste will be greater. Therefore, the VHS model predicts that larger waste volumes will tend to produce a greater impact on the underlying groundwater. Specifically, as the amount of waste increases up to about 8,000 cubic yards (yd[3]), the model predicts a corresponding decrease in contaminant dilution between the point of release and the hypothetical receptor well.⁹⁹ However, for waste volumes greater than 8,000 yd[3], the VHS model predicts no further reduction in dilution. EPA has acknowledged that wastes above this amount "probably would have a greater than predicted impact at the compliance point," but asserts that it has insufficient information to modify the model to account for larger quantities of waste.¹⁰⁰

For wastes generated on an ongoing basis, the Agency seriously compounds the model's underestimation with respect to larger waste quantities by considering the amount of waste generated in a single year. EPA itself recently acknowledged that large quantities of waste (such as those accumulating over multiple years) that leach a contaminant at a particular level will have a greater impact on an underlying aquifer than a small amount of waste leaching at the same (or even higher) level.¹⁰¹ In other words, as waste volume increases, hazard potential also increases. Nonetheless, EPA's analysis of ongoing delistings is based solely on data from one year of waste production.

EPA responded to criticism of its approach by stating that it considers "the use of annual or one-time waste volumes to be sufficiently conservative since it is a reasonable worst-case for a petitioner to dispose of one year's waste in a single landfill cell at one time."¹⁰² EPA further claimed that delisted wastes subsequently generated and disposed of would probably be placed in additional landfill cells, mixed with other wastes, and "due to this long-term mixing the effect on the underlying aquifer would be reduced."¹⁰³

This reasoning is far from persuasive. First, EPA's own discussion of the volume limitations of the VHS model implicitly acknowledges that the model underestimates the impact on an underlying aquifer of waste volumes in excess of 8,000 yd[3], which cannot reasonably be assumed to be disposed of in a single trench or cell of the dimensions assumed by the model. The same argument applies to wastes generated over more than one year: the model cannot predict the impact of more than a single trench filled with at most 8,000 yd[3] of waste.¹⁰⁴ Second, EPA is apparently assuming that different cells overlie different aquifers; yet the first year's cell does not stop leaching just because the second year's cell also begins to do so.

Moreover, even if one were to accept these highly questionable assumptions, there is simply no basis for EPA to assume (especially in a reasonable worst-case scenario) that waste generated in different years will go to different cells. EPA offers no data indicating that most municipal landfills operate cells with only a single year of capacity, or that these calls are hydraulically isolated from one another. Indeed, the vast majority of existing municipal landfills lack any liners whatsoever,¹⁰⁵ supporting an assumption opposite that made by EPA.

Thus, EPA's approach is likely to cause future compliance-point concentrations to be significantly under-estimated, even if the initial waste volume is within the useful range of the VHS model. Moreover, some of EPA's recent delisting proposals exceed the model's useful range even in a single year because they involve wastes having a volume over 8,000 yd[3] annually.¹⁰⁶ Other proposed delistings involve wastes that would reach the 8,000-yd[3] useful maximum within a few years.¹⁰⁷

[19 ELR 10567]

Finally, even aside from the model's volumetric limitations, delistings as currently granted for ongoing wastes appear to have no quantitative restrictions keyed to waste volumes.¹⁰⁸ As part of a delisting petition, a generator must estimate the annual volume of waste produced and affirm the validity of that estimate under penalty of perjury. However, after a delisting is granted, there are no express limits on a generator's ability to increase the annual waste volume. An increase may well occur, for example, following an expansion of the underlying production process. Although a delisting automatically expires if the underlying process (and at least potentially the waste) is qualitatively changed, no expiration is triggered by a quantitative change, regardless of its magnitude. Indeed, generators need not even report annual volumes of delisted wastes.

Improper Use of the VHS Model for Evaluating Wastes in Surface Impoundments

EPA's implementation of the delisting program is also flawed by the Agency's use of the VHS model to evaluate wastes disposed of in surface impoundments. Recently, EPA has used the model to evaluate seven delisting petitions involving surface impoundments.¹⁰⁹ EPA claims that the VHS model is "conservative in all of its assumptions" and therefore provides "an additional margin of safety" that justifies its application to surface impoundments even though the model was designed to simulate leaching from a landfill.¹¹⁰

The Agency's rationale for using the VHS model in the absence of a surface-impoundment-specific model has several weaknesses. To begin with, EPA argues that the increased hydraulic head (i.e., pressure of overlying liquid that enhances leachate movement) expected in a surface impoundment will be at least partially offset by several other mechanism: sorption¹¹¹ and retardation of dissolved contaminants, and clogging of impoundment bottom materials, all of which EPA claims will result in slower migration of contaminants from a surface impoundment than would occur in a landfill.¹¹² These assumptions range from highly questionable to demonstrably wrong.

