01/28/88

The meaning of "endangers" or "imminent and substantial endangerment" of drinking water sources under sections 1421 and 1431 of the Safe Drinking Water Act, especially in the context of Class V UIC wells


To:  Steven Pardieck, UIC Section Chief, Region 9 and Laurie Kermish,
Office of Regional Counsel, Region 9

From:  Christopher Sproul, Office of Regional Counsel, Region 9

Re:  The meaning of "endangers" or "imminent and substantial endangerment"
of drinking water sources under sections 1421 and 1431 of the Safe
Drinking Water Act, especially in the context of Class V UIC wells.

Date:  January 28, 1988

I.  THE NEED FOR A DEFINITION OF ENDANGERMENT FOR CLASS V ENFORCEMENT

    The Safe Drinking Water Act ( SDWA ) requires the EPA to ensure that
states "prevent underground injection which endangers drinking water
sources."  SDWA, Section 1421(b)(1), 42 USC Section 300.  1/

  1/ Specifically, the EPA must list in the Federal Register all states
    "for which in ( its ) judgment a State underground injection control
    program may be necessary to assure that underground injection will not
    endanger drinking water sources."  SDWA, Section 1422(a), 42 USC
    Section 300h-1(a).  After a state has been listed, it must establish
    an "Underground Injection Control" ( UIC ) program which will prevent
    underground injection well operations from endangering underground
    sources of drinking water ( "USDWs" ).  If a state fails to act, then
    the EPA must establish a program instead.  SDWA, Section 1422(c).  All
    states have been listed, and either have a state-administered or
    EPA-administered UIC program.  40 CFR Section 144.1(e).

          All UIC programs in the states must "contain minimum
    requirements for effective programs to prevent underground injection
    which endangers drinking sources."  SDWA, Section 1421(b)(1).  40 CFR
    Section 144.12 and 145.11(a)(6) implement this provision.  In this
    roundabout fashion, the SDWA thus bans endangerment of USDWs.

In administering this statutory mandate, the EPA has adopted regulations
requiring Class I, II, and III wells to be built and operated to certain
technological standards.  In theory, compliance with these standards
should prevent endangerment of USDWs from these wells.  These standards
simplify enforcement

                                   - 2 -

actions against Class I, II, and III wells.  The agency can issue
compliance orders, seek administrative penalties, and refer a civil or
criminal action to the Justice Department based on evidence that a well
has failed a mechanical integrity test, lacks proper casing and cementing,
and so on, without proving that the technical failure did or might leak
contaminants into a USDW in dangerous amounts.

    The EPA has not adopted comparable technology-based regulations for
Class V wells.  Accordingly, enforcement against a Class V well requires
establishing that the well is endangering a USDW.  2/

  2/ This statement needs one qualification:  the EPA can require a Class
    V well to obtain a permit based on a finding that protection of a USDW
    requires this action.  The permit could specify technological
    controls.  Violation of these permit conditions could then be the
    basis for an enforcement action.

Thus, if the Program is to identify Class V injection cases which can be
successfully prosecuted, it must understand the legal requirements for
establishing "endangerment" of a USDW.

II.  "ENDANGERMENT" UNDER THE SDWA:  THE STATUTORY LANGUAGE AND JUDICIAL
      DISCUSSION

    Section 1421(d)(2) of the SDWA defines underground injection which
endangers USDWs:

                                   - 3 -

    Underground injection endangers drinking water sources if such
    injection may result in the presence in underground water which
    supplies or can reasonably be expected to supply any public water
    system of any contaminant, and if the presence of such contaminant may
    result in such system's not complying with any national primary
    drinking water regulation or may otherwise adversely affect the health
    of persons.

42 USC Section 300h(d)(2).  Using this clause in enforcement actions
requires resolution of two issues:  1, what are "underground water which
supplies or can reasonably be expected to supply" public water systems
and, 2, what causal link between an injection and USDW contamination
and/or detrimental health effects must the EPA establish to prevail
against a well operator.

    A.  What Aquifers are Protected under the SDWA?

    On its face, section 1421(d)(2) indicates that a USDW need not be in
use as a source of drinking water before it is subject to SDWA protection
--as long as it "can reasonably be expected to supply any public water
system," a USDW is protected.  A recent decision from the U.S. Court of
Appeals for the First Circuit implicitly supports this interpretation.
Natural Resources Defense Council v. EPA, 824 F.2d 1258 ( 1st Cir. 1987 ).
In NRDC v. EPA, the First Circuit vacated the EPA's High Level Waste
( HLW ) regulations adopted pursuant to the Nuclear Waste Policy Act
( NWPA ) and remanded them to the EPA for further consideration.  These
regulations established "individual protection requirements," or the
maximum radiation exposure that any member of the public could receive
from groundwater contamination due to the underground disposal of high
level

                                   - 4 -

nuclear waste.  EPA set these individual protection requirements higher
than the MCL for radiation set by EPA regulations under the SDWA.  The
court held that placing canisters of nuclear reactor wastes underground is
the underground injection of fluids, and is thus subject to the SDWA's
prohibition of endangering USDWs.  824 F.2d at 1270-71.  The court next
concluded that the HLW regulations authorized endangerment of USDWs, "the
individual protection requirements will permit repositories to be built
and used for the disposal of HLW which will, judged by the stricter
standard of the SDWA, `endanger' drinking water supplies."  824 F.2d at
1276.  The court, applying the standard of review specified by the
Administrative Procedure Act ( "APA" ), held that this made the HLW
regulations "arbitrary and capricious" because the agency offered no
explanation for allowing higher levels of radiation contamination of
groundwater under the NWPA than the SDWA.  824 F.2d at 1281.  3/

  3/ The court found this sufficient basis to vacate the HLW regulations
    and remand them to the agency for further consideration, deeming it
    unnecessary to decide whether the HLW regulations also violate the APA
    by being "contrary to law" because they authorize contamination in
    excess of MCLs.  824 F.2d at 1280 n.11.

          The court speculated that there might be reasons not occurring
    to the court why greater contamination due to HLW disposal should be
    allowed than the EPA's SDWA regulations would allow, thus making the
    HLW regulations not contrary to law.  824 F.2d at 1282.  The court
    found that until the agency articulated such reasons, the HLW
    regulations remained arbitrary and capricious.

