vvEPA
             United States
             Environmental Protection
             Agency
            Office of Emergency and
            Remedial Response
            Washington DC 20460
EPA540P-90.0Q1
May 1990
             Superfund
Field Test of the Proposed
Revised Hazard Ranking
System (MRS)

-------
                        EPA/540/P-90/001
                        May 1990
      FIELD TEST OF THE
   PROPOSED REVISED HRS
      Site Assessment Branch
 Hazardous Site Evaluation Division
U.S. Environmental Protection Agency
          August 18,1989

-------
                              Notice

This document has been reviewed in accordance with the U.S. Environmen-
tal Protection Agency's peer and administrative  review policies and ap-
proved for publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
                                11

-------
                                  EXECUTIVE SUMMARY
 The Superfund Amendments  and Reauthorization Act  of 1986 (SARA) requires  the U.S.
 Environmental Protection Agency (EPA) to revise the Hazard Ranking System (HRS) so that, to the
 maximum extent feasible, it accurately assesses the relative risks associated with actual or potential
 releases of hazardous substances from a site.  The Conference  Report on SARA (H. Rep. 962, 99 Cong.,
 2d Sess. at 199-200 (1986)) makes clear that this mandate does not require detailed risk assessments,
 but directs EPA to rank sites as  accurately as feasible based  simply  on information available from
 preliminary assessments and site inspections  consistent with the goal of "expeditiously" identifying
 candidates for response actions. In addition, SARA specifically directs EPA to:

      •   Assess how surface water contamination affects the human food chain and recreational
          use of surface water.

      •   Consider potential contamination of ambient air as well as actual contamination.

      •   Give a high priority to sites which have contaminated principal drinking water supplies.

      •   Consider the quantity, toxicity,  and concentrations of hazardous constituents in wastes
          generated primarily by combustion of coal or other fossil fuels (e.g., fly ash, bottom ash,
          slag waste).

 In response to these mandates, EPA  proposed HRS revisions  for public comment on December 23,
 1988 (53 FR 51962).   The HRS is used to assess the relative  risks posed by releases or threatened
 releases of hazardous substances from  a site, and  as a means of identifying releases as national
 priorities for further investigation  and  possible remediation.  The HRS assigns  numerical values
 (according to prescribed rules) to factors that  characterize the  potential of any given release to  cause
 harm to public health or the environment. The values are then combined and used to yield a single
 score that is designed to indicate the potential hazard posed by each site relative to other sites.

 EPA initiated a field test of the proposed HRS revisions1 to help assess the costs and implementation
 concerns associated with the modifications. The field test had several major objectives:

      •  To  test the feasibility  of implementing the new and expanded proposed revised HRS
         factors.

      •  To determine the resources required (i.e., costs and technical hours) for specific tasks under
         the proposed revised HRS.

      •  To assess the availability of information that would  be needed for the evaluation of sites
         with the proposed revised HRS and to identify difficulties with its use .

The Agency tested the proposed revised HRS by performing inspections at 29 sites nationwide.  Sites
were not randomly selected, but were  primarily  selected with  characteristics that would  help
evaluate the proposed new components of the HRS.
'In actuality, a "draft" version of the proposed  revised HRS was tested.  Differences between the
"draft" proposed revised HRS and the proposed revised HRS are slight, and essentially all components
of the proposed revised HRS were tested. The differences are described in Section 1.5.
                                            111

-------
Upon completion of the inspections, field test participants prepared scoring packages for each site
using the proposed  revised MRS instructions.  EPA HQ staff reviewed these scoring  packages and
visited  each EPA  Regional office to exchange information on implementation concerns, data
collection methodologies, and scoring issues associated with the proposed MRS revisions. Other field
test materials (e.g., planning documents and project-related reports) were also used to obtain test
results.

Generally, the field  test achieved  its stated objectives.  The proposed  MRS  revisions generally
functioned as anticipated and could be evaluated using an appropriate level of effort (LOE).  Data to
support the  various proposed revised MRS  factors usually  could be collected using information
obtained from the site inspections. The primary purpose of this report is to discuss those findings or
issues regarding the proposed revised HRS that were brought forward during the field test.

Project participants noted that many of the  proposed  revisions represented improvements over the
current HRS and should allow for more complete and accurate relative assessments of the risks posed
by sites. These improvements include the following:

      •  Hazardous waste quantity.  The tiered  approach to evaluating this  factor provides new
         mechanisms for evaluating hazardous waste quantity when complete information on the
         quantities of hazardous substances or hazardous wastes is known.  New mechanisms are
         also available for estimating hazardous waste  quantity in the absence of complete
         information on the types and amounts of hazardous substances deposited.

      •  Potential releases to the air pathway.  The proposed air  pathway provides  a means of
         evaluating the threat posed by potential releases of contaminants in addition to observed
         releases.

Several other proposed revisions were also identified as improvements,  but were thought by project
participants to warrant some additional review prior to final rule-making. Among these are:

      •  Distance/dilution weighting of targets.   The attenuation  of contaminants in the
         environment is reflected in the application of distance and  dilution weights applied to
         potentially contaminated  targets.  Dilution weighting  factors as  currently proposed,
         however, may not accurately represent the degree of contaminant dilution in large water
         bodies.  Also, study results suggest that the  distance weighting  factors applied to nearby
         populations under the onsite exposure pathway's nearby target population factor may
         require  additional review.  This factor  often scored very  high and may not realistically
         reflect the propensity of nearby populations to come into contact with contaminants at a
         site.

      •  Human food chain threat.  The proposed surface water pathway provides mechanisms for
         evaluating the threat to humans  posed by  actual and potential contamination of the
         aquatic human food chain.  The human food chain target population  factor was found to
         be generally difficult to evaluate and often resulted  in disproportionately high human
         food chain threat scores, particularly for sites located near the coast or near water bodies
         with low average flows. The target distance limit, dilution weighting factor, and human
         food chain production factor contributed to the difficulties associated with evaluating
         human food chain threat targets.

      •  Human  recreation threat targets.   The  proposed  surface water pathway also provides a
         means of evaluating the threat posed by actual and  potential  contamination of human
         recreation areas.  The human recreation target population factor was generally difficult
         and very time consuming to  evaluate.  This factor appeared to be disproportionately
         difficult to evaluate relative to its impact on scores.  The target distance limits for
                                            IV

-------
          evaluating  recreation  use  population, which  are  determined  from  the
          accessibility/attractiveness factor, contributed to the problems associated with evaluating
          human recreation threat targets.

      •   Onsite exposures. The proposed onsite exposure pathway evaluates the threats associated
          with direct, physical contact with hazardous wastes or  contaminated  soil.   Resident
          populations, nearby populations, and terrestrial sensitive  environments are the targets
          which may be affected by such contamination.  While project  participants felt that the
          addition of this  pathway does much  to  improve the evaluation of the  relative  risks
          associated with sites, some difficulties in scoring the pathway were noted.  Some project
          participants had  difficulty in documenting the total area of surficial  contamination at a
          site, identifying contaminated resident population properties, and evaluating both high-
          risk and total resident populations.

 Project participants also recommended that simplification of the proposed revised HRS be pursued,
 particularly in terms of the instructions for scoring factors; project participants often had difficulty in
 interpreting the requirements of the proposed revisions.  For example, while the tiered approach to
 evaluating hazardous waste quantity was viewed as a  more effective use of data, hazardous waste
 quantity was perhaps the single most difficult factor to evaluate.

 The costs associated with data collection, sampling,  evaluation, and administrative requirements in
 support of revised HRS scoring were found to vary widely among sites.  These costs are summarized  by
 pathway in Table ES-1. Note that these costs represent the entire sequential process of pre-remedial
 site evaluation, that comprehensive evaluations were performed for all pathways at most sites, and
 that the sites themselves were primarily selected for specific characteristics of interest from the
 perspective of field testing.  As such, these costs may not be representative of the costs associated
 with site inspections occurring on the greater universe of CERCLA sites. Costs are discussed in detail in
 Section 5.

 Finally, scores for the field  test sites were compared under both the  proposed revised HRS and the
 current  HRS.  Scoring results are discussed in Section 6. Significant scoring results include the
 following:

      •  Under the proposed  revised HRS, surface water tended to be the highest scoring  pathway
         for the field test sites.  Under the current HRS, the  ground water pathway tended to score
         highest.

      •  Proposed revised  HRS surface water pathway scores were usually dominated by the human
         food chain threat.

      •  Surface water and air pathway scores were generally  higher using the proposed revised
         HRS than with the current HRS.

      •  Ground water pathway scores were generally lower with the proposed revised HRS than
         with the current HRS.

      •  Overall site scores for the field test sites were generally higher under the proposed revised
         HRS than under the current HRS.

The ability to extrapolate these results to the greater universe of CERCLA sites is limited because the
field test sites were primarily selected to test specific  components of the proposed  revised HRS.
However, these results do provide a useful measure of how actual  environmental data perform
within the framework of the proposed revised HRS.
                                            v

-------
                                  TABLE ES-1
                          SUMMARY OF SITE COSTS
                          (Based on Field Test Sites)
                                    LOE Hours         Dollars        CLP Samples
                                     Range            Range           Range
 General Tasks                        480 -1,500     $39,000 - 84,000       3-17
 Site/Source/Waste Characterization      80-530      $20,000-79,000       10-46
 Air Pathway                          20-270        $900-44,000        0-23
 Ground Water Pathway               50 -1,360     $3,000 -156,000       0 - 40
 Surface Water Pathway                10-290        $500-42,000        0-28
 Onsite Exposure Pathway               10-340        $300-32,000        0-15

 Total Site*                           970-3,310    $100,000-311,0000     34-98
*  Represents ranges of actual total site costs for field test sites, not the sum of individual
   line items.
Note:  These values are based upon findings from field test sites that were primarily
       selected to test specific features of the proposed MRS.  As such, these values are not
       necessarily representative of the greater universe of CERCLA sites.
                                      VI

-------
                                 CONTENTS


SECTION                                                                   PAGE

1   INTRODUCTION  	 1-1
    1.0    Background  	
    1.1    SARA and the MRS  	
    1.2    The Current MRS  	
    1.3    The Proposed Revised MRS
    1.4    Design of the Field Test ..
    1.5    Model Version Used in the Field Test Project
-1
-2
-3
-3
-8
-9
    1.6   Outline for Remainder of Report  	  1-9
2   METHODOLOGY  	 2-1

    2.0    Introduction  	 2-1
    2.1    Project Management and Specific Responsibilities  	 2-1
    2.2    Site Selection  	 2-1
    2.3    Site Inspections  	 2-3
    2.4    Initial Proposed MRS Scoring Packages  	 2-4
    2.5    Other Field Test Work Products  	 2-5
    2.6    General Schedule  	 2-6
3   CROSS-CUTTING ISSUES AND FINDINGS  	  3-1

    3.0   Introduction  	  3-1
    3.1   Identification and Characterization of Sources  	  3-1
    3.2   Evaluation of Toxicity, Mobility, and Persistence Factors  	  3-5
    3.3   Hazardous Waste Quantity  	  3-14
    3.4   Sampling Strategy	  3-19
    3.5   Data Significance  	  3-22
    3.6   Evaluation of Actual and Potential Proposed
         HRS Targets   	  3-23
4  INDIVIDUAL PATHWAY ISSUES AND FINDINGS 	  4-1

   4.0   Introduction  	  4-1
   4.1   Proposed Air Pathway  	  4-1
   4.2   Proposed Ground Water Pathway  	  4-2
   4.3   Proposed Surface Water Pathway 	  4-6
   4.4   Proposed Onsite Exposure Pathway  	  4-13
                                     vn

-------
                          CONTENTS (concluded)
SECTION
5  COST ANALYSIS
   5.0   Introduction  	  5-1
   5.1   Methodology  	  5-1
   5.2   Database Development  	  5-2
   5.3   Discussion  	  5-3
   5.4   Summary of Site Costs  	  5-26
6  FIELD TEST MRS SCORING RESULTS 	  6-1

   6.0   Introduction  	  6-1
   6.1   Field Test Scores Under the Proposed HRS  	  6-2
   6.2   Comparison of Proposed and Current HRS Scores  	  6-12
   6.3   Scoring Summary  	  6-17
7  MAJOR SOURCE MATERIALS  	  7-1
                                   Vlll

-------
                                        FIGURES
FIGURE                                                                               PAGE

 1-1      Ground Water Migration Pathway 	  1-4
 1-2      Surface Water Migration Pathway 	  1-5
 1-3      Air Migration Pathway  	  1-6
 1 -4      Onsite Exposure Pathway  	  1-7

 2-1      Example of Field Test Process  	  2-2
 2-2      General Project Schedule  	  2-7

 3-1      Air Pathway: Containment Factor Values  	  3-4
 3-2      Air/Ground Water/Onsite Exposure Pathways: Toxicity Factor Values  	   3-6
 3-3      Surf ace Water Pathway: Drinking Water/Environmental
          Threats - Toxicity Factor Values  	  3-7
 3-4      Air/Ground Water Pathways: Mobility Factor Values  	   3-9
 3-5      Surface Water Pathway: Persistence FactorValues  	   3-10
 3-6      Surface Water Pathway: Drinking Water/Human Food Chain
          Threats - Toxicity/Persistence Factor Values  	   3-11
 3-7      Surf ace Water Pathway: Recreation/Environmental Threats -
          Toxicity/Persistence Factor Values  	   3-12
 3-8      Air/Ground Water Pathways: Toxicity/Mobility Factor Values 	   3-13
 3-9      Air/Ground Water Pathways: Hazardous Waste Quantity Factor Values  	   3-17
 3-10     Surface Water/Onsite Exposure Pathways: Hazardous Waste Quantity
          Factor Values  	   3-18
 3-11     Air Pathway: Targets Population Factor Values 	   3-28
 3-12     Ground Water Pathway: Targets Population FactorValues  	   3-29
 3-13     Surface Water Pathway: Drinking Water Threat-
          Targets Population Factor Values 	   3-30
 3-14     Surface Water Pathway: Recreation Threat - Targets Population Factor Values  	   3-31
 3-15     Onsite Exposure Pathway: Resident Population Threat - Targets Factor Values  	   3-32
 3-16     Onsite Exposure Pathway: Nearby Population Threat-Targets Factor Values  	   3-33
 3-17     Frequency of Sensitive Environment Types  	   3-35
 3-18     Air Pathway: Targets Sensitive Environment Factor Values 	   3-37
 3-19     Surface Water Pathway: Environmental Threat - Targets Factor Values 	   3-38
 3-20     Air Pathway: Targets Use Factor Values  	   3-40
 3-21     Ground Water Pathway: Targets Use Factor Values   	   3-41
 3-22     Surface Water Pathway: Drinking Water Threat-Targets Use Factor Values   	   3-42
 3-23     Surface Water Pathway: Human Food Chain Threat-Targets Use Factor Values  ...   3-43

 4-1      Surface Water Pathway: Food Chain Threat-Targets Population Factor Values  ...   4-10
 4-2      Surface Water Pathway: Recreation Threat - Targets Population Factor Values  	   4-12
 4-3      Onsite Exposure Pathway: Nearby Population Threat - Use Factor Values  	   4-15

 5-1      Total Site LOE Hours 	  5-4
 5-2      Total Site Dollar Cost  	  5-5
 5-3      Total Site CLP Samples   	  5-6
 5-4      Proportion of Total Site LOE, by Category  	  5-8
 5-5      Proportion of Total Site Dollars, by Category 	  5-9
 5-6      Proportionate Distribution of CLP Samples  	   5-10
                                          IX

-------
                                 FIGURES (concluded)

FIGURE                                                                              PAGE

 5-7     General Tasks LOE Hours 	   5-11
 5-8     General Tasks Dollar Cost  	   5-12
 5-9     Site, Source, and Waste Characteristics LOE Hours  	   5-14
 5-10    Site, Source, and Waste Characteristics Dollar Cost 	   5-15
 5-11    Ground Water Pathway LOE Hours  	   5-16
 5-12    Ground Water Pathway Dollar Cost  	   5-17
 5-13    Surface Water Pathway LOE Hours  	   5-19
 5-14    Surface Water Pathway Dollar Cost  	   5-20
 5-15    Air Pathway LOE Hours 	   5-22
 5-16    Air Pathway Dollar Cost 	   5-23
 5-17    Onsite Exposure Pathway LOE Hours  	   5-24
 5-18    Onsite Exposure Pathway Dollar Cost   	   5-25
 5-19    Proportion of Total Site LOE 	   5-28
 5-20    Proportion of Total Site Dollars  	   5-29

 6-1     Proposed HRS Site Scores 	   6-3
 6-2     Proposed HRS Air Pathway Scores  	   6-4
 6-3     Proposed HRS Ground Water Pathway Scores 	   6-5
 6-4     Proposed HRS Surface Water Pathway Scores  	   6-6
 6-5     Proposed HRS Onsite Exposure Pathway Scores 	   6-7
 6-6     Proposed HRS Surface Water Pathway: Drinking Water Threat Scores  	   6-8
 6-7     Proposed HRS Surface Water Pathway: Food Chain Threat Scores 	   6-9
 6-8     Proposed HRS Surface Water Pathway: Recreation Threat Scores  	   6-10
 6-9     Proposed HRS Surface Water Pathway: Environmental Threat Scores  	   6-11
 6-10    Proposed and Current HRS Scores  	   6-13
 6-11     Air Pathway Scores:  Proposed and Current HRS  	   6-14
 6-12    Ground Water Pathway Scores:  Proposed and Current HRS   	   6-15
 6-13    Surface Water Pathway Scores:  Proposed and Current HRS  	   6-16

-------
                                        TABLES
TABLE                                                                               PAGE

 3-1      Percentage of Sites Receiving Maximum Toxicity, Mobility,
          or Persistence Factor Values  	  3-8
 3-2      Percentage of Sites Receiving Maximum Combined Toxicity/Persistence or
          Toxicity/Mobility Factor Values 	  3-14
 3-3      Hazardous Waste Quantity Measures Used at the Field Test Sites  	  3-19
 3-4      Conditions Necessary to Document an Observed Release  	  3-23
 3-5      Automated Databases Used to Support Revised MRS Factors  	  3-25

 4-1      Persistence Values 	  4-8

 5-1      Summary of Site Costs  	  5-27

 6-1      Average and Median Proposed MRS Pathway and Site Scores  	   6-2
 6-2      Average and Median Proposed HRS Surface Water Threat Scores   	   6-12
 6-3      Average and Median Current HRS Pathway and Site Scores  	   6-12
 6-4      Average and Median Pathway and Site Scores Under the Proposed
          and Current HRS  	  6-17
                                          XI

-------
                                        SECTION 1

                                     INTRODUCTION
1.0   BACKGROUND

In 1980, Congress enacted the Comprehensive Environmental Response, Compensation, and Liability
Act (CERCLA), commonly called the Superfund, in response to the dangers posed by uncontrolled
releases of hazardous substances, pollutants, or contaminants into the environment that occur from
hazardous waste sites or facilities.  To implement Section 105 of CERCLA, the U.S. Environmental
Protection Agency (EPA) revised the National Oil  and Hazardous Substances Pollution Contingency
Plan(NCP). Section 105 required EPA to establish:

       "criteria for determining priorities among releases or threatened  releases [of
       hazardous substances] throughout the United States for the purpose  of taking
       remedial action and, to the extent practicable, taking into account the potential
       urgency of such  action, for the purpose  of taking removal action.  Criteria and
       priorities ... shall  be based upon relative risk or danger to public health or welfare or
       the environment... taking into account to the extent possible the population at risk,
       the hazard potential of hazardous substances at such  facilities, the potential for
       contamination of drinking water supplies, the potential  for direct human contact
       [and] the potential for destruction of sensitive ecosystems..."

To meet this requirement and help set priorities,  EPA developed and adopted the Hazard Ranking
System (HRS) as Appendix A of the NCP on July  16, 1982 (47  FR 31180).  The HRS is used  in the
Superfund program to evaluate sites for possible placement on the National Priorities List (NPL). The
NPL includes those sites that appear to pose the most serious threats to  public health or the
environment, and that appear to warrant remedial  investigation and possible cleanup under CERCLA.
The Superfund Amendments and Reauthorization Act of 1986 (SARA) required EPA to revise the HRS
so that, to the  maximum extent feasible, it  accurately assesses the relative risks of sites.  The
Conference Report on  SARA (H. Rep. 962, 99 Cong., 2d Sess. at 199-200 (1986))  makes clear that this
mandate does not require detailed risk  assessments,  but directs EPA to rank sites as accurately as
feasible based simply  on information available from preliminary assessments  and site inspections
consistent with the goal of "expeditiously" identifying candidates for response actions.  In addition,
SARA specifically directed EPA to consider contamination of the human food chain, drinking water
wells, surface water recreation use, and potential contamination of ambient air. In response to these
mandates, EPA  proposed HRS revisions for public comment  on December 23, 1988 (53 FR 51962-
52081).

The Agency gathers data to support an HRS score through a series of site investigations consisting of
increasingly focused  data collection efforts.  This series of data  collection efforts is intended to
provide more detailed information and confidence in determining whether a hazardous waste site
(or a hazardous substance facility) is likely to pose risks to public health, welfare, or the environment.

In revising the HRS, EPA is proposing a number of changes to the current HRS which are designed to
better assess the risk posed  by each site relative  to other sites.  The process for developing the
proposed revisions to the HRS reflects  the Agency's efforts to improve this assessment using
information that could reasonably be collected. The proposed H^S revisions are intended to provide
a scoring system that will allow for the expeditious scoring of sites, yet still  increase the accuracy of
the scoring system.  However, the proposed revised HRS is not designed to be used as a quantitative
risk assessment tool.
                                           1-1

-------
In 1987, the Agency initiated a field test to help assess the costs and implementation concerns
associated with the proposed revisions to the HRS. The field test had several major objectives:

      •   To test the feasibility of implementing the new and proposed revised HRS factors.

      •   To determine the resources required (i.e., costs and technical  hours)  for specific site
          inspection and scoring tasks under the proposed revised HRS.

      •   To assess the availability of information that would be needed for the evaluation of sites
          with the proposed revised HRS and to identify difficulties with its use.

To meet these objectives, EPA tested the proposed revised HRS by performing site inspections at 29
sites nationwide.  (The limitations of the field test are discussed  within Sections 1.4 and 6.0.)  To
accomplish this,  EPA selected sites with characteristics that would help test  the proposed  new
components of the HRS.

Generally, the field test achieved the objectives stated above.  The proposed HRS revisions generally
functioned as anticipated and could be evaluated using an appropriate level of effort (LOE).  Data to
support the various proposed revised HRS factors usually could  be collected using information
obtained from the site inspections.  As such, the primary purpose of this report is to discuss those
findings or issues regarding the proposed revised HRS  that were  brought forward during the field
test.

1.1    SARA AND THE HRS

SARA requires that the HRS be reviewed  and modified (if appropriate)  to  account for several
considerations.  Specifically, Section 105 of CERCLA, as amended by SARA, requires that  the HRS
consider, to the extent feasible, the following:

      •   The damage to natural resources associated with releases or threatened releases that may
          affect the human food chain.

      •   Contamination or potential contamination of the ambient  air that is  associated with
          releases or threatened releases.

      •   The human health risks associated with actual or potential contamination of surface water
          used for recreation or as potable water.

Additionally, Section 118 of CERCLA, as amended by SARA, states that a high priority shall be given to
including  on the  NPL those facilities where the release of hazardous substances or pollutants has
resulted in the contamination or closing of drinking water wells or a principal drinking water supply.
Section 125 requires that the HRS be revised with respect to facilities that contain substantial volumes
of wastes  generated primarily by combustion of coal or other fossil fuels (e.g., fly ash,  bottom ash,
slag waste, and waste from control of flue gas emissions).  For such facilities, the  assessment must
consider the following:

      •   The  quantity, toxicity, and  concentrations  of  hazardous constituents  present  in such
          wastes.

      •   The extent of, and the potential for, release of such constituents into the environment.

      •   The degree of risk to human health and the environment posed by such constituents.
                                             1-2

-------
Finally, SARA requires EPA to modify the HRSsothat, "to the maximum extent feasible, [it] accurately
assesses the relative degree of risk to human health and the environment posed by sites and facilities
subject to review."

To meet these requirements, EPA proposed a number of changes to the current MRS.  In the proposed
revisions, all of the factors considered in the current HRS have been revised in some way, with some
"new" factors having been added and some "old" factors deleted.

1.2   THE CURRENT HRS

The current HRS, promulgated in 1982, evaluates the  relative threat of a site over five pathways.  The
scores for three pathways - ground  water, surface  water, and air -- are combined  into an overall
migration  score that is the primary consideration in placing a site on the NPL.  The two other
pathways  -- direct contact and fire/explosion -- are evaluated to determine the need for immediate
or emergency removal action.

The current  HRS uses a structured value analysis approach to scoring sites.  This approach assigns
values to factors related to or indicative of risk from migration of hazardous substances from the site.
A scale of numerical rating values is provided for each factor, and a value is assigned to each factor
based on conditions at the site. Individual values  are then weighted.  The factors are grouped into
three factor categories (release, waste characteristics, and targets) and are combined to obtain the
factor category  scores.  Each factor category has a maximum  value, as does each  of the component
factors within the category.

The relevant factor category values are multiplied  together within each pathway and normalized to
obtain a pathway score.  If all three pathway scores are low, the HRS score will be low. However, the
HRS score will  be relatively high  even if only one pathway  score is high.  EPA considers  this  an
important requirement because some dangerous  sites pose  threats through  only one migration
route.

1.3   THE PROPOSED REVISED HRS

The proposed revised HRS retains the same basic approach as the current HRS, while incorporating
SARA requirements as well as other improvements identified by the Agency (Figures 1-1 through 1-4).
The proposed changes that could have a significant impact on data collection activities include:

      •  Structure - The revisions propose to retain the ground  water, surface water, and air
         pathways, and add a  fourth pathway, onsite  exposure.  The  direct contact and
         fire/explosion pathways will be dropped.