First, as a result of the increased hydraulic head, the movement of leachate from a typical surface impoundment is many times greater than that from a comparably situated landfill.¹¹³ EPA asserts, without any support, that this effect can be offset by "sorption and retardation of dissolved contaminants with the aquifer solids encountered through migration in the ground water."¹¹⁴ As EPA's own description indicates, these two mechanisms operate not in the unit, but beneath it. Regardless of their effectiveness in reducing contaminant levels reaching a receptor, there is no basis for assuming that they have a greater effect beneath a surface impoundment than beneath a landfill. Thus, these mechanisms are entirely irrelevant in comparing surface impoundments to landfills; accordingly, they provide no basis for EPA's assertion that they justify the use of the VHS model for a surface impoundment because they would offset the effect of increased hydraulic head.

Moreover, EPA explicitly rejected consideration of attenuation factors when devising the VHS model:

Based on the Agency's evaluation, it was believed that assuming no attenuation is a reasonable worst-case…. Inorganic contaminants, such as metals, are substances which undergo transformation and complexation, but do not degrade in the environment. Thus, they are likely to escape into the environment…. [N]umerous studies have demonstrated that, when a solution containing a metal is continuously applied to a soil column, the metal eventually "breaks through." That is, the concentration of metal entering is exactly equal to the concentration exiting.¹¹⁵

EPA has never offered any rationale for assuming a greater degree of attenuation underneath a surface impoundment than underneath a landfill, and no such rationale seems likely to be defensible.

EPA also envisions clogging as a new mechanism operating in surface impoundments that would counteract increased hydraulic head. While clogging can be presumed to act within the unit, EPA has never presented any evidence of the magnitude of this effect or of its greater effectiveness in a surface impoundment than in a landfill. To the extent that clogging reduces leachate migration, there is no reason to assume that it would not also take place in a landfill, where significant portions of the waste are likely to be comprised of small particles. Indeed, EPA [19 ELR 10568] has recently acknowledged the comparability of clogging beneath landfills and surface impoundments.¹¹⁶

In sum, the only justifiable assumption is that the increased hydraulic head in a surface impoundment will cause higher compliance-point concentrations than would be expected from a landfill. EPA's use of the VHS model is thus particularly inappropriate in such situations.

Lack of Verification Testing

All of the shortcomings outlined above are exacerbated by yet another factor: EPA typically accepts a petitioner's data without verifying it, though it conducts site visits to spot-check some petitioners' data. While all petitioners must submit an affidavit averring to the accuracy of the statements made in the petition, it is not evident whether this is an effective compliance tool; as the U.S. General Accounting Office has noted, "[b]ecause of the added costs involved in handling a waste if it is designated as hazardous, it is in the interest of a facility to obtain a delisting."¹¹⁷

As of 1986, the Agency had found errors at 70 percent of the few facilities it had visited.¹¹⁸ The General Accounting Office reported that EPA program managers characterized the problems found on the site visits "as ranging from minor inaccuracies to blatant misrepresentations."¹¹⁹ GAO observed that "[w]ith so few site visits, there is little to deter a petitioner from submitting inaccurate information in order to obtain a delisting."¹²⁰

Recommendations and Conclusions

As implemented to date, EPA's delisting program suffers from significant shortcomings that may well result in the delisting of wastes that still meet RCRA's definition of hazardous wastes: ones that "may … pose a substantial present or potential hazard to human health or the environment when improperly treated, stored, transported, or disposed of, or otherwise mismanaged."¹²¹ Those shortcomings include EPA's failure to consider seriously the potential for exposure through pathways other than consumption of contaminated groundwater; the presumption that the delisted wastes will be disposed of in municipal landfills at which the only source of leachate will be the delisted waste itself and that no other source of contaminants will affect the hypothetical well; and the use of a fate and transport model that cannot evaluate large quantities of wastes and that assumes the nearest drinking water well is 500 feet away.

Moreover, the delisting program is being implemented in a fashion that allows the undercutting of objectives of other important RCRA programs, particularly those involving land disposal restrictions and clean closure. While EPA has acknowledged those disparities, it dismisses them as unimportant, with apparently no concern for the long-range programmatic implications.

All of these problems are exacerbated by the permanent nature of delistings now granted by the program: currently, delistings, like diamonds, are forever. There is no mechanism at all for a waste to be reevaluated despite subsequent changes in relevant regulatory levels of concern, improvements in the VHS model, or development of more rigorous pre-disposal treatment requirements.

EPA plainly has the authority and ability to correct many of these problems. If it fails to do so, Congress has the opportunity to hone the delisting process during the upcoming RCRA reauthorization and to ensure that the delisting program is not implemented in a fashion that circumvents other RCRA programs, such as the land ban and closure programs. Moreover, the program should be restructured to provide for limited-term delistings that, when up for renewal, could incorporate improvements in hazardous waste technologies and regulations.

