The URL can be used to link to this page

Home My WebLink AboutNC0004978_FINAL_Marshall_CSA_Report_2018_20180129161P synTerra 2018 COMPREHENSIVE SITE ASSESSMENT UPDATE Site Name and Location: Marshall Steam Station 8320 East Carolina Highway 150 Terrell, North Carolina 28682 Groundwater Incident No.: Not Assigned NPDES Permit No.: NC0004987 Date of Report: January 31, 2018 Permittee and Current Duke Energy Carolinas, LLC Property Owner: 526 South Church St. Charlotte, NC 28202-1803 980-373-2129 Consultant Information: SynTerra 148 River Street, Suite 220 Greenville, South Carolina 29601 (864) 421-9999 Latitude and Longitude of Facility: N 36.42167/W 81.83500 N SF'`•y�I 2r O 7 >F ri -der, NC LG 2546 Aoject Geologi��,�, L}cENSF Kat Webb, NTT 132§[,4i O Proje Si �02@ ypj �l OG15� W �! a 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra DIVISION OF WATER RESOURCES Certification for the Submittal of a Comprehensive Site Assessment Responsible Party and/or Permittee: Duke Energy Carolinas, LLC Contact Person: Paul Draovitch Address: 526 South Church_ Sweet City: Charlotte State: NC Zip Code: 28202 Site Name: Marshall Steam Station Address: 8320 East North Carolina Highway 15IJ. City: Terrell State: NC Zip Code: 28682 Groundwater Incident Number (applicable): NA/ Coal Ash Management Act CSA I, Brian Wilker, a Professional Engineer/Professional Geologist (circle one) for SynTerra Corporation (firm or company of employment) do hereby certify that the information indicated below is enclosed as part of the required Comprehensive Site Assessment (CSA) and that to the best of my knowledge the data, assessments, conclusions, recommendations and other associated materials are correct, complete and accurate. (Each item must be initialed by the certifying licensed professional) 1. Y>L2 The source of the coal combustion residuals (contamination) has been identified. A list of all potential sources of the coal combustion residuals (contamination) is attached. 2. Imminent hazards to public health and safety have been evaluated. 3. _� —Potential receptors and significant exposure pathways have been identified, 4. —Geological and hydrogeological features influencing the movement of groundwater have been identified. The chemical and physical character of the contaminants has been identified. 5. The CSA sufficiently characterizes the cause, significance and extent of groundwater and soil contamination associated with the regulated coal ash management areas such that a groundwater Corrective Action Plan can be developed. If any of the above statements have been altered or items not initialed, provide a detailed explanation (See Attached). Failure to initial any item or to provide written justification for the lack thereof will result in immediate return of the CSA to the responsible party. A RO , o %7 �av dE N N�5 ►44►1I►►►u1, a►+�� P:\Duke Energy Carolinas\18. MARSHALL\CSA Update January 2018\FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Explanation of altered or items not initialed: Item 1. The CSA was specifically designed to assess the coal ash management areas of the facility. Sufficient information is available to prepare the groundwater corrective action plan for the ash management areas of the facility. Data limitations are discussed in Section 11 of the CSA report. Continued groundwater monitoring at the Site is planned. Item 2. Imminent hazards to human health and the environment have been evaluated. The NCDEQ data associated with nearby water supply wells is provided herein and is being evaluated. Item 5. The groundwater assessment plan for the CSA as approved by NCDEQ was specifically developed to assess the coal ash management areas of the facility for the purposes of developing a corrective action plan for groundwater. Other areas of possible contamination on the property, if noted, are anticipated to be evaluated separately. P: \ Duke Energy Carolinas \ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra WORK PERFORMED BY OTHERS • HDR Engineering, Inc. (HDR) of the Carolinas prepared the reports referenced herein under contract to Duke Energy. • The reports were sealed by licensed geologists or engineers as required by the North Carolina Board for Licensing of Geologists or Board of Examiners for Engineers and Surveyors. • The evaluations of hydrogeologic conditions and other information provided in this updated Comprehensive Site Assessment (CSA) are based in part on the work contained in the HDR documents and on sampling activities performed by Pace Analytical Services after the submittal of the HDR documents. The evaluations described in this paragraph meet requirements detailed in 15A NCAC 02L .0106(g). • SynTerra relied on information from the HDR reports as being correct. • The seal of the licensed geologist for this CSA applies to activities conducted and interpretations derived after the HDR reports were submitted. This submittal relies on the professional work performed by HDR and references that work. P: \ Duke Energy Carolinas \ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra EXECUTIVE SUMMARY ES.1 Source Information Duke Energy Carolinas, LLC (Duke Energy) currently owns and operates Marshall Steam Station (MSS or the Site), a coal-fired electricity -generating facility in the city of Terrell, Catawba County, North Carolina. The Comprehensive Site Assessment (CSA) update was conducted to refine and expand the understanding of subsurface conditions and evaluate the extent of impacts from historical management of coal ash in accordance with the North Carolina Coal Ash Management Act (CAMA). This CAMA CSA update contains an assessment of site conditions based on a comprehensive interpretation of geologic and sampling results from the initial site assessment and geologic and sampling results obtained subsequent to the initial assessment. Available groundwater data from monitoring wells associated with the Federal Coal Combustion Residuals Rule (CCR Rule) compliance program are also considered in data interpretations. However, the CCR data has not been fully incorporated into the analysis of this CSA due to the data only becoming available as of mid -January 2018. For example, analytical results from CCR Rule -specific monitoring wells are included on isoconcentration maps and analytical summary tables, but not integrated into detailed mathematical analysis, such as piper plots, box -and -whisker plots or background statistical calculations. MSS contains four units that generate electricity; operation of Unit 1 began in 1965, and operation of Unit 2 began in 1966, with each generating 350 megawatts. Operation of Unit 3 began in 1969, and operation of Unit 4 began in 1970, with each generating 648 megawatts. The MSS ash basin contains ash generated from the historic and active coal combustion at the Site. An earthen dike was constructed in 1965 at the confluence of Holdsclaw Creek that historically entered the Catawba River to create the ash basin. The MSS ash basin is located north of the steam station. The Catawba River was dammed by the construction of the Cowans Ford Dam between 1959 and 1964 to create Lake Norman, the largest man-made water body in North Carolina. Fly ash and bottom ash from MSS were sluiced to the ash basin from approximately 1965 until 1984. Fly ash precipitated from flue gas and bottom ash collected in the bottom of the boilers were sluiced to the ash basin using conveyance water withdrawn from Lake Norman. Since 1984, fly ash has mainly been disposed of in the on -site dry ash landfills and the sluicing of bottom ash to the ash basin has continued. The basin has a dendritic shape consisting of coves of deposited ash, dikes that impound ash in portions of the basin, and four main areas of ponded water. The area contained within the ash basin waste boundary is approximately 394 acres. Page ES-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra CAMA related source areas that lie partially or completely within the ash basin waste boundary include the Dry Ash Landfill (Phase II) (Permit 1804-INDUS-1983), Industrial Landfill No. 1 (Permit 1812-INDUS-2008), and the Photovoltaic (PV) Structural Fill. Other landfill areas located at the station beyond the ash basin waste boundary include the Dry Ash Landfill (Phase I) (Permit 1804-INDUS-1983), Flue Gas Desulfurization (FGD) Landfill (Permit 1809-INDUS-), Demolition Landfill (Permit 1804-INDUS-1983), and Asbestos Landfill (Permit 1804-INDUS-1983). Assessment results indicate the ash thickness in the areas of investigation varied. Ash was observed up to 85 feet thick within the ash basin, 111 feet within the dry ash landfill (Phase II), and 71 feet within the PV structural fill. The majority of ash contained in the ash basin is saturated. The contact between ash and underlying soil was distinct in each boring as physical intrusion of ash into the underlying soils appeared to have been negligible. Assessment findings determined that ash sluiced to, and accumulated within, the ash basin is the primary source of impact to groundwater. CCR in the additional ash management areas have also contributed to groundwater impact. The inferred extent of constituent migration from these sources based on evaluation of constituent concentrations greater than both water quality standards and background is shown on Figure ES-1. A detailed evaluation of constituent migration is included in this CAMA CSA update report. ES.2 Initial Abatement and Emergency Response Duke Energy has not conducted emergency response actions because groundwater impacts from the ash basin do not present an imminent hazard to human health or the environment requiring emergency response. Abatement activities conducted in preparation for ash basin closure include the design and construction of new lined retention basin systems to handle certain stormwater and wastewater at the Site. In addition, Duke Energy is in the process of converting to dry handling of bottom ash (anticipated by late 2018 or early 2019). ES.3 Receptor Information In accordance with North Carolina Department of Environmental Quality (NCDEQ) direction, CSA receptor survey activities include listing and depicting all water supply wells (public or private, including irrigation wells and unused wells) within a 0.5-mile radius of the ash basin compliance boundary. ES.3.1 Public Water Supply Wells Four public water supply wells were identified within a 0.5-mile radius of the MSS ash basin compliance boundary. One well is no longer in use; one is owned Page ES-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra by Duke Energy; the remaining two wells are classified as transient, non - community and are located south of the Site. ES.3.2 Private Water Supply Wells Water supply wells have been identified during receptor surveys. NCDEQ coordinated sampling at 38 water supply wells between February and October 2015. Ten (10) additional water supply wells near MSS were sampled between September 2016 and February 2017 by Duke Energy. A review of the analytical data for the 48 private water supply wells sampled indicated several constituents were detected above 2L or Interim Maximum Allowable Concentrations (IMAC) standards, including pH (33 wells), iron (five wells), manganese (two wells), zinc (one well), total dissolved solids (TDS) (one well), and vanadium (39 wells). Water supply wells are hydraulically upgradient of the ash basin. The lack of boron and low levels of CCR-related constituents (such as chloride and sulfate) indicate that ash basin operations have not impacted water supply wells. ES.3.3 Surface Water Bodies MSS is located in the Catawba River Basin on the upper end of Lake Norman. Groundwater influenced by the ash basin flows toward Lake Norman and the unnamed tributary that flows north to south on the eastern side of the ash basin. The surface water results collected from Lake Norman do not indicate that impacted groundwater associated with the MSS ash basin is causing 2B exceedances in Lake Norman. Lake Norman is used as a water supply for the greater Charlotte area. The two closest intakes for municipal water supply are located approximately 3.8 miles upstream and 5.6 miles downstream of MSS. ES.3.4 Land Use The area surrounding MSS generally consists of residential properties, undeveloped land, and Lake Norman. Properties located within a 0.5-mile radius of the MSS ash basin compliance boundary generally