SA8-EEC-86-010
4906
REPORT
on the review of
ALTERNATE CONCENTRATION LIMIT GUIDANCE
together with
TWO CASE STUDIES DEMONSTRATING THAT GUIDANCE
by the
Environmental Engineering Committee
Science Advisory Board
U. S. Environmental Protection Agency
February, 1986
U.S. Environmental Protection Agency
Region V, Library
230 South Dearborn Street
Chicago, Illinois 60604
-------
*
'**
OFFICE OF
THE ADMINISTRATOR
. UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON. D.C. 20460"
SMay 8, 1986
Honorable Lee M. Thomas
Administrator
U. S. Environmental Protection Agency
401 M Street, S. W.
Washington, D.C. 20460
Dear Mr. Thomas:
In June, 1985, the Office of Solid Wastes (OSW) asked the Science
Advisory Board (SAB) to review draft Agency guidance for the establishment
of Alternate Concentration Limits (ACL) for RCRA facilities to protect
human health and the environment. The review was assigned to the SAB's
Environmental Engineering Committee. In October, 1985, OSW also requested
the Committee to review two case studies demonstrating applications of
that guidance.
The Committee has completed its preliminary review, and is forwarding
its report to you. The Committee identified, at this time, only obvious
technical errors or omissions; OSW has informed us that it will seek a
more comprehensive scientific review when it prepares a final draft of
the ACL guidance.
The Committee has made a number of specific technical recommendations
for improving both the guidance and the case studies. We would appreciate
an Agency response to this advice.
Thank you for the opportunity to present our views on these issues.
Sincerely,
Raymdnd C. Loehr
Chairman, Environmental
Engineering Committee
Science Advisory Board
-- _^^
Norton Nelson
Chairman, Executive Committee
Science Advisory Board
cc: J. Winston Porter
Marcia Williams
Vernon Myers
Terry Ycsie
-------
SECTION I
INTRODUCTION
Background
The Resource Conservation and Recovery Act (RCRA) regulations .require that the
owner/operator of a hazardous waste disposal facility delineate any plume of
contamination that has entered the ground water from a regfllated unit.- This
includes identifying the concentration of each Appendix VIII constituent in
the plume. If any of the hazardous constituents are present .in concentrations
which exceed established concentration limits (Interim Primary Drinking Water
Standards) or background levels, then a corrective action plan must be submit-
ted. The owner/operator is not required, however, to submit this plan if he
can demonstrate that alternative concentration limits (ACL) at the point of
compliance will protect human health and the environment.
EPA has been developing, for several years, guidance on how these ACL's can
be determined. In August, 1984, Mr. John Skinner, then the Director of the
Office of Solid Waste, asked the Science"Advisory Board (SAB) to review
drafts of this guidance as soon as it was prepared by his Office, as well as
several case studies demonstrating applications of the proposed guidance.
The Environmental Engineering Committee of the SAB was assigned the task of
conducting the review. Several information briefings were given to the Com-
mittee by Mr. Vernon Myers, the project officer, but the draft guidance was
not delivered to the Committee until June, 1985, at which time they were
asked to make only a cursory review of the guidance and to inform the Agency
if there were any glaring problems which needed immediate resolution. It
was understood that a more complete review of the guidance would be done
when the Notice of Availability of the document was published in the Federal
Register.
In October of 1985, Mr. Myers sent the first two case studies to the Committee
for detailed review. This report is the results of both a cursory review
of the draft ACL guidance, and of the detailed review of the first two case
studies.
Committee Review Procedures
A Subcommittee of the Environmental Engineering Committee, consisting of Mr.
Richard Conway (Chairman), Dr. Donald O'Connor, Dr. Ben Ewing, Dr. Mitchell
Small and Dr. Joseph T. Ling, was organized to conduct the review. The ini-
tial briefing on the Guidance was presented to the Committee by Mr. Burnell
Vincent, Office of Solid Waste, at a meeting on August 16-17, 1984 (the
meeting was announced in the Federal Register on July 31, 1984, page 30596).