    In reaching this result, the court implicitly indicated that the SDWA
protects not only aquifers currently used by public water systems, but
aquifers that could be used in the future.  In

                                   - 5 -

rejecting an argument that Congress was unconcerned with HLW canisters
when it enacted the SDWA, the court observed that Congress was generally
concerned with avoiding the problem of "(u)nusable ground water" rather
than regulating specific practices.  824 F.2d at 1271.  The court
elaborated that in enacting the SDWA, Congress's "overall intent ( was )
to protect future supplies of drinking water against contamination."  Id.
( emphasis added ).  For support, the court cited this passage from
relevant legislative history:

    It is the Committee's intent that the definition be liberally
    construed so as to effectuate the preventative and public health
    protective purposes of the bill.  The Committee seeks to protect not
    only currently-used sources of drinking water, but also potential
    drinking water sources for the future....

H.R.Rep. No. 1185, 93d Cong. 2d Sess., reprinted in 1974 U.S. Code &
Admin. News at 6484, quoted by NRDC v. EPA, 824 F.2d at 1271.

    Assuming that other circuits will follow NRDC v. EPA, which seems
likely given the legislative history quoted above, enforcement can proceed
whenever injection endangers "potential" drinking water sources or
"future" drinking water supplies.  To prove that a given aquifer is a
"potential" or "future" source, the agency could cite planning documents
supplied by public water systems indicating an intent to tap a given
aquifer in the future, or perhaps with models constructed from projections
of future population growth and development indicating that a given
aquifer will likely be used in the future.  The burden of developing this
data, however, would likely preclude the agency

                                   - 6 -

from litigating or pursuing contested administrative actions in all but
the most serious class V contamination cases.

    Alternately, the agency could base enforcement on its regulations
defining USDWs, reasoning that any aquifer within the definition is per se
reasonably expected to supply a public water system.  See 40 CFR Sections
144.3, 143.2.  4/

  4/ These regulations define USDWs as all aquifers containing less than
    10,000 ppm TDS that are large enough to supply water for 25 persons
    daily for at least 60 days per year.

A Class V operator might seek to escape SDWA liability by arguing that the
aquifer the he or she has endangered can not reasonably be expected to
supply any public water system, EPA regulations to the contrary
notwithstanding.  The agency's definition of a USDW, however, would
probably be upheld by a court and be preclusive on whether a given aquifer
can reasonably be expected to supply a public water system.  5/

  5/ The agency could seek to bar such a defense raised in an enforcement
    proceeding by relying on SDWA Section 1448.  Section 1448 provides
    that someone challenging an agency rule or regulation must file a
    petition with the U.S. Court of Appeals for the D.C. Circuit within 45
    days of the regulation's promulgation unless the petition is "based
    solely on grounds arising after the expiration of such period."
    Section 1448 further provides that regulations shall not be otherwise
    "subject to judicial review in any civil or criminal proceeding for
    enforcement or in any civil action to enjoin enforcement."  Thus, on
    its face, SDWA Section 1448 seems to bar any defense in an enforcement
    action that would require judicial rejection of an agency rule, such
    as the agency's rule defining a USDW.

          In law, however, courts have a way of replacing what seem like
    obvious propositions with convoluted and obscure ones.  In Adamo
    Wrecking Co. v. United States, 434 U.S. 275 (1978), the EPA criminally
    indicated an individual who demolished an asbestos laden building
    without complying with the applicable EPA regulation, 40 CFR Section
    61.22(d)(2)(i).  The Clean Air Act authorizes criminal prosecutions
    for violation of EPA promulgated regulations specifying "emission
    standards."  At the time, the agency had deemed 40 CFR Section
    61.22(d)(2)(i) an emission standard.  The defendant sought to defend
    against the indictment by arguing that Section 61.22 is not an
    "emission standard" as Congress intended the term when it enacted the
    Clean Air Act.

          In language essentially identical to the SDWA, the Clean Air Act
    provides for exclusive review of EPA regulations promulgated under the
    Act via timely petition in the U.S. Court of Appeals for the D.C.
    Circuit.  The EPA sought to rely on this provision of the Clean Air
    Act to bar the defendant's defense.  The Supreme Court, however, sided
    with the defendant.  The Court acknowledged that the Clean Air Act
    expressly barred judicial review of EPA rulemaking in civil or
    criminal proceedings for enforcement.  The Court, however, in
    mysterious reasoning, decided that this express bar prevented a
    criminal defendant from challenging whether the regulation is
    "arbitrary, capricious, or supported by the administrative record,"
    but did not bar the defendant from challenging whether EPA's
    designation of the regulation as an "emission standard" was consistent
    with what Congress intended "emission standard" regulations to cover.
    Specifically, the Court found it a proper question "whether the
    regulation which the defendant is alleged to have violated is on its
    face an `emission standard' within the broad limits of the
    congressional meaning of that term."  434 U.S. at 285.  How this is
    not judicial review of agency rulemaking is beyond me, but the Supreme
    Court is not final because it is right, it is right because it's
    final.

          The significance of Adamo depends on how one interprets its
    obscure reasoning.  On the one hand, interpreting language that is
    identical to SDWA section 1448, the Court did say that the Clean Air
    Act bars judicial review of the substance of a regulation.
    Accordingly, the EPA might be able to resist any attack on its USDW
    definition by claiming  that such attack concerns the merits of what
    is a USDW.  On the other hand, regardless of what the Supreme Court
    said it was doing ( which I find largely unintelligible ), what the
    Supreme Court did was to hold that the EPA could not regulate the
    demolition of asbestos laden buildings under Congress's grant of
    authority to set emission standards.  The Court determined that the
    demolition of a building does not cause an "emission" within the
    meaning of the relevant section of the Clean Air Act.  Looked at this
    way, there seems to be no principled reason why the Court could not
    also void enforcement for injection into an aquifer with 9,000 ppm TDS
    on the theory that EPA's definition of a USDW does not comport with
    Congress' understanding of what is an "underground drinking water
    source" within SDWA section 1421(d)(2).

