^€0 ST4*.
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
MAR I 8 1994
OFFICE OF
WATER
Dear Colleague:
EPA's Office of Wastewater Enforcement and Compliance (OWEC)
formed a workgroup in 1991 to address the issue of implementing
and enforcing water quality-based effluent limits (WQBELs) set
below analytical detection/quantitation levels. Members included
representatives from EPA Headquarters, Regions, and States
(attachment 1). The purpose of this letter is to transmit the
product of that group, the draft "National Guidance for the
Permitting, Monitoring, and Enforcement of Water Quality-based
Effluent Limitations Set Below Analytical Detection/Quantitation
Levels" (attachment 2), for your review and comment.
Previous EPA guidance (Dioxin Strategy - 5/90 & TSD - 3/91)
recommended that minimum levels (MLs) be used to evaluate
compliance with WQBELs set below quantitation. This guidance
goes a step further and makes recommendations for calculating
interim MLs when promulgated MLs for a particular analytical
method do not yet exist, and for reporting results below
quantitation. The purpose of this document is to encourage
national consistency in this area by providing recommendations
for addressing those WQBELs until analytical chemistry has
advanced enough to measure to them.
The four major elements in this guidance are: 1) a permit
limit is expressed as the calculated WQBEL; 2) the ML is used as
the quantitation level and is included in the permit as a
footnote to the WQBEL; 3) where a promulgated ML is not
available, an "interim ML" is calculated using a factor of 3.18
times the method detection limit (MDL); and 4) analytical results
below the ML are reported as "0."
The workgroup for this project includes representatives from
EPA Headquarters, EPA Regions, EPA Office of Research and
Development, and five States (FL, MD, NJ, TX & WI). The first
working draft of this document was released to the workgroup and
the Regions for review in December 1992. A draft was released to
the same group in October 1993. The two major areas we received
comments on were: how to develop interim MLs, and how to report
analytical results below quantitation.
The rationale and justification for the major
recommendations included in the document are as follows:
RacycMfftocydaM*
C\ WrtttdonMpwiruicortiici
CjC7 « rvrrQM dm
-------
1. Permit limits are expressed aa the-calculated hobbl
• This statement is in conformance with the Clean Water Act
(CWA) and EPA regulations at 40 CFR 122.44(d) that require
limits to be protective of water quality standards.
2. ML is used as the quantitation level and Is included in the
permit as a footnote to WOBEL
• The Strategy for the Regulation of Discharges of PCDDs and
PCDFs from Pulp and Paper Mills to Waters of the United
States (May 21, 1990) recommends that the permit writer
specify the ML as the compliance evaluation level in permits
that limit dioxin.
• March 1991, EPA expanded its guidance on quantitation levels
in the Technical Support Document for Water Quality-based
Toxics Control to include all other pollutants.
• The ML is a concept originated by EPA's Office of Water
Engineering and Analysis Division in developing the 1600
series analytical methods.
• The ML is the concentration in a sample that is equivalent
to the lowest calibration standard analyzed by a specific
analytical procedure.
• MLs were developed to cope with the high level of
variability among laboratories, analytes, and matrices when
developing MDLs.
3. Where promulgated MLs are not available, an "intoiHw irr." in
calculated bv multiplying 3.18 times the MDL.
MDL Component of Interim ML:
• The MDL is the cornerstone of the ML. The MDL has withstood
court challenges and has generally been accepted by the
regulated community. It is also defined in EPA regulations
at 40 CFR Part 136, Appendix B.
• Resources are not available at this time to derive a new
endpoint for the methods.
• MDLs are being developed for metals, and MDLs are not too
expensive to generate.
Factor of 3.18i
• The relationship between the ML and MDL varies between
2.7 - 5.0, depending on the analyte. Multiple laboratory
confirmation indicates that MLs are closer to 2.7 - 3.3
times the MDL.
2
-------
• The factor of 3.18 is scientifically based, consistent and
documented by the American Chemical Society (ACS) and the
International Union of Pure and Applied Chemistry (IUPAC)
for a limit of quantitation.
• Multiplying 3.18 times the MDL contains the practical
procedures necessary for calibrating analytical instruments
and making environmental measurements.
• The EPA Office of Water Analytical Methods Staff used this
factor in deriving MLs for the 1600 series methods.
• The development of interim MLs should be based on the same
principles as the MLs. It should contain the operational
procedure of the MDL. A factor of 3.18 times the MDL should
be applied in order to derive interim MLs.
• We recommend that 3.18 be used until a promulgated factor is
derived, additional MLs are derived, or the scientific basis
of 3.18 changes.
4. Analytical results which fall below the ML should be
reported as "0."
• Defensible compliance tool: This approach avoids taking an
enforcement action based on results falling in a region of
uncertainty (below the ML). Where the analytical value is
at or above the ML (i.e., non-zero) it provides certainty
that a violation has indeed occurred. (This approach is
successfully being used by Region VI for dioxin limits.)
• Will result in consistency of reporting values below
quantitation and calculating monthly averages.
• Easy and cost effective to implement. Requires no change to
reporting or tracking systems (e.g. PCS).
• The majority of commenters on October 1993 draft favored the
use of zero.
The recommendations presented in this guidance will not
satisfy everyone's concerns, however, they are based on current
regulations, science, and policy and they are defensible. Please
keep in mind that this document is guidance only and it does not
offer a permanent solution to the issue of implementing and
enforcing WQBELS' below quantitation. The approach presented is
an interim solution and, as science and technology in this area
evolves, it will need to be revised.
3
-------
Please send your comments to Jackie Romney (mail code: 4203)
or Kathy Smith (4202) of my staff by Friday, April 29, 1994. At
this time, we are also sending the document for review by EPA
Regions, States, workgroup members, and other external groups.
We will consider all comments received by April 29 and expect to
have a final document in June 1994.
Sincerely yours*,
%UW IjJJn
Michael B. Cook
Director
Office of Wastewater Enforcement
and Compliance
Attachment
4
-------
PRATT
MARCH 22, 1994
NATIONAL GUIDANCE FOR THE
PERMITTING, MONITORING, AMD ENFORCEMENT 07
WATER QUALITY-BASED EFFLUENT LIMITATIONS
SET BELOW ANALYTICAL DETECTION/QUANTITATION LEVELS
I. PURPOSE
Regions and States have expressed a need for guidance
regarding the implementation and enforcement of water quality-
based effluent limitations (WQBELs) set below analytical
detection levels. The problem is that science has not kept pace
with policy in this area. The purpose of this document is to
provide interim guidance until analytical chemistry can measure
such low levels.
This guidance contains interim techniques for establishing
compliance at the WQBEL. Existing EPA guidance and this guidance
recommend that minimum levels (MLs) be used to make a compliance
determination. The ML is a concept originated by the Office of
Water, Engineering and Analysis Division, in developing the 1600
series methods. Currently, methods 1624 and 1625 are the only
EPA methods promulgated under Section 304(h) that have MLs.
However, other 1600 series methods have been or are presently
being proposed and contain MLs. The most notable proposed method
is Method 1613, for chlorinated dioxins and furans. Use of this
method has been required for permit compliance monitoring since
its proposal in February 1991. EPA's 600 series methods contain
method detection limits (MDLs), not MLs. This document provides
guidance for developing interim MLs from MDLs in those instances
in which MLs do not exist. These interim MLs are intended to b«
temporary—to be used only until minimum levels are promulgated.
This approach addresses the immediate need to provide interim
quantitation levels to determine compliance with WQBELs.
Alone, this guidance does not adequately address values
below analytical detection. Progress in advancing technology
must continue, including development of minimum levels, new
analytical methods with lower detection limits, and better
quality assurance/quality control techniques. This progress will
require prioritization and commitment of resources. Only with
that commitment and the subsequent advancement of technology will
we truly achieve national consistency in addressing WQBELs below
detection levels. As science and technology in this area
evolves, this document will need to be revised.
The spirit behind this guidance is that even in the absence
of more sensitive promulgated methods, laboratories should do
their best to measure as closely as possible to WQBELs. Since as
an agency, our ultimate goal is to protect the Nation's surface
-------
waters, and eventually achieve the goal of "no discharge of
toxics in toxic amounts," we recommend that every effort be made
to use the best available techniques for measuring these low
levels.
This document was developed using current information and is
EPA's approach to using 40 CFR Part 136 and other analytical
methods to achieve WQBELs set below analytical detection levels.
In addition, this document represents guidance. As such, it is a
general statement of policy. It does not establish or affect
legal rights or obligations. It does not establish a binding
norm and is not finally determinative of the issues addressed.
Agency decisions in any particular case will be made by applying
the law and regulations on the basis of specific facts and actual
action.
In some cases, this guidance reiterates statutory or
regulatory requirements and provides citations to the relevant
statutory or regulatory provisions. Otherwise, this guidance
makes recommendations only. These recommendations are not
accompanied by statutory or regulatory citations.
The recommendations in this document will not be appropriate
in all situations. It is the responsibility of the regulatory
authority to use all available information to assess the
situation and then to implement additional or more stringent
requirements, as necessary, to protect aquatic life and human
health from the impacts of pollutants being discharged in amounts
below analytical detection.
II. EXBCUTXVB 8UMMARY
As a result of the increasing use of WQBELs in National
Pollutant Discharge Elimination System (NPDES) permits, a
significant number of permits now contain limits that fall below
the detection/quantification capability of current analytical
technology for certain analytes. A major concern to permit
writer, permittee, and enforcement authority is how to:
• Incorporate calculated limits below detection levels
into permits
• Monitor effluent and require that analytical methods
used in effluent monitoring measure the lowest
accurately quantifiable level possible
• Report analytical results when data fall below detection
levels
• Determine compliance with limits that fall below
detection levels
2
-------
• Track analytical results
• Facilitate advances in technology
The EPA Offices of Wastewater Enforcement and Compliance (OWEC)
and Office of Science and Technology (OST) have developed this
guidance to address these concerns.
This document includes the five following basic alaments:
• WQBELs are to be imposed as NPDES permit limits vhen
necessary to protect the designated uses of the
receiving waters, even when the WQBEL is less than the
analytical detection level.
• The quantification level to be used in analyzing samples
and reporting the results is the ML. Guidance for
incorporating MLs into NPDES permits, developing ML*
where they are not currently available, and translating
analytical values below the KL into the discharge
monitoring report (OMR) data for compliance evaluations
is provided in this document.
• The existing analytical data reporting and tracking
mechanisms, the DMR and permit compliance system (PCS),
respectively, will continue to be used with no changes
to their structures. This document provides guidance
and examples for reporting analytical values that are at
or below the ML. The ML has an indirect role in
compliance evaluations. The ML is the threshold for
recording the analytical data that will then be used to
calculate and report summary information on the DMR.
• Enforcement criteria are consistent with those set forth
in the EMS guidelines, and compliance evaluations
continue to be made directly between the DMR information
and the permit limit (which is the WQBEL in cases where
water quality is the most stringent limitation).
• Important permitting and compliance issues, such as
addressing matrix interferences, using internal outfaile
in permit writing, driving technology to achieve lower
detection levels, specifying analytical protocols in
MPD23 permits, and determining the ability of
laboratories to achieve specified MLs are also addressed
in this guidance.
III. BACKGROUND
At present, EPA Regions and NPDES-authorized States use a
wide variety of approaches for establishing water quality-based
permit limits, including the imposition of MDLs, Practical
3
-------
Quantification Levels (PQLs), Minimum Quantification Levels
(MQLs), Quantification Levels (QLs), MLs, Practical Detection
Levels (PDLs), Limits of Detection (LODs), and/or setting levels
as an unspecified non-detect. While the rationale supporting
these approaches may differ, the basis for each is shared, that
is, the need to establish a level in the permit that can be
measured for evaluating compliance.
The Office of Water issued Final Guidance ozt Section 304(1);
Listing and Permitting of Palp and Paper Mills (March 15, 1989)
(hereinafter the 304(1) Guidance) recommending that where
calculated WQBELs are less than the detection level for the
specified analytical method, the calculated WQBEL should be
included as a requirement of the permit. This guidance is in
conformance with the Clean Water Act (CWA) and EPA regulations at
40 CFR 122.44(d) that require limits to be protective of water
quality standards. This guidance also stated that the detection
level of the analytical method should be the threshold for,
compliance/noncompliance determinations. The detection level
issue was also raised in the Strategy for the Regulation of
Discharges of PCDDs and PCDFs from Pulp and Paper Mills to Waters
of the Unites States (May 21, 1990) (hereinafter the Dioxin
strategy). This modified the March 15, 19A9 guidance by
recommending that the permit writer specify the ML as the
compliance evaluation level in permits that limit dioxin.
In March 1991, EPA further expanded its guidance on
detection levels in the Technical Support Document tor Water
Quality-based Toxics Control (hereinafter the TSD). On page ill
of the TSD, EPA recommended applying the concepts contained in
the Dioxin Strategy to analytical detection levels for all
pollutants. Despite the support for the use of MLs, obstacles
exist that prevent their wider use. These obstacles include the
availability of MLs, the historical misuse of detection and
quantification levels, and the inconsistent definitions
representing these levels.
17. INTRODUCTION
This guidance outlines the responsibilities of both the
permittee and the permitting authority to ensure accurate
monitoring, reporting, and tracking as analytical technology
advances to meet lover detection levels necessary to protect
water quality. Specific guidance is included to assist the
permit writer and compliance officer in writing and enforcing
permits with WQBELs that fall below detection levels.
This guidance is designed to provide an interim way to deal
with analytical results that fall below detection until advances
in technology are made that allow lower detection levels and thus
increase the ability to ensure protection of the Nation's waters,
as well as to encourage source reduction that will limit the
4
-------
discharge of the pollutants. Figure 1, which appears on the next
page, demonstrates the problems created by WQBELs that fall below
detection levels. It depicts how future technical advances in
analytical chemistry detection and quantification levels can
resolve the problems created by WQBELs that are below these
levels.
This document was developed with significant input from the
Regions, States, and other EPA Offices. EPA's Office of Research
and Development (ORD) is currently conducting a method validation
study for trace metals by inductively-coupled plasma (ICP)
spectroscopy. That study is designed to address issues related
to method detection limits. ORD's definitive study should
produce an appropriate data base for defining and evaluating
detection limits and quantification levels as well as state-of-
the-art MDLs and MLs for toxic metals.
This guidance is consistent with EPA's May 1990 Dioxin
Strategy as well as other pertinent permitting and enforcement
documentation (e.g., Quarterly Noncompliance Report (QNCR)
guidance, the TSO, and the Permit Compliance System (PCS) Policy
Statement).
In summary, the goals of this guidance are: (1) to outline
objectives for achieving consistency in establishing permit
pollutant limitations for pollutants that are set below detection
and quantification levels, taking into consideration the current
capabilities and uncertainties of science and technology, and (2)
to provide specific guidance on implementing the objectives of
this approach in a rational and defensible manner.
This document is organized in four sections: (1) Key
Definitions, (2) Permitting Principles and Strategy, (3)
Compliance Monitoring and Enforcement Principles and Strategy,
and (4) Additional Permitting and Enforcement Strategy Issues.
The appendices provide: (1) a list of available quantitation/
detection levels taken from 40 CFR Part 136, (2) procedures for
reviewing discharge-specific MDLs, (3) recommended permit writing
boilerplate language, (4) recommended permittee procedures for
reporting, (5) compliance evaluations in accordance with this
guidance, and (6) recommended permit writer procedures for
calculating WQBELs at internal outfalls.
V. XBY DBVXVXTX0H8
The following terms are used in this document; they are
consistent with EPA regulations and existing EPA guidance.
Detection Limit (DL)
The lowest concentration or amount of the target analyte
that can be determined to be different from zero by a single
5
-------
FIGURE 1
DETECTION/QUANTIFICATION OF
WQBELS
WQBEL
WQBEL
WQBEL
FUTURE
LIMITS
B«st Available Technology limits, by definition, are s«e at
concentration* that arm quantifiable. Water Quality-baaed Effluent
Limitations, to bm protective, Arm often amt belour the Minimum
Levml and aoamtiamm belotf the Method Detection Level. In the
future, am analytical chemistry improves, thm MLa and MDLs will
become more sensitive and approach the lover WQBSL.
6
-------
measurement at a stated level of probability.
Interim Minimum Level (Interim ML)
The interim ML is calculated when a method-specified ML does
not exist. It is equal to 3.18 times the method-specified
MDL rounded to the nearest multiple of 1, 2, 5, 10, 20, 50,
etc. The interim ML should be used until an analytically
developed ML can be established.
Method Detection Limit (MDL)
The minimum concentration of an analyte that can be measured
and reported with 99 percent confidence that the analyte
concentration is greater than zero as determined by a
specific laboratory method (40 CFR Part 136). [An MDL is
usually determined in reagent water, but a discharge-
specific MDL is sometimes developed using effluent. The use
of reagent water is for two purposes: (l) reagent water as
a reference matrix is available to all laboratories who can
then reproduce the work, and (2) it is necessary to
determine the MDL at very low concentrations of analyte. It
may not be possible to find concentrations this low in a
wastewater effluent.] The procedure for determination of an
MDL is in 40 CFR Part 136, Appendix B.
Minimum Level (ML)
The concentration at which the entire analytical system oust
give a recognizable signal and acceptable calibration point.
The ML is the concentration in a sample that is equivalent
to the concentration of the lowest calibration standard
analyzed by a specific analytical procedure, assuming that
all the method-specified sample weights, volumes, and
processing steps have been followed. Quantifying
measurements below the ML requires extrapolation of the
calibration relationship below the range of data used to
establish the calibration, such an extrapolation is not a
preferred practice and leads to greater uncertainty in the
quantitative result.
Minimum Quantification Level (MQL)
An MQL ia a concentration identified by EPA Region 6 baaed
on a thorough evaluation of current literature and
analytical capabilities to be used until more analytically
developed MLs can be established.
Practical Quantification Level (PQL)
A quantification limit that is approximately 3 to 10 tiaee
the MDL and that is greater than the concentration of a
7
-------
chemical that can be detected by current laboratory methods,
thereby providing a less demanding, "practically-based"
analytical limit. This definition of the PQL is different
from the PQL defined by the Office of Solid Waste, and
published in earlier versions of SW-846 methods. The PQL is
not used in this guidance.
Quantification Level (QL)
A measurement of the concentration of a chemifcal obtained by
using a specified laboratory procedure, at a specified
concentration above the detection level. Examples are the
ML and PQL.
Water Quality-Based Effluent Limitation (WQBEL)
The limitation developed in a wastewater discharge permit to
control all pollutants that will cause, have the reasonable
potential to cause, or contribute to an excursion above any
State or Federal water quality standard or criteria. WQBELa
are imposed in NPDES permits when it has been determined
that more stringent limits than technology-based effluent
guidelines must be applied to a discharge to protect the
designated use of the receiving waters.
The Best Available Technology Economically Achievable, by
definition, sets limits at concentrations that are
quantifiable. WQBELs, to be protective, are often set below
the ML and sometimes below the MDL. As analytical chemistry
technologies improve, MLs and MDLs are likely to be lowered
such that measurement of the WQBEL becomes possible.
Summary of Concepts:
LESS THAN (<) MDL Not Detected
GREATER THAN OR EQUAL TO (>)
MDL BUT LESS THAN (<) ML Not Quantified
GREATER THAN OR EQUAL Quantified (Numerical
TO (>) ML Result Reported)
VI. PESMZTTZVa PRINCIPLES AHD STRATEGY FOR WATER QUALITY-BASED
EFFLUENT LIMITS THAT FALL BELOW ANALYTICAL METHOD DETECTION
LEVELS
A. Permitting Principles
This permitting strategy is based on the following
principles:
8
-------
• Principle I: Permit requirements must be protective of
water quality.
• Principle II: Permits must be written to avoid
ambiguity and ensure enforceability.
• Principle ills Permit requirements must support the
achievement of the lowest accurately quantifiable level
possible when f/QBBLs fall below detection levels.
B. Permitting Strategy
EPA'a NPDES regulations at 40 CFR Part 122 state that permit
limitations must reflect the most stringent of technology-based
or water quality-based controls, or other standards required by
the CWA. Once it has been determined that a WQBEL is needed to
protect water quality, consideration must bo given to the
capability of available analytical chemistry to measure at the
WQBEL. Permitting authorities must increasingly contend with
enforcing WQBELs that fall below what can be measured using
current analytical chemistry techniques.
Reasonable Potential fRP)
The first step in developing a water quality-based limit Is
to determine whether a limit is needed. Permitting authorities
must determine whether a discharge causes, has the reasonable
potential to cause, or contributes to an excursion above a water
quality standard. The procedures used in making this
determination are discussed in Chapter 3 of the TSD. When
monitoring results are at levels above analytical quantification
levels, they are easily compared to the water quality standard to
determine reasonable potential.
However, when monitoring results are below analytical
quantification levels, a decision must be made on how to
interpret the data to determine RP. In this case, necessity for
the reasonable potential procedure should be based on the
likelihood that the pollutant is present in the discharge(s)
resulting from operation of the facility, once this
determination is made, the following guidance offers
recommendations on how to interpret data to make a reasonable
potential analysis.
