Superfund: Half the Sites Have All Cleanup Remedies in Place or Completed
(Letter Report, 07/30/1999, GAO/RCED-99-245).

Pursuant to a legislative requirement, GAO provided information on the
Environmental Protection Agency's (EPA) Superfund Program, focusing on:
(1) the status in the Superfund cleanup process of the sites on EPA's
list of the nation's most hazardous sites, called the National
Priorities List (NPL); and (2) for the 609 NPL sites GAO included in its
review where cleanups have not been completed, what work has been
accomplished, what work remains, and when the remaining work will be
completed.

GAO noted that: (1) as of June 30, 1999, a majority of the 1,231 NPL
sites had progressed to the later steps in the Superfund cleanup
process; (2) at some 595 sites, the cleanup had been completed, all
remedies were in place to achieve cleanup, or there was no further need
for the site to be included on the NPL; (3) at another 424 sites, at
least one of the remedies had been selected, was under way, or had been
completed, according to EPA site managers; (4) at an additional 179
sites, no remedies had yet been selected; (5) however, managers reported
that short-term cleanup actions had addressed some of the contamination
at a majority of these sites; (6) at sites where cleanups were not
complete, a significant amount of work had already been accomplished,
and the remainder of the work was scheduled to be completed in the near
future; (7) two-thirds of the remedies selected for the 609 sites in
GAO's survey were under way or completed, the managers reported; (8)
these remedies had already addressed large portions of the contaminated
soil, solid material, and liquid wastes found at the sites; and (9) EPA
expected to complete work for a majority of the planned or ongoing
remedies for the 609 sites by 2002.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-99-245
     TITLE:  Superfund: Half the Sites Have All Cleanup Remedies in
	     Place or Completed
      DATE:  07/30/1999
   SUBJECT:  Environmental monitoring
	     Pollution control
	     Waste disposal
	     Environmental policies
	     Surveys
	     Hazardous substances
IDENTIFIER:  Superfund Program
	     EPA National Priorities List

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed product.                         **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************

Cover
================================================================ COVER

Report to the Ranking Minority Member, Committee on Commerce, House
of Representatives

July 1999

SUPERFUND - HALF THE SITES HAVE
ALL CLEANUP REMEDIES IN PLACE OR
COMPLETED

GAO/RCED-99-245

Status of the Superfund Program

(160438)

Abbreviations
=============================================================== ABBREV

  EPA -
  NPL -
  CERCLA -
  CERCLIS -
  GAO -

Letter
=============================================================== LETTER

B-283212

July 30, 1999

The Honorable John D.  Dingell
Ranking Minority Member
Committee on Commerce
House of Representatives

Dear Mr.  Dingell: 

Almost 20 years after the Congress established the Superfund
hazardous waste program, about half (595) of the 1,231 sites in the
program either are cleaned up or have the methods--remedies--in place
to achieve cleanup.  The Congress is now debating how much longer and
how many more resources it will take to select remedies and complete
cleanups at the remaining sites.  To assist in this debate, you asked
us to determine (1) the status in the Superfund cleanup process of
the sites on the Environmental Protection Agency's (EPA) list of the
nation's most hazardous sites, called the National Priorities List
(NPL),\1 and (2) for the 609 NPL sites we included in our review
where cleanups have not been completed, what work has been
accomplished, what work remains, and when the remaining work will be
completed.\2

To answer these questions, we analyzed data from EPA's Superfund
management information system.  In addition, for the 609 NPL sites
included in our review, we used Internet technology to administer a
survey to the responsible EPA site managers.  These sites were those
that EPA had not yet designated in its information system as
ï¿½construction complete.ï¿½ This term means that, for a particular site,
all selected remedies have been constructed or implemented, all
immediate risks have been addressed, and all long-term risks are
under control.  The results of our survey for each site are reported
at http://www.gao.gov/RCED-99-245/.  Because sites differ in size; in
the type, amount, and location of their contamination; and in the
number and types of remedies under way or selected, it is difficult
to respond to the second objective and summarize what work has been
accomplished and what work remains across sites.  For example, it is
difficult to measure and compare the percentage of contamination
addressed at two different sites through two very different
remediesï¿½at one, placing in the deed to the property a restriction
that limits its future use and at the other, using microorganisms to
biologically treat hundreds of cubic yards of contaminated soil. 
Therefore, we responded to the second objective by summarizing the
work accomplished and remaining by individual remedies rather than by
entire sites. 