          In any case, Adamo might be limited to a criminal enforcement
    context.  In Adamo, the Supreme Court did not expressly comment on
    whether an EPA regulation could also be scrutinized for consistency
    with an authorizing statute in a civil enforcement proceeding, but it
    seemed to indicate that the fact that criminal prosecution was
    involved was significant factor in its holding.  434 U.S. at 548.

          Assuming that Adamo would allow a criminal and even a civil
    enforcement action defendant to challenge the EPA's definition of a
    USDW, the EPA would have a good chance of withstanding a challenge on
    the merits.  The standard of judicial review of an agency regulation
    interpreting a statutory term or phrase is somewhat obscure.  Whether
    the agency regulation is consistent with the statute is a question of
    law, and thus is not necessarily reviewed with the same deferential
    standards applied to agency findings of fact.  Courts declare that
    congressional intent is controlling, but given that congressional
    intent is often unclear, this often does not dispose of issues.  When
    congressional intent is not obvious, the degree of judicial deference
    to an administrative interpretation of a statute seems to vary in no
    discernable pattern.  Compare Chevron, U.S.A. v. Natural Resources
    Defense Council, 467 U.S. 837, 843-45, reh'g denied, American Iron and
    Steel Inst. v. Natural Resources Defense Council, 105 S. Ct. 28 (1984)
    with SEC v. Sloan, 436 U.S. 103, 117-18 (1978); Barnett v. Weinberger,
    818 F.2d 953, 960-64 ( D.C. Cir. 1987 ).

          Whatever the standard of judicial review, the EPA's USDW
    definition will probably be upheld.  To begin, the EPA's specification
    that aquifers large enough to serve 25 persons enjoy SDWA protection
    seems clearly consonant with congressional intent.  Section 1401(4) of
    the SDWA defines a public water system as any system regularly
    supplying 25 persons, indicating that Congress was concerned with
    water supplies of that size.  The added specification that aquifers
    with less than 10,000 ppm TDS are protected might be subject to more
    serious challenge.   Obviously, no one is going to drink water
    containing such a level of TDS.  I assume, however, that the agency
    developed evidence during notice and comment proceedings by evidence
    that such waters can be rendered drinkable by treatment.  If this is
    the case, then the TDS provision is likely to be upheld even if it is
    subject to judicial review in an enforcement proceeding.

                                   - 8 -

    B.  What Constitutes Legally Actionable Contamination or Risk of
Contamination of a USDW?

    Section 1421's definition of "endangerment" bans activities which "may
result" in contamination of a USDW.  On its face, the SDWA thus authorizes
enforcement without proof of actual contamination.  NRDC v. EPA and
numerous other authorities indicate that this is the case.  See, e.g.,
United States v.

                                   - 9 -

Price, 688 F.2d 204, 213-14 ( 3d Cir. 1982 ) ( interpreting section 1431
of the SDWA ).

    Contamination occurs if, (1), injection "may result" in a public water
system failing to comply with maximum contaminant levels ( MCLs ) set by
the EPA for various contaminants pursuant to section 1412 of the SDWA or,
(2), regardless of whether an MCL will be violated, an injection may
"otherwise adversely affect the health of persons."

    The agency could prevail in an enforcement action by demonstrating
that well operation will or may cause tap water from an existing public
water system to violate an MCL.  See SDWA, Section 1412(b)(4).  6/

  6/ Section 1412(b)(4) requires the EPA to establish maximum contaminant
    goals "at the level at which no known or anticipated adverse effects
    on the health of persons occur and which allows an adequate margin of
    safety."  The EPA must then set maximum contaminant levels which are
    as close to these maximum contaminant goals "as is feasible."  The
    latter are the enforceable standards.  An argument could be made that
    because MCLs are to be set with an adequate margin of safety to the
    extent feasible, contamination just above an MCL does not necessarily
    endanger human health.  Such an argument seems unlikely to succeed,
    though, because even decreasing a margin of safety probably
    constitutes endangering human health as the term "endangering" has
    been used judicially.  See the discussion infra of judicial treatment
    of "endangerment."

Furthermore, NRDC v. EPA indicates that the agency might prevail by
demonstrating a reasonable chance that a public water system might in the
future seek to tap a currently unused aquifer and the water as taken out
of the ground will exceed MCLs.  7/

  7/ Simply showing that an injectate fluid has contaminants in excess of
    MCLs is probably insufficient, however.  Instead, the agency probably
    must at least offer a theory connecting the injection and deleterious
    ambient levels of contaminants in the aquifer.  Provided such a theory
    is offered, though, the agency can probably prevail even without
    showing that injection will cause the ambient water quality of the
    aquifer to violate MCLs.  The legislative history indicates this to be
    the intent of Congress.  See the discussion infra of H.R. Rep. No.
    1185.

                                  - 10 -

    The agency can also show "contamination" without reference to MCLs by
demonstrating that injection "may otherwise endanger the health of
persons."  To demonstrate this, the agency need not show that someone will
drink water contaminated by an injection well and thereby suffer risk of
harm.  NRDC v. EPA indicates that it is sufficient to demonstrate that if
someone drank the water from an aquifer without its being treated, he or
she would risk ill health.

    In NRDC v. EPA, the agency argued that its HLW regulations did not
violate the SDWA or regulations promulgated thereunder because the agency
could apply drinking water standards to ensure that no public water system
delivered contaminated water to members of the public.  The court rejected
this argument, noting that the EPA's approach would burden local water
companies with cleaning up drinking water sources.  The court reasoned
that "While placing this burden on the local water companies may prevent
... contaminated water from being improperly used as drinking water, it
will not prevent the future endangerment of drinking water supplies, which
is a declared purpose of the Safe Drinking Water Act."  824 F.2d at 1281
( emphasis in the original ).  The court's approach is supported by
legislative

                                  - 11 -

history, which the court quoted:

    The Committee was concerned that its definition of "endangering
    drinking water sources" also be construed liberally.  Injection which
    causes or increases contamination of such sources may fall within this
    definition even if the amount of contaminant which may enter the water
    source would not by itself cause the maximum allowable levels to be
    exceeded.  The definition would be met if injected material were not
    completely contained in the well, and if it may enter either present
    or potential drinking water sources, and if it ( or some form into
    which it might be converted ) may pose a threat to human health or
    render the water source unfit for human consumption.