consist of undeveloped land and Lake Norman to the east, undeveloped land and residential properties located to the north and west, portions of the MSS site (outside the compliance boundary), undeveloped land, and residences to the south, and commercial properties to the southeast along North Carolina Highway 150. No change in land use surrounding MSS is currently anticipated. ES.4 Human Health and Ecological Risk Assessment This CAMA CSA provides an update to the 2016 human health and ecological risk assessment. There is no evidence of unacceptable risks to humans and wildlife at MSS attributed to CCR constituent migration in groundwater from the ash basins. The 2016 risk assessment identified potential risks under a Page ES-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra hypothetical recreational and subsistence fisher exposure scenario and to the great blue heron (for selenium and vanadium). This risk assessment update supports that the fisher risks were overestimated based on exposure and modeled fish tissue uptake assumptions. Further, surface water concentrations of selenium and vanadium did not exceed ecological screening values, and thus do not pose risks to the great blue heron. This update to the human health and ecological risk assessment supports a risk classification of "Low" for groundwater -related considerations. ES.5 Sampling/Investigation Results This CAMA CSA includes evaluations of the hydrogeological and geochemical properties of soil and groundwater at multiple depths and distances from the ash management areas. ES.5.1 Background Concentration Determinations Naturally occurring background concentrations were determined using statistical analysis for both soil and groundwater. Statistical determinations of provisional background threshold values (PBTVs) were performed in strict accordance with the revised Statistical Methods for Developing Reference Background Concentrations for Groundwater and Soil at Coal Ash Facilities (statistical methods document) (HDR and SynTerra, 2017). The current background monitoring well network consists of wells installed within three flow zones — shallow, deep, and bedrock. Background datasets for each flow system used to statistically determine naturally occurring concentrations of inorganic constituents in soil and groundwater are provided herein. As of October 11, 2017, NCDEQ approved a number of the statistically derived background values; however, others are still under evaluation and thus considered preliminary at this time. Background results may be greater than the PBTVs due to the limited valid dataset currently available. The statistically derived background threshold values will continue to be adjusted as additional data becomes available. ES.5.2 Nature and Extent of Contamination Site -specific groundwater constituents of interest (COIs) were developed by evaluating groundwater sampling results with respect to 2L/IMAC and PBTVs. In addition, the distribution of constituents in relation to the ash basin, co - occurrence with CCR indicator constituents such as boron, and likely migration directions based on groundwater flow direction are also considered in determination of groundwater COIs. The following list of groundwater COIs has been developed for MSS: Page ES-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 410 Antimony 67' Manganese ,61P Arsenic 67 Molybdenum 101 Barium 101 Nickel 17 Beryllium 17 Selenium 07 Boron 07 Strontium ,61P Cadmium 67' Sulfate ,61P Chloride 167 TDS ,67 Chromium (total) 67 Thallium ,67 Cobalt 67 Vanadium 01 Iron At MSS, boron is a key indicator of CCR groundwater impacts. Boron, in its most common forms, is soluble in water, and has a very low soil -water partitioning coefficient (Kd), indicating the constituent is highly mobile in groundwater. Boron is detected at concentrations greater than the 2L and PBTV beneath and downgradient (east) of the ash basin waste boundary, but is not detected in background groundwater. Therefore, the detection of boron in groundwater provides a close approximation of the distribution of CCR-impacted groundwater. The area farthest downgradient at which boron is detected at a concentration greater than the PBTV is interpreted as the leading edge of the CCR-derived plume moving downgradient from the source area. This leading edge is observed in the shallow and deep flow zones east of the ash basin between the Dry Ash Landfill (Phase I) and the unnamed tributary. In the bedrock flow units, boron is detected above 2L and the PBTV in downgradient wells near or beyond the ash basin compliance boundary east of the ash basin. The shallow and deep flow zones indicate COIs at concentrations greater than the 2L/IMAC or PBTV beneath and downgradient of the ash basin. Bedrock groundwater also exhibits CCR impacts, though to a lesser extent, and appears to be limited to areas beneath the Dry Ash Landfill (Phase II) and the ash basin dam. For the most recent valid sampling results available, bedrock monitoring wells installed beneath the ash basin or downgradient indicate exceedances of either PBTV or 2L/IMAC, whichever is greater, for barium, boron, chloride, chromium, iron, manganese, molybdenum, strontium, sulfate, and TDS. It is anticipated that additional monitoring wells for vertical delineation may be installed to support the groundwater corrective action planning process. Page ES-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra ES.5.3 Maximum Contaminant Concentrations (Source Information) The source at MSS includes CCR sluiced to and contained within the ash basin. Ash pore water samples collected from wells installed within the ash basin and screened in the ash layers have been monitored since 2015. The concentrations of detected constituents have been relatively stable with minor fluctuations. The ash basin is a permitted wastewater system; therefore, comparison of ash pore water within the wastewater treatment residuals (ash) to 2B or 2L/IMAC is used for source area information. Maximum observed COI concentrations at MSS, based on the September 2017 sampling event, are detected in ash pore water within the ash basin waste boundary and groundwater beneath the Dry Ash Landfill (Phase II). Soil samples collected below the ash/soil interface within the ash basin indicate arsenic, barium, chromium (total), iron, molybdenum, nickel, selenium, strontium, sulfate and vanadium concentrations in at least one soil sample exceeded the calculated soil PBTV and the NCDEQ Preliminary Soil Remediation Goals (PSRG) Protection of Groundwater (POG) value (if applicable). ES.5.4 Site Geology and Hydrogeology The subsurface at MSS is comprised of a surficial unit (soil, fill and reworked soil, alluvium, and saprolite), a transition zone, and fractured bedrock. The transition zone is comprised of partially weathered rock that is gradational between saprolite and competent bedrock. The bedrock is dominantly mica gneiss, meta -granite, and quartz-sericite schist. Shallow bedrock is fractured; however, only mildly productive fractures (providing water to wells) were observed within the top 50 - 75 feet of bedrock. The majority of fractures are relatively small (e.g., close and tight) and appear to be limited in connectivity between borings. Yields from pumping or packer testing are low. Groundwater exists under unconfined or water table conditions throughout the Site. Generally, groundwater is contained within fractures (secondary porosity) of the underlying bedrock. The hydrogeologic characteristics of the ash basin environment are the primary control mechanisms on groundwater flow and constituent transport. The stream valley in which the ash basin was constructed is a distinct slope -aquifer system in which flow of groundwater into the ash basin and out of the ash basin is restricted to the local flow regime. Localized topographic relief results in adjacent groundwater divides associated with the natural ridges separating historic draws. Groundwater flows generally from the northwest to southeast Page ES-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra across the Site. Active sluicing contributes to free-standing water within the ash basin and is controlled downgradient by the ash basin dam and the National Pollutant Discharge Elimination System (NPDES) outfall/discharge to Lake Norman (east side of ash basin). ES.6 Conclusions and Recommendations The investigation described in the CSA presents the results of the assessments required by the CAMA and 2L. Ash sluiced to, and accumulated within, the ash basin is determined to be a source of groundwater impacts at MSS. The assessment investigated the Site hydrogeology, determined the direction of groundwater flow from the ash basin, and determined the horizontal and vertical extent of impacts to groundwater and soil sufficient to proceed with preparation of a corrective action plan (CAP). Impacts to groundwater in all three flow zones have been identified beneath and downgradient of the ash basin at MSS. Boron is considered the CCR-impact indicator constituent detected in groundwater at concentrations greater than the PBTV and the 2L standard. Bedrock groundwater exhibits CCR impacts to a lesser extent, and appears to be limited to areas beneath the Dry Ash Landfill (Phase II) and northern portion of the ash basin dam. Additional data collection to support groundwater modeling and long- term monitoring is anticipated to support the corrective action planning process. Secondary sources have been identified in soil beneath the ash basin. Shallow soil impacts are anticipated to be addressed through basin closure and the CAP. Surface water receptors downgradient of the ash basin (e.g. Lake Norman) demonstrate compliance with 2B standards, with the occasional exception of dissolved oxygen, chloride, TDS, arsenic, selenium, cadmium (D), copper (D), and lead (D). Localized influence from NPDES permitted outfalls are likely contributing to these exceptions. The surface water results collected from Lake Norman do not indicate that impacted groundwater associated with the MSS ash basin is causing 2B exceedances in Lake Norman. Additional surface water and sediment data collection is anticipated to support the evaluation of potential monitored natural attenuation (MNA) in the area of the groundwater plume discharge into surface water. Information evaluated as part of the updated CSA indicates that the identified water supply wells are not impacted by ash basin operations; constituent concentrations detected in water supply wells are deemed to not be sourced from the ash basin. MSS's ash basin is currently designated as "Intermediate" risk under CAMA, meaning that closure of the ash basin is required by 2024. The updated CSA has determined no unacceptable risks to human health or wildlife from exposure to groundwater, surface Page ES-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra water, or sediment impacts related to the ash basin. The private water supply wells located upgradient of the Site are not impacted by the ash basin. A "Low" risk classification and closure via a cap -in -place scenario are considered appropriate as alternative water supplies are being provided in accordance with G.S. 130A- 309.213.(d)(1) of House Bill 630. A preliminary evaluation of groundwater corrective action alternatives is included in this CSA to provide insight into the CAP preparation process. The evaluation within the CAP will include predictive groundwater modeling to evaluate the effectiveness associated with various options. The source control (closure) options will be evaluated to determine the most technically, economically, and sustainably feasible means of controlling the CCR as a source to the groundwater flow system. For MSS, the primary source control (closure) method anticipated to be evaluated in the CAP consists of a cap -in -place scenario. This method would utilize an engineered, low permeability cover system over the ash basin to minimize infiltration and reduce leaching from the source area to groundwater. If a "Low" risk classification is determined, a well -designed capping system is expected to minimize ongoing constituent migration to groundwater. In addition to source control measures, the CAP will evaluate measures to address groundwater conditions associated with the ash basin. Groundwater corrective action by MNA is anticipated to be a remedy further evaluated in the CAP. As warranted, a number of viable groundwater remediation technologies such as phytoremediation, groundwater extraction, or hydraulic barriers may be evaluated based upon short-term and long-term effectiveness, implementability, sustainability and cost. Results of the evaluation, including groundwater flow and transport modeling, and geochemical modeling, will be used for remedy selection in the CAP. Page ES-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx p NORTH - AND POINT RD -d I . INDUSTRIAL �r .