      •  Hazardous Waste Quantity - The proposed "tiered" approach to evaluating the hazardous
         waste quantity factor would allow use of the actual amount of hazardous constituents if
         these data are available (e.g., analytical results, processing records).  If such data are not
         available,  the quantity of waste  deposited at the site that contains  hazardous substances
         could be used.  If quantities are not available, then the volume or area  of the sources
         present at the site could  be used.

      •  Target Distances - With some exceptions, the  distance over which human populations,
         sensitive environments,  or other targets  are evaluated will be measured from the sources
         of contamination at  the site (or from the probable point where contaminants enter a
         surface water body),  rather  than measuring from the extent of contamination.  Also, the
         distances over which targets are evaluated have been increased for the ground water and
         surface water pathways.
                                            1-3

-------
                                       FIGURE 1-1
                        GROUND WATER MIGRATION PATHWAY
Current HRS
 Release          X           Waste Characteristics        X         Targets

 Observed Release              Q  Hazardous Waste Quantity           n  Ground Water Use
       or                     D  Toxicity/Pers/'stence                 Q  Distance to Nearest
 Route Characteristics                                                   We///Population Served
 Q Depth to Aquifer of Concern
 n Net Precipitation
 n Permeability of Unsatu rated Zone
 D Physical State
 n Containment
Revised HRS
 Likelihood of     X            Waste Characteristics         X        Targets
 Release

 Observed Release               Q  Hazardous Waste Quantity*         n  Ground Water Use*
       or                     D  Toxicity/MOBILITY                 O  Population*
 Potential to Release                                                 D  MAXIMALLY EXPOSED
 O  Depth to Aquifer/                                                   INDIVIDUAL
    HYDRAULIC                                                    D  WELLHEAD
    CONDUCTIVITY                                                     PROTECTION AREA
 n  Net Precipitation
 O  SORPTIVE CAPACITY
 O  Containment
 Items in italic under Current HRS have been dropped or replaced.
 Items in caps under Revised HRS are new. Most items not in caps have been revised significantly.

 * Factor based on several sub-factors.
                                            1-4

-------
                                      FIGURE 1-2
                       SURFACE WATER MIGRATION PATHWAY
Current HRS
Release X
Observed Release
or
Route Characteristics
n Facility Slope/
Intervening Terrain
Q One Year, 24 Hour Rainfall
n Physical State
n Distance to Nearest Surface
Q] Containment
Waste Characteristics X
Q Hazardous Waste Quantity
n Toxicity/Persistence





Water

Targets
D Surface Water Use
D Population Served/
Distance to Nearest
Intake Downstream
n Distance to a Sensitive
Environment



Revised HRS

Likelihood of X
Release
Observed Release
or
Potential to Release
OVERLAND FLOW
Q Containment
O RUNOFF*
Q Distance to Surface Water
Drinking Water Threat
Waste Characteristics X

fj Hazardous Waste Quantity*
n Toxicity/Persistence






Targets

Q Surf ace Water Use*
Q Population*
n MAXIMALLY EXPOSED
INDIVIDUAL



POTENTIAL TO RELEASE BY FLOOD
n CONTAINMENT (FLOOD)
D FLOOD FREQUENCY




4.

Likelihood of X
Release

(same as above)

Human Food Chain Threat
Waste Characteristics X

n Hazardous Waste Quantity*
Q Toxicity/Persistence/
BIOACCUMULATION

Targets

n FISHERY USE
D POPULATION*

+

Likelihood of X
Release

(same as above)

Recreational Threat
Waste Characteristics X

Q Hazardous Waste Quantity*
n Toxicity/Persistence/DOSE
ADJUSTING FACTOR

Targets

D POPULATION*


41

Likelihood of X
Release

(same as above)

Environmental Threat
Waste Characteristics X

n Hazardous Waste Quantity*
D ECOSYSTEM
TOXICITY/Persistence

Targets

n Sensitive
Environments

Items in italic under Current HRS have been dropped or replaced.
Items in caps under Revised HRS are new. Most items not in caps have been revised significantly.
* Factor based on several sub-factors.
                                         1-5

-------
                                       FIGURE 1-3
                               AIR MIGRATION PATHWAY
Current HRS
 Release          X

 Observed Release
Waste Characteristics

n  Hazardous Waste Quantity
n  Toxicity
n  Reactivity and
    Incompatibility
Targets

D  Land Use
D  Population Within
    4-Mile Radius
D  Distance to Sensitive
    Environment
Revised HRS
 Likelihood of     X
 Release

 Observed Release
       or
 POTENTIAL TO RELEASE
 D  SOURCE TYPE
 D  SOURCE MOBILITY*
 n  SOURCE CONTAINMENT
Waste Characteristics
   Hazardous Waste Quantity*
   Toxicity/MOBILITY*
Targets
   Land Use
   Population
   MAXIMALLY
   EXPOSED INDIVIDUAL
   Sensitive
   Environments
 /terns in italic under Current HRS have been dropped or replaced.
 Items in caps under Revised HRS are new. Most items not in caps have been revised significantly.

 * Factor based on several sub-factors.
                                           1-6

-------
                                    FIGURE 1-4
                           ONSITE EXPOSURE PATHWAY
Revised MRS*
 LIKELIHOOD OF
 EXPOSURE

 D  OBSERVED
    CONTAMINATION
  RESIDENT POPULATION THREAT

WASTE CHARACTERISTICS    X


D TOXICITY
TARGETS
                                                          Q
   HIGH RISK POPULATION
   TOTAL RESIDENT
   POPULATION
   TERRESTRIAL SENSITIVE
   ENVIRONMENTS
 LIKELIHOOD OF            X
 EXPOSURE

 D ACCESSIBILITY/
   FREQUENCY OF USE
 D HAZARDOUS WASTE QUANTITY
  NEARBY POPULATION THREAT

   WASTE CHARACTERISTICS    X


   n  TOXICITY
   TARGETS
      POPULATION WITHIN
      1 MILE
   The current MRS includes a direct contact pathway, but that pathway is not used in calculating the
   overall HRS migration score.
                                       1-7

-------
     •   Source Characteristics - The proposed  MRS places additional emphasis on defining and
         characterizing the hazardous waste sources at a facility.  Sources are evaluated separately
         for containment, hazardous waste quantity, waste characteristics, and their potential to
         release hazardous substances.

     •   Actual vs. Potential Exposure - Those targets (e.g., human populations, individuals,
         sensitive environments, drinking water supplies) that have actually been exposed to site
         contaminants will receive higher scores than those potentially exposed. This approach will
         better reflect the differential exposures and risks to these targets.

     •   Dilution/Distance Weighting - The proposed revised MRS would apply distance weighting
         (or for the surface water pathway, dilution weighting) to human populations, individuals,
         sensitive environments, and other targets  potentially  exposed  to contamination.
         Therefore, target distance from the site may be important.

     •   Sensitive Environments - Actual and potential environmental impacts will receive greater
         emphasis, target distance limits are increased, and the types of sensitive environments to
         be considered have been expanded.

     •   Surface Water Pathway - This pathway has been revised  to include four surface water
         "threats" or "subpathways" -- drinking water, human food chain, recreation, and sensitive
         environments. New factors to assess flood potential have also been incorporated. Targets
         are weighted based on dilution or flow characteristics of surface water bodies.

     •   Air Pathway - An observed release will no longer be required to score the air pathway; a
         mechanism to evaluate the potential for an air release has been included in response to
         Section 105 of SARA.  Targets are weighted based on distance from the site.

     •   Onsite Exposure Pathway - This new pathway evaluates actual or potential exposure to
         contaminated soils or shallow wastes associated with the site. Soil sampling of residential,
         school, or day care property may be important.

Note that additional changes (including revisions to the changes discussed  above) are possible as a
result of public comment on the proposed rule.

1.4   DESIGN OF THE FIELD TEST

Due to the design of the field test, the sites tested were not randomly selected and are not necessarily
representative of the greater universe of CERCLA sites. The group of sites included:

     •   Sites with features  the current HRS could not address (e.g., human food chain exposures,
         onsite exposures, potential  air releases).

     •   High-scoring and low-scoring sites under the current HRS.

     •   Different types of sites and sources (e.g., landfills, surface impoundments, waste piles).

A random sample of sites from the universe of sites normally evaluated  under CERCLA might have
been ideal for the field test.  However, a random sample that would  be statistically valid may have
required many more sites, and the amount of time, effort, and money required to test such a random
sample of sites would be prohibitive.

Depending  on the purpose of the analysis, the number of sites tested (29)  may not provide a
statistically valid sample. This limitation was recognized in the design of the field test; however, the
                                            1-8

-------
Agency believes that the field test data regarding the proposed revised HRS are useful and will allow
EPA to draw broad conclusions regarding the proposed revisions.

1.5     MODEL VERSION USED IN THE FIELD TEST PROJECT

As indicated above, the overall  objective of the field test project was to test the  proposed revised
HRS, particularly the new or expanded  HRS factors.  The effort focused on the November  10, 1987
version of the "draft" proposed revised HRS rule, rather than the revisions proposed on December 23,
1988. Differences between these versions include the use of the well head protection area factor for
the ground water pathway as opposed to  the sole source aquifer factor, and greater  resolution in
defined levels of actual contamination.  Other than these differences, essentially all components of
the proposed revised HRS were tested.   For the remainder  of this report, the "draft" or proposed
revised HRS will be referred to as the proposed HRS.

1.6   OUTLINE FOR REMAINDER OF REPORT

The remaining portions of the report are briefly described below:

Section 2 -  "Methodology" explains the general  methodology  and corresponding planning
documents and work products (i.e., "deliverables")  used to obtain field test results.  This section
provides background  information  as to how  the project  was managed and discusses  the
responsibilities  of those offices  and individuals involved in  the field test.  In addition. Section 2
describes what happened during the project.

Section 3 - "Cross-Cutting Issues and Findings" identifies six significant implementation concerns
regarding the proposed HRS: (1) identification and characterization  of sources; (2) evaluation of
toxicity, mobility, and persistence factors;  (3) evaluation of the hazardous waste quantity factor; (4)
sampling strategy used to document observed releases and to obtain other analytic  information; (5)
data significance; and (6) evaluation of actual and potential proposed HRS targets.

Section 4 - "Individual Pathway  Issues and  Findings" identifies implementation issues and discusses
selected items for the air, ground water, surface water, and onsite exposure pathways exclusive of
those issues addressed in Section  3.

Section 5 - "Cost Analysis" provides information on the resources required to collect data in support
of the proposed HRS at the sites tested. Both overall cost averages and ranges, as well as specific data
element costs, are presented.  Section  5 identifies  cost categories and specific elements which
contribute most significantly to site inspection costs and the preparation of proposed HRS scoring
packages.

Section 6 - "Field Test HRS Scoring  Results" discusses the observed impact of various proposed  HRS
factors and categories on  the overall HRS score, as well as on individual pathway scores.  This
information is based solely on sites involved  in the project.

Section 7 - "Major Source Materials" provides citations for source documents and pertinent reference
materials.
                                            1-9

-------
                                        SECTION 2

                                     METHODOLOGY
2.0   INTRODUCTION
This section describes the field test and explains the general  methodology and  corresponding
materials used to obtain results and findings (Figure 2-1). It also provides information on how the
project was managed and discusses the responsibilities of those offices and individuals involved in the
field test.  For each site, a number of work products or deliverables were prepared through a
combined Regional EPA/Field Investigation Team (FIT) effort.  (FITs, one in each Region, are groups of
multidisciplinary,  professional technical personnel who constitute EPA's primary capability for
inspecting waste sites at the pre-remedial stage.) These materials form the basis for the results and
findings presented in this report.

2.1    PROJECT MANAGEMENT AND SPECIFIC RESPONSIBILITIES

Overall  guidance concerning the  development and performance  of project requirements was
provided by the Agency to better ensure that consistent approaches were employed across the field
test sites. In addition to specifying project objectives, EPA  HQ organized all contractors' involvement;
coordinated continuing EPA Regional participation; reviewed project  planning documents and work
products;  assured expedited Contract  Laboratory  Program  (CLP)  access and turnaround (e.g.,
analytical results were generally provided within 30 days of sample collection); and  participated in
and focused project deliberations.

All  10 EPA Regional offices  and their  FIT offices were responsible for carrying  out planning,
background and  desktop data collection activities, field work, and reporting requirements for site
inspections.  For  each Region, one  person from the FIT office served as the project focal point.  A
counterpart was  also appointed within the EPA Regional office.   Each FIT office established site
managers and appropriate teams to conduct the inspections.  The Regional EPA/FIT offices also
worked  together to prepare and review initial proposed MRS scoring packages, to  determine the
resources required (costs,  hours, and elapsed time) for specific site inspection tasks, to prepare site-
specific summary letter reports discussing field test results and findings, and to assist in disseminating
information between all project participants within each Region.

During the start-up of the  field test,  several mechanisms were established to assist in project
management and coordination.  These included:  periodic progress  reports, teleconferences, and
reviews of Regional EPA/FIT work plans and initial proposed MRS scoring packages. These mechanisms
were intended to disseminate information and clarify project requirements, as well as to address
specific concerns raised by the Regions.

FIT offices prepared site-specific work plans and proposed MRS scoring packages and forwarded them
to the respective EPA Regions,  EPA HQ, and the various contractors.  Feedback on these materials was
provided to the Regional EPA/FIT off ices.

2.2    SITE SELECTION

The sites tested in the project were not selected randomly, but were sites that the Regions had
already planned to work on in 1988, or sites that had  specific features EPA wanted to test using the
proposed MRS  (e.g., potential  human food chain or onsite exposure  concerns).  Each Region was
encouraged to suggest five to  eight sites, along with brief site descriptions, for EPA HQ to consider.
To minimize disruptions, the Agency made maximum use of current Regional  EPA/FIT activities with
                                            2-1

-------
            FIGURE 2-1
   EXAMPLE OF FIELD TEST PROCESS
           SELECT SITES
   ATTEND ORIENTATION PROGRAM
          PREPARE WORK
          PLANS, OTHER
       PLANNING DOCUMENTS
    INITIATE SITE INSPECTIONS
    AND BACKGROUND DATA
          COLLECTION
        RECEIVE VALIDATED
        SAMPLING RESULTS
PREPARE INITIAL MRS
SCORING PACKAGES
   DISCUSS HRS SCORING PACKAGES
       FINALIZE HRS SCORING
            PACKAGES
PREPARE SUMMARY REPORTS, CURRENT
  HRS PACKAGES, AND COST FORMS
       PREPARE FINAL REPORT
               2-2

-------
some relatively minor modifications for the field test. Potential candidate sites had one or more of
the following features:

      •  A proposed MRS scoring package could be completed by August 31,1988.

      •  A site inspection (SI) had been completed; minimal follow-up could address the additional
         data needed to support the  proposed  HRS (e.g., human food chain threat, recreation
         threat, release potential for the air pathway, and onsite exposure pathway).

      •  An SI was underway; work  plans could be adjusted  to address the proposed HRS data
         elements.

      •  SI activities were planned to start shortly; field work could  be completed and sampling
         results received by July 15,1988.

      •  Field work had resulted  in a site score less than 28.50 (under the current HRS), but  unique
         site characteristics existed which could be addressed under the  proposed HRS.

EPA chose to exclude Federal facilities and  sites regulated  under Subtitle C of the Resource
Conservation and Recovery Act (RCRA)  from the list of candidates. The  Agency selected two to four
sites from each Region for the field test based on available time, site characteristics, and Regional
priorities.

2.3   SITE INSPECTIONS

A three-day orientation program for project participants was held in December 1987.  Two days were
devoted to the proposed HRS.  The remaining time was used to provide an overview  of the field
testing project.  Specific  project responsibilities were discussed in  detail, along with the  general
schedule and work assignments.  Planning considerations for developing site inspection work plans
were also introduced. Finally, several factors and  issues were identified as priorities for field testing
the  "implementability"  of the proposed HRS.  Some of these items required methodologies
significantly different from those used under the current HRS and during previous site inspections.

In  the next step, the Regional EPA/FIT  offices  prepared work plans  for collecting  information to
support the proposed HRS for each site.  These work plans were necessary to ensure that the site
inspection and data collection effort would meet the objectives of the project. Generally, the plans
described the proposed methodology for collecting the necessary data for the proposed HRS, along
with the procedures utilized for  gathering the information needed for other related project
objectives.  Budget, schedule, milestones, key personnel, and  other similar planning considerations
were also covered.

The plans  provided specific details  for all  field  work (e.g.,  sampling  strategies and site
reconnaissances), background data collection activities, and other related tasks.  The plans also
outlined alternative procedures for collecting information to support selected proposed HRS factors.
Each work plan was reviewed by EPA HQ. The Regional EPA/FIT offices revised the work plans to
incorporate EPA HQ comments before the site inspections started at most  sites.

The site inspections conducted during the field test were primarily designed to collect the data
necessary for preparing proposed  HRS scoring  and  documentation packages.  The  scope  of these
inspections generally depended  on how much  data  had previously been collected. To fully test the
proposed HRS, the  Agency supported innovative approaches for collecting data. Also, the  Regions
were encouraged to collect data for every factor in the proposed HRS, including release potential
factors, even in cases where an observed release had been documented for that pathway.
                                            2-3

-------
Site inspections included collection of additional background data (i.e., beyond those required for the
current HRS), onsite/offsite reconnaissances, and field work. Typical background data collection tasks
involved reviewing archival aerial photography; gathering additional target population information;
evaluating   geologic reports, well logs,  and other available data; collecting streamflow data for
nearby surface water; and obtaining information on fishery production, recreational surface water
use, and sensitive environments within the appropriate target distance limits.

Onsite/offsite reconnaissance was conducted to determine the type of containment present at the
site,  to measure source dimensions  and  distances to targets/receptors, to evaluate  the size  of the
drainage area and predominant land use in the vicinity of the site, and to document other proposed
HRS factors through visual observations, photographs, and detailed notes.

Field work  focused on  collection of samples (i.e.,  air, ground water, surface water, soil, sediment,
leachate, and source material) to support the proposed HRS.   In some instances, field analytical
screening techniques and geophysical surveys were performed to target CLP sample locations and to
install  monitoring wells.  Monitoring wells were installed  (or were already in existence) at
approximately 60 percent of the sites in the project. For some sites, tasks normally beyond the scope
of previous  inspections were performed, including the use of ground water seepage  meters, drive
points, and temporary monitoring wells to collect ground water samples; collection of mollusks and
finfish to analyze for human food chain contamination; and use of a  computer model to assist in
estimating the hazardous waste quantity factor.

2.4   INITIAL PROPOSED HRS SCORING PACKAGES

The Regional EPA/FIT office prepared initial proposed HRS scoring packages for each site, using the
proposed HRS instructions and a draft version of the proposed HRS documentation record received
during the  orientation  program.  EPA HQ and FIT contractor support staff visited one site in each
Region during the field work to familiarize themselves with site characteristics.  In addition, a
preliminary reference table  was developed for use in evaluating the various proposed  HRS waste
characteristics factors (e.g., toxicity, mobility, persistence).

EPA HQ also sponsored computer searches of national databases to support the determination of
various proposed HRS factor values. Thegoalsof this activity were to:

      •  Identify the databases having relevant HRS-related information.
      •  Determine which HRS factors the data could support.
      •  Provide site-specific data in a form that could be used to assign factor values.
      •  Provide instructions for future  use of the databases by  Regional EPA/FIT offices and others
         involved in HRS scoring.

These databases are discussed further in Section 3.6.

EPA HQ, model developers,  and contractor support staff evaluated each scoring package, prepared
written comments, and met to discuss general and  site-specific issues. These comments generally fell
into three major categories:  (1) implementation concerns; (2) model development/design issues; and
(3) QA/QC documentation requirements.

The review of Regional EPA/FIT planning documents and work products represented a significant
effort to exchange information on these issues, as well as on data collection methodologies.  EPA HQ
staff and contractors visited each Region after they had reviewed the  initial proposed HRS scoring
packages submitted  by the  Region.  The Regions and  FIT provided feedback to EPA HQ on the
proposed HRS revisions and how they felt the revisions "captured" the relative risks at the  sites
tested.
                                             2-4

-------
Approximately one day was devoted to each site during the Regional visit.  The Regional EPA/FIT
project manager described each site briefly; provided an overview of field work; identified significant
technical, data  collection, and  implementation issues; and discussed other relevant topics.  A
thorough discussion followed, focusing first on sources found  at the site and the quantity of
hazardous wastes associated with these sources, and concluding with a review of each pathway and
its factors.

2.5   OTHER FIELD TEST WORK PRODUCTS

After each Regional visit, EPA HQ forwarded a list of follow-up items for each initial scoring package
to the Regional EPA/FIT office.  These items were derived from reviews  of each package, including
review of  QA and model development issues, and discussions during the Regional  meeting.  In
response, the Regional EPA/FIT offices provided the revised documentation to EPA HQ and discussed
other issues as necessary. These responses completed the HRS scoring process for each site involved in
the field test.

The Regional responses helped EPA HQ to better understand the various innovative data collection
methods employed at the sites, as well as to develop more "refined" proposed HRS scores. These
scores (both individual factor values and overall pathway scores) will be  used to help  fine-tune the
proposed HRS, and have  provided input for  the development of a  proposed  HRS  database for
additional Agency studies.

Each Regional EPA/FIT office developed a summary report  for each  site.  This report discussed
findings  gathered throughout the project and focused  on the  Regional experience in  testing the
proposed HRS.

The Regional EPA/FIT offices also prepared an abbreviated version of a current HRS scoring package
consisting of scoresheets along with an abridged documentation record.  This information permitted
an initial examination of the scoring impact of factors or categories that  have changed between the
current and proposed HRS. These results are discussed in Section 6.

To meet one of the field test objectives -- determine the resources required for specific site inspection
and scoring tasks under the proposed HRS -- a cost information form was prepared for each site. The
form provided a basis for estimating resources (i.e., overall costs, unit or factor costs, and hours)
required  for specific project management and  data collection  tasks.   The form also gathered
information concerning the types of alternative data collection procedures used to support selected
proposed HRS factors (e.g., special field tasks, computer databases, and "desktop" information).

Generally, all data collection and project administrative functions were considered in completing the
cost form, including:

      •  General FIT technical or project management activities.

      •  Collection of background data and review of existing data.

      •  Onsite and offsite reconnaissance.

      •  Sampling activities.

      •  Special field investigation tasks (e.g., drilling  boreholes, conducting geophysical surveys,
         installing monitoring wells).

      •  Proposed HRS scoring package preparation.
                                            2-5

-------
The cost forms provided sufficient detail to cover all significant tasks undertaken to score a site and
represented the "best estimate" of the costs incurred to support proposed  HRS scoring at each site.
The resources required to collect data in support of the proposed HRS are discussed in Section 5.

2.6   GENERAL SCHEDULE

The field test required approximately 18 months (Figure 2-2).  The overall task was to perform site
inspections needed to prepare initial and final proposed HRS packages for 29 sites nationwide. The
most time-consuming activity, completion of the final packages,  involved a series of EPA HQ and
contractor reviews for each package, as well as visits to all 10 EPA Regions.
                                            2-6

-------
                                                                 FIGURE 2-2
                                                       GENERAL PROJECT SCHEDULE
1987


NOV    DEC
1988


JAN
                     FEB    MAR    APR   MAY   JUN     JUL    AUG     SEP    OCT    NOV    DEC
                                                                                                 1989
                                                                                                 JAN    FEB    MAR    APR   MAY    JUN     JUL
              Project Scoping/Onentation/Site Selection
                                    Preparation/Review of Site-Specific Study Plans
                                           Site Inspections/Background Data Collection/Field Work
                                                                       Initial Proposed MRS Scoring Packages
                                                                                           EPA Regional Visits
                     Final Proposed HRS Scoring Packages
                                                                      Cost Analysis
                                  Field Testing Analysis of Proposed HRS Issues
                                                                                                  Final Report

-------
                                       SECTION 3

                        CROSS-CUTTING ISSUES AND FINDINGS
3.0   INTRODUCTION
Project participants identified several cross-cutting issues (i.e., issues that apply to two or more
pathways) while collecting information to support the proposed MRS. These issues, which represent a
combination of implementation concerns and model development items, are applicable throughout
the proposed MRS scoring process.  The issues include:

      •  Identification and characterization of sources.

      •  Evaluation of toxicity, mobility, and persistence factors.

      •  Evaluation of the hazardous waste quantity factor.

      •  Sampling strategy to document  observed releases and to obtain other  analytic
         information.

      •  Data significance (i.e., conditions necessary to document an observed release).

      •  Evaluation of actual and potential proposed HRS targets (i.e., human populations, sensitive
         environments, and resource uses).

This section discusses significant findings and results relating to these  issues. Specific examples are
provided where appropriate. Some of these issues were unresolved during the field test and are the
subject of ongoing Agency studies,  including the development of revised  HRS  guidance as well as
other analyses and considerations to further improve the accuracy and implementation of the model.

3.1    IDENTIFICATION AND CHARACTERIZATION OF SOURCES

The proposed HRS differs from the current HRS  by placing additional emphasis on defining and
characterizing the sources of contamination found at a facility.  The current HRS generally measures
target distances from the locations where hazardous substances (the term "hazardous substances"
refers to CERCLA hazardous substances, pollutants, and contaminants as defined in CERCLA Sections
101(14) and  101(33),  as amended) were originally  deposited and the locations where hazardous
substances have  migrated.   For the proposed HRS, target distances are generally measured from
sources. A source is defined as any  area where a hazardous substance has been deposited, stored,
disposed of, or placed.