Congress should also create a fee system, payable by petitioners, to finance the delisting program. Unlike other aspects of the RCRA program, delisting confers a distinct economic benefit upon the generator who presents a successful delisting petition. While society as a whole may gain from a well-operated delisting program — primarily through the reduction of the overall amount of waste destined for the nation's limited hazardous waste treatment and disposal capacity — the generator's individual benefit is far more immediate. Given EPA's abysmal record in meeting other RCRA deadlines,¹²² the delisting program should not be allowed to compete for resources that are needed for more urgent tasks. By relieving competition for resources, a fee-based system may even result in faster review of delisting petitions, thus benefiting generators as well. In addition, the fee should cover routine independent testing by EPA or its contractors to verify the data submitted in the delisting petition. Such an approach is the only way to ensure the integrity of the delisting program.

The delisting program plays an important role in allowing wastes that present no hazard to avoid being subject to the strictures and expense of RCRA's hazardous waste regulations. But an unduly narrow evaluation of delisting petitions can lead to precisely the same sort of problems as those created by failing to bring wastes within the hazardous waste system in the first place. Such errors do "little more … than to contribute to future burdens" on the nation's hazardous waste cleanup program.¹²³

1. H.R. REP. NO. 1491, 94th Cong., 2d Sess. 3, reprinted in 1976 U.S. CODE CONG. & ADMIN. NEWS 6238, 6241.

2. RCRA §§ 1002-11002, 42 U.S.C. §§ 6901-92k, ELR STAT. RCRA 001-050.

3. See RCRA §§ 3001-18, 42 U.S.C. §§ 6921-39, ELR STAT. RCRA 010-025; 40 C.F.R. pts. 260-270 (1988). Specifically, a hazardous waste is statutorily defined as any solid waste that may:

(A) cause, or significantly contribute to an increase in mortality or an increase in serious irreversible, or incapacitating reversible, illness; or

(B) pose a substantial present or potential hazard to human health or the environment when improperly treated, stored, transported, or disposed of, or otherwise managed.

RCRA § 1004(5), 42 U.S.C. § 6903(5), ELR STAT. RCRA 005. The process for listing hazardous wastes is discussed infra at notes 16-25 and accompanying text.

4. As implemented to date, the federal regulation of nonhazardous wastes amounts to little more than a ban on open dumping. See RCRA §§ 4004(a), 4005(a), 42 U.S.C. §§ 6944(a), 6945(a), ELR STAT. RCRA 028; see also U.S. ENVIRONMENTAL PROTECTION AGENCY, REPORT TO CONGRESS: SOLID WASTE DISPOSAL IN THE UNITED STATES (1988) [hereinafter REPORT TO CONGRESS].

5. RCRA § 3001(a), 42 U.S.C. § 6921(a), ELR STAT. RCRA 010.

6. 45 Fed. Reg. 33073, 33076-77 (May 19, 1980) (current version codified at 40 C.F.R. § 260.22 (1988)).

7. 45 Fed. Reg. 33084, 33116 (May 19, 1980).

8. U.S. GENERAL ACCOUNTING OFFICE, HAZARDOUS WASTE: EPA HAS MADE LIMITED PROGRESS IN DETERMINING THE WASTES TO BE REGULATED 38 (1986).

9. 45 Fed. Reg. at 33077 (former version of 40 C.F.R. § 260.22(m)).

10. Hazardous and Solid Waste Amendments of 1984, Pub. L. No. 98-616, 98 Stat. 3221 (1984).

11. H.R. REP. NO. 198, 98th Cong., 2d Sess. 19, reprinted in 1984 U.S. CODE CONG. & ADMIN. NEWS 5576, 5578.

12. Id.

13. Id. at 20, 1984 U.S. CODE & ADMIN. NEWS at 5579.

14. Id. at 58, 1984 U.S. CODE & ADMIN. NEWS at 5576, 5617.

15. See REPORT TO CONGRESS, supra note 4.

16. This section greatly simplifies a complex topic. For additional background, see R. FORTUNA & D. LENNETT, HAZARDOUS WASTE REGULATION: THE NEW ERA (1987); Garelick, EPA's Definition of Solid Waste: Making Distinctions Between Shades of Gray, 17 ELR 10349 (1987).

17. See 40 C.F.R. § 261.10 & App. VIII (list of hazardous constituents) (1988).

18. See infra at notes 23-25 and accompanying text.

19. See 40 C.F.R. § 261.31 (1988) (non-specific wastes); 40 C.F.R. § 261.32 (1988) (specific wastes).

20. 40 C.F.R. § 261.3(c)(2) (1988) (derived-from rule); id. § 261.3(a)(2)(iv) (1988) (mixture rule). Both rules were promulgated as part of the original 1980 RCRA regulations package; litigation on each was stayed for a considerable time, but is now proceeding. Shell Oil Co. v. Environmental Protection Agency, No. 80-1532 (D.C. Cir. June 12, 1989) (order requiring submission of petitioners' briefs by Sept. 5, 1989).