--------------------
\1 We did not include sites that are on federal property and are
being managed by federal agencies other than EPA.  These sites are
known as federal facilities.  EPA had 156 such sites on the NPL as of
July 1, 1999, and had removed another 9 of these sites from the NPL
since the beginning of the program. 

\2 We did not include 27 sites because they were added to the NPL
after we began our review or were transferred to programs other than
Superfund for cleanup. 

   RESULTS IN BRIEF
------------------------------------------------------------ Letter :1

As of June 30, 1999, a majority of the 1,231 NPL sites had progressed
to the later steps in the Superfund cleanup process: 

  -- At some 595 sites, the cleanup had been completed, all remedies
     were in place to achieve cleanup, or there was no further need
     for the site to be included on the NPL. 

  -- At another 424 sites, at least one of the remedies had been
     selected, was under way, or had been completed, according to EPA
     site managers.\3

  -- At an additional 135 sites, no remedies had yet been selected;
     however, managers reported that short-term cleanup actions had
     addressed some of the contamination at a majority of these
     sites.\4

At sites where cleanups were not complete, a significant amount of
work had already been accomplished, and the remainder of the work was
scheduled to be completed in the near future: 

  -- Two thirds of the remedies selected for the 609 sites in our
     survey were under way or completed, the managers reported. 

  -- These remedies had already addressed large portions of the
     contaminated soil, solid material, and liquid wastes found at
     the sites. 

  -- EPA expected to complete work for a majority of the planned or
     ongoing remedies for the 609 sites by 2002. 

--------------------
\3 For 44 sites, our survey data showed that EPA had signed a record
of decision on the remedies selected; however, EPA cleanup managers
did not provide data on the type of remedies chosen or their
progress. 

\4 Of the remaining 33 sites, we included 6 in our survey but did not
receive enough data on them to categorize their status.  We excluded
4 because they were transferred to programs other than Superfund and
23 because they were added to the NPL after we initiated our survey. 

   BACKGROUND
------------------------------------------------------------ Letter :2

The Congress passed the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA) in 1980, thereby
establishing the Superfund program to clean up highly contaminated
hazardous waste sites.  To decide whether to include a site on the
NPL, EPA uses its hazard ranking system to review available data for
the site and determine whether it presents high enough health or
environmental risks to qualify for a long-term cleanup under CERCLA. 
If so, and the relevant state environmental agency agrees, EPA will
include the site on the NPL.  Under CERCLA, parties responsible for
the contamination at an NPL site are liable for conducting the
cleanup.\5 If these parties do not initially agree to do so, EPA can
fund the cleanup and subsequently compel the parties to reimburse the
agency for its costs. 

Once a site is on the NPL, EPA or responsible parties, with EPA's
oversight, generally conduct two studies--an extensive investigation
of the risks the site poses and an evaluation of alternative remedies
to address these risks.  Figure 1 outlines the steps in the Superfund
cleanup process. 

   Figure 1:  Superfund Cleanup
   Process

   (See figure in printed
   edition.)

Source:  EPA. 

After these studies have been completed, EPA selects one or more
cleanup actions, called remedies, to conduct at the site, such as
excavating and treating contaminated soil.  An individual site can
have from 1 to more than 50 remedies.  EPA formally documents its
selection in one or more records of decision.  Then, either EPA or
responsible parties design and construct the cleanup methods and
technologies.  When the construction is finished and all risks are
under control, EPA classifies the site as construction complete. 
This category also includes sites whose cleanup actions do not
involve actual construction, such as sites that use a deed
restriction.  Finally, this category may include sites where
long-term cleanup actions have been implemented but cleanups are not
yet complete.  For example, a system may have been constructed to
pump and treat groundwater, but parties will have to operate it for
more than 30 years until contaminants in the water are reduced to
acceptable levels.  Once EPA, in consultation with the state in which
the site is located, has determined that the work at the site has
achieved the desired cleanup goals, the site can be deleted from the
NPL. 