In the current HRS, sources (i.e., the different means of storage or disposal at the site) are generally
only evaluated for assigning containment factor values and waste characteristics.   Under  the
proposed HRS,  however, careful characterization of each  source found at the site is important for
several reasons. For example, sources are evaluated separately not only for containment, but also for
hazardous waste quantity, waste characteristics,  and their potential to release specific hazardous
substances.  Some factors in  the proposed HRS  depend on the types of sources at the site.  For
example, when evaluating potential to release in the proposed air pathway,  sources must meet
minimum size requirements to receive a source type factor value other than zero.  Also,  when
evaluating the hazardous waste quantity factor  based on volume or area, different equations are
used for the different types of sources identified at the site.
                                           3-1

-------
The site inspections performed during field testing included characterizing the types of sources found
at the site.  Field teams measured source  dimensions, looked for evidence of containment,  and
determined waste disposal or storage methods.  This information was needed to identify types of
sources to be evaluated for the proposed HRS.  Sampling was also conducted to delineate source
boundaries and  characterize the types of hazardous substances  associated with  each source.  This
sampling was also necessary to support objectives for additional samples collected (e.g., to document
an observed release or demonstrate actual contamination of nearby targets).

Four significant items related to the identification and characterization of sources were identified:

      •  Definition of source boundaries across all pathways.
      •  Description of source types present at the site.
      •  Containment descriptions for the air, ground water, and  surface water pathways.
      •   Point in time at which the site is scored (i.e., initial vs. current conditions).

Although the last item has implications throughout all pathways, as well as for several cross-cutting
issues and individual proposed HRS factors, it will only be discussed  in this section.

3.1.1     Definition of Source Boundaries

The boundaries of a "facility" under CERCLA are generally defined by those locations where wastes
have been deposited, stored, disposed of, or placed, or have come to be located. At several sites,
project participants were unsure of how to treat this last category --  areas contaminated by the
migration of wastes.  For example, wastes spilled  at one location could be transported via rainfall or
erosion away from the source.  It was unclear whether these contaminated  areas should be
considered as part of the source for proposed HRS scoring purposes (e.g., for measurement of target
distances).  In the same manner, spills and  leaks  from drum storage or other source areas may  also
contaminate the nearby ground surface.  Whether such areas are considered as part of the storage
area, or as a separate source, has an impact on site definition and scoring.

Another issue brought forward  during testing involved consistent definition of source boundaries
across all pathways.  Project participants commented that the size of the source may be different for
each pathway under the proposed HRS.  This was particularly true for the onsite exposure pathway,
where definition of the site was  difficult because  many field test personnel felt that the data
obtained from their site inspections did not provide a  high level of confidence in establishing the
areal extent of contamination.  Participants were unclear on how to determine the area of surficial
contamination  for  this pathway at several sites;  this determination is the primary  means of
differentiating between onsite/resident targets and offsite/nearby targets.  The proposed HRS does
not clearly indicate whether the source boundaries established for the air, ground water, and surface
water pathways are the same as those used for the onsite exposure  pathway.

3.1.2     Description of Source Types

Adequately scoring the site  involves describing and characterizing the  sources  of  hazardous
substances.  Field test personnel raised  several issues when  attempting to select the appropriate
source type (e.g., landfill, waste pile, surface impoundment) for contamination at a site. Although a
variety of methods were used, difficulties  arose concerning how to select source type for several
situations. Specific cases included:

      •  Wastes found inside buildings and structures (whether abandoned or active).

      •  Wastes present within inactive surface impoundments that have since been filled in.
                                            3-2

-------
       •   Areas potentially contaminated from spillage or leakage from  another waste-disposal
          location.

       •   Contaminated streams and their sediments.

       •   Seeps or leachate from below-ground sources.

 During the Regional visits, EPA HQ staff and project participants spent considerable time discussing
 issues relating to source type descriptions. The consensus was that better definitions are necessary to
 differentiate between the various source types (e.g., waste piles vs. contaminated soil).

 Under the air pathway, the proposed HRS provides conditions for defining  and evaluating multiple
 sources as a single source if they meet all of the following criteria:

      •   Sources of the same type.
      •   Sources containing the same hazardous substances.
      •   Sources with the same containment characteristics.

 Participants found these criteria were too restrictive considering the information collected during the
 site inspections and the conditions found at the sites tested. At many sites, it was difficult without
 extensive sampling to  determine  whether multiple storage areas  contained the same hazardous
 substances.  Field test personnel were also unclear as to how to evaluate sources that were part of
 another source (e.g., buried drums within a landfill).

 3.1.3      Containment Descriptions

 Except for the onsite exposure pathway, all pathways in  the proposed HRS evaluate containment
 factors for the likelihood of release category. The containment factors in the proposed ground water
 and surface water pathways address more types of  sources  than does the current HRS, and the rating
 scale  was expanded to provide more meaningful discrimination among sites.  However, some field
 test personnel remarked that the containment descriptions were too specific, given the types of sites
 normally evaluated in Superfund.  All sites tested  received the maximum containment factor value
 (10 points) for the proposed ground water and surface water pathways.

 For the proposed air pathway, the containment factor assesses the potential for sources to release
 both gaseous and particulate emissions.  Nearly all sites tested received the maximum containment
 factor value for gaseous containment, as well as for overall source containment (3 points) for this
 pathway.  Values were somewhat more widely distributed for particulate containment (Figure 3-1).
 Most  of the issues identified concerning the air  containment factor involved the containment
 descriptions, specifically those  dealing with the thickness  of uncontaminated soil cover for source
 types  such as landfills, contaminated  soil, or waste piles. Project participants were sometimes unclear
 as to  how to identify the thickness of cover and how to evaluate  sources with varying degrees of
 containment. At some  sites, the uncontaminated soil cover varied in thickness across the surface of
 the source being evaluated, and it was unclear what value should be assigned for gas containment. It
 was also difficult to evaluate containment for wastes inside buildings since this type of source does
 not appear to be well defined by either the current or proposed HRS.

 3.1.4     Initial vs. Current Conditions

The point in time at which the site is scored received much discussion during field testing. Timing
 plays  a critical role in defining and characterizing sources.  Interim response actions or partial
cleanups (e.g., removal  of leaking  drums) may change the  types of  sources present at the site.  The
source identification issue cannot be separated from the issue of initial vs. current site conditions.
                                            3-3

-------
                                      FIGURE 3-1
                                    AIR PATHWAY:
                           CONTAINMENT FACTOR VALUES
                             (Based on 28 Field Test Sites)
Number
of Sites
28
24
20
16
12
8
4
n
• Gas Containment

Paniculate Containment
vy//, Source Containment




1
^^^ [
                      0123

                                        Assigned Values


Note:    These values are based upon findings from field test sites that were primarily selected to
        test  specific features of the proposed MRS.  As such, these values are not necessarily
        representative of the greater universe of CERCLA sites.
                                          3-4

-------
Other scoring factors affected by timing include containment, hazardous waste quantity, site-specific
waste characteristics, observed releases, levels of contamination, target distances, and a number
others.

Several project participants commented that the proposed air and onsite exposure pathways should
be evaluated on current conditions because they felt current conditions may be more representative
of the threats posed by these pathways.  For  example, at one site, drums  in a storage area were
removed during an  emergency response action.  If the drums were removed with their  contents
intact, it may not be appropriate to include them in the hazardous  waste quantity evaluation.
However, if the contents of the drums have leaked to surrounding soils, it may be appropriate to
include the removed drums in the evaluation of hazardous waste quantity.

3.2   EVALUATION OF TOXICITY, MOBILITY, AND PERSISTENCE FACTORS

Both the current and proposed HRS include a waste characteristics factor category in all pathways.
For the  proposed  air, ground  water, and  surface water pathways, two  factors  are  included:  (1)
toxicity combined with either mobility or persistence; and (2) hazardous waste quantity. The onsite
exposure pathway considers toxicity only and does not include hazardous waste quantity in the waste
characteristics factor category.  This section discusses general issues and findings relating to:

      •  Distribution of toxicity, mobility, and persistence factor values for each pathway.
      •  Use of the proposed HRS hazardous substance reference table.

3.2.1     Distribution of Toxicitv. Mobility, and Persistence Factor Values

Both the current and proposed HRS  toxicity rating systems are intended to distinguish the relative
hazard associated with different hazardous substances.  The toxicity factor in the proposed HRS
considers acute toxicity, as well as carcinogenic and chronic noncarcinogenic toxicity, and the rating
scale was slightly expanded to provide more meaningful discrimination among sites.  The  field test
results provide indications of the extent of discrimination provided by the proposed  HRS toxicity
factor among the sites tested.

The percentage of sites receiving maximum toxicity, mobility, or persistence values among the  various
proposed pathways  and surface water threats is summarized in Table  3-1.  At approximately 80
percent of the sites tested1, the proposed ground water and onsite exposure pathways received the
maximum toxicity factor value of 5 (Figure 3-2).

In addition, about 80 percent of the sites recorded a maximum toxicity factor value for the proposed
surface water drinking water threat (Figure 3-3). In contrast, about 40 percent of the sites received
the maximum toxicity (or ecotoxicity) factor value for the surface water human recreation and
environmental threats.
'All averages presented in this report were calculated based on the number of sites for which the
particular pathway or factor was evaluated.  Not all pathways were evaluated for each site due to:
the absence of nearby surface water or the absence of an overland flow segment to surface water (six
sites); the lack of observed contamination as defined in the onsite exposure pathway (one site); or
the assignment of a value of zero for the containment factor under a particular pathway (one site).
                                            3-5

-------
                                      FIGURE 3-2
                 AIR/GROUND WATER/ONSITE EXPOSURE PATHWAYS:
                              TOXICITY FACTOR VALUES
                              (Based on Field Test Sites)
Number
of Sites
  Note:
24


20


16


12


 8


 4


 0
                                                                          23   23
                       Air Pathway

                       Ground Water Pathway

                       Onsite Exposure Pathway
           012345

                                Assigned Values


 These values are based upon findings from field test sites that were primarily selected to
 test  specific features of the proposed MRS.  As such, these values are not necessarily
 representative of the greater universe of CERCLA sites.
                                         3-6

-------
                                      FIGURE 3-3
                             SURFACE WATER PATHWAY:
        DRINKING WATER/ENVIRONMENTAL THREATS - TOXICITY FACTOR VALUES
                             (Based on 23 Field Test Sites)
Number
of Sites
24


20


16


12


 8
                       Drinking Water Threat

                       Environmental Threat
   Note:    These values are based upon findings from field test sites that were primarily selected to
           test  specific features of the  proposed HRS.  As such, these values are not necessarily
           representative of the greater universe of CERCLA sites.
                                         3-7

-------
                                       TABLE 3-1
       PERCENTAGE OF SITES RECEIVING MAXIMUM TOXICITY, MOBILITY, OR
                             PERSISTENCE FACTOR VALUES
                               (Based on Field Test Sites)
         Pathway or Threat
                                    Toxicity
Mobility or Persistence
 Air
 Ground water
 Surface water drinking water
 Surface water human food chain
 Surface water human recreation
 Surface water environmental
 Onsite exposure
                                     57%
                                     79%
                                     78%
                                     57%
                                     43%
                                     43%
                                     82%
        75%
        83%
        96%
        83%
        65%
        87%
Note:  These values are based upon findings from field test sites that were primarily selected to test
       specific features of the proposed MRS. As such, these values are not necessarily representative
       of the greater universe of CERCLA sites.

The proposed air and ground water pathways combine toxicity ratings for individual substances with
mobility factor values.  Over 75 percent of the sites tested received the  maximum mobility factor
value (3 points) for either the proposed air or  ground  water  pathways (Figure 3-4).  The proposed
surface water pathway combines toxicity ratings with  persistence factor  values.  Each of the four
surface water threats receives a specific persistence value.  Persistence is assessed in terms of half-lives
for hazardous substances and the type of water body (e.g., river, ocean,  lake) adjacent to targets.
Over 95 percent of the sites recorded a maximum persistence factor value for the proposed drinking
water threat (Figure 3-5).

The percentage of sites recording maximum combined toxicity/persistence or toxicity/mobility factor
values among the various proposed pathways  and  surface water threats is given in Table 3-2.  A
larger percentage of the sites received maximum toxicity/persistence  factor values for the surface
water drinking water threat than for any of the  other surface water threats (Figures 3-6 and 3-7). At
least 60 percent of the sites tested received maximum toxicity/mobility factor values for the proposed
ground water pathway (Figure 3-8).
3.2.2
Use of the Proposed MRS Hazardous Substance Reference Table
During the early stages of the field test, a hazardous substance reference table was developed for use
in evaluating the various waste characteristics factors in the proposed HRS.  The table included look-
up values for approximately 130 hazardous substances selected because they were  frequently
detected at NPL sites.

Project participants noted that a number of data gaps exist in the table, particularly with respect to
health- and ecological-based benchmarks. A few substances found at the sites tested were not
contained in the table, including munitions/explosives, brand-name pesticides, and  isomers of PCBs.
Also, infectious materials were not included. Finally, some participants remarked that the  assignment
of waste characteristics values throughout the proposed HRS was time consuming.
                                           3-8

-------
                                        FIGURE 3-4
                            AIR/GROUND WATER PATHWAYS:
                                MOBILITY FACTOR VALUES
                                 (Based on Field Test Sites)
          28


          24


          20


Number    16
of Sites

          12


           8


           4


           0
                   Air Pathway

                   Ground Water Pathway
24
    Note:
           01                   23

                             Assigned Values


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed HRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                            3-9

-------
                                     FIGURE 3-5
                            SURFACE WATER PATHWAY:
                           PERSISTENCE FACTOR VALUES
                            (Based on 23 Field Test Sites)
Number
of Sites
28
24
20
16
12
8
4
^1 Environmental Threat

2%: Recreation Threat


Drinking Water Threat
	

1 11 1
Note:
                                                                  22
               0123

                                 Assigned Values


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed MRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                        3-10

-------
                                       FIGURE 3-6
                               SURFACE WATER PATHWAY:
DRINKING WATER/HUMAN FOOD CHAIN THREATS - TOXICITY/PERSISTENCE FACTOR VALUES
                               (Based on 23 Field Test Sites)
Number
of Sites
24


20


16


12


 8


 4


 0
  Drinking Water Threat

J Human Food Chain Threat
                                                                           18
                  0-10  11-20  21-30  31-40  41-50   51-60  61-70  71-80  81-90 91-100

                                         Assigned Values


   Note:    These values are based upon findings from field test sites that were primarily selected to
           test specific features of the  proposed MRS.  As such,  these values are  not necessarily
           representative of the greater universe of CERCLA sites.
                                           3-11

-------
                                     FIGURE 3-7
                            SURFACE WATER PATHWAY:
  RECREATION/ENVIRONMENTAL THREATS - TOXICITY/PERSISTENCE FACTOR VALUES
                            (Based on 23 Field Test Sites)





Number
of Sites



i f.

10

8

6
4
2
n

• Environmental Threat

[ j Recreation Threat
	







1

                                                                     10
                   0-10  11-20 21-30  31-40  41-50  51-60 61-70  71-80  81-90  91-100

                                       Assigned Values
Note:    These values are based upon findings from field test sites that were primarily selected to
        test  specific features of the proposed MRS.  As such, these values are not necessarily
        representative of the greater universe of CERCLA sites.
                                        3-12

-------
                                                            FIGURE 3-8
                                                AIR/GROUND WATER PATHWAYS:
                                               TOXICITY/MOBILITY FACTOR VALUES
                                                    (Based on Field Test Sites)
Ul
         24



         20


         16


Number  12
of Sites

          8


          4


          0
                      Note:
                                              Air Pathway

                                              Ground Water Pathway
                   0-10   11-20    21-30   31-40   41-50   51-60   61-70   71-80   81-90   91-100

                                              Assigned Values


              These values are based  upon findings from field test sites that were primarily selected to
              test specific features of the  proposed MRS.  As such, these values are not necessarily
              representative of the greater universe of CERCLA sites.

-------
                                      TABLE 3-2
  PERCENTAGE OF SITES RECEIVING MAXIMUM COMBINED TOXICITY/PERSISTENCE
                       OR TOXICITY/MOBILITY FACTOR VALUES
                               (Based on Field Test Sites)
         Pathway or Threat                Toxicity/Mobility            Toxicity/Persistence

 Air                                          34%
 Ground water                                62%
 Surface water drinking water                     -                          78%
 Surface water human food chain                   -                          43%
 Surface water human recreation                   -                          35%
 Surface water environmental                      -                          30%

Note:  These values are based upon findings from field test sites that were primarily selected to test
       specific features of the proposed MRS. As such, these values are not necessarily representative
       of the greater universe of CERCLA sites.

3.3    HAZARDOUS WASTE QUANTITY

The current HRS evaluates the hazardous waste quantity factor based on the total amount of
hazardous substances, pollutants, and contaminants at the facility, excluding any wastes associated
with totally contained sources (i.e., those having a containment value of zero for the pathway being
evaluated). The proposed HRS establishes a tiered approach involving three subsidiary factors:

      •  Hazardous constituent quantity.
      •  Site wastestream quantity.
      •  Site disposal capacity.

The three factors are then evaluated based on some or all of the following measures:

      •  Quantity of CERCLA hazardous substances deposited.
      •  Quantity of wastes deposited containing hazardous substances.
      •  Source volume.
      •  Source area.

Nearly all project participants felt that hazardous waste quantity was  the  most difficult factor to
evaluate in the proposed HRS.  Some found  the instructions confusing while others noted the
increased effort  necessary to evaluate the factor.  However, most participants remarked  that the
tiered approach is a significant improvement over the current HRS. They felt that it permitted a more
thorough evaluation of waste quantities  and  provided new mechanisms for estimating hazardous
waste quantity using the volume or area of sources in the absence of other information on hazardous
waste quantity.

Three significant items related to hazardous waste quantity were identified during testing:

      •  Analytic requirements for estimating waste quantity.
      t  Nonanalytic data collection for documenting waste quantity.
      t  Distribution of hazardous waste quantity factor values.
                                          3-14

-------
3.3.1      Analytic Requirements for Estimating Hazardous Waste Quantity

The proposed HRS differs from the current MRS in the way that analytic evidence (e.g., sampling
results) may be utilized to evaluate the hazardous waste quantity factor. For example, concentration
data can now be used to estimate the quantity of hazardous constituents present.  Furthermore,  if
information concerning the quantity of waste containing hazardous substances deposited at a facility
is inadequate, the sizes of the sources identified  can be used to determine the amount of hazardous
substances potentially deposited. In some cases, samples may  be collected to document the source
volume or area.  However, some project  participants commented that the type, number, and
distribution of samples necessary to document these calculations required further explanation.

For several sites, it was unclear whether analyses of subsurface samples could be used to calculate the
depth  of a source.  If available, this depth could be multiplied by the surficial area of the waste-
disposal location to determine source volume. At some test sites, subsurface samples were collected
at various intervals from boreholes and during the installation of monitoring wells.  The uncertainty
with this approach is whether the resulting subsurface contamination represents the waste as
deposited or the extent of vertical migration. Also,  not all samples analyzed indicated the presence
of contamination.

Another issue involved delineation of surficial contamination.  In the onsite exposure pathway, the
total extent of areal contamination at the site,  including the migration of wastes,  is used in
evaluating likelihood of exposure. In all other proposed pathways, source area without including the
areal extent of contamination is used to evaluate the hazardous waste quantity factor, providing no
better  information is available. At several project  sites, field  test personnel collected shallow soil
samples in an attempt to establish the size of the contaminated source area.  Difficulties with this
approach included the following:

      •  The number  and distribution  of samples necessary  for  defining  the extent of
         contamination.

      •  The criteria for inferring contamination between sampling locations.

      •  The high cost of collecting and analyzing a large number of samples.

      •  The attribution of the detected contaminants to site operations.

      •  The criteria for establishing concentrations significantly above background.

Project participants were also uncertain about the data needed to calculate hazardous constituent
quantities.  The tiered approach allows the use of analytic data to estimate hazardous substance
concentrations. Most field testing personnel felt that obtaining such data would be difficult, costly,
and beyond the scope of a site inspection,  and would  not be available on a routine basis.  Others
commented that data from a single sampling event would provide only a "snapshot" of current
conditions. However, several field test  personnel collected source samples (or used previous analytic
results) at their sites in an attempt to determine the concentration of hazardous constituents.  This
approach worked best when wastes in a source were  relatively homogeneous, or when detailed
records and comprehensive  analytic results were available for a  specific wastestream  over the
operational history of the site.
                                           3-15

-------
3.3.2     Nonanalvtic Data Collection for Documenting Hazardous Waste Quantity

Data collection not involving analysis of samples can be used to assist in documenting the hazardous
waste quantity factor.  At some of the sites tested, data collection methodologies were used that
were not often employed for the current HRS.  For example, disturbed areas were identified at several
project sites by reviewing aerial  photographs.  Historical  photos were  particularly  helpful in
reconstructing the sizes and types of sources originally present. These photos assisted field teams in
developing sampling strategies to meet the analytic requirements for calculating waste quantity.
Several field test personnel commented that it is unclear if aerial photography alone would provide
sufficient information to evaluate the hazardous waste quantity factor.

Geophysical investigations-- including electromagnetics, magnetometer surveys, electrical resistivity,
and seismic refraction - were also performed at several test sites.  Much of the geophysical work was
conducted to identify subsurface anomalies  prior to drilling.   Other  project  participants  used
geophysical techniques in attempts to delineate areas of subsurface waste disposal, although they
were uncertain whether this approach provided adequate documentation for the hazardous waste
quantity factor.

3.3.3     Distribution of Hazardous Waste Quantity Factor Values

In the proposed HRS, the hazardous waste quantity factor is assigned a value between 10 and  100 for
all pathways except onsite exposure (where it is a measure of likelihood of exposure), and except in
cases where complete data are available for the hazardous constituent quantity subsidiary factor. For
these exceptions, the hazardous waste quantity factor is assigned a value between zero and  100.

About 60 percent of the sites tested received maximum hazardous  waste quantity factor values for
the proposed air, ground water, and surface water pathways, versus approximately 40 percent for the
onsite exposure pathway (Figures 3-9 and 3-10).

All four measures (hazardous substances quantity, waste quantity as deposited, source volume, and
source area) were evaluated to determine the hazardous waste quantity factor (Table 3-3). The most
frequent waste quantity measures actually used for scoring sites were source volume (38 percent) and
hazardous substances quantity (31  percent).  The waste quantity as deposited and source  volume
measures resulted in the greatest range of hazardous waste  quantity factor values.  For all field test
sites using the site or wastestream hazardous substances quantity factors,  the maximum value of 100
was assigned. Three sites received the default value of 10 for the hazardous waste quantity factor. As
a general rule, sites where actual information  was available on the amount of  wastes deposited
scored  higher than sites where source volume  or source area was  used to evaluate the hazardous
waste quantity factor.

3.3.4     Other Comments

Other comments were made regarding the evaluation  of the hazardous waste quantity factor and
the establishment of criteria to avoid possible double counting  of the quantity of waste present.
Also, field test personnel pointed out that some divisors (Table 2-14 in the proposed HRS) used in the
equations for evaluating the hazardous waste quantity   factor may require adjustment. Some
participants felt that the divisors for calculating source area and source volume are unreasonably
large for some source types.  For example, approximately  80 acres of contaminated  soil must be
documented at a site in order for the site to receive a hazardous  waste quantity factor value higher
than the default value of 10.
                                           3-16

-------
                                     FIGURE 3-9
                          AIR/GROUND WATER PATHWAYS:
                  HAZARDOUS WASTE QUANTITY FACTOR VALUES
                              (Based on Field Test Sites)
Number
of Sites
tt
20
16
12
8
4
n
•






J 3
I
Note:
                          Air Pathway

                          Ground Water Pathway
               0-10  11-20 21-30  31-40  41-50  51-60  61-70  71-80  81-90  91-100
                              Assigned Values


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed HRS. As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                        3-17

-------
                                       FIGURE 3-10
                     SURFACE WATER/ONSITE EXPOSURE PATHWAYS:
                     HAZARDOUS WASTE QUANTITY FACTOR VALUES
                                (Based on Field Test Sites)
Number
of Sites
  Note:
          20
          16
          12
           8
               Onsite Exposure Pathway

               Surface Water Pathway
                                                                               17
       0-10   11-20  21-30  31-40  41-50  51-60  61-70  71-80  81-90  91-100

                               Assigned Values

These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed MRS.  As such, these  values are not necessarily
representative of the greater universe of CERCLA sites.
                                          3-18

-------
                                        TABLE 3-3
                      HAZARDOUS WASTE QUANTITY MEASURES
                                USED AT FIELD TEST SITES
              Measure                Frequency Used     Average Value     Range of Values
 Hazardous substances quantity              31%                95              52-100
 (site, wastestream, or source)

 Waste quantity as deposited                 14%                47               0 -100
 (wastestream or source)
 Source volume                            38%                80               0-100

 Source area                               17%                69              28-100

Note:  At some sites tested, more than one hazardous waste quantity measure was used to evaluate
       the hazardous waste quantity factor.  In these cases, the measure that contributed most
       significantly to the hazardous waste quantity factor value is portrayed in this table.  The use of
       measures that resulted  in values less than 10 have been given the default value of 10 for the
       purpose of  developing average values where appropriate.  Note also that these values are
       based upon findings from field test sites that were selected primarily to test specific  features
       of the proposed HRS. As such, these values are not necessarily representative of the greater
       universe of CERCLA sites.

3.4   SAMPLING STRATEGY

The collection  of environmental  samples (e.g., air, ground water, surface water,  soil)  was an
important part of the site inspections performed during testing.  In  the current HRS, sampling is
generally limited to documenting observed releases, identifying hazardous substances of concern,
and determining locations from which to measure target distances. The proposed HRS provides more
opportunities for using analytic data to help characterize the degree of risk associated with sites.  For
example, analytic evidence may be utilized to evaluate the hazardous waste quantity factor and to
document targets actually exposed to contamination.  However, Contract Laboratory Program (CLP)
sample analyses are costly (averaging $1,000 or more per sample) and project participants were
unclear as to requirements for determining data quality and numbers of samples, alternative analytic
approaches (e.g., field screening), and the appropriateness of sampling for scoring purposes.