When determining reasonable potential, actual analytical
results should be used whenever possible. When an analytical
result is below the ML, the Agency recommends that the permit
writer set this result equal to zero because the result is not
quantifiable.
When the effluent data consist of non-quantifiable/non-
detectable values, the regulatory authority should consider other
9
-------
that limits be protective of water quality. The use of the ML as
an analytical chemistry performance standard provides an
unambiguous and rational means to demonstrate that the best
chemistry available at the tine of permit issuance is being used
in conjunction with WQBELs that are below quantification levels.
The Agency is recommending the use of the ML when WQBELs are
below analytical detection and quantification levels to address
the problems associated with recording and reporting. EPA
believes that the use of MLs encourages permittees to maintain
the highest quality assurance/quality control (QA/QC) standards,
as well as strict standard operating procedures, which, in turn,
will ensure that the self-monitoring data reported to the
permitting and enforcement authority will be as accurate as
possible.
The ML should be included in a footnote to the WQBEL in the
NPDES permit. These MLs are to be used as quantification levels
for recording and reporting purposes and are not to be confused
with the compliance level, which is the WQBEL. As part of this
guidance, a summary of available quantification and detection
levels has been provided in Appendix A. In the absence of
promulgated MLs, interim MLs should be used. The basis for the
interim ML is the MDL, and the procedure for calculating interim
MLs follows on page 12.
EPA recognizes that not all analytical methods approved
under 40 CFR Part 136 have MDLs or MLs and therefore recommends
that permit writers explicitly require analyses in accordance
with the most sensitive technique within those approved methods.
The method improvement provisions in the quality control section
of some methods (e.g., 40 CFR 136, Appendix A) allow the analyst
to use more sensitive techniques than those given in the method.
In addition, 40 CFR Parts 13 6.4 and 136.5 provide procedures for
using alternate test procedures (ATPs) which may allow for a
lower level of detection.
Permit writers should also require calibration of laboratory
equipment with standards that include a concentration equivalent
to the ML. The Agency believes that it is appropriate to specify
the analytical method since specified MLs are based on a specific
laboratory technique, and facilities must perform analyses in
accordance with that technique or an analogous technique to
achieve the ML. Additionally, since the ML is defined as
equivalent to the lowest calibration standard, it is appropriate
to specify that the ML be used as the lowest calibration
standard.
The Agency recommends that reporting requirements in permits
specify that actual analytical results be reported whenever
possible. When analytical results are below the minimum level
that can be quantified (i.e., below the ML or interim ML), the
11
-------
Agency recommends reporting "zero." This recommendation provides
a two-fold advantage: (1) it ensures a margin of relief to the
permittee seeking to avoid false positives which lead to
violations, and (2) in the cases where the analytical value is
non-zero, it provides certainty to the compliance personnel that
a violation has indeed occurred where such is noted on the OMR.
This recommendation will not satisfy everyone's concerns.
Some people will interpret this approach to imply 'that EPA
believes that there are no pollutants in results that fall below
the ML. In the context of this guidance, zero is a numeric
symbol selected by the Office of Wastewater Enforcement and
Compliance to represent non-measurable data and thereby
facilitate consistency in DMR calculations and reporting.
EPA recognizes this approach introduces some bias to the
data and does not drive technology to achieve lower detection
levels. In developing this guidance, a number of different
approaches were considered. The use of zero was chosen because
it is defensible and is successfully being used by the Agency to
issue and enforce permits with dioxin limits, and because it can
be implemented without changes to EPA monitoring and reporting
systems. This approach is meant to be an interim solution which
should be periodically re-evaluated and improved as improvements
in technology occur.
EPA believes compliance with permit limitations that are
derived from "no discharge" or "no detectable amounts" effluent
limitations guidelines are to be determined by qualifying samples
as less than the MDL. For these limitations, which are
established to essentially prohibit any addition of a pollutant
to the receiving water, the standard of compliance is set at the
minimum level of detection as opposed to a level of quantitation
(e.g., the ML). The application of the MDL in instances of no
discharge or no detectable amounts limitations will provide
assurance to the 99th percentile against a false positive.
Permit limitations based on such effluent limitations should
clearly establish the standard of compliance at the MDL in the
fact sheet and state compliance with the no discharge standard
will be established by reporting analytical results below the
approved MDL as zero on the DMR.
This approach is required for situations where the discharge
standard has been establish to effectively prohibit any further
addition of a specific pollutant to waters of the U.S. One
example of this situation is at 40 CFR Part 423, the effluent
limitations guidelines for the steam electric category. This
effluent guideline has several limitaions which establish a no
discharge in detectable (not quantifiable) amounts, specifically
the prohibition on the discharge of PCBs and priority pollutants
other than chromium and zinc in chemicals added for cooling tower
maintenance (see 40 CFR 423.13(a) and 423.13(d)(1)). Compliance
12
-------
with permit limitations based on these effluent guidelines would
be shown by analytical results demonstrating that there was no
pollutant of concern in detectable amounts, i.e. at levels at or
above the MDL. Results at or below the MDL would be reported as
zero on the DMR.
Permits should require that the permittee provide
information that will be needed when reissuing the permit. In
support of permitting Principle II. EPA recommends that the
permit writer require the permittee to report the 'achieved ML and
the number of occurrences during the month when analytical
results were below the ML (i.e., number reported as zero) in the
comment field of the DMR form or on an attachment to the DMR
form. This allows gathering of sufficient information for use in
reasonable potential calculations, as well as for other usee.
EPA recommends that permits that have pollutant limits for
aquatic life that are set below analytical detection levels
include the requirement to perform whole effluent toxicity (WET)
tests. WET testing provides the only mechanism available to
measure the cumulative toxic effects of one or more pollutants
and counteract the potentially misapplied "benefit of the doubt"
provided to a facility by the translation of data less than the
ML.
Additionally, EPA supports the use of other measures to
protect water quality. For example, mercury being discharged in
amounts below quantitation may be suspected of impacting the
benthic community. In such situations, the regulatory authority
should go beyond the recommendations in this guidance and require
bioaccumulation or sediment studies, pollution
reduction/prevention plans, or other measures to protect water
quality.
As part of this guidance, EPA has provided, in Appendix C,
boilerplate language that includes requirements for setting
limitations and specifying monitoring and reporting.
Basis for MLaz
Analytical data are characterized by many different teres.
A true detection level connotes the lowest concentration that can
be detected; The quantitation level connotes the lowest
concentration that can be quantified reliably^ This guidance
uses the quantitation level as a threshold for compliance with
water quality-based limits.
For purposes of this guidance, the use of the ML as the
compliance level has been established. EPA's position that the
ML is a valid scientific and regulatory concept is consistent
13
-------
with the "May 1990 Dioxin Permitting Strategy" and the "1991
TSD".
The ML is defined as the lowest concentration that gives
recognizable signals and an acceptable calibration point. It is
the equivalent concentration of the lowest calibration standard
analyzed by a specific analytical procedure, assuming that all
the method-specified sample weights, volumes, and processing
steps have been followed. MLs areanalyte and method specific
and are established during the development and validation of the
method. The 1600 Series EPA methods are currently the only
Agency methods that explicitly include MLs.
MDL Component:
An interim ML is needed when a promulgated ML is not
available for a pollutant. The following discussion explains why
the interim ML should be based on the MDL, and how it should be
derived.
The MDL is a well-established part of the foundation of the
NPDES program, as defined in 40 CFR Part 136. MDLs have
withstood court challenges and are generally accepted by the
regulated community. Attempting to use a level other than the
MDL as the basis for determining the quantitation level would be
resource intensive and would have extensive repercussions.
EPA's procedure for determining the MDL calls for making an
initial estimate of the detection limit using one of the
following: (1) the concentration value that corresponds to an
instrument signal/noise in the range of 2.5 to 5, (2) the
concentration equivalent to three times the standard deviation of
replicate instrumental measurements of the analyte in reagent
water, (3) that region of the standard curve where there is a
significant change in sensitivity, i.e., a break in the slope of
the standard curve, or (4) instrumental limitations. If the
initial MDL is suspected of being too low or too high, the MDL
procedure can be iterated to determine the reasonableness of the
initial value and to arrive at a final value.
In the region of the MDL, the absolute variance of
analytical measurements (measured as the standard deviation)
remains approximately constant. Above the ML, the absolute
variance increases with increasing concentration, whereas the
relative variance (measured as the relative standard deviation;
coefficient of variation) remains approximately constant (i.e.,
the variance is proportional to the concentration). Between the
MDL and the ML, there is a region of transition from constant Co
proportional variance as the concentration increases. Because
the variance is approximately constant in the region of the MDL,
changes in test concentrations in this region will result in a
constant MDL value. However, as the test concentration increases
14
-------
significantly above the MDL, a point will be reached at which the
calculated MDL will begin to rise. Because the ML is based on
the MDL, the ML procedure requires iteration of the spiking and
measurement processes until the calculated MDL is within a factor
of 5 of the spike level. This iteration assures that the MDL,
and therefore the ML, will not be overstated.
As the MDL is determined by the analyses of a series of
replicate samples, its applicability to single sample analysis is
open to question. The MDL may not take into accoiiht oias,
analyst proficiency, inter-laboratory variability, matrix
effects, and other considerations that affect analytical results.
Furthermore, because the statistical derivation of the MDL is
designed to minimize the occurrence of false negative results, it
has been suggested that using the MDL as the compliance threshold
may lead to unacceptable levels of false positive results, and
thus potential permit violations.
MDL values determined in a clean matrix such as reagent
water are useful indicators of a laboratory's capabilities in a
clean matrix, but may not necessarily be achievable in other
matrices, such as NPDES effluents. However, EPA's experience has
shown that effluents from well-designed and well-operated "best
available technology (BAT)" treatment systems behave much as
reagent water in the analytical process, primarily because the
concentrations of the analytes of interest and of interfering
substances are reduced to undetectable or very low levels.
Therefore, although there may be a few instances in which MDL
values in reagent water do not reflect MDL values in NPDES
effluents, the number of instances can be expected to be small.
The procedure for determining the MDL and ML in this guidance can
be applied to either reagent water, NPDES effluent, or other
matrices to arrive at an ML that reflects the detection and
quantitation levels that can be achieved under actual operating
conditions.
If the MDL is set as a compliance threshold, any
concentration above the MDL would be considered a violation when
the permit limit was below the MDL. The uncertainty of
measurements between zero and the ML, and the certainty of
measurements at or above the ML, makes the MDL unreliable for
compliance determinations. If the MDL does not offer the level
of certainty needed to make a compliance determination, the
question becomes, what level will offer a reasonable level of
certainty?
Relationship Between Promulgated ML and the MDL
First, it is important to look at the definition and the
nature of the Limit of Quantitations (LOQ) as defined by the
American Chemical Society (ACS). The LOQ, as adopted by ACS and
by the International Union of Pure and Applied Chemistry (IUPAC),
15
-------
is a statistically based level that has been defined as the
"level above which quantitative results may be obtained with a
specified degree of confidence1*. The ACS recommends an LOQ set
at 10 standard deviations above the average level of a well
characterized blank. These 10 standard deviations correspond to
an (estimated) quantitative uncertainty of ± 30 percent at a 99
percent confidence level.
In order to compare the ACS LOQ with the EPA ML, it is
important to first understand the nature of the EPA MDL. As
defined in Appendix of B to 40 CFR Part 136, the MDL is "the
minimum concentration of a substance that can be ...measured and
reported with 99 percent confidence that the analyte
concentration is greater than zero". The HDL is statistically
derived from data based on the analysis of at least seven
replicate samples containing a known concentration of each
analyte of interest. Specifically, the MDL is calculated by
multiplying the standard deviation of the seven replicates by
3.14. The 3.14 value is the multiplier used for seven replicates
according to the student's t-test.
The OW Office of Science and Technology (OST) currently uses
the MDL to calculate the ML. OST has recently refined this
approach by using a multiplier of 3.18 which reflects the 10
standard deviations afeove zero that is used by ACS and normalizes
the ML to reflect the difference produced by the 3.14 multiplier
used in MDL determinations. As a result, both the ML and the LOQ
are statistically derived values that correspond to an estimated
uncertainty of + 30 percent at 99 percent confidence. The
remaining difference between the ML and LOQ is that the ML ie
rounded to allow calibration at practical levels, as described
below, and because rounding simplifies entries and compliance
calculations on the DMR.
For the sake of being consistent with ACS guidelines for
reporting results beyond the calibration range of the instrument,
OST utilizes the ML as the lowest point on the calibration curve.
Because calibration of instruments at exact numbers such as 4.86
or 13.9 is not practical, OST rounds the calculated ML to the
nearest multiple of 1, 2, 5, 10, 20, 50, 100, 200, 500, etc.
Rounding is done because it simplifies instrument calibrations
and reduces the potential for error in the preparation of
calibration solutions.
The above approach is the basis for developing the
promulgated MLs and the same rationale was used to develop
interim MLs.
Developing an Interim ML
In the absence of promulgated MLs and the definition of the
MDL concept, EPA believes that interim ML values can be derived
16
-------
most effectively as a multiple of the existing MDL value for a
given analyte.
Thus, the interim ML should be developed by multiplying the
published MDL for the analyte from a specific analytical method
approved under Section 304(h) or previously approved for use by
the permitting authority by 3.18, and rounding the calculated ML
as explained above.
For metals, the soon-to-be-proposed MDLs will, be used to
calculate the MLs. When neither the ML nor the MDL is available,
3.18 times the best estimate of the detection level should be
used. When a range is given instead of a detection level, the
lower end of the range should be used.
EPA's rationale for selecting a factor of 3.18 is based on the
following:
• The MDL is defined as "the minimum concentration of a
substance that can be measured and reported with 99%
confidence that the analyte concentration is greater than
zero..." (40 CFR 136, Appendix B). The MDL is equal to 3.14
times the standard deviation of seven replicate
measurements.
• The American Chemical Society has defined the Limit of
Quantitation (LOQ) as the level at which a sample can be
reasonably measured (quantified) at 10 standard deviations
above the average blank measurement using graphical and
statistical techniques. Since the MDL is equal to 3.14
standard deviations about the replicate measurements,
dividing this into 10 provides a multiplier of 3.18 between
the MDL and the calculated ML.
• The calculated ML is then rounded to 1, 2, 5, 10, 20, 50,
etc. Rounding is consistent with the established scientific
and mathematical principle of rounding a number after all
calculations, not before.
• Rounding the result to the nearest whole number also makes
the preparation of calibration standards a straight forward
process, given that the standards are typically prepared by
serial dilution of a stock solution.
• Similarly, rounding simplifies reporting results, as
analytes that are not detected will be associated with a
whole number value.
17
-------
Proposed fJew Science for Deriving Detection and Quantitation
Levels
The American Chemical Society (ACS), American Standard
Testing Methods (ASTM), and EPA (Office Ground Water & Drinking
Water) are currently developing an alternative approach for
deriving detection and quantitation levels. The approach would
go beyond the existing LODs and LOQs and would work toward
reducing the false negatives and false positives around the
detection and quantitation levels. This effort i's in its
formative stages and has not yet been peered reviewed or
documented. Until an alternative approach is developed and
documented, our guidance is consistent with the ACS'a current
recommendation to use a multiplier of 3.18 times the HDL to
calculate a LOQ (which is equivalent to our interim ML). As
science changes in this area, our guidance should be reviewed anc
revised as needed.
711. COMPLIANCE MONITORING AHD ENFORCEMENT PRINCIPLES AMD
STRATEGY POR WATER QUALITY-BASED B77L0ENT LIMITATION BSLOW
ANALYTICAL METHOD DETECTION LEVELS
A. Enforcement Principles
This enforcement strategy is based on four principles:
• Principle I: Permittees are responsible tor attaining
and maintaining permit compliance and for the quality of
the monitoring data they submit.
• Principle II: Enforcement authorities will evaluate
self-monitoring data quality to ensure program
integrity.
• Principle III: Enforcement authorities will assess
compliance through discharger data reviews and confirm
compliance through inspections, audits, and other
Independent monitoring or review activities.
• Principle IV: Enforcement authorities will enforce
effluent limits, reporting and other requirements, and
compliance schedules to ensure the protection of water
quality.
B. Enforoeaent strategy
When a facility's WQBEL is set below detection, the facility
is in compliance with the limit when it has demonstrated that it
is capable of measuring concentrations at the reguired ML using
the designated analytical method (Principle I). To ensure the
enforceability of WQBELs when analytical values are below the ML,
and because the region below the ML is one of greater uncertainty
18
-------
than the region at or above the ML, EPA recommends that values
below the ML be represented as zero for determining maximums and
averages for. DMR submittal.
EPA recognizes that this approach is a compromise that
introduces some bias into the data, and does not ensure
compliance with WQBELs. There is an indirect relationship
between the ML and the compliance level. The WQBEL is used to
determine compliance; MLs are used as the analytical measure by
which a laboratory determines the concentration of an analyte in
a given sample. When the analysis of a sample indicates that the
analyte is present above some detection level, some quantitative
assessment must be made of the concentration in order for it to
be used for DMR reporting purposes. Once the quantifiable level
has been achieved, that value can be used tc report data on the
DMR. Consistent with previous EPA guidance, such as the March
15, 1989 Section 304(1) Guidance and the May 21, 1990 Dioxin
Strategy, this approach involves the use of a threshold value
(i.e., ML) for purposes of reporting DMR data for compliance
evaluations. EPA Region VI has been successfully using a similar
approach since 1992.
The major advantage to this approach is that, in situation*
of high uncertainty, it gives the permittee a small margin of
relief to avoid false positives leading to violations. The
purpose of this approach is to provide us with a temporary,
consistent way to report values below quantification levels until
better analytical technology is developed.
Based on this guidance, EPA sees no need to modify the PCS
database to accommodate data where analytical results are at less
than detectable levels. As with all self-monitoring data that
are submitted on DMRs, these data will also be entered into PCS.
Data shall be reported as positive numbers or as zero. Mo less
than (<) or minus (-) signs accompanying the data or alpha
characters (such as rtND" for non-detect) representing the data
are deemed necessary, based on this approach. Use of zero and
positive numbers will address the current problem of inconsistent
reporting of these data. For permits that require reporting a
maximum value, the number of analytical results exceeding the KL
and the maximum value of the reporting period will be entered.
For average values, the calculated average as it appears on the
DMR will b* entered.
Any reported DMR value showing a concentration above the
specified WQBEL is considered an effluent violation when the
stated ML or interim ML for the pollutant in the most sensitive
analytical method promulgated at 40 CFR Part 136 is at or below
the WQBEL. When the stated ML or interim ML for that method le
above the WQBEL, an effluent violation has occurred when the
concentration is equal to or greater than the ML.
19
-------
The ML is not to be misconstrued as the level by which
compliance is determined. In support of Enforcement Principle
II, the ML is used to ensure the sensitivity of the permittee's
analytical method. In general, any violation involving WQBELs
established at levels less than quantifiable are considered to be
of concern to the regulatory authority and should receive
thorough review. This is oecause the pollutants that are covered
under the description of "non-detectable" have been deemed to
have potential water quality impacts at such low concentrations
that detection alone is of concern. Appendices D'and E provide
examples of the monitoring, reporting, tracking, and the
enforcement process.
To ensure protection of water quality from pollutants of
concern, the permit writer should specify in the permit any
methods to be used by the permittee to demonstrate compliance
with WQBELs included in the permit. Clear permit requirements
are essential for effective compliance evaluation and enforcement
of WQBELs. Facilities that are not in compliance with their
permit requirements must be required to return to compliance
promptly. EPA's approach to enforcement (Enforcement Principles
III and IV) is set forth below:
• Effluent Violations: Though any violation that has been
deemed to cause a water quality impact may be placed on
the "Quarterly Non-compliance Report (QNCR)," the QNCR
regulations contained as an appendix to 40 CFR Part
123.46 dictate the following minimum:
The application of EPA's "Enforcement Management Syscea"
(EMS) guidelines treats these parameters according to
their group designation: Group I (Technical Review
Criteria (TRC) of 1.4 times the monthly average effluent
limit) and Group II (TRC of 1.2 times the monthly
average effluent limit). A reportable noncompliance
(RNC) status is achieved and the violation placed on the
QNCR if a facility has two exceedances of either 1.4
times the monthly average effluent limit or greater
(Group I pollutants) or 1.2 times the monthly average
effluent limit or greater (Group II pollutants) in 6
months, or four exceedances or more of any magnitude in
a 6-month period. These criteria apply to all Group I
and Group II pollutants, defined below.
just as all other QNCR-reportable permit effluent
violations are considered significant noncompliance
(SNC) and must receive timely and appropriate
enforcement activity, these limits are held to the stae
standards. In accordance with the EMS, when a WQBEL hae
been violated and the determination has been made that
the violation has the potential to cause a water quality
impact, enforcement action should be prompt. Initial
20
-------
Group I Pollutants — TRC =1.4
Biochemical Oxygen Demand
Chemical Oxygen Demand
Total Oxygen Demands
Total Organic carbon
other oxygen Demand
Total Suspended Solids (Residues)
Total Dissolved Solids (Residues)
Other Solids
Inorganic Phosphorus
Compounds
Inorganic Nitrogen Compounds
Other Nutrients
NBAS
NTA
Oil and Grease
Other Detergents or Algicides
Calcium
Chloride
Fluoride
Magnesium
Sodium
Potassium
Sulfur
Sulfate
Total Alkalinity
Total Hardness
Other Minerals
Aluminum
Cobalt
Iron
Vanadium
Group II Pollutants — TRC » 1.2
Other Metals Not Specifically Listed Under Group r
Cyanide
Total Residual Chlorine
All Organics Except Those Specifically Listed Under Group I
activity may be in the form of a 308 letter requesting
additional information about the violation. The formal
enforcement action that follows will be tailored to the
circumstances surrounding the facility to achieve the
most efficient return to compliance. The action may be
an administrative order (with or without penalties) or a
judicial action, depending on the severity and duration
of the violation. Either approach should require
compliance by a specified date met via a compliance
schedule requiring milestone achievement by certain
dates and progress reports at frequent intervals.