Subsequent status briefings were given by Mr. Vernon Myers, the ACL project
officer at regularly scheduled meetings of the EEC on February 26-27, 1985
(announced in the Federal Register February 11, 1985, page 5672) and on June
13-14, 1985 (announced in the Federal Register May 29, 1985, page 21936.
The draft guidance was furnished to the Subcommittee in June 1985. On
October 15, 1985, Mr. Myers also furnished the Subcommittee copies of two
-------
-2-
case studies intended to demonstrate application of the guidance, and to
serve as models to aid in implementing the June, 1985, draft of the guidance.
The Subcommittee held a meeting on December 3, 1985 {announced in the Federal
Register, November 13, 1985, page 46829) to review both the draft guidance and
the two case studies. The public had been invited to comment on the proposed
guidance and the case studies, but only one organization, the law firm of
Heron and Burchette, Washington, D. C., availed themselves of the opportunity. _
This report was drafted by members of the ACL Subcommittee,^with some additional
background information provided by Mr. Harry C. Torno, Executive Secretary to
the Environmental Engineering Committee. It was then reviewed and approved
by the Committee after modification. -•
-------
-3-
SECTION II
REVIEW OF THE DRAFT ACL GUIDANCE AND CASE STUDIES A AND B
Draft ACL Guidance Document
The SAB was requested to review the early draft (dated Jur>e 1985) of this
guidance document; only obvious omissions or errors were soyght, as a more
detailed review of the final draft destined for the Federal Register will be
requested from the SAB.
While the draft ACL guidance is in general well done, the following
improvements are recommended:
1. Executive Summary: An addition to the list of considerations regard-
ing surface water should be: "Surface water models that predict factors like
initial mixing zone,.final dilution, partitioning, bed effects" (p. E-3).
2. Introduction: The concept that "all Agency published exposure levels
for the protection of human health or the environment can be used as ACLs"
should be modified to include those Agency-approved values that have been
peer reviewed or subjected to public comment (p. 5).
3. Chapter III-"Hydrogeologic Characteristics": Consideration should-
be given to allowing a simplified approach of using worst-case assumptions
of hydraulic conductivity and porosity instead of detailed laboratory and
field testing of each stratigraphic unit. For example, assume a complex
situation has a conductivity of sand and a porosity of clay. This will re-
duce study costs, but still be protective of the environment.
4. Chapter IV-"Ground-Water Flow Direction": Consider allowing the
assumption without detailed study that flow on a long term net basis will be
toward surface water in obvious cases like a peninsula.
5. Chapter VII-"Proximity of Surfaca Water and Groundwater Users": Many
ACLs rest upon the dilution afforded by down-gradient surface waters. Conse-
quently the discussion of surface water models should be expanded to explain
factors like the applicability of mixing zones, the selection of low flow
rate, partitioning to suspended solids, benthos toxicity, resuspension of
deposited solids and transformation.
6. Chapter X-HPotential Health Risks": Consider allowing this part of
the study to be omitted when the proposed ACL is less than existing properly
published health-based water quality criteria. However, the inhalation path-
way specified on page 58 should not be overlooked when it has a significant
chance of contributing to the daily intake.
Some general guidance is needed to encourage applicants for ACLs to con-
centrate their effort in the critical areas and not spend efforts on factors
of low sensitivity.
-------
... -4-
i
Case Studies A and B
General
These case studies are intended to assist applicants for alternate con-
centration limits (ACL) in the preparation of the ACL demonstration in
accordance with ACL Guidance Criteria, Part 1. They may also be used by permit
writers for guidance in review of applications. Although the case studies are -
based on real situations, they are intended to be hypothetical1ACL demonstra-
tions which exemplify the organization, the logic and the detail required for
supporting information. Most important in these case studies is the methodology
used in the assessment of the health and ecological risks associated with
proposed contaminant concentration levels.