21. The acutely hazardous designation has special significance. While most hazardous wastes are essentially exempt from regulation when generated in quantities of less than 100 kilograms a month, the threshold is one kilogram a month for acutely hazardous wastes. See 40 C.F.R. § 261.5(a), (e)(1) (1988).

22. 40 C.F.R. § 261.33 (1988).

23. 40 C.F.R. §§ 261.20-261.24 (1988).

24. The metals are arsenic, barium, cadmium, chromium, lead, mercury, selenium, and silver; the pesticides are endrin, lindane, methoxychlor, toxaphene, 2,4-D, and 2,4,5-TP. 40 C.F.R. § 261.24 (1988). For the last several years, EPA has been actively (albeit very slowly) working on an expanded characteristic for toxicity, the so-called organic toxicity characteristic (OTC); this characteristic would be assessed via the toxicity characteristic leaching procedure (TCLP). The OTC/TCLP was proposed in June 1986, 51 Fed. Reg. 21648. In October 1988, the Agency estimated that the final OTC/TCLP, covering 38 additional chemicals, would be issued in December 1988, but the final regulation has not been issued as of this writing. In a recent court filing, EPA proposed a schedule under which it would issue the final OTC in November 1989. See EPA's Motion for Partial Summary Judgment & Lowrance Affidavit P12, Environmental Defense Fund v. Reilly (D.D.C. filed June 12, 1989) (No. 89-0598).

25. 40 C.F.R. § 262.11 (1988).

26. Although EPA requires petitioners to demonstrate that the wastes do not exhibit any characteristics at the time of the delisting (see 40 C.F.R. § 260.22(c)(1), (d)(3), (e)(3) (1988)), this does not absolve petitioners from the responsibility of determining whether the waste exhibits a characteristic after it has been delisted. In practice, it generally should not exhibit a characteristic, since delisting evaluations should be performed on representative waste samples; however, generators are responsible for determining whether the waste does in fact exhibit a characteristic. See 40 C.F.R. 260.22(a)(2) (1988).

27. See RCRA § 3001(f), 42 U.S.C. § 6921(f), ELR STAT. RCRA 011. See generally Compton and Patterson, Delisting Hazardous Wastes — Do the RCRA Amendments Spell Relief?, 14 ELR 10374 (1984).

28. See supra note 9 and accompanying text.

29. RCRA § 3001(f), 42 U.S.C. § 6921(f), ELR STAT. RCRA 011.

30. Id.

31. See U.S. GENERAL ACCOUNTING OFFICE, supra note 8, at 36 ("EPA reports that only about 22 percent of the temporary delistings it issued meet the new criteria and will be granted a final delisting").

32. State authorization is provided for by RCRA § 3006, 42 U.S.C. § 6926, ELR STAT. RCRA 019; see 40 C.F.R. pts. 271, 272 (1988).

33. See U.S. GENERAL ACCOUNTING OFFICE, supra note 8, at 39.

34. Id.

35. See Comments of the Environmental Defense Fund re: Proposed Grant of Delisting Petition Submitted by Waste-Tech Services, Inc., Kimball Nebraska Thermal Treatment Unit, submitted as letters to Mr. Ken Kolthoff, Nebraska Department of Environmental Control (Feb. 15, Mar. 13, and May 11, 1989). Those comments identified several areas of concern, including the state's use of a site-specific fate and transport model.

36. 40 C.F.R. § 260.22(a) (1988).

37. H.R. REP. No. 198, 98th Cong., 2d Sess. 58, reprinted in 1984 U.S. CODE CONG. & ADMIN. NEWS 5576, 5617.

38. 40 C.F.R. § 260.22(d)(1) (1988).

39. EPA has published a guidance manual for generators who wish to petition for delistings. ENVIRONMENTAL PROTECTION AGENCY, OFFICE OF SOLID WASTE, PETITIONS TO DELIST HAZARDOUS WASTES: A GUIDANCE MANUAL (Apr. 1985). For information on obtaining the manual, see 50 Fed. Reg. 21607 (May 28, 1985).

40. The Agency can dismiss outright a grossly deficient petition. 53 Fed. Reg. 6822 (Mar. 3, 1988).

41. Before publishing a proposed denial, EPA contacts the petitioner and offers him or her the opportunity to withdraw the petition. This allows EPA to save some resources and allows the petitioner to avoid any possible negative publicity from publication of a proposed denial. See Silverman, Delisting Hazardous Wastes Under RCRA: A Response to Compton and Patterson, 15 ELR 10006, 10007 n.12 (1985).