Enforcement may also encourage a facility to pursue
pollution prevention activities or, when appropriate, a
Toxicity Identification Evaluation (TIE)/Toxicity
21
-------
Reduction Evaluation (TRE) as long as this effort does
not delay compliance.
• Schedule Violations: Permit writers have no latitude to
grant extensions for compliance with WQBELs, unless
State regulations specify such and the permit writer
knows that the facility will need time to comply (e.g.,
where construction to provide additional treatment is
needed). Where State regulations do not allow for
compliance extensions, compliance schedules will need to
be set forth as part of enforcement actions (i.e.,
enforcement orders). Permit writers may, however,
establish compliance schedules in permits for conducting
TIEs/TREs and/or pollution prevention studies,
establishing best management practices, or other such
special conditions. Where such a schedule is
established, EPA recommends the use of milestone dates
and the submission of routine progress reports to
evaluate progress and minimize delays. The following
milestones are considered SNC when 90 or more days
overdue: start construction, end construction, and
attain compliance with permit. Slippage of these
milestones of 90 days or more must be reported on the
QNCR.
Violations of schedules of 90 days or more, or reporting
violations in which progress report due dates are missed
by 30 days or more, are signs that the facility may not
meet a final compliance date. These violations should
be addressed with formal enforcement actions to quickly
resolve the delay and retain the expected schedule.
• Reporting or other violations: Treatment of reporting
violations is consistent with existing EMS guidelines.
Submittal of data determined to be invalid (i.e.,
failing to meet the specified ML) is a reporting
violation equivalent to nonsubmittal of the data. The
facility must perform a second analysis and submit the
corrected data.
• Administrative Penalty Orders (APOs): APOs may be used
in conjunction with an Administrative Order for permit
violations. The CWA Civil Penalty Policy should be used
for determining which class of APO to pursue.
The EMS guidelines for addressing violations of pollutant
limits or excursions that have potential to cause a water quality
impact require an enforcement action (Administrative Order or
State equivalent mechanism, judicial action, etc. with a
compliance schedule and date of final compliance) to be taken.
22
-------
Noncompliance may be resolved (designated as having been
returned to compliance on the QNCR) when a facility has returned
to compliance from a particular violation or as a whole with the
requirement# of its permit for one full quarter. Time]y and
appropriate enforcement actions in accordance with the EMS are
required to prompt a facility to return to compliance. A failure
to take prompt enforcement action (within two quarters of a
facility appearing on the QNCR as being in SNC) will require the
facility to be placed on the Exceptions List.
The EPA-recommended boilerplate language is designed to
provide for the reporting of data on DMRs that are consistent and
that facilitate compliance evaluation, development of NPDES
permit limits, and preparation of studies and reports. The data
provided for compliance evaluation purposes will reflect
noncompliance only when noncompliance has indeed occurred. An
example of a compliance evaluation is shown in Appendix E.
VII. ADDITIONAL RECOMMENDATIONS
To improve the quality of analytical data, EPA recommends
the use of clean techniques and extraction/concentration
techniques to aid in the lowering of detection and quantitation
levels, and the use of data quality objectives to facilitate an
understanding of results required.
A. Clean Techniques
The term "clean techniques'* refers to practices used to
reduce contamination and interference during the accurate and
precise measurement of trace level pollutants. Data quality eey
be compromised due to contamination of samples during collection,
preparation, storage, and analysis. Clean techniques begin with
the field sampling procedures and continue through all sample
processing and analysis steps. Examples of clean techniques
include wearing latex or polyethylene gloves during all step# of
the sampling and sample preparation process, using acid-washed
plastic equipment for trace metals, and analyzing blanks.
Although the Agency has not yet issued protocols for clean
techniques, specifications for apparatus and materials, quality
control, and sample collection, preservation, and handling are
included for each analytical method in 40 CFR Part 136, Appendix
A, and these procedures, while not exactly analogous to the clean
techniques described in the literature for trace metals, are
minimum requirements for sampling and analysis. The Agency
believes that field and laboratory use of clean techniques say be
critical for lowering the detection and quantitation levels of
analytical methods to achieve or approach WQBELs.
EPA has included general, interim guidance on the use of
clean techniques as part of the Office of Water Policy and
Technical Guidance on Interpretation and Implementation of
23
-------
Aquatic Life Metals Criteria issued on October 1, 1993. EPA
plans to develop protocols on the use of clean techniques and is
coordinating this effort with the United States Geological
Survey. Draft protocols are expected in early 1994.
B. Sample Hrtraction/Concentration
Concentrating the sample by evaporation of water is a
technique that has been shown to lover detection ^Levels when
analyzing for specific chemicals. For example, volume reduction
during digestion has been shown to aid in the determination of
metals in the nanogram per liter range. Extracting samples and
concentrating the sample extract prior to instrumental analysis
is an established, universally accepted technique used to enhance
the range of detectability for organic compounds.
The following questions are commonly asked by the permitting
authority in reference to concentrating samples:
1) Are the techniques for concentrating samples allowed under 40
CFR Part 136? and
2) How easily are the techniques employed?
For metals analyses, concentrating samples can be easily
accomplished through the existing regulations under 40 CFR Part
13 6. References to concentration techniques in the regulations
exist for some of the analytical methods, but not all. Where a
method does not include procedures for concentrating samples, a
method modification would be required.
Concentrating samples is a proven, highly successful
technique that may be utilized to lower analytical laboratory
detection levels. Concentration of samples is straightforward;
however, problems sometimes occur prior to analysis, such as
matrix interferences, precipitation of some of the components, or
solidification. There are we11-documented solutions to each of
these problems, although solutions differ from facility to
facility, and amcng production processes using different
chemicals within a facility. Among the solutions is the use of
other instrumentation methods. For example, an ion
chromatographic technique may be substituted for the Technicon
analyzer for minerals/nutrients analysis.
In thft absence of methods in 40 CFR Part 136 that can
measure below the detection level, EPA recommends methodologies
for concentrating and extracting samples and the use of cleanup
or clean techniques. Such techniques allow the permittee to
measure as close as possible to the WQBELs, thereby meeting the
goal of this guidance.
24
-------
C. Data Quality Objectives
In support of the development of technology-based limits and
WQBELs and determining compliance with those limits, SPA
recommends the use of data quality objectives (DQOs).
The "DQO Process" is a series of planning steps designed to
ensure that the type, quantity, and quality of environmental data
used in decision making are appropriate for the intended
application. It is a planning tool that allows decision maxers
to define their data requirements and acceptable levels of
decision errors during planning, before they collect data.
EPA's Quality Assurance Management Staff recently issued an
interim final document entitled, Guidance for Planning for Data
Collection in Support of Environmental Decision Making Using the
Data Quality Objectives Process (EPA QA/G-4). This document
describes seven distinct steps that make up the DQO Process, the
value of such a process, when it should be used, and who should
participate.
EPA's Office of Wastewater Enforcement and Compliance also
issued guidance titled nNPDES Permit Writer's Guide on Data
Quality Objectives" (December 3, 1990).
VIII. ADDITIONAL PERMITTING AND KN70RCKMENT ISSUES
While EPA believes that the above guidance for permitting
and enforcing WQBELs that fall below detectable levels is
appropriate, there are several issues of concern.
A. Using Internal outfalls to Ensure Compliance with WQBBLa
Potential Issue; Often effluent limits are below detection
levels because of dilution following treatment of the
wastewater which contains only trace levels of pollutants
(e.g., non-contact cooling water). Where technology-based
limits are applied, setting limits at an internal outfall at
the end-of-process is common. However, since WQBELs are
applied at the end-of-pipe, an internal outfall appears to
be inappropriate.
Approach; EPA fully supports the use of internal outfalls
to determine compliance even where permit limits are water
quality-based. EPA believes that internal outfalls can b«
especially appropriate where a pollutant of concern is very
concentrated at an internal outfall and is co-mingled with
discharges from other outfalls prior to discharge, causing
dilution or matrix interference. Also, there are internal
and external censor devices that are easily installed on the
25
-------
treatment system pipes and that allow sampling to be
conducted before the wastewater is diluted.
Plow variability for internal waste streams is an issue
that can be accounted for by installing a flow meter that
allows flows to be quantified and eliminates the
"uncertainty" for nonlinear tlow.
Appendix F of this guidance provides an example of the
use of internal outfalls in determining compliance with
WQBELs. which, if applied at the final outfall, would be
belcw detectable levels. The example in Appendix F shows
that assuming no contribution of pollutants from the
dilution waste stream will demonstrate whether the
concentrated waste stream would result in a violation of the
WQBEL.
Rationale; Internal waste stream limits are used to account
for pollutants prior to dilution. Applying limits at an
internal outfall to determine compliance with WQBELs is
consistent with 40 CFR 122.45(h), which allows for the use
of an internal outfall where dilution or interference makes
detection of the pollutants impracticable at the otherwise
established monitoring point.
[This is consistent with Permitting Principles I and II. J
B. Driving the Technology to Achieve Lover Oeteotion Levels
Potential Issue; Providing permittees with an ML may result
in the use of the ML as an easily met performance level and
thus not drive technology to achieve lower detection levels.
Approach: The ML is a quantification level used in
reporting analytical results, not a compliance evaluation
level. Facilities are required to comply with the WQBEL.
This is demonstrated in Appendix E, which shows examples of
compliance evaluations.
EPA recommends that, in expectation of improved analytical
method*, permit writers include generic re-opener clauses
that are sufficient to provide permitting authorities the
means- to reopen, modify, and reissue the permit as
analytical methods are revised and/or promulgated. At a
minimum, improved analytical methods and quantification
levels should be incorporated in the permit at the time of
reissuance.
Rationale; EPA believes that WQBELs set below detection
levels, regardless of the use of MLs, will encourage
development of improved analytical methods and better
quality assurance/quality control techniques. In some
26
-------
cases, facilities will violate the WQBEL simply when
detecting a pollutant. Thus, any exceedance of the ML will
be a violation. Technology will be driven to lower
detection levels when facilities make efforts to quantify
pollutants below the ML in order to avoid exceedances of
WQBELa. This is also true when trying to negate a
demonstration of reasonable potential and thus eliminate the
need for a limit altogether.
[This is consistent with Permitting Principle III.]
C. Specifying Analytical Methods in HPDBS Permits
Potential Issue: Facilities may challenge EPA's authority
for requiring analyses by a specific method and calibration
at a specific concentration.
Approach: EPA believes it is appropriate and necessary to
include specific language for reporting compliance with the
HL, the quantification level used by the permittee, and the
use of the ML concentration as a calibration standard.
Rationale: EPA already limits the specific methods that nay
be employed to those proposed or promulgated under Section
304(h) or approved by the permitting authority in advance.
In the case of many pollutants, there is a limited set of
approved methods. Specificity ensures that the ML can be
measured and that the laboratory is conducting analyses by
the most sensitive method available. Pursuant to 40 CFR
122.43, permit writers may include additional conditions In
a permit that "provide for and assure compliance with all
applicable requirements of the CWA and regulations." Baaed
on this authority, permit writers should establish special
monitoring and reporting requirements pertaining to WQBELa
that are less than detectable levels.
[This is consistent with Permitting Principle II and
Enforcement Principle I.]
D. Dealing With Matrix Interferences
Potential Issue: Matrix interferences may, in some cases,
result: in laboratories being unable to meet the stated or
interim MLs.
Approach: In most cases, EPA believes that wastewater of
high quality will have vary little matrix interference. EPA
acknowledges that some facilities will have matrix
interference. Appendix B to 40 CFR Part 136 sets forth
procedures that facilities aust follow when establishing
discharge-specific MDLs. Procedures for reviewing
27
-------
discharge-specific MDL and ML submittals are provided in
Appendix B to this guidance.
Rationale: EPA acknowledges that some facilities may be
unable to meet the specified MLs due to matrix
interferences. However, EPA believes that it is important
to distinguish between instances when MLs are not achieved
due to poor laboratory technique and when matrix
interferences dc indeed occur. To make this determination,
guidelines and procedures under which facilities must
conduct studies and, if appropriate, develop discharge-
specific MLs must be set forth. Without such, facilities
are likely to develop less stringent MDLs for use as
quantification levels that have no technical basis.
[This is consistent with Permitting Principle II and
Enforcement Principle I.]
B. Providing Time Extensions for Meeting the Specified
Analytical Method and Achieving the ML
Potential Issue: Many facilities may be unable to meet the
ML specified in the permit, particularly where they have
laboratory equipment that performs analyses in accordance
with an alternate test procedure approved under 40 CFR Part
136.4.
Approach: EPA recognizes that facilities may experience
delays in meeting specified analytical requirements for
certain parameters until in-house capabilities are
established. However, it is recommended that permitting
authorities not provide compliance extensions for permittees
to gain the capability of measuring the ML using a specified
method. If permittees do not meet the required methods and
ML specified in the permit, schedules for meeting thea
should take place through enforcement action which may be
issued concurrently with the permit.
Rationale: Since analytical capabilities for conducting
analyses in accordance with specified methods are readily
available, the Agency feels that compliance schedules are
not necessary. Schedules, when set forth, should include
milestone dates to be fully enforceable and avoid ambiguity.
[This is consistent with Permitting Principles I and II, and
Enforcement Principles I and IV.]
7. Determining Whether Failure to Achieve MLs Is Caused by Peer
Lab Performance or other Factors
Potential Issue: While specifying monitoring and reporting
requirements in a permit is appropriate and should result In
23
-------
consistency in permitting as veil as compliance from the
majority of permittees, this will not assist EPA in
determining whether exceedances of the ML are actual
exceedances of the WQBEL or the result of other problems
such as poor laboratory technique or some other problem.
Approach: Permittees are responsible for knowing and
understanding the terms and conditions of their permit,
attaining and maintaining compliance, and ensuring the
quality of monitoring data collected and submitted, where
permittees experience problems, they must identify the cause
and develop a solution. EPA does, however, recommend use cf
oversight mechanisms to identify actual or potential
problems.
In accordance with established procedures, major permitted
facilities and minor facilities of concern will be inspected
at minimum frequency of once per year. EPA recommends that
inspections of permittees that discharge pollutants that arm
limited below the ML should be particularly thorough and
conducted by inspectors with analytical experience. All
permittees should comply with QA/QC procedures for sample
collection, storage, transport, analysis, instrument
maintenance, calibration, etc. The possibility of not
achieving the specified ML and thus being noncompliant as a
result of sample contamination or technician error is
greater at low pollutant concentrations. Sampling and
handling procedures and proper QA/QC are addressed in 4 0 cm
Part 136 for each method. A work group of the Environmental
Monitoring Management Council is addressing this aspect of
analytical testing in its Agency Standardization of wethod*.
EPA recommends that laboratory procedures and records be
evaluated during inspections and audits to ensure that the
data are generated under acceptable procedures and that
submitted self-monitoring data are reflective of the
underlying raw data.
Rationale: Inspections/audits are used to verify the
permittee laboratory's capability to perform analyses
necessary to reach the required ML. EPA maintains, however,
that requiring facilities to report the ML that they achieve
will assist EPA in identifying facilities that may not be
meeting the ML. These factors can then be used to
subsequently target facilities for in-depth performance
audit inspections. This will allow EPA to determine whether
problems are caused by poor laboratory technique or matrices
interference.
[This is consistent with Permitting Principle II and
Enforcement Principle III.]
29
-------
G. Addressing Potential Water Quality Problems That Kay Not Be
Addressed Where Analytical Data Cannot Be Quantified
Potential Issue; Despite the measures proposed by EPA as
part of this guidance to more accurately and consistently
quantify pollutant discharges, water quality problems still
may occur.
Approach! EPA acknowledges that measures intended to
demonstrate compliance with WQBKLs will not necessarily
ensure compliance with water quality. Thu3,» EPA raccamends
that permits that have pollutant limits for aquatic life
below analytical detection levels should include the
requirement to perform WET testing in the permit.
Additionally, EPA supports the use of other measures to
protect water quality including bioaccumulation or sediment
studies, pollution reduction/prevention plans, or other such
special requirements.
Any information provided to the permitting and enforcement
authority from EPA or State sources, citizens groups, etc.,
that indicate water quality problems may be used to initiate
a more thorough investigation of the facility. For example,
fish/shellfish tissue samples that demonstrate a
bioaccumulative effect of pollutants otherwise not detected
or a chemical mass balance that shows the release of large
quantities of pollutants in a discharge may be used to
initiate additional activities that will further quantify
pollutants. In response to noted problems, EPA may conduct
an inspection, issue a request for information pursuant to
Section 308 of the CWA, or require studies. Ultimately, the
permit may be reopened to include more stringent
requirements that result in water quality protection and/or
a compliance agreement may be negotiated that includes
pollution prevention/reduction goals.
Rationale: Permits must be protective of water quality.
[This is consistent with Permitting Principle I and
Enforcement Principle ill.]
VIII. 80MllftT
This guidance includes the five following basic elements:
• WQBELs are to be imposed as NPDES permit limits when
necessary to protect the designated uses of the
receiving waters, even if the WQBEL is less than the
analytical detection level.
30
-------
• The quantification level to be used in analyzing samples
and reporting the results is the ML. Guidance for
incorporating MLs into NPDES permits, developing MLs
where they are not currently available, and translating
analytical values belov the ML into OMR data for
compliance evaluations is provided in this guidance and
the accompanying appendices.
• The existing analytical data reporting and tracking
mechanisms (the OMR and PCS, respectively) will continue
to be used with no changes to their structures. This
document provides guidance and examples for reporting
analytical values that are at or below the ML. The role
of the ML in compliance evaluations is not direct;
rather, it is the threshold for recording the analytical
data, which will then be used to calculate and report
summary information on the DMR.
• Enforcement criteria are consistent with those set forth
in the EMS guidelines, and compliance evaluations
continue to be made directly between the DMR information
and the permit limit (which is the WQBEL in cases where
water quality is the most stringent limitation).
• Important permitting and compliance issues, such as
matrix interference, the use of internal outfalls in
permit writing, driving technology to achieve lower
detection levels, specifying analytical protocol in
NPDES permits, and the ability of laboratories to
achieve specified MLs, are also addressed in this
guidance.
31
-------
APPENDIX A
MILS, IDLa AMD OTHER ANALYTICAL QUANT I TAT ION
LEVELS Ft* EPA APPCOK0 ANALYTICAL NEfHCBS
-------
l»2F«l*fbon> (PMM. IM AnNiMi,
Cfi Hiiji nil '*
00
fij
«» W •* AnnrtewiT
OCMi im OvMta r r 1 T Ml hot#
OftAn'
ftO
iU
CCHO «03 «J AartK**'
GCM* 10* 'Vih^ Cni|i ¦.!< Ken Mta to
fioktttn by Cat 11^^
Ch««vMiv'
M
4 4
fti
COM «i« Tu^ka'
a
GCMC Oil 1ft f\a9^to Oiam CaafMd^
QCU* IIM THtgm Qrtmm Ct i»i 1 *
OMba*
fifMCMMC,' f\«9> ll«p
MnilJw
•0
44
SLfift
GCMO IA ImAM* Aotfi*
GCM9 IIA *MnMi Hurt Cniarti fev
OCMfiO^llll U>Tl4li
cutentMinNMUi
HJkl
fi-U
at
CCH» Kl TtfotM NiMlM-
0Qi»WNp^
0
OGMS oan& *ry«|^b 0i|«k rnfn—n.
10
on
• 0
OCMD «0I T>nii> MMIm'
OCftftft 114 ftftl llii"
ocrv m tu-m
DflWto*"
»
i.h iiytiiutwiiic
to
PCtt
1 •
OCttft •» 0* Aclfe*
OGM l«n Of«lD Cm I. f li w
Mpi Him*
»
GOMiailll Iiii ¦ « (iiiMk
r—»n mil, faUt ftmi i *k. Qm
taMfiMofofcofimio
10
i •
GCftU «2S iMMul* AAto*
GCM0 101ft *»«* C>M'n—' by
Iiii^i 01«m'
ft
OCMflfrllll iMlMyk^ CilMUM
Co«paunrt», HaoMOfe, Cm
fib i fifaPtOMftb"
t.l-dtaMMMthan*
fiX
0 01
J •
0CHS0 001 T
UMI M4 •» i»i af'
a
OCMS 0311b T i *ib 0«ra friM|iri
-------
iMttfttl-HlMMOIt-DIIMMDriCT MEW
ACHIEVABLE DETECTION LEVELS
(Continued)
OROAMZATtON |
40 CFH Part Itt Approval
i
OMi-CM*
USOS*
AfiTM4
MM*
r- •: r—a
rrrtfcd
MM
CM
4WU
•MM
EK
*•**
M
Ml
M
BCM t
M
001
M
OOMS l«M >Mi Oim* Cbbii »ii b* teapa
s
aOM»«1W iy«|MM <*9*to 1 ¦¦in Ml*.
