iSISZi
S PPO^°
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D C 20460
r»T"w
*
OCT 15 1993	OCT I 2 I993
U&ZW®0USW''
OFFICE OF
Vrtill-U. .oivjN	SOLIO WASTE AND EMERGENCY RESPONSE
OSWER Directive 9320.2-07A
MEMORANDUM	]
1	/"\
/,
SUBJECT: Additional Guidance on j'Worst1 Sijtes" and "NPL Caliber
Sites" to Assist in SAClf Implementation j]
FROM:
Richard J. Guimond,	/ Jj J//./*" '«i-
Assistant Surgeon General-'/ "USPHS " '
Acting Assistant Administrator
TO:	Addressees
OBJECTIVES
This document is intended to assist the Regions by giving
clear guidance as to what constitutes NPL caliber sites and to
assist in minimizing the potential for "false positive" NPL
packages. It also sets forth the actions needed to support our
efforts to implement SACM and encourage appropriate data
gathering to support NPL listing and RI/FS decisions. The end
result should be that we take more early actions to reduce risks
and clean up sites, that the RIs to support long term actions
begin earlier in the process, and that the risks of starting
action earlier in the assessment process are acceptable. This
document supplements guidance issued by OERR on August 26, 1993
(OSWER Directive 9320.2-072).
BACKGROUND
The focus of the Superfund program under the Superfund
Accelerated Cleanup Model (SACM) continues to be reducing risk at
sites posing the worst health and environmental threats. Through
integrated assessment and the Regional Decision Teams' (RDTs)
actions to identify worst sites and take early and long-term
actions, SACM focuses the Regions' efforts on one list of the
worst sites requiring response action. This list of worst sites
would include NPL caliber sites as well as other sites that may
be eligible for removal actions.
Regional concerns have been raised regarding the need to
better define the term "NPL caliber sites." The Regions have
been encouraged to begin RIs at prospective NPL sites as soon as
they identify them rather than waiting until a site has been
proposed for the NPL. Some Regions have expressed concerns about
the risks involved. We recognize that this approach carries some

-------
-2-
risk that the site will not score above 28.5 on the Hazard
Ranking System (HRS). The Regions clearly want to avoid spending
considerable resources and time preparing HRS packages and
starting RI/FS's on sites that do not deserve priority remedial
action.
There are, moreover, some basic constraints that the Agency
will always face in selecting priority sites:
•	PA/SIs and the HRS are screening processes designed to
make decisions on thousands of potential sites each year, based
on limited data, to save assessment costs. In dealing with such
large numbers, the potential for false positives (sites that
initially appear to qualify for the NPL but, upon further review,
do not) cannot be totally avoided.
•	NPL listing is a legal process that entails comment and
potential legal challenge. Interested parties have the right to
present data during the comment period to refute the site score.
In some instances, this new information may result in a
conclusion that the site does not qualify for the NPL.
Given these constraints, EPA can still take action to better
define priority sites and reduce the number of potential false
positives. While it is not possible to eliminate the risk of
false positives, we can reduce those risks to acceptable levels.
IMPLEMENTATION: ACTIONS TO BETTER DEFINE NPL CALIBER SITES
We have previously issued guidance ("Guidance on Setting
Priorities for NPL Candidate Sites," OSWER Directive 9203.1,
October 28, 1992) encouraging Regions to set priorities for
screening among the worst sites for NPL listing. Regions should
be assessing their CERCLIS inventory of sites to identify those
posing the greatest present and potential threats, and therefore
most deserving the time, resources, and priority involved in NPL
listing. Such priority screening should result in NPL packages
more likely to score above the cut-off score and should identify
sites posing the greatest health and ecological risks.
In addition, the fact sheet "Assessing Sites Under SACM —
Interim Guidance" (OSWER Directive 9203.1-051, Vol. 1 No. 4
December 1992) offers examples of NPL caliber sites. Those
examples include sites where:
Public drinking water supplies are contaminated
with a hazardous substance
Private wells are contaminated with a hazardous
substance above a health-based benchmark
Soils on school, daycare center, or residential
properties are contaminated by a hazardous substance
above background levels