During field testing, several sampling techniques were used to document observed releases or
observed contamination for each  proposed HRS pathway. In addition to  detecting contaminant
releases, sampling was used to help document:

      •   Location, size, and contents of sources.
      •   Waste quantity.
      •   Actual contamination at targets.

This section discusses significant implementation  issues and other limitations relating to sampling to
document observed releases that were encountered during the field test.
                                          3-19

-------
3.4.1     Sampling to Document Observed Releases

Both the current and proposed HRS encourage sampling when hazardous substance releases can
likely be attributed to the site.  An observed release can be scored under either of the following
conditions:

      •  Concentrations of contaminants in a media (e.g., ground water, surface water) that are
         attributable to the site significantly exceed background concentrations in that media.

      •  Substances  that are observed being  discharged directly into pathway media from the
         facility can be demonstrated to contain hazardous constituents.

The criteria for documenting observed releases were a significant  issue during field testing and are
discussed in Section 3.5.  Sampling plans were generally designed to collect analytic data for all
pathways of concern.  For the proposed  air pathway,  samples were collected upwind and downwind
in the breathing zone near a suspected source. For the  proposed ground water pathway, samples
collected from monitoring wells,  residential wells, or municipal wells were  compared with other
samples collected in areas believed to be relatively free of contaminants. Samples were collected at
the probable  point of entry and at upstream and downstream locations for the proposed surface
water pathway. The addition of the onsite exposure pathway to the proposed HRS requires sampling
to document surficial  contamination; this pathway is only evaluated when actual contamination is
observed to be present.

3.4.1.1   Proposed Air Path way

Direct air sampling was conducted at 10 of 29 sites tested.  Field teams used planning criteria common
to all sampling approaches:  background samples were  chosen to establish the concentration  of
hazardous substances in the  atmosphere exclusive of any possible releases from the site;  samples
were taken under the same  meteorological conditions  (i.e., wind speed and direction,  relative
humidity, atmospheric pressure, and temperature); and samples were collected in the "breathing"
zone. Equipment utilized during testing  included:  high volume samplers, PM-10  samplers,
polyurethane  foam (PUF) samplers, low volume  personal air samplers, and Fourier-transform Infrared
(FTtR) samplers.

One limitation related to air sampling was recorded during the field test and this involved seasonal
weather conditions. Several project participants opted not to sample because of low  probability for
detecting releases under  less than ideal  conditions (e.g., cold  temperatures and snow  cover).
Additionally, it was not clear that all sampling  equipment utilized would provide data of sufficient
precision to support a documented air release

3.4.1.2   Proposed Ground Water Pathway

Field teams employed both traditional and innovative ground water sampling techniques during
testing, and used planning criteria common to all approaches:  when available, known localized
ground water flow conditions were considered in the selection of sample locations; potential aquifer
interconnections were evaluated  to  decide which  aquifers were  "threatened";  density  of
contaminants of concern (e.g., floaters or sinkers) was considered in deciding which aquifer intervals
to sample; and permanent or temporary wells were installed only if existing monitoring or residential
wells were deemed unusable for sampling, in terms of location or distance from the site.

Two limitations related  to ground water sampling were recorded  during the field test.  First, some
project  participants were  concerned with contaminating underlying aquifers when  using portable
drills or power augers to install temporary wells.  Second, some field test personnel  identified
                                           3-20

-------
 situations where shallow ground water was possibly being released to nearby surface water bodies.
 Participants were unclear as to how to develop sampling strategies to document such releases.

 3.4.1.3    Proposed Surface Water Pathway

 In the current MRS, surface water sampling is generally limited  to documenting observed releases,
 identifying hazardous substances of concern, and determining  locations  from which to measure
 target distances.  In the proposed MRS, more opportunities are  available to use analytic data from
 surface waters as a result of  modifications to the surface water  pathway.    For example, proposed
 surface water targets may be evaluated  based on actual or potential  contamination.  Therefore,
 opportunities may exist to sample drinking water intakes, as well as fisheries, recreational areas, and
 sensitive environments suspected of showing actual contamination.

 Surface water sampling approaches utilized during the field test consisted of collecting grab samples
 at the probable point of entry to surface water, target samples along the in-water surface water
 migration path, and downstream sediment samples.  Other samples that were collected included:

      •   Shallow well point samples/seepage meter samples.   At some sites, project participants
          were unable to identify an overland flow segment to  surface water.  However, potential
          ground water discharges  to surface water were felt to be possible  in several  cases.
          Therefore,  some field teams attempted to document actual  contamination across the
          ground water/surface water interface using well points or seepage meters.

      •   Biological samples.  Some field test personnel conducted sampling of aquatic species to
          document  either an observed release to  surface water or actual human food chain
          contamination.  Electroshocking techniques were used to collect fish for tissue analysis at
          several sites.  Crabs were collected  in traps at one site; and oysters, mussels, and shrimp
          were sampled at another.

 Two limitations regarding surface water sampling were identified during testing:

      •   Use of health- and  ecological-based benchmarks.   Participants noted that  these
          benchmarks are usually based  on aqueous concentrations of contaminants.  Therefore,
          some  felt that sediment and aqueous samples should always  be collected in pairs to
          maximize data usefulness.

      •   Use of Food and Drug Administration (FDA) action levels.  The proposed MRS requires
          comparison of sample results with published FDA action levels in order to document actual
          human food chain  contamination.  However, several field test personnel  remarked that
          these criteria are not available for most hazardous substances.

 3.4.1.4    Proposed Onsite Exposure Pathway

 In the proposed HRS,  the onsite exposure  pathway can be scored only if contamination is observed;
therefore, sampling is essential to confirm surficial contamination and to determine the boundaries
for evaluating the  resident and  nearby population threat target populations.  Field  teams used
conventional sampling techniques to characterize surficial  contamination.  For example, several
project participants used sample grids to define the area of onsite contamination and to determine
waste quantity for the nearby population threat. Shelby tube samplers were used to collect samples
to depths of two feet. Hand-held or power augers were found to be useful at sites with hardened
surface soils. Composite samples were sometimes taken for large areas of expected contamination,
particularly where grab samples at numerous grid points would be cost-prohibitive.
                                          3-21

-------
Field teams also  used  field analytical screening analyses to define  the extent of surficial
contamination, evaluate waste quantity, and document the presence or absence of contamination on
nearby properties. Screening samples were not solely relied on for scoring; however, these samples
were used to select sample locations for subsequent CLP analyses.

Project participants identified several  significant implementation issues or limitations regarding
onsite exposure sampling strategies:

      •  Areal  extent of surficial contamination.  Many field test personnel felt that the data
         obtained from  their site inspections did  not provide a high level  of confidence  in
         establishing the areal extent of contamination.  Several participants were uncertain as  to
         how to determine the areas of surficial contamination (e.g.,  how many samples must be
         taken).  This is important because the area of surficial contamination is the primary means
         of differentiating between onsite/resident targets and offsite/nearby targets.

      •  Depth of surficial contamination. Project participants were uncertain as to what sampling
         strategies to use when only portions of a source contained hazardous substances within
         two feet of the surface.  Several sites had  partially  buried sources, and inferring
         contamination  between "positive" sample points in these situations could result  in
         erroneous waste quantity and target calculations.

      •  Inferred contamination for resident population threat.  The proposed MRS implies that it is
         unnecessary to  sample each residence, school,  or day  care  facility suspected of being
         contaminated.  Points of contamination  established through direct sampling, combined
         with transport mechanisms and topography or surface conditions which  indicate that these
         sampling points are connected, should be adequate.  However, several project participants
         were unclear as to how to connect sample points to infer contamination for resident target
         populations.

3.5   DATA SIGNIFICANCE

In the current MRS, specific guidance is not provided  as to what constitutes a "significantly higher
than background" hazardous substance release.  In the proposed HRS,  however,  several  conditions
are provided through which detected concentrations of hazardous substances can be considered
"significantly" above  background (Table 3-4).  The detection  limit is defined to include the CLP
contract-required detection limit (CRDL) or quantitation limit (CRQL), the laboratory instrument
detection limit (IDL), and the analytical method-specific detection limit (MDL).

Some project participants felt the conditions for documenting "significance" may be unrealistic. They
questioned whether  these  conditions could adequately address  the  following uncertainties
associated with all  measured concentrations:

      •  Uncertainty associated with sampling procedures.
      •  Uncertainty associated with the natural background variability of metals.
      •  Uncertainty associated with laboratory analytical procedures.

As an alternative, some project participants suggested comparison of concentration data with health-
based levels.  However, health-based levels are  not available for all media  for many hazardous
substances.   In addition, not all routine analytical services  (RAS) CRDLs  are below health-based
standards (e.g., maximum  contaminant levels (MCLs) under the Safe Drinking Water Act).  Field test
personnel also remarked that based on the limited sampling performed during the site inspections,
concentrations detected may not be representative of concentrations  that would be detected  at
other times or at other sampling locations.
                                           3-22

-------
                                       TABLE 3-4
          CONDITIONS NECESSARY TO DOCUMENT AN OBSERVED RELEASE
 If background concentration is:
Observed release occurs if detected
concentration is:
 Not detected.
 Greater than or equal to the detection limit, but
 less than 2 times the detection limit.
 Greater than or equal to 2 times the detection
 limit.
Greater than or equal to 3 times the detection
limit.

Greater than or equal to 3 times the applicable
background concentration or greater than  or
equal to 4 times the detection limit, whichever is
less.

Greater than or equal to 2 times the applicable
background concentration.
Several field test personnel suggested the need for careful review of laboratory data.  The range of
precision of sample values should be considered rather than the strict comparison of absolute values.
Also, some personnel felt that the range of allowable precision for CLP analysis may preclude scoring
a 2x release, especially near the detection limit.

3.6   EVALUATION OF ACTUAL AND POTENTIAL PROPOSED HRS TARGETS

Both the current and proposed HRS include a targets factor category for the air, ground water, and
surface water pathways.  The proposed HRS adds an onsite exposure pathway which also includes a
targets factor category.  For each pathway in the proposed HRS, all or most of the following factors
are evaluated for the targets factor category:  human populations, maximally exposed individuals
(MEI), sensitive environments, and resource uses (e.g., land, ground water, fishery).

This section  addresses general issues and findings relating to the proposed HRS targets factor
category. Three factors - population, sensitive environments, and  resource use - are discussed for
each proposed pathway as appropriate. Significant issues and findings for other target factors (i.e.,
surface water human food chain threat  and human recreation threat targets, onsite exposure
pathway target populations) are presented in Section 4.

3.6.1  Target Population Factors

Several target population  factors primarily related to human health are evaluated in the proposed
HRS.  The proposed air and ground water pathways each include one target population factor; the
surface water pathway contains three -- for the drinking water, food chain, and recreation threats.
Three factors are also included in  the onsite exposure pathway - high-risk resident population, total
resident population, and nearby population.  Each pathway has a prescribed target distance within
which target populations are evaluated.  (The proposed HRS also adds human health factors to the
proposed air, ground water, and surface water pathways reflecting risks to the MEI; that is, those
individuals likely to be exposed  to  the highest concentrations of hazardous  substances.)   Five
significant issues or findings were identified regarding the various population factors:

      •   Quality of demographic information needed to document target populations.

      •   Evaluation of onsite human target populations  for the proposed air and onsite exposure
         pathways.

      •   Use of distance weighting factors.
                                          3-23

-------
      •  Use of distance weighting factors.

      •  Use of health-based benchmarks for populations actually exposed to contamination when
         evaluating target populations.

      •  Distribution of target population factor values.

3.6.1.1   Quality of Demographic Information

In  the context of demographic information for documenting target population factors in  the
proposed HRS,  "quality" primarily reflects the ability to calculate relatively accurate population
estimates. A variety of methods -- including topographic maps, aerial photographs, general census
data,  automated databases, visual  surveys,  specialized U.S.  Bureau of Census information, and
contacts with state and local planning agencies - were utilized to estimate target populations during
field testing. For each method, field test personnel identified concerns that could potentially result in
a loss  of accuracy when calculating target populations.  For example, house counts from topographic
maps  were not feasible for some urban sites and were outdated for other sites tested.  Project
participants also were concerned with using general census tract data.  Census tracts sometimes were
located within multiple target distances (e.g., 1/4 -1/2 miles and 1/2 - 1  mile), and participants needed
to make assumptions as to how to apportion  populations between  these intervals;  uniform
distribution of populations within census tracts was generally assumed.

Project participants generally remarked that documenting target populations for the proposed HRS
was very time consuming compared to the current HRS, particularly for the air population factor and
the surface water recreation target population factor.  For example, in the proposed air pathway, the
total  number  of residents, students,  and workers residing  within the various target  distance
categories are all counted. Some field test personnel were unable to determine student and worker
populations  due to a lack of available  information.  Others expressed uncertainties about possible
double-counting of individuals who both lived and worked within the prescribed target distance.  For
the proposed surface water recreation threat,  the target population may be evaluated up to a
maximum distance of 125 miles. Most project participants felt that there was no reasonable way to
determine this population except through the use of automated databases.

The current HRS assumes a multiplier of 3.8 persons per residence as a default when calculating target
populations.  In the proposed HRS, the default multiplier is based on the most recent U.S. Census
factor for the  number of persons per  residence for the  county in which  target populations  are
located. Nearly all project participants considered this an improvement.

Several national automated databases (Table 3-5) were searched to  support the determination of
various proposed HRS factors (including target populations) as part of field testing.  The Graphical
Exposure Modeling System (GEMS) provided target population estimates for the proposed air, onsite
exposure, and surface water  recreation threat pathways.  Developed by EPA's Office  of Toxic
Substances,  GEMS contains procedures for manipulating  1980 census data geographically.
Population data are graphically displayed over user-specified  circular distances.   Populations  are
assigned to the centroid of each census district.

Project participants noted that the GEMS population estimates developed for sites tested had several
limitations.  First, GEMS was relatively unsuccessful in providing accurate data for close-in  distances
(less than one mile from the site). Because populations closest to the site are weighted more heavily
than those farther away, this limitation is important. Second, GEMS was not helpful in determining
populations  in rural or sparsely populated areas, where census districts are larger than in urban areas.
Because the  centroid of a larger district may not fall within the appropriate distance category,  GEMS
often overstated populations or indicated  no resident population when in  fact  there was a
population.
                                            3-24

-------
                                       TABLE 3-5
         AUTOMATED DATABASES USED TO SUPPORT REVISED MRS FACTORS
      Reference Database
        Link Programs
       Reference Databases
          HRS Factors
     Database May Support
 Air Pathway

 GEMS (Graphical Exposure
   Modeling System)
 Census Bureau
 MEI, population
 Ground Water Pathway

 NAWDEX (NAtional Water
   Data Exchange)
 GWSI (Ground Water Site
  Inventory)
                            FRDS (Federal Reporting Data
                             System)

                            Wellfax
MEI, well depth, water use, karst
aquifer, depth to aquifer, aquifer
thickness, hydraulic conductivity,
sorptive capacity

MEI
                               Population
 Surface Water Pathway

 GEMS

 PATHSCAN
 STORE! (STOrage and
  RETrieval)
Census Bureau

WSDB (Water Supply Data Base)
IFD Plot (Industrial Facility
 Discharge Plot)
GAGE file
REACH file

Bios Survey File
Fish Kill File
Recreation population

MEI, stream flow rates, water use
Fishery use
 Onsite Exposure Pathway

 GEMS
Census Bureau
                                                          Population
Note:  GEMS, FRDS, PATHSCAN, WSDB, IFD Plot, GAGE File, REACH File, STORET, Bios Survey File, and
      Fish Kill File are maintained by the U.S. Environmental Protection Agency; NAWDEX and GWSI
      are maintained by the U.S. Geological Survey; Wellfax is maintained by the National Water Well
      Association.
                                         3-25

-------
However, GEMS did provide some advantages over the other data collection methods. GEMS proved
more useful for estimating populations in urban areas compared to other methods, and was a quick
and  reasonable method for estimating the recreation target population in the proposed  surface
water pathway.

For the proposed ground water pathway, the Ground Water Site Inventory (GWSI) and Federal
Reporting Data System (FRDS) databases were searched to support  the target population factor.
Project participants generally remarked that the data provided on well locations (both domestic and
public-supply) were often outdated and inaccurate.  For example, GWSI listed some inactive wells and
failed to identify many active wells within the ground water target  distance.  FRDS data were not
current, and locations sometimes were not provided for municipal wells.

The  PATHSCAN database was utilized to identify the target  population for the proposed surface
water drinking water threat target population.  The database was helpful in locating public water-
supply intakes along the surface water migration path. However, project participants usually had to
verify the drinking water target population associated with these intakes.

Overall, most field test personnel were uncertain about the quality of demographic information from
automated databases.  While these databases may be useful :n  identifying potential target
populations,  further analyses are being conducted to determine whether estimates from these
databases are adequate for scoring sites.

3.6.1.2   Evaluation of Onsite Human Target Populations

The  proposed  HRS evaluates onsite  human  target populations  in the air  and onsite exposure
pathways.  (This section restricts discussion to these specific pathways; the proposed ground water
and  surface water  pathways also evaluate onsite drinking water populations.)  Onsite  populations
are generally weighted more heavily than populations farther away.  For the proposed air pathway,
the onsite population consists of residents, students, workers, and other persons who are regularly
present.  For the onsite exposure  pathway, the onsite population is evaluated  as the resident
population - that is, people living or  attending school or day care centers where there is observed
contamination. Two groups are counted in this pathway: high-risk population (children under seven
years of age) and total resident population (all others).

Project  participants raised several issues regarding documentation of onsite populations. For
example, the criteria for defining onsite target populations for the air pathway were unclear.  At
some sites, an active manufacturing facility was located adjacent to an area of soil contaminated from
the plant's waste management procedures.  Field test personnel questioned whether the  plant's
employees should be considered as an onsite population.

Another issue raised by project participants involved exclusion of onsite workers from total resident
population in the onsite exposure pathway.  The participants noted  that this is the only proposed
pathway that does not consider worker populations. The sfinal difficulty pointed  out was that
documenting resident populations required increased community relations, as well as the collection
of specific information on the occupants of each household (ages of children) with observed property
contamination.
                                           3-26

-------
3.6.1.3    Use of Distance-Weighting Factors

The current MRS weights the population factor by distance only in the air pathway.  The proposed MRS
incorporates dilution/distance-weighting in several target population factors to account for the
attenuation  of hazardous substances in  the environment.  The proposed air and  ground water
pathways use distance-weighting factors to evaluate targets potentially at risk from  releases of
hazardous substances from a site.   These factors are  intended to  represent the reduced
concentrations of contaminants as distance from the site increases.  For the onsite exposure pathway,
resident target populations are more heavily weighted than nearby target populations, which are
distance-weighted. The proposed surface water pathway evaluates potentially exposed targets using
dilution-weighting factors based  on the flow characteristics of the water available for dilution.
Dilution-weighting factors are discussed further in Section 4.3.

Several project participants felt that the distance-weighting of target populations improved the
relative accuracy of the proposed HRS. Some participants also noted  that the distance-weighting
factors resulted in proposed ground water target population values  being  lower than values under
the current HRS.  In addition,  the distance-weighting factors used for the proposed air target
population factor are such that, for distances beyond one mile, very large populations must  be
present to significantly contribute to target values.

Some project participants remarked that the distance-weighting factors used for the nearby target
population factor (onsite exposure pathway) were too high,  possibly resulting  in  overestimated
factor values.  This observation may relate in part to the unattractiveness  of these sites to draw
nearby populations.

3.6.1.4    Use of Health-Based Benchmarks

The current  HRS does not consider whether specific concentrations  of hazardous substances in
drinking water supplies are above health-based benchmarks, but only whether the contaminants are
significantly above background levels.  As a result,   populations known to be exposed to hazardous
substances and those potentially exposed are treated in the same way in  the current HRS.  In the
proposed  HRS, the ground water pathway and  surface water  drinking water threat give greater
weight to target populations whose drinking water wells or intakes are contaminated by hazardous
substances attributable to the site. The  evaluation of these target populations  includes a factor based
on the Federal primary drinking water  regulations,  or on  some other health-based benchmark  if no
standard exists.

During the site inspections performed in the field test, a few instances were identified where people
were  drinking water from contaminated wells,  but no surface water drinking water intakes were
found to be contaminated above health-based benchmarks.  Most project participants commented
that the scope of the site inspection permitted only limited sampling of nearby drinking water-supply
wells. Therefore, the target population identified  as exposed above health-based benchmarks was
not expected to be very large, unless associated with contaminated municipal water-supply wells.

Participants commented that health-based benchmarks do not exist for several hazardous substances
(e.g., cyanide, naphthalene, 1,1-dichloroethane) and that benchmarks are normally not available for
sediments.

3.6.1.5   Distribution of Target Population Factor Values

For the sites tested, target population factor values are generally clustered at the low end  of the scale
(Figures 3-11 through  3-15).  Exceptions are the nearby  target population  factor for  the onsite
exposure pathway (Figure 3-16) and the  target population factor of the food chain threat for the
                                           3-27

-------
                                     FIGURE 3-11
                                   AIR PATHWAY:
                       TARGETS POPULATION FACTOR VALUES
                             (Based on 28 Field Test Sites)
f.\J
10
1C

1 ")
Number ' *
of Sites 1 n
I U
8
c
A
T

n






--~




18




>.:;-..•;.
;:-: ;/;•.











4
2 2
; 	 , , , 1 1

i . .: i.V:. : 1 ;.. :.! 1 . .1
Note:
                 0-10
                 11-20   21-30
31-40
41-50    51-60   61-70
>70
                                        Assigned Values
These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed HRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                         3-28

-------
Ul
tsj
VO
Number
of Sites
20
18

14

10
 8
 6

 2
 0
                                                         FIGURE 3-12
                                                 GROUND WATER PATHWAY:
                                            TARGETS POPULATIpN FACTOR VALUES
                                                 (Based on 29 Field Test Sites)
                                    18
                                                            1
                                                   1
                                   0-20    21-40    41-60   61-80  81-100  101-120  121-140 141-160 161-180  181-200
                                                           Assigned Values
                     Note:     These values are based upon findings from field test sites that were primarily selected to
                              test  specific features of the proposed MRS.  As such, these values are not necessarily
                              representative of the greater universe of CERCLA sites.

-------
                                                          FIGURE 3-13
                                                  SURFACE WATER PATHWAY:
                               DRINKING WATER THREAT - TARGETS POPULATION FACTOR VALUES
                                                  (Based on 23 Field Test Sites)
                 Number
                 of Sites
Ul
o
24


20


16


12


 8
                                    21
                                             1
                                                                            1
                      Note:
         0-20   21-40  41-60 61-80  81-100  101-120  121-140 141-160  161-180   181-200


                               Assigned Values


   These values are based upon findings from field test sites that were primarily selected to
   test  specific features of the proposed MRS.  As such, these values are not necessarily
   representative of the greater universe of CERCLA sites.

-------
                                                          FIGURE 3-14
                                                  SURFACE WATER PATHWAY:
                                  RECREATION THREAT - TARGETS POPULATION FACTOR VALUES
                                                  (Based on 23 Field Test Sites)
Ul
00
Number
of Sites
18
16
14
12
10
 8
 6
 4

 0
                                    17
                      Note:
                   0-20    21-40   41-60  61-80  81-100  101-120  121-140  141-160 161-180   181-200
                                          Assigned Values
              These values are based upon findings from field test sites that were primarily selected to
              test  specific  features of the proposed HRS.  As such, these values are not necessarily
              representative of the greater universe of CERCLA sites.

-------
                                        FIGURE 3-15
                               ONSITE EXPOSURE PATHWAY:
                 RESIDENT POPULATION THREAT - TARGETS FACTOR VALUES
                                (Based on 28 Field Test Sites)
          28


          24


          20


Number    16
of Sites

          12


           8
    Note:
       22
      0-10  11-20  21-30   31-40  41-50  51-60  61-70  71-80  81-90  91-100

                              Assigned Values


These values are based upon findings from field test sites that were primarily selected to
test specific features of the  proposed HRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                            3-32

-------
                                                            FIGURE 3-16
                                                  ONSITE EXPOSURE PATHWAY:
                                     NEARBY POPULATION THREAT - TARGETS FACTOR VALUES
                                                   (Based on 28 Field Test Sites)
CO
U>
                 Number
                 of Sites
16

14

12

10

 8

 6
                                                                                                         14
                                                                                          1
         0-10    11-20   21-30
                                                          31-40    41-50   51-60   61-70   71-80    81-90   91-100

                                                           Assigned Values
                       Note:    These values are based upon findings from field test sites that were primarily selected to
                                test specific features of the proposed  HRS.  As such, these values are not necessarily
                                representative of the greater universe of CERCLA sites.

-------
surface water pathway (discussed in Section 4.3.4).  These results may primarily relate to the use of
distance-weighting factors, along with a general lack of target populations for some sites.

3.6.2     Sensitive Environment Factors

Sensitive environment factors are evaluated in each proposed HRS pathway except ground water.
The current HRS includes only the highest-scoring sensitive ecosystem when more than one exists
within the appropriate target distance, while the proposed HRS sums values for all ecosystems within
the target distance, and most environments are weighted for distance or dilution.   The  types of
sensitive environments considered for each proposed pathway have been substantially expanded as
well.

Three significant issues or findings were identified  regarding the sensitive environment factors
during testing:

      •  Definition and evaluation of sensitive environments.
      •  Use of Natural Heritage Program rankings.
      •  Distribution of sensitive environment factor values.

