|
The purpose of the Triad Approach field investigation at Area Requiring Environmental Evaluation (AREE) 34 was to bring closure to the investigation phase of this site and to provide a complete conceptual site model (CSM) to allow subsequent remedial decisions to be made. A CSM is a basic description of how contaminants enter a system, how they are transported within the system, and where routes of exposure to organisms and humans occur. Through systematic project planning (SPP) and using real-time measurement systems to perform a HRSC under a DWS, the project team was able to perform 1 week of field work in November 2003 to collect sufficient data to resolve the identified uncertainty in the CSM. As a result of this project, a proposed site remedy was tentatively agreed upon by the Army and the site regulators (the Virginia Department of Environmental Quality [VDEQ]) within 2 months of completion of the Triad field investigation. The majority of the project team that executed this project had not previously used the Triad Approach.
| Site Name | Vint Hill Farms Station Area Requiring Environmental Evaluation (AREE) 34 |
| Location | Warrenton, VA |
| Site Type |
Contaminated Aquifer - Contamination Source Unknown |
| Project Lead Organization | U.S. Army |
| Project Lead Type |
U.S. Army Lead |
| Regulatory Lead Program |
Base Realignment and Closure (BRAC) |
| Triad Project Status | Field Program Completed |
| Reuse Objective Identified |
Yes |
| Proposed Reuse: |
Commercial/Industrial |
The former Vint Hill Farms Station (VHFS) is located approximately 40 miles southwest of Washington, D.C., in Fauquier County, Virginia, and approximately 11 miles northeast of Warrenton, Virginia. While active, VHFS primarily functioned as an Army installation engaged in communications intelligence. VHFS closed in September, 1997, pursuant to the Base Realignment and Closure (BRAC) program. All of the former VHFS property has been transferred to the Vint Hill Economic Development Authority (VHEDA) or to Fauquier County as part of a public benefit conveyance. The former VHFS property was transferred in three parcels, with the final transfer occurring on May 22, 2003.
Prior to the final deed transfer, AREE 34, located in the large parking lot behind Building 2400, was found to have low-level chlorinated volatile organic compound (VOC) contamination in the saturated overburden and underlying fractured bedrock aquifer. However, the source and extent of the VOC contamination detected in the saturated overburden material at AREE 34 was unknown. Further, since one of the drinking water production wells serving the former VHFS property is located approximately 400 feet away from AREE 34 in a hydraulically downgradient direction with respect to the general direction of groundwater flow in the bedrock aquifer, the VDEQ wanted to quickly determine the threat, if any, to the drinking water supply. The fact that the property had already been deeded to the VHEDA resulted in increasing pressure to resolve the contamination issue as quickly as possible. The site is part of an on-going redevelopment project involving a sustainable, mixed-use community.
The initial CSM for the AREE 34 site was based on historical information and data from previous investigations. Hydrogeologic data indicate the groundwater in the saturated overburden moves slowly to the northwest (<0.4 feet/year). Groundwater in the fractured bedrock aquifer flows to the east-southeast. Chlorinated VOC contamination in the saturated overburden had migrated into the fractured bedrock aquifer along fractures.
Chlorinated VOC contamination in the saturated overburden is sporadic, apparently resulting from one or more small spills. More than one source of contamination was suspected because previous investigation data indicated PCE as the predominant contaminant in some areas while carbon tetrachloride (CCl4) was the predominant contaminant in other areas. The saturated overburden consists of mainly fine-grained materials with small occurrences of coarser-grained material. These were determined to be the controlling factors in the highly heterogeneous distribution of contamination in the saturated overburden.
Although the site geology is highly heterogeneous, previous investigations indicated that the total area of contamination in the saturated overburden is likely fairly small (approximately 0.5 acre). However, the final permanent monitoring well installations during the supplemental remedial investigation (SRI) indicated that the groundwater contamination in the saturated overburden was not bounded in the upgradient (southeast) direction. The VOC contamination in MW34-4, the most upgradient saturated overburden monitoring well on site, was found to contain the highest concentrations of chlorinated VOC contaminants. This was interpreted to mean that there was a potential source of chlorinated VOCs upgradient of this location.