Case Study A and Case Study B represent cases which are so straightfor-
ward that the minimal nature of the health and ecosystem risks is virtually
self-evident. In such circumstances, an ACL demonstration should not require
such elaborate documentation. The preface for each case study should wake the
point that the case study is intended to be an example of the procedure for
preparing an ACL application for situations which may not be so self-evident.
Case Study A
Case Study A represents an extreme situation selected to demonstrate how
the ACL Guidance Criteria procedures under S264.94(b) would be applied in
arid hydrogeological and meteorological conditions. It involves a hazardous
waste treatment, storage and disposal site in the western US where evapo-
transpiration exceeds precipitation, where groundwater is too saline for use
as drinking water, and where the site is so remote that there is no human
exposure.
The contaminant transport modeling is valid based on a simplified worst-
case model; this is adequate to demonstrate the premise that groundwater
transport will be very slow (8xlQ~l ft/yr, conservatively) and degrading
constituents will degrade to negligible concentrations before traveling even
a short distance from the site.
The Introduction, Section 1, should indicate clearly the ACL.sought by
the applicant. It was not until Section 6, Alternate Concentration Limits,
of Case Study A that the reader learns that the request is for establishing
the ACL at the limit of solubility for each contaminant (see p. 40). This is
appropriate for organic compounds (if co-solvent effects can be neglected),
but Table 2-1 lists 10 metals which are hazardous constituents suspected to
be present at the site. The solubility of metals lepends on other chemical
species. The solubility that will be used in setting the ACL in the case of
these metals needs to be specified.
The conclusion reached in Case Study A that the ACL's should be the
saturation level is essentially based on the assertion or "demonstration"
that there is no present or future use of the groundwater.
Hence there would be no reason that potential pollution of the groundwater at
any level could have a significant health or ecological impact. However, the
saturation concentration for dieldrin cited on page 40 is extremely high
-------
-5-
(0.5 percent). The calculations based on the presumed transport conditions
indeed demonstrate that negligible concentrations of this compound (10-43
ppb) would occur in the groundwater. There may be, however, real reasons
to be wary of setting the ACL at such high concentrations as the saturation
level. Other health or ecological risks not foreseen "in this analysis could
be devasting with such high concentrations. If hydrologic and geologic
factors change, or future use of groundwater not now foreseen should develop,
these computations could be drastically changed. The case study should
incorporate a warning that such high concentrations ought to be subjected
to other tests of reasonableness. Perhaps a factor of safety should be.
applied to allow for unforeseen conditions. Alternately an ACL might be
based on a concentration level which is reasonably attainable under current
conditions even though it may be some orders of magnitude less than the sa-
turation level. It just does not seem prudent to accept such a high value
for an ACL on the basis of a single calculation or scenario.
The discussion of the transport and the health risk is insufficient
insofar as metals are concerned. The document emphasizes that the case is
not based on delay in hazard to humans or the environment. Since metals are
not degraded and their transport is retarded but not prevented (breakthrough
would eventually occur), the principal pr-otection is based on the presumption
that the highly saline groundwater can never be used for drinking water. The
case could be strengthened by further discussion of the metals. It is con-
ceivable that future population migration into this currently remote region
together with new technology for desalinization could lead to some future use
of the saline groundwater. Discussion of the effect of desalinization pro-
cesses in the removal of metals would make the case stronger More attention
to the nondegraded metals and the effect of chemical reactions which may
change the species solubility is recommended.
In this example ACLs for only organics are sought, as perhaps the metals
will be limited to background or drinking water levels. If so, this should be
stated.