M Amwrtk, to
M
Ui
• •
>
ooif^ini r>iwiin^.
rud BaBM^ Qa Q»—m
to
B.M
cchm> mi »» rhmflftan"
a
oau frim oro««o r—m r*,.
I«MI to Ovmmq9«mM«i
OMm*
1
l
j
10
BJ
II
iMp* OUk»a
•
QCHt^HI iMMMd Ufiifc
GMfW*. WanMMi. Qm
2-«MMOMfc(t «tl|l M(W
M-WI
10
202
*
oonanift iih t II I' |
*
6CMI 0 91 tft 7N»p^n f nm»n n *.
i
UCAC U3»;3 *l«* IMud hv Ltf^v Mi*m^
W**« N^gimai Wn#onafcn«i to Wm'
m
111
• »
-*«• «« - r ^
»
1
i iMHini aM mm ri ¦ m ii
i
UiV DAC I ml Mi»od to rbvv* to
*«• If Oaa 1
-------
2f®»-O»J»-O»-4O16-OIM«N0CllCT ^
ACHIEVABLE DETECTION LEVELS
(Continued!
ORGANIZATION
40 CFR P*t 136 Approve1
OASi-CtN*
U&OS*
ASTM*
MUUM
if
•w
ym
kMM
tot
(Mi
•MM
KM
teaU
—
10
1.14
u&
1 0
04
QCtC 111 *QMhM
MMBtt 1 TumII Winrnlnm'
ft
UCMI 0-3110 Iw/Mii 14 tiamftte
fnwto frllrtlil*
00M0 t«» *MaMi Qi»biIi Cii hii Ii kf
OCMO 410 «n* Mil'
ocro ooa immm'
ft
GCM0 O-011I Imdh •* (i»«U4
CaafoM^ I«M RaoMOk, Om
Opaa*«aa«to*
la:
\ M
ft*
44
00
ft
0CMft»»t10 'IiiiMi ¦ « UimMi
»0
III
ftJLA
OCMO Ml Imm Idl «« Mfe'
OCMO 10X0 *0mM* 0*
ft
OCMO O31I0 liti I4>
121
fcii
to
rxHSfi >D|
»
ocmo a>110
OCMIMfMiMito*
~.a iw a owm
KC
11*
01
OCffi OM "Hmi*
QCM0 IA ImIMO Aa*'
1
0CM0O0117 ~AsM bk«MM Cuiii>bmi *¦¦
to
irP
0
QQMOS1H ImMmfI tnwiMi
Otr«MM0«f4aAMw IpaonMVk'
taM»» Rim i*
«tfp wM"rlif'i
20
1 0
an
0CM0 oso *0mMu** «M asm*'
WnNfNMiMi-
OCMO 1020 ImMO 0i««4d Cnw»n »*i fey
tooMp* OOMtat"
ft
OCMO O>110 •Obiff*
r«wm II*. M Wiimii ¦ Gm
CMn«ig^lOi*lwi O^Btowte'
ao
GCM0 10a 'f\ 1 liMl Ck|«ito CnwywMi by
ft
OCMO 0-3110 iMMytd tuiouMi
CMpoMO, (M4 Oaumw**. 0«
CtaHMMv^tatAlw iiiiim i>te*
•0
94
u
OC»© 404 ftMMfc'
HI M4 I4J% ft«HI tl >1 ¦! If
t
GCMft O >11 f -JUM iitMM rwNaii^,
» I ¦ ». o«» i **<—
"
••
OJJ
«ft
MM t«n IMMW mmm m II 1 <1 III if
i
wut» Aim 'AM lUMM 1 —1 ¦ 11
i
oil o CiAtf 'Imi Vmm Ior nunofe n
fc* G«ft Vh#iad
A*
-------
V0l-O*Jl-OJ-*O»6-Oti\NONDfIECT WW
ACHIEVABLE DETECTION LEVELS
(Continued)
OROAM2ATION
40 CHINA 136 Approved1
OMt-CIH4
u&os*
AftTM4
W«-
m.
MU
B
•MM*
101
MU
MmM
DM.
1mm
iftotitaMl
Mi ll-ftiitaql ptMix
to
U
u
aawWTNiWw'r mm to
80M lltt *anM* Of«to Cin^u iw* kv
ft
QCMO OHIO *1 Oil id bmiih
CMfM*, 1«rf to
kuryt kaatyl fMMati
10
SJ1
I*
ft
OCMSOftlia 1i oil Id Ijrt iriill
CMfMOh 1«ai IhmA^ to
pMhrirta
10
fiJI
11
hni| i f»m*
ft
QCMO 0-91 If ImHOnd
Cm^mO, Tout AM*«Ak, to
1
I
w
90
U
octc o» *n< i lOh
CMfMll, 1M ¦¦ II ¦ ¦*». to
"* "J" '
10
ill
t 0
DOM ion 'ImM* m urti bf
mini Hi
ft
QCMl o-ini lairra n (ivmi
fi §1 <1. tmt* ¦<**. to
iumwiyi *MMa»
10
ftja
i •
aas «a TTitiim f«r
QCMO on ImMM* artf Aoi»*
OCMft lift IniMMi Oia* Cnwm iH hf
hH)i OftikA*
ft
QCMO OHIO 1 n i 01 id tmniili
CmpmO, 1«M ImmM. to
Mom M ¦«¦¦¦«¦
10
70
UU
QCMO on 111 nil 1* Act*"
OOtt 10n InMMl 0r»«* fi i, Hi *
ft
1
OCMO O-ftllO •QiiiOO ¦ « U» i Hill
mow ftltlt TiNiiniii *—m
MmiIm 0MU. f«4 ¦ r Hi**
banio lal yyrans
10
to
0 013
QCMO on -ftiuHh mm «w AoMa"
GCMO ion *0«wM (kf«0 rn|ii r 01 toy
ft
1
QCMOO-ftllO UnoMk
^new ft iJtL*a
*»0nn«>iw tflOt, lout
Wwiwai ikM* ciiflwiBy»m*
S.44nialkmMhiM
MntsIM ko>MtM|
QQft*
40
Qflii
GGMS on «* A«*'
ft
GCMftO-Mlft iilWM
1
a
1
m
II
UU
t
i tm • »m 1mim« I— iii mm
a->
-------
\0i-oa»-o»-4oi6-oi». «ocrtcT mw
ACHIEVABLE DETECTION LEVELS
(Continued)
OftOAMZATION
40 CfRPwt 1M Appfovtd1
EM8|»-C1M*
UgGfi*
ASTM*
MM
jwij
MM
is
*•
•mm
COL
—
COL
W-»
IhM
ufcr)—n»
to
u
841
atmwMiMA «**<*•
maw #10 TMra^f Arm* Hrtmrfen'.
ooms im iiMii Oi«*
ft
l
aan^ini imiii ¦« uimo
rin in i it, i*d an i inaii, to
tpmoMffc'
IflOW *SU» *NVW*S Aia»MQ
IPNAI. I«li
P»« aipwWn*
10
M
U
ocms in immmm mm*"
IMV> «*!¦"
ft
OCMS O-ftl 1ft 1 litti >4 UlnMi
ri MMI fc. 1o«4 Mini rail, Cm
¦Bimm
10
1 0
ut
0CMI til 1wi»* 1^ Mi Mi'
nfUVY tlfi Tymrtt* tomife ttrtraMtar.
ft
1
QCttftO-ftltl tiiiih ni tuiHrt
rfflOM Ollll TM|i>nii fcawa
H»4nBitaw mi. iMMi^ >V^
r»tein w im**
to
4 1
JL£Z1
GCMft Uft In nm M tail'
iflfflY llg Tttrnrtw ftm— Hiiwt Hn"
GCMfi fM *MhMI (k|M C—B in» *
ft
t
GCM »II1I *Mk«4 limrih
r n hi | n »iwnlMB IftttJ. I«M Akkw* Ate,
r»fiM 1*4 C»IMMpite'
mm il 1 Mi )iww
ift
1 t
UU
dfHI 4J% *l i — i m ±mm'
~
UUM4IUI fc «•«»***
A 6
-------
\I62FU>I-OS33-03-40I6-OI3)NOMOCTICT NEW
ACHIEVABLE DETECTION LEVELS
(ContiniMd)
OAOAMZA HON
40 CfR Part 130 Approved*
EM6L-CJN*
UfiOfi*
A6TM*
Poiuum
m
w»
hMtatf
(01
MU
•MM
IK
MU
ftMbwl
PVMM
10
14
UL
KUmiMMaMi m*Aam'
WlCUV <10 Tefrm*# «hm* IMscvtm'
ft
«
COM 02118 UvnuMi
rinn nik Will mill, Ca
tflCUV frtlll i*ibi JWomMb
m»iil i ¦ fMU. 1«4 honi^ Htg^
iU.XLMu.Ul.uti.
10
fctt
«.i
DM»*
9
0OM0 »t1l» •>!! ill rni HI nil.
I«M Oil i i Oh. to Cta>w»a»»i0i*lMi
1
0etc own *l«i MiOH tar U» MnhnidB
infcwm
10
•.0
U.
oatiw^i ii in*
GCM> 002 Aimto*
0CMB tin *V 1 m Ol»M riBgi.H h fcnwn
OliMl *
a
0CMS »I11I ft«rM (k»^D rnwpmiHia.
latf Oin Uto. to
a:
U1
i •
gcmm> ooi fuoMbb nimtoi'
s
com ami Tw^ti ik«MB cw^gm*
0GMIM Ti > MM*
so:
U
1 •
GCHfiD SOI MttMfem'
s
OCMt OSIIft (k0M finn»ii n^.
j
1
|
s
1
2 4~dtaMMiaftianQi
10
Ul
t-f
CgPOMft—fc-
oaam
i
QOfta »}1I7 *Ui ti>—on Cmpsui*.
M RmmM, to fliwiiii
I
sac OAtt *l«i Hi < .11 to mm* h
W«» *
an
fiJd.
u
oatt)oSl^^^fS£^:^8lt^aP^~
*
OCMiaim *ni|i oil 0r»«*> **-~n ni no.
Talrf " . to n»l«mnim»Mw
i t
u
in
nrrtffi mi Timi— -m-m*
2.4-*Nd*^wu|
10
JLU
11
GCH) 004 -flwoto-
0CM1 on "toafttaufr* Mi A«U»*
GCMI 1«» 'ImmMi lk|M» fc*
lnwpi Ota4
i
OOHM1U "A«M |i««lw rni»8iin».
TM 0i l l Ifc. to CkflMMV^Wto
GCMi 116 ImMm* a# Aofe*
ln^i OMbu*
B
GCMSO)111 -MlMi
C««pM^ foul Hilly ii Oil, to
••
ULL
• •
0
CCMOI1II li *li ¦« »
Si
911
m
u
'«* ¦ m i , , ¦ ».
A
-------
iMFUH-OMl-Ol-MIB-OIVNaNDCTtCT NEW
ACHIEVABLE DETECTION LEVELS
(Continued)
OHOAMZATIOM
40 CFRPmn 1M Approver
OML-CtN*
usos*
ASTM*
MuUM
m
wt
•MM
COL
IMM
KM.
fc»A|
MoHkmI
•ttlyfeMOM
M
M
U
ooaa im ortMti ci«»iiKi *
DMm*
>
GCMft frllll Oi»««fc> Cnipnii—.
fliuiw i>h'
to
«*
&u
OCMO «3ft Iiii^Ii Hb wtf AaM»*
ft
t
ocus o-aiio *o m ¦* Uvn^n
sars«u^Vii i ii i» jinw^n
>»*¦¦¦>¦ !¦ nttl. IM Oi ir II . *gfr-
pwim^w imi cum iBniir
4^ManpllM|l >*»mil
•Oar
10
u
«t
gchop on
QOil Ml 1m»h idfc «ni AMU*
OIAm'
ft
GCMS 0-3110 hiwO»
lii. loM Qui ¦ Hli. to
a
1»
Li
OCM Oil
(Ktfiffl.3aatofa wi&mL
OCMIIUA "fariNM ftfMfe rii«r » »V
ft
OCMI 03110 1mAm4 CilMOfc
ri Ij ||. ImIInmM, to
Mi II iMiiliKHHl mtm
w
u
1.1
GCHfiO oil
QCM 01% m+ Aom'
>i OMm1
6
GCU0 0-3110 till mi »«
IM to
Ma 12-oMwoathaivt
matfwM
to
u
k.1
GCHfiO 011
POM oa *»—mmm+ m+ Acttv*
OGMfl l«A 'Im*M f ¦ wpw n* to
0
OCMI 0-3110 *0iii^ »* fillip
rnw>n«^i. M to
BHtkylMM oMmM*
to
Bit
U
IKWD tD1
3
oatnnfini
IMdlll bWbiMi
tui
SJt
GOtOD 001 ¦¦ ¦*» ftftfiflm"
3
iM^KtBHwOto. (Ub Onafcf^KalMwi
1
1
a:
U1
3
QCM 0-311ft 'napiM OiO0* fn—niiiiifc
f»in»n i i«|p%
n
Lll
KfifiQ MITWomm rf«K«tont"
•0
ua
«>
nn>0 AM Ti||i ifci'
a
CCMI»)11ft Tkf^k (k(ra rnw»wfc,
••WNM*'
1
GCiC 030/3 *I«»J yiM (aw
w*«N H*pr«M Mmlani n Wtm'
11
«
1
UUC 0M'3 'l«M Urtnl ** low
Ka^m n^iyminii n
A-a
-------
vl&tfVOt-OMl-Ol-Mlft-OIJWONDClfCT NIW
ACHIEVABLE DETECTION LEVELS
Continuftdl
OftOAMZATlON
40 CF ft Port 1M A|*N9V«f
fMSi-ON1
usas*
A6TM4
PoAuim*
P
I0L
MU
cot
MU
cWwiwWuniww—Wmw
u:
tm
0CH» ]11ft OiMta Cii—wwOk
t
aac o»n "i«i »i> i or nn
0CM0 IM TtffNH*'
M
flCfC til XlttMrf
ft
OCHOO-1UO tMMMtri trt-Ofc
OCkMOl IhMniO
KM mB *MnM Oh— CiMKh *v
10
fiifl
0
OCMI»IU0 lmni) iU Ummn+m
1 10i*A
sac D3000 "Imi MiOi* tv Or««MMita>
iHfhMM
10
»i
U
OCFO m H>w 1 ¦ iQni hgftow1
GCMO in -IMKMM Wtf AM*
0CU0 1010 InM* b*
ft
QOHIftllll bt mail
Cm^omOk, loM RmmOIb. G«
10
11
10
GCMB l» MiAflO*.
ft
1
QCM* 0-1110 linUli >« Cii«NOk
mew ftiiti tamwm
nil mi, imi ii i ^
UwW n>—nmmtm'
iMmin
to
so
11
0M> COO •* Mmm*
ft
ocMOOiuo in mo x
n i, pIi. l-*im ¦ Oil. Qm
l-nhnptanol
10
&•
SSSm?Ull"iO Mri AoMi*
1
4-
-------
i\lfi2FUI-i m*
1 10 «o/l
cac 03000 M«m IM«i to
h» II i mW«m*
u
KM •« 1 I On 1* «ntf A4M*
"""
ow
0.014
nnrn, ,,m,-
00ft
aac 0-31M mra
(
*
i
1L21
0.000
to
OflWW ^iiTtiTV^' rn*****9
anririn
o.w
00
041
cac »IIM *>>
to
(w»»wMb» ipmlrti
u:
ajiu
tt
cac m orMiKNBihi rMMR «« rar
0CM» Ub IwUfctO ar* AM*
0 01
GGKC O3104 'lk|M(t*k* m*
m* Otoil I, 0« CMtmrnm m*mm
1 IOi^I
cac DDK Mitwl to IViwrtiiw
0 01'
POM
1 H»«»A
Gac DX81 -Iwt to Tkonn^i 1 1
M*«NC
tail
Uttl
« I
1 to m%f\
U3i* OK/<4 *twi tout to ftpnHnB
rMvri»> * Wiv'
ul.
~ ••
vr.rzrr::—
« Id f^i
uCA* DOM 'taw *w+o4 to (k«mM«ra
kM
• •
1 li «%/!
L.U . UJOM liitrt ha
-------
EM62rvOI-OM3-0>-«OI»-OI3WMNOCTECT NEW
ACHIEVABLE DETECTION LEVELS
(Continued)
ORGANIZATION
40 CFR Part 1M Approved1
BMSL-CIN*
usos*
ASTM*
PMUUM
i»
ff
MiM
UM.
«0U
KOt
Ml
•MM
r»iMi
IC
MKWtmnwMifrwWiirtrai'
041
0CIC O-S104 'OH—1O1I11 m*
te OrMMfefvM'
1
oac MtM *T«t IMM4 te
Uplw M fCM te WMr'
KB-I1M
mi
I1X»*
a
a
octc m -oiMMHMrta MM xmr
OAt
QCCC »I1M 'OimMi
1
oac MU< *T«m MaM lv
UpteMjte rCM te W«te*
Kfrllll
0.01
QCCC 0-S1M md
¦rfDMNi. toOnnafef^MB*
9
oac OMI 1m Mi>¦11 m r ¦ 1 «* rca»-
0CM0 M *»¦!!» AM'
001
oac a SIM
ObhMl 0m
1
oac 03Ut Mm Uwim* te f»mi imu
Kpterih #CM te MM1
rc»iMo
•4
a*
oac m '0t»w<»N ri ¦ 11 a* rcfe-
GCMO Ml c*l teste*
001
oac 09IM 'OnwJ^w
1
sac D)t)( -imm Ml—< te N^iinfci — <
Mte^te fCtei te W*»*
V.
U1
001
oac o-jkm "QiwiioiIii m*
wmtmm+mt, toQNM>f«Mi'
B0-M00
m
oac 0RN *Tm Mm 1 *¦ Hitmmmmtm
(TCDOt
sab:
sjbi
acn« «1» •»»! rmir
•
OCMOO-Stt* *ln ili.irf
ri mi 1 ^ i«miMMk te
L^dr'""" *WWV"d *" MMU *ro®m,a 40 «*'•« 1M feu ottMt (Mkadt UtM arm not NPDES ippnwid. bill My b« u««d
1 A '
1*1 *¦
a o ol 40 CM F«rt 13fl. Umtetftwtf *0U «nd ML«
A
-------
ACHIEVABLE DETECTION LEVELS-ABBREVIATIONS'
ORGANIZATION: The custodial organization responsible for the method cooes are:
ASTM American Society for Testing Materials
EMSL-CIN EPA Environmental Monitoring and Support Laboratory in Cincinnati, Ohio
USGS U.S. Geological Survey Techniques of Water Resources Investigations
METHOD APPARATUS: As derived from the METHOD. The following codes are used:
CVAA Cold vapor atomic absorption spectrometry
FIAA Flame atomic absorption spectrometry
GCEC Gas chromatography with electron capture detector
GCF1D Gas chromatography with flame ionization detector
GCHRMS Gas chromatography with high resolution mass spectrometry
GCHSD Gas chromatography with halogen-specific detector (HaU, O.I.,
microcoulometnc, electrolytic conductivity)
GCMS Gas chromatography/mass spectrometry
GCPID Gas chromatography with nitrogen-photoionization detector
HPLC High performance liquid chromatography
HPLCUV HPLC with an ultra-violet detector
METHOD TITLES: Where a method number is assigned in the table, the full title of the
method is given here.
801 "Purgeable Halocarbons," EPA EMSL-Cin
602 "Purgeable Aromaocs." EPA EMSL-Cin
603 "Acrolein and Acrykxutnle.* EPA EMSL-Cin
604 "Determination of Pentachlorophenoi Salt in Wastewater." EPA EMSL 0»
605 "Benzidines," EPA EMSL-Cin
606 "Phthalate Ester." EPA EMSL-Cin
607 -Nitrosoamines." EPA EMSL-Cin
608 "Organochlorme Pesnodes and PCBs," EPA EMSL-Cin
609 "Nitroaromatica and laoehorone," EPA EMSL-Cin
610 "Polynuciear Aromatic Hydrocarbons." EPA EMSL-Cin
*u
-------
E\l62Ftf>t-0«)*-O*-«OI6-OI>\MC»U)CTECT NEW
ACHIEVABLE DETECTION LEVELS
(Continued)
S.Mathoda may b« found h "Mathodi lof Aiutyih of Inotginlc SubtUiMM In W>l« and Fluvial Stdknmti" U.6. D«fX. ol IM Interior. U.S. Geological Survay, Opwi fik
Rafxart IMtS.