3.6.2.1   Definition and Evaluation of Sensitive Environments

Given the expanded types of sensitive environments considered in the proposed HRS, several project
participants were unclear as to how to define and evaluate sensitive environments within the target
distances of the sites.  The types of sensitive environments that often presented difficulties were:

         Wetlands.
         Critical habitats for Federally designated endangered or threatened species.
         Spawning/feeding areas and migratory pathways critical for maintenance of a fish species.
         Fish hatcheries.
         Other areas administered by national and state agencies (e.g., forests, memorials, parks).

Figure 3-17 displays the types of sensitive environments that occurred most frequently at sites tested.

A variety of data collection methods were utilized to identify sensitive environments, including
topographic maps, National Wetlands  Inventory maps, visual surveys.  Natural Heritage  Program
rankings, and contacts  with Federal, state, and local agencies.  Field  test personnel noted  that
classifying and analyzing sensitive environments is more time-consuming with the proposed  HRS than
with the current HRS.

Project participants also had  difficulty  evaluating some sensitive environments identified within the
target distances. For example, for the proposed air pathway,  sensitive environments  are weighted
based on their distance from any onsite emission source.  At some sites, contiguous wetlands were
scattered over more than one distance category,  raising the question of evaluating these  wetlands
for each distance interval or only for the interval closest to the site.

Finally,  project participants  had difficulty defining sensitive environment boundaries.   For the
proposed surface water environmental threat, field test personnel were unclear as to how to
determine boundaries between wetlands.   During field testing, it was also sometimes difficult to
document the actual  locations of sensitive ecosystems, particularly ones without fixed geographical
positions (e.g., the range used by endangered or threatened species, as well as nesting locations).
Some environmental agencies (e.g., U.S. Fish and Wildlife Service) were reluctant to provide specific
locations of sensitive environments for fear of compromising the welfare of these environments.
                                            3-34

-------
Percent
of Sites
                                         FIGURE 3-17
                       FREQUENCY OF SENSITIVE ENVIRONMENT TYPES
                                  (Based on Field Test Sites)
                      55
48
40
32
24
16
8
n









21

14 14 14
10

                  Wetlands    Habitats    Refuges
Spawning   Migratory     State-
  Areas    Pathways   Designated
                        Lands
                                                Types
     Note:    These values are based upon findings from field test sites that were primarily selected to
             test  specific features of the proposed HRS.  As such, these values are not necessarily
             representative of the greater universe of CERCLA sites.
                                             3-35

-------
3.6.2.2   Natural Heritage Program Rankings

The proposed MRS incorporates the use of the Nature Conservancy's Natural Heritage Program (NHP)
rankings to assist in evaluating sensitive environments. The NHP was established to provide precise
scientific information in  support of preserving  biological diversity.  In cooperation with  state
governments, the Nature Conservancy maintains ecological inventories of plants, animals, and
communities that are uncommon and threatened on a state or national basis.  These inventories
classify  "elements of diversity" which are  individual species or communities of species and  their
habitats. The elements are ranked by a system in which such factors as rarity and vulnerability are
considered.

Many project participants accessed NHP information in evaluating sensitive environments.  However,
NHP rankings were used to support scoring for only 20 percent of the sites tested. Some NHP data are
partially based on "sightings" that occurred many years ago. Therefore, additional investigation was
required to verify old information.  Project participants also noted that the  NHP databases are not
entirely consistent among the states,  primarily due to  the length of time that the various state
databases have been maintained.

Two other issues were raised during  field testing.  First, NHP normally does not identify the range of
a species' habitat.  Therefore, while  NHP may locate part of the species' habitat, it generally cannot
establish the geographic boundaries  of the entire  range. Second, some states do not normally assign
national rankings to species found in the United States,  but rather only to distinguish species that are
rare in other parts of the Western  Hemisphere.

3.6.2.3   Distribution of Sensitive Environment Factor Values

The air targets sensitive environment factor values are  clustered at the low end of the scale (Figure
3-18). These results primarily relate to the distance-weighting factors, along with the general lack of
sensitive environment targets for some sites.

For the  proposed surface water pathway, most environmental threat targets  factor values were also
relatively low (Figure 3-19).  Higher  assigned values were recorded for those sensitive environments
subject to actual contamination.   Actual contamination (i.e., concentrations exceeding ecological-
based benchmarks based on ambient water quality criteria) of one  or more  types of  sensitive
environments occurred for approximately 20 percent of the sites tested.

For the  onsite  exposure pathway, no terrestrial sensitive environments were identified during field
testing.

3.6.3     Resource Use Factors

The proposed HRS evaluates four  resource use factors.  The proposed air and ground water pathways
each include one factor; the surface water pathway includes a resource use factor for both the
drinking water threat and human food chain threat.  Each pathway has a  prescribed target distance
within which resource uses are  evaluated.   Two significant  issues  or  findings were identified
regarding the resource use factors during testing:

      •  Definition and evaluation of resource use types for each use factor.
      •  Distribution of resource use factor values.

3.6.3.1   Definition and Evaluation of Resource Use Types

Several  project participants were uncertain as to how to define  and evaluate resource  use  types
within the target distances.  For the proposed air pathway, it was unclear whether the site itself
                                            3-36

-------
Number
of Sites
20

18

16
                                        FIGURE 3-18
                                      AIR PATHWAY:
                    TARGETS SENSITIVE ENVIRONMENT FACTOR VALUES
                                (Based on 28 Field Test Sites)
                    18
                                 1
                                           1
1
1
                  0-10   11-20  21-30  31-40  41-50  51-60  61-70  71-80  81-90  91-100

                                           Assigned Values


   Note:    These values are based upon findings from field test sites that were primarily selected to
            test specific features of the proposed MRS. As such, these values are not necessarily
            representative of the greater universe of CERCLA sites.
                                            3-37

-------
                                    FIGURE 3-19
                            SURFACE WATER PATHWAY:
                ENVIRONMENTAL THREAT - TARGETS FACTOR VALUES
                            (Based on 23 Field Test Sites)
16

10

Number 19
of Sites
8
A
2
n
—



—
~

—

15



.':::• ..v...
'• ''•:• .
•'? f-.':
: • $• . ' '<•' '
.' ••_. .. ,•••' .
' viv'';






4
1111 ;
;^:=P-^-;I ^ tt/,:
Note:
   0-10  11-20 21-30 31-40  41-50  51-60  61-70  71-80 81-90 91-100  101-110  111-120

                                  Assigned Values


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the  proposed HRS.  As such, these  values are not necessarily
representative of the greater universe of CERCLA sites.

-------
should be assigned a land use value. Some participants encountered prime agricultural land (based
on Soil Conservation Service soil  surveys) that was  no longer used for agricultural purposes.
Generally, the definitions of the land use types required subjective interpretation on the part of field
test personnel.

In the  proposed ground water pathway and surface water drinking water threat, the resource use
factors are  divided into two subcomponents -- drinking water  use  and "other" water use.  For
drinking water use, project participants requested clarification of minimum hook-up requirements.
For "other" water use, some personnel encountered situations where wells or intakes supplied water
for fish hatcheries.  It was unclear whether this use should  be evaluated as a resource use.
Uncertainties also existed for the fishery use factor in the proposed surface water human food chain
threat, where field test personnel were generally unable to distinguish between subsistence fishing
and recreation/sport fishing.

3.6.3.2   Distribution of Resource Use Factor Values

For the proposed air and ground water pathways, resource use factor values for the sites tested are
clustered at the high end  of the scale  (Figures 3-20 and  3-21).   For the proposed surface water
pathway, there is a somewhat broader distribution of  resource use factor values (Figures 3-22 and
3-23). This may be due in part to instructions in the proposed HRS that require at least one of the two
surface water use factors for a watershed to be given a value of zero.
                                           3-39

-------
                                          FIGURE 3-20
                                        AIR PATHWAY:
                                 TARGETS USE FACTOR VALUES
                                  (Based on 28 Field Test Sites)
Number
of Sites
24


20


16


12


 8


 4
                                                                                     21
                                                 1       1
                   0123
                                4       5      6      7      8      9      10

                                 Assigned Values
     Note:
   These values are based upon findings from field test sites that were primarily selected to
   test  specific features of the  proposed HRS. As such, these values are not necessarily
   representative of the greater universe of CERCLA sites,
                                              3-40

-------
                                          FIGURE 3-21
                                  GROUND WATER PATHWAY:
                                 TARGETS USE FACTOR VALUES
                                  (Based on 29 Field Test Sites)
Number
of Sites
tU
24
20
16
12
8
4
n

	




.
;
1 ' "' I '•:.:. 1
23


                      0-10     15-20
25-30      35-40

Assigned Values
45-50
     Note:     These values are based upon findings from field test sites that were primarily selected to
              test  specific features of the proposed MRS. As such, these values are not necessarily
              representative of the greater universe of CERCLA sites.
                                              3-41

-------
                                        FIGURE 3-22
                                SURFACE WATER PATHWAY:
                  DRINKING WATER THREAT - TARGETS USE FACTOR VALUES
                                (Based on 23 Field Test Sites)
Number
of Sites
IU
1C

12
m
I U
8
e
A
*t
2

n


	
__^





16




• - - . ; ' • ••.;-.' ';.
'..: ..A' ;'•',

' •. ••' - :'"





4
2
r™mtTT,,m^,,,. _ i ...-..•...

:-^-:--i-,. I - ,-•• 1 :
                     0-10
15-20         25-30        35-40

        Assigned Values
45-50
    Note:    These values are based upon findings from field test sites that were primarily selected to
            test  specific features of the proposed HRS.  As such, these values are not necessarily
            representative of the greater universe of CERCLA sites.
                                             3-42

-------
                                    FIGURE 3-23
                            SURFACE WATER PATHWAY:
            HUMAN FOOD CHAIN THREAT - TARGETS USE FACTOR VALUES
                            (Based on 23 Field Test Sites)
               12
               10
                8
     Number
     of Sites
                           10
                                       30
40
50
                                        Assigned Values
Note:    These values are based upon findings from field test sites that were primarily selected to
        test  specific features of the proposed HRS.  As such, these values are not necessarily
        representative of the greater universe of CERCLA sites.
                                        3-43

-------
                                       SECTION 4

                    INDIVIDUAL PATHWAY ISSUES AND FINDINGS
4.0   INTRODUCTION

Project participants identified several pathway-specific issues while gathering information to support
the proposed MRS. These issues, which represent a combination of implementation concerns and
model development items, are generally exclusive of those items discussed in Section 3.  The issues or
items include:

      •  Selected potential to release factors within each proposed MRS pathway (likelihood  of
         release factor category).

      •  General considerations unique to each proposed pathway  (e.g., aquifer boundaries,
         streamflow estimations, delineation of the hazardous substance migration path for surface
         water, surface water persistence values).

      •  Evaluation of surface water human food chain threat and human recreation threat targets.

      •  Evaluation of onsite exposure pathway target populations.

This section describes significant findings and results for issues relating to the proposed air, ground
water, surface  water,  and onsite  exposure pathways.  Specific examples  are  provided  where
appropriate.  Some of these issues  were unresolved during the field test and are the subject for
ongoing Agency studies, including the development of revised MRS guidance as well as other analyses
and considerations to further improve the accuracy and implementation of the model.

4.1   PROPOSED AIR PATH WAY

In the current MRS, the air pathway  is evaluated only if an observed release of hazardous substances
to the atmosphere can be established; otherwise, the pathway score is zero. As mandated by CERCLA
Section 105(a)(8)(A), as amended by SARA, the proposed  air pathway is to consider, to the extent
possible,  factors to evaluate potential  air contamination in the absence of data  to  support an
observed release. The proposed factors are dependent on the type of source (source type factor), the
physical/chemical properties of the hazardous substances at the  source (source mobility factor), and
the degree to which the release of hazardous substances is inhibited (source containment factor).
The source mobility and source containment factors are evaluated based  on  the  potential for the
release of either gaseous or particulate hazardous substances emissions.

Nearly all project participants favored   the proposed air pathway's mechanism for evaluating
potential to release.  Some  remarked that this revision allows  a more accurate assessment of the
relative degree of risk that a site presents. Few comments were recorded for the potential to release
factors; most field test personnel found these factors to be straightforward and relatively simple  to
evaluate.  However, two issues were identified regarding the potential to release factors during
testing:

      •  Definition of minimum size requirements for sources.
      •  Evaluation of potential releases.
                                           4-1

-------
4.1.1     Minimum Size Requirements for Sources

Under the potential to release section of the proposed air pathway, each source on the site which
meets minimum size requirements is assigned a nonzero factor value for source type. Minimum size
requirements are based on the quantity of hazardous substances potentially deposited on the site.  As
a result, some participants had difficulties evaluating minimum size requirements.  These difficulties
were similar to those associated with the hazardous waste quantity factor evaluation (Section 3.3).

Others also commented  that minimum size requirements should  not be based  on the amount of
waste present within a source. Some noted that small quantities of highly toxic hazardous substances
can cause adverse effects, including acute as well as long-term impacts.

4.1.2     Potential Releases

The source  mobility factor included under the proposed  air pathway's potential to release  section
reflects the relative tendency of hazardous substances to be released as gases or  particulates from a
source. The gas and particulate mobility factors are evaluated independently and then  combined to
produce the source mobility factor value for each source at the site.  The gas and particulate
containment factors are  also assigned separate values,  with the higher value serving as the source
containment factor value. Some project participants noted that this approach may not be as accurate
as separate evaluations for potential gaseous and particulate releases.  Others emphasized that the
potential to release factors should  reflect the different characteristics associated with  gaseous and
particulate emissions and should not be combined in this manner.

In the proposed air pathway, the Thornthwaite precipitation effectiveness  (P-E)  index  is utilized to
rate particulate mobility. This index is a surrogate for  the antecedent moisture content of  surface
material which, in turn, is used as a  measure of the  relative mobility of particulates.  However, some
field test personnel commented that both wind speed and  particle size at the surface may  be
important considerations for potential particulate releases.  Others noted that particulates may  be
released via indiscriminate burning  or spontaneous combustion; the source mobility factor does not
account for these types of situations.

Another issue regarding  potential particulate releases involved the distance-weighting factors  used
to evaluate population and sensitive  environments for the proposed air targets factor category.
Some project participants remarked that these distance weights are based on gaseous  releases and
may not be appropriate for sources that contain only particulate hazardous substances.

4.2    PROPOSED GROUND WATER PATHWAY

As in the current HRS, the proposed  HRS considers both  an observed release or a potential to  release
for  the ground  water pathway.   An observed release to ground  water for both the  current and
proposed HRS is demonstrated when direct deposition of material containing  hazardous substances
into ground water is observed,  or when  ground  water samples show a significant increase in
contaminant concentrations over background levels and those contaminants can be attributed to the
site.

The proposed ground water potential to release factor category is comparable to the combined route
characteristics/containment factor categories in the current HRS. For the proposed HRS, however, the
potential to release evaluation encompasses a number of new or modified scoring factors.  Major
revisions to the current  HRS  approach to evaluating potential ground water releases, which  have
been incorporated in the  proposed HRS, can be summarized as follows:

      •  Annual net precipitation is calculated based on the sum of monthly values, with a monthly
         net precipitation value of  zero assigned for negative months.
                                            4-2

-------
      •  Geologic data for scoring potential to release factors can generally be collected only within
         two miles of the site, rather than within three miles as under the current MRS.

      •  The depth to aquifer factor assigns a value greater than zero to all aquifers, not just to
         aquifers within 150 feet of the ground surf ace.

      •  Thickness-weighted hydraulic  conductivity for the entire geologic interval is  used as
         opposed  to the  hydraulic conductivity of the single layer  with  the  lowest hydraulic
         conductivity in that interval.

      •  Depth to aquifer and hydraulic conductivity are combined in a matrix to produce a single
         factor value.

      •  A sorptive capacity factor has been added.

      •  Factors in both the potential to release and targets factor categories have been modified to
         account for special properties of karst aquifers.

      •  The containment factor has been revised and expanded.

Several significant  issues regarding the proposed ground water pathway were raised during  field
testing:

      •  Evaluation of potential ground water  releases  (i.e., thickness-weighted  hydraulic
         conductivity and sorptive capacity).

      •  Evaluation of site geologic setting.

      •  Determination of aquifer boundaries.

      •  Identification of karst aquifers.

      •  Consideration of ground water flow direction.

4.2.1     Evaluation of Potential Ground Water Releases

Evaluation of thickness-weighted hydraulic conductivity  (T/HC) and sorptive capacity (SC) are new
factors under the proposed HRS.   Look-up tables for  permeability and clay/carbon content of
different soil/rock types are provided in the proposed rule for scoring these factors.  At several sites,
field teams collected core samples for laboratory analysis of T/HC and SC.  Lab results and the table
values generally produced the same factor values.

Several project participants commented  that the least  permeable  layer in a stratigraphic  interval
apparently dominated the  T/HC value. Similarly, the least permeable layer sometimes appeared to
influence the SC factor (e.g., low permeability silt and clay layers have a high clay and carbon
content). Thus,  some participants noted that the thickness-weighted approach in the proposed HRS
may not provide enhanced discrimination over the current HRS  approach (which evaluates only the
lowest hydraulic conductivity layer in a given stratigraphic sequence).

Finally, a number of project participants commented on the poor quality of well logs available for
evaluating ground  water potential to release factors.  Stratigraphic descriptions from residential
drinking well or monitoring well logs in  the vicinity of the site  being evaluated were often judged
inadequate to accurately assess depth to aquifer, thickness-weighted hydraulic conductivity, and
                                            4-3

-------
sorptive capacity.  This deficiency prompted several project managers to propose site-specific boring
programs.

4.2.2     Evaluation of the Site Geologic Setting

The majority of the proposed  MRS potential to release factors (e.g., depth to aquifer,  thickness-
weighted hydraulic conductivity, and sorptive capacity) require geologic information representative
of site conditions. The proposed rule allows for subsurface conditions within two miles of the site to
be evaluated when scoring potential to release. A number of project participants felt, however, that
"regional" or "desktop" geologic data (e.g., Federal/state geologic reports) may not be adequate at
many sites for scoring  purposes, and it may be necessary to gather site-specific  data through
drilling/soil boring programs.

Several sites tested  had "complex" structural geology in the form of tilted  bedrock layers and
pervasive jointing and faulting. Stratigraphic thicknesses, rock types, and depths to aquifers varied
significantly over short distances. At these sites, project participants stressed the  need for site-specific
geologic data to evaluate potential to  release.  In  areas of relatively "simple" layer-cake geology,
exhibiting uniform Stratigraphic thicknesses and continuous formations over large distances, project
participants pointed out that regional geologic data would generally be adequate for scoring release
potential.

Site-specific  geologic and  hydrogeologic  investigations are costly in terms of  drilling subcontract
dollars and field  investigation team resources  to  oversee  drilling operations  (see Section 5.3.4).
When drilling was justified during testing, project participants  employed field  sample screening
analyses and surface geophysical techniques to select appropriate boring/monitoring well locations
and to minimize the total number of sample locations required. This phased data collection approach
was generally successful.

4.2.3     Determination of Aquifer Boundaries

Neither the current nor  proposed MRS  provide specific criteria for the evaluation of aquifer
boundaries such as interconnections and  discontinuities.  Under the proposed  MRS, however,  some
guidance is provided to assist in identifying interconnections.  Interconnections may exist when the
following apply:

      •  Geologic data do not demonstrate the presence of low hydraulic conductivity layers (e.g.,
         two orders of magnitude lower than the overlying layer) or confining layers between
         aquifers.

      •  Withdrawals of water from one aquifer affect water levels in the other aquifer.

      •  Migration of constituents from one aquifer to another aquifer has been observed.

These conditions are generally evaluated within a  two-mile  radius  of the site.   Most  project
participants felt that proposed HRS criteria  for evaluating interconnection are  an  improvement,
though unequivocal data to support the specified conditions were rarely available for the sites tested.
Several  field test personnel  suggested employing pump tests  at sites  where interconnection is
uncertain, but could be critical to scoring; however, costs could be prohibitive. In areas with multiple
NPL-candidate sites, participants felt the benefits of regional pump test data could be worth the cost
and effort.

Several project participants suggested the need for more complete and better definition of the terms
"aquifer" and "aquifer discontinuity."  One issue is how thick a layer of low hydraulic conductivity
must be in order to constitute a discontinuity.  Participants commented that this type of information
                                             4-4

-------
would rarely be available without a detailed hydrogeologic investigation. Another is whether a
stratigraphic interval that appears to contain significant low permeability layers is a single aquifer or
multiple aquifers when wells in the vicinity appear to be drawing water across those layers.  A third
is whether aquifer and  aquitard distinctions  should be based on  use rather than on
hydrostratigraphy. Better definitions, and a more thorough discussion of aquifer conditions,  might
limit the subjectivity involved in documenting aquifer interconnections and discontinuities.  Correctly
documenting communication between aquifers can have a significant impact on the calculation of
target populations.   In the proposed MRS, targets are evaluated based  on multiple aquifers, where
appropriate, rather than on a single aquifer as with the current HRS.

4.2.4      Identification of Karst Aquifers

The current HRS does not provide a separate approach in the ground water pathway for evaluating
sites in karst terrain.  Under the proposed HRS, however, factors in both the potential to release and
target factor categories are modified to account for the rapid  transport and low contaminant
attenuation properties of karst.  Most participants favored this approach; however, many felt that
the proposed rule is unclear as to the circumstances under which an aquifer can be considered karst.
Uncertainties are associated with whether karst terrain (e.g.,  caverns,  springs, sinkholes) must be
present to assume a karst aquifer, as well as  with what constitutes sufficient documentation of karst
conditions (e.g., the number of karst terrain features within a four-mile radius of the site) and how
dissolutioned limestone/dolomite/gypsum aquifers lacking surface expressions of karst topography
should be  treated. Two sites tested involved underlying karst aquifers.

4.2.5      Consideration of Ground Water Flow Direction

Neither the current nor proposed HRS directly considers  the  direction of ground water  flow  in
determining and differentiating among  populations that may be  affected by the migration of
hazardous substances. If available, information regarding the direction of ground water flow near
the site can be used to identify sampling locations (e.g., placement of monitoring wells or sampling
of existing wells) and to help determine whether hazardous substances detected in ground water can
be attributed to the site  being evaluated.   However, the proposed  HRS  indirectly considers flow
direction in  evaluating target populations by including a mechanism that accounts  for direction of
substance  migration  in ground water. This is accomplished by assigning higher weights to  those
populations exposed to  drinking water contamination either above or  below health-based
benchmarks (Section 3.6.1.4).  Some project participants  remarked  that this mechanism  was an
improvement for the proposed ground water pathway.

Although  the site inspections performed during the  field test were not designed to determine the
direction of ground water flow, measurements of ground water  levels were normally made from
monitoring wells that existed in the vicinity of the field test sites. Permanent monitoring wells were
available for 18 of the 29 sites tested. The number of wells installed generally ranged from three to
nine; the average was about four or five wells. For many sites, although more data concerning site-
specific hydrogeology were collected during the field test than would be gathered under a typical site
inspection conducted for  the current HRS, several project participants  commented that one-time
measurements of ground water levels did not provide an accurate  interpretation  of the  general
direction of ground water flow. For example, some participants pointed out that ground water may
flow in several directions from a site (e.g., due to mounded ground water beneath landfills) and that
the flow direction near a site may differ from the general flow direction throughout the ground
water target distance limit  (e.g., due to  pumping effects  of other wells). Others  noted that this
uncertainty would be compounded where multiple aquifers existed beneath a site.  Finally, several
field test personnel felt that accurately determining flow direction would  be beyond the scope of a
typical site inspection without expending the additional time and  cost necessary to document
direction with some level of confidence.
                                            4-5

-------
4.3   PROPOSED SURFACE WATER PATH WAY

The proposed MRS divides the surface water pathway into four evaluations representing threats to
human health through drinking water, the human food chain, and recreational water use, as well as
threats to the environment. As in the current HRS, the likelihood of release category is evaluated as
an observed release or as a potential to release.  For the proposed HRS, potential to release has two
components:  overland  flow and flooding.  The flooding component is new and the overland flow
component has been significantly revised.  Each proposed surface water threat is analyzed separately
for waste characteristics (except hazardous waste quantity) and targets. The target distance for these
threats has been extended to  15 miles downstream for both flowing and static water.  Targets
exposed to actual contamination are assigned higher values than those potentially exposed; targets
that are potentially exposed are dilution-weighted based on the flow characteristics of the water
available for dilution.

These major changes included in the proposed surface water pathway raised several significant issues
during field testing:

          Delineation of the surface water migration path (i.e., use of the target distance limit).
          Use of dilution-weighting factors.
          Use of surface water persistence values.
          Evaluation of human food chain threat targets.
          Evaluation of human recreation  threat targets.

4.3.1      Delineation of the Surface Water Migration Path

The surface water migration path includes an overland flow segment and an in-water segment that
hazardous substances would travel as they migrate away from the site.  In most cases, the distance
limit for evaluating proposed surface water targets begins at the probable point of entry to surface
water and extends for 15 miles along the in-water segment. For surface waters such as lakes, oceans,
and bays, no flow direction is presumed and the distance limit is applied as an arc.

Project participants identified several issues relating to delineation of the surface water migration
path.  For some sites, it was unclear whether  storm sewers and other man-made drainage systems
should be included as part of the overland flow segment.  Other participants encountered cases
where no apparent overland flow segment was present;  however, ground water discharges to
surface water potentially existed.  This type of discharge may occur in situations where the depth to
ground water is relatively shallow.  Field test personnel commented that the proposed surface water
pathway does not address this potential release to surface water.  A number of project participants
saw this as a deficiency  and suggested that criteria for potential horizontal  migration of ground
water could be added to the proposed surface water pathway.  The threat associated with ground
water plumes that have not yet reached surface water was felt to be significant at several sites tested.