- - LANDFILL #1 LANDFILL - - - r: ASBESTOS LANDFILL _ ASH LANDFILL (PHASE II) PV ASH STRUCTURAL BASIN - } FILL BORROW - - - -x AREA ASH�t,. LANDFILL 4 (PHASE I) _ ACCESS ROAD _.,..- STRUCTURAL FILL,r - s r h �� -- - ASH BASIN } FGD LAKE NORMAN Z �, - f RESIDUE -_ LANDFILL ,f •r,�.' �- 1} Lw ~r MARSHALL •, STEAM - STATION LEGEND SHALLOW AREA OF CONCENTRATION IN GROUNDWATER ABOVE NC2L (SEE NOTE 5) DEEP ASH BASIN WASTE BOUNDARY APPROXIMATE LANDFILL WASTE BOUNDARY APPROXIMATE STRUCTURAL FILL BOUNDARY GENERALIZED GROUNDWATER FLOW DIRECTION BEDROCK ■ WATER SUPPLY WELL LOCATION STREAM WITH FLOW DIRECTION --'— DUKE ENERGY PROPERTY BOUNDARY VISUAL AID ONLY - NOTE: DEPICTION NOT TO SCALE FIGURE ES-1 1. SEPTEMBER16AERIAL2017.A RIALDATEDOCTHOTOGRAPHY BER2D2016. OGLE EARTH PRO ON SEPTEMBER 1 B, 201 ]. AERIAL DATED R ^ �A�r^ �Q WnT`.. APPROXIMATE EXTENT OF IMPACTS 148 RIVER STREET, SUITE 220 WHEELER 2. STREAMS OBTAINED FROM AMECFOSTERWHEELERNRTR,MAY 2015. FOSTER W NR M GREENVILLE, SOUTH CAROLINA 29601 MARSHALL STEAM STATION 3. GENERALIZED GROUNDWATER FLOW DIRECTION BASED ON SEPTEMBER 11, 2017 WATER LEVEL DATA. 4. PROPERTY BOUNDARY PROVIDED BY DUKE ENERGY. PHONE 869999 wwwsynterracorrracorp.com DUKE ENERGY CARO LI N AS, L LC TERRELL, NORTH CAROLINA /` DUKE � Y ENERG5. DRAWN BY: BENJAMIN YOUNG DATE:12/12/2017 PROJECT MANAGER: TODD PLATING GENERALIZED AREAL EXTENT OF MIGRATION REPRESENTED BY NCAC 02L LAYOUT: ES1 EXCEEDANCES OF MULTIPLE CONSTITUENTS IN MULTIPLE FLOW ZONES. CAROLINAS OS/15/20189:09 AM P:\Duke Ener Pro ress.1026 OOGIS BASE DATA Marshall Ma docs GSA -Su Iement2 3DGSM Marshall_3D_ES1.dw 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra TABLE OF CONTENTS SECTION PAGE ES.1 SOURCE INFORMATION....................................................................................... ES-1 ES.2 INITIAL ABATEMENT AND EMERGENCY RESPONSE ................................ ES-2 ES.3 RECEPTOR INFORMATION.................................................................................. ES-2 ES.3.1 Public Water Supply Wells................................................................................... ES-2 ES.3.2 Private Water Supply Wells................................................................................. ES-3 ES.3.3 Surface Water Bodies............................................................................................. ES-3 ES.3.4 Land Use.................................................................................................................. ES-3 ESA HUMAN HEALTH AND ECOLOGICAL RISK ASSESSMENT ...................... ES-3 ES.5 SAMPLING/INVESTIGATION RESULTS.......................................................... ES-4 ES.5.1 Background Concentration Determinations...................................................... ES-4 ES.5.2 Nature and Extent of Contamination.................................................................. ES-4 ES.5.3 Maximum Contaminant Concentrations (Source Information) ...................... ES-6 ES.5.4 Site Geology and Hydrogeology......................................................................... ES-6 ES.6 CONCLUSIONS AND RECOMMENDATIONS................................................ ES-7 1.0 INTRODUCTION.........................................................................................................1-1 1.1 Purpose of Comprehensive Site Assessment........................................................1-1 1.2 Regulatory Background...........................................................................................1-2 1.2.1 Notice of Regulatory Requirements (NORR)...............................................1-2 1.2.2 Coal Ash Management Act Requirements....................................................1-3 1.2.3 Coal Combustion Residuals Rule...................................................................1-4 1.3 Approach to Comprehensive Site Assessment.....................................................1-5 1.3.1 NORR Guidance................................................................................................1-5 1.3.2 USEPA Monitored Natural Attenuation Tiered Approach ........................1-5 1.3.3 ASTM Conceptual Site Model Guidance.......................................................1-6 1.4 Technical Objectives.................................................................................................1-6 1.5 Previous Submittals..................................................................................................1-7 Page i P: \ Duke Energy Carolinas \ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra TABLE OF CONTENTS (CONTINUED) 2.0 SITE HISTORY AND DESCRIPTION.....................................................................2-1 2.1 Site Description, Ownership, and Use History ..................................................... 2-1 2.2 Geographic Setting, Surrounding Land Use, and Surface WaterClassification................................................................................................. 2-2 2.3 CAMA-related Source Areas...................................................................................2-3 2.3.1 Ash Basin............................................................................................................2-4 2.3.2 Dry Ash Landfill............................................................................................... 2-4 2.3.3 FGD Landfill......................................................................................................2-5 2.3.4 Industrial Landfill No. 1................................................................................... 2-5 2.3.5 Photovoltaic Farm Structural Fill.................................................................... 2-6 2.4 Other Primary and Secondary Sources..................................................................2-6 2.4.1 Additional Landfills......................................................................................... 2-6 2.4.2 Secondary Sources............................................................................................ 2-6 2.5 Summary of Permitted Activities........................................................................... 2-7 2.6 History of Site Groundwater Monitoring.............................................................. 2-7 2.6.1 Ash Basin Voluntary Groundwater Monitoring .......................................... 2-8 2.6.2 Ash Basin NPDES Groundwater Monitoring...............................................2-8 2.6.3 Ash Basin CAMA Groundwater Monitoring ................................................ 2-9 2.6.4 Landfill Groundwater Monitoring................................................................. 2-9 2.7 Summary of Assessment Activities........................................................................ 2-9 2.8 Summary of Initial Abatement, Source Removal, or other CorrectiveAction.................................................................................................... 2-11 3.0 SOURCE CHARACTERISTICS................................................................................. 3-1 3.1 Coal Combustion and Ash Handling System.......................................................3-1 3.2 General Physical and Chemical Properties of Ash ............................................... 3-2 3.3 Site -Specific Coal Ash Data..................................................................................... 3-4 4.0 RECEPTOR INFORMATION.....................................................................................4-1 4.1 Summary of Receptor Survey Activities................................................................4-1 4.2 Summary of Receptor Survey Findings.................................................................4-3 4.2.1 Water Supply Lines..........................................................................................4-4 4.2.2 Public Water Supply Wells..............................................................................4-4 4.2.3 Private Water Supply Wells............................................................................4-4 Page ii P: \ Duke Energy Carolinas \ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall-CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra TABLE OF CONTENTS (CONTINUED) 4.3 Private Water Well Sampling..................................................................................4-4 4.4 Surface Water Receptors.......................................................................................... 4-7 5.0 REGIONAL GEOLOGY AND HYDROGEOLOGY...............................................5-1 5.1 Regional Geology......................................................................................................5-1 5.2 Regional Hydrogeology........................................................................................... 5-2 6.0 SITE GEOLOGY AND HYDROGEOLOGY............................................................6-1 6.1 Site Geology............................................................................................................... 6-1 6.1.1 Soil Classification.............................................................................................. 6-2 6.1.2 Rock Lithology.................................................................................................. 6-2 6.1.3 Structural Geology............................................................................................6-3 6.1.4 Soil and Rock Mineralogy and Chemistry .................................................... 6-4 6.1.5 Geologic Mapping............................................................................................. 6-5 6.1.6 Effects of Geologic Structure on Groundwater Flow ................................... 6-6 6.2 Site Hydrogeology.................................................................................................... 6-6 6.2.1 Hydrostratigraphic Layer Development....................................................... 6-7 6.2.2 Hydrostratigraphic Layer Properties............................................................. 6-7 6.3 Groundwater Flow Direction.................................................................................. 6-8 6.4 Hydraulic Gradient................................................................................................... 6-9 6.5 Facility Soil Data..................................................................................................... 6-10 6.5.1 Soil Beneath Ash Basin................................................................................... 6-10 6.5.2 Soil Beyond Waste Boundary and Within Compliance Boundary.......... 6-11 6.5.3 Comparison of PWR and Bedrock Results to Background ....................... 6-11 6.6 Hydraulic Conductivity......................................................................................... 6-12 6.7 Groundwater Velocity............................................................................................6-12 6.8 Contaminant Velocity............................................................................................. 6-13 6.9 Slug Test and Aquifer Test Results...................................................................... 6-14 6.10 Fracture Trace Study Results................................................................................. 6-15 6.10.1 Methods............................................................................................................6-16 6.10.2 Results...............................................................................................................6-17 Page iii P: \ Duke Energy Carolinas \ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra TABLE OF CONTENTS (CONTINUED) 7.0 SOIL SAMPLING RESULTS...................................................................................... 