4.Annual Book ol ASTM Slwdw* Vokmmm 11.02. ASTM. PA. 1991. PuMeatlon Cod* Number 0111029110.
BJntaffcn Mia m Jwttd kf mm wpwimrt m 1.11 Um*» Uva low mi MOL irtnlwMi usktg an EPA tppravid mtdwd.
A-
-------
ACHIEVABLE DETECTION LEVELS-ABBREVIATIONS
(Continued)
611
612
613
624
625
1613
1624
1625
02580
03vow
03371
D3534
03973
METHOD SUFFIX
0-3104
0-3113
0-3 Vt 5
0-3117
"Haloether," EPA EMSL-Cin
"Chlorinated Hydrocarbons,.' EPA EMSL-Cin
"2,3,7,8-Tetrachlorodibenzo-p-Oioxin," EPA EMSL-Cin
"Purgeables," EPA EMSL-Cin
* Determination of Pentachlorophenoi Salt in Wastewater," EPA EMSLCm
*Tatra- through Octa-Chlorinated Dioxins and Furans by Isotope Dilution,*
EPA OWRS. ITO
"Volatile Organic Compounds by Isotope Dilution GCMS,* EPA OWRS. ITD
'Semivolatile Organic Compounds by Isotope Dilution GCMS/ EPA
OWRS. ITO
"Test Method for Phenols in Water by Gas-Liquid Chromatography,*
Annual Book of ASTM Standards, Section 11, ASTM
'Test Method for Organichlorine Pesticides in Water," Annual Book of
ASTM Standards, Section 11, ASTM
'Test Methods for Nitriies in Aqueous Solution by Gas-Liquid
Chromatography," Annual Book of ASTM Standards, Section 11. ASTU
"Test Method for Pofychlorinated Biphenyls (PCBs) in Water," Annual
Book of ASTM Standards, Section 11, ASTM
"Test Method for Low-Molecular Weight Halogenated Hydrocarbons
Water," Annual Book of ASTM Standards, Section 11, ASTM
The suffix to the METHOD. The suffix is specific to the sample fraction,
matrix, and level. Suffixes are defined as follows:
"Organochlorine and Organophosphorus Compounds, Total Recoverable
and Dissolved, Gas Chromatographic." Methods for the Determination o*
Organic Substances m Water and Fluvial Sediments, United Stated
Geological Survey (USGSI
"Polynuclear Aromaoc Hydrocarbons (PNA), Total Recoverable. High-
Performance Liquid Chromatographic," Methods for the Determination of
Organic Substances « Water and Fluvial Sediments, USGS
"Purgeable Organic Compounds, Total Recoverable, Gas
Chromatographtc/Mass Soectrometric,' Purge and Trap, Methods for the
Determination of Organic Substances in Water and Fluvial Sediments.
USGS
'Acid Extraction Compounda, Total Recoverable, Gss
Chromatographic/Maee Soectrometric,* Methods for the Determination o*
Organic Substances « Water and Fluvial Sediments, USGS
t-u
-------
ACHIEVABLE DETECTION LEVELS-ABBREVIATIONS
(Continued)
"Base/Neutral Extractable Compounds, Total Recoverable, Gas
Chromatographic/Mass Spectrometry," Methods for the Determination of
Organic Substances in Water and Fluvial Sediments, USGS
a-is
-------
ACHIEVABLE DETECTION LEVELS-ABBREVIATIONS
(Continued!
METHOD SUFFIX The suffix to the METHOD. The suffix is specific to the sample fraction
matrix, and level.
Suffixes are defined aa follows:
Suffix
Frac
Matrjx
Laval
AW
Acid
Water
6N
Base/neutral
BNW
Base/neutral
Water
CHS
Combined
Solid
High
W
Water
The following codes are used:
EDL Estimated detection limit
MDL Method detection limit [49 £fl 43234 (Appendix Bl)
ML Minimum Level - used in EPA OWRS Industrial Technology Division
programs; definition of the minimum level that must give recognizee*
mass spectra and acceptable calibration points [see footnote 2 on TaM-2
of Method 1624, Revision B (49 £B 43234)].
A-14
-------
APPENDIX B
GUIDANCE FOR PERMIT WRITERS AND THE PERMITTEE
ON THE DEVELOPMENT AND REVIEW OF
DISCHARGE-SPECIFIC METHOD DETECTION LIMITS
-------
March 7 ,1994
CHAPTER 1 — WHY ARB DISCHARGE-SPECIFIC KDLS NECESSARY?
1.1 UNDERSTANDING BPA'8 STRATEGY
The continuing implementation of EPA'a National Pollutant
Discharge Elimination System (NPDES) program has resulted in the
development of water quality-based effluent limitations (WQBELs)
which are below analytical detection levels. EPA's October 20,
1992 document, National Strategy on the Permitting, Monitoring,
and Enforcement of Water Quality-Based Effluent Limitations Set
Below Analytical Detection Levels ("the strategy") addresses the
issuance of permits that have limits set below the detection
levels of current analytical chemistry technology. EPA believes
that the implementation of this strategy will result, to the
greatest extent possible, in the protection of water quality.
As part of this strategy, EPA has specified that the value
in the NPDES permit should be expressed as the calculated WQBELs
regardless of the ability to measure to the level of the WQBELs.
Permit limits that are expressed as calculated WQBELs are
consistent with the requirements found in 40 Code of Federal
Regulations (CFR) 122.44(d) which indicate that limits be
protective of water quality standards.
The purpose of this guidance is to clarify the provisions in
40 CFR Part 136, Appendix B, for developing site-specific Method
Detection Limits (MDLs). Specifically, this guidance addresses
those circumstances when the analytical methods do not provide
permittees with the capability to measure to levels as low as the
WQBEL when determining compliance. For example, a permittee's
analytical laboratory may be able to achieve an MDL of only 1.0
mg/L for the analyt* of interest in the permittee's discharge due
to interferences; however, the MDL provided in an analytical
method in 40 CFR Part 136 may be 0.02 mg/L, and the monthly
average and daily maximum WQBELs say be 0.0005 mg/L and 0.001
mg/L, respectively. In this situation, compliance cannot be
Permit Writer'* Guide to Discharge-Specific MDL Ivaluation
Bl-l
-------
March 7,1994
assured because the analyses provided by the laboratory do not
reach the level of the WQBELs. Simply detecting to a level of
1.0 mg/L, or even to the MDL of 0.02 mg/L does not demonstrate
compliance with the WQBELs. EPA recognizes that in some cases,
permittee laboratories may fail to reach low, but achievable,
quantitation levels. Other times, even when quantifying to the
lowest levels possible, permittee laboratories may be unable to
reach the level of the effluent limitations.
To address this problem, EPA's strategy specifies that in
cases where limits are below quantifiable levels, an interim
Minimum Level (ML) should be established. The ML is a term that
originated in the EPA 1600 Series methods, and is defined as the
concentration in a sample that is equivalent to the concentration
of the lowest calibration standard analyzed by a specific
analytical procedure, assuming that all the method-specified
sample weights, volumes, and processing steps have been followed.
Because the majority of EPA methods for wastewater analysis do
not contain method-specified MLs, EPA has developed an approach
for calculating an interim ML for use in permit compliance. That
approach often requires the calculation of a discharge-specific
MDL.
The interim ML is designed to be used as a quantitation
level that facilities must meet in analyzing samples and
reporting results. The ML is not to be confused with the basis
for determining permit compliance, which is the WQBEL. The ML's
role in compliance evaluations is not a direct relationship;
rather, it la the threshold for recording the analytical data
which will then be summarized and reported on Discharge
Monitoring Reports (OMRs). The values reported on the DMRs will
be used for determining compliance.
Bl-2
Permit Writer'¦ Cuide to Di*eh*rg«-Sp«cific MDL Ivtluation
-------
March 7,1994
1.2 UHDERSTANDINQ DIBCHARQH-SPBCI7IC MDLs
The MDL is defined in 40 CFR Part 135, Appendix E, as "the
minimus concentration of an analyte that can be measured and
reported vith 99% confidence that the analyte concentration is
greater than zero as determined by a specific laboratory method"
%
(40 CFR Part 136). As specified in 40 CFR Part 136, the MDL may
be determined in reagent (blank) water because reagent water is
available to all laboratories. The MDL also may be determined in
the wastewater of interest, provided that the analyte
concentration is adjusted to be within 1-5 times the estimated
detection limit.
In this strategy, EPA has established a procedure for
estimating the interim ML of an analyte from a discharge-specific
MDL. Those procedures are described in Chapter 3 of this
guidance.
EPA acknowledges that in a wastewater discharge matrix,
permittees may not be able to accurately quantify to the level of
the MDL determined in reagent water on all occasions; therefore.
EPA addressed this concern through the use of a conversion factor
when developing the compliance threshold. This multiplier Is
3.18 times the MDL. The calculated ML is then rounded to the
nearest multiple of 1, 2, 5, 10, 20, 50, 100, 200, 500, etc.. as
provided in Section IV of this guidance. When following
appropriate analytical procedures and incorporating appropriate
quality assurance and quality control practices, most permittees
should be able to attain the interim MLs, contained in EPA's
strategy, in their discharge-specific matrix.
In some cases however, laboratories may be unaoia to ee«t
the EPA-specified MDLs, due to matrix interferences. In other
cases, inappropriate laboratory techniques and poor adherence to
laboratory and quality control procedures will result in
permittees failing to meet EPA-specified MDLs. EPA believee tnat
Permit Writar'a Guida to Dlachar9a-Sp«cific MDL Evaluation
-------
March 7,1994
in cases where true matrix interference can be demonstrated, a
discharge-specific MDL is justified as the basis for establishing
a site-specific ML. EPA does not believe that laboratory error
or the use of an inappropriate analytical method should be used
as a basis for seeking discharge-specific MDLs. Therefore, it is
important to distinguish the cause of the failure to achieve EPA-
specified MDLs so that the strategy will work as planned.
Matrix interferences occur when constituents in a sample
other than those being measured cause deviations of the
analytical test results. Matrix interferences generally occur as
a result of three situations: (l) a substance is similar to the
analyte being measured and cannot be distinguished from the
analyte, thus producing a higher measured value; (2) a substance
reacts with the analyte being measured and produces a lower
measured value; and (3) a substance reacts with chemical reagents
used in the analysis, and causes a higher or lower measured
value.
The evaluation of potential matrix interferences require*
both modest investments in analytical testing and data evaluation
by the permittee, and a slight increase in the level of oversight
performed by the permitting authority. Thus, the determination
that a discharge-specific MDL needs to be established should be
based on technical goals consistent with the EPA strategy. This
guide is designed to help permit writers in making this
determination by providing guidelines and procedures under which
facilities must conduct studies and, if appropriate, develop
discharge-specific MDLs. EPA believes that such guidelinee will
help meet the goals of the strategy, while precluding the
development of discharge-specific MDLs without technical
justification.
81-4
Permit Writer's Cuid* to Oiach«rg«-Sp*cific MDL Iv«lu«ttsn
-------
March 7,1994
CHAPTER 2 — HOW DO PERMITTEES DEMONSTRATE THAT
DISCHARGE-SPECIFIC MDLs ARB WARRANTED?
Prior to proceeding with the discharge-specific MDL
development process, the need for discharge-specific MDLa muse be
demonstrated. Through this demonstration process, the permitting
authority can identify those facilities that warrant discharge-
specific MDLs due to matrix interferences. In this process, it
is the permittee's responsibility to request discharge-specific
MDLs and demonstrate that such an MDL is warranted. It is the
permit writer's responsibility to evaluate the adequacy of
permittee submittals and determine whether discharge-specific
MDLs should be granted based on the submitted information.
CPA provided flexibility for dealing with matrix
interferences when it promulgated the analytical methods for the
measurement of pollutants in 40 CFR Part 136 (49 £B 43234). The
major flexibility options are found in the Preamble, including a
mechanism for obtaining approval of an alternate test procedure
on a nationwide or discharge-specific basis (40 CFR Parts 136.4
and 136.5). The alternate test procedure process was intended to
encourage the development of new analytical methods by instrument
manufacturers, permittees, and individual laboratories. However,
such new methods may not be used for compliance monitoring
purposes without completing the approval process.
The allowance for flexibility in the promulgated methods it
intended to give analysts a number of options for resolving
analytical problems unique to specific waste waters. The
Preamble to the publication of the 600 and 1600 Series Methods in
40 CFR Part 136 on October 26, 1984 contains a lengthy discussion
of the need for and allowances for flexibility in these methods.
The Response to Comments section also includes a discussion that
reads, in part:
"Analysts may also usm their discretion In selecting cleanup
procedures. EPA has also relaxed the strict protocol tor
Permit Writer's Guide to Diacharge-SpecLfIc MDL Evaluation
>2-1
-------
March 1,1994
sample extract concentration . . . Accordingly, within the
scope of the GC, HPLC, and GC/MS test procedures, analyses
are permitted sone discretion in selection of concentration
techniques.¦
In addition, the 600 and 1600 Series Kethods for organic
pollutants contain a statement to the effect that the analyst is
permitted to modify the method to "improve separations or lower
the costs of analyses" provided that the results are not less
precise or less accurate than the results obtained using the
unmodified method. As a result, analysts may use their
discretion in selecting packed or open tubular columns, operating
temperatures, carrier gas or solvent flow rates, and type of
detector, as well as cleanup procedures and extract concentration
procedures, within guidelines outlined in each method and in the
Preamble to the regulation.
EPA recognizes that there may be a few intractable saaple
matrices that do not yield to extensive analytical efforts. In
this case, the analyst must document for EPA the steps taken by
the analyst, the solutions found, and the instances in which «
given matrix did not yield to known analytical techniques. EPA
does not wish to learn that "the sample couldn't be analyzed" and
that an analyst made no attempt to reduce the matrix
interferences. EPA will view the lack of an attempt to overcome
matrix interferences as in direct conflict with the spirit of
flexibility permitted by the regulation and the methods.
EPA has set forth guidelines by which permittees may request
discharge-specific MDLs, therefore, permit writers can evaluate
discharge-specific HDL requests in a consistent and thorough
manner. The suggested roles of the permittee and permit writer
in this process are discussed below. To provide more information
to both the permittee and the permit writer in determining how to
demonstrate that discharge-specific MDLs are appropriate, axaeple
solutions to matrix interference problems, and case histories of
industry claims of matrix interferences have been provided. The
B2-2
Panait Writer's Gulda to Discharge-Specific HDL Evaluation
-------
March 7,1994
examples are taken directly from Guidance on Evaluation,
Resolution, and Documentation of Analytical Problems Associated
vith Compliance Monitoring, June 1993 {EPA 821-3-93-001)
(Monitoring Guidance). Copies of this document may be obtained
from the Office of Water, Engineering and Analysis Division,
Office of Science and Technology, 401 M St., SW, Washington, DC
20450.
2.1 PERMITTEE'S ROLB
To demonstrate that discharge-specific MDLs are warranted,
each permittee must:
Demonstrate that the permittee's laboratory is capable
of achieving the EPA-specifled HDL in reagent (blank)
water? that is, the permittee should test spiked
reagent water to illustrate that the EPA-specifled KDL
can be met by the analytical laboratory;
Demonstrate that a reasonable attempt to resolve the
matrix interference has been made using the approaches
suggested in the Monitoring Guidance; and
Demonstrate that the EPA-specifled MDL cannot be
achieved in the permittee's effluent.
These demonstrations require that the permittee's role be
primarily one of analytical testing, because qualitative
assessments of laboratory capabilities and matrix problems are
not sufficient. Thus, an analytical testing program must be
implemented by the permittee and must consist of the following:
Testing of spiked reagent water to illustrate that the
EPA-specified MDL can be met;
Testing of effluent discharge to demonstrate that
matrix interferences occur and that the EPA-specifled
MDL for reagent water cannot be met in the effluent;
and
Testing of the approaches suggested in the MonitorIng
Guidance, and testing of other analytical approaches to
demonstrate that matrix interferences cannot be
overcome.
Parmit Writer'¦ Ouida to Diacharga-Spaclfie KDL Evaluation
•J-)
-------
March 7,1994
For each analyte for which discharge-specific HDLs are
requested, permittees should conduct tasting of samples of spiked
reagent water and samples of effluent discharge. The permittee
may not assume that matrix interferences for one analyte ara
necessarily an interference for another analyte. The number of
samples and replicates that are required for this demonstration
may differ based on the permittee's situation.
In many cases, the permittee may employ a commercial
laboratory for its compliance monitoring analyses. In these
instances, the role of the permittee is to: 1) establish the
testing program outlined above, 2) have the testing program
implemented by the laboratory, 3) collect all the relevant
information from the laboratory, and 4) use it to demonstrate to
EPA that.a discharge-specific MDL is warranted. When a
commercial laboratory is employed for such a testing program, it
is imperative that the same laboratory that was used to
demonstrate that the EPA-specified MDL can be achieved in raaqent
water also be used to demonstrate that the EPA-specified HDL
cannot be achieved in the permittee's effluent. Ideally, both
demonstrations also should involve the same analytical
instrumentation and personnel. EPA recognizes that this may not
always be practical in a commercial laboratory setting. However.
EPA expects that where this is not practical, the instruments and
personnel used for both demonstrations will have equivalent
capabilities. For instance, it would not be appropriate to use a
more sensitive instrument for the reagent water testing and a
less sensitive instrument for the effluent testing.
Ultimately, the permittee lust demonstrate to the permit
writer's satisfaction that a discharge-specific MDL is warranted;
also, this demonstration must be supported by specific analytical
data. The following checklist of laboratory data, which is
required to support a claim that a permittee has been unable to
measure pollutants due to matrix interferences, was taken froe
B2-4
P«rait Writer's Guide to Dischtrga-Spacifie MDL
-------
March 7,1994
the Monitoring Guidance; however, this checklist, shown in
Exhibit 2-1/ has been modified to also include data related to
the analysis of metals.
2.2 PERMIT WRITER1S ROLE
The permit writer's role in the process of demonstrating
that a discharge-specific MDL is warranted can be summarized into
three specific functions:
Recognizing situations in which discharge-specific mol
development may be necessary;
Specifying prerequisite requirements and time frames
for permittees to submit demonstrations of the need for
discharge-specific MDLs; and
Evaluating the adequacy and accuracy of permittee
demonstrations of the need for discharge-specific MDLs.
A discussion of these roles follows.
At some point in the permitting process, the permit writer
may recognize that reported analytical data do not achieve the
EPA-specified MDLs. This may occur prior to drafting the permit
as a result of reviewing DMR or other data. It is also possible
that the permit writer may be well into the permit development
stage prior to noting that, based on data such as that submitted
in a permit application form, the interim MLs specified in EFA'a
strategy will not be met. A permittee's inability to meet the
EPA-specified interim MLs may be the result of one of two
situations: (1) the laboratory simply is not measuring to the
lowest level of sensitivity, namely the EPA-specified interis ML;
or (2) matrix interferences or laboratory error result in the
laboratory not measuring to the EPA-specified interim ML. Permit
writers have few options outside of requiring a permittee to both
re-analyze for parameters of concern, and then resubmit data that
meets the EPA-specified interim ML (to distinguish between these
two situations). Because parameter-specific analyses may be
Parmit Writer'* Guide to Discharga-SpcclfIc MDL KvaluatLon
-------
March 7,1994
EXHIBIT 2-1t CHECKLIST OF LABORATORY DATA TO DEMONSTRATE
MATRIX INTERFERENCE
Type of Laboratory Data
Kxamplee
Method nunber of the base method
used for the measurement.
* Method number from 40 crx Part 136
• Changes made to the BPA method
should be described, including
supporting data that demonstrates
equivalent performance
Detailed narrative discussing the
problem* with the analysis,
corrective actions taken, and the
changes made to the base method
identified above.
• Permittee must require analytical
chemist to thoroughly document
problems encountered and solutions
attempted
Summary level report or data
reporting forms giving the
pollutants for which analyses were
conducted and the concentratione
detected, for the pollutants that
were not detected, the detection
limits or estimated detection
limits, must be provided.
• Results should be provided for
each field sample analysed,
Including all dilutions and
reanalyses
• The meane for estimating detection
limits must be provided
• If flags are used in the
laboratory report, the definitLone
must be provided with the data
Summary of all quality control
results required by the base method.
Including but not limited to the
examples shown.
• Instrument tuning
• Calibration
• Calibration verification
• Initial preciaion tfidTrecovery
• Ongoing precision and recovery
• QC check standard
• Matrix spike, matrix spike
duplicate results
• Surrogate recoveries
• Labeled compound recoveries for
isotope dilution methode
• Blank results
• Serial dilutions
• Method of standard additions, if
applied
• Interference check standards
• Quality control charts and liaies
B2-6
Permit Writer's Guide to Discharge-Specific MDL
ion
-------
March 7,1994
EXHIBIT 2-It CHECKLIST 07 LABORATORY DATA TO DEM0N8TRAT8
XATRIX INTERFERENCE (continued)
Type of Laboratory Data
Examples
Raw data that allow an independent
reviewer to validate each
determination and calculation
performed by the laboratory,
including but not limited to the
examples ahown.