-------
-3-
A hazardous substance is detected above background in
an off-site air release in a populated area.
•	A highly toxic substance known to bioaccumulate (e.g.,
PCBs, mercury, dioxin, PAHs) is discharged into surface
waters.
Sensitive environments (e.g., critical habitats for
endangered species) are contaminated with a hazardous
substance above background levels.
Some percentage of sites that have those characteristics will,
upon review, not score above 28.5 due to the small number of
targets, small waste quantity, etc. Expressed in general terms,
sites where significant human exposures to hazardous substances
have been documented or where sensitive environments have become
contaminated should be considered NPL caliber sites.
A preliminary evaluation should be performed to determine
whether significant human or environmental exposure is likely,
and thus a score of 28.5 would result. If the results of this
preliminary evaluation show that the score will only marginally
exceed 28.5, the Region should confirm the adequacy of scoring
data. If additional information is required, this information
could be requested from the PRP or obtained from other sources.
If any additional information on the type of contamination, its
constituents, pathway and transport characteristics, or other
factors is determined to be appropriate, it should be provided to
the RDT for use in confirming the importance of the site.
The RDT will review all potential NPL sites to ensure
appropriate data are gathered to support the HRS package, as well
as for use in preparing the risk assessment and determining the
need for long-term remediation. Once the data have been
gathered, the RDT will ensure the site is one of the priority
"worst sites" in the Region before submitting the NPL package to
Headquarters or initiating a response action. For setting
Regional priorities, the RDT can consider concentrations of
hazardous substances, risk assessment information where
available, human exposure or potential exposure, States'
priorities, environmental equity and other community concerns.
The RDT should ensure that the priority setting approach set
forth in the "Guidance on Setting Priorities for NPL Candidate
Sites" is followed. The attached flow chart on integrated
assessments depicts the role of the RDT and the various checks
built into the assessment process.
The following actions should also help us to focus our
efforts on the highest priority NPL sites in the future:
•	For those SACM pilot sites where the scores were found to
be lower than the Regions had expected, Headquarters (HQ) will
work with the Regions to evaluate what caused the problem - i.e.

-------
-4-
what changed from the early priority determination to the later
listing issues. Changing chemical toxicity/characteristics
during package development is an example of such a problem. HQ
recommends that the Regions develop draft HRS scores based on
several of the worst chemicals present. This would support both
having a worst site as well as likelihood of scoring above 28.5
in the NPL package. The Regions should also recognize the risks
of basing an HRS package on a single substance.
•	Several Regions (e.g. Region IV, VIII, and X) have
developed and implemented Regional geographical information
systems to correlate data on aquifers, populations, sensitive
environments, etc. This helps them screen sites for priority
assessments. This should be a fruitful area for cross-Regional
cooperation in how to incorporate environmental, geophysical, and
population data to better screen for worst sites likely to be
placed on the NPL.
•	Actual environmental contamination receives higher weight
in scoring than potential contamination under the revised HRS.
Since this puts priority on sampling and analysis results,
Regions should coordinate chemical analysis and sampling among
all the potential data users and do this early in the process.
The Delivery of Analytical Services (DAS) focus on a customer
service function for analytical services will greatly assist in
guiding the screening process to those sites with definitive data
on contaminated pathways.
•	SACM calls for an integrated assessment process, and we
have established HQ-Regional work groups for that purpose. One
workgroup is considering data quality needs of the assessment,
removal, and remedial processes. We will issue fact sheets later
this year to allow a better common understanding and common use
of data to support SACM implementation in FY 94. This should
encourage integrated decision-making, emphasis on worst sites,
and the ability to support either early or long-term response
action.
•	Cross-training of site assessment and removal staff is
also critical in promoting a consistent process for screening
sites for NPL caliber potential. Region IV has a one day
training course on each program to assist this integration. All
Regions need to address this need.
CONCLUSION
Our efforts to better define NPL caliber sites are
critically dependent on Regional experiences to date and our
interaction with our Regional clients. Our national conferences
and continuing SACM workgroups should continue to provide us with
that interaction.

-------
-5-
The major concern addressed here is how EPA will identify
the worst sites that are most worthy of either early action or
long- term action as part of SACM. Actions to support
identifying worst sites include: (1) focus on actual
contamination; (2) using more than one chemical for HRS scoring;
(3) use of GIS; (4) coordinating analytical data collection; and
(5) training across the programs. We should be moving towards
the same criteria for worst sites, regardless of whether we use
removal or remedial authorities to address them. Our efforts to
integrate EE/CA data requirements with HRS/RI/FS data
requirements should allow us to pursue both approaches
concurrently until we're sure which authority we want to use.
Attachment
ADDRESSEES:
Regional Waste Management Division Directors
ESD Directors
Superfund Branch Chiefs
cc: Tim Fields
Bruce Diamond
Sally Seymour
OERR Division Directors

-------
Integrated Assessment
No Further Action
Report
| Received/Iteleaae
Identified
i
S»mpl» (Optional)
No Further Action
k
Initial Screening
for Emergency
Qlgfbdjty
Bl« Search
(1/ ttae ilkswt)
J
Initial Field
lnveaHgatfcm
(Reeon)
Review
3»WDidelo« 01
Further Action
CUmIc
Emergency
Ketpnut
I

Complete the
Mend

Report
Do Removal
Time Critlcil
Removal?
Y«
I
Sampling
Needed?
Ye.
Definitely > 18.S
Integrated
Sampling
Plan
Si/Removal
Awetitntnt
S» TipUng Report
>28 S
I Do Removal

-------