--------------------
\5 Responsible parties may include, among others, waste generators,
waste haulers, and site owners and operators. 

   AT HALF THE SITES, CLEANUPS HAD
   BEEN COMPLETED OR ALL REMEDIES
   WERE IN PLACE TO ACHIEVE
   CLEANUP
------------------------------------------------------------ Letter :3

As of June 30, 1999, EPA had classified about half, or 595, of the
1,231 sites placed on the NPL since the beginning of the program as
either construction complete (419)ï¿½all remedies were in place and all
risks under controlï¿½or as no longer needing to remain on the NPL
(176).  Figure 2 highlights the status of the sites that were placed
on the NPL. 

   Figure 2:  Status of NPL Sites
   in the Superfund Cleanup
   Process

   (See figure in printed
   edition.)

Note:  The shaded area represents sites (1) where cleanups had been
completed or all remedies were in place to achieve cleanup or (2)
that had been removed from the NPL. 

Source:  EPA and GAO's survey data. 

The remaining half, or 636 sites, had progressed through varying
steps in the Superfund cleanup process, according to EPA site
managers.  At 54 sites, all remedies selected to date had been
completed.  Some of these sites may have additional remedies
implemented in the future, and some may soon be categorized as
construction complete.  Another 370 sites had progressed to the later
steps of the Superfund cleanup process and had at least one remedy
selected, being implemented, or completed.  For 44 sites, our survey
data showed that EPA had signed a record decision on the remedies
selected; however, EPA cleanup managers did not provide data on the
type of remedies chosen or their progress.  Data in EPA's Superfund
managment information system indicate that remedies are or will soon
be under way at some of these sites.  Some 135 sites had no remedies
selected; however, 115 of them had progressed to the step in the
cleanup process that involves study to identify risks and appropriate
cleanup alternatives, and 112 had undergone short-term cleanup
actions to address some contamination.\6 Finally, 33 sites had
recently been added to the NPL or transferred to other programs for
cleanup. 

Two factors are helping EPA complete the cleanup process at more
sites now than in the past.  First, the majority of NPL sites have
been in the Superfund program for several years and have moved to the
later steps in the cleanup process.  In March 1997, we reported that,
on average, it was taking more than 10 years to construct the
remedies at Superfund sites.\7 More recently, EPA stated that for
sites added to the NPL in the 1990s, it was taking an average of 8
years to reach this milestone.  Second, we also reported that since
1994, the agency's priority has been to finish cleaning up sites that
are already in the program rather than add new ones.  Consequently,
the agency has moved funds from site assessment to construction. 

--------------------
\6 Short-term cleanup actions called removals are also undertaken at
sites to mitigate immediate and significant threats, such as those
stemming from contaminated drinking water or unrestricted access to
sites.  These actions are generally of a short-term and emergency
nature, such as providing alternative drinking water supplies and
cleaning up chemical spills caused by transportation accidents. 

\7 Superfund:  Times to Complete the Assessment and Cleanup of
Hazardous Waste Sites (GAO/RCED-97-20, Mar.  31, 1997). 

   AT SITES WHERE CLEANUPS WERE
   NOT COMPLETE, TWO-THIRDS OF THE
   REQUIRED WORK WAS UNDER WAY OR
   DONE
------------------------------------------------------------ Letter :4

As of June 30, 1999, EPA and private parties had selected a total of
2,046 remedies to construct and implement across the 609 sites in our
survey.\8 Site managers reported that construction or implementation
was complete for 617 of these remedies, was under way for another
617, and had not begun for 724.  Managers did not provide the data
needed to categorize the 88 remaining remedies.  About a third of the
remedies address contaminated soil; another third address
contaminated groundwater; and the remainder address contaminated
solid wastes, such as wastes generated from processing raw materials;
sediments; surface water; debris; sludge; leachate (runoff); liquid
waste, such as wastewater generated from industrial processes; and
air. 