42. See 40 C.F.R. pt. 261, App. IX (1988).

43. Since some petitions are withdrawn by the petitioner before a proposed dispositionis published (see supra note 41), this critique of EPA's program is based solely on the published notices available to the authors.

44. During the same time period, EPA also published final decisions on several delistings proposed prior to 1988. Because the authors did not participate in reviewing those proposals, those delistings are not discussed in this Article.

45. The model was published at 50 Fed. Reg. 7896 (Feb. 26, 1985). In early 1988, the D.C. Circuit held that EPA had given "the effect of a rule to its 'VHS model' … without having exposed the model to the comment opportunities required for rules by the Administrative Procedure Act." McLouth Steel Products Corp. v. Thomas, 838 F.2d 1317, 1319, 18 ELR 20473, 20474 (D.C. Cir. 1988). To cure this procedural defect, EPA has re-proposed the model in the context of each individual delisting proposal published since the McLouth ruling.

46. In evaluating several delisting petitions for wastes located in surface impoundments, EPA used the VHS model because it has no model for surface impoundments; as discussed below, there are reasons for questioning the appropriateness of this approach. See infra text accompanying notes 109-116.

47. The VHS model has only two petition-specific input variables: the volume of the waste and the concentrations of contaminants in a waste extract. The concentrations are determined in one of two ways. For contaminants covered by the EP toxicity test (i.e., the eight metals and six pesticides discussed supra in note 24), the EP test is performed on waste samples and the resulting maximum extract concentrations are used. For other organic contaminants, an organic leaching model is used. That model estimates the concentration of contaminants that would be present in leachate as a function of the concentration of those contaminants in the waste and their aqueous solubilities. The organic leaching model was published as a final rule in late 1986. 51 Fed. Reg. 41082 (Nov. 13, 1986).

48. The approximately 150 chemicals examined in the delisting methodology are compiled in the Docket Report on Health-Based Regulatory Levels and Solubilities Used in the Evaluation of Delisting Petitions (June 8, 1988) (unpublished document, part of EPA delisting program docket).

49. 53 Fed. Reg. 50551 (Dec. 16, 1988) (emphasis added). Substantially similar language appears in other delisting notices.

50. See also infra text accompanying notes 51-60 (discussing EPA's failure to consider total constituent levels).

51. 40 C.F.R. § 260.22(a)(1) (1988) (emphasis added).

52. Environmental Protection Agency, Background Document, Resource Conservation and Recovery Act, Subtitle D — Hazardous Waste Management, § 3001 — Identification and Listing of Hazardous Waste, Book 11, at 82-83 (1980) (unpublished background document for electroplating wastes) [hereinafter Background Document].

53. Id. at 94-95 (emphasis added). For discussion of the EP toxicity test, see supra note 24.

54. Background Document, supra note 52, at 90.

55. Id. at 121 (listing document for spent waste cyanide solutions and sludges).

56. See 53 Fed. Reg. 15704 (May 3, 1988) (U.S. Nameplate Co.), final exclusion granted 53 Fed. Reg. 37759 (Sept. 28, 1988); 53 Fed. Reg. 26283 (July 12, 1988) (North American Philips Consumer Electronics Corp.), final exclusion granted 54 Fed. Reg. 15938 (Apr. 20, 1989); 53 Fed. Reg. 26455 (July 13, 1988) (Goodyear Tire and Rubber Co.), final exclusion granted 53 Fed. Reg. 47692 (Nov. 25, 1988); 53 Fed. Reg. 36070 (Sept. 16, 1988) (Clay Equipment); 53 Fed. Reg. 45106 (Nov. 8, 1988) (Marquette Electronics, Inc.), final exclusion granted 54 Fed. Reg. 15935 (Apr. 20, 1989); 53 Fed. Reg. 47731 (Nov. 25, 1988) (Brush Wellman, Inc.); 53 Fed. Reg. 48655 (Dec. 2, 1988) (Boeing Commercial Airplane Co.).

57. 53 Fed. Reg. 36070, 36073 (Sept. 16, 1988) (Clay Equipment).

58. See 53 Fed. Reg. 47731 (Nov. 25, 1988) (Brush Wellman, Inc.).

59. See 53 Fed. Reg. 29058 (Aug. 2, 1988) (waste generated by Vulcan Materials Co.); 53 Fed. Reg. 45112 (Nov. 8, 1988) (waste generated by Occidental Chemical Corp.); 53 Fed. Reg. 49680 (Dec. 9, 1988) (waste generated by BF Goodrich Intermediates Co.).

60. Environmental Protection Agency, Listing Background Document: Chlorine Production 20 (no date) (unpublished manuscript contained in EPA rulemaking docket for the listing of chlorine production wastes) (emphasis added).