7-1 7.1 Background Soil Data............................................................................................... 7-1 7.2 Secondary Sources.................................................................................................... 7-2 8.0 SEDIMENT RESULTS.................................................................................................8-1 8.1 Sediment/Surface Soil Associated with AOWs.................................................... 8-1 9.0 SURFACE WATER RESULTS.................................................................................... 9-1 9.1 Comparison of Exceedances to 2B Standards.......................................................9-2 9.2 Discussion of Results for Constituents without Established 2B Standard ....... 9-3 9.3 Discussion of Surface Water Results...................................................................... 9-4 10.0 GROUNDWATER SAMPLING RESULTS............................................................10-1 10.1 Background Groundwater Concentrations.........................................................10-2 10.1.1 Background Dataset Statistical Analysis.....................................................10-3 10.1.2 Piper Diagrams — Background Wells...........................................................10-5 10.2 Downgradient Groundwater Concentrations.....................................................10-5 10.2.1 Monitoring Wells Beneath Ash Basin..........................................................10-5 10.2.2 Monitoring Wells Downgradient of Ash Basin..........................................10-6 10.2.3 Piper Diagrams - Wells Beneath and Downgradient from theAsh Basin...................................................................................................10-6 10.2.4 Radiological Laboratory Testing...................................................................10-7 10.3 Site -Specific Exceedances (Groundwater COIs).................................................10-7 10.3.1 Background Threshold Values (PBTVs)......................................................10-7 10.3.2 Applicable Standards.....................................................................................10-8 10.3.3 Additional Requirements...............................................................................10-9 10.3.4 Marshall COIs................................................................................................10-10 11.0 HYDROGEOLOGICAL INVESTIGATION..........................................................11-1 11.1 Plume Physical and Chemical Characterization................................................11-1 11.1.1 Plume Physical Characterization..................................................................11-1 11.1.2 Plume Chemical Characterization................................................................11-7 11.2 Pending Investigation(s)......................................................................................11-27 Page iv P: \ Duke Energy Carolinas \ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall-CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra TABLE OF CONTENTS (CONTINUED) 12.0 RISK ASSESSMENT..................................................................................................12-1 12.1 Human Health Screening Summary....................................................................12-2 12.2 Ecological Screening Summary.............................................................................12-4 12.3 Private Well Receptor Assessment Update.........................................................12-5 12.4 Risk Assessment Update Summary.....................................................................12-5 13.0 GROUNDWATER MODELING RESULTS...........................................................13-1 13.1 Summary of Flow and Transport Model Results...............................................13-2 13.2 Summary of Geochemical Model Results...........................................................13-3 14.0 SITE ASSESSMENT RESULTS................................................................................14-1 14.1 Nature and Extent of Contamination...................................................................14-1 14.2 Maximum COI Concentrations.............................................................................14-4 14.3 Contaminant Migration and Potentially Affected Receptors ...........................14-6 15.0 CONCLUSIONS AND RECOMMENDATIONS.................................................15-1 15.1 Overview of Site Conditions at Specific Source Areas......................................15-2 15.2 Revised Site Conceptual Model............................................................................15-2 15.3 Interim Monitoring Program.................................................................................15-3 15.3.1 IMP Implementation.......................................................................................15-4 15.3.2 IMP Reporting.................................................................................................15-4 15.4 Preliminary Evaluation of Corrective Action Alternatives...............................15-4 15.4.1 CAP Preparation Process...............................................................................15-5 15.4.2 Summary..........................................................................................................15-7 16.0 REFERENCES...............................................................................................................16-1 Page v P: \ Duke Energy Carolinas \ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall-CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF TABLES 2.0 Site History and Description Table 2-1 Well Construction Data Table 2-2 NPDES Groundwater Monitoring Requirements Table 2-3 Summary of Onsite Environmental Incidents 3.0 Source Characteristics Table 3-1 Soil/Material Properties for Ash Table 3-2 Ash, Rock, and Soil Composition Table 3-3 Cation/Anion Sample Results Table 3-4 Ash Sample SPLP Results 4.0 Receptor Information Table 4-1 Private Supply Well Property Information 6.0 Site Geology and Hydrogeology Table 6-1 Soil Mineralogy Results Table 6-2 Soil Chemistry Results, Elemental Composition and % Oxides Table 6-3 Soil and Ash Parameters and Analytical Methods Table 6-4 Ash Basin Surface Water, Pore Water and Seep Parameters and Analytical Methods Table 6-5 Transition Zone Mineralogy Results Table 6-6 Transition Zone Results, Elemental Composition and % Oxides Table 6-7 Whole Rock Chemistry Results, Elemental Composition and % Oxides Table 6-8 Petrographic Analysis Summary Table 6-9 Historic Water Level Data Table 6-10 Horizontal Groundwater Gradients and Velocities Table 6-11 Vertical Hydraulic Gradients Table 6-12 Hydrostratigraphic Layer Properties - Horizontal Hydraulic Conductivity Table 6-13 Hydrostratigraphic Layer Properties - Vertical Hydraulic Conductivity Table 6-14 Historic Slug Test Permeability Results Table 6-15 In -Situ Hydraulic Conductivity Results Table 6-16 Estimated Effective Porosity/Specific Yield and Specific Storage for Upper Hydrostratigraphic Units (A, F, S, M1, and M2) Table 6-17 Total Porosity, Secondary (Effective) Porosity/Specific Yield, and Specific Storage for Lower Hydrostratigraphic Units (TZ and BR) Page vi P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF TABLES (CONTINUED) Table 6-18 Field Permeability Test Results Table 6-19 Laboratory Permeability Test Results 7.0 Soil Sampling Results Table 7-1 Provisional Background Threshold Values for Soil Table 7-2 Potential Secondary Source Soil Analytical Results 10.0 Groundwater Sampling Results Table 10-1 Groundwater Provisional Background Threshold Values Table 10-2 Charge Balance Summary 11.0 Hydrogeological Investigation Table 11-1 Data Inventory Summary 13.0 Groundwater Modeling Results Table 13-1 Summary of Kd Values from Batch and Column Studies 15.0 Discussion - Conclusion and Recommendations Table 15-1 Sampling Parameters and Analytical Methods Table 15-2 Interim Monitoring Program List Page vii P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES Executive Summary Figure ES-1 Approximate Extent of Impacts 1.0 Introduction Figure 1-1 Site Location Map 2.0 Site History and Description Figure 2-1 Marshall Plant Vicinity Map Figure 2-2 1893 USGS Topographic Map Figure 2-3 NPDES Flow Diagram Figure 2-4 Site Layout 3.0 Source Characteristics Figure 3-1 Known Sample of Ash for Comparison Figure 3-2 Elemental Composition for Bottom Ash, Fly Ash, Shale, and Volcanic Ash Figure 3-3 Coal Ash TCLP Leachate Concentration Ranges Compared to Regulatory Limits Figure 3-4 Piper Diagram — Ash Pore Water Wells 4.0 Receptor Information - Need to confirm this section Figure 4-1 Ash Basin Underground Utility Map Figure 4-2 USGS Map with Water Supply Wells Figure 4-3 Water Supply Well Locations Figure 4-4 Water Intake Location Map 5.0 Regional Geology and Hydrogeology Figure 5-1 Tectonostratigraphic Map of the Southern and Central Appalachians Figure 5-2 Regional Geologic Map Figure 5-3 Interconnected, Two -Medium Piedmont Groundwater System Figure 5-4 Piedmont Slope -Aquifer System 6.0 Site Geology Figure 6-1 Figure 6-2 Figure 6-3 Site Geologic Map General Cross Section A -A' General Cross Section B-B' Page viii P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 6-4 General Cross Section C-C' Figure 6-5 Shallow Water Level Map - March 2017 Figure 6-6 Deep Water Level Map - March 2017 Figure 6-7 Bedrock Water Level Map - March 2017 Figure 6-8 Shallow Water Level Map - September 2017 Figure 6-9 Deep Water Level Map - September 2017 Figure 6-10 Bedrock Water Level Map - September 2017 Figure 6-11 Potential Vertical Gradient Between Shallow, Deep, and Bedrock Zones Figure 6-12 Topographic Lineaments and Rose Diagram Figure 6-13 Aerial Photography Lineaments and Rose Diagram 7.0 Soil Sampling Results Figure 7-1 Potential Secondary Source Soil Analytical Results 9.0 Surface Water Results Figure 9-1 Piper Diagram - Surface Water 10.0 Groundwater Sampling Results Figure 10-1 Piper Diagram - Background Wells Figure 10-2 Piper Diagram - Wells Beneath Ash Basin Figure 10-3 Piper Diagram - Downgradient Wells 11.0 Hydrogeological Investigation Figure 11-1 Isoconcentration Map - Antimony in Shallow Groundwater Figure 11-2 Isoconcentration Map - Antimony in Deep Groundwater Figure 11-3 Isoconcentration Map - Antimony in Bedrock Groundwater Figure 11-4 Isoconcentration Map - Arsenic in Shallow Groundwater Figure 11-5 Isoconcentration Map - Arsenic in Deep Groundwater Figure 11-6 Isoconcentration Map - Arsenic in Bedrock Groundwater Figure 11-7 Isoconcentration Map - Barium in Shallow Groundwater Figure 11-8 Isoconcentration Map - Barium in Deep Groundwater Figure 11-9 Isoconcentration Map - Barium in Bedrock Groundwater Figure 11-10 Isoconcentration Map - Beryllium in Shallow Groundwater Figure 11-11 Isoconcentration Map - Beryllium in Deep Groundwater Figure 11-12 Isoconcentration Map - Beryllium in Bedrock Groundwater Figure 11-13 Isoconcentration Map - Boron in Shallow Groundwater Page ix P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 11-14 Isoconcentration Map - Boron in Deep Groundwater Figure 11-15 Isoconcentration Map - Boron in Bedrock Groundwater Figure 11-16 Isoconcentration Map - Cadmium in Shallow Groundwater Figure 11-17 Isoconcentration Map - Cadmium in Deep Groundwater Figure 11-18 Isoconcentration Map - Cadmium in Bedrock Groundwater Figure 11-19 Isoconcentration Map - Chloride in Shallow Groundwater Figure 11-20 Isoconcentration Map - Chloride in Deep Groundwater Figure 11-21 Isoconcentration Map - Chloride in Bedrock Groundwater Figure 11-22 Isoconcentration Map - Chromium (VI) and Chromium (Total) in Shallow Groundwater Figure 11-23 Isoconcentration Map - Chromium (VI) and Chromium (Total) in Deep Groundwater Figure 11-24 Isoconcentration Map - Chromium (VI) and Chromium (Total) in Bedrock Groundwater Figure 11-25 Isoconcentration Map - Cobalt in Shallow Groundwater Figure 11-26 Isoconcentration Map - Cobalt in Deep Groundwater Figure 11-27 Isoconcentration Map - Cobalt in Bedrock Groundwater Figure 11-28 Isoconcentration Map - Iron in Shallow Groundwater Figure 11-29 Isoconcentration Map - Iron in Deep Groundwater Figure 11-30 Isoconcentration Map - Iron in Bedrock Groundwater Figure 11-31 Isoconcentration Map - Manganese in Shallow Groundwater Figure 11-32 Isoconcentration Map - Manganese in Deep Groundwater