To gauge the environmental impact of these cleanup actions to date,
it is important to consider not only the number but also the type of
remedies implemented.  CERCLA establishes a preference for remedies
that permanently treat the contamination and thus help to eliminate
the risks it poses.  According to our survey results, only 7 percent
of the soil remedies but 39 percent of the groundwater remedies
permanently treat the contamination.  Overall, more than half of
these treatment remedies were under way or completed.  About another
half of the soil remedies and another quarter of the groundwater
remedies are classified as support actions because they make it
easier to treat the contamination.  For example, directional wells
provide a treatment technology with better access to contaminated
groundwater, and soil washing extracts contaminants so that they can
be treated.  About two-thirds of both the soil and groundwater
remedies that support treatment were also under way or completed. 
Table 1 shows the status of various types of remedies, by
environmental medium. 

                                         Table 1
                         
                         Type and Status of Selected Remedies, by
                           Environmental Medium, as of June 30,
                                           1999

                                          Status of remedy
                     ----------------------------------------------------------
                          Selected           Under way           Completed
                     ------------------  ------------------  ------------------
Medium/type of                 Percent\                                             Total
remedy                 Number         a    Number   Percent    Number   Percent  number\a
-------------------  --------  --------  --------  --------  --------  --------  --------
=========================================================================================
Soil                      227        35       179        27       251        38       657
Permanent treatment        16        34        19        40        12        26        47
Support actions\b          98        35        69        25       110        40       277
Disposal                   23        32        14        19        36        49        73
Containment                41        42        31        32        26        27        98
Stabilization              18        50        10        28         8        22        36
Institutional              23        34        14        21        31        46        68
 controls
Other\c                     8        14        22        38        28        48        58
=========================================================================================
Groundwater               279        44       211        33       141        22       631
Permanent treatment       116        47        88        36        41        17       245
Support actions            58        36        59        38        41        26       157
Monitoring                 35        58        23        37         4         6        62
Containment                14        37         9        24        15        39        38
Discharge                   3        27         7        64         1         9        11
Institutional              24        69        10        29         1         3        35
 controls
Other                      30        36        15        18        38        46        83
=========================================================================================
Other media\d             215        33       212        32       226        35       653
=========================================================================================
Total                     721       100       602       100       618       100     1,941
-----------------------------------------------------------------------------------------
Note:  Some percentages do not total 100 because of rounding. 

\a We did not receive sufficient data to categorize 105, or 5
percent, of the remedies included in this table. 

\b These include actions that support the implementation of treatment
remedies. 

\c These include numerous other approaches selected to address
contamination, such as relocating populations and providing
alternative drinking water supplies. 

\d These include all other media addressed by these actions. 

Source:  GAO's survey data. 

Other remedies leave the contamination in place but reduce its risks
by containing or stabilizing it.  For example, a layer of impermeable
clay may be placed over contaminated soil so that the contaminants do
not migrate.  Of such remedies, about 56 percent for soil and 63
percent for groundwater were under way or completed.  Still other
remedies, called institutional controls, limit the risk from exposure
to contaminants but do not involve construction.  These include a
wide range of options, from fencing, to deed restrictions, to
restrictions on recreational uses.  Of such remedies, about
two-thirds for soil and one-third for groundwater were under way or
completed. 