Based on the initial CSM, the largest uncertainty to be addressed by the Triad project was to locate the source(s) of chlorinated VOCs in the saturated overburden. This information was needed to determine the long-term threat to the fractured bedrock aquifer, which was apparently in communication with the contaminated saturated overburden. Because the saturated overburden is encountered at shallow depths ranging from 4.77 to 14.13 feet below ground surface (bgs) in the contaminated portion of AREE 34, the location of the source(s) of VOC contamination in the saturated overburden also needed to be addressed due to the potential threat to construction workers and the potential for vapor intrusion (VI) into any newly constructed buildings which may exist on the site in the future.
The project team agreed that high-resolution chlorinated VOC data were required to address the uncertainty in the initial CSM. It was also agreed that speciation of individual VOC compounds was needed to determine the potential locations of different chlorinated VOC sources in the saturated overburden. Due to budget limitations, the vertical spatial boundary of this investigation was limited to the saturated overburden.
During the Triad field investigation, a variety of on site, real-time measurement systems (i.e., MIP and DSITMS), in combination with conventional soil and groundwater sample collection and analysis techniques, using a DWS were used to build strong collaborative data sets in order to refine the CSM for AREE 34. Each data set collected during the Triad field investigation collaborated with all of the other data sets to build an accurate final CSM as documented in the Summary Report. Utilizing the collaborative data sets, the following observations regarding the VOC contamination in the overburden at AREE 34 were made to refine the CSM:
The initial on-site DSITMS results of the passive diffusion bag (PDB) groundwater samples provided higher vertical resolution of contaminant distribution within the existing monitoring well screens sampled. These data confirmed the initial CSM assumption of highly heterogeneous distribution of contaminants in the subsurface at the site. In addition, the PDB groundwater sample results from MW34-4 were used to determine that the MIP/Electron-Capture Detector (ECD) could provide data of adequate sensitivity for intended on-site decisions.
The CSM evolved by the end of the Triad field investigation adequately described chlorinated VOC contaminant distribution within the overburden at AREE 34. The sampling conducted during the Triad field investigation, the RI, and the SRI has established the boundaries of the VOC contamination at AREE 34.
By using on-site, real-time measurement systems (i.e., the MIP/ECD and DSITMS), thousands of data points were collected during the 1 week Triad field investigation, creating a high-density data set that provided a high level of confidence that project team members required to agree the investigation of the site was complete. The result of this HRSC effort was a mature CSM for the site that confirmed the initial CSM with a high degree of certainty. That is, the VOC contamination present in the overburden at AREE 34 is the result of two or more localized sources, possibly small spills of solvent in the immediate vicinity of AREE 34, and is limited to a 0.5 acre area. This final CSM was supported by the presence of CCl4 as the primary VOC contaminant at grid point J16 and the lack of CCl4 at grid point H14.
Off-site laboratory analytical data were required for risk assessment purposes. Because of the high level of confidence in the CSM achieved as a result of the HRSC with the MIP/ECD and the on-site DSITMS, the project team was able to limit the number of soil and groundwater samples collected for off-site laboratory analysis while ensuring that the samples were fully representative of specific site conditions. Six samples were split for on-site DSITMS analysis by EPA Method 8265 and off-site laboratory analysis by EPA Method 8260 (four PDB groundwater samples and two temporary monitoring well groundwater samples). The results of these analyses indicated a very strong correlation between the two data sets (r2 = 0.89) and 46% low bias for the DSITMS results compared to the results by EPA Method 8260. The cause of the low bias for the on-site DSITMS results could not be determined as all quality control (QC) results for both methods were within acceptable criteria. This low bias for the on-site DSITMS results is also not consistent with numerous other split data sets between the two analytical methods (for example, see Davis et al., 1998). However, the on-site DSITMS data were adequate to determine the location(s) of highest VOC contamination for off-site laboratory analysis to support risk assessment. Since the two data sets indicated a very strong correlation and consistent bias, the use of the on-site DSITMS data for field decision-making was not adversely affected. Therefore, on-site DSITMS analytical results were shown to provide reliable information on the extent of VOC contamination and on the locations of the highest concentrations of VOCs in the saturated overburden at AREE 34, thus providing an appropriate mechanism for ensuring that representative samples were collected for the risk assessment.