H.R.Rep. No. 1185, 93d Cong. 2d Sess., reprinted in 1974 U.S. Code &
Admin. News at 6484, quoted by NRDC v. EPA, 824 F.2d at 1271.

    Also, subsection 1401(1)(D)(i) of the Act, inter alia, directs EPA to
adopt "primary drinking water regulations" which ensure "the minimum
quality of water which may be taken into ( a public water ) system"
( emphasis added ).  By thus expressing a concern for water quality before
being taken into a public water system, Congress further demonstrated
intent to regulate activities that might make an aquifer undrinkable
without treatment.

    Establishing endangerment when injection will cause contamination with
substances no prudent person would drink, but for which there are no MCLs
( e.g., toluene, xylene, ethylbenzene, methylene chloride, etc. ) does
require some creativity.  Outside the realm of MCLs, the SDWA and
accompanying regulations do not indicate how grave harm to human health
must be before it is

                                  - 12 -

legally cognizable.  8/

  8/ One general point from the case law is worth noting here:  the agency
    will face less judicial scrutiny and greater chance of prevailing if
    it can offer a reasonable hypothesis that activity threatens to cause
    cancer as opposed to other human health risks.  As the court in
    Environmental Defense Fund v. EPA, observed:

    Courts have traditionally recognized a special judicial interest in
    protecting the public health, particularly where `the matter involved
    is as sensitive and fright-laden as cancer.'  ... Where the harm
    envisaged is cancer, courts have recognized the need for action based
    upon lower standards of proof than otherwise applicable.

    598 F.2d 62, 88 ( D.C. Cir. 1978 ) ( citation omitted ).

Similarly, the SDWA and the regulations do not indicate how unfit for
human consumption water must be before liability attaches.  Guidance,
however, can be found from cases interpreting endangerment of human health
as the phrase is used in other environmental statutes.

III.  ENDANGERMENT AS DEFINED BY CASES ARISING UNDER OTHER ENVIRONMENTAL
      STATUTES

    Perhaps the best way to understand what the agency must do to prove
endangerment is to examine past cases where the agency succeeded in
proving endangerment.  Reserve Mining Co. v. EPA and Ethyl Corp. v. United
States are two cases that can serve as models of successful endangerment
arguments.  9/

  9/ Reserve mining was an enforcement action, and accordingly can serve
    as an enforcement action model.  Ethyl Corp., however, concerned a
    challenge to EPA rulemaking and is thus less analogous to a Class V
    UIC enforcement action.  The most notable difference is that in a
    judicial challenge to an EPA regulation, the EPA's definition of
    endangerment is itself "law" to which the court gives deference.  See
    note 12 infra.

          Ethyl Corp. still usefully indicates what evidence a court will
    consider sufficient to establish endangerment.

                                  - 13 -

    A.   Reserve Mining Co. v. EPA

    In Reserve Mining Co. v. EPA, a U.S. district court found and the U.S.
Court of Appeals for the Eighth Circuit agreed that Reserve Mining
Company's daily discharge of 67,000 tons of taconite mining tailings into
Lake Superior endangered the health of persons within the meaning of
subsections 1160(c)(5) and (g)(1) of the Federal Water Pollution Control
Act.  10/

  10/ As these sections existed prior to 1972 amendments of the Act.

See 514 F.2d 492 ( 8th Cir. 1975 ) ( en banc ).  Subsections 1160(c)(5)
and (g)(1) authorized a court to issue "such judgment as the public
interest and the equities of the case may require" "in the case of
pollution of waters which is endangering the health or welfare of
persons."  Applying these provisions, the court of appeals ordered Reserve
Mining to cease discharging taconite tailings into Lake Superior and find
a suitable on-land disposal method for the tailings "within a reasonable
time" or shut down.  514 F.2d at 535-38.

    Reserve Mining Co.'s operations accounted for 12% of the total iron
ore produced in the United States.  The company, which had an annual
payroll of $31,700,000, was by far the largest employer in Silver Bay,
Minnesota, which depended virtually entirely on the company for its
economic base.  514 F.2d at 536.  Against these proven benefits of
Reserve's operations, the threat to human health from its Lake Superior
discharge was speculative.

                                  - 14 -

    Communities surrounding Lake Superior took their drinking water from
the lake.  The EPA offered indirect evidence that Reserve's discharge
ended up in these communities' drinking water.  The discharge contained
minerals of the amphibole family.  Water samples taken from the lake
near Reserve's outfall and from public water system also contained
amphibole fibers, and this concentration increased the closer the sample
to Reserve's outfall.  The EPA also offered evidence, which Reserve
disputed, that natural sources of amphibole could not have produced the
concentrations found.  Reserve's discharge contained particles of the
mineral cummingtonite-grunerite which the EPA contended is chemically and
morphologically indistinguishable from amosite asbestos.  Epidemiological
studies have linked occupational exposures to amosite asbestos to
asbestosis, lung cancer, and, to a lesser extent, gastrointestinal cancer.
These studies, however, concerned workers who were exposed to amosite
asbestos dust in ambient air.  EPA offered the theory that excess
gastrointestinal cancer in asbestos workers is caused by their swallowing
the asbestos fibers that they inhale.  EPA argued accordingly that
drinking water laden with asbestiform fibers also might cause cancer.

    The EPA's case had several weaknesses.  To begin, the agency could
not prove that Reserve's cummingtonite-grunerite fibers had the same
pathogenic property as amosite asbestos.  While epidemiological studies
have conclusively shown amosite asbestos to be pathogenic, the property of
amosite asbestos which renders

                                  - 15 -

it pathogenic is unknown.  Thus, the morphological and chemical similarity
of Reserve's cummingtonite-grunerite to amosite asbestos only made it a
suspected, not a proven pathogen.  Even then, existing data showed that
asbestos dust is a health hazard, leaving in doubt whether ingesting
fibers in water is dangerous.  Next, the level of exposure to asbestos of
workers who contracted cancer or asbestosis suffered is essentially
unknown, and essentially there is no evidence equating doses of asbestos
to incidences of disease.  On top of this uncertainty, the EPA did not
have any firm evidence about the quantity of cummingtonite-grunerite
fibers that the average citizen might be ingesting as a result of
Reserve's discharge.  Thus, the EPA could not offer evidence that citizens
were being exposed to potentially pathogenic asbestiform fibers at levels
known to cause disease.  Furthermore, a tissue study of deceased residents
of a nearby Lake Superior community showed no traces of cummingtonite
-grunerite fibers.  Also, a National Cancer Institute study of the area
revealed no excess cancers of the types associated with asbestos exposure.