• Sample miabera or other
identifiers
• Extraction/digestion dates
• Analysis dates and times
• Sequence of analyses and run logs
• Sample volume
• Extract volume prior to and after
each cleanup step
• Final extract volume prior tc
injection
• Digestion volume
• Titration volume
• Percent solids or percent moisture
¦ Dilution data
• Instruments and operating
conditions
• GC/MS, GC, AA, and ZCP operating
conditions
• Chromatograms, ion current
profiles, bar graph spectra,
library search results
• Quantitation reports, data system
outputs, and other data used to
link raw data to rasults reported
• AA and ZCP peak profiles and raw
data
« Direct instrument readouts
• Laboratory bench shsats, logbook
pages
Example calculations that allow the
data raviawtr to determine how tha
laboratory usad tha raw data to
arrive at tha final rasults.
• Examples of both detected and
nondetected compounds
• Show all adjustments for sample
volume, dry weight, etc.
For CC/K3 and othar instruments
involving data systems, tha
permittee should ba prepared to
submit raw data on nagnatic disk
upon request by IPJfc.
Names, titles, addresses, and
talaphona numbers of tha analysts
who performed tha analyses, ind tha
quality assuranca officer who will
verify the analysea.
Summary of steps taken to overcome
matrix interferencea
• Referencee to sections in the
Honitoring Guidance
• Details of other approaches used
to overcome interferences
Permit Writer's Guide to Discharge-Specific KOL Evaluation
-------
March 7,1994
costly and oftan point to the need to develop discharge-specific MDLs, Lt nay
be more reasonable for the permitting authority to require the permittee to
conduct analytical testing as part of discharge-specific KDL demonstrations
(discussed below).
Ultimately, the review of theee demonstrations may result in additional
burdens being placad on permittees and permitting staff. Thus, whan it
becomes avidant that discharge-specific MDLs may be needed, lt la to the
permitting authority's advantage to have a well-planned program in place that
stipulates the permittee's rola in the process of demonstrating the need for a
discharge-specific MDL. Por example/ establishing prerequisite standard
proceduras for permittses in a short guide may help to initially explain the
demonstration process, and therefore, may help to avoid a deficient submittal
that could result in resource axpandituras by both parmittaas during revisions
and permit writers during reavaluation. Another idea is to require the
submittal of data in a simple format that can be easily reviewed. Soma
additional considerations to help ansuro tha completeness and accuracy of
submittals include prerequisite requirements for the following!
The simultaneous submittal of all necessary data,
A statement indicating that tha laboratory in which analyses for
this demonstration package ware performed is tha same as the
laboratory that will ba used in subsequent testing;
A statement indicating that tha samples used during the
demonstration process are representativa of tha discharge;
A statement indicating that all data that were generated as part
of this demonstration are being submitted;
Specification of tha laboratory method(s) used; and
Reference to the sections of the Monitoring Guidance that were
applied in attempts to overcoma matrix interference problems.
As previously indicated, the permittea's responsibilities will primarily
entail analytical testing. Permit writers ehould consider prescribing tha
desired amount of analytical testing to help avoid deficient submittals. Pour
approaches suitable for different analytical testing requirements are
described ia Kxhibit 2-2.
B2-8
Permit Writer's Guide to Discharge-Specific MDL
-------
March 7,1994
EXHIBIT 2-21 DECISION MAXIMO PROCKS* FOB B8QUIKIHO TBSTIKO
I Analytical Requirement
Variables
Perform a single analysis each of
one aliquot of reagent water and one
aliquot of effluent discharge
• Laboratory performance not
questionable
Matrix interference very |
likely
Perform analyses of several
replicates of reagant water and a
single sample of effluent discharge
Laboratory performance
questionable
Matrix interference very
likely
Perform analyses of a single sample
of reagent water and several samples
of effluent discharge
Laboratory performance not
questionable
• Matrix interference
questionable
Perform analyses of several
replicates of reagent water and
eeveral aliquots each of the
different samples of effluent
discharge
Laboratory performance
questionable
• Matrix interference
questionable
The permit writer can determine if a potential matrix interference
exists by comparing the HDL determined in reagent water to tha"HDL detsreirtsd
in the permittee'* discharge. If the MDL in the discharge ia greater than
twice the MDL in reagent water, then a potential matrix interference exist a.
In comparing the MDL* determined in reagent water and in the permittee •
discharge, the permit writer should evaluate data that reflect the quality of
the laboratory as well as the MDLs currently achieved by the permittee.
Information available in making an assessment of this type can be obtained
through review oft
Inspection reports* inspection reports such as compliance
evaluation inspections and performance audit inspections
¦ay point to laboratory performance problems. PAIS also may
provide overviews of procedural problems with individual
parameters which may Include matrix interference problems or ia
some eases, may show that an analytical method less sensitive
nseded is currently being used. Information on quality assort***
and quality control (QA/QC) measures also may point to perform****
problems.
DOB OA program daCat The DMB QA program was established to
periodically evaluate the quality of data submitted on DMA* tr
requiring laboratories to perform analysee on samplee containing
unknown quantities of analyses. At least annually, these
samples are sent to permittees. The resulting OMR QA proqra* u«•
Permit Writer's Culde to Discharge-Specific MDL evaluation
». *
-------
March 7,1994
may provid* trends showing problems in the Analyses for particular
parameters.
DMM uxd ptait application form rfatsi This data indicates
pexaittae laboratory performance. Fast OMR data that are at
concentrations abova tha SPA-specified interim ML may hava bean
questioned by tha enforcement or permitting staff. Ae a result,
matrix interference concarna may hava bean previously addressed.
Data from tha DKR and application form nay show trends where
particular parameters are abova IPX-specified interim MLs while
others are not. This may point to matrix interference.
Laboratory curt ill cation progruta > Many states hava labontory
certification programs that require that all analyaaa for
parameters of concern be certified. These programe vary widely.
Soma may involve laboratory inspection programs and more frequent
OMR QA requirements, while others do not. However, some
information may be gained from records of these programs.
If the permittee has employed a number of laboratories for tasting in
the recent past, these sort of historical records may be less valuable.
Permit writers may find it useful to consult with personnel from the state
laboratory or tha KPA Regional laboratory who may have experience with the
analysis of the permittee's effluent or with the methods in question.
Once tha permittee has completed analytical testing in accordance with
the permit writer's instructions, the request to develop a discharge-specific
HDL must be submitted for review and approval. At this point, the permit
writer must review these requests for completeness and accuracy. Here, cne
permit writer's primary responsibility is ensuring that matrix interferences
are tha true cause of a laboratory's inability to reach the quantifiable
level. This evaluation requires a certain level of expertise in laboratory
data evaluation because it involves a thorough rsview of analytical data
sheets. Again, consultations with state and BFA Regional laboratory personnel
may be useful.
BFA highly encourages permit writsrs to be proactive in providing
permitteee with support and guidelines in the conduct of this demonstration
process. Chapter 3 of this document provides a discussion on how the
discharge-specific MLs are developed. Permit writers should consider
eetablishingproceduree in which demonstrations are approved prior to
permittees being authorised to proceed with discharge-specific MDL
development. Ultimately! such proceduree will help ainiaite resource burdene
on both permit writers and permittee*.
3.3 BXAMPLBS OF SOLUTIONS TO KAXaXS PBOeLSKS
Some example solutions to matrix problems are described below. Thee*
examples are taken from ffuirfance on fvajuacion, Rmsolution, and Document41 im
82-10
Permit Writer's Guide to Discharge-Specific HDL Evaluation
-------
March 7,1994
of Analytical Problaaa Associated with Compliance Monitoring (Monitoring
Guidance). Other solutions Addressing specific parameters and using
laboratory techniques of interest, may be found by conducting literature
searches. The examples provided here are intended to help the analyst
understand how to develop solutions to matrix problems by utilising the
flexibility provided by the CPA methods. Any modifications made to BPA
methods within the scope of the allowed flexibility oust be tmafd by
performing the initial demonstration of capabilities (the "(tart up" tests)
and by meeting the performance specifications described in the 40 CFR Par*-. 136
methodo prior to using the modified method for compliance monitoring.
The issue of increased laboratory costs for resolving matrix
interference problems is also addressed in the document cited above. Some
estimated incremental costs associated with cleanup techniques and other
approaches are provided, ranging from no increased costs (associated with the
use of GC with selective detector in place of GC/MS), to as much as $500 per
sample (associated with the use of isotope dilution GC/MS in Methods 1624 and
1625. However, costs may vary, depending on the laboratory, the number of
sampleo, laboratory capacity, and other factore.
Examples of solutions to matrix Interferences in samplee containing
volatile organic pollutantss
Ob* ot selective OC detectors. The CWA 304(h) methods for volatllas
include SPA methods 601, 602, 603, 624, and 1624. When effluent lialts
are in the 10 nq/h or greater range, the selective GC detectors allow
detection levele well below those achieved by mass spectrometers. The
specificity provided by electrolytic conductivity detector and by the
photoionization detector allow detection of the halogenated and aromatic
analytes, respectively, in complex matrices.
Conceiitraced aaaplaat micro-extraction and gaa chromatography with
selective detector*. The selective GC detectors in Methods 601 and 602
provide sensitivity that is 10 to 100 times greater than that required
to detect the analytes of interest. Some of this sensitivity can be
used to substitute micro-extraction in place of purge-and-trap. The
advantage of micro-extraction is that the pH of the water can be
adjusted to attempt to keep the interferencee in solution while the
analytes of Interest are extracted.
Samplm dilation. Methods 601 and 602 can achieve method detection
limits of less than 1 pg/L for many volatile analytes. The added
sensitivity of selective GC detectors can be used to overcome matrix
problems by diluting the sample by a factor of 10 to 100. Bven with
such dilutions, the pollutante can be detected at the levels required,
and the effecte of Interferencee can be reduced or eliminated.
Isotope dilution. Method 1624 employs stable lsotopically labeled
analoge of the pollutante ae internal standards in the analyeis. The
use of these labeled compounde frequently permits the pollutant to be
determined in the presence of Interferences. This Is due to the unique
spectrum of the labeled compound which can be located in the presence of
these interferences. The pollutant's spectrum, can then be located by
reference to the labeled compound.
Permit Writer's Guide to Discharge-Specific MDL evaluation
• 3-11
-------
March 7,1994
Sxaaples of solutions to matrix Interferences In samples containing
semivolatile organic pollutants includei
Oae of selective OC detectors. SPA methods 604 through 612 employ gas
chromatography with selective detectors and high performance liquid
chromatography with an ultraviolat (UV) datactor to dataet pollutants in
tha presence of intarfarancaa. As with volatile*, tha addad sensitivity
of tha salactiva detectors permits tha sample to be dilutad by a factor
of 10 to 100 while allowing dataction of tha analytaa at tha afflusnt
Limits spacifiad in most parmits.
Use of pa cbangm, A vary powerful aeans of saparating tha pollurants of
intarast from interferences is to adjust tha pB of tha sample to keep
tha intarfarancas in solution whila allowing tha pollutants to ba
axtractad in an organic solvant. For example, nautral pollutants can ba
•xtractad at aithar low or high pH.
del permeation (alnm exclusionf chromatography* Thia technique is
dascribad in Ravision C of Method 1625. Tha sama tachniqua is usad In
tha Suparfund Contract Laboratory Program (CLP) mathoda and SW-846
methods, and has baan shown to ba affactiva for removing lipids and high
molacular waight intarfarancaa that can dagrada GC and mass ipactroMtar
parformanca.
<7am of adsorbents florlsll, alumina, and silica gel. Thaaa adsorbanca
ara affactiva in saparating nautral spaciaa from polar intarfarancaa.
For polar analytas of intarast, tha adsorbant must ba avaluatad to
determine if tha analyta will ba racovarad. Tha laval of activation of
tha adsorbant playa a major rola in thia racovary process.
Isotope dilution. Mathod 1625 parmits determination of pollutants la
tha prasanca of intarfarancaa in saaivolatila saaplaa in tha same v«y
dascribad for volatilaa. In addition, tha rangaa of racovariaa of (ha
1abalad analogs parmittad in tha mathod allow for quantitation o( tha
pollutant whan intarfarancaa reduce tha afficiancy of tha extraction
If any of thasa tachniquaa ara appliad to tha analysis of wasta water
saaplas undar tha allowanca for flaxibility in 40 CPU Part 136 mathoda, tfteir
usa must ba supportad by performing tha initial daaonstration of capabilities
using thasa modifications. Thia demonstration must take placa before the
modified method is appliad to tha analysis of saaplaa, and must meet the
performance spacificationa of tha original method.
2.4 CASB BXROKXXS Of CLAIMS OP KAX1XX IHTB*FOUWC*«
Tha case histories presented below were adapted from the Guidance oe
Evaluation, Resolution, and Documentation of Analytical Problaas Associated
with Compliance Monitoring (Monitoring Guidance). These case historiee
demonstrate cases in which the permittees and/or the contract laboratoriee
were using incorrect analytical mathoda, did not follow tha required
procedurea in 40 CPR Part 136, did not submit data neceaaary to document that
the methods were being followed, ox did not submit documentation regarding the
nature of interferences and the attempts (if any) to resolve these
interferences. These ease historias are included here to provide examples of
B2-12
Permit Writsr's Guide to Discharge-Specific MDL Evaluation
-------
March 7,1994
the type of response that permit writers nay receive froo dischargers who
claim that they are unable to meet permit limits due to interferences. The
identities of the dischargers and the analytical laboratories arc not
disclosed.
Cam* Bistory 1 - Th* informatloa mimittmd by thm mamlyticml laboratory
reveaietf iaeenMiatmuelmm with thm stated analytical mmtbodm. The discharger
allowed the laboratory to use either methods alternate to 40 CFR Part 136, or
ta use modifications to methods 624 and 625, Alternate methods are allowed
under 40 CFR Parts 136.4 and 136.5, provided that the facility submits the
alternate methods to EPA's Environmental Monitoring and Support Laboratory m
Cincinnati, Ohio (EMSL-Ci) for approval. Otherwise, alternate methods are hoc
allowed. No reference to alternate msthods approved by BMSL-Ci was found.
If methods 624 and 625 were modified as allowed under 40 era Put ;J6,
these modifications were not documented and equivalence wae not demonstrated.
Modifications that were made by the laboratory to these methods includedi
Combining acid and base/neutral fractions;
Using a fused silica capillary column for the analysis; of acid
and base/neutral fractions;
Employing alternate internal standard*;
Employing alternate surrogates;
Achieving higher detection limits;
Using fewer matrix spike compounds; and
Using matrix spike amounts inconsistent with regulatory
compliance, background, or method-specified levels.
EPA recognizes that the use of multiple internal standards and a fuse*
capillary column for the base/neutral/acid fraction represent improvementsi
however it does not believe that combining fractions, higher detection luute.
alternate matrix spike compounds, and matrix spike amounts which are
inconsistent with background or regulatory compliance levels represent
improvements. On the contrary, these changes degrade method performance, aetf
are therefore in violation of both the spirit and letter of the flexibility
permitted la the 600/1600 Series 40 CPR Part 136 organic methode.
The detection limits reported for the semivolatiles were, for the meet
part, twice the HLe given in method 1625 and were approximately 10-20 tLmee
the KOLs given in Method 625. Mo explanation for the increased detection
levele was given, nor could the limits be derived in any meaningful fasiuoe
from the data provided. The laboratory aade no attempt to clean up the
¦amplee using pif change, gel permeation chromatography, or the other
techniques in the 600/1600 Series aethode or the draft Monitoring Guidance
that had been provided to the permittee.
Permit Writer's Guide to Discharge-Specific HDL Evaluation
aj-1)
-------
March 7,1994
Cut almtorj 2 - Tbm permittee provided inmuflicimat information tor «
rfetaiietf reviev of tbm *n&lytic*l results. The discharger in this cat#
history subadtted samples to a contract laboratory for analyses by a CC/KS
method which failed to produce useful results. The discharger and/or the
laboratory attributed tha problems to larga concantrations of acatona in tha
discharge. However, thia problem could not be confined from the information
provided. The discharger's analytical laboratory proposed an approach to
overcome the interference problems. This solution was to employ Method* *01
and 602 for the volatiles analysis. Because these methods are both .i-cra
sensitive and more selective than a GC/MS method, the analytes regulated
should be measurable in the presence of a large concentration of acetone. Tha
diiichargar chose to ignore the laboratory's proposal and submitted a claim o?
matrix interference. SPA believee that the approach suggested by the
laboratory is workable and appropriate, and should have been attempted.
Furthermore, the documentation of the laboratory results should include the
material listed in Sxhlbit 2-1, Checklist of Laboratory Data.
Case Blatory 3 - Tbm permittee autmittmd latter* and reports from
amvmrtl contract .laboratories, vitA only soma of tbm (fata apmclfimd la txhlblt
2-1, G&ecJtliat of Laboratory Oata. Data items that were submitted included
instrument tunes, run chronologies, chromatogram, calibration data,
calibration verification data, results for blanks, quantitation reports for
samples, and matrix spike data runs against the QC limits for Methods 624 *nd
625. Missing were the initial praciaion and recovery (XPR) data that
demonstrate method equivalence.
The seaivolatile matrix spike data were puzzling. The results of the
unspiked samples indicated that some of the acids and base/neutrals were not
detected, yet the results for the spiked samples showed large concentrations
of some of these analytes that were not spiked into the samples. The
volatiles matrix spike had been diluted by a factor of 200 and spiked after
dilution. Diluting and spiking will not show matrix interferencea, and tftwa
these data are of no value in evaluating the undiluted sample results.
Case IIstory 4 - Tbm permittee submitted an—ry reports from tbm
analytical literatory. None of the materials submitted contained the
information required, as listed in Ixhiblt 2-1, Checklist of Laboratory 9«t«.
and none of the materials contained explanations of the nature of the
interferences found or descriptions of attempts to overcome the interferences.
Undoubtedly, permit writers have or will encounter situations not
described in these case histories. However, the similarity among these
82-14
Permit Writer's Cuide to Oischarge-Specific MDL
-------
March 7,1994
that will likely apply to other situations is the lack of documentation
provided by the permittees whan making claims of matrix interferences or other
analytical difficulties. The information necessary to evaluate such clams is
outlined in the checklist in Exhibit 2-1, and includes the information that is
necessary to daoonstrate that a matrix interference is present and to
demonstrate the performance of any method modifications made under the
allowance for flexibility in 40 CPU Part 136.
Permit Writer's Guide to Discharge-Specific HDL Kvaluation
•i-H
-------
March 7,1994
CHAPTK* 3 - HOW utl DlSCHJUtOB-SPSCirXC MDLS DEVELOPED?
One* the permittee haa deoonstratad to the permit writer's aatiafactlon
that a discharge-specific KDL is warranted, the process of davaloping
discharga-apecific MDLa bagina. In this procaaa, it is tha permittee'a
raaponaibility to davalop a technically-baaed diacharga-apacifie MDL and uaa
this to determine an appropriate interim KL. As in tha dempnatration of
whether the peraittee warrante a discharge-opecific KDL, the permit writer'a
role is one of oversight. The perisit writer oust evaluate the adequacy of the
permittee auboittale and determine whether or not the discharge-specific MDL
developed by the permittee ia appropriate and ahould be granted. Information
contained in this document doee not differ from that given in 40 CPR Part 136,
and ia intended to provide guidance on tha applicability of tha regulation to
the development of diacharge-apacific MDLa when matrix interferencea ara a
problem.
3.1 THS PERMITTEE'¦ ROLE
EPA has aet forth a procedure for the development of discharge-apeciflc
MDLa in Appendix B to 40 CFR Part 136. This procedure can be used with a wide
variety of aaaple types ranging from reagent water containing an analyta to
wastewater containing an analyte. However, in the development of a diacharge-
specific MDL, the matrix must be waatewater. The MDL procedure was designed
for applicability to a broad variety of physical and chemical-sethode, and aa
auch, the procedure ia device- and inatrument-independent. However, the
calculated KDL value is essentially a "anapehot" of laboratory capability at
the time that the MDL was determined. Therefore, it is imperative that
diacharge-apecific MDLs be developed ia »*ch laboratory in which the
permittee's samples will be analysed for the aaslytes in question. Given tha
coat of developing discharge-specific MDLs, it may be to the peraittee'a
advantage to limit the number of laboratories that are employed for routine
analyses.
Permitteee who wiah to develop diacharge-apecific MDLs muse follow the
procedures set forth in Appendix 8 to 40 CFR Part 136. A aunoary of the
diacharga-apaciflc MDL development procaaa ia provided belowi
Stmp It Make an initial eatimate of the detection limit;
Stmp 2t Prepare a laboratory standard in the discharge-specific
matrix that containa an acceptable concentration of the analyte;
Stmp 3t Take a minimum of seven aliquots of the discharge aaaple
and process these sample aliquots through the entire analytical
method, recording the maaauraaants;
Permit Writer's Guide to Discharge-Specific MDL (valuation
SJ- 1
-------
March 7,1994
8tmp 4t Calculate tha variance and the standard deviation of the
replicate MaiurMMnti;
Stmp St Calculate tha KDL; and
Stmp St Verify tha reasonableness of tha estimated datactlon
limit and tha subsequent KDL determination.
