Standardizing the standards

Much has been written about the use of risk-based cleanups at sites in state-run voluntary cleanup programs. Many states have established, uniform, risk-based cleanup standards for soil and groundwater to remediate brownfields - real estate stigmatized by real or perceived contamination - and other sites to levels that are consistent with the intended property use. Additionally, the American Society of Testing and Materials (ASTM) has developed guidance on the development and use of risk-based cleanup criteria at underground storage tank (UST) sites, and is working on guidance for chemical release sites. The advantage of these guidelines is that they are user-friendly, incorporating numeric reference tables where one can compare the site data with published cleanup levels to see if a site is "clean."

All of these efforts have enabled the cleanup of numerous sites to "safe" levels as defined by technical professionals (as in the case of ASTM) or state agencies (as in the case of state voluntary cleanup programs). But what about sites that are being investigated and cleaned up under the jurisdiction and oversight of the U.S. Environmental Protection Agency (EPA)?

Although risk-based cleanups are encouraged, the EPA does not have uniform, risk-based cleanup standards for its sites under the Resource Conservation Recovery Act (RCRA) or the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA), commonly referred to as Superfund. The status of risk-based cleanups at EPA cleanup sites is reviewed along with a RCRA corrective action case history that applies the principles of site-specific risk assessment to justify a "no further action" determination for the site.

Overview of EPA risk-based cleanups
EPA-directed cleanups generally fall into two major categories. These include cleanups at National Priorities List (NPL or CERCLA) sites and RCRA closure or corrective action sites. At these sites, the investigation and cleanup procedures and techniques are very prescriptive; however, the manner in which the cleanup levels are derived is not. Derivation of the cleanup levels is subject to much interaction and negotiation between the parties.

There are two main components to the derivation process. The first is the establishment of the delineation level. The delineation level is the level to which the extent of the contamination is defined. Cleanup levels are then defined as the levels of contaminants allowed to remain at the site without further remedial action.

Historically, sites within the RCRA program have been required to utilize either background levels (for naturally occurring constituents), detection limits or maximum contaminant levels (MCLs) for synthetic constituents as cleanup levels for soil and groundwater. These cleanup standards are very conservative; most sites cannot be remediated to these standards because of technological or economic limitations. There are provisions under RCRA for the establishment of site-specific, alternate concentration limits (ACLs), which can be risk-based. However, few facilities have established ACLs.

Historically, Superfund has used comprehensive risk assessments to establish cleanup levels for NPL sites by evaluating all exposure pathways and calculating a site-specific safe level within the 10-4 to 10-6 risk range. The procedures for such a risk assessment are detailed in the EPA guidance document Risk Assessment Guidance for Superfund (RAGS), which is used to develop risk-based preliminary remediation goals (PRGs) for these sites. The PRGs are the precursors to cleanup levels, and can be used to determine if a site warrants additional investigation and/or remediation. It is important to note that RAGS is for human-health based risk assessment; an ecological assessment is often also utilized to determine if the PRGs computed by RAGS are also ecologically protective. The problem with the historical approach to deriving cleanup levels for RCRA and Superfund sites is that the process is not user-friendly, and involves numerous exposure assumptions and contaminant specific calculations.

EPA is attempting to change this process to encourage more cleanups under its RCRA and CERCLA programs. In fact, EPA Region III and Region IX have each published risk-based concentration (RBC) and PRG reference tables to assist in the development of delineation and cleanup levels for sites in these regions. These tables have been developed using EPA's Soil Screening Guidance, published in 1996. This guidance provides relatively simple methodology to calculate risk-based soil screening levels (SSLs) that are used to determine if a site needs further investigation. Although the first sentence of the SSL fact sheet states that "SSLs are not national cleanup standards," the SSL guidance provides a roadmap to establish risk-based cleanup levels at Superfund and RCRA sites.