Figure 11-33 Isoconcentration Map - Manganese in Bedrock Groundwater Figure 11-34 Isoconcentration Map - Molybdenum in Shallow Groundwater Figure 11-35 Isoconcentration Map - Molybdenum in Deep Groundwater Figure 11-36 Isoconcentration Map - Molybdenum in Bedrock Groundwater Figure 11-37 Isoconcentration Map - Nickel in Shallow Groundwater Figure 11-38 Isoconcentration Map - Nickel in Deep Groundwater Figure 11-39 Isoconcentration Map - Nickel in Bedrock Groundwater Figure 11-40 Isoconcentration Map - Selenium in Shallow Groundwater Figure 11-41 Isoconcentration Map - Selenium in Deep Groundwater Figure 11-42 Isoconcentration Map - Selenium in Bedrock Groundwater Figure 11-43 Isoconcentration Map - Strontium in Shallow Groundwater Figure 11-44 Isoconcentration Map - Strontium in Deep Groundwater Figure 11-45 Isoconcentration Map - Strontium in Bedrock Groundwater Figure 11-46 Isoconcentration Map - Sulfate in Shallow Groundwater Figure 11-47 Isoconcentration Map - Sulfate in Deep Groundwater Page x P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 11-48 Isoconcentration Map - Sulfate in Bedrock Groundwater Figure 11-49 Isoconcentration Map - Total Dissolved Solids in Shallow Groundwater Figure 11-50 Isoconcentration Map - Total Dissolved Solids in Deep Groundwater Figure 11-51 Isoconcentration Map - Total Dissolved Solids in Bedrock Groundwater Figure 11-52 Isoconcentration Map - Thallium in Shallow Groundwater Figure 11-53 Isoconcentration Map - Thallium in Deep Groundwater Figure 11-54 Isoconcentration Map - Thallium in Bedrock Groundwater Figure 11-55 Isoconcentration Map - Vanadium in Shallow Groundwater Figure 11-56 Isoconcentration Map - Vanadium in Deep Groundwater Figure 11-57 Isoconcentration Map - Vanadium in Bedrock Groundwater Figure 11-58 Isoconcentration Map - pH in Shallow Groundwater Figure 11-59 Isoconcentration Map - pH in Deep Groundwater Figure 11-60 Isoconcentration Map - pH in Bedrock Groundwater Figure 11-61 Isoconcentration Map - Specific Conductance in Shallow Groundwater Figure 11-62 Isoconcentration Map - Specific Conductance in Deep Groundwater Figure 11-63 Isoconcentration Map - Specific Conductance in Bedrock Figure 11-64 Antimony Analytical Results Cross Section A -A' Figure 11-65 Antimony Analytical Results Cross Section B-B' Figure 11-66 Antimony Analytical Results Cross Section C-C' Figure 11-67 Arsenic Analytical Results Cross Section A -A' Figure 11-68 Arsenic Analytical Results Cross Section B-B' Figure 11-69 Arsenic Analytical Results Cross Section C-C' Figure 11-70 Barium Analytical Results Cross Section A -A' Figure 11-71 Barium Analytical Results Cross Section B-B' Figure 11-72 Barium Analytical Results Cross Section C-C' Figure 11-73 Beryllium Analytical Results Cross Section A -A' Figure 11-74 Beryllium Analytical Results Cross Section B-B' Figure 11-75 Beryllium Analytical Results Cross Section C-C' Figure 11-76 Boron Analytical Results Cross Section A -A' Figure 11-77 Boron Analytical Results Cross Section B-B' Figure 11-78 Boron Analytical Results Cross Section C-C' Figure 11-79 Cadmium Analytical Results Cross Section A -A' Figure 11-80 Cadmium Analytical Results Cross Section B-B' Page xi P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 11-81 Cadmium Analytical Results Cross Section C-C' Figure 11-82 Chloride Analytical Results Cross Section A -A' Figure 11-83 Chloride Analytical Results Cross Section B-B' Figure 11-84 Chloride Analytical Results Cross Section C-C' Figure 11-85 Chromium Analytical Results Cross Section A -A' Figure 11-86 Chromium Analytical Results Cross Section B-B' Figure 11-87 Chromium Analytical Results Cross Section C-C' Figure 11-88 Cobalt Analytical Results Cross Section A -A' Figure 11-89 Cobalt Analytical Results Cross Section B-B' Figure 11-90 Cobalt Analytical Results Cross Section C-C' Figure 11-91 Iron Analytical Results Cross Section A -A' Figure 11-92 Iron Analytical Results Cross Section B-B' Figure 11-93 Iron Analytical Results Cross Section C-C' Figure 11-94 Manganese Analytical Results Cross Section A -A' Figure 11-95 Manganese Analytical Results Cross Section B-B' Figure 11-96 Manganese Analytical Results Cross Section C-C' Figure 11-97 Molybdenum Analytical Results Cross Section A -A' Figure 11-98 Molybdenum Analytical Results Cross Section B-B' Figure 11-99 Molybdenum Analytical Results Cross Section C-C' Figure 11-100 Nickel Analytical Results Cross Section A -A' Figure 11-101 Nickel Analytical Results Cross Section B-B' Figure 11-102 Nickel Analytical Results Cross Section C-C' Figure 11-103 Selenium Analytical Results Cross Section A -A' Figure 11-104 Selenium Analytical Results Cross Section B-B' Figure 11-105 Selenium Analytical Results Cross Section C-C' Figure 11-106 Strontium Analytical Results Cross Section A -A' Figure 11-107 Strontium Analytical Results Cross Section B-B' Figure 11-108 Strontium Analytical Results Cross Section C-C' Figure 11-109 Sulfate Analytical Results Cross Section A -A' Figure 11-110 Sulfate Analytical Results Cross Section B-B' Figure 11-111 Sulfate Analytical Results Cross Section C-C' Figure 11-112 Thallium Analytical Results Cross Section A -A' Figure 11-113 Thallium Analytical Results Cross Section B-B' Figure 11-114 Thallium Analytical Results Cross Section C-C' Figure 11-115 Total Dissolved Solids Analytical Results Cross Section A -A' Figure 11-116 Total Dissolved Solids Analytical Results Cross Section B-B' Figure 11-117 Total Dissolved Solids Analytical Results Cross Section C-C' Page xii P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 11-118 Vanadium Analytical Results Cross Section A -A' Figure 11-119 Vanadium Analytical Results Cross Section B-B' Figure 11-120 Vanadium Analytical Results Cross Section C-C' Figure 11-121 Solid Phase - Groundwater Interaction Data Map 12.0 Screening -Level Risk Assessment Figure 12-1 Human Health and Ecological Exposure Areas 14.0 Discussion - Assessment Results Figure 14-1 Time versus Concentration - Central Basin Antimony Figure 14-2 Time versus Concentration - Downgradient Antimony Figure 14-3 Time versus Concentration - Central Basin Arsenic Figure 14-4 Time versus Concentration - Downgradient Arsenic Figure 14-5 Time versus Concentration - Central Basin Barium Figure 14-6 Time versus Concentration - Downgradient Barium Figure 14-7 Time versus Concentration - Central Basin Beryllium Figure 14-8 Time versus Concentration - Downgradient Beryllium Figure 14-9 Time versus Concentration - Central Basin Boron Figure 14-10 Time versus Concentration - Downgradient Boron Figure 14-11 Time versus Concentration - Central Basin Cadmium Figure 14-12 Time versus Concentration - Downgradient Cadmium Figure 14-13 Time versus Concentration - Central Basin Chloride Figure 14-14 Time versus Concentration - Downgradient Chloride Figure 14-15 Time versus Concentration - Central Basin Chromium Figure 14-16 Time versus Concentration - Downgradient Chromium Figure 14-17 Time versus Concentration - Central Basin Cobalt Figure 14-18 Time versus Concentration - Downgradient Cobalt Figure 14-19 Time versus Concentration - Central Basin Iron Figure 14-20 Time versus Concentration - Downgradient Iron Figure 14-21 Time versus Concentration - Central Basin Manganese Figure 14-22 Time versus Concentration - Downgradient Manganese Figure 14-23 Time versus Concentration - Central Basin Molybdenum Figure 14-24 Time versus Concentration - Downgradient Molybdenum Figure 14-25 Time versus Concentration - Central Basin Nickel Figure 14-26 Time versus Concentration - Downgradient Nickel Page xiii P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 14-27 Time versus Concentration - Central Basin Selenium Figure 14-28 Time versus Concentration - Downgradient Selenium Figure 14-29 Time versus Concentration - Central Basin Strontium Figure 14-30 Time versus Concentration - Downgradient Strontium Figure 14-31 Time versus Concentration - Central Basin Sulfate Figure 14-32 Time versus Concentration - Downgradient Sulfate Figure 14-33 Time versus Concentration - Central Basin Thallium Figure 14-34 Time versus Concentration - Downgradient Thallium Figure 14-35 Time versus Concentration - Central Basin Total Dissolved Solids Figure 14-36 Time versus Concentration - Downgradient Total Dissolved Solids Figure 14-37 Time versus Concentration - Central Basin Vanadium Figure 14-38 Time versus Concentration - Downgradient Vanadium Figure 14-39 Groundwater Concentration Trend Analysis Antimony In All Flow Layers Figure 14-40 Groundwater Concentration Trend Analysis Antimony In Surface Water Figure 14-41 Groundwater Concentration Trend Analysis Arsenic In All Flow Layers Figure 14-42 Groundwater Concentration Trend Analysis Arsenic In Surface Water Figure 14-43 Groundwater Concentration Trend Analysis Barium In All Flow Layers Figure 14-44 Groundwater Concentration Trend Analysis Barium In Surface Water Figure 14-45 Groundwater Concentration Trend Analysis Beryllium In All Flow Layers Figure 14-46 Groundwater Concentration Trend Analysis Beryllium In Surface Water Figure 14-47 Groundwater Concentration Trend Analysis Boron In All Flow Layers Figure 14-48 Groundwater Concentration Trend Analysis Boron In Surface Water Figure 14-49 Groundwater Concentration Trend Analysis Cadmium In All Flow Layers Page xiv P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 14-50 Groundwater Concentration Trend Analysis Cadmium In Surface Water Figure 14-51 Groundwater Concentration Trend Analysis Chloride In All Flow Layers Figure 14-52 Groundwater Concentration Trend Analysis Chloride In Surface Water Figure 14-53 Groundwater Concentration Trend Analysis Chromium (Total) In All Flow Layers Figure 14-54 Groundwater Concentration Trend Analysis Chromium (Total) In Surface Water Figure 14-55 Groundwater Concentration Trend Analysis Cobalt In All Flow Layers Figure 14-56 Groundwater Concentration Trend Analysis Cobalt In Surface Water Figure 14-57 Groundwater Concentration Trend Analysis Iron In All Flow Layers Figure 14-58 Groundwater Concentration Trend Analysis Iron In Surface Water Figure 14-59 Groundwater Concentration Trend Analysis Manganese In All Flow Layers Figure 14-60 Groundwater Concentration Trend Analysis Manganese In Surface Water Figure 14-61 Groundwater Concentration Trend Analysis Molybdenum In All Flow Layers Figure 14-62 Groundwater Concentration Trend Analysis Molybdenum In Surface Water Figure 14-63 Groundwater Concentration Trend Analysis Nickel In All Flow Layers Figure 14-64 Groundwater Concentration Trend Analysis Nickel In Surface Water Figure 14-65 Groundwater Concentration Trend Analysis Selenium In All Flow Layers Figure 14-66 Groundwater Concentration Trend Analysis Selenium In Surface Water Figure 14-67 Groundwater Concentration Trend Analysis Strontium In All Flow Layers Figure 14-68 Groundwater Concentration Trend Analysis Strontium In Surface Water Page xv P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF FIGURES (CONTINUED) Figure 14-69 Groundwater Concentration Trend Analysis Sulfate In All Flow Layers Figure 14-70 Groundwater Concentration Trend Analysis Sulfate In Surface Water Figure 14-71 Groundwater Concentration Trend Analysis Thallium In All Flow Layers Figure 14-72 Groundwater Concentration Trend Analysis Thallium In Surface Water Figure 14-73 Groundwater Concentration Trend Analysis Total Dissolved Solids In All Flow Layers Figure 14-74 Groundwater Concentration Trend Analysis Total Dissolved Solids In Surface Water Figure 14-75 Groundwater Concentration Trend Analysis Vanadium In All Flow Layers Figure 14-76 Groundwater Concentration Trend Analysis Vanadium In Surface Water Figure 14-77 Comprehensive Soil and Sediment Data Figure 14-78 Comprehensive Groundwater Data Figure 14-79 Comprehensive Surface Water and Area of Wetness Page xvi P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF APPENDICES Appendix A Regulatory Correspondence NCDEQ Expectations Document (July 18, 2017) Completed NCDEQ CSA Update Expectations Check List NCDENR NORR Letter (August 13, 2014) NCDEQ Background Location Approvals (October 11, 2017) NCDEQ Background Dataset Review (July 7, 2017) NCDEQ PBTV Approval Attachments (September 1, 2017) NCDEQ Correspondence — Revised Interim Monitoring Plan (December 20, 2017) Appendix B Comprehensive Data Table Comprehensive Data Table Notes Table 1- Groundwater Results Table 2 - Surface Water Results Table 3 - AOW Results Table 4 - Soil and Ash Results