Step 1 involves making an initial aatimata of tha dataction limit uaing
tha parmittaa'a bast judgaaant. Ona of tha four following techniques sat
forth in Appendix B to 40 CFR Part 136 muat ba uaad whan formulating an
initial aatimatat
Determine tha eoncantration valua that eorraaponds to an
instrument signal/noiaa in tha ranga of 2.5 to 5;
Datarmina tha eoncantration aquivalant to thraa timaa tha standard
daviation of raplicata instrumental measurements of tha analyta in
raagant (blank) water/
Locata tha ragion of tha standard curva whara thara is a
significant changa in sansitivity (i.e., a braak in tha slop* of
tha atandard curva); and
Utilize tha instrumantal limitations.
In Stmp 2, a wastawatar discharga sampla should b« obtainad and analytad
to datarmina if tha analyta is in tha recommended ranga of 1 to 5 timas tha
aatimatad dataction limit. If so, it can ba usad for davaloping an HDL. If
not, and tha maaaurad laval of analyta ia laaa than thia recesaended ranqev a
known amount of analyta should ba addad to bring tha laval of analyta to tha
recommended ranga of batwaan ona and fiva timaa tha astimatad dataction limit.
On tha othar hand, if tha maasurad laval of analyta is graatar than fiva tiaaa
tha astimatad dataction limit, thara ara two optiona. Tha first option is to
obtain, if poasibla, anothar waatawatar discharga sampla with a lowar laval of
analyta. Tha othar option ia to usa tha sampla as is, aa long aa tha analyta
laval doaa not axcaad tan timas tha IPA-specified KDL. Whan tha maasurad
laval of analyta ia batwaan 5 and 10 timaa tha aatimatad dataction laval, it
ia important to raalisa that tha resulting discharge-specific KDL may not be
truly reflective of lower analyta eoncantrationa. Thia is because the
variance of th» analytical method is a function of concentration.
Stmp J involves analysing a minimum of aaven aliquots of tha wastewater
discharge sample. This is the core of the discharge-specific KDL development
process. Strict adherence to procedures and well-kept records are essential.
If a blank measurement is required by the analytical method to calculate
the measured level of analyte (e.g., for a spectrophotOMtric method involving
a color change), then a separate blank is analyted for each sample aliquot.
B3-2
Permit Writer'a Guide to Discharge-Specific KDL Evaluation
-------
March 7,1994
Each blank measurement should b« subtracted from the respective sample
uaiurnunt to obtain the final result. Theee blank-subtracted results are
than used to calculate tha MDL. (Notes This blank subtraction only applies
to those 40 C7R Part 136 methods that explicitly require it. It does not
apply to those methods that require tha preparation of a method blank
concurrent with tha sample preparation (i.e., it does not apply to the 600 or
1600 Series EPA methods))*
Before performing all seven analyses, it may be economically *.i
n
k t
n-1
and
S - (S*)w
MDL ¦ ,.€ . ^ (S)
Permit Writer's Guide to Discharge-Specific MDL Evaluation
¦ i-i
-------
March 7,1994
where tfruwm " the studenta' t value appropriate for a 99% confidence level
and n-1 degreaa cif freeduoi.
Stap 6 la the itaratlva procadura in 40 CPR Part 136 that muat ba uaad
to varify tha reasonableneae of tha KOL. Zt la alao tha atap that ia moat
oftan overlooked. Tha calculated MDL valua ia compared to tha eetia&ted
dataction limit datarainad in Stap 1. If tha aplking laval uaad was not
within a factor of five of tha calculated MDL, than the calculated MDL value
ia not valid. The reaaon for thia test ia that while the abaolute variability
in an analytical meaeurement increaaea with increaaing concentration, the
ralative variability embodied in the atandard daviaeion dacreaaaa with
increaaing concentration. Thus, samplae apiked at the very high concentration
nay raault in a vary low MDL valua that ia not rapreaantative of tha
capabilitiaa of the analytical procedure at the lower liait of ita
aenaitivity.
In addition, the procedurea in Appendix B 40 CFR Part 136 contain
inatructiona for an optional iterative procedure to evaluate the reasonable of
multiple MDL deterninationa. Given the coat concarne that exiat in discharge-
¦pacific MDL deterninationa, thia optional procedure ia not diacuaaed here.
Once the permittee haa developed a diacharge-apecific MDL for each
analyte, thia MDL ia tranalated into a calculated interia ML by multiplying-
the diacharge-apecific MDL by a factor of 3.IB. The calculated interia KL it
rounded to 0.1, 0.2, 0.S, 1, 2, 0.S, 1, 2, 5, 10, 20, SO, etc. to produce the
final interim ML. Rounding aimpllfiee both the calibration of analytical
instruments and the reporting of reaulta on the DMR.
There are no praciae rulee to apply to the lavel of effort that must be
expended in thia procedure. However, it ie the permittee's responsibility to
gather the appropriate data that will substantiate the discharge-apecifle CI.
In developing the discharge-specific MDL, it ie the permittee's responsibility
to ensure that the proper proceduree have been followed, that adequate data
have been--collected, and that sufficient documentation ia provided to justify
the discharge-specific MDL.
3.2 PSXMIT NRITBX'S SOLI
The permit writer'a role in the diacharge-apecific MDL development is
one of overaight and evaluation. To fulfill thia role, the permit writer
fully underat&nd tha diacharga-apeclfle MDL development requirements and the
procesa by which tha MDLs are used to generate the interia KLe. It alao
B3-4
Permit Writer's Cuide to Discharge-Specific MDL Evaluation
-------
March 7,1994
requires that the permit writer either pontii or obtain the technical
expertise raquirad to avaluata tha supporting data, determine tha adequacy of
assumptions, and interpret the data results. As noted in Chapter 2, during
this evaluation, the permit writer nay wish to consult with others, including
personnel from state and BPA Regional laboratoriee.
It is imperative that the data provided by the permittee for evaluation
allow the perait writer to reach the >aae conclusions and develop the *une
discharge-specific MDL. As such, the perait writer should be prepared to take
the data submitted by the permittee and calculate a discharge-specific MDL for
verification purposee.
As part of this process, the permit writer may be expected by the
permittee to act as a technical resource in understanding procedures for
discharge-sposific MDL development. Additionally, the review of these
requests may result in burdens being placed on permitteee and permitting
staff. As discussed previously under Section 2.2, it may be advantageous for
tha permitting authority to have a set program that involves roles and
responsibilities, and procedures for implementation, establishing standard
procedures for permittees similar to those outlined in Section 3.1 may be
useful, as would the use of a standard report format/outline or standard
auaoary form. Exhibit 3-1 provides one example of such a standard format.
Other suggestions that may ease the burden on the permittee and permit writer
have been highlighted in Section 2.2.
Permit Writer's Guide to Discharge-SpecifIc KDL Evaluation
¦J-*
-------
March 7,1994
EXHIBIT 3-ii DIICHAJW1-SPECIFIC NDL DKVZLOPMBfT WORKSHEET
DiaCHJUWX-SPICIFIC KDL development worksheet
Facility NiMt Dates
Facility Addresst
Psrmit Humbert
Pollut*nti
Sample tutrix*
Initial Estimate of XDL Based oot (Check one)
a. Concentration value that correspond* to an instrument
signal/noise in the rang* of 2.5 to 5
b. Concentration equivalent to three times the standard
deviation of replicate instrumental measursoents of the
analyte in reagent water
c. The region of the standard curve where there la a
significant change in sensitivity, i.e.# a break in the
slope of the standard curve
d. The instruments! limitations, which are described below.
ANALYTICAL TUT RESULTS
ROPllClf BllMlt Difference*
Concentration Units
• Mote tbafc results for the blank and the blank-subtracted
result (difference) only apply to those 40 C7R Part 136
methods that explicitly require a blank correction.
& »¦ __Spiking Level ____
S -
t • ____ as* a-I degrees of freedoa
for S* and t. A minimum of
7 replicatee are required.
KDL -
B3-6
Permit writer's Guide to Discharge-Specific HDL EvtU«io«i
-------
APPENDIX C
RECOMMENDED NPDES PERMIT WRITING BOILERPLATE LANGUAGE
-------
RECOMMENDED PERMIT WRITING BOILERPLATE LANGUAGE
This appendix provides recommended boilerplate language to use vhen
including WQBELtt in permits. The example used represents a
hypothetical situation in which the discharger's permit contains a
WQBEL for mercury. In addition, the example only addresses this
specific limit and the portions of the permit that are diractly
related. This example is not intended to represent an entire
permit, nor is it intended to be a recommendation to establish a
limit for mercury.
c-i
-------
RECOMMENDED PERMIT WRITING BOILERPLATE LANGUAGE
A. EFFLUENT LIMITATIONS AND MONITORING REQUIREMENTS
1. During the period beginning on the effective date of this permit and lasting through the
expiration date, the Permittee is authorized to discharge from outfall(a) serial number(s)
001: noncontact cooling water, boiler blowdown, and treated process water to the Merrie
River.
Such discharge shall be limited and monitored by the Permittee as specified below:
EFFLUENT CHARACTERISTICS
Flow (HGD)
Mercury, Total (1)
DISCHARGE LIMITATIONS
(lbs/day) Other Units (Specify)
Monthly Daily Monthly Daily
Average Maximum Average Maximum
N/A
0.0084
N/A
0.025
Report
0.1 /ig/L
Report
0.3 /tg/L
MONITORING REQUIREMENTS
Measurement
Frequency
Continuous
1/week
Sample
Type
Recorder
Grab
(1) The water quality-based effluent limitations for this parameter are not quantifiable using EPA
approved analytical methods. Thus, EPA has set forth a reporting threshold to measure the highest,
acceptable quantification level for this parameter,
discharge of total mercury in excess of the WQBELs.
in this permit.
This reporting threshold does not authorize the
For more information, see special condition A
C-2
-------
RECOMMENDED PERMIT WRITING BOILERPLATE LANGUAGE (Continued)
A. Conditions Applicable to WQBELs Below Detection Levels
1. For purposes of reporting, the Permittee shall use the
reporting threshold equivalent to the ML. The ML is
defined as the concentration in a sample equivalent to
the concentration of the lowest calibration standard
analyzed in a specific analytical procedure assuming
that all the method-specified sample weights, volumes,
and processing steps have been followed. As such, the
permittee must conduct analyses in accordance with the
method specified below and must utilize a standard
equivalent to the concentration of the ML specified
below:
ML and Lowest
Calibration
Parameter Analytical Method Concentration
Mercury 245.2 0.2 /xg/L
2. For purposes of reporting on the discharge monitoring
report, actual analytical results should be reported
whenever possible. All analytical values at or above
the ML shall be reported as the measured value, when
results can not be quantified, values below the ML
shall be reported as "0".
3. In the "Comment Section" of the DMR, the permittee
shall report the lowest calibration standard used, the
ML achieved, and the number of times non-detectable
results were reported as "0".
C-3
-------
APPENDIX D
RECOMMENDED PERMITTEE PROCEDURES FOR REPORTING
-------
PERMITTEE PROCEDURES FOR REPORTING
RECOMMENDED PERMITTEE PROCEDURES FOR REPORTING
When reporting analytical data on Discharge Monitoring Rpport
(DMR) forms, permittees should follow the instructions set forth
in the EPA Office of Water publication entitled NPDES Self-
Monitoring System User Guide, March 1985. Reporting procedures
also have been summarized in the general instructions given on
the DMR report form. These instructions are provided toelo^.
DMR General Instructions
1. If form has been partially completed by preprinting,
disregard instructions below pertaining to preprinted
information.
2. Enter "PERMITTEE NAME/MAILING ADDRESS" (and facility
name/location, if different). Enter "PERMIT NUMBER," and
"DISCHARGE NUMBER" where indicated. (A separate form is
required for each discharge.)
3. Enter beginning and ending dates for "MONITORING PERIOD"
covered by form where indicated.
4. Enter each "PARAMETER" as specified in monitoring
requirements of permit.
5. Enter "SAMPLE MEASUREMENT" data for each parameter under
"QUANTITY AND "QUALITY" in units specified in the permit.
"AVERAGE" is normally arithmetic average (geometric avaraqe
for bacterial parameters) of all sample measurements for
each parameter obtained during "MONITORING PERIOD."
"MAXIMUM" and "MINIMUM" are normally extreme high and lov
measurements obtained during "MONITORING PERIOD." (NOTE to
municipals with secondary treatment requirement, enter 30-
day average of sample measurements under "AVERAGE" and «nt«r
maximum 7-day average of sample measurements obtained durirxj
monitoring period under "MAXIMUM."
6. Enter "PERMIT REQUIREMENT" for each parameter under
"QUANTITY" and "QUALITY" as specified in permit.
7. Under "NO. EX" enter the number of sample measurements
during monitoring period that exceed the maximum and/or
day or monthly average (and/or are lower than the mini sua)
permit requirement for each parameter. If none, enter ").m
8. Enter "FREQUENCY OF ANALYSIS" both as "SAMPLE MEASUREMEXT•
(actual frequency of sampling and analysis used during
monitoring period) and as "PERMIT REQUIREMENT" specified in
D-l
-------
PERMITTEE PROCEDURES FOR REPORTING
permit (e.g. enter "CONT." for continuous monitoring; "1/7"
for one day per week, "i/30" for one day per month;
for one day per quarter, etc.)
9. Enter "SAMPLE TYPE" both as "SAMPLE MEASUREMENT" (actual
sample type used during monitoring period) and as "PERMIT
REQUIREMENT" (e.g. enter "GRAB" for individual sample;
"24HC" for 24-hour composite; "N/A" for continuous
monitoring, etc.).
10 WHERE VIOLATIONS OF PERMIT REQUIREMENTS ARE REPORTED, ATTACH
A BRIEF EXPLANATION TO DESCRIBE CAUSE AND CORRECTIVE ACTIONS
TAKEN. REFERENCE EACH VIOLATION BY DATE.
11. If no discharge occurs during the monitoring period, enter
"NO DISCHARGE" across form in place of data entry.
12. Enter "NAME/TITLE OF PRINCIPAL EXECUTIVE OFFICER" with
"SIGNATURE OF PRINCIPAL EXECUTIVE OFFICER OR AUTHORIZED
AGENT." Also enter "TELEPHONE NUMBER" and "DATE" at bottom
of form.
13. Mail the signed discharge monitoring report to the Office(s)
in accordance with the date(s) specified in the permit.
Retain a copy for your records.
14. More detailed instructions for use of this DMR form may be
obtained from the Office(s) specified in the Dermit.
In accordance with EPA's national Guidance for the Permitting,
Monitoring, and Enforcement of Water Quality-Based Effluent
Limitations Set Below Analytical Detection/Quantification Levis,
the following guidance should be followed when reporting
analytical data below detectable levels:
• Substitute "0" as the value for analytical test results
below the ML;
• Use "0," for results below the ML when determining the
average and maximum values in accordance with the DKJt
instructions;
• Record the determined average and maximum values in
accordance with DMR instructions; and
• Record in the comment section of the DMR, the lowest
calibration standard used, and the ML achieved, and the
number of times non-detectable results were reported as
zero.
D-2
-------
PERMITTEE PROCEDURES FOR REPORTING
This appendix provides two sets of example data to demonstrate
how MLs are determined and how average concentrations and
loadings should be calculated when the permit limits are at or
below the detection/quantification level. Using these same
example data, determinations of exceedances also have been
provided. These data have been summarized on example DKRs
included near the end of this appendix.
Using the ML as a Reporting Threshold
This example presents hypothetical permit conditions in Table l
on the next page, in which WQBELs are below analytical detection
limits. For each pollutant, the values to report on the DMR will
depend on the ML; thus, it is important to correctly identify the
ML. Table 1 depicts the following situations:
• For pyrene, an ML is specified in the approved method.
The permittee must achieve this ML in its analyses and
must base its reporting calculations on this ML.
• For dieldrin, no ML is specified in the approved
method. The permittee has therefore calculated an
interim ML equal to 3.18 times the MDL (and then
rounded to an appropriate calibration dilution). The
permittee must achieve this ML in its analyses and aust
base reporting calculations on this ML.
• For benzene, an ML, is specified in the analytical
method, but the facility is unable to achieve the ML
because of matrix interference. Due to problems with
matrix interfernce, the permittee has developed a
discharge-specific MDL, and correspondingly, an ML
equal to 3.18 times the discharge-specific MDL. In
this case the facility has previously demonstrated to
the permitting authority that a discharge-specific KL
was warranted. Note that the benzene example is
expanded to demonstrate the calculation of monthly
averages and the compliance evaluation.
Calculating Monthly Average Concentrations Using Example
Analytical Data
A set of hypothetical data is provided in Table 2, to which all
procedures refer. These data are fictional and are intended to
serve as an example only. The example represents a situation
where water quality-based permit limits are required for benzene;
however, it also is applicable to any other pollutant. Both tne
D-3
-------
Table 1
Determining the ml
Pollutant
Method
1(#*g/L
)
HDL
(H 9/L)
Method
Specified
ML (/ig/L)
Interim
ML1
(^g/L)
Discharge
Specific
ML2 (fjg/L)
Permit Limits
Daily Month. Daily Month.
Max. Avg. Max. Avg.
(nq/h) (/ig/L) (lb/d) (lb/d)
Pyrene
1625
1.93
10
NA
NA
5
5
0.42
0.42
Dieldrin
608
0. 002
None
1
NA
0.1
ng/L
0.1
ng/L
8.3 x
10~6
8.3 X
10~6
Benzene
1624
4 . 43
10
NA
20
10
10
0.83
0.83
NA - Not Applicable
Specified in Permit
Specified in Permit and previously approved by permitting authority following a
demonstration of a discharge specific MDL (from which an interim ML was derived) such as
described in Appendix B.
^Because Methods 1625 and 1624 •¦ploy th« concept of an ML, a single laboratory MDL
determination (such a« appear* in the 600 eeriaa methods) was not published in Method 1625.
U-4
-------
Tabla 2
Example Monthly Laboratory Analytical Data-Concentrations
Parameter: Benzene
EPA Method: 1624
ML: 20 /ig/L
Date
Result (M9/L)
Date
Result (pj/L)
9/1/93
< 20
12/2/93
< 20
9/3/93
< 20
12/5/93
< 20
9/6/93
< 20
12/9/93
< 20
9/10/93
< 20
12/12/93
< 20
9/13/93
< 20
12/15/93
< 20
9/17/93
< 20
12/18/93
< 20
9/20/93
< 20
12/21/93
20
9/24/93
< 20
12/22/93
20
9/27/93
< 20
12/23/93
< 20
12/25/93
< 20
12/28/93
< 20
12/31/93
< 20
10/1/93
< 20
1/3/94
< 20
10/5/93
< 20
1/5/94
< 20
10/8/93
20
1/9/94
< 20
10/12/93
30
1/12/94
< 20
10/15/93
30
1/15/94
< 20
10/19/93
40
1/19/94
< 20
10/22/93
30
1/22/94
< 20
10/23/93
20
1/25/94
30
10/26/93
< 20
1/28/94
30
10/29/93
< 20
11/1/93
< 20
2/1/94
30
11/4/93
< 20
2/2/94
50
11/7/93
< 20
2/4/94
40
11/11/93
< 20
2/7/94
30
11/14/93
< 20
2/10/94
< 20
11/18/93
< 20
2/14/94
< 20
11/21/93
30
2/17/94
< 20
11/25/93
40
2/20/94
< 20
11/28/93
< 20
2/24/94
< 20
2/27/94
< 20
0-5
-------
PERMITTEE PROCEDURES FOR REPORTING
daily maximum and monthly average limits are below the ML, which
in this case, is a discharge-specific ML. The steps for
calculating a monthly average concentration for reporting
purposes and compliance evaluations follow.
1. Determine the number of samples analyzed during the month
for a given parameter.
Using the example data provided in Table D-2, the number of
samples analyzed were:
September = 9 December = 12
October = 10 January = 9
November = 9 February = 10
2. Determine the number of results that were found to be below
the ML (nondetectable).
Using the example data provided in Table D-2, the number of
results below the ML were:
September = 9 December = 10
October » 4 January = 7
November = 7 February =¦ 6
3. Calculate the monthly average concentrations, substituting
zero for the results belov the ML (nondetectable).
Average Sum of all results
Concentration = Number of samples taken during the month
Using the data shown in Table D-2, the monthly average
concentrations for benzene are calculated as follows:
September (9)(0) = 0.0 nq/L
9
October f4W0) + f3M30) + f2H20) + flM40) = 17
10
November (7) (0) + fl) (301 + (1) (40) =» 7.8 ttq/L
D-6
-------
PERMITTEE PROCEDURES FOR REPORTING
December (101(01 + 2(20)
12
3.3 nq/L
January (71(0) + 2(301
9
6.7 ng/L
February (61 (01 + 2(301 + (11 (401 + (11 .'501 = 15 nq/L
Calculating Monthly Average Loadings Using Analytical Data
1. Calculate the monthly average loadings, substituting zero
for the results below the ML (nondetectable). Using the
same six months of data shown in Table 2, the loadings for
benzene are calculated as shown in Table 3. The conversion
factor from concentration in nq/L to loading in lb/day is
0.00834. While loading limits for benzene are uncommon,
these calculations are transferrable to any pollutant.