Project participants also raised issues concerning the in-water segment of the surface water migration
path.  For sites located near tidally influenced areas, it was difficult to document tidal reversals for
evaluating upstream  targets (e.g., sensitive  environments,  fishery resources).  Field  test personnel
were unclear as to how to determine the extent of the tidal run in these situations. In addition, some
participants commented that the  15-mile target distance may be too long for linking possible surface
water contamination to a site, particularly in coastal settings. Others remarked that assessing targets
along the entire in-water  segment was time consuming and  did  not have a  significant impact on
surface water target factor values when  larger bodies of water were involved.  Several participants
suggested the target distance should be  shorter if an observed surface water release cannot be
established.
                                            4-6

-------
4.3.2      Dilution-Weighting Factors

The proposed surface water pathway uses dilution-weighting  factors to evaluate targets;  these
factors reflect the expected  decreased concentration as hazardous substances are diluted in surface
waters.  Values for potentially exposed targets are multiplied by dilution factors assigned according
to the average annual flow of surface water at the target.  Seven flow categories (e.g., minimum
perennial stream, major river, ocean) are available from which to choose a dilution-weighting factor.

Field test personnel identified several difficulties regarding the estimation of streamflow and the
selection of dilution-weighting factors.  For several sites, gauging station information was not
available to establish streamflow.  Project participants were unsure whether field techniques or
visual observations could reliably  be used to estimate average annual flow. Some commented that
these approaches were not very precise and questioned the level  of accuracy needed; others felt that
such estimates could often  be made within the order-of-magnitude range of the factor.  Other
difficulties in determining flow occurred with:

      •   Intermittent water bodies.
      •   Reservoirs, lakes, and ponds.
      •   Water bodies controlled by engineered structures (e.g., dams, spillways).
      •   Tidally-influenced systems (e.g., estuaries, marshes, and other waters near coastal areas).

Project participants also remarked that perhaps another flow category should be included in the list
of dilution-weighting factors. They felt the dilution-weighting factor for very large rivers such as the
Mississippi and Columbia, which have average discharges in excess of 250,000 cubic  feet per second
(cfs), should be numerically lower than the 0.001 factor given for  "major rivers," which is based on
flows greater than 10,000 cfs.

Several field  test personnel  felt that surface water targets closer to the site should receive  higher
values (i.e., that targets should be weighted by distance as well as by dilution).  At some sites, the
dilution-weighting factors resulted in  targets receiving the same  value regardless of whether they
were located 1 mile or 15 miles from the  probable point of entry to surface water.  In  addition, the
dilution weights used in the proposed surface water pathway to evaluate targets located along major
rivers and the oceans generally resulted in these targets receiving relatively low factor values.  For
example, 50 critical habitats subject to potential contamination  must be located along the in-water
segment of such surface water bodies for the factor to be assigned at least 1 point.

4.3.3      Use of Surface Water Persistence Values

The current MRS evaluates the persistence of a hazardous substance based on biodegradation.  In the
proposed  HRS, persistence is based  not only on biodegration but also on four additional  decay
processes: hydrolysis, photolysis, volatilization, and free-radical oxidation.  The rate of decay for each
process is defined by the half-life  of the substance.  The persistence value is assigned based on this
half-life and on the type of water  body (e.g., river, lake, ocean) between the probable point of entry
and surface water threat targets (i.e., drinking water intakes, fisheries, recreation areas, and sensitive
environments).

Several project participants felt that the persistence values assigned to certain hazardous substances
were inaccurate.  Examples of such persistence values, derived through the proposed HRS evaluation
procedures, are given in Table 4-1.  For the substances listed in  the table, participants expected
maximum persistence values for both types of water categories since these compounds are generally
considered to be highly persistent in sediments.  Some field test personnel commented  that the
evaluation of persistence for hazardous substances in the proposed HRS tends to be associated  with
laboratory processes (e.g., free-radical oxidation, photolysis) rather than the fate of  chemicals in the
surface water environment.  For example, sorption of substances onto particulates,  and subsequent
                                             4-7

-------
sedimentation, is a mechanism that may result in the availability of contaminants to surface waters
for a much greater period of time than would be predicted using the proposed HRS approach.  This
type of mechanism is not addressed by persistence under the proposed HRS.

                                        TABLE 4-1
                                  PERSISTENCE VALUES
                                      Persistence Values by Water Category
    Substance
                                Rivers/Oceans/Great Lakes                      Lakes

 PCBs (Aroclor)                                 3                                2

 TCDD(Dioxin)                                 2                                1

 Chlordane                                    3                                2


4.3.4     Evaluation of Human Food Chain Threat Targets

The current HRS addresses human health risks associated with the food chain through several factors
in the target categories of the ground water, surface  water, and air pathways.  However, none of
these factors explicitly evaluates effects to humans via the aquatic food chain.  The proposed surface
water pathway includes a separate threat evaluation  to assess risks from consumption of fish and
shellfish taken from surface waters within the migration path.  This evaluation consists of  three
factor categories:  likelihood  of release, waste characteristics, and targets.  The targets category
includes two factors --  population and fishery use (the  fishery use  factor is discussed  in Section
3.6.3.1).  The population factor consists of two components:  actual and potential food chain
contamination.  The human food chain population value is based on human food chain production
(i.e., the annual harvest or yield (in pounds) of human food chain organisms from each  fishery under
evaluation) and a bioaccumulation potential factor.  Production is calculated by  estimating the
quantity of food  chain  organisms harvested from fisheries along the surface water migration  path.
Dilution-weighting factors are used to determine the potential human food chain contamination
component of population.  Actual and potential contamination populations are added  to assign the
target population factor value.

Nearly all project participants felt that food chain population was one of the more difficult factors to
evaluate  in the proposed HRS.  Some were unable to obtain actual  data on  human food chain
production and pointed out that the standing crop default values (Table I-5  in the proposed HRS)
were incomplete.  Others had difficulties with regard to the definition and characterization of
fisheries,  particularly for tidally-influenced systems.  However, several field test personnel remarked
that the evaluation of  human  food chain risks is a significant improvement over the  current HRS.
Previously, most sites located near coastal areas received low surface water pathway scores because of
the emphasis of the current  HRS on drinking water. The inclusion of the human food chain threat in
the proposed HRS generally increased surface water pathway scores for coastal sites involved in the
field test.

Four significant issues relating to the human food chain threat were identified during testing:

     •  Definition and characterization of fisheries.
     •  Evaluation of human food chain production.
     •  Criteria for actual  human food chain contamination.
     •  Possibly high values associated with the human food chain target population.
                                            4-8

-------
Project participants encountered situations where the definition and characterization of fisheries
along the surface water migration path were difficult.  For example, several  hatcheries were
withdrawing surface water within the target distance for use in propagating and raising fish. In some
cases, these fish were not released to nearby surface water but were transported elsewhere for
release.  Participants questioned whether fish hatcheries met the definition of a fishery for scoring
purposes.  Others had difficulty with delineating fishery boundaries, particularly for migratory fish
such as salmon.

Another issue raised during testing involved the evaluation of human food chain production. A
variety of data collection methods were utilized to estimate productivity.  These methods included
actual catch or harvest  information,  historical stocking rate data, landings data, standing  crop
information, and default values for standing crop.  For each of these  methods, participants
encountered difficulties that could  potentially result in a loss of accuracy when calculating annual
human food chain production.   For example, some productivity data consisted of all food chain
species, including fish not normally consumed  by humans (e.g., menhaden, alewives).  Other
participants were uncertain about  using landings data for commercial catch because the actual
harvest locations may be outside the target distance limit.

In the absence of actual  data on yield or productivity (including stocking rate data), the revised  HRS
includes standing crop default values to estimate  human food chain production. These values were
used for approximately 50 percent of the sites tested that had fishery evaluations.  Several field test
personnel  commented that the table of standing crop default values was incomplete. For example,
values were not available for specific fishery habitats (e.g., ponds, streams, lakes) within some states.
For other fisheries, several values were available.  In such situations, participants were unclear as to
which value should be selected for human food chain production. Others had difficulty when a range
of values (e.g.,  200-300 pounds  per acre) was provided. Overall, most field test personnel  favored
simplifying the table of standing crop default values.

The third issue involved the criteria for documenting actual human food chain contamination. In the
proposed  HRS,  actual  contamination  may be demonstrated  for fisheries along the surface water
migration path based on either of the following conditions:

      •   There is attributable contamination in  fish  tissue above  Food and  Drug Administration
          (FDA) action level standards, and the hazardous substance  that exceeds the FDA action
          level has been documented in an observed release from the site.

      •   A fishery has been closed and the hazardous substance that caused the  closing has been
          documented in an observed release from the  site.

Several participants felt these criteria were too restrictive for documenting actual human food chain
contamination.  For example, FDA action levels were available  for very  few of the hazardous
substances included in the proposed  HRS hazardous substance reference table.  Other field test
personnel  encountered fisheries that were closed due to problems unrelated to hazardous substances
(e.g., high bacterial  counts).  Some suggested the use of state benchmarks, ambient water quality
standards, or observed  release  criteria to demonstrate actual  human food chain contamination.
Participants also proposed the inclusion of additional  levels of  food chain contamination similar to
those used in the proposed ground  water (population factor)  and  surface water  (sensitive
environments factor) pathways.  These levels could, for example,  be  applied to situations where
fisheries are located between  the probable point of entry to surface water and downstream
contaminated sediments.

The fourth issue identified during testing involved frequent high values associated with the human
food chain target population.   Over 50 percent of  the sites  with fishery evaluations received
maximum  or near-maximum human food chain population factor  values (200 points) (Figure 4-1).
                                            4-9

-------
                                           FIGURE 4-1
                                  SURFACE WATER PATHWAY:
                  FOOD CHAIN THREAT - TARGETS POPULATION FACTOR VALUES
                                  (Based on 23 Field Test Sites)
       12



       10



        8



        6

Number
of Sites  4
        0
       Note:
 0-20     21-40   41-60  61-80  81-100   101-120  121-140  141-160  161-180   181-200

                               Assigned Values


These values are based upon findings from field test sites that were primarily selected to
test specific features of the proposed MRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.

-------
Population values were generally high for sites near coastal areas or near water bodies with  low
average annual flows.  Several possible mechanisms were suggested  as contributing to these high
values:

      •   Target distance that is too extensive for coastal areas.

      •   Use of the dilution-weighting factors specified for large water  bodies, which may  not
          adequately account for the extent of dilution expected.

      •   Use of the standing crop default values.

      •   Use of the single highest bioaccumulation potential factor (e.g., bioconcentration factor).

4.3.5      Evaluation of Human Recreation Threat Targets

The current HRS does not evaluate threats to human health through recreational water use except as
a subfactor of surface water use.  The proposed surface water pathway includes a separate threat
evaluation to assess the risk from dermal contact, inhalation, and ingestion during recreational
activities in surface water (i.e., swimming or fishing).  This evaluation  consists of three factor
categories: likelihood  of release, waste characteristics, and targets.  The targets category includes
one factor, population, which is dose adjusted for each  recreation area within the target distance
limit.  The population factor consists of two components:

      •   The estimated number of visits to a recreation area that is actually contaminated.  This
          number is   based on the  distribution of  populations  around the  area,  an
          accessibility/attractiveness factor (i.e., the type  of area present, such as  a boat ramp,
          marina, or beach), and a dose adjusting factor.

      •   The estimated number of visits to a recreation area threatened  by contamination.  This
          number is derived as above, and is adjusted by a dilution-weighting factor.

The value for the highest scoring recreation area is used  as the population factor  value for  the
recreation threat.

Project participants commented that the human recreation target population was a difficult factor to
evaluate in the proposed HRS.  Many felt that the approach for determining population was
inaccurate and time consuming.  Others noted that, even though considerable effort was required to
collect the data to evaluate recreation target population, factor values were assigned low scores for
most sites tested (Figure 4-2), and the result contributed little to pathway and overall site scores. Still
others commented that the target distances were too long.  Some participants  remarked that  the
population factor did  not provide meaningful discrimination among recreation areas.   The most
significant issues included:

      •   Definition and characterization of recreation areas.

      •   Use of the accessibility/attractiveness factor to provide relative differentiation among
          recreation areas.

      •   Use of actual recreation population data (e.g., number of visitors per year).

Several field test personnel encountered situations where the definition and characterization of
recreation areas within the target distance were difficult.  For some sites, recreation activities were
identified along the in-water segment that did not meet the strict definition of a  recreation activity.
Project participants pointed out that the proposed  HRS does not consider recreation  involving
                                           4-11

-------
Number
of Sites
18

16



12

10

 8

 6

 4



 0
                                          FIGURE 4-2
                                 SURFACE WATER PATHWAY:
                 RECREATION THREAT- TARGETS POPULATION FACTOR VALUES
                                 (Based on 23 Field Test Sites)
                   17
                  0-20   21-40  41-60  61-80  81-100   101-120  121-140  141-160   161-180  181-200

                                               Assigned Values
     Note:    These values are based upon findings from field test sites that were primarily selected to
              test specific features of the proposed MRS. As such, these values are not necessarily
              representative of the greater universe of CERCLA sites.

-------
 canoeists, boaters, sailors, or outdoorsmen (i.e., the definition is restricted to fishing and swimming).
 Other participants had difficulties with delineating recreation area boundaries and determining the
 access point from which to measure target populations, particularly when several miles of surface
 water were  used for recreation/sport fishing.  Finally, some participants were unclear as to the
 definition of a recreation area subject to actual contamination.  For example, difficulties arose in
 cases where the recreation area was located between the probable point of entry to surface water
 and downstream contaminated sediments.

 The second  issue regarding recreation population involved the use of the accessibility/attractiveness
 factor.  In the proposed MRS, the accessibility/attractiveness factor determines the distance over which
 the target population is estimated for each recreation area.  Project participants commented that the
 accessibility/attractiveness factor  does not  reflect the relative attractiveness of the actual  surface
 water body, but is generally based on capital  use and access improvements for specific types of
 recreation areas.  For example, a recreation area could be defined by a bridge crossing a  river, if
 waterfront access to the public is provided.  Some  field test personnel remarked that this definition
 may be unreasonable if only limited recreation activities occur in the vicinity of the bridge.  Others
 noted that the number of access points to the water body should be considered because  greater
 accessibility provides more opportunities for recreational water use.

 As mentioned previously for the human food  chain threat,  project participants felt that target
 distances for estimating the recreation population were too long.  These distances range from 10 to
 125 miles, depending on the type of recreation area  identified along the surface water migration
 path.  Some commented that  it is not realistic  to assume that a waterfront picnic area will attract
 populations from a  125-mile radius, especially  when  more attractive recreation areas are present
 within the radius.  As a result, several field test  personnel suggested using actual visitor data (when
 available) to determine the recreation threat  target population.   For many private and  public
 recreation areas, this information was readily available and provided a more accurate assessment of
 the population at risk. Other participants remarked that perhaps a population default value could be
 developed for areas where no visitor data could be obtained, such as locations used for sport fishing
 and local swimming holes.

 4.4   PROPOSED ONSITE EXPOSURE PATHWAY

 The new onsite exposure pathway evaluates threats to human health and the environment through
 direct, physical  contact with hazardous wastes or  contaminated soil.  These  threats include the
 resident population -- those targets (people  or  terrestrial sensitive environments) within  the
 boundaries of a contaminated  property -- and the nearby population - those people living within a
 one-mile travel distance of the site. Both evaluations consist of three factor categories:  likelihood of
 exposure, waste characteristics, and targets.

 A number of project participants felt the onsite exposure pathway was the easiest to evaluate among
 the four proposed HRS pathways.  Several issues associated with this pathway have been discussed in
 other sections of this report. These issues include:

      •  Definition of onsite (or resident) target populations (Section 3.6.1.2).

      •  Analytic requirements for calculating waste quantity (i.e., delineation of the area of
         surficial contamination) (Section 3.3.1).

Two other significant issues and findings were identified during field testing:

      •  Evaluation of likelihood of exposure.
      •  Evaluation of resident and nearby target populations.
                                            4-13

-------
4.4.1     Evaluation of Likelihood of Exposure

Field test personnel made several comments regarding the evaluation of the likelihood of exposure
factor category. For the resident population threat, likelihood of exposure is evaluated based only on
the presence of observed contamination, not on potential to  release.  The  criteria for observed
contamination require analytic evidence of hazardous substances in soils or sources containing
shallow wastes (i.e., on  or  above the surface,  or not  more than two feet below the surface).  In
addition, contamination  may be inferred for areas between the site and the locations used to provide
the analytic evidence.  Based on these criteria, project participants questioned whether leachate,
sediment, or  seepage  samples could  be used to document observed contamination.  Others
commented that the criteria for contamination do not include a containment factor.  At one test site,
a thin layer of asphalt covered shallow contaminated soils;  participants felt that this situation was not
representative of observed contamination.

For the nearby population threat, likelihood of exposure is based on the quantity of hazardous waste
on the site and the site's accessibility/frequency of use.  The quantity of hazardous waste is primarily
expressed as  the total areal extent of  contamination; areal extent  serves as an indicator of  the
probability  of human contact with shallow wastes.  As discussed  in  Section 3.3.1, participants
encountered  several difficulties in determining the extent of contamination.  Other additional
uncertainties involved:

      •  Definition of the contaminated area to be included for determining waste quantity.

      •  Treatment of sources where only a portion of the source contains wastes within two feet of
         the surface (e.g., thickness of cover material varies across the surface of a landfill).

The accessibility/frequency of use factor is evaluated based on the physical characteristics of the site,
along with the type of  property (e.g.,  park, playground, school) exhibiting contamination. Some
participants commented that there was no assigned value for observed  contamination found solely
on residential  property.   In addition, others  remarked that this factor does not account  for
contaminated property or land used heavily for recreation but not formally designated for public use.
Some project participants felt this factor should  be  more related  to the characteristics of  the
contaminated land in attracting neighboring populations.

The accessibility/frequency of use factor values for sites tested are generally clustered at the high end
of the scale of values (Figure 4-3).

4.4.2     Evaluation of Resident and Nearby Target Populations

The second significant issue regarding the proposed onsite exposure pathway was the evaluation of
resident and  nearby target  populations. Some field test personnel had difficulty with the definition
of terrestrial  sensitive environments which are assessed under the resident population threat.  For
example, participants were unclear whether areas associated with Federally designated endangered
species such as waterfowl met the eligibility criteria for terrestrial sensitive environments.  At two
sites, endangered birds were nesting or flying in locations above the areal  extent of contamination
but were not actually seen on the ground. For these cases, field test personnel questioned whether
these areas should be evaluated as terrestrial sensitive environments.

Project participants also had difficulties regarding  the evaluation of resident human target
populations.  At some sites, field test personnel  encountered obstacles in obtaining access to nearby
potentially contaminated properties, particularly residences.  Without such access, documenting the
presence of additional observed contamination for these properties was difficult.   In addition,
participants sometimes did  not have enough information  prior to the site inspection to assume that
nearby properties might be contaminated. Consequently, sampling plans for several sites tested did
                                            4-14

-------
                                        FIGURE 4-3
                               ONSITE EXPOSURE PATHWAY:
                    NEARBY POPULATION THREAT - USE FACTOR VALUES
                                (Based on 28 Field Test Sites)
          22
          16
Number
of Sites
          10

           8

           6

           4

           2

           0
                                                            21
                                 25            50           75

                                         Assigned Values
100
    Note:    These values are based upon findings from field test sites that were primarily selected to
            test  specific features of the proposed MRS.  As such, these values are not necessarily
            representative of the greater universe of CERCLA sites.
                                            4-15

-------
not include the collection of samples from nearby properties, and resident human target populations
were not always evaluated. Finally, some project participants highlighted a difficulty associated with
inferring contamination of properties located between sample "hits." These participants noted that,
due to the potentially large impact of the resident population target factor on pathway and overall
site scores, scores could be overestimated unless each target property that is scored as contaminated
is documented based on analytic results.

Under the nearby population threat, the target factor is evaluated based on the population within a
one-mile travel distance from the site. The travel distance is measured along a straight line unless
natural barriers, such as rivers, are present.  Some participants were unclear as to the definition of
natural barriers to travel; others were  in favor of possibly including physical or man-made barriers to
travel (e.g., highways, topographic features, elevated railroads).
                                             4-16

-------
                                        SECTION 5

                                     COST ANALYSIS
5.0   INTRODUCTION
This section provides an analysis of the cost data generated from the field test sites.  The principal
objective of the analysis is to identify a "best estimate" of the average and range of costs incurred to
support the data requirements of the proposed HRS at the field test sites. Costs are defined in terms
of dollars, FIT technical level of  effort (LOE) hours,  and number  of analytic  samples.  Subsidiary
objectives include the identification of specific cost categories and cost elements that contributed
most significantly to overall costs at these sites and may, therefore, be candidates for the formulation
of cost reduction strategies. However, no attempt has been made to evaluate  the cost effectiveness
of individual data elements in terms of their relative contribution to scoring requirements, or the cost
effectiveness  of different approaches to  collecting equivalent data. These are  considerations that
might also influence the development of cost reduction strategies.

During the field test, FIT personnel tracked  and recorded  the  costs associated with their site
inspections.  A standardized cost  reporting form was  used for this purpose, requesting specific costs
for over 200 individual data elements at a level of detail corresponding to proposed HRS factors and
subfactors.  A uniform  charge rate for FIT technical  LOE was assumed at $50 per hour.  This rate
includes direct labor, equipment, travel  and living  expenses, expendables,  and overhead.  Not
included in this rate are subcontract charges, which were reported on an actual cost basis.  Contract
Laboratory Program (CLP) charges for full Target Compound List (TCL) sample analysis were assumed
to be $1,000  per sample  (i.e., full organic  and inorganic fraction  analyses) and special analytical
services (SAS) were assumed at $1,500 per sample (e.g., very low detection limits; high  concentration
samples; analysis for special, non-TCL substances).  (SAS charges also include full organic and
inorganic fraction analyses.) These figures approximate the current  (1989) average costs experienced
nationwide under the CLP.  In cases where something less than full TCL analysis was performed,
and/or actual  analytic costs were readily accessible, the actual costs were used. Note that SAS charges
for fast turnaround were not included.

5.1   METHODOLOGY

Prior to analysis, the cost data submitted by FIT were subjected to a data refinement process. Several
factors made the development and application of such a process  necessary.  These factors, and
associated data refinement objectives, are described below.

      •  Persons responsible for  tracking and reporting costs associated with a site (usually the site
         manager) tended to have different ideas as to what specific cost data should be reported
         and where on the form these  data should be placed.  There  was a need,  therefore,  to
         improve  the uniformity of reporting and the assumptions employed, both within and
         between individual cost packages.

      •   The field test  was, in fact, a test of the proposed HRS itself, its data requirements, and the
         means available to satisfy those requirements.  As such, FIT personnel were encouraged  to
         test new sources of information, innovative methods of acquiring data, and even
         duplicative means of obtaining the same data.  In many cases, the reported costs thus
         included  elements that would  not likely be a part of future site inspection activities  in
         support of the proposed  HRS,  but were performed primarily for testing purposes.  Such
         elements were isolated and excluded from the analysis.
                                            5-1

-------
      •  Sites had different levels of available information pre-dating the field test.  For some sites,
         extensive investigations had already  been performed by FIT,  EPA's Technical Assistance
         Teams (TATs), state agencies, or potentially responsible parties (PRPs).  At these sites, the
         availability of existing analytic data and in-place monitoring wells often had a substantial
         effect in reducing FIT cost expenditures during the field test.  By contrast, other sites had
         very little data available prior to the field test.  In order to normalize costs, it was thus
         necessary to establish a common "starting point"  for all sites. For this purpose, the cost of
         developing previously available data which were useful to FIT and were consistent with FIT
         data needs in support of proposed MRS scoring (whether such data had been previously
         developed by FIT or by other entities)  were estimated and included in overall site costs. In
         this way, all sites were costed as if FIT had been the sole investigatory entity, all sites were
         costed from a common "starting  point" (of little available pre-existing data),  and the
         reported costs better  reflect the true total cost of data acquisition to support proposed MRS
         scoring. Pre-existing data which were not representative of the type of data that FIT would
         develop, or were not  consistent with the data requirements of the proposed  HRS, were not
         costed for inclusion in the analysis.

      •  At some sites, inspections may have been constrained  by time  and/or weather limitations.
         For example, some project participants felt that air sampling would have been conducted
         at their sites at any  other time of the year  (most of the field test  site inspections, by
         necessity, occurred during the winter). For these sites, the costs of air sampling  programs
         were estimated and included in the analysis, again, in order to better reflect the true total
         cost of data acquisition. At other  sites, time and  weather conspired to cut short sampling
         or well installation programs. In such cases, the full cost of planned activities  was estimated
         and used for the analysis.

      •  Most of the personnel involved in the field test felt that,  as a result of  the  field test
         experience, they had learned a great deal about  the proposed HRS, how it operates, and
         what its data requirements are.   In some cases,  project participants felt that, given this
         higher  level of  experience and  comfort with the model, some aspects of their site
         inspections might have  been conducted differently.  For such sites, these  considerations
         were reflected in the reported costs in order to establish a common "ending  point" among
         sites that represents, as much as possible, a  level of data availability and  quality that is
         representative of the  requirements for site scoring with the proposed HRS.

The data refinement process consisted of a detailed examination and evaluation of the raw,  "as
reported" data, followed by an in-depth  discussion  with the project participants responsible  for
reporting site costs.  These discussions covered  the full range of  cost elements  and included
evaluation  of cost reasonableness or typicality;  data collection methodologies;  and  sampling
strategies, alternatives, and lessons learned.  On the basis of these discussions, the dollars, LOE hours,
and/or number of samples associated with  some specific cost elements were adjusted, within the
context of the objectives outlined above, and with the concurrence of  the FIT representative.  While
this data  refinement process served to "normalize" costs and establish a common framework for cost
reporting, the process should not be construed as one of "optimization." The resultant costs reflect
the views of individual FIT personnel as to appropriate data collection and evaluation  procedures for
their individual sites;  "real world" delays and inefficiencies occasionally occurred and are reflected in
the reported costs; and least cost methodologies were not always employed.