As another measure of accomplishment for some cleanups, we were able
to obtain information on the amount of contamination addressed.\9 We
obtained such information for a number of remedies that addressed
contamination in soil, other solid material, and some liquids. 
Because EPA site managers provided this information in different
measures, such as cubic yards and tons, we could not calculate
overall how much of a particular medium had been addressed or
remained to be addressed.  However, site managers estimated that more
than half of 9.6 million cubic yards of soil needing to be excavated
had been excavated, as well as most of an additional 1.4 million tons
of soil.  While treatment had been completed for only 12 percent of
3.2 million cubic yards of soil, almost 89 percent of another 0.6
million tons of soil had been treated.  Figures 3 and 4 show EPA site
managers' estimates of how much soil and other solid material had
been excavated, treated, and contained and how much remained to be
addressed.  These estimates, which the managers provided when data
were available, are expressed in cubic yards and in tons. 

   Figure 3:  Cubic Yards of Soil
   and Other Solid Material
   Excavated, Treated, and
   Contained

   (See figure in printed
   edition.)

Source:  GAO's survey results. 

   Figure 4:  Tons of Soil and
   Other Solid Material Excavated
   and Treated

   (See figure in printed
   edition.)

Note:  In addition, 4,000 tons of soil were contained and 5,000 tons
of other solid material remain to be contained. 

Source:  GAO's survey results. 

In addition, according to our survey results, more than half of 6,350
acres needing to be contained had been addressed, and almost 60
percent of 20.5 billion gallons of liquid wastes, other than
groundwater, needing treatment had also been addressed. 

--------------------
\8 For our universe of sites, we initially included any nonfederal
NPL site that had not been designated as construction complete as of
Aug.  1998.  We subsequently deleted any site that had reached this
designation by June 30, 1999.  This resulted in a total survey
universe of 609 sites.  Since Aug.  1998, EPA has added another 23
sites to the NPL that we did not include in our survey. 

\9 EPA site managers could not provide us with data on the amount of
contamination addressed by all remedies.  For example, they could not
provide data on the volume of contaminated groundwater addressed
because they could not estimate how many gallons of water might exist
in large underground aquifers.  Furthermore, systems may have to pump
out, treat, and return groundwater to an aquifer for 30 or more years
until the level of contaminants is low enough to meet cleanup
standards.  Therefore, the number of gallons of groundwater pumped is
not a meaningful measure of accomplishment for groundwater cleanups. 

   SHORT-TERM CLEANUP
   ACCOMPLISHMENTS
------------------------------------------------------------ Letter :5

Besides these longer-term cleanup remedies, EPA and responsible
parties had implemented shorter-term removal actions at 62 percent of
the sites in our survey, further addressing contamination.  EPA and
parties had completed 1,596 removal actions and were implementing 321
more actions.  Several of these removal actions were restrictions
imposed on the use of a site to reduce the risk of exposure to
contamination.  However, a number of these actions had also removed a
large amount of contamination and, in some cases, had treated it. 
For example, our survey data show that removal actions treated about
2.5 billion gallons of liquid wastes, excavated 2.6 million cubic
yards of contaminated soil, and treated another 0.2 million cubic
yards of soil, helping to reduce the risks of exposure to
contamination. 

   EXPECTED COMPLETION DATES FOR
   REMAINING WORK
------------------------------------------------------------ Letter :6

According to our analysis of our survey results, EPA site managers
estimated that construction or implementation would soon be complete
for most of the remedies that had been selected or were in progress. 
See figure 5. 

   Figure 5:  Expected Completion
   Dates for Construction of
   Remedies

   (See figure in printed
   edition.)

Note:  Of a total of 2,046 remedies in our survey, EPA cleanup
managers provided the expected completion dates for 1,314 of the
1,341 remedies that were planned or under way.  Managers reported
another 617 remedies were completed and did not provide sufficient
data to categorize the 88 remaining remedies. 

Source:  GAO's survey data. 

EPA site managers expected the majority of the remedies that had been
selected or were under way to be constructed or implemented by the
end of fiscal year 2002.  In addition, they expected 82 percent of
the selected remedies to be completed by the end of fiscal year 2005
and 87 percent of them to be completed by the end of fiscal year
2010.  Similarly, they expected a majority of the remedies that treat
and thus directly address contamination in soil and groundwater to be
constructed or implemented by fiscal year 2003.  In general, the
remedies expected to extend beyond fiscal year 2005 addressed
contaminated groundwater (e.g., monitoring it, pumping and treating
it, and allowing the contamination in it to naturally attenuate, or
thin out, over a period of years) or required long-term operations
and maintenance. 