    The court of appeals concluded that whether Reserve's discharge would
cause disease lay "on the frontiers of scientific knowledge," making
"proof with certainty ... impossible."  514 F.2d at 519-20.  The court
nevertheless felt that Congress intended the courts to grant relief in
such situations when the EPA offered "evidence of potential harm" that
"gives rise to a reasonable medical concern for the public health."  514
F.2d at

                                  - 16 -

528-29.  The court reasoned that Congress used the term endangering "in a
precautionary or preventive sense."  The court further explained ( quoting
language from another opinion ) that determining whether activity
endangered human health required risk analysis:

    "Endanger, ... is not a standard prone to factual proof alone.
    Danger is a risk, and so can only be decided by assessment of risks.
    ...
          A risk may be assessed from suspected, but not completely
    substantiated, relationships between facts, from trends among facts,
    from theoretical projections from imperfect data, or from probative
    preliminary data not yet certifiable as `fact.'"

Id.  ( citation omitted ).

    Applying this risk analysis to Reserve, the court evaluated the
evidence as failing to establish "that the probability of harm is more
likely than not."  514 F.2d at 520.  Instead, the record revealed that
Reserve's discharge "under an acceptable but unproved medical theory may
be considered as carcinogenic."  514 F.2d at 529.  The court deemed this
sufficient to create a reasonable medical concern over the public health,
its definition of endangerment.  See also United States v. Vertac, 489 F.
Supp. 870, 880-885 ( E.D. Ark. 1980 ) ( applying Reserve to an action
brought under the FWPCA ).

    B.   Ethyl Corp. v. United States

    In Ethyl Corp. v. United States, the U.S. Court of Appeals for the
D.C. Circuit sustained EPA's regulations requiring the phased reduction of
lead in gasoline.  541 F.2d 1 (1976) ( en banc ), cert. denied, 426 U.S.
941.  The EPA adopted these

                                  - 17 -

regulations pursuant to section 211(c)(1)(A) of the Clean Air Act, which
authorizes the agency to control or prohibit any fuel or fuel additive "if
any emission products of such fuel or fuel additive will endanger the
public health or welfare."

    In analyzing whether EPA regulations were consistent with section
211(c)(1)(A), the court thus had to define endangerment.  The court
explained its understanding of the term:

    Case law and dictionary definition agree that endanger means something
    less than actual harm.  When one is endangered, harm is threatened; no
    actual injury need ever occur.  Thus, for example, a town may be
    "endangered" by a threatened plague or hurricane and yet emerge from
    the danger completely unscathed.  A statute allowing for regulation
    in the face of danger is, necessarily, a precautionary statute.
    Regulatory action may be taken before the threatened harm occurs;
    indeed, the very existence of such precautionary legislation would
    seem to demand that regulatory action precede, and optimally, prevent,
    the perceived threat.

    541 F.2d at 13 ( emphasis original ) ( footnotes omitted ).

    Like the Eighth Circuit in Reserve Mining, the D.C. Circuit found that
endangerment must be determined in any given case by risk analysis.  The
court explained that risk analysis should be performed on a case-by-case
assessment of possible harm discounted by its probability:  11/

  11/ For the lawyers reading this--Learned Hand lives]

    While the dictionary admittedly settles on "probable" as its measure
    of danger, we believe a more sophisticated case-by-case analysis is
    appropriate.  Danger ... is not set by a fixed probability of harm,
    but rather is composed of reciprocal elements of risk and harm, or
    probability and severity....  That is to say, the public health may
    properly be found endangered both by a lesser risk of a greater harm
    and by a greater risk of a lesser harm.

541 F.2d at 18.

                                  - 18 -

    Like the Eighth Circuit in Reserve Mining, the D.C. Circuit found that
the required risk analysis is satisfied by "a reasonable hypothesis
supported by evidence."  541 F.2d at 45-47 & n.96.

    In Ethyl Corp., the EPA supported its regulations with medical
evidence that excess lead in the human bloodstream causes lead poisoning
that can be fatal.  The EPA further offered evidence that lead is absorbed
from ingesting lead-laden substances or breathing lead-laden air.  The
agency offered evidence that 90% of lead in the air comes from automobile
emissions.  The EPA concluded that lead in gasoline endangers human health
both because it adds lead to the air people breathe and because it adds
lead to dust that children with pica ( the tendency to ingest non-food
substances ) might swallow.

    There were some weaknesses to the agency's conclusions.  To begin, the
agency had no evidence linking lead in the air to corresponding levels of
bloodstream lead in human populations.  Lead comes from many dietary
sources, is fungible once it is in the bloodstream, and thus it is
impossible to trace the source of the lead present in any given
individual's bloodstream.  Moreover, substantial uncertainty exists what
level of bloodstream lead causes illness.  The only population to suffer
widespread ill-health from lead poisoning are young children.  The
evidence, however, indicates that children with lead poisoning have
consumed peeling lead-based paint particles.  The agency theorized that
leaded gasoline could have nevertheless

                                  - 19 -

contributed to children's lead poisoning or might contribute in the future
because lead emissions settle in roadside dust which children also might
have consumed or might consume in the future.  The EPA offered studies
showing that dust lead concentrations in residential sites in cities
averaged 1,613 ppm compared to a normal range of 2-200 ppm.  The agency
had no direct evidence that any given child had ever eaten lead
-contaminated dust, however.

    The Eighth Circuit found that the agency's hypothesis about roadside
dust posing a threat to children to be reasonable and supported by
evidence of elevated levels of lead in city dust and children's tendency
to pica.  Added to EPA evidence that air-borne lead comes from automobile
emissions and that air-borne lead is absorbed into the bloodstream, the
court deemed EPA had sufficiently shown "a significant risk of harm to the
health of urban populations, particularly to the health of city children."
12/

  12/ In Reserve Mining, the district court was asked in an enforcement
     action to enjoin a disposal activity on the basis that it constituted
     endangerment.  The court independently judged whether Reserve's
     activities "endanger human health," turning only to the statute,
     legislative history, and principles from case law for authoritative
     guidance.