The computation is based on a conservative "worst-case" estimate of the
inflow flux of contaminant to the aquifer (p. 31) but does not make clear how
this value is used in the steady state transport model. Was it assumed that
the entire contaminant content of the landfill was instantaneously inserted
into the groundwater flow? If so, how can this instantaneous event be incor-
porated into a steady-state transport model? What is the sustained rate of
input for the initial condition in the steady-state model? Or, was the inflow
flux assumed to be the product of the groundwater flow rate and the saturation
concentration of the contaminant, as is also suggested by the discussion on
page 31? If so, low reasonable is this assumption in view of the total
contaminant inventory within the landfill? Why are some of the available.
unsaturated flow models for vertical distribution not used for estimating the
flux of contaminant from the landfill to the groundwater? The assumption that
the initial concentration in the groundwater is the saturation value represents
a worst-case assumption, but it may be so unrealistic that the solution to
the problem is meaningless.
This analysis considers the landfill disposal unit but ignores the 200-
acre land treatment units. The combined effect of the land treatment units
in addition to the landfill on the groundwater pollution risk should at least
-------
-6-
be discussed. The analysis is also based on groundwater consumption for
drinking water as the exposure pathway. No other pathways are considered.
Inhalation of airborne contaminants may be a significant human exposure in
the case of the land treatment units. To what extent must multiple exposure
pathways be considered in ACL demonstrations?
The Case Study is strongly based on the arid meteorology,, the remote
location, and the-saline character of the groundwater. In view of the near-
permanence of groundwater pollution, there is some need for consideration of
possible future changes in land use (population in-migration, industrial
plant location with possible groundwater utilization, etc.f or future changes
in water use (new desalinization technology, etc.).
Some more specific comments follow:
o Ownership of the surrounding land by the US Government is stated (p. 3),
Who owns the 800-acre site?
o Evaporation rates are estimated at 60x precipitation rate (p. 7).
This does not" seem to agree with Table 3-1.
o The permeability value selected is cited as being conservative (p. 31);
all results should be presented, not just the one selected, to allow
this selection to be evaluated.
o Only one transformation and retardation value is given (p. 32); the
less conservative values should be cited and referenced.
o The effect of the high salt content on transformation and retardation
should be discussed (p. 32-38).
o The effect of acclimation on transformation should be discussed as
well as specifying whether the half-life constants were obtained with
acclimated or unacclimated microorganisms (p. 37).
o The reporting of computed concentrations at infinitesimally small
values (e.g., 10"43 ppm or ppb on p. 37 and 40; 10'59 ppb on p. 41;
10-532 on p> 53) may imp]y to some that there exists an ability to
compute and differentiate between these values, some of which may
require dilution volumes exceeding the dimensions of the solar system.
Anything below 10"^ ppb should simply be called "less than one ppq".
o Table 5-2 on p. 36 requires clarification. The indicated values are
actually "adsorption" (or "distribution", or "partition") coefficients,
not "retardation" coefficients as indicated in the title. Also, it
should be noted that computation of Kd from K0c assumes a fraction
of organic carbon in the soil of foc - 0.001, i.e.,
Kd = foc Koc
Was the assumed foc of the soil at the site ever specified?
-------
-7-
With the provisos expressed above, especially those concerning the se-
lection of solubility limits, the report is a scientifically sound evaluation
of potential groundwater problems at the indicated waste disposal site. The
level of detail and impact criteria selected appear to be appropriate, given
the essential absence of a beneficial use for the aquifer surrounding the
site.. The selection and implementation of the predictive model and identifi-
cation of parameters appear to be reasonably well founded. With the improve-
ments recommended," it can be useful in demonstrating how the ACL Guidance
document (Part I) can be applied in a specific case.
Case Study 8 .
This document presents an example of the arguments used by Facility B
for Alternative Concentration Limits (ACLs) under 40 CFS 264.94 for five
hazardous constituents that have been detected in the groundwater near the
facility. The five constituents are o-dichlorobenzene, p-dichlorobenzene,
ethylbenzene, toluene, and chlorobenzene. The concentration limits for all
other hazardous constituents will be background concentrations or, where ap-
plicable, the maximum concentration levels specified in Table 1 of 40 CFS
264.94.