Cleanup activity in the Superfund program will not cease after the
remedies in our survey have been constructed or implemented.  Either
responsible parties or states will need to continue operating a
portion of the remedies--some for 30 years or more--until all
contamination has been addressed as required in the records of
decision for the applicable sites.  In addition, CERCLA requires EPA
to periodically monitor the effectiveness of remedies where
contaminants are left on-site.  EPA also has to monitor cleanups by
private parties to ensure that settlement agreements reached with
them are being implemented.  Furthermore, sites that are currently in
the program's early stages will require cleanup work after EPA
selects remedies for them.  For example, EPA site managers estimated
that 88 percent of any new records of decision they could identify
for the sites in our universe would be signed no later than the end
of fiscal year 2001.  While some of these decisions will find that no
further action is required because all contamination has been reduced
to acceptable levels of risk, other decisions will select new
remedies that will take several more years to design and implement. 
Additionally, EPA recently placed another 23 sites on the NPL that we
did not include in our survey; these sites will entail some number of
additional records of decision and new remedies to implement. 

Finally, EPA will also add some sites to the NPL in the future,
further extending the cleanup workload.  How many sites EPA will add
is difficult to predict, especially now that the states are assuming
responsibility for more sites.  In November 1998, for example, we
reported that EPA and the states still had to negotiate whether the
federal or a state government would manage cleanups for another
several hundred sites awaiting cleanup decisions.\10 In April 1999,
we also reported that the states are now better able to manage
cleanups and often choose to do so rather than refer sites to EPA for
Superfund consideration.\11 As evidence of this trend, EPA,
responding to a 1995 congressional request, began requiring its
regions to obtain the appropriate governor's concurrence before
adding a site to the NPL.  We reported that as of February 1999,
governors had opposed the listing of 31 sites and supported the
listing of 123.  Because of this increased state involvement, EPA
expects to add fewer sites to the NPL each yearï¿½at most, about 40
sites per year, compared with 76, on average, from the late 1980s to
the early 1990s. 

--------------------
\10 Hazardous Waste:  Information on Potential Superfund Sites
(GAO/RCED-99-22, Nov.  30, 1998). 

\11 Superfund:  Progress Made by EPA and Other Federal Agencies to
Resolve Program Management Issues (GAO/RCED-99-111, Apr.  29, 1999). 

   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :7

To help ensure the validity and reliability of our data, we designed
our survey so that when an EPA manager accessed it, data available in
the agency's Superfund management information system would
automatically be placed in the appropriate response boxes.\12 Such
data included background information on a site, the types of remedies
that had been selected or were under way, and the actual or projected
dates of completion for these remedies.  We asked each manager to
verify the accuracy of these data and to provide original data in
response to any remaining questions, particularly for questions on
the amount of contamination addressed, since these data are not
included in EPA's information system.  We also asked each manager to
indicate the source of the original data provided, such as formal
site records or personal knowledge.  We did not independently verify
the accuracy of the data the managers reported to us because the time
and cost to do so would have been prohibitive for the large number of
sites in our survey.  We conducted our review from April 1998 through
July 1999 in accordance with generally accepted government auditing
standards.  Appendix I includes a more detailed discussion of our
survey's scope and methodology. 

--------------------
\12 This system is the Comprehensive Environmental Response,
Compensation, and Liability Information System, version 3, more
commonly known as CERCLIS 3. 

   AGENCY COMMENTS
------------------------------------------------------------ Letter :8

We provided a copy of the information on each site to the respective
EPA site manager and asked each manager to provide us with any
necessary changes.  We were able to update the data for 93 percent of
the sites in our survey.  We were unable to update the data for the
remaining sites because we did not receive responses from the
managers.  In addition, we met with officials in EPA's Office of
Emergency Response and Remediation who manage the Superfund program,
including the Director of the Planning, Analysis, and Resources
Management Center.  We reviewed the results of our summary analyses
with these officials.  In general, they were appreciative of the
amount of data we had collected on the program and said that the
information would help them as they assessed the future of the
program. 