          By contrast, in Ethyl Corp., the court of appeals reviewed EPA
     regulations adopted via informal rulemaking pursuant to section 4 of
     the Administrative Procedure Act ( APA ).  The court thus reviewed
     the EPA's authoritative definition of endangerment.  The APA provides
     that judicial review of regulations promulgated by an agency is
     limited to determining whether they are "arbitrary, capricious, an
     abuse of discretion, or otherwise not in accordance with law."  5
     USC Section 706(2)(A).  This standard of review is a highly
     deferential one, and presumes agency action to be valid.  Overton
     Park v. Volpe, 401 U.S. 402 (1971); Ethyl Corp., 541 F.2d at 34.  A
     reviewing court must uphold agency regulations is they have "a
     rational basis."  Bowman Transportation, Inc. v. Arkansas-Best
     Freight Systems, Inc., 419 U.S. 281, 290 (1974); Ethyl Corp., 541
     F.2d at 35 n.74.  The courts are not always clear or consistent in
     explaining what they do in determining whether regulations have a
     rational basis.  A frequent explanation is that courts must engage in
     a "searching and careful" "substantial inquiry into the facts" and
     determine whether agency action was "based on a consideration of the
     relevant factors."  Overton Park, 401 U.S. at 415, 416; Ethyl Corp.,
     541 F.2d at 34-35.  In other words, courts must probe whether
     conclusions are logical--internally consistent and take into account
     all data.

          A judicial EPA enforcement action against a Class V well based
     on the SDWA's prohibition of endangerment would be analogous to the
     enforcement action in Reserve Mining rather than the promulgation of
     lead regulations tested in Ethyl Corp.  In a Class V enforcement
     suit, the EPA would have to prove to a preponderance of the evidence
     and a court would independently judge whether a well endangered human
     health.  The agency would lack the benefit of the APA's arbitrary and
     capricious standard, under which an EPA rule defining endangerment
     would be sustained if it had a rational basis.

          The standards applied by a court in reviewing an administrative
     compliance or penalty order duly issued by the EPA pursuant to
     section 1423(c) of the SDWA would differ from that applied in a
     judicial enforcement action, however.  Whatever findings of fact the
     agency made ( through an ALJ or the Administrator ) would be reviewed
     to determine whether they are supported by "substantial evidence in
     the record, taken as a whole," or whether the decision "constitutes
     an abuse of discretion."  SDWA, Section 1423(c)(6).  This standard is
     in theory less deferential than the APA's arbitrary and capricious
     standard.  Though in theory the "substantial evidence" standard
     creates a heightened judicial review, cases interpreting the same
     phrase in the APA indicate that this standard still creates
     considerable deference to agency actions.  Indeed, in practice there
     is little substantive difference in the arbitrary and capricious and
     substantial evidence standards.  Applying either standard, courts say
     they cannot substitute their judgment for that of the agency, but
     must limit themselves to testing whether the conclusions reached by
     an agency are consistent with the evidence relied upon.  Compare
     Universal Camera Corp. v. NLRB, 340 U.S. 474 (1951); Consolo v.
     Federal Maritime Comm'n, 388 U.S. 607 (1966); SEC v. New England
     Elec. System, 390 U.S. 207 (1968) with e.g., Ethyl Corp. at 36-37 &
     n.78.

          As for conclusions of law ( which probably includes a working
     definition of "endangerment" ), the standard of judicial review is
     less clear.  I have not researched cases of specifically treating
     agency legal conclusions rendered in administrative adjudications,
     but cases discussing agency determinations of legal issues in
     rulemaking proceedings probably indicate the judicial approach--which
     is, I repeat, less than clear.  Compare Chevron, U.S.A. v. Natural
     Resources Defense Council, 467 U.S. 837, 843-45, reh'g denied,
     American Iron and Steel Inst. v. Natural Resources Defense Council,
     105 S. Ct. 28 (1984) with SEC v. Sloan, 436 U.S. 103, 117-18 (1978);
     Barnett v. Weinberger, 818 F.2d 953, 960-64 ( D.C. Cir. 1987 ).

541 F.2d at 47-48.

                                  - 21 -

    C.   Cases Under RCRA Section 7003 and CERCLA Section 106

    Section 7003 of RCRA authorizes injunctive relief when the handling,
storage, treatment, transportation or disposal of hazardous waste "may
present an imminent and substantial endangerment to health or the
environment."  Similarly, section 106 of CERCLA authorizes injunctive
relief when "there may be an imminent and substantial endangerment to the
public health or welfare or the environment because of an actual or
threatened release of a hazardous substance from a facility."  42 USC
Section 9606.  There are a number of cases granting injunctive relief
pursuant to these provisions against owners, operators, and users of
hazardous waste disposal sites which interpret the meaning of the phrase
"imminent and substantial endangerment."

    As discussed in more detail below, section 1431 of the SDWA also
provides for enforcement in instances of "imminent and substantial
endangerment," whereas section 1423 authorizes enforcement only on a
showing of "endangerment."  Thus, these RCRA and CERCLA cases are most on
point for an action under section 1431.