Table 5 - Sediment Results Table 6 - SPLP Results Table 7 - CCR Results Appendix C Site Assessment Data HDR CSA Appendix H — Hydrogeological Investigation Soil Physical Lab Reports Mineralogy Lab Reports Slug Test Procedure Slug Test Reports Historic Permeability Data Field Permeability Data Fetter -Bear Diagrams — Porosity Historic Porosity Data Estimated Seasonal High Groundwater Elevations Calculation HDR CSA Supplement 2 Slug Test Report / Slug Test Results UNCC Soil Sorption Evaluation Addendum to the UNCC Soil Sorption Evaluation Page i P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF APPENDICES (CONTINUED) Appendix D Receptor Surveys Drinking Water Well and Receptor Survey Report Supplement to Drinking Water Supply Well and Receptor Survey Dewberry Report - Permanent Water Supply Proposal to DEQ Appendix E Supporting Documents Stantec Report WSP Maps Appendix F Boring Logs, Construction Diagrams, and Abandonment Records Appendix G Sample Characterization Source Characterization - HDR CSA Appendix C Soil and Rock Characterization - HDR CSA Appendix D Surface Water and Sediment Characterization - HDR CSA Appendix F Groundwater Characterization - HDR CSA Appendix G Includes Duke Low Flow Sampling Plan Field, Sampling, and Data Analysis Quality Assurance / Quality Control - HDR CSA Appendix E Appendix H Background Determination Site specific 2017 CSA PBTV Report Appendix I Risk Assessment Appendix J Lab Reports Page ii P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra LIST OF ACRONYMS NCDENR Title 15A, Subchapter 2B. Surface Water and Wetland 2B Standards NCDENR Title 15A, Subchapter 2L. Groundwater Classification 2L and Standards ADD Average Daily Dose AOW Areas of Wetness APS Aquifer Protection Section ASTM American Society for Testing and Materials BR Bedrock Flow Layer BGS Below Ground Surface CA Critical Area CAMA Coal Ash Management Act CAP Corrective Action Plan CCR Coal Combustion Residuals CFR Code of Federal Regulations COI Constituent of Interest CSA Comprehensive Site Assessment CSM Conceptual Site Model D Deep Flow Layer DWM Divison of Waste Management DWQ Divison of Water Quality DWR Division of Water Resources EMP Effectiveness Monitoring Plan ESV Ecological Screening Value FGD Flue Gas Desulfurization GAP Groundwater Assessment Work Plan GCL Geosynthetic Clay Liner GIS Geographic Information Systems HAO Hydroxide Phases of Aluminum HFO Hydroxide Phases of Iron HQ Hazard Quotient IHSB Inactive Hazardous Sites Branch IMAC Interim Maximum Allowable Concentrations Page iii P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra IMP Interim Monitoring Plan MCL Maximum Contaminant Level MNA Monitored Natural Attenuation MRL Method Reporting Limit MSS, Plant, or Site Marshall Steam Station NCDENR North Carolina Department of Environment and Natural Resources NCDEQ North Carolina Department of Environmental Quality NCDHHS North Carolina Department of Health and Human Services NORR Notice of Regulatory Requirements NPDES National Pollution Discharge Elimination System NTUs Nephelometric Turbidity Units NURE National Uranium Resource Evaluation PBTV Provisional Background Threshold Value POG Protection of Groundwater PSRG Preliminary Soil Remediation Goal PV Photovoltaic PWR Partially Weathered Rock PWS Public Water Supply RBC Risk -Based Concentration REC Recovery RQD Rock Quality Designation RSL Regional Screening Level S Shallow Flow Layer SCM Site Conceptual Model SMCL Secondary Maximum Contaminant Level SPLP Synthetic Precipitation Leaching Procedure TCLP Toxicity Characteristic Leaching Procedure TDC Total Dissolved Solids TOC Total Organic Carbon TRV Toxicity Reference Values TZ Transition Zone UNC University of North Carolina UNCC University of North Carolina - Charlotte USDA-SCS U.S. Department of Agriculture Soil Conservation Service Page iv P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra USEPA United States Environmental Protection Agency USGS United States Geological Survey UTL Upper Tolerance Limit Page v P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 1.0 INTRODUCTION Duke Energy Carolinas, LLC (Duke Energy) owns and operates the Marshall Steam Station (MSS, the Plant, or the Site), located at 8320 NC Highway 150 East in Terrell, Catawba County, North Carolina (Figure 1-1). Operations, as a coal-fired generating station, began at MSS in 1965. Four coal-fired units presently are in operation. Coal combustion residuals (CCR) consisting of bottom and fly ash material from MSS have historically been managed in the Site ash basin, located north of the station adjacent to Lake Norman. Dry ash has been disposed of in other areas at the Site, including the dry ash landfill units (Phases I and II) and Industrial Landfill No. 1. Flue gas desulfurization (FGD) residue (i.e., gypsum) and fly ash were disposed of in the FGD Residue Landfill, which was placed in intermediate closure in October 2015. Fly ash was also used as structural fill in the photovoltaic (PV) structural fill and beneath portions of the Industrial Landfill No. 1. Discharge from the ash basin to Lake Norman is permitted by the North Carolina Department of Environmental Quality (NCDEQ)1 Division of Water Resources (DWR) under the National Pollutant Discharge Elimination System (NPDES) Permit NC0004987. 1.1 Purpose of Comprehensive Site Assessment This Comprehensive Site Assessment (CSA) update was conducted to refine and expand the understanding of subsurface geologic/hydrogeologic conditions and evaluate the extent of impacts from historical management of coal ash in the ash basin in accordance with Coal Ash Management Act (CAMA). This CAMA CSA update contains an assessment of Site conditions based on a comprehensive interpretation of geologic and sampling results from the initial Site assessment and geologic and sampling results obtained after the initial assessment. This CAMA CSA update has been prepared in coordination with Duke Energy and NCDEQ in response to requests for additional information, including additional sampling and assessment of specified areas. Data associated with monitoring wells installed for the purpose of compliance with 40 C.F.R. § 257.90 (the CCR Rule) is provided as information. Groundwater data collected for the purpose of compliance with the CCR Rule will be evaluated in accordance with the requirements and schedule provided therein. This CAMA CSA update was prepared in conformance with the most recently updated CSA table of contents provided by NCDEQ to Duke Energy on September 29, 2017. In response to the NCDEQ request for additional information, this submittal includes the following: I Prior to September 18, 2015, the NCDEQ was referred to as the North Carolina Department of Environment and Natural Resources (NCDENR). Both naming conventions are used in this report, as appropriate. Page 1-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra h Review of baseline assessment data collected and reported as part of CSA activities 167 A summary of NPDES and Coal Ash Management Act (CAMA) groundwater monitoring information h A summary of potential receptors, including results from water supply wells h A description and findings of additional assessment activities conducted since submittal of the CSA Supplement reports h An update on background concentrations for groundwater and soil h Definition of horizontal and vertical extent of CCR constituents in soil and groundwater based on NCDEQ approved background concentrations �7 An update of human health and ecological risk assessment to evaluate the existence of imminent hazards to public health, safety, and the environment 1.2 Regulatory Background The CAMA of 2014 directs owners of CCR surface impoundments in North Carolina to conduct groundwater monitoring, assessment, and remedial activities, if necessary. The CSA was performed to collect information necessary to evaluate the horizontal and vertical extent of impacts to soil and groundwater attributable to CCR source area(s), identify potential receptors, and screen for potential risks to those receptors. 1.2.1 Notice of Regulatory Requirements (NORR) On August 13, 2014, North Carolina Department of Environment and Natural Resources (NCDENR) issued a Notice of Regulatory Requirements (NORR) letter notifying Duke Energy that exceedances of groundwater quality standards at 14 coal ash facilities owned and operated by Duke Energy had been reported. Those groundwater quality standards are part of 15A NCAC 02L (2L) .0200 Classifications and Water Quality Standards Applicable to the Groundwaters of North Carolina. The NORR stipulated that for each coal ash facility, Duke Energy was to conduct a CSA. The NORR also stipulated that before conducting each CSA, Duke was to submit a Groundwater Assessment Work Plan (GAP or Work Plan) and a receptor survey. In accordance with the NORR requirements, a receptor survey was performed to identify all receptors within a 0.5-mile radius (2,640 feet) of the ash basin compliance boundary, and a CSA was conducted for each facility. The NORR letter is included in Appendix A. Page 1-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 1.2.2 Coal Ash Management Act Requirements The CAMA of 2014 — General Assembly of North Carolina Senate Bill 729 Ratified Bill (Session 2013) (SB 729) requires that ash from Duke Energy coal plant sites located in North Carolina either (1) be excavated and relocated to fully lined storage facilities, or (2) go through a classification process to determine closure options and schedule. Closure options can include a combination of excavating and relocating ash to a fully lined structural fill, excavating and relocating the ash to a lined landfill (on -site or off -site), and/or capping the ash with an engineered synthetic barrier system, either in place or after being consolidated to a smaller area on -site. As a component of implementing this objective, CAMA provides instructions for owners of coal combustion residuals surface impoundments to perform various groundwater monitoring and assessment activities. Section §130A-309.209 of the CAMA ruling specifies groundwater assessment and corrective actions, drinking water supply well surveys and provisions of alternate water supply, and reporting requirements as follows: (a) Groundwater Assessment of Coal Combustion Residuals Surface Impoundments. — The owner of a coal combustion residuals surface impoundment shall conduct groundwater monitoring and assessment as provided in this subsection. The requirements for groundwater monitoring and assessment set out in this subsection are in addition to any other groundwater monitoring and assessment requirements applicable to the owners of coal combustion residuals surface impoundments. (1) No later than December 31, 2014, the owner of a coal combustion residuals surface impoundment shall submit a proposed Groundwater Assessment Plan for the impoundment to the Department for its review and approval. The Groundwater Assessment Plan shall, at a minimum, provide for all of the following: • A description of all receptors and significant exposure pathways. • An assessment of the horizontal and vertical extent of soil and groundwater contamination for all contaminants confirmed to be present in groundwater in exceedance of groundwater quality standards. • A description of all significant factors affecting movement and transport of contaminants. Page 1-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra • A description of the geological and hydrogeological features influencing the chemical and physical character of the contaminants. (2) The Department shall approve the Groundwater Assessment Plan if it determines that the Plan complies with the requirements of this subsection and will be sufficient to protect public health, safety, and welfare; the environment, and natural resources. (3) No later than 10 days from approval of the Groundwater Assessment Plan, the owner shall begin implementation of the Plan. (4) No later than 180 days from approval of the Groundwater Assessment Plan, the owner shall submit a Groundwater Assessment Report to the Department. The Report shall describe all exceedances of groundwater quality standards associated with the impoundment. 