EPA officials made several general points about the results of our
summary analyses.  First, they recommended we emphasize that the
dates EPA site managers provided to us represent the dates the site
managers expect the remedies to be constructed or implemented, not
the dates they expect the cleanups themselves to be complete.  The
EPA officials believed it was important to make this distinction so
that decisionmakers would recognize that additional time would be
required to complete cleanups.  We revised our report to emphasize
this point.  Second, the officials acknowledged that, for the
percentage of remedies that permanently treat contamination, the
agency had publicly reported a higher rate than we determined through
our survey.  According to the officials, this difference occurred
because EPA's rate includes both treatment remedies and support
actions taken to make treatment easier.  We did not categorize
support actions as treatment remedies because support actions only
temporarily manage the contamination. 

---------------------------------------------------------- Letter :8.1

As arranged with your office, unless you publicly announce its
contents earlier, we plan no further distribution of this report
until 30 days after the date of this letter.  At that time, we will
send copies to Senator Max Baucus, Senator Christopher S.  Bond,
Senator John Chafee, Senator Frank Lautenberg, Senator Barbara
Mikulski, and Senator Robert C.  Smith and to Representative Thomas
Bliley, Representative Sherwood L.  Boehlert, Representative Robert
A.  Borski, Representative Alan B.  Mollohan, Representative James L. 
Oberstar, Representative Michael G.  Oxley, Representative Bud
Shuster, and Representative James T.  Walsh in their capacities as
the Chairs and Ranking Minority Members of the Senate and House
Committees and Subcommittees with jurisdiction over the Superfund
program.  We are also sending copies of this report to Carol M. 
Browner, Administrator, EPA.  Copies will also be made available to
others upon request. 

If you or your staff have any questions, please contact me at (202)
512-6111.  Key contributors to this report are listed in appendix II. 

Sincerely yours,

David G.  Wood
Associate Director, Environmental
 Protection Issues

OBJECTIVES, SCOPE, AND METHODOLOGY
=========================================================== Appendix I

At the request of John D.  Dingell, Ranking Minority Member of the
House Committee on Commerce, we were to determine (1) the status in
the Superfund cleanup process of the sites on the Environmental
Protection Agency's (EPA) list of the nation's most hazardous sites,
the National Priorities List (NPL),\1 and (2) for the 609 NPL sites
we included in our survey where cleanups have not been completed,
what work has been accomplished, what work remains, and when the
remaining work will be completed. 

To answer these questions, we obtained and analyzed EPA's data on
nonfederal NPL sites.  We used this information to determine the
total number of sites that (1) have been placed on the NPL since the
beginning of the program, (2) have been deleted from the list, and
(3) have completed the construction or implementation of all
necessary remedies and are candidates for deletion. 

In addition, we developed an automated survey instrument that we sent
to EPA site managers to obtain information on the remaining NPL sites
that EPA had not designated as ï¿½construction completeï¿½ as of August
1998.  The survey comprised several sections of questions that
addressed

  -- the background of the site, including its name, size, and
     location; information on prior activities that caused
     contamination at the site; the contaminants present; and the
     condition of the site at the time EPA added it to the NPL;

  -- the status of the site in the Superfund cleanup process, such as
     the estimated or actual dates of completing risk and feasibility
     studies, deciding on the cleanup actions to be taken, or
     designing the cleanup methods to be used; and

  -- the nature and extent of any short-term (removal) and long-term
     (remedial) cleanup actions taken at the site and the parties
     conducting these actions. 