Further, by comparing the high-density data generated during the Triad field investigation with previously collected groundwater data from temporary and permanent monitoring wells, off-site laboratory temporary monitoring well groundwater analytical results collected during the Triad field investigation were shown to be representative of site conditions. Since off-site laboratory temporary monitoring well groundwater results were available for the two locations having the highest concentrations of VOCs (i.e., grid points H14 and J16), these results were deemed adequate and sufficient for evaluating risks associated with the VOC contamination in the overburden groundwater at AREE 34, thereby eliminating the need for additional permanent saturated overburden monitoring wells.
Based on the results presented above, the objectives of the Triad field investigation were achieved as follows:
An important observation related to this Triad field investigation relates to the utility of the real-time measurement systems. At the outset of the field work, it was unclear whether the MIP could detect the levels of VOC contamination present at AREE 34. Contingency plans were built into the AREE 34 Work Plan to address this uncertainty so that the objectives of the Triad field investigation could be achieved without use of the MIP, if necessary. However, comparisons of on-site DSITMS soil and groundwater analytical results to the MIP/ECD logs showed excellent correlation for both detections and non-detections, proving the MIP/ECD capable of meeting project data quality objectives (DQOs). The MIP/ECD was not only effective in identifying contaminated areas, but it was also able to clearly locate "clean" areas.
Application of the Triad Approach at AREE 34 was deemed successful by the regulators and stakeholders. The collaborative data set, which took only 1 week in the field to generate using real-time measurement systems, filled the data gaps and confirmed the initial CSM. Confirmation of the CSM subsequently led to the rapid development of a proposed remedy for the site that included deed restrictions and long-term monitoring (LTM) of groundwater in the saturated overburden and the bedrock aquifer. This proposed remedy was tentatively agreed to by VDEQ within 2 months of completion of the Triad field investigation, with formal agreement due following the completion of a focused feasibility study/proposed plan (FS/PP) and installation and initial sampling of the LTM wells. Of significant note is that the final CSM developed during the Triad field investigation sufficiently characterized the contamination in the saturated overburden to allow selection of a proposed remedy which has been deemed protective of the bedrock aquifer.
The objectives were to:
The Triad offered a cost-effective approach to gather the high-resolution information necessary to ensure that confident decisions could be made in an efficient and timely manner. Investigation costs and mobilizations were minimized.
The cost and schedule savings using the Triad Approach were significant. The Army had spent over $500,000 and 3 years working to arrive at the point where the Triad Approach was implemented. From that point on, the Army spent approximately $75,000 in developing the Systematic Planning Outline and AREE 34 Work Plan, and another $135,000 for the field work, all of which was accomplished in 11 months. Including the final analysis and Summary Report, the total Triad effort cost approximately $260,000 and took 15 months. At the end of this expedited time frame, a tentative agreement on a proposed remedy for the site was reached with the regulators.
As a comparison, had this site continued using the conventional iterative investigation approach used previously, it would have likely taken at least another 3 years and $500,000, assuming the sampling plan successfully resulted in samples being collected in the right places. This assumption is based on the need to put in at least another 26 temporary monitoring wells followed by installation and sampling of an additional 4 to 6 permanent monitoring wells based on the results of the temporary monitoring wells. These costs were avoided by using the Triad Approach as it investigated the designated area quickly and provided high-resolution data that supported CSM confirmation. By eliminating almost 2 years of conventional investigation, the Triad Approach not only saved money on investigative costs, it saved money on project lifecycle costs by reducing the costs of regulatory and Army efforts to manage the site while an extended investigation occurred. In addition, by resulting in a quicker resolution at AREE 34, the regulatory and Army personnel were allowed to focus their time and energy on other environmental cleanup sites.
The initial challenge was to quickly bring everyone onto an even playing field with respect to what was currently known about the site and the initial CSM. At the same time, the project team members had to be quickly educated about the Triad process. This was accomplished through numerous teleconferences and the development of a Systematic Planning Outline (see Electronic Documentation at the end of this profile). This Systematic Planning Outline provided a wide variety of information, including:
The initial CSM established during the systematic planning stage found that contamination existed in the saturated overburden groundwater as a result of one or more small spills of solvent in the immediate vicinity of AREE 34. The presence of VOC contamination in the bedrock aquifer existed as a result of transport of the contaminants from the saturated overburden since the overburden serves as a reservoir for slow recharge to the bedrock aquifer along fractures.