61. Indeed, EPA itself acknowledged the importance of these pathways in originally listing certain wastes. For example, in listing an emission-control dust, EPA stated that "airborne exposure to lead, chromium, or cadmium particulates escaping from mismanaged emission control dusts is another pathway of concern. These minute particles could be dispersed by wind if dusts are piled in the open, placed in insecure landfills or improperly handled during transportation." Background Document, supra note 52, at 761.

62. See, e.g., U.S. PUBLIC HEALTH SERVICE, AGENCY FOR TOXIC SUBSTANCES AND DISEASE REGISTRY, THE NATURE AND EXTENT OF LEAD POISONING IN CHILDREN IN THE UNITED STATES: A REPORT TO CONGRESS (July 1988).

63. See, e.g., U.S. EPA, Estimating Exposures to 2,3,7,8-TCDD (external review draft document, March 1988).

64. 54 Fed. Reg. 5081 (Feb. 1, 1989) (VAW, Inc.); 54 Fed. Reg. 11706 (Mar. 22, 1989) (Roanoake); 54 Fed. Reg. 15935 (Apr. 20, 1989) (Marquette Electronics); 54 Fed. 19888 (May 9, 1989) (CF&I Steel); 54 Fed. Reg. 21941 (May 22, 1989) (Bethlehem Steel).

65. 54 Fed. Reg. at 5082, 11708, 15936, 19989, 21943.

66. 54 Fed. Reg. at 11708, 11989, 21943.

67. 54 Fed. Reg. at 5082, 15936.

68. 54 Fed. Reg. at 11708, 11990, 15936, 21943.

69. RCRA § 1002(b)(7), 42 U.S.C. § 6901(b)(7), ELR STAT. RCRA 004.

70. RCRA § 3004(m)(1), 42 U.S.C. § 6924(m)(1), ELR STAT. RCRA 014.

71. RCRA § 3004(m)(2), 42 U.S.C. § 6924(m)(2), ELR STAT. RCRA 014.

72. 51 Fed. Reg. 1602, 1611 (Jan. 14, 1986).

73. Id.

74. Id. at40572, 40578 (Nov. 7, 1986) (emphasis added).

75. To date, two groups of treatment standards have been promulgated, covering roughly two-thirds of the listed waste categories. See 53 Fed. Reg. 31138 (Aug. 17, 1988); 54 Fed. 26594 (June 23, 1989). EPA faces a mandatory deadline of May 1990 for issuing the final group of standards. See RCRA § 3004(g)(4)(C), 42 U.S.C. § 6924(g)(4)(C), ELR STAT. RCRA 014. Several challenges to the regulations that have been promulgated so far remain pending. See Hazardous Waste Treatment Council v. Environmental Protection Agency, No. 86-1657 (D.C. Cir. petition for review filed Dec. 3, 1986); Hazardous Waste Treatment Council v. Reilly, No. 87-1492 (D.C. Cir. petition for review filed Sept. 16, 1987); and American Petroleum Institute v. Environmental Protection Agency, No. 88-1606 (D.C. Cir. petition for review filed Aug. 19, 1988).

76. Some questions have arisen as to whether a delisted waste remains subject to the land ban, or whether the waste is beyond the reach of the land disposal restrictions, since it is no longer deemed hazardous. In the latter event, an equally notable anomaly would exist: for two physically and chemically identical wastes, one of which had been delisted but the other of which had not, the former could be placed in a hazardous waste landfill without meeting the BDAT standards but the latter could not.

77. Compare 54 Fed. Reg. 1189 (Jan. 12, 1989) (proposal) (Perox); 54 Fed. Reg. 11706 (Mar. 22, 1989) (Roanoke); 54 Fed. Reg. 15935 (Apr. 20, 1989) (Marquette); 54 Fed. 19888 (May 9, 1989) (CF&I); 54 Fed. Reg. 31675 (Aug. 1, 1989) (Clay Equipment) with 53 Fed. Reg. 31138 (BDATs).

78. 40 C.F.R. pt. 264, subpart G (1988).

79. 40 C.F.R. § 264.228 (1988) (surface impoundments); id. at 264.258 (waste piles).

80. 40 C.F.R. § 264.228(a), 264.258(a) (1988).

81. See 52 Fed. Reg. 8704 (Mar. 19, 1987).

82. See 40 C.F.R. § 264.258 (1988) (closure of waste piles); 40 C.F.R. § 264.228 (1988) (closure of surface impoundments).

83. 52 Fed. Reg. 8704, 8706 (Mar. 19, 1987).