To better understand the cleanup process and the types of data
available to respond to our objectives, we met with cleanup managers
in EPA's Office of Emergency Response and Remediation, who are
responsible for the cleanup program and CERCLIS 3, and with Superfund
cleanup managers from EPA's Region V.  We also talked to officials at
the U.S.  Army Corps of Engineers who manage cleanup work to obtain
information on available technologies.  We reviewed supporting
documentation, including Superfund manuals, CERCLIS 3 documentation,
and information from EPA on the various approaches available for
cleaning up sites. 

Once we designed the survey questions, we conducted two pretests,
making necessary revisions.  We administered the pretests to EPA
regional staff responsible for managing cleanups at individual sites,
called remedial project managers, in regions I and V. 

We then designed an electronic survey instrument that would be posted
on GAO's home page on the Internet.  As part of this design and at
EPA's request, we obtained available data from CERCLIS 3 for each
site that provided answers to some of our survey questions, including
data on the site's background and on the types and dates of cleanup
actions.  CERCLIS 3 does not contain information on the amount of
contamination addressed.  We designed the survey so that when a site
manager accessed it electronically, any relevant CERCLIS data would
appear in the appropriate response boxes.  To accomplish this
electronic link, we worked with EPA staff responsible for maintaining
CERCLIS 3.  We also worked on-site with the contractor that EPA uses
to support this database.  We asked each manager to verify that the
CERCLIS 3 data were accurate and to make any necessary corrections
and additions. 

To ensure security and data integrity, we provided each manager
responsible for sites in our universe with a password that would
allow the manager to access and complete a survey for each of the
manager's sites.  No one else could access that survey or edit its
data.  Also, after transmitting a completed survey to GAO, the
manager could not change any of the data but could ask us to make any
necessary changes.  We conducted 11 pretests with site managers in
regions III, IV, and V to ensure that the managers could easily
access and complete the surveys and transmit the data electronically. 

We made the survey accessible to managers on April 6, 1999.  We
designed the survey so that if a respondent had a question, that
person could immediately send us an electronic message and we would
provide an answer.  We also handled a number of phone calls from
respondents to assist them in completing the survey. 

We initially included in our review any site on the NPL that, as of
August 1998, EPA had not designated as construction complete. 
However, we eliminated many sites from our universe after learning
through our survey responses that, for example, a site had been
completed after our cutoff date or had been referred to another
cleanup program.  Therefore, we are reporting on a final universe of
609 NPL sites. 

We did not independently verify any of the data that the managers
provided because the time and costs required to do so would have been
prohibitive.  However, we did ask the managers to identify their
sources for the different types of data they provided.  For example,
we asked the managers to indicate which of the following sources they
used to supply cleanup data: 

  -- the signed record of decision for the site,

  -- formal status reports generated during the cleanup,

  -- formal reports submitted to verify the completion of a cleanup
     method's construction,

  -- the CERCLIS 3 database,

  -- other EPA personnel with knowledge of the site, and/or

  -- personal knowledge. 

To ensure the consistency and accuracy of our data, we sent a
completed table of survey results for each site to the relevant
manager and asked the manager to review the table to make sure that
it accurately represented the data that the manager had provided.  We
received responses for 568 of the 609 sites in our survey, or 93
percent.  We revised these tables and our database as appropriate
following the managers' reviews.  Therefore, our site data are
current as of June 30, 1999. 

We conducted our review from April 1998 through July 1999 in
accordance with generally accepted government auditing standards. 

--------------------
\1 We did not include sites that are on federal property and are
being managed by federal agencies other than EPA.  These sites are
known as federal facilities.  EPA had 156 such sites on the NPL as of
July 1, 1999, and had removed another 9 of these sites from the NPL
since the beginning of the program. 

GAO CONTACTS AND STAFF
ACKNOWLEDGMENTS
========================================================== Appendix II

GAO CONTACTS

David G.  Wood (202) 512-6111
Eileen Regen Larence (202) 512-6510

ACKNOWLEDGMENTS

In addition to those named above, Alice Feldesman, Mitch Karpman,
Pauline Lichtenfeld, Lynn Musser, Judy Pagano, Steve Palincsar, and
Derek Updegraff made key contributions to this report. 

*** End of document. ***