Development of the Systematic Planning Outline was an iterative process, and all project team members provided input during frequent teleconferences in which the entire project team participated. The Systematic Planning Outline provided a broad view of all potential data gaps based on the initial CSM as well as possible investigative tools to fill the data gaps. From this broad listing, the project team had the task of agreeing upon the final list of data gaps that had to be addressed to get a remedy in place, as well as the tools to fill those data gaps. This aspect of the project was the most difficult, even using the Triad approach, because of the number of project team members involved. The Systematic Planning Outline was the key to bridging the differences between all parties, as it provided a common foundation for the decision-making process and a means to communicate. Once the data gaps were agreed upon, the project objectives were to fill the data gaps quickly, efficiently, and cost effectively.
VDEQ personnel were intimately involved from the beginning of the Triad planning. However, due to the size of the project team, VDEQ was hesitant to communicate freely. To address this problem, additional "sidebar" meetings involving only the Army and VDEQ were held to ensure VDEQ's agreement with the data gaps, and thus the Triad field investigation objectives. As it turned out, the Systematic Planning Outline became an invaluable tool in these "sidebar" discussions between the Army and VDEQ. Based on the information presented in the Systematic Planning Outline, the VDEQ concurred with the initial CSM. Based on the initial CSM and the current and planned uses of the AREE 34 land, VDEQ's primary concerns were impact on the nearby drinking water production well, risk to construction workers during site development, and VI into buildings constructed on the site in the future. Based on these concerns, VDEQ wanted to see a highly focused list of data gaps filled to complete the investigation phase and proceed to the remedy for the site. These data gaps became the objectives of the Triad field investigation and were considerably focused compared to the initial list of data gaps identified in the Systematic Planning Outline. The data gaps, and thus objectives of the Triad field investigation, were as follows:
Since the AREE 34 SRI found that the highest VOC contamination was present in the most upgradient saturated overburden monitoring well (i.e., MW34-4), the first data gap listed was the primary gap that had to be filled.
The Systematic Planning Outline played a significant role in quickly bringing all project team members to a common understanding of existing site conditions and knowledge. It was the foundation for the development of the initial CSM and the identification of the uncertainty remaining in the initial CSM. Using the Systematic Planning Outline as a communication tool, the initial CSM and uncertainty in it were constantly refined in an iterative fashion until agreement was reached among project team members as to the final list of data gaps to be addressed and the appropriate investigation tools to be used to rapidly and cost effectively fill those data gaps. For example, the size of the area to be investigated was dramatically reduced as the planning process proceeded and the project team members were able to re-evaluate the known data using the Systematic Planning Outline. The application of the Triad Approach to this project, including the early involvement of the regulators, led to much quicker agreement regarding data gaps than would have been achieved by traditional investigative approaches.
The complex nature of the site led to a large degree of uncertainty associated with sample representativeness. Therefore, the AREE 34 DWS included the use of a variety of data collection tools, including real-time measurement systems and conventional sample collection and analysis techniques. The combination of tools was selected to address sampling uncertainties, generate collaborative data sets, and provide data suitable for use in risk assessment. The real-time measurement systems selected for the project included the MIP and the on-site DSITMS. The MIP, equipped with an ECD to measure chlorinated compounds and a Photoionization Detector (PID) to measure aromatic compounds, was selected to provide continuous in situ VOC sensing on a real-time basis. The MIP was also equipped to measure electrical conductivity (EC) on a continuous basis as it was advanced into the subsurface to monitor changes in subsurface lithology. Operation of the MIP, as specified in the AREE 34 Work Plan, was designed such that a portion of the MIP vapor effluent would be split, and the VOCs would subsequently be captured on a sorbent trap. While the ECD and PID measured total VOCs, the sorbent trap samples could be analyzed using the on-site DSITMS to speciate the VOCs to gain a better understanding of the distribution of individual VOCs across the site in real-time.
The DSITMS was selected as an on-site analytical tool for the Triad field investigation because, in addition to analyzing sorbent trap samples, it could provide real-time (3-minute analysis) quantitative measurement of individual VOCs in groundwater and soil samples which were collected using conventional techniques. Per the AREE 34 DWS, soil samples would be collected from soil cores produced using the Geoprobe® direct push technology (DPT) drill rig that was also used to advance the MIP. Groundwater samples would be collected from temporary monitoring wells and from PDB samplers. The PDB samplers were selected as a tool to determine vertical stratification of VOC contamination within the screened intervals of selected permanent monitoring wells at the site and to collaborate the results of the MIP/ECD.