Facility B is located in the Piedmont region of the southeastern United
States, at the confluence of River A and Creek B. The report presents a sum-
mary of the hydrogeology, exposure pathways, contaminant levels, contam-
inant transport and Proposed ACLs.
This review is primarily directed to the contaminant levels and transport
and the proposed ACLs, the determination of which was based on the hydraulic
and contaminant transport in the groundwater and river.
o The groundwater flow and transport model, which is vertically inte-
grated and two dimensional, is appropriately applied to reproduce
groundwater levels. Although approximations were necessarily made
to simplify the systems, the computed water levels are in qualitative
agreement with those observed, from which is determined the flow to
each of the compliance wells and to the river and creek, which are
boundaries for the area. The general conformance of observed and
computed water levels indicates that reasonable values of the various
coefficients (e.g., conductivity, permeability, porosity, and recharge)
were assigned. The difficulty of assigning precise values of these
parameters is recognized, considering the heterogeneous composition
of the aquifer. Although the observed and computed groundwater levels
are in qualitative agreement, it would be advisable to present a more
direct comparison in order to provide a more quantitative validation
of the hydrodynamic model.
o The next phase of the analysis presents a discussion of the relevant
characteristics and constituents, i.e., electrical conductivity and
the organic chemicals. Although the Appendix contains the appropri-
ate models for calculation of the spatial distribution of these con-
stituents, no such model runs were conducted. The observed distribu-
tion of total dissolved solids (TOS) is particularly suitable to use
for further independent validation of the groundwater transport
model.
-------
-8-
The section on the contaminant transport model provides the necessary
background information and input data, but does not present any model
runs of the spatial distribution of the organic chemicals. The obser-
vations of the total chlorobenzenes provide a good basis for the cali-
bration of the contaminant transport model. As in the case of IDS,
when such data are available it should be mandatory to use these
observations for the purpose of calibration/validation of the model's
when such-distribution is critical to the ACL application (which is
not the situation in Case B, but will be true at other times). This
counsel should be added to Part I of the guidance document.
The crux of the analysis lies in the mass balance relation (Equation
6-1). This equation contains the flows and concentrations of both
surface and subsurface regimes, all of which are assignable with the
exception of the groundwater flow. The manner in which this flow is
computed is valid and relevant, but obviously subject to significant
variation depending on the coefficients assigned to the groundwater
model. Any variation in the groundwater flow inversely affects the
ACL in a linear fashion. Since the other terms in Equation 6-1 are
directly specified, the concentration is sensitive only to the ground-
water flow.
In the final analysis, there is no need in this case for extensive
modelling of the groundwater. The problem is essentially concerned
with conditions in the surface water, which serves as a source of
potable water and a fishery resource. As indicated in the previous
section, the critical factor in the calculation of the ACLs is the
ground water flow. This flow may also be determined from a hydraulic
balance of measured river flows at two locations in the river. The
incremental flow (cfs/mile), which varies with the magnitude of the
river flow, may be evaluated from published flow records for the
various hydrologic regions of the country. The determination of
the flow therefore does not necessarily require a groundwater calcu-
lation. It is, however, desirable to carry on both analyses to
provide an interlocking check on the flow balance.
Since the problem is essentially concerned with water quality con-
ditions in the river, it would be appropriate to direct more attention
to a contaminant transport model in that type of surface water system,
Factors such as partitioning to suspended and bed solids, decay and
volatilization, and lateral and vertical mixing, should be considered.
Depending on the dimensions of the river and the distance downstream
to nearest water use, a mixing zone model may have to be developed,
in which both transverse and vertical variations are considered. If
these considerations are relevant in a given case, the ACLs would be
lower because the entire cross-sectional area may not be available
for dilution, resulting in a more conservative assessment.
The statement "there is no evidence that surficial groundwater flows
under the river," should be further substantiated and explained with
respect to the possible contamination of the groundwater west of the
river.