1.2.3 Coal Combustion Residuals Rule On April 17, 2015, the United States Environmental Protection Agency (USEPA) published 40 Code of Federal Regulations (CFR) Parts 257 and 261: Hazardous and Solid Waste Management System; Disposal of Coal Combustion Residuals from Electric Utilities; Final Rule (USEPA, April 2015). This regulation addresses the safe disposal of CCR as solid waste under Subtitle D of the Resource Conservation and Recovery Act and is referred to herein as the CCR Rule. The CCR Rule became effective on October 19, 2015. The rule provides national minimum criteria for "the safe disposal of CCR in new and existing CCR landfills, surface impoundments, and lateral expansions, design and operating criteria, groundwater monitoring and corrective action, closure requirements and post closure care, and recordkeeping, notification, and internet posting requirements. Monitoring wells have been installed and monitored in accordance with the CCR Rule. The most recent data available from the CCR groundwater monitoring well network is provided on isoconcentration maps and cross -sections herein. The CCR Rule data will otherwise be evaluated in accordance with the requirements and schedule outlined therein. The CCR Rule establishes requirements for a phased groundwater monitoring program consisting of detection monitoring and, if necessary, assessment monitoring and corrective action. The first phase of monitoring to comply with the CCR Rule is to include at least eight rounds of Detection Monitoring for Appendix III and Appendix IV constituents (described below) prior to October 2017. This Detection Monitoring may be followed by Assessment Monitoring. The USEPA considers several parameters to be leading indicators that provide an Page 1-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra early detection of whether constituents are migrating from a CCR unit (e.g. Appendix III constituents). Appendix III constituents include: boron, calcium, chloride, fluoride, pH, sulfate, and total dissolved solids (TDS). If sampling results from the Detection Monitoring phase indicate that there are statistically significant increases over background concentrations for Appendix III constituents, the Final Rule requires the facility (or Site) to begin Assessment Monitoring. The parameters required for Assessment Monitoring are antimony, arsenic, barium, beryllium, cadmium, chromium, cobalt, fluoride, lead, lithium, mercury, molybdenum, radium 226 and 228 (combined), selenium, and thallium (USEPA 2015; Appendix IV constituents). If the results of assessment monitoring indicate that corrective action may be warranted, the CCR Rule requires additional assessment of corrective measures and potentially corrective action monitoring. 1.3 Approach to Comprehensive Site Assessment The CSA has been performed to meet NCDEQ requirements associated with potential site remedy selection. The following components were used to develop the assessment. 1.3.1 NORR Guidance The NORR requires that the site assessment provide information to meet the requirements of 2L .0106 (g). This regulation lists the items to be included in site assessments conducted pursuant to Paragraph (c) of the rule. These requirements are listed below and referenced to the applicable sections of this CSA. 15A NCAC 02L .0106(g) Requirement CSA Sections) (1) The source and cause of contamination Section 3 (2) Any imminent hazards to public health and safety, as defined in Sections ES.2 G.S. 130A-2, and any actions taken to mitigate them in and 2.8 accordance with Paragraph (f) of this Rule (3) All receptors and significant exposure pathways Sections 4 and 12 (4) The horizontal and vertical extent of soil and groundwater Sections 7, 10, contamination and all significant factors affecting contaminant 11, 13, 14 transport (5) Geological and hydrogeological features influencing the Sections 6, 11, movement, chemical, and physical character of the contaminants and 15 1.3.2 USEPA Monitored Natural Attenuation Tiered Approach The assessment data is compiled in a manner to be consistent with "Monitored Natural Attenuation (MNA) of Inorganic Contaminants in Groundwater" Page 1-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra (EPA/600/R-07/139) (USEPA, October, 2007). The tiered analysis approach discussed in this guidance document is designed to align site characterization tasks to reduce uncertainty in remedy selection. Evaluation of MNA as a viable remedy will be included in the corrective action planning process. The tiered assessment data collection includes information to evaluate: ,67 Active contaminant removal from groundwater and dissolved plume stability ,67 The mechanisms and rates of attenuation 07 The long-term capacity for attenuation and stability of immobilized contaminants 101 Anticipated performance monitoring needs to support the selected remedy 1.3.3 ASTM Conceptual Site Model Guidance The American Society for Testing and Materials (ASTM) E1689-95 generally describes the major components of conceptual site models, including an outline for developing models. To the extent possible, this guidance was incorporated into preparation of the Site Conceptual Model (SCM). The SCM is used to integrate site information, identify data gaps, and determine whether additional information is needed at the site. The model is also used to facilitate selection of remedial alternatives and effectiveness of remedial actions in reducing the exposure of environmental receptors to contaminants (ASTM, 2014). 1.4 Technical Objectives The objectives of CSA activities fall into one of the following categories: y Determine the range of background groundwater quality from pertinent geologic settings (horizontal and vertical) across a broad area of the Site. 47 Evaluate groundwater quality from pertinent geologic settings (horizontal and vertical extent of CCR leachate constituents). t7 Evaluate groundwater quality from pertinent geologic settings (horizontal and vertical extent of CCR leachate constituents). h Establish perimeter (horizontal and vertical) boundary conditions for groundwater modeling. Page 1-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra �7 Provide source area information, including ash pore water chemistry, physical and hydraulic properties, CCR thickness, and residual saturation within the ash basins. '67 Address soil chemistry in the vicinity of the ash basins (horizontal and vertical extent of CCR leachate constituents in soil) compared with background concentrations. h Determine potential routes of exposure and receptors. h Compile information necessary to develop a groundwater Corrective Action Plan (CAP) protective of human health and the environment in accordance with 2L. 1.5 Previous Submittals Detailed descriptions of the Site operational history, the Site conceptual model, physical setting and features, geology/hydrogeology, and results of the findings of the CSA and other CAMA-related work are documented in full in the following: 17 Comprehensive Site Assessment Report — Marshall Steam Station Ash Basin (HDR, September 8, 2015a) 0 Corrective Action Plan Part 1 — Marshall Steam Station Ash Basin (HDR, December 7, 2015b) 0 Corrective Action Plan Part 2 (included CSA Supplement 1 as Appendix A) — Marshall Steam Station Ash Basin (HDR, March 3, 2016a) 47 Comprehensive Site Assessment Supplement 2 — Marshall Steam Station Ash Basin (HDR, August 4, 2016b) (U.S. Department of Agriculture, 1975) Page 1-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 2.0 SITE HISTORY AND DESCRIPTION An overview of the Marshall Steam Station setting and operations is presented in the following sections. 2.1 Site Description, Ownership, and Use History MSS is a coal-fired electricity -generating facility located on the west bank of Lake Norman near the town of Terrell, Catawba County, North Carolina. The address of MSS is 8320 NC Highway 150 E, Terrell, NC. The entire Site, approximately 1,446 acres in area, is owned by Duke Energy. The Site is thought to have been largely undeveloped prior to Duke Energy ownership. MSS is a four -unit, coal-fired plant that generates electricity. Coal is delivered to the station by a railroad line. Operation of Unit 1 began in 1965, and operation of Unit 2 began in 1966, with each generating 350 megawatts. Operation of Unit 3 began in 1969, and operation of Unit 4 began in 1970, with each generating 648 megawatts. Improvements to the Plant since 1970 have increased the electric generating capacity to 2,090 megawatts. The MSS ash basin, which contains ash generated from the historic and active coal combustion at the Plant, is situated with MSS to the south, topographic divides located along Sherrills Ford Road to the west, Island Point Road to the north, and Duke Energy property to the east (Figure 2-1). The ash basin, approximately 394 acres in size, was constructed with an earthen dike. A 500-foot compliance boundary for the routine groundwater monitoring well network encircles the ash basin, co -located with the property boundary on the western edge of the Site and extending to Lake Norman on the eastern edge of the Site. Fly ash and bottom ash from MSS was managed in the ash basin from approximately 1965 until 1984. Fly ash precipitated from flue gas and bottom ash collected in the bottom of the boilers was sluiced to the ash basin using conveyance water withdrawn from Lake Norman. Since 1984, fly ash has been disposed of in the on -site landfills and bottom ash has continued to be sluiced to the ash basin. The ash handling system is described further in Section 3.1. The air pollution control ("scrubber") system for the coal-fired units at MSS includes a FGD system that was placed into operation in 2007.Other areas of the Site are occupied by facilities supporting the production or transmission of power (one switchyard and associated transmission lines), the FGD wastewater treatment system, and the gypsum handling station (associated with the FGD system). Page 2-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra In addition to the power plant property, Duke Energy owns and operates the Catawba- Wateree Hydroelectric Project (Federal Energy Regulatory Commission Project No. 2232). Lake Norman reservoir, part of the Catawba-Wateree project, is used for hydroelectric generation, a source of cooling water for MSS and Duke Energy's McGuire Nuclear Station, municipal water supply, and recreation. 2.2 Geographic Setting, Surrounding Land Use, and Surface Water Classification The MSS is situated in a semi -developed area on the west bank of Lake Norman in the southeastern corner of Catawba County, NC. A description of the physical setting of the MSS is provided in the following sections. Geographic Setting The MSS is located on the west bank of Lake Norman near the town of Terrell, Catawba County, North Carolina. Natural topography at the Site generally slopes downward from historic topographic divides to the ash basin and toward Lake Norman. Lake Norman encompasses approximately 32,000 acres at a full pond elevation of 760 feet2. The entire MSS is approximately 1,446 acres in area. An 1893 United States Geological Survey (USGS) topographic map depicting the site prior to construction of the ash basin is shown on Figure 2-2. Surrounding Land Use The area surrounding MSS generally consists of residential properties, undeveloped land, and Lake Norman. Properties located within a 0.5-mile radius of the MSS ash basin compliance boundary generally consist of undeveloped land and Lake Norman to the east; undeveloped land and residential properties located to the north and west; portions of the MSS site (outside the compliance boundary), undeveloped land, and residences to the south; and commercial properties to the southeast along North Carolina Highway 150. Meteorological Data According to the U.S. Department of Agriculture Soil Conservation Service (USDA-SCS) soil survey (1975), the average summer temperature in Catawba County is 78°F and the average daily maximum temperature is 88°F. During winter, the average temperature is 42°F and the average daily minimum temperature is 31°F. Precipitation is well distributed throughout the county and throughout the year, and averages approximately 49.2 inches per year. Much of the rainfall during the growing season (April to November) comes from 2 The datum for all elevation information presented in this report is NAVD88. Page 2-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra thunderstorms. The average relative humidity in midafternoon is approximately 67 percent, with humidity reaching higher levels at night. The prevailing wind is from the northwest, and average wind velocity is approximately 8 miles per hour (USDA-SCS 1975). Surface Water Classification Surface water features located on the Site are shown on Figure 2-1. The Site is located along the shores of Lake Norman, which is part of the Catawba River watershed. The ash basin is adjacent to Lake Norman. Surface water classifications in North Carolina are promulgated in Title15A NCAC Subchapter 2B. The surface water classification for the Catawba River is Class WS-IV; B and CA waters. Class WS-IV waters are protected as water supplies that are generally in moderately to highly developed watersheds. Class B waters are protected for all Class C uses in addition to primary recreation. Primary recreational activities include swimming, skin diving, water skiing, and similar uses involving human body contact with water where such activities take place in an organized manner or on a frequent basis. A Critical Area (CA) is described as being within 0.5 miles of Class B waters that drain to water supplies as measured from the normal pool elevation of the reservoir. Point source discharges of treated wastewater are permitted pursuant to Rules .0104 and .0211 of this Subchapter (2B). Local programs to control nonpoint sources and storm water discharges of pollution are required for all Class C uses (i.e., freshwaters protected for secondary recreation, fishing, aquatic life including propagation and survival, and wildlife). Two local municipal water supply intakes are located on Lake Norman: Mooresville (approximately 3.8 miles upstream from the Site) and Lincoln (approximately 5.6 miles downstream from the Site). 