    When the facts and outcomes of cases where courts applied an "imminent
and substantial endangerment" standard are compared to

                                  - 22 -

the facts and outcomes of cases under a mere "endangerment" standard,
there seems to be little real difference.  United States v. Northeastern
Pharmaceutical, 579 F. Supp. 823 ( W.D. Mo. 1984 ), aff'd 810 F.2d 726
( 8th Cir. 1986 ), when compared to Reserve Mining and Ethyl Corp.,
exemplifies this.  In Northeastern Pharmaceutical, wastes containing
dioxin where placed in a trench disposal site.  A subsequent EPA survey
found no dioxin in nearby wells, but did find dioxin migration 30 inches
into subsurface strata of the trench.  The agency offered evidence that
the soil underneath the trench was "such that particles, water and
leachate may move rapidly down through the soil to the water table below."
The agency also offered an expert's testimony that the underlying aquifer
supplied nearby wells.  The agency also introduced the results of a dye
test that indicated movement of dye from boreholes in a nearby stream to
nearby wells.  This test did not directly demonstrate possible paths for
the dioxin wastes as the boreholes were not in places known to be exposed
to the dioxin, but it did show the general permeability of the area soil.
The district court concluded, "Because of the soil conditions, there was a
substantial likelihood of the hazardous wastes in the trench ... entering
the environment and going into the ground water system; whereupon, the
contaminants may have come into contact with members of the public who may
have been adversely affected ..."  579 F. Supp. at 833.  The court found
this sufficient to establish imminent and substantial endangerment.  579
F. Supp at

                                  - 23 -

846.  13/

  13/ The defendants did not appeal this finding, though they did appeal
     other aspects of the decision.  See 810 F.2d at 749 and United States
     v. Conservation Chemical Co., 619 F. Supp. 162, 195 n.8 ( D.C. Mo.
     1985 ).

The district court, echoing Ethyl Corp., explained that determining
whether an imminent and substantial endangerment exists requires "a
case-by-case assessment of the relationship between the magnitude of risk
and harm arising from the presence of the hazardous waste."  579 F. Supp.
at 846.

    Courts interpreting "imminent and substantial endangerment"
universally agree that the EPA need not establish actual harm to health or
the environment, proof of risk or harm will suffice.  For example, in
United States v. Price, the court of appeals noted that RCRA section 7003
and SDWA section 1431 "have enhanced the courts' traditional equitable
powers by authorizing the issuance of injunctions when there is but a risk
of harm, a more lenient standard than the traditional requirement of
threatened irreparable harm."  688 F.2d 204, 211 ( 3rd Cir. 1982 ); see
also Northeastern Pharmaceutical, 579 F. Supp. at 846 n.28.

    Though case under either an "endangerment" or an "imminent and
substantial endangerment" standard seem to be highly similar, the courts
do seem to employ a slightly more stringent analysis ( at least in
theory ) under the latter standard.  The courts, respecting the dictionary
meaning of the word "imminent," consistently indicate that while
environmental statutes using the phrase allow enforcement without proof of
actual harm and with only proof of risk of harm, there must be proof of "a
present

                                  - 24 -

threat to the public health or environment."  See, e.g., United States v.
Price, 688 F.2d at 214 ( emphasis added ).  Courts discussing "imminent
and substantial endangerment" cite two passages from legislative history
as the starting points to determine congressional intent in using the
phrase.  The House Committee Report accompanying section 1431 of the SDWA
states:

    "(I)mminence" must be considered in light of the time it may take to
    prepare administrative orders or moving papers to commence and
    complete litigation and to permit issuance, notification,
    implementation, and enforcement of administrative or court orders to
    protect the public health.

          Furthermore, while the risk of harm must be "imminent" for the
    Administrator to act, the harm itself need not be.  Thus, for example,
    the Administrator may invoke this section when there is imminent
    likelihood of the introduction into drinking water of contaminants
    that may cause health damage after a period of latency.

H.R. Rep. No. 1185, 93rd Cong., 2d Sess. 35-36, reprinted in 1974 U.S.
Code & Cong. Ad. News. 6454, 6487-88, cited by Northeastern
Pharmaceutical, 579 F. Supp. 846 n.28; United States v. Reilly Tar &
Chemical Corp., 546 F. Supp. 1100, 1109-11 ( D. Minn. 1982 ).  This
language, though vague and even circular, seems to suggest that Congress
intended to use "imminent" literally, and thus restrict enforcement under
section 1431 to activities threatening to contaminate USDWs "soon."  While
how little time can pass before a prospective event is still "soon," is
subjective.  Most people ( and judges ) probably would agree that events
more than a few years away are not "soon" to occur.

    Legislative history behind recent amendment to RCRA section 7003,
however, suggests a more expansive meaning to the phrase "imminent
endangerment:"

                                  - 25 -

    An endangerment is "imminent" and actionable when it is shown that it
    presents a threat to human health or the environment, even if it may
    not eventuate or be fully manifest for a period of many years--as may
    be the case with drinking water contamination, cancer, and many other
    effects.  United States v. Price ... and United States v. Reilly Tar
    & Chemical Co., ( 546 F. Supp. 1100, 1109-10 ).

S. Rep. No. 284, 98th Cong., 1st Sess., at 59 ( Oct. 28, 1984 ).

    The courts have not rigorously analyzed what events must occur "soon"
( e.g., percolation of contaminants to subsurface areas near an aquifer,
actually into the aquifer, into drinking water taken from a well, etc. ),
and how "soon" is soon enough before a risk of harm is deemed "imminent."
The closest courts come are statements such as the following:  "(A)n
endangerment is `imminent' if factors giving rise to it are present, even
though the harm may not be realized for years."  United States v.
Conservation Chemical Co., 619 F. Supp. 162, 193-94 ( D.C. Mo. 1985 ).  "A
hazard may be `imminent' even if its impact will not be apparent for many
years."  Northeastern Pharmaceutical, 579 F. Supp. at 846 n.28.  "An
`imminent hazard' may be declared at any point in a chain of events which
may ultimately result in harm to the public.  It is not necessary that the
final anticipated injury actually have occurred prior to a determination
that an `imminent hazard' exists.  United States v. Ottati & Goss, Inc.,
630 F. Supp. 1361, 1394 ( D.N.H. 1985 ) quoting Environmental Defense Fund
v. EPA, 465 F.2d 528, 525 ( D.C. Cir. 1972 ).

    The significance of this for UIC enforcement is that the agency
probably faces a somewhat higher, though even less

                                  - 26 -

defined standard when proceeding with enforcement under SDWA 1431 versus
1423.

IV. ENFORCEMENT UNDER SDWA SECTION 1423 VERSUS SECTION 1431

    The Program has asked for guidance on when enforcement should proceed
under SDWA section 1423 versus section 1431.  My advice is that
enforcement should proceed under section 1431 when operation of an
injection well located in a primacy state presents such a clear and
present danger that the agency deems it necessary to halt the injection
immediately.  In all other cases, including injection posing immediate and
urgent hazard in a nonprimacy state, the agency should rely on section
1423.