The AREE 34 Work Plan also included contingency plans to prevent work stoppage if certain technical problems arose in the field. For example, in the event the MIP (along with on-site DSITMS analysis of the sorbent traps) was unable to detect the low levels of VOCs present at AREE 34, the AREE 34 DWS provided a contingency plan to use conventional techniques for collecting soil and groundwater samples for on-site DSITMS analysis. Although the contingency plan specified an alternative to the MIP, the same decision logic for selecting sampling locations still applied.
The AREE 34 Work Plan provided decision logic to be used during implementation of the Triad field investigation. The decision logic focused on the primary objective of the Triad field investigation which was to identify the highest concentration of VOCs in the overburden soil and groundwater. The purpose of the decision logic was to provide a guide to the field team to assist with real-time decision-making as new information became available. The decision logic provided a guide for selecting subsequent sampling locations based on a comparison of new results with the results obtained near monitoring well MW34-4. MW34-4 was used as a "benchmark" for decision-making since it was the location of the highest known concentration of VOCs in the saturated overburden at the outset of the Triad field investigation. The decision logic was a crucial tool in maximizing data collection efficiency because it allowed the project team to pre-plan field decisions to be made based on a variety of hypothetical scenarios. Thus, as data were generated during the field work, the decision-making process was streamlined for the field team. Since it would not be practical to plan for all possible scenarios, the decision logic directed the field team to communicate with the rest of the project team before proceeding under certain less likely scenarios.
The largest uncertainty associated with meeting the project objectives was sample support due to the previously documented subsurface stratigraphic heterogeneity within the saturated overburden. Sample support is a term that describes the physical properties of the sample that help determine what the analytical concentration results will be. Such physical properties include, but are not limited to, particle size, volume, and orientation of the sample collected and the volume of sample extracted for analysis. The term sample support in groundwater sampling includes the volume of water mixed in a screened interval that provides the sample. When the length of the screened interval brackets heterogeneous contaminant distributions in the subsurface, sample mixing occurs, resulting in variable contaminant concentrations which are often not representative for the types of decisions that need to be made at the site. This was the reason for collection of scale-appropriate, high-resolution data with the MIP/ECD and the DSITMS. High-resolution data collected using real-time field measurement techniques allowed the project team to address the uncertainty associated with sample support.
The large volume of data collected in a very short time frame posed data management and data communications challenges. During execution of the Triad field investigation, some of the core technical team members were deployed and others were available, on call by telephone. Since only some of the core technical team members and none of the key decision makers were deployed to the field, data generated during the Triad field investigation were posted nightly to the secure project website established and maintained by Columbia Technologies, the MIP services provider. The data generated during each day of the field effort were available the following morning on the project website for review by off-site key decision makers and core technical team members. Columbia Technologies also generated 3-D data visualizations of the MIP results and the on-site DSITMS groundwater results which were posted to the project website to help the project team members quickly digest the large volume of data collected. Two pre-planned teleconferences for the entire project team were used to review the data and modify the investigation strategy based on the revised CSM. Additionally, unscheduled teleconferences between selected off-site core technical team members and key decision makers were anticipated in the AREE 34 Work Plan and were used in the final 2 days of the Triad field investigation to ensure that all necessary data were collected prior to demobilization. As a result of the various teleconferences held during the 1 week Triad field investigation, all project team members were able to participate in the real-time decision-making process.
TQRS pending
The data provided by the real-time measurement systems was used, sometimes immediately, to support site decisions. Therefore, QC of the data collected was critical to the quality of site decisions. The quality of the data sets collected during the Triad field investigation was assured by adherence to Standard Operating Procedures (SOPs) and QC protocols proscribed in the AREE 34 Work Plan. The MIP was determined to be operating within acceptable control limits by analysis of aqueous QC check standards at midday and at the end of each day. This system response check also ensured that the MIP/ECD system was operating in a stable manner. The DSITMS was determined to be operating within acceptable control limits for soil, groundwater, and sorbent trap samples by performing daily quantitative calibration, continuing calibration check standards, blank analyses, and matrix spike analyses as specified in the SOPs and the Quality Assurance Project Plan (QAPP).
Columbia Technologies SmartData Solutions® system and website reporting.
November 17 to 24, 2003
336-454-0745To update this profile, contact Cheryl T. Johnson at Johnson.Cheryl@epa.gov or (703) 603-9045.