Monthly average loading is calculated as follows:
Table 3 illustrates the calculation that converts the
pollutant concentration data in nq/L into a loading in
lb/day. Using these converted values, the monthly average
loadings for benzene are calculated as follows:
September (91(01 = 0 lb/d
October (41(01 + 1.70 + 2.53 + 2.53 + 3.27 + 2.50 + 1.68
10
Monthly
Average
Loading
sum of [(each result in tig/h) (flow in
MGD1 10.00834 (lbl/(gall(uglVl
Number of samples taken during the month
11
10
> 1.42 lb/d
November (71(01 + 2.6 + 3.4
9
0.67 lb/d
December (101 (01 + 1.67 + 1. 68 =»
0.28 lb/d
12
D-7
-------
PERMITTEE PROCEDURES FOR REPORTING
January f7)fQ) + 2«5 + 2.5 ~ 0.56 Ib/d
9.0
February rewoi + 2.48 + 4.09 + 3.30 + 2.52 - 1.24 lb/d
10
D-a
-------
Table 3
Example Monthly Analytical Data - Loadings
Parameter: Benzene j
EPA Method:
1624
ML: 20 uq/L |
Date
Result
No. to use in
(^g/L)
calculations
Flow
Quantity 1
(Ug/L) x (MGD) >
conversion
= (lb/d)
| 9/1/23
< 20
0
10.1
0.00834
0 !
9/3/93
< 20
0
10.4
0.00834
0 |
9/6/93
< 20
0
9.9
0.00834
0 I
9/10/93
< 20
0
9.8
0.C0834
0
9/13/93
< 20
0
10.1
0.00834
0
9/17/93
< 20
0
10.2
0.00834
0
9/20/93
< 20
0
10.3
0.00834
0
9/24/93
< 20
0
10.2
0.00834
0
9/27/93
< 20
0
10.1
0.00834
0
10/1/93
< 20
0
9.7
0.00834
0
10/5/93
< 20
0
10.4
0.00834
0
10/8/93
20
20
10.1
0.00834
1.70
10/12/93
30
30
10.1
0.00834
2.53
10/15/93
30
30
10.2
0.00834
2.53
10/19/93
40
40
9.8
0.-00834
3.-2 7
10/22/93
30
30
10.0
0.00834
2.50
10/23/93
20
20
10.1
0.00834
1.68
10/26/93
< 20
0
10.0
0.00834
0
10/29/93
< 20
0
10.2
0.00834
0
11/1/93
< 20
0
9.9
0.00834
0
11/4/93
< 20
0
10.0
0.00834
0
11/7/93
< 20
0
10.2
0.00834
0
11/11/93
< 20
0
10.0
0.00834
0
11/14/93
< 20
0
10.4
0.00834
0
11/18/93
< 20
0
10.2
0.00834
0
11/21/93
30
30
10.4
0.00834
2.6
11/25/93
40
40
10.2
0.00834
3.4
11/28/93
< 20
0
10.0
0.00834
0
D-9
-------
Tabla 3
Bxaapl* Monthly Analytical Data - Loadings (Continued)
1 — ¦¦¦¦—¦ ' 1—" ¦ =
Parameter*. Benzene
EFA Method:
1624
ML: 20 jug/L
Date
Result
No. to use in
(Wf/D
calculations
Flow
Quantity
(jig/L) >
C (MGD) >
conversion
- (lb/d)
12/2/93
< 20
0
9.8
0.00834
0
12/5/93
< 20
0
9.8
0.00834
a
12/9/93
< 20
0
9.7
0.00834
0
12/12/93
< 20
0
9.8
0.00834
0
12/15/93
< 20
0
9.8
0.00834
0
12/18/93
< 20
0
10.0
0.00834
0
12/21/93
20
20
10.0
0.00834
1.67
12/22/93
20
20
10.1
0.00834
1. 68
12/23/93
< 20
0
10.0
0.00834
0
12/25/93
< 20
0
9.9
0.00834
0
12/28/93
< 20
0
9.8
0.00834
0
12/31/93
< 20
0
9.7
0.00834
0
1/3/94
< 20
0
9.7
0.00834
0
1/5/94
< 20
0
9.8
0.00834
0
1/9/94
< 20
0
9.8
0.00834
0
1/12/94
< 20
0
9.9
0.00834
0
1/15/94
< 20
0
10.1
0.'00834
0
1/19/94
< 20
0
10.0
0.00834
0
1/22/94
< 20
0
9.9
0.00834
0
1/25/94
30
30
10.0
0.00834
2.5
1/28/94
30
30
10.0
0.00834
2.5
2/1/54
30
30
9.9
0.00834
2.41
2/2/94
50
50
9.8
0.00834
4.09
2/4/94
40
40
9.9
0.00834
3. 30
2/7/94
30
30
10.0
0.00834
2.52
2/10/94
< 20
0
10.1
0.00834
0
2/14/94
< ao
0
10.2
0.00834
0
2/17/94
< 20
0
10.2
0.00834
0
2/20/94
< 20
0
10.0
0.00834
0
2/24/94
< 20
0
10.1
0.00834
0
2/27/94
< 20
0
9.9
0.00834
0
D-10
-------
PERMITTEE PROCEDURES FOR REPORTING
Determining Exceedances
Using the example data provided in Tables 2 and 3 and comparing
these data to the WQBELs in Table 4 below, the following
compliance determinations can be made. Notice that in this
example, both the daily maximum and the monthly average are equal
and are below the ML.
Table 4
Water Quality-Based Effluent
Limitations for Benzene
Concentration
(Mg/L)
Loading
(lbs/dayj
Daily maximum
10
0.83
Monthly average
10
0.83
The number of exceedances reported on the DMR refer to the number
of sample measurements during a monitoring period that exceed the
maximum permit limit requirement for each parameter. When an
NPDES permit contains limits for both the concentration and the
loading for a parameter1, it is possible to violate both liaiti
with one sample measurement. However, this is reported as one
exceedance.
Table 5 summarizes the exceedances to be reported on the DMR
forms on the following pages. Of note is the fact that for this
example, the daily maximum is exceeded whenever a result greater
than the ML occurs. However, if more than one sample is analyzed
on a particular day (i.e. multiple grab samples are collected)
then the daily maximum is calculated as the average of saaplee
within the 24 hour period. As a result, a concentration found
above the ML would not necessarily result in an exceedance of the
daily maximum limit.
Sample completed DMRs follow this discussion. The daily maxiaua
concentrations, calculated monthly averages, and loadings are
presented along with the number of exceedances. As per
instructions in the beginning of this appendix, the lowest
calibration standard and the KL achieved is also noted in the
comments section. Since this is a fictional' example, it does not
include attachments explaining the exceedances. In reality, the
permittee should attach a brief explanation to describe the c«u««
and corrective actions taken, with each violation referenced by
date.
D-ll
-------
Table 5
Exceedances of Benaene Limits
Date
Daily Max.
Cone.
Potentially
Exseeding
the Limit
(Mg/D
Daily Max.
Loadings
Potentially
Exceeding
the Limit
(lb/d)
Mo. Avg.
Cono.
(M9/L)
Mo. Avg. 1
Loadings 1
(lb/d) |
September
9/1/93-
9/30/92
0
0
0
•
Number of Exceedances of Daily Max: 0 |
October
10/8/93
20
1.70
-
-
10/12/93
30
2.53
-
-
10/15/93
30
2.53
-
-
10/19/93
40
3.27
-
-
10/22/93
30
2.50
-
-
10/23/93
20
1.68
-
-
10/1/93-
10/31/93
-
-
17
1.42
Number of Exceedances of Daily Max: 6
November
11/21/93
30
2.6
—
-
11/25/93
40
3.4
-
-
11/1/93-
11/30/93
-
-
7.8
0.67
8 Number of Exceedances of Daily Max: 2
December
12/21/93
20
1.67
-
-
12/22/93
20
1.68
-
-
12/1/93-
12/31/93
-
-
3.3
0.28
Number of Exceedances of Daily Max: 2
0-12
-------
Table 5
Exceedances of Bensene Liaits (Continued)
Date
r
Daily Max.
Cono.
Potentially
Exceeding
the Liait
(Mg/L)
Daily Max.
Loadings
Potentially
Exceeding
the Limit
(lb/d)
Mo. kvg.
Cono.
(M9/L)
1
Mo. Avg. fl
Loadings 5
(lb/d) |j
January j
1/25/94
30
2.5
-
1
1/28/94
30
2.5
-
-
1/1/94-
1/31/94
-
-
6.7
0. 56
Number of Exceedances of Daily Max: 2
February
2/1/94
30
2.48
-
-
2/2/94
50
4.09
-
-
2/4/94
40
3.30
-
-
2/7/94
30
2.52
-
-
2/1/94-
2/28/94
-
-
15
1.24
Number of Exceedances of Daily Max: 4
D-13
-------
fu> «•""'/ J, >^i tyytf ?«*/ ,-~/' "i'Y
L^rf1' ~n7 '"
LVO
7
0/
£4
MM
91 fO
T~
in
il
35
<1*'9
ttom
-5-455 >£¦•' f
JOtn
X3»
wmw
>¦ L,.;A.d l-u_
IS
I®
I
'Jri&V
1(W
i«»ov oinnoHipv m nrvi^o
lAim^pi iv#2
. i «•'
".•sir
..:• ij '•/ v
' ,r'* •'.
«' . I 1
tn_- JoL
«'¦ ¦ ¦'' -
h1' ::.X^
:JI
o
«F?"vi
i ¦ r
jJli
jr
*« * ^
i?i •..{
a. i. i:
*' -
n l*rf" 1 11'
/ 7 / u. / ¦» ,' -/»•« •••/" • •'' r. 'V •) * 1 /
I I
**•
#•« | >f# ••
¦MWtl < *
l*M •! •!«
J^iry j * • «««
I riAU/Wr*
i.AKiV.lL
• -'•*•- 'i n
¦ !;»r.«h?i.-'
i«-£
J-' ^':Jfe
wmx«m araoAv
~77FiT)
NWivimaaiwa ma Mnvno (*wo W) »)
UUOI«nn » mi*linn •uononit«ujpMu nuoK MUtUu#'. W
dm)
phi
60Vi
• IMTI
• - 'im-'1 '
sMfeiEfe
r. ''* :' •
'• «¦ - '•
ji
i *•
•¦.
y°i
at-* o
iM&l
¦•i .*« '$»:j
¦*
Jli-
o
I -oj
(nw)
te-iC-Ol m*anima*a\r
»ooo-omx cncno
fWOiMy woj
ZuHmW
¦¥«i| | *yt» ow mva
no juuiwvne ('1*0 W f J
IVMVl
il
IMtNWOtVIN
inimvtMtw
IIMVM
iiimiMinm
iimpm
iWlllMfWVIH
rwnvi
IWfl
fir to
aOIMlrf ONIMOXINOH
HOW
'—Z Z Z Z '^P^^DZZIU. Dl'
N0UV9O1
umv<
l
(WHO ) 1M043U 9MMO1M0M B9UVHOSIO
< T1QJN) Mil*** NOU'MH I) kMVhOMO IWITT1M IVWIlmi
MMVINIMtfU ~] - " iTHtT ~ ~V^P *1
\oooooosrr -- v.— wir*',T*yr'T ****
—T^T—^ -brpywrnf^wf 3^3, JSKJC. ,£5
•r*WJ MMooy/iNVN aftiu»w««
-------
HMTTII NAXE/AOOeeiS /tefte*
*pam*m 1_ kALiii£JLl*cr doo-ff 1 Anpprt/w
&£.«., Hv-AJg llyAr 12JLL
MAHOMAL raU.UT*M OIBCM*m« «l »HfW> StriTIM (NFDtSt
OlSCHARCe M0MT0IMN6 REPORT I OM«)
(Hi) tltttl
rot* AppHWMl-
Qtmm 2040*0004
Approve# «*p*me 10-31-44
fAourr ___. ..
LOCAT!OW_ Ca n_
HONITORINO PIRIOU
VfAMl HO OAT
¦PIE: iwid ImIhiwom >ww
tnta term
i4 Cknt Onlr) OUALT* oil COMCDfTRATIOM
I IM-iff l*tfJ) (14411
(j an o*r) quantity om ld*p»w
UL111 U±£U
ui in
it* mi
CQfJT
h . r.r i
lb/4
Beni£fi£
*, ;-r
r /»¦. ¦« .a;
i 4,> * M . I
¦mm
r-^Sh'l '•
rr7-PTr~in
Mitii
a*
IF'Affi.i-;;
.. r M.f y.*
LArilf
PFifPn
M3.
• 1*1
c? |W IV,
Kjcunvr
I COM I ——» |*«a«
'A Lf^'U*4 /rl*uts ftfiX»K-J rfS wo 1 r,^ri a~t of
-------
VMMTIII NAMK/AOONMt fbtMi
Apa/*m flood US OQQQ00 I
frk-CuA^IJlZJ2JU
NATIOMAL KU.UMM MC1MM) « MMATtOM svariM (NFDtS •
DISCHARGE MOMTOftINCIKMRT I Dull
(t-lit «/M*>
1*?^!! - _ rum MO I PA* I |t«A»| MO 1 OAT
" ktlZ-Wn*
MONITOR I NO PKNIOD
Font >^nw»l
QM NO 2MO-OOM
r^—rr
>.» . .1
fe'g' •>.•
IMT
MAWMUM
(o7\~
vS
;r« •>»
,.< »¦ in"
i[iy imji.
' 4 r» U\K * ,*•
^LMjL
'-».r un-p'^r
IMIII
AI6P
(4 Old Oitff J OOAUTV OK COMCKNKIATION
(U-iit (*4fJ) (!«•«/>
¦ ft jjUjr ft
• VTTTT'i
.'J'F ¦
,J"VW
AVSMAOK
iiisiiii
t; f~ _
>'• •,
t- ;¦ .tw ».
-iiE-ii
LSJtii?
•v®
»
* . « / ( / «4 /» »' #1
»»• u «M is«t • rT—n>i11
!•« 10 W »ui f f^P .Mill MM
-¦¦ > »*«4 m i^mwi «
!#• M MilM
« •« »« •• I d»<
fmm;
MAKMUM
;J-i -Ti
• i>K» , - '• ' "1
V it;
- -gfigg.-l !
»;/•», 't : i .«
&.5i , «.
\ 4'< 1:
•» • I
g 4Cuin'r
orcein OA A^TXQMUIO AOCNI
<4^«f
rt
/A-
ii
fp
z/wfc
IS
57 r*v; ••» i'
^p,JS.:S fc"
M&
iiSi
5f>
aWA
I.SST
6>rr«6
jL7,.U|
m
fi
e a r c
n
¦i
A l'- J- C'^ IL—
r n »i J* '(f* L ^ "T' i I A . ^7 -frVLTS
'*7 .. Cl(- *i tufA^
-------
COMPLIANCE EVALUATION
The permittee must periodically submit a discharge
monitoring report (DMR) to the permitting authority for its
review to determine whether there are violations of NPDES permit
requirements. Any violation of a NPDES permit is a violation of
the Clean Water Act for which the permittee is liable. EPA uses
a policy of enforcement discretion in deriding on a response to a
violation. Compliance history, type, frequency, and magnitude of
the violation are some of the factors considered in deciding what
an appropriate enforcement action should be. Enforcement actions
can be Informal (e.g., phone call) or formal (e.g., issue an
Administrative Order). EPA's Enforcement Management System
(1989) and Guidance for Preparation of Quarterly and Semi-Annual
Noncompliance Reports provides recommendations for evaluating
noncompliance and determining appropriate enforcement responses.
Benzene Example
The recommendations in the "National Guidance for
Permitting, Monitoring, and Enforcement of Water Quality-based
Effluent Limitations Set Below Analytical Detection1* were applied
to evaluate compliance in the benzene example. Non-detect values
were reported as 0 on the DMR, and 0 represented those values In
the calculation of monthly averages. The ML is identified and
the number of non-detect values for that month are identified In
the comment section of the DMR. Based on all the information
included in this Appendix, example appropriate enforcement
responses are discussed below. These are meant to be
illustrative, not prescriptive. The permitting authority has
discretion in determining an appropriate response.
«Ap»flmhAr
Compliance status: Facility is in compliance with daily maximta
and monthly average limits.
Violations: none
Enforcement response; none
Ootober
Compliance status: Facility is non-compliant.
Violations; 6 exceedances of daily maximum limit; monthly
average limit exceeded.
Enforcement response: exceeds the violation review action
D-17
-------
criteria (VRAC) outlined in the EMS which requires these
violations to be reviewed by a professional. Due to dramatic
difference between Sept. and Oct., the permitting authority
called the facility to inform them of the violation and see if
there is an explanation. No explanation was provided so a letter
was sent (under section 308 of the Clean Water) requiring the
facility to provide additional monitoring data or other
information (any change in process) to determine the increase in
the discharge of benzene.
Any limit violation suspected by th« permitting authority to have
the potential to cause a water quality or health problem exceeds
can be considered significant noncompliance (SNC) and reported on
the quarterly noncompliance report (QNCR).
Hnpawhar
Compliance status: Facility is non-compliant.
Violations: 2 exceedances of daily maximum limit
Enforcement response: no additional action taken this month
Compliance status: Facility is non-compliant.
Violations: 2 exceedances of daily maximum limit
Enforcement response: no additional action taken this month
January
Compliance status: Facility is non-compliant.
Violations: 2 exceedances of daily maximum limit
Enforcement response: DMR data and additional data collected by
308 letter indicate a chronic problem with this facility meeting
its limits for benzene. Notice of violation sent to the
facility.
February
Compliance status: Facility is non-complaint.
0-18
-------
Violations: 4 exceedances of daily maximum limit; monthly
average limit exceeded.
Enforcement response: 4 or more violations of daily maximum
exceed the VRAC so the violations are professionally reviewed.
Also, this is the second month within 6 months that the monthly
average was exceeded so a professional review of those violations
is required.
The monthly average violations were evaluated to see if they
should be reported on the QNCR. They should be reported on the
QNCR if the monthly average limit violation is greater than or
equal to the product of the technical review criteria (TRC) times
the effluent limit, and occurs 2 months within a 6 month period.
Benzene is an organic pollutant and is therefore classified as a
Group II pollutant with a TRC ® 1.2.
First month of violation was October: 17 > (1.2 x 10)
17 > 12
Second month of violation was February: 15 > 12
This facility is in significant non-compliance (SNC) because
benzene discharged from this facility exceeded the product of TRC
times the monthly average limit 2 months in a 6 month period.
There are also chronic exceedances of the daily maximum limit.
This facility is listed on the QNCR and an Administrative Order
is issued.
D-19
-------
APPENDIX E
RECOMMENDED PERMIT WRITER PROCEDURES FCR CALCULATING WATER
QUALITY-BASED EFFLUENT LIMITS AT INTERNAL WASTESTREAMS
E-l
-------
RECOMMENDED PERMIT WRITER PROCEDURES FOR CALCULATING WATER
QUALITY-BASED EFFLUENT LIMITS AT INTERNAL WASTESTREAMS
With the continuing imposition of WQBELs in permits, many
facilities are noting the feasibility in combining discharges into
one outfall prior to discharge. One advantage of this practice is
that combining vastestreams minimizes costs associated vith sample
collection and analyses at multiple outfal»ls. Combined
vastestreams may also provide greater dilution for specific
pollutants, thus, the WQBELs are more easily met. However, in some
cases, dilution of a given vastestrean vith other wastewaters may
result in matrix interferences and/or concentrations of pollutants
below detectable levels. Where a wastestream is diluted by other
vastevaters, EPA recommends that the permit writer consider
establishing a WQBEL for the individual wastestrearn. The following
approach sets forth several steps for the calculation of a WQBEL at
an internal outfall.
STEP 1: Locate an accessible sampling point of the concentrated
wastestream, C. If not immediately accessible, the
permit writer may require the construction of an
accessible monitoring point pursuant to 40 CFR 122.4 5(h).
STEP 2: Determine the long term average flow at the internal
sampling point (QJ
STEP 3: Determine the long term average flow of all dilution
vastestreams (Sum of QJ
STEP 4: Assuming that the dilution wastestream is contributing no
tangible amount of the pollutant, calculate the
concentration at the internal sampling point (Cc) which
must be met in accordance with the following formula:
c- WQBEL ~ Q4
0.
An example of this calculation is provided on the following page.
E-2
-------
RECOMMENDED PERMIT WRITER PROCEDURES FOR CALCULATING WATER
QUALITY-BASED EFFLUENT LIMITS AT INTERNAL WASTESTREAMS
Continued
Dilution
Wastestreams
Concentrated
Wastestrean
Q,, - 2 MGD
Q* - 8 MGD
Q, ¦ 0.1 MGD
&
Point Where Interr.<
WQBEL, Ce, Applies
Race i v 1 ng
water
Point
where
WQBEL
Applies
Combined Outfall to
Receiving Water
Q, - (2 + 8 + 0.1) - 10. l
3
C- WQBEL * O,
0.
C« 0.0002 mq/1 • 10.1 MGD
0.1 MGD
C#- 0.02 ag/1
------- |