The SSL guidance and regional reference tables provide numeric, risk-based screening levels that are determined to be protective of human health, based on a prescribed land use and risk tolerance. In general, these approaches use published contaminant-specific toxicological data and conservative default values as input parameters into the RAGS risk assessment equations to derive a screening level. Changing the input parameters to better reflect industrial or residential exposure routes and pathways can determine industrial and residential screening levels. More site-specific risk assessment can be performed, if necessary, by inputting site-specific input parameters into the RAGS equations to yield more site-specific screening or cleanup levels. The following case history describes a simple screening process at a RCRA site to determine if RCRA corrective action is warranted.

RCRA corrective action case history
A steel manufacturing facility in EPA Region IV was required to undergo RCRA corrective action as a requirement of its RCRA post-closure permit. As a part of this permit, the facility was required to investigate numerous solid waste management units (SWMUs) for the presence of soil contamination. A RCRA facility investigation (RFI) workplan was prepared that described the investigation and data evaluation techniques to be used to determine if the SWMUs were impacted. The application of risk-based delineation and cleanup levels was proposed in this workplan and accepted by EPA. The Region III and SSL guidance referenced above was used to establish risk-based delineation levels for shallow and deep soil. Risk-based concentrations (RBCs) from the Region III reference tables were used to evaluate shallow soil data, while SSLs from the EPA guidance were used to evaluate deep soils. Soil ingestion was the primary pathway of concern for the shallow soil, while leaching to groundwater was the pathway of concern f or deep soil. Exposure via dermal contact and air emissions was eliminated from consideration by site-specific conditions.

The workplan included drilling a total of 63 borings throughout the SWMUs and analyzing over 200 samples for various hazardous constituents. Upon receipt of the laboratory data, a four-step data evaluation process was undertaken to determine if a SWMU required further investigation or remediation. This four-step process is shown in Figure 1.

In the first step, the data was compared to background delineation levels for naturally occurring constituents and detection limits for synthetic constituents. Background delineation levels were computed as twice the naturally occurring background concentrations, as determined by on-site sampling of background areas. If the data showed concentrations that were less than these levels, no further action was required, and the parameter was eliminated from consideration. If concentrations were greater than these levels, the area was considered to be impacted, and the evaluation proceeded to step 2.

In step 2, the contaminant concentrations were compared to the RBCs. If the results exceeded the RBCs, the area was then subjected to further evaluation and potential remediation. Additionally, if the data exceeded the background delineation level, groundwater quality data from that SWMU was reviewed to determine if the parameter of concern was present in the groundwater. If present, then the groundwater pathway was considered, and the deep soil data was compared to the SSLs. If the SSLs were not exceeded, that parameter was eliminated from consideration. If the SSL was exceeded, the area was then subjected to further evaluation and potential remediation.

The data evaluation process for this site revealed that of over 200 samples collected, 159 samples exceeded the background delineation levels, demonstrating that each of the SWMUs had been impacted. Further evaluation using the EPA risk-based values revealed that only three samples exceeded the industrial RBCs, 10 exceeded the residential RBCs and four exceeded the SSLs. The risk-based evaluation reduced the number of samples for further evaluation from 159 down to 17, with most results only slightly over the RBCs or SSLs. This information was submitted to EPA, which concluded that, based on the low frequency of detection and the relative magnitude of the exceedances, no further action was required.

Summary
Risk-based cleanups are receiving more and more attention at both the state and federal agency levels. The use of user-friendly reference tables makes data evaluation and data reduction much simpler, without compromising environmental protection. These approaches enable facilities to focus resources on sites that truly represent an environmental threat, while providing an exit mechanism for those sites that do not. Efforts by EPA Regions III and IX should be used to promote national cleanup standards that can be used throughout the country.

Click here to post comments about this topic, and read what others have to say.

This article originally appeared in the December, 1999 issue of Environmental Protection magazine, Vol. 10, Number 12, pp. 38-41.

This article originally appeared in the 12/01/1999 issue of Environmental Protection.

comments powered by Disqus