    Under section SDWA 1423, the EPA can bring an enforcement action
"(w)henever the Administrator finds ... that any person subject to any
requirement of any applicable underground injection control program ... is
violating such requirement."  Section 1423(a)(1) and (a)(2).  This section
thus authorizes an enforcement action for violation of section 1421's
prohibition on endangerment of USDWs, which is a requirement of all
underground injection programs in all 50 states.  The section imposes one
procedural limitation:  in primacy state, the EPA must give 30 days notice
to the state before commencing enforcement, and can only begin enforcement
if "the State has not commenced appropriate enforcement action."  Section
1423(a)(1).  Accordingly, section 1423 is an inappropriate vehicle for

                                  - 27 -

enforcement in a primacy state where the Program perceives the threat from
injection activity to be so serious that it must be halted immediately.
In non-primacy states, however, section 1423 does not require the EPA to
notify the state before commencing enforcement, and thus there does not
seem to be any procedural objection to using 1423 to obtain even immediate
relief, such as a judicial temporary restraining order ( "TRO" ) halting
injection.

    Section 1431, entitle "Emergency Powers," authorizes the EPA to take
enforcement actions when well injection  14/  "may present an imminent and
substantial endangerment to the health of persons."

  14/ Or any other activity that may contaminate a Public Water Supply or
     USDW.

The section imposes a more limited procedural requirement than does
section 1423:  the EPA is required "To the extent ( the Administrator )
determines it practicable in light of such imminent endangerment" to
"consult with the State and local authorities to confirm the correctness
of the information on which action proposed to be taken under this
subsection is based and to ascertain the action which such authorities are
or will be taking."  Unlike section 1423, section 1431 thus apparently
does not require any time to elapse between notice to a state and EPA
enforcement in a primacy state.  Moreover, in an extreme emergency, the
EPA is apparently authorized to dispense with notice altogether.
Furthermore, state inaction is not a

                                  - 28 -

prerequisite to an EPA action.  15/

  15/ As for non-primacy states, section 1431 is ambiguous on whether EPA
     is required to consult with State and local governments.  Section
     1431 requires consultation with all State and local "authorities,"
     and does not specify that EPA must consult only with primacy states.
     Yet, it seems incongruous that EPA must consult with nonprimacy
     states when undertaking emergency enforcement under section 1431 when
     the agency will necessarily be in a hurry and need not consult when
     undertaking enforcement under section 1423 when it might be able to
     proceed more leisurely.  Perhaps the agency could escape the need to
     consult with a nonprimacy state in a 1431 action by arguing that the
     nonprimacy state is not an "authority" for SDWA purposes when it has
     not had SDWA authority delegated to it.

Thus, in true emergency situations in a primacy state, section 1431 is the
proper enforcement vehicle.

    Other than this one limited procedural advantage under section 1431,
enforcement under section 1423 is preferable.  To begin, the agency can
secure broader remedy under section 1423.  The same injunctive relief is
available under either section, but section 1423 authorizes larger
penalties and in more situations.  Section 1431 authorizes courts to issue
TROs, preliminary, or permanent injunctions when "necessary to protect the
health of persons."  Section 1423 does not expressly authorize injunctive
remedies, but clearly indicates that courts have this power under the
section by stating a U.S. district court "shall have jurisdiction to
require compliance with any requirement of an applicable underground
injection program or with an ( administrative compliance ) order...."
TROs and injunctions are the understood vehicle for courts to require
actions of

                                  - 29 -

parties.  16/

  16/ Section 1423 also adds, "The ( appropriate district ) court may
     enter such judgment as protection of public health requires."  This
     indicates a congressional intent to grant courts their full equity
     jurisdiction to issue injunctions.  See United States v. Price, 688
     F.2d at 214 ( "Congress, in the endangerment provisions of RCRA and
     SDWA sought to invoke nothing less than the full equity powers of the
     federal courts...." ).

As for penalties, section 1423 authorizes up to $25,000 per day per
violation in a judicial proceeding, $10,000 per day per violation for non
-ClassII wells, and $5,000 per day for Class II wells in an administrative
proceeding ( up to a $125,000 ceiling ).  By contrast, section 1431 only
authorizes $5,000 per day per violation in a judicial proceeding brought
to enforce an administrative compliance order.  17/

  17/ Strange logic of Congress--up to $25,000 per day penalty for
     violating a 1423 non-emergency order, but only $5,000 per day penalty
     for what seems like the more serious offense of violating a 1431
     emergency order.

    As noted above, section 1423 also provides a slightly easier
substantive standard for the agency to meet.  Under 1423, the agency need
only prove "endangerment"; under 1431, the agency must prove "imminent and
substantial endangerment."

V.  GROUNDWATER STUDIES AS A REMEDY UNDER THE SDWA

    The Program is concerned whether it can order a groundwater quality
study under the SDWA.  United States v. Price indicates that the answer is
yes.  In Price, the EPA sought a preliminary injunction ordering the
operators of a landfill which had received hazardous wastes to fund an
extensive geohydrological

                                  - 30 -

study to determine the extent of groundwater contamination due to leakage
from the fill.  The EPA relied on both RCRA and SDWA section 1431 for
authority for the injunction.  The district court denied the preliminary
injunction, reasoning that it lacked authority to issue the injunction.
On appeal, the Third Circuit Court of Appeals rejected the district
court's interpretation of RCRA and the SDWA, and held that both statutes
provide authority to issue such an injunction.  688 F.2d at 211-12.  18/

  18/ The court of appeals declined to reverse the district court,
     however, indicating that the district court could still exercise its
     discretion and withhold issuing a preliminary injunction in favor of
     determining the appropriate injunctive relief after trial.

    Apart from seeking a judicial injunction, the EPA could also issue an
administrative order requiring a groundwater study.  40 CFR section 144.27
authorizes the agency to "require the owner or operator of any well
authorized by rule ... to submit information as deemed necessary by the
Regional Administrator...."  Section 144.27 adds that such information may
include groundwater monitoring and periodic reports.  Section 144.27,
however, is only applicable to EPA administered programs.
Return to Enforcement Policies