-------
-9-
o The allowable portion of ACL Is a major component in the calculation
of maximum concentration for each of the constituents. These percent
ages are allotted based on "prevalence" of the contaminants (p. 4-8).
It would be appropriate to justify more fully this proportioning,
which has a direct effect on the alternate concentration limits.
o The following items are noted for improvement and correction:
1. Equation 4-4 (p. 4-7) has a typographical error
2. The equations (unnumbered) on the top of p. 5-3 are typed
incorrectly. Rather than Qx = 0 = etc. and Qx = L = etc., it
should read Q(at x=o) = etc , and Q(at x=L) = etc. Also, it
should be noted that positive Q means flow in the +x direction,
and negative Q means flow in the -x direction.
3. Equation (6-1) on p. 6-3 is written incorrectly, but the cor-
rect version appears to have been used for the calculations.
It should read:
Ca(Qu + Qgw)-QuCu
ACL = - - -
4. The isopleth map on p. 2-12 does not seem consistent with the
data in Table 2-2 (p. 2-10).
In summary, tf.e document reflects an understanding of the necessary
elements of a determination of ACLs and presents a satisfactory analysis of
the groundwater flow and human exposure concentrations, but an inadequate
assessment of the TDS and contaminant concentrations. More attention should
be directed to the river conditions; otherwise this report generally is a
good engineering study with sufficiently realistic and conservative assumptions,
-------
rv
230 South Dearborn Street
Chicago, Illinois €0604
-------
-10-
SECTION III
SUMMARY/CONCLUSIONS
I. Part I; Information Required in ACL Demonstrations
The general guidance document is well done for an eaHy draft; the
suggested improvements the SAB noted include more emphasis on surface-water
modeling as it often is a sensitive factor, peer review of Agency-published
acceptable exposure levels, and encouraging the utilization of conservative
assumptions in areas of low sensitivity rather than spending resources to
develop high-reliability estimates.
II. Part II: Case Studies A and B
Both Case Study- A and B represent generally useful demonstrations of
implementation of the ACL Guidance Document, although both are in need of
improvement before issuance. Because of the particular extreme circumstances
of the two case studies, the full range of models demonstrated may not have
been necessary far these evaluations, but they do serve to demonstrate the
methodologies. There are a number of technical problems, including omission
of necessary information and discussion in certain instances (such as the
identification of the inflow flux rate assumed in Case Study A), the need for
more (and earlier) emphasis on critical components of the analysis (such as
the stream model calculation in Case Study B), inconsistencies in certain
tables and figures, and typographical errors in both reports. These are
detailed in the reviews of each study. .These problems should be relatively
easy to correct, and the resulting reports should provide clear and effect-
ive guidance for the ACL implementation.
There is one additional area of concern raised by the reports. In both
cases, derived concentrations of the organic chemical ACL for the sites are
very high. These are based on the high level of dilution and (for Case Study
A) decay provided by the surrounding aquifer or stream. The result is that
the ground water immediately below or adjacent to the site is permitted to be
thoroughly degraded by the ACL chemicals. We question this approach, even
though the narrowly stated goal of the ACL, to protect human health and the
environment, is apparently satisfied. This concern is compounded by the possi-
bility of future changes in land use or other unforseen circumstances that may
make these ground waters more accessible to human or ecosystem exposure. The
suggestion is that there may be some minimum level of protection which should
be afforded to all ground waters, even those without direct beneficial use.
We recognize that this is more an issue of social or political choice, rather
than scientific credibility, and as such may be outside the province of the
SAB. However, there is a concern that this important social issue may be
masked and hidden by the scientific methodology developed for the ACL process.
The Agency is encouraged to recognize this possibility when finalizing the ACL
requirements.
-------
US. Environmental Protection Agency
Region V, Library
230 South Dearborn Street v
Chicago, fffinols 60604
------- |