5.2   DATABASE DEVELOPMENT

Following the refinement process, all data were  entered into a computerized database.  The database
was built from data elements corresponding, or analogous, to proposed  HRS factors and subfactors.
These were used to build costs  to the factor category, pathway, and site levels.  Pathway-level data
totals consist of the four proposed HRS pathways (air, ground water,  surface water, and  onsite
                                            5-2

-------
exposure) plus two other categories which have been defined for analytical purposes: general tasks;
and site, source, and waste characterization.

The general tasks category consists of cost elements that tend to be non-pathway-specific and non-
data-generating.  These include project planning  and management; mobilization, demobilization,
and travel; data validation;  preparation of the proposed MRS scoring package and documentation;
and collection of QA/QC samples (blanks and duplicates, which tend to be media-specific rather than
pathway-specific).

The components of the site, source, and waste characterization category are cost elements that are
data-generating but also tend to be non-pathway-specific. These data elements tend to be collected
and evaluated within the context of site/source/waste characterization rather than within the context
of individual  pathway evaluation.  Included are:  identifying and evaluating sources and source
containment; identifying site hazardous substances; evaluating hazardous waste quantity and other
pathway-specific waste characteristics factors;  and collecting non-pathway-specific, multipurpose
samples (such as soil or source samples which can be applied to  the identification of  hazardous
substances, establishment of near-surface  contamination, and  evaluation of hazardous waste
quantity).

Totals for the four proposed MRS pathways include the costs associated with  pathway-specific
subcontracts,  likelihood of release, targets evaluation, and pathway-specific environmental samples
(those associated with observed release and targets).

5.3   DISCUSSION

The following subsections present and discuss the  results of the cost analysis, which is based on 24
field test sites (five  sites were  not included in the analysis due to data constraints).  Costs are
examined on the basis of  site and  pathway totals, and  include  technical  LOE  hours, dollar
expenditures, and number of analytic samples. LOE hours represent the time involved in planning
and conducting the site inspection, acquiring and evaluating  data, scoring the site and preparing the
initial scoring package and documentation record, and internal FIT review and revision. Not included
are EPA reviews and FIT response to such reviews.  Dollar totals include  the costs of FIT LOE,
subcontracted work, and CLP analysis.  In the discussions of proposed HRS  costs that follow, the actual
range and average values derived from the refined cost data for the  24  sites are given, though LOE
has been rounded to the nearest ten  hours and dollar costs are generally rounded to the  nearest
thousand. These figures are discussed in the following subsections and summarized in Section 5.4
below.

5.3.1     Total Site Costs

The range of total site  LOE was 970 to 3,310 hours, with an average of 1,860 hours.  Only two sites,
however, were in excess of 2,320 hours and three sites were at or below 1,120 hours; most sites fell  in
the range of 1,320 to 2,320 hours. The distribution of site LOE  is shown in Figure 5-1.

The dollar cost ranged from $100,000 to $311,000 (Figure 5-2), and averaged $176,000.  Only two sites
were in the range above $220,000 and three sites were below $120,000; most sites were within the
range of $130,000 to $220,000.

The number of CLP samples costed at field test sites ranged from 34 to 98 (Figure 5-3) and averaged
63. Field screening (FASP) was conducted at nine sites. Theoretically, FASP should reduce the number
of samples requiring CLP analysis by providing in-field identification of samples that are  in  fact
contaminated. One might expect the sites where FASP techniques were used to be among those with
the lowest number of CLP samples.  However, this was not always the case:  four of the FASP sites
(employing 11 to 46 FASP samples) required fewer than the average number of CLP samples (a range
                                           5-3

-------
Number
of Sites
Within
 Range
                                          FIGURE 5-1
                                  TOTAL SITE LOE HOURS
                                 (Based on Field Test Sites)
                 StO  10-  11-  12-  13-  14- 15- 16- 17- 18- 19- 20-  21-  22-  23-  24.  25-  26-  27- 28-  29-  30-  31-  32-  >33
                     11  12  13  14  15  16  17  18 19  20  21  22  23  24   25  26  27  28  29  30   31  32  33


                                   Range of LOE Hours (x100)


 Note:     These values are based upon findings from field test sites that were primarily selected to
           test  specific features of the proposed MRS.  As such, these values are not necessarily
           representative of the greater universe of CERCLA sites.
                                              5-4

-------
Number
of Sites
Within
 Range
           0
                                           FIGURE 5-2
                                  TOTAL SITE DOLLAR COST
                                  (Based on Field Test Sites)
                SI10  110-  120-  130-  140- 150- 160-  170-  180- 190- 200-  210-  220- 230- 240-  250-  260- 270- 280-  290- 300-  >310

                     120  130  140  150 160  170  180  190 200  210  220  230 240  250  260  270 280 290  300  310
                                   Range of Dollar Costs (x1,000)
Note:     These values are based upon findings from field test sites that were primarily selected to
          test  specific features of the proposed HRS.  As such, these values are not necessarily
          representative of the greater universe of CERCIA sites.
                                                5-5

-------
                                        FIGURE 5-3
                                TOTAL SITE CLP SAMPLES
                                (Based on Field Test Sites)
Number
of Sites
Within
 Range
          0
                  <35
36-
40
                             41-
                             45
46-
50
51-
55
56-
60
61-
65
66-
70
71-
75
76-
80
81-
85
86-
90
91-
95
                                                                                        >95
                                    Range of CLP Samples

 Note:     These values are based upon findings from field test sites that were primarily selected to
          test specific features of the proposed HRS.  As such, these values are not necessarily
          representative of the greater universe of CERCLA sites.
                                            5-6

-------
of 38 to 44); two of the FASP sites (with 26 and 57 FASP samples) had about average numbers of CLP
samples (64 and 65); and three of the FASP sites (having 22 to 81 FASP samples) involved higher than
average CLP sampling (77 to 91). These results may be attributable to unfamiliarity with the field
screening concept and its implementation, since this is a relatively new technique being employed at
sites and full implementation of FASP in terms  of equipment and capabilities has not yet been
realized.

Figures 5-4 through 5-6 show the average  proportion of total site LOE, total site dollars, and CLP
samples contributed by each of the six cost categories.  Note that, while all sites included samples for
site, source, and waste characterization, and for QA/QC (general tasks) purposes, not all sites involved
sampling within all four proposed HRS pathways.  The percentages illustrated in Figure 5-6 do not
reflect this subtlety but, rather, show the  proportionate representation,  by category, of all  CLP
samples collected during the field test.

It is of interest to note that, combined, the two categories of general tasks and site, source, and  waste
characterization constituted the majority of average total site costs for all three types of costs:


                                       LOE           Dollars        CLP Samples
                                    (% of Total)      (% of Total)       (% of Total)
         General Tasks                   52              35               17
         Site, Source, and Waste
         Characterization                 Jj>              23               37_
                                        67              58               54

5.3.2      General Tasks

Among the  six  reporting categories,  general tasks was the most costly in terms of both LOE  and
dollars.  LOE ranged  from 480 to 1,500 hours (Figure 5-7), and averaged 960 hours. Most sites fell
within the range of 600 to 1,300 hours; only one was below this range and two were above.  In
relative terms, as noted above,  general tasks accounted for  fully one-half (52 percent) of average
total site LOE, ranging from 39 to as much as 70 percent.

The dollar cost of general tasks  ranged from $39,000 to $84,000, and averaged $59,000.  Figure 5-8
shows the distribution of sites over this range.  General tasks accounted for about one-third of
average total site dollars.

Among the components of general  tasks, project planning and management was the largest source
of both LOE and dollar costs.  LOE for this element ranged from 100 to 870 hours, with all but one site
within the range of 100 to 600 hours. The average was 340 hours, which is 36 percent of average total
general tasks LOE and 18 percent of average total site LOE. In dollar terms, the range was $5,000 to
$43,000, with all but one site in the  range of $5,000 to $30,000. The 24 sites averaged $17,000, or 29
percent of average total general tasks and 10 percent of average total  site dollars.  During the data
refinement discussions with FIT personnel, almost all of the project participants interviewed felt that
a significant "learning curve" associated with the proposed HRS was attached to the cost of project
management, and the raw figures for individual  sites were accordingly adjusted downward through
further discussion. The figures reported here are the lower, "refined" figures.

Mobilization, demobilization, and travel ranged from 90 to as much as 700 hours, though all but one
site fell into the range of 90 to about 420 hours.  The average was 240  hours. This cost element was
highly variable between sites and is a function not only of distance of the site from the FIT office, but
also of team size and the number of field trips made to the site.
                                            5-7

-------
                                      FIGURE 5-4
                   PROPORTION OF TOTAL SITE LOE, BY CATEGORY
                              (Based on Field Test Sites)
                              Air Pathway
                                  5%
Onsite Exposure Pathway
         2%
          Surface Water Pathway
                  8%
   Ground Water Pathway
           18%
Site, Source, and Waste Characterization
               15%
                              General Tasks
                                 52%
Note:    These values are based upon findings from field test sites that were primarily selected to
         test specific features of the proposed MRS.  As such, these values are not necessarily
         representative of the greater universe of CERCLA sites.
                                         5-8

-------
                                      FIGURE 5-5
                 PROPORTION OF TOTAL SITE DOLLARS^ BY CATEGORY
                               (Based on Field Test Sites)
                                         Onsite Exposure Pathway
                           Air Pathway             2%
                               6%
   Surface Water Pathway
           13%
  Ground Water Pathway
         21%
General Tasks
    35%
            Site, Source, and Waste Characterization
                          23%
Note:    These values are based upon findings from field test sites that were primarily selected to
        test  specific features of the  proposed MRS.  As such, these values are not necessarily
        representative of the greater universe of CERCLA sites.
                                          5-9

-------
                                     FIGURE 5-6
                  PROPORTIONATE DISTRIBUTION Qf CLP SAMPLES
                              (Based on Field Test Sites)
                        Air Pathway
                            6%
                                        Onsite Exposure Pathway
                                                 4%
Ground Water Pathway
        14%
        General Tasks
      (QA/QC Samples)
            17%
                                                                      Site, Source, and Waste
                                                                         Characterization
                                                                              37%
                        Surface Water Pathway
                                22%
 Note:    These values are based upon findings from field test sites that were primarily selected to
         test  specific features of the proposed MRS. As such, these values are not necessarily
         representative of the greater universe of CERCLA sites.
                                         5-10

-------
                                         FIGURE 5-7
                                GENERAL TASKS LOE HOURS
                                 (Based on Field Test Sites)
Number
of Sites
Within
 Range
                     <5    5-6    6-7    7-8    8-9   9-10   10-11   11-12  12-13   13-14  14-15  >15
                                    Range of LOE Hours (x100)
    Note:    These values are based upon findings from field test sites that were primarily selected to
            test  specific features of the proposed MRS.  As such, these values are not necessarily
            representative of the greater universe of CERCLA sites.

-------
                                             FIGURE 5-8
                                   GENERAL TASKS DOLLAR COST
                                      (Based on Field Test Sites)
Number
of Sites
Within
 Range
6
c
J
4
•3
•)
1
n


	









<40











40-
45











45-
50











50-
55











55-
60











60-
65





65- 70- 75- >80
70 75 80
                                      Range of Dollar Costs (xl ,000)
        Note:    These values are based upon findings from field test sites that were primarily selected to
                test  specific features of the proposed MRS. As such, these values are not necessarily
                representative of the greater universe of CERCLA sites.

-------
 Data validation costs were found to vary by the number of samples and analytic fractions involved,
 and by Regional differences in data validation requirements.  In some Regions,  data validation is
 performed directly by FIT; in other Regions it is performed by EPA's Environmental Services Division
 (ESD) or is subcontracted to a third party. To normalize these differences, data validation was costed,
 in all cases, at a rate estimated as that which would be required if the work were performed by FIT. In
 most cases, FIT was able to supply such estimates based on previous data validation experience.  For
 the field test sites, data validation ranged from 50 to 380 hours and averaged about 180 hours. Most
 sites required between 100 and 300 hours.

 Preparation of the proposed MRS documentation and scoring package for tested sites ranged from 60
 to 400 hours (though only two sites were above  250 hours), with an average of  150  hours. These
 hours  represent the development of scores from  collected  data and the preparation of
 documentation. They do not include the acquisition of data necessary to support factor scores; data
 acquisition LOE is reflected within the individual pathway factors. Nor do they include EPA Regional
 and Headquarters review of submitted packages or FIT response to such reviews.  Like project
 planning and management, most project participants interviewed felt that there was a significant
 learning curve involved here and that the experience of having learned the proposed MRS during the
 course of the field test would result in more efficient site scoring at future sites. The figures reported
 here reflect this view.

 The final element within general tasks is QA/QC samples, and LOE for these was generally modest.
 LOE ranged from less than 10 to 150 hours, though only one site involved more than 65 hours.  The
 range of dollar costs was $5,000  to $24,000, and this was keyed directly to the number of QA/QC
 samples undergoing CLP analysis (a range of 3 to 17, average of 11).

 5.3.3      Site. Source, and Waste Characterization

 The LOE required for the site, source, and waste characterization category at the 24 sites represented,
 on average, 15 percent of total site LOE.  The range was 80 to 530 hours with an average of 280 hours
 (Figure 5-9). Only one site required  less than 130 hours.  Collection of CLP samples required about
 one-third of the LOE expended in this category, while all other activities accounted for two-thirds.

 Among the six cost categories, site, source, and waste characterization was second highest in terms of
 dollar cost, representing nearly one-quarter of average total site dollars. Figure 5-10 shows the cost
 distribution which ranged from about $20,000 to $79,000, averaging $40,000. Only one site involved
 more than $64,000.  Costs associated with CLP samples (LOE and analytic charges) in this category
 accounted for 76 percent of average total dollar cost; the number of samples ranged from 10 to 46
 and averaged 24.  All but two sites involved 32 or fewer samples for  site, source, and waste
 characterization.

 5.3.4     Proposed Ground Water Pathway

While all of the proposed MRS pathways may require environmental sampling, the ground water
pathway also often required the installation of monitoring wells or boreholes at field test sites. As a
result, among the proposed MRS pathways, ground water was by far the largest contributor to total
site LOE, representing 18 percent of average total site  LOE -- which is more than the other three
pathways combined.  Dollar costs were  also considerably higher than those of the other  pathways
and, at 21 percent of the site total,  were about equivalent to the other three pathways combined.
Total ground water  pathway LOE ranged from a low of 50 to a high of 1,360 hours (Figure 5-11).
Dollar costs ranged from under $3,000 to $156,000 (Figure 5-12). The inclusion and extent of drilling
programs were a major factor in these wide ranges of costs.

Drilling programs were costed at 15 of the  24 sites.  Monitoring wells were  involved at 13 sites,
boreholes at two sites, and one site had  both. The number of wells installed generally ranged from
                                           5-13

-------
                                      FIGURE 5-9
              SITE, SOURCE, AND WASTE CHARACTERISTICS LOE HOURS
                              (Based on Field Test Sites)
Number
of Sites
Within
 Range
J
4
3
2
1
0








<100


















101- 151- 201- 251- 301- 351-
150 200 250 300 350 400


401-
450





451- >500
500
                                   Range of LOE Hours
 Note:    These values are based upon findings from field test sites that were primarily selected to
          test specific features of the proposed MRS.  As such, these values are not necessarily
          representative of the greater universe of CERCLA sites.

-------
                                     FIGURE 5-10
             SITE, SOURCE, AND WASTE CHARACTERISTICS DOLLAR COST
                              (Based on Field Test Sites)
J
4
Number
of Sites 3
Within
Range
2
1
0







































































<25 25- 30- 35-
30 35 40










40-
45



























45- 50-
50 55










55-
60

















60- 65- 70- >75
65 70 75
                               Range of Dollar Costs (x 1,000)
Note:    These values are based upon findings from field test sites that were primarily selected to
        test  specific features of the proposed MRS.  As such, these values are not necessarily
        representative of the greater universe of CERCLA sites.
                                         5-15

-------
Number
of Sites
Within
Range
                                        FIGURE 5-11
                         GROUND WATER PATHWAY LOE HOURS
                                 (Based on Field Test Sites)
                                                                   E
Sites without drilling

Sites with drilling
                  <1    1-2    2-3   3-4    4-5   5-6   6-7   7-8    8-9  9-10  10-11  11-12  12-13  >13
                                  Range of LOE Hours (x100)
 Note:     These values are based upon findings from field test sites that were primarily selected to
          test specific features of the proposed HRS.  As such, these values are not necessarily
          representative of the greater universe of CERCLA sites.
                                            5-16

-------
                                       FIGURE 5-12
                       GROUND WATER PATHWAY DOLLAR COST
                                (Based on Field Test Sites)



Number
of Sites
Within
Range





6
c
j
4

3


2
£.
1
1
0
	


































1
1
1
1
1
n

<10 10- 20-
20 30
                                                                      Sites without d ri 11 i ng
                                                                      Sites with drilling
                               30-
                               40
40-
50
50-
60
60-
70
70-
80
80-
90
90-
100
100-
110
110-
120
120-
130
130-
140
140-
150
                                                                                          >150
                                 Range of Dollar Costs (x 1,000)
Note:    These values are based upon findings from field test sites that were primarily selected to
         test  specific features of the proposed  MRS.  As such, these values are not necessarily
         representative of the greater universe of CERCLA sites.
                                            5-17

-------
three to nine, though one site involved 14 wells. The average was about four or five wells  No more
than three boreholes were drilled at any one site.  For these 15 sites, LOE ranged from 140 to 1,360
hours and averaged 480 hours. The one site involving more than 850 hours included an unusually
comprehensive drilling program (the 14 wells).  The dollar cost at these sites ranged from $22,000 to
$156,000 and averaged $59,000.  As shown in  Figure 5-12, only the site with the extensive drilling
program exceeded $100,000.

For the nine sites where a drilling program was not costed, total ground water  pathway LOE ranged
from 50 to 240 hours, with an average of 120 hours.  Six of these were in the range of 50 to 100 hours.
Dollar costs ranged from about $3,000 to $26,000, though all but one site were in the range of $3,000
to $17,000. The average was about $11,000.  The higher-cost sites generally involved larger numbers
of samples from nearby domestic and/or municipal wells.

The implementation of a drilling program clearly had significant impacts on overall  costs, in terms of
both LOE and dollars.  The sites where drilling occurred required subcontract administration and in-
field supervision of the borehole/well installation which involved 40 to 660 LOE hours,  along with
$5,000 to $73,000 in FIT costs and subcontractor charges. Sites where drilling occurred also tended to
have a higher number of ground water samples and  associated costs, averaging 11  ground water
samples. These ranged from  two to 40, with all but two sites having 15 or fewer samples.  By
comparison, sites without drilling ranged from zero to 16 ground water samples (from municipal and
residential wells) and averaged around five or six.  All but two  of these sites had eight  or fewer
samples. Dollar costs associated with sample collection and  CLP  analysis averaged $16,000 at sites
with boreholes and/or monitoring wells and $6,000 at sites without.

5.3.5     Proposed Surface Water Pathway

The surface water pathway was next in overall cost, contributing eight percent of average  total  site
LOE and 13 percent of total dollars. Figures 5-13 and 5-14 show the distributions.

Of the 24 sites  included in  the cost database, four were shown not to have  viable surface water
pathways (e.g., lack of nearby surface water, no overland flow route or flood potential); one of these
nevertheless involved extensive sampling.  LOE for  these four sites ranged from 10 to 87 hours;  the
three involving six or fewer samples were all under 50 hours.  The dollar cost for the three sites with
few samples ranged from $500 to $7,000. The dollar cost at the fourth site was $ 18,000.

For the  20 sites that did have viable surface water  pathways, LOE ranged from 70 to 290 hours and
averaged 170 hours. Most sites fell into the range of about 70 to about 220 hours. The dollar cost of
these 20 sites showed a range from $10,000 to $42,000 with an average of about  $25,000.   The
broad range of dollar costs is primarily attributable to the range of samples, from a low of six to as
many as 28 (average of 15). Sampling to establish an observed release involved the full range of six to
28 samples at these sites. In many cases, the surface water body  itself constituted  a  recreational or
environmental target and sampling for observed release served  the dual purpose of sampling for
potentially contaminated targets. Specific sampling of discrete surface water targets was limited at
the field test sites, as illustrated below:


                                                           Range of        Average No.
                                         No. of Sites        Samples         of Samples

    Observed Release                          20              6-28              14
    Drinking Water Targets                       1                2                2
    Human Food Chain Targets                   2              2-4                3
    Recreation Targets                          3              2-4                3
    Environmental Targets                       3              2-5                4
                                            5-18

-------
                                        FIGURE 5-13
                         SURFACE WATER PATHWAY LOE HOURS
                                 (Based on Field Test Sites)
         4
Number
of Sites  2
Within
 Range
Sites without a viable pathway
Sites with a viable pathway
                <25
26-
50
51-
75
76-
100
101-
125
126-
150
151-
175
176-
200
201-
225
226-
250
251-
275
                                                                                     >275
                                      Range of LOE Hours

 Note:     These values are based upon findings from field test sites that were primarily selected to
          test specific features of the proposed HRS.  As such, these values are not necessarily
          representative of the greater universe of CERCLA sites.
                                            5-19

-------
                                     FIGURE 5-14
                     SURFACE WATER PATHWAY DOLLAR COST
                              (Based on Field Test Sites)
Number
of Sites
Within
Range
u
7
6
5
4
3
2
1
n








• si,

                      Sites without a viable pathway

                      Sites with a viable pathway
             <5
                     5-10
10-15
                                        15-20
                 20-25
                                                         25-30
30-35
                                                                          35-40
                  >40
                               Range of Dollar Costs (x1,000)
Note:     These values are based upon findings from field test sites that were primarily selected to
         test  specific features of the proposed MRS. As such, these values are not necessarily
         representative of the greater universe of CERCLA sites.
                                         5-20

-------
 5.3.6     Proposed Air Pathway

 The LOE range for the air pathway was 20 to 270 hours, averaging 90 hours (Figure 5-15). Fifteen sites
 involved no collection of analytic air samples and these ranged from about 20 to about 90 hours and
 averaged about 50 hours.  Two sites outside of this range involved  FASP sampling at 140 and 170
 hours.  Seven sites involved CLP air sampling and these showed a wide range from 60 to 270 hours
 with an average of 150 hours. The majority of these, however, were clustered between 110 and 140
 hours.

 For sites without CLP air sampling and analysis, dollar costs were modest at less than $5,000 for 15 of
 the  17 sites (averaging less than $3,000); the two sites involving FASP air sampling and analysis had
 costs between $7,000 and about $8,000. The seven sites with CLP sampling and analysis ranged from
 $8,000 to $44,000  and averaged $25,000.  Five of these, however, were in the range of $8,000  to
 about $27,000 (Figure 5-16). The primary determinant of the range of dollar costs was the number of
 samples collected.  This ranged from five to 23 and averaged  13;  the five sites under $25,000 ranged
 between five and 16 samples, averaging 10 samples.

 5.3.7     Proposed Onsite Exposure Pathway

 At the field test sites, the onsite exposure pathway was generally the least costly of the proposed HRS
 pathways to evaluate. This observation is at least partly due to the fact that much of the information
 required to evaluate this pathway could be collected for, or in conjunction with, data collection for
 other pathways. The attribution of costs was probably somewhat related to the physical structure  of
 the cost reporting  form and the means by which project participants kept track of costs.  Since the
 onsite pathway is new to the HRS, is unfamiliar, and was placed at the end  of the cost reporting form,
 much of the cost associated with onsite pathway factors could easily have been attributed to other
 pathways where that information was also employed.  For example, nearby  population data were
 generally  derived from air pathway target population data, and accessibility/frequency of use could
 be assessed  during a reconnaissance or via general observations which  may have been costed
 elsewhere.  Costs associated  with  other factors, including onsite pathway-specific waste quantity,
 toxicity, and sampling to establish onsite observed contamination (with the  exception of samples
 taken from resident population properties) have  been aggregated under site,  source, and  waste
 characterization. The result is that costs specifically attributable to the onsite exposure pathway were
 generally  quite modest at these sites.  LOE and dollar cost  distributions for the onsite  exposure
 pathway are shown in Figures 5-17 and 5-18, respectively.

 There were some instances where FIT felt that a resident population threat was  possible, yet sampling
 was not conducted due to access problems or community relations concerns. In such cases, the costs
 of such  samples were estimated and added during the data refinement  process.  For 15  sites, no
 sampling of adjacent residential, school or day care properties, or terrestrial sensitive environments,
 was costed; nine sites did include the cost of such sampling.  The sites  without resident population
 samples ranged in  LOE from less than 10 to about  30 hours.  The dollar costs ranged from $300 to
 $1,700 and averaged $800.

 The number of resident population samples costed at the other nine sites ranged from one to 15 and
 averaged six. LOE ranged from less  than 20 to 340  hours and averaged 80 hours.  Dollar costs
 remained proportional to the number of samples and ranged  from about $3,000 to $32,000 with an
 average of $10,000.  Most field test sites involved eight or fewer samples, less than 100 hours of LOE,
and total dollar costs under $11,000. Only one site - which had a  very high resident population
threat, a number of contaminated  residential properties,  and  associated  community  relations
concerns-- involved higher levels of sampling, LOE, and dollar costs.
                                           5-21

-------
                                      FIGURE 5-15
                              AIR PATHWAY LOE HOURS
                               (Based on Field Test Sites)
u
C
j
4
Number
of Sites 3
Within
Range
2
£.
1
n

















































•
•LI
<25 26- 51- 76-
50 75 100
                                                               E
                Sites without CLP samples
                Sites with CLP samples
                                            101-
                                            125
126-
150-
151-
175
176-
200
201-
225
226-
250
                                                                                      >250
                                      Range of LOE Hours
Note:    These values are based upon findings from field test sites that were primarily selected to
         test  specific features of the  proposed HRS.  As such, these values are not necessarily
         representative of the greater universe of CERCLA sites.
                                          5-22

-------
                                      FIGURE 5-16
                             AIR PATHWAY DOLLAR COST
                               (Based on Field Test Sites)
Number
of Sites
Within
Range
IO
16
14
12
10
8
6
4
2
n













                                                                B
                                            Sites without CLP samples

                                            Sites with CLP samples
               <5
5-10
10-15     15-20     20-25     25-30
30-35
35-40    >40
                                       Range of Dollar Costs (xl.OOO)
Note:    These values are based upon findings from field test sites that were primarily selected to
         test  specific features of the proposed HRS.  As such, these values are not necessarily
         representative of the greater universe of CERCLA sites.