84. Id. (emphasis added).

85. Id. at 8707.

86. Id.

87. See REPORT TO CONGRESS, supra note 4. There are over 27,000 industrial subtitle D units. Id. at 36. EPA concluded that existing federal and state regulations for these facilities are inadequate, pointing out that "Federal and some State solid waste regulations lack the following essential requirements: location criteria, appropriate design criteria, ground-water monitoring, corrective action, closure and post-closure care, and financial responsibility." Id. at ES-2. In particular, use of design controls at industrial waste facilities is "very limited" and numerous violations of state standards have been documented. Id. at ES-2 and 37-38. Moreover, available data on these facilities are sketchy at best. Id. at ES-3. To help fill some of these data gaps, EPA has proposed to collect basic information on industrial facilities. See 53 Fed. Reg. 33314, 33398 (Aug. 30, 1988) (proposed rule requiring completion of two-page form by industrial facilities). That final rule is expected to be issued in late 1989.

88. See 53 Fed. Reg. 48655, 48657 (Dec. 2, 1988) (petition by Boeing Commercial Airplane Co. to exclude "residually contaminated soils … remaining after the excavation of a sludge pile and underlying soils" containing electroplating sludge.) A similar approach might potentially be used to circumvent landfill closure requirements based on the limited analysis required for delisting. In addition, EPA recently granted a petition that involved untimely closure activities. 54 Fed. Reg. 31675 (Aug. 1, 1989) (Clay Equipment). Though it acknowledged the delay, EPA stated it "believes that whether or not Clay violated the closure regulations should not directly effect the delisting decision." Id. at 31679.

89. See, e.g., 53 Fed. Reg. 47731, 47735 (Nov. 25, 1988) (discussing conservative aspects of VHS model as basis for using that model to evaluate surface impoundments); 50 Fed. Reg. 7895, 7899 (Feb. 26, 1985) (describing model as conservative).

90. For example, the model assumes no attenuation or degradation of constituents during transit to the aquifer.

91. See 50 Fed. Reg. 7896 (Feb. 26, 1985) ("The following discussion assumes disposal at an unlined municipal landfill").

92. The current criteria for municipal solid waste landfills are found at 40 C.F.R. pt. 257 (1988). Amended criteria were proposed in mid-1988. 53 Fed. Reg. 33314 (Aug. 30, 1988). Those criteria are scheduled for final promulgation in December 1989. 53 Fed. Reg. 42537 (Oct. 24, 1988) (semi-annual regulatory agenda).

93. 50 Fed. Reg. 7898 (Feb. 26, 1985).

94. In 1988, EPA estimated that there were about 6,500 municipal solid waste landfills in the United States. REPORT TO CONGRESS, supra note 4, at 13.

95. These figures were derived by analyzing a partial printout of the actual subtitle D (solid waste landfill) risk model input values. The printout was appended as Attachment 3 to the Comments of the Environmental Defense Fund on EPA's Proposed Solid Waste Disposal Criteria (Docket No. F-88-CMLP-FFFFF) (filed Nov. 20, 1988).

96. 53 Fed. Reg. 48655 (Dec. 2, 1988) (Boeing Commercial Aircraft Co.).

97. 54 Fed. Reg. at 11708 (Mar. 22, 1989) (Roanoke); id. at 19890 (May 9, 1989) (CF&I Steel); id. at 21944 (May 22, 1989) (Bethlehem Steel).

98. Id. at 11708, 19890, 21994.

99. 54 Fed. Reg. 31475, 31479 (Aug. 18, 1989) (grant of final delisting to BF Goodrich). Originally, EPA stated that the VHS model indicated very little additional dilution above 2,000 yd[3]. 50 Fed. Reg. 48886, 48899 (Nov. 27, 1985). EPA now states that its original description of the VHS model was "not accurate" in indicating virtually no further dilution above 2,000 yd[3]; rather, according to EPA, the useful maximum is 8,000 yd[3]. 54 Fed. Reg. 31479. In fact, very little additional dilution is predicted as waste volume increases from 2,000 to 8,000 yd[3]: the dilution factor only decreases by 30 percent despite a four-fold increase in waste volume. Id. Hence, for practical purposes, EPA's original statement is still accurate.

100. 54 Fed. Reg. at 31478.

101. Id.

102. Id.

103. Id.

104. EPA further attempts to rationalize its position by stating that "the reason that the dilution factor remains constant after the waste volume exceeds 8,000 cubic yards is a function of the assumptions made in the disposal unit dimensions for the VHS model." Id. While true, this statement is circular reasoning: when setting the disposal unit dimensions, EPA explicitly stated they were inadequate to account for larger volumes of waste, but asserted that the Agency "does not have sufficient information to assess this impact, and will therefore continue to consider the scenario of a single 40-ft wide trench." In sum, for large volumes of waste, the fact remains that the VHS model understates the environmental consequences of delisting.

105. U.S. ENVIRONMENTAL PROTECTION AGENCY, OFFICE OF SOLID WASTE, CENSUS OF STATE AND TERRITORIAL SUBTITLE D NON-HAZARDOUS WASTE PROGRAMS (1986).

106. 53 Fed. Reg. 23661, 23665 (June 23, 1988) (Goodyear) (estimated production of 50,000 yd[3]); 53 Fed. Reg. 29067, 29070 (Aug. 2, 1988) (Clay Equipment Corp.) (estimated ongoing production of 34,500 yd[3] of waste annually).