-------
                                      FIGURE 5-17
                       ONSITE EXPOSURE PATHWAY LOE HOURS
                               (Based on Field Test Sites)
         14


         12


         10


Number   8
of Sites
Within
 Range    °


          4


          2


          0
g
Sites without resident population samples
Sites with resident population samples
                 <25   26-   51-    76-   101-   126-  151-  176-  201-  226-  251-  276-  301-
                      50    75    100   125   150   175   200   225   250   275   300   325
                                     >325
                                   Range of LOE Hours
 Note:     These values are based upon findings from field test sites that were primarily selected to
          test specific features of the proposed HRS.  As such, these values are not necessarily
          representative of the greater universe of CERCLA sites.
                                          5-24

-------
                                        FIGURE 5-18
                        ONSITE EXPOSURE PATHWAY DOLLAR COST
                                 (Based on Field Test Sites)
Number
of Sites
Within
 Range
1 U
14
12
10
8
6
4
2
n
—








                                                      Sites without resident population samples
                                                      Sites with resident population samples
                   <5
5-10      10-15
15-20
20-25
25-30     >30
                                 Range of Dollar Costs (x1,000)
  Note:    These values are based upon findings from field test sites that were primarily selected to
           test specific features of the  proposed HRS.  As such, these values are not necessarily
           representative of the greater universe of CERCLA sites.
                                            5-25

-------
5.4   SUMMARY OF SITE COSTS

Table 5-1 summarizes the range of costs, and their averages, as discussed in preceding sections.  The
24 field test sites required between 970 and 3,310 LOE hours, involved 34 to 98 CLP samples, and cost
a total of $100,000 to $311,000.  Since not all sites required a full evaluation of each pathway, Table
5-1 presents the costs incurred for the common  alternatives,  i.e., sites with or without: CLP air
sampling; monitoring well or borehole installation; a viable surface water pathway;  and resident
population sampling.

One of the most significant influences on the overall cost of these site inspections was the number of
CLP samples collected at a site.  While the LOE involved in sample collection and handling (sample
management, packaging, chain of custody paperwork, shipping)  represents only 16 percent of total
site LOE (Figure 5-19), the dollar cost associated with samples (which is mostly CLP analytic charges)
represents nearly one-half (48 percent) of average total site dollar cost (Figure 5-20). Any reductions
in CLP sample collection within the proposed  MRS pathways, and for purposes of site, source, and
waste characterization, could have significant impacts on overall site costs.

A second major cost element was the installation of monitoring wells which, as discussed in Section
5.3.4, tends to be very expensive in terms of both dollars and LOE, and also results in greater numbers
of CLP samples.  This can be readily seen in the comparisons given in Table 5-1. Limiting the number
of wells installed could reduce overall costs substantially.

The largest cost center for LOE seems to be concentrated among  the non-pathway-specific and non-
data-generating elements of the general tasks category, which accounts for one-half of average total
site LOE.  This cost analysis has attempted to account for the*  effects of a proposed MRS "learning
curve" as it impacts the elements of project planning and  management and proposed HRS scoring
and documentation; there may or may not be additional efficiencies to be found here as personnel
become more accustomed to working with the proposed model,  (tt should also be pointed out that,
while no attempt was made  to account for similar learning efficiencies among the individual data
elements within the proposed HRS pathways, the possibility of their realization seems likely).

Other factors which bear consideration when interpreting these cost results include the following:

      •   As has been indicated elsewhere in this report, the sites  involved  in the field test were not a
         random selection from among the universe of Superfund sites.  Some sites were selected
         because they had particular characteristics that were of interest from the perspective of
         proposed HRS testing.  It is likely that this particular group of  sites is, as a  whole,  more
         complex than a random grouping would be.

      •   The site inspections conducted during the field test were quite comprehensive, and each
         pathway was fully evaluated at most sites. This may not be necessary at all sites, and early
         recognition of low scoring potential for a particular  pathway or threat could be  useful in
         focusing FIT efforts and controlling costs.

      •   As explained in Section 5.1, the costs reported here represent all costs incurred for all data
         development activities that occurred at a site,  regardless of  source (FIT, TAT, PRPs, state
         agencies), provided that those data were both representative of the type of  data that FIT
         would develop and were consistent with FIT data  needs in support of  proposed  HRS
         scoring. Consequently, while the costs reported here are meant to be representative of the
         total cost of data acquisition at these sites, they may  in some  cases overstate the direct
         costs.
                                            5-26

-------
General Tasks
Site/Source/Waste Characterization
Air Pathway
    w/CLP sampling
    w/o CLP sampling
Ground Water Pathway
    w/drilling
    w/o drilling
Surface Water Pathway
    w/viable pathway
    w/o viable pathway
Onsite Exposure Pathway
    w/resident population sampling
    w/o resident population sampling
                                                      TABLE 5-1
                                              SUMMARY OF SITE COSTS
                                              (Based on Field Test Sites)
                                              LOE Hours
                                    Dollars
                                                CLP Samples
Range
480-
80-
60-
20-
140-
50-
70
10
20
10
1.500
530
270
170
1.360
240
290
-90
-340
-30
Average
960
280
150
60
480
120
170
40
80
20
Range
$39.000 -
$20,000 -
$8.000 -
$900-
$22.000 -
$3.000 -
$10,000
$500-
$3,000
$300-
84,000
79.000
44,000
8,000
156,000
26.000
- 42,000
18,000
32,000
1.700
Average
$
$
$
$
$
$
$
$
$
$
59,000
40.000
25.000
3,000
59.000
11,000
25.000
7.000
10,000
800
Range
3-17
10-46
5-23
NA
2-40
0-16
6-28
0-14
1-15
NA
Average
11
24
13
NA
11
5
15
5
6
NA
 Total Site*
970-3,310
1,860
$100,000-311.000     $176,000
34-98
63
*  Represents actual total site costs for the 24 sites, not the sum of individual line items.
Note:  These values are based upon findings from field test sites that were primarily selected to test specific features of the proposed MRS
       As such, these values are not necessarily representative of the greater universe of CERCLA sites.

-------
                                    FIGURE 5-19
                         PROPORTION OF TOTAL SITE LOE
                             (Based on Field Test Sites)
                Sample Collection
                      16%
                                                         All Other Tasks
                                                             84%
Note:     These values are based upon findings from field test sites that were primarily selected to
         test specific features of the proposed HRS.  As such, these values are  not necessarily
         representative of the greater universe of CERCLA sites.
                                         5-28

-------
                                      FIGURE 5-20
                         PROPORTION OF TOTAL SITE DOLLARS
                               (Based on Field Test Sites)
                                                    Sam pie Col lection
                                                          9%
          All Other Tasks
               52%
                                                                   CLP Analysis
                                                                       39%
Note:    These values are based upon findings from field test sites that were primarily selected to
        test  specific features of the proposed HRS.  As such, these values are not necessarily
        representative of the greater universe of CERCLA sites.
                                          5-29

-------
The process of collecting information and evaluating sites for the purpose of supporting
MRS  scoring is one  that  would typically involve a series of increasingly focused
investigations.  This was not the case with the field test sites, nor was this approach
reflected in the cost reporting and refinement process. Rather, the costs discussed here (as
has been described above and in Section 5.1) have been normalized to reflect a common
framework for comparison which  approximates the entire sequential process of site
investigation and evaluation (starting from  "little available  data").  The costs presented
here should be viewed within this context.
                                   5-30

-------
                                        SECTION 6

                           FIELD TEST MRS SCORING RESULTS
 6.0   INTRODUCTION

 This section discusses the MRS scoring results for the 29 sites involved in the field test. Data generated
 from  the site inspections performed  during field testing were used to develop both current and
 proposed MRS scores. Among the sites included in the field test were the following:

      •  Sites with features not addressed by the current HRS (e.g., human food chain exposures,
         potential air releases, onsite exposures, human exposures through recreational  surface
         water use).

      •  Sites with different types of sources (e.g., landfills, surface impoundments, waste piles).

      •  High-scoring and low-scoring sites under the current HRS.

 The initial proposed HRS scoring package for each site was reviewed  by the Agency and several of its
 contractors. Written comments were prepared for each initial scoring package and forwarded to the
 Regional EPA/FIT  office  upon conclusion of the respective Regional visit. These comments often
 consisted of items that were derived from the  initial  reviews,  including scoring package
 documentation requirements.  However, the review process for field test sites was not as extensive
 (nor as well defined) as the review process for sites being considered for possible placement on the
 NPL.

 Significant  highlights regarding the field test  HRS scoring results that are presented in this section
 include:

      •  The distribution of proposed HRS site scores for the field test sites.

      •  The distribution of individual pathway scores under the proposed HRS for the field test
         sites.

      •  The distribution of surface water threat scores (i.e., drinking water,  human  food chain,
         human recreation, and environmental) under the proposed  HRS for the field test sites.

      •  The comparison between current and proposed HRS  site scores,  as well  as individual
         pathway scores, for the field test sites.

      •  The general characteristics of those field test sites that scored relatively high or low under
         individual proposed HRS pathways.

Although the field test results provide the Agency with  useful information, there are significant
limitations  on the conclusions regarding HRS scores that can be drawn from these results.  The sites
tested were not randomly selected but were in part selected for potential risks to public health and
the environment that are not evaluated under  the current HRS. Therefore, the proposed HRS scores
associated with these sites are  not necessarily  representative of the scores that other CERCLA sites
would generate.

Due to the small number of sites tested and the  fact that the sites tested were not  randomly selected,
the field  test results should not be construed to fully  reflect  the general relationships  between
proposed HRS scores and current HRS scores.
                                            6-1

-------
6.1   FIELD TEST SCORES UNDER THE PROPOSED HRS

The distribution of proposed HRS site scores for sites tested is shown in Figure 6-1. The distributions
of the scores for the individual proposed HRS pathways are shown in Figures 6-2 through 6-5. Not all
pathways were evaluated for each site due to:

      •  The absence of nearby surface water or the absence of an overland flow segment to
         surface water (six sites).

      •  The lack of observed contamination as defined in the onsite exposure pathway (one site).

      •  The assignment of a value of zero for the containment factor under a particular  pathway
         (one site).

Overall scores for the 29 sites ranged from 15 to  71, with an  average of 46 and a median of 53.
Average and median  scores for the individual proposed  HRS pathways and overall site scores are
summarized in Table 6-1.

                                       TABLE 6-1
        AVERAGE AND MEDIAN PROPOSED HRS PATHWAY AND SITE SCORES
                              (Based on Field Test Sites)
Pathway or Site Score
Air
Ground water
Surface water
Onsite exposure
Overall site score
No. of Sites
Tested
28
29
23
28
29
Average Score
22
40
65
40
46
Median Score
22
31
62
25
53
Range
3
4
13
0
15
of Scores
- 47
- 100
- 100
- 100
- 71
Note:  These values are based upon findings from field test sites that were primarily selected to test
       specific features of the proposed HRS.  As such, these values are not necessarily representative
       of the greater universe of CERCLA sites.

For the sites tested, the proposed surface water pathway had the highest average and median scores
while the proposed air pathway had the lowest average and median scores.  The air pathway also had
the smallest range of individual pathway scores under the proposed HRS.

For the proposed surface water pathway, the distributions of scores for each surface water threat
among the sites tested are given  in Figures 6-6 through  6-9.  Average and median scores for the
various surface water threats are summarized in Table 6-2.  Scores for these threats are normalized to
a scale of 0-100 in these figures for the purposes of comparison.  Of the 23 sites for which the surface
water pathway was evaluated, most scored relatively low for the drinking water, human recreation,
and environmental threats; the human food  chain threat scores were higher because about  60
percent of the sites tested had significant human food  chain target values (Figure 4-1).  In nearly  all
cases, the presence of nearby surface waters resulted in the identification of some human food chain
targets since most waters supported  aquatic life.  Targets were  not always present for the other
surface water threats. As a result, for sites tested, the remaining surface water threats normally did
not have target values that were as high as those target values for the human food chain threat.
                                           6-2

-------
Number
of Sites
         10
          8
          0
                                         FIGURE 6-1
                                PROPOSED MRS SITE SCORES
                                (Based on 29 Field Test Sites)
                 0-10    11-20   21-30   31-40   41-50  51-60  61-70  71-80   81-90  91-100

                                             Site Scores
  Note:     These values are based upon findings from field test sites that were primarily selected to
           test  specific features of the proposed MRS.  As such, these values are  not necessarily
           representative of the greater universe of CERCLA sites.
                                             6-3

-------
                                       FIGURE 6-2
                         PROPOSED HRS AIR PATHWAY SCORES
                              (Based on 28 Field Test Sites)
         12


         10


          8


          6
Number
of Sites
Note:
         0-10  11-20  21-30  31-40  41-50  51-60  61-70   71-80   81-90   91-100

                             Pathway Scores


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed  HRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                           6-4

-------
Number
of Sites
 Note:
                                       FIGURE 6-3
                  PROPOSED MRS GROUND WATER PATHWAY SCORES
                              (Based on 29 Field Test Sites)
          8
          0
        0-10   11-20   21-30  31-40   41-50  51-60  61-70  71-80  81-90  91-100

                                  Pathway Scores


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed MRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                           6-5

-------
                                            FIGURE 6-4
                       PROPOSED MRS SURFACE WATER PATHWAY SCORES
                                   (Based on 23 Field Test Sites)
         10
          8
Number
of Sites
                  0-10   11-20   21-30  31-40  41-50  51-60  61-70   71-80  81-90  91-100

                                            Pathway Scores
      Note:    These values are based upon findings from field test sites that were primarily selected to
               test specific features of the proposed HRS.  As such, these values are not necessarily
               representative of the greater universe of CERCLA sites.
                                                6-6

-------
                                            FIGURE 6-5
                      PROPOSED HRS ONSITE EXPOSURE PATHWAY SCORES
                                   (Based on 28 Field Test Sites)
         10
          8
Number   6
of Sites
          0
        Note:
  0-10  11-20  21-30  31-40  41-50  51-60  61-70  71-80  81-90  91-100

                           Pathway Scores

These values are based upon findings from field  test sites that were primarily selected to
test  specific features of the proposed HRS. As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                                6-7

-------
Number
of Sites
                                      FIGURE 6-6
                     PROPOSED MRS SURFACE WATER PATHWAY:
                          DRINKING WATER THREAT SCORES
                             (Based on 23 Field Test Sites)
20
18
16
14
12
10
 8
 6
 4
 2
 0
                  0-10  11-20  21-30   31-40  41-50  51-60  61-70  71-80 81-90 91-100
                                          Threat Scores
 Note:    These values are based upon findings from field test sites that were primarily selected to
         test  specific features of the proposed MRS.  As such, these values are  not necessarily
         representative of the greater universe of CERCLA sites.
                                          6-8

-------
                                       FIGURE 6-7
                      PROPOSED MRS SURFACE WATER PATHWAY:
                              FOOD CHAIN THREAT SCORES
                              (Based on 23 Field Test Sites)
Number
of Sites
                 0-10   11-20  21-30  31-40  41-50  51-60  61-70  71-80  81-90  91-100

                                           Threat Scores


  Note:    These values are based upon findings from field test sites that were primarily selected to
          test  specific features of the proposed MRS.  As such, these  values are not necessarily
          representative of the greater universe of CERCLA sites.
                                           6-9

-------
Number
of Sites
20
18
16
14
12
10
 8
 6
 4
 2
                                     FIGURE 6-8
                     PROPOSED MRS SURFACE WATER PATHWAY:
                            RECREATION THREAT SCORES
                             (Based on 23 Field Test Sites)
                  0-10  11-20  21-30   31-40  41-50  51-60  61-70  71-80  81-90   91-100
                                          Threat Scores
Note:     These values are based upon findings from field test sites that were primarily selected to
         test  specific features of the proposed MRS.  As such, these values are not necessarily
         representative of the greater universe of CERCLA sites.
                                         6-10

-------
Note:
                                     FIGURE 6-9
                    PROPOSED HRS SURFACE WATER PATHWAY:
                         ENVIRONMENTAL THREAT SCORES
                            (Based on 23 Field Test Sites)
Number
of Sites


18
16
14
12
10
8
6
4
2
n
••




—
—







nil FfSllPP! F~^~l ••'-•:•• .
        0-10   11-20  21-30  31-40   41-50  51-60   61-70  71-80  81-90  91-100

                                Threat Scores


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed HRS. As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                        6-11

-------
                                      TABLE 6-2
     AVERAGE AND MEDIAN PROPOSED HRS SURFACE WATER THREAT SCORES
                            (Based on 23 Field Test Sites)
Threat
Drinking water
Human food chain
Human recreation
Environmental
Average Score
5
49
9
22
Median Score
0
52
1
3
Range of Scores
0 - 100
0 - 100
0 - 67
0 - 100
Note:  These values are based upon findings from field test sites that were primarily selected to test
       specific features of the proposed HRS. As such, these values are not necessarily representative
       of the greater universe of CERCLA sites.


6.2   COMPARISON OF PROPOSED AND CURRENT HRS SCORES

Under the current HRS, overall scores for the 29 sites ranged from 3 to 51, with an average of 30 and a
median of 30.  Average and median scores for the individual current HRS pathways and overall site
scores are summarized in Table 6-3. For the field test sites, site scores were generally higher under the
proposed HRS than under the current HRS.  Only two sites scored 51 or above under the current HRS;
16 scored 51 or above under the proposed HRS (Figure 6-10).


                                      TABLE  6-3
        AVERAGE AND MEDIAN CURRENT HRS PATHWAY AND SITE SCORES
                             (Based on Field Test Sites)

Pathway or Site Score
Air
Ground water
Surface water
Overall site score
No. of Sites
Tested
28
29
23
29

Average Score
4
48
9
30

Median Score
0
50
9
30

Range of Scores
0 - 42
0 - 100
0 - 16
3 - 51
Note:  These values are based upon findings from field test sites that were primarily selected to test
       specific features of the proposed HRS. As such, these values are not necessarily representative
       of the greater universe of CERCLA sites.


For the field test sites, the ground water pathway in the current HRS had the highest average and
median scores, while the current HRS air  pathway had the lowest average and median scores. The
surface water pathway had the smallest range of individual pathway scores under the current HRS.

Comparative distributions of  current and proposed HRS pathway scores are given in  Figures 6-11
through 6-13.  Average and median pathway scores as well as overall site scores under the proposed
and current HRS are summarized in Table 6-4.
                                         6-12

-------
Number
of Sites
 Note:
 12


 10


  8


  6


  4
           0
                                      FIGURE 6-10
                         PROPOSED AND CURRENT HRS SCORES
                              (Based on 29 Field Test Sites)
                                                                        Key:
                                                                               Current HRS

                                                                               Proposed HRS
                  0-10    11-20  21-30  31-40  41-50  51-60  61-70  71-80   81-90   91-100

                                            Site Scores
These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed HRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                          6-13

-------
Dumber
of Sites
Note:
  28


  24


  20


  16


  12


   8
                                     FIGURE 6-11
                                AIR PATHWAY SCORES:
                            PROPOSED AND CURRENT MRS
                             (Based on 28 Field Test Sites)
                                                                      Key:
                                                                           Current MRS

                                                                           Proposed MRS
           0-10   11-20  21-30  31-40  41-50  51-60  61-70  71-80  81-90  91-100

                                   Pathway Scores


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed HRS. As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                         6-14

-------
Number
of Sites
 Note:
                                     FIGURE 6-12
                         GROUND WATER PATHWAY SCORES:
                            PROPOSED AND CURRENT MRS
                             (Based on 29 Field Test Sites)
                                                                      Key:
                                                                          Current MRS

                                                                          Proposed MRS
        0-10   11-20   21-30  31-40   41-50  51-60  61-70   71-80  81-90  91-100

                                  Pathway Scores

These values are based upon findings from field test sites that were primarily selected to
test  specific features of the proposed  MRS.  As such, these values are not  necessarily
representative of the greater universe of CERCLA sites.
                                         6-15

-------
Number
of Sites
14


12


10


 8


 6


 4


 2


 0
                                     FIGURE 6-13
                         SURFACE WATER PATHWAY SCORES:
                            PROPOSED AND CURRENT HRS
                             (Based on 23 Field Test Sites)
                                                                      Key:
                                                                           Current HRS

                                                                           Proposed HRS
  Note:
        0-10  11-20  21-30   31-40  41-50  51-60  61-70  71-80  81-90 91-100

                                 Pathway Scores


These values are based upon findings from field test sites that were primarily selected to
test  specific features of the  proposed HRS.  As such, these values are not necessarily
representative of the greater universe of CERCLA sites.
                                          6-16

-------
                                      TABLE 6-4
                AVERAGE AND MEDIAN PATHWAY AND SITE SCORES
                     UNDER THE PROPOSED AND CURRENT HRS
                              (Based on Field Test Sites)
Pathway or
Site Score
Air
Ground water
Surface water
Onsite exposure
Overall site score

Proposed
22
40
65
40
46
Averaqe Scores

HRS Current HRS
4
48
9
NA
30
Median
Proposed HRS
22
31
62
25
53
Scores
Current HRS
0
50
9
NA
30
Note:  These values are based upon findings from field test sites that were primarily selected to test
       specific features of the proposed HRS. As such, these values are not necessari ly representative
       of the greater universe of CERCLA sites.


6.3    SCORING SUMMARY

This scoring analysis, conducted by comparing HRS scores for sites evaluated under both the current
and proposed HRS, is based on the 29 field test sites.  For these sites:

      •  Surface water  scores were highest among the individual pathway  scores under the
         proposed HRS while, under the current HRS, ground water scores were highest among the
         individual pathway scores.

      •  Proposed surface water  pathway scores were, in turn,  usually dominated  by the human
         food chain threat.

      •  Surface water and  air pathway scores were generally higher with the proposed HRS than
         with the current HRS.

      •  Ground water pathway scores were generally lower with the proposed HRS than with the
         current HRS.

      •  Overall site scores were generally higher with the proposed HRS than with the current HRS.

Sites scoring higher than the  average and median proposed air pathway scores among the field test
sites generally had maximum  hazardous waste quantity factor values, significant target population or
sensitive environment factor  values (but not necessarily both together), and documented observed
air releases or maximum gas mobility factor values. Sites scoring lower than the average and median
proposed air pathway scores generally had relatively smaller target  population or  sensitive
environment factor values.

For the proposed ground water pathway, sites scoring higher than the average and median pathway
scores among  the field test sites generally had maximum  hazardous waste quantity factor values,
significant potential ground water target population or Level I population (i.e., those populations
which are drinking water contaminated above health-based benchmarks) factor values, and
documented observed ground water releases or maximum depth to aquifer/hydraulic conductivity
matrix values.  Sites scoring lower than the average and median proposed ground water pathway
                                          6-17

-------
scores generally had smaller potential target population factor values, lower hydraulic conductivity
factor values, or lower MEI factor values.

Field test sites scoring higher than the  average and median proposed surface water pathway scores
were generally characterized by higher hazardous waste quantity factor values, higher distance to
surface water factor values, and documented observed surface water releases. This group of sites also
had at least two surface water threats that received significant scores, one of which was always the
human food chain threat. The highest scoring surface water threats among the sites tested generally
had significant target population (i.e., drinking water or recreation) or sensitive environment factor
values specific to those  threats.  In many cases, these targets also met the proposed MRS criteria for
actual contamination (human recreation and environmental threats only).

Sites that scored lower under the proposed surface water pathway  generally had  at least three
surface water threats that received small scores, two of which were usually the human food chain and
environmental threats.  In most cases, the lack of targets or the small hazardous waste quantity factor
values for these threats resulted in lower scores.

For the onsite exposure pathway, the field test sites that scored higher than the average and median
pathway scores were generally characterized  by significant resident or nearby  human target
population factor values and higher waste quantity factor values. Sites that scored lower generally
had relatively smaller resident or nearby target population factor values.

Because the field test sites were primarily selected to test new components included in the proposed
HRS, the ability to extrapolate the field test results to the greater universe of CERCLA  sites is limited.
Overall site scores in  the field test, for example,  tended to be higher under the proposed HRS than
under the current HRS, but this would not necessarily hold true for CERCLA sites in general. However,
the scoring results of this study do provide a useful measure of how actual environmental data
perform within the framework of the proposed HRS.
                                            6-18

-------
                                       SECTION 7

                              MAJOR SOURCE MATERIALS
U.S. Environmental Protection Agency.  1982. National Contingency Plan. Appendix A; Final Rule. 47
Federal Register 32220. Washington, D.C. July 16.
        . 1988. Background Information:  Proposed Revisions to Hazard Ranking System. HW-10.5.
Washington, D.C. November.
        . 1988. Technical Support Document: Revised Hazard Ranking System. Washington, D.C.
December.
        . 1988.  National Contingency Plan. Appendix A; Proposed Rule. 53 Federal Register 51962.
Washington, D.C. December 23.
	. 1989.  SI/HRS Information Bulletin. Issue No. 2. OSWER Directive No. 9200.5-302.
Washington, D.C. April.
                                                   «U.S. GOVERNMENT PRINTING OFFICE :1 990 -71.8 -1 5 9/0 0 it 7 3

                                           7-1

-------