107. See, e.g., 53 Fed. Reg. 26455, 26458 (July 13, 1988) (Goodyear) (estimated ongoing production of 1,920 yd[3] of waste annually); 54 Fed. Reg. 1189, 1192 (Jan. 12, 1989) (Perox, Inc.) (estimated ongoing production of 4,800 yd[3] of waste annually). The final delisting recently granted to BF Goodrich, 54 Fed. Reg. 34175 (Aug. 18, 1989), also raises this issue, since BF Goodrich's estimated annual production of waste is 6,420 yd[3]. See 53 Fed. Reg. 49680, 49684 (Dec. 9, 1988).

108. In several delisting proposals, EPA's preamble has suggested that changes in volume would require submission of a new petition. 53 Fed. Reg. 54117 (Nov. 8, 1988) (Occidental); 53 Fed. Reg. 47736 (Nov. 25, 1988) (Brush Wellman II); 54 Fed. Reg. 30413 (July 20, 1989) (USX). In other proposed and final delistings, however, any mention of volume restrictions is conspicuous by its absence. Moreover, none of the final delistings contain volume limits as part of the actual regulatory language.

109. See 53 Fed. Reg. 15704 (May 3, 1988) (U.S. Nameplate Co.); 53 Fed. Reg. 36070 (Sept. 16, 1988) (Clay Equipment Corp.); 53 Fed. Reg. 37601 (Sept. 27, 1988) (Merck & Company); 53 Fed. Reg. 37803 (Sept. 28, 1988) (Weirton Steel Corp.); 53 Fed. Reg. 37808 (Sept. 28, 1988) (Brush Wellman Corp.); 53 Fed. Reg. 47731 (Nov. 25, 1988) (Brush Wellman Corp.); 53 Fed. Reg. 50550 (Dec. 16, 1988) (Fisher Guide).

110. See, e.g., 53 Fed. Reg. 50550, 50554 (Dec. 16, 1988).

111. Sorption refers to the taking up or holding of substances by physical or chemical means.

112. See, e.g., 53 Fed. Reg. 50550, 50554 (Dec. 16, 1988).

113. Specific assumptions and calculations are spelled out in Comments of the Environmental Defense Fund on a Proposed Exclusion of Hazardous Wastes in Response to a Delisting Petition Submitted by Brush Wellman, Inc., EPA RCRA Docket No. F-88-BWEDFFFFF (Jan. 9, 1989); 53 Fed. Reg. 47731 (Nov. 25, 1988) (proposal). The comments also identify a number of additional reasons for questioning the applicability of the VHS model to surface impoundments.

Other EPA studies have demonstrated that fluxes even from lined surface impoundments may be as much as 16 times higher than from lined landfills. See, e.g., EPA Office of Solid Waste Land Disposal Branch, Draft Interim Report #3 of the Liner/Location Analysis, Table 4-5 (July 1983). That report further states that "[o]ther factors being equal, for example, the release rates of surface impoundments during the operational period tend to be significantly higher than those from landfills. Such higher initial rates serve to speed the movement of leachate through the unsaturated zone to the water table." Id. at 1-6.

114. 53 Fed. Reg. 47731, 47735, n.3 (Nov. 25, 1988) (emphasis added).

115. 50 Fed. Reg. 48903 (Nov. 27, 1985). Elsewhere, EPA noted that

[f]ifty soil dispersion models have been evaluated to predict the concentration of contaminants exiting a soil column. These models generally predict an attenuation of at most fifty percent for toxicants in the leachate of a waste continuously applied to a solid column. In fact, many of the models evaluated predict no attenuation at all, since eventually a soil will become saturated and unable to absorb any more of a particular toxicant.

Id. at 7895, 7897 (Feb. 26, 1985).

116. 54 Fed. Reg. 31675 (Aug. 1, 1989) (Clay Equipment) ("it is reasonable to expect clogging to occur in both the surface impoundment and landfill disposal scenario").

117. U.S. GENERAL ACCOUNTING OFFICE, supra note 8, at 42.

118. Id. at 41.

119. Id. at 42.

120. Id. at 36.

121. RCRA § 1004(5)(B), 42 U.S.C. 6903(5)(B), ELR STAT. RCRA 005.

122. See Environmental Defense Fund v. Reilly, No. 89-0598 (D.D.C. filed Mar. 8, 1989). In two motions filed on June 12, 1989, and June 23, 1989, EPA admitted its failure to meet two dozen separate RCRA deadlines and proposed a compliance schedule stretching into the 21st century.

123. H.R. REP. No. 198, 98th Cong., 2d Sess. 20, reprinted in 1984 U.S. CODE CONG. & ADMIN. NEWS 5576, 5579.