2.3 CAMA-related Source Areas CAMA provides for groundwater assessment of CCR surface impoundments defined as topographic depressions, excavations, or diked areas formed primarily of earthen materials, without a base liner, and that meet other criteria related to design, usage, and ownership (Section §130A-309.201). At MSS, the groundwater assessment incorporated the ash basin, dry ash landfills (Phases I and II), and PV structural fill (Figure 2-1). Collectively, the ash basin and various landfills are referred to herein as ash management areas. Page 2-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 2.3.1 Ash Basin The initial MSS ash basin was constructed in 1965 by building an earthen dike at the confluence where Holdsclaw Creek historically entered the Catawba River. The earthen dike was constructed to impound water, and ash was sluiced to the unlined basin. In general, the ash basin is located in a historical depression formed from Holdsclaw Creek and small tributaries that fed the creek. The basin has a dendritic shape consisting of coves of deposited ash, dikes that impound ash in portions of the basin, and four main areas of ponded water. The area contained within the ash basin waste boundary is approximately 394 acres. Coal ash from MSS was managed in the ash basin from approximately 1965 until 1984. Fly ash precipitated from flue gas and bottom ash collected in the bottom of the boilers were sluiced to the ash basin using conveyance water withdrawn from Lake Norman. Since 1984, fly ash has mainly been disposed of in the on -site dry ash landfills (described below) and the sluicing of bottom ash to the ash basin has continued. While FGD residue is not placed in the ash basin, contact storm water and leachate from the FGD landfill, along with FGD wastewater treatment system effluent, are routed to the ash basin. The FGD residue produced by the air treatment system at MSS is primarily gypsum (CaSO4•H2O) and is sold for re- use or disposed of in one of the on -site landfills. Bottom ash is sluiced to concrete pits, where the water is allowed to decant and then flow to the ash basin via a discharge canal. Bottom ash is then excavated from the pit and discharge canal that flows to the ash basin, and is sold for off -site beneficial reuse or used for roads at the ash basin facility. During operations, the sluice water/ash slurry (and other flows) is discharged into the southwest portion of the ash basin. As required by CAMA, Duke Energy plans to implement closure and remediation of the MSS ash basin. As activities are completed to meet the requirements of G.S. 130A-309.213.(d)(1) of House Bill 630, a pending "Low" risk ranking will allow the primary CAMA-related source, or coal ash in the ash basin, to be capped in place. An engineered cap would reduce infiltration through the covered area, thereby reducing the potential of leaching of Constituents of Interest (COIs) into the groundwater underlying the closed basin. 2.3.2 Dry Ash Landfill Two unlined ash landfill units, referred to as the Marshall dry ash landfill (NCDEQ Division of Solid Waste Permit No. 1804-INDUS), are located adjacent to the east (Phase I) and northeast (Phase II) portions of the ash basin. Phase I Page 2-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra contains approximately 280,000 tons of fly ash, which was placed from September 1984 through March 1986. Placement of ash in the Phase II areas began around March 1986 and was completed in 1999. Phase II contains approximately 4,900,000 tons of fly ash. The approximate boundaries of Phase I and II units are shown on Figure 2-1. The landfill units were constructed prior to the requirement for lining industrial landfills and were closed with a soil cover system. 2.3.3 FGD Landfill The FGD landfill (NCDEQ Division of Solid Waste Permit No. 1809-INDUS-) is located to the west of the ash basin. The landfill was constructed with an engineered, single -liner system. In general, the topography of this landfill site slopes from the west-northwest to the east-southeast toward the MSS ash basin. The landfill is permitted to receive the following types of waste generated at Duke Energy Corporation facilities: FGD residue (gypsum), clarifier sludge, fly ash, bottom ash, construction and demolition waste, asbestos waste, mill rejects (pyrites), waste limestone material, land clearing and inert debris, boiler slag, ball mill rejects, sand blast material, and coal waste. Contact storm water and leachate have been collected and piped to the ash basin; this water will be re- routed to a lined retention basin in 2018. The landfill is currently in interim closure with a 12-inch soil cover system in place. A traditional capping system with a geosynthetic clay liner (GCL) is planned to be installed for closure by December 31, 2018. 2.3.4 Industrial Landfill No. 1 Industrial Landfill No. 1 (NCDEQ Permit No. 1812-INDUS-2008) is located adjacent to the north portion of the ash basin. The landfill was constructed with a leachate collection and removal system and a three -component liner system in which the components consist of a primary geomembrane, secondary geomembrane (with a leak detection system between them), and GCL and soil liners. Historically, the landfill has been permitted to receive the following types of waste generated at Duke Energy Corporation facilities: fly ash, bottom ash, FGD residue, FGD clarifier sludge, asbestos material, land clearing and inert debris, coal mill rejects, waste limestone material, boiler slag, construction and demolition waste, sand blast material, ball mill rejects, coal waste, and pyrites. The landfill was constructed over portions of residual material and over portions of the ash basin. The subgrade for portions of this landfill were constructed of fly ash under the structural fill rules found in 15A NCAC 13B .1700 et seq. Contact Page 2-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra storm water and leachate have historically been collected and piped to the ash basin; this water will be re-routed to a lined retention basin in 2018. 2.3.5 Photovoltaic Farm Structural Fill The photovoltaic farm structural fill (PV structural fill) was constructed of fly ash, under the structural fill rules found in 15A NCAC 13B .1700 et seq., and bottom ash, under Duke Energy's Distribution of Residuals Solids Permit issued by NCDENR Division of Water Quality (DWQ), and is located adjacent to and partially on top of the northwest portion of the ash basin. The PV structural fill, used for renewable energy production, contains a solar panel field on the south portion of the structural fill unit. Placement of dry ash in the structural fill began in October 2000. The structural fill is unlined and was completed and closed with a soil cover system in February 2013. 2.4 Other Primary and Secondary Sources 2.4.1 Additional Landfills The demolition landfill (NCDEQ Permit No. 1804-INDUS-1983) is located adjacent to the north portion of the ash basin directly north of the dry ash landfill (Phase II). The landfill received construction and demolition waste from MSS starting in September 1984. The landfill is unlined and was closed with a soil cap in 2008. The asbestos landfill (NCDEQ Permit No. 1804-INDUS-1983) is located adjacent to the north portion of the ash basin and adjacent to the demolition landfill. The landfill received asbestos waste from MSS and other Duke Energy facilities starting in December 1987. The landfill is unlined and was closed with a soil cap in 2008. 2.4.2 Secondary Sources Regulation 15A NCAC 02L .0106 (f)(4) requires that the secondary sources that could be potential continuing sources of possible pollutants to groundwater be addressed in the CAP. At MSS, the soil located below the ash basin, dry ash landfills, and PV structural fill could be considered a potential CAMA-related secondary source. Further evaluation of soil beneath the ash management areas as a potential secondary source is presented in Section 7.3. CSA activities include an assessment of the horizontal and vertical extent of constituents related to the CCR impoundments and observed at concentrations greater than 2L or background concentrations. If groundwater assessment Page 2-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra indicates constituent exceedances related to sources other than the CCR impoundments, those sources will be addressed as part of a separate process in compliance with the requirements of 2L. MSS continues to maintain a coal pile in the area located between the coal-fired units and the active ash basin. 2.5 Summary of Permitted Activities Duke Energy is authorized to discharge wastewater from MSS to receiving waters designated as the Catawba River in accordance with NPDES Permit NC0004987. The NPDES program regulates wastewater discharges to surface waters to ensure that surface water quality standards are maintained. The NPDES permitting program requires that permits be renewed every five years. The most recent NPDES permit became effective October 1, 2016, and expires on September 30, 2021. As part of the permit renewal, the facility identified seeps and collected seep samples. The seeps were incorporated into the permit as outfalls. The permit requires surface water monitoring, including continued sampling of seep outfalls, as part of the permit conditions. In compliance with CAMA, MSS is required to decommission the ash basin. It is anticipated that through this process existing seeps will potentially dry out or significantly reduce in flow volume and/or concentrations of CCR constituents. Duke Energy is also permitted to discharge storm water to the Catawba River in accordance with NPDES Draft Permit NCS000548 dated May 15, 2015. Any other point source discharge to surface waters of the state is prohibited unless it is an allowable non -storm water discharge or is covered by another permit, authorization, or approval. An NPDES flow diagram for MSS is included as Figure 2-3. Two permitted landfills (the FGD Residue Landfill and Industrial Landfill No. 1), one closed dry ash landfill consisting of two units (dry ash landfill Phases I and II), one closed demolition and construction debris landfill, one closed asbestos landfill, and one fly ash structural fill unit (PV structural fill) are located partially or wholly outside the ash basin footprint. 2.6 History of Site Groundwater Monitoring The location of the ash basin voluntary and compliance monitoring wells, the ash basin waste boundary, and the compliance boundary are shown on Figure 2-4. The compliance boundary for groundwater quality at the MSS ash basin is defined in accordance with Title 15A NCAC 02L .0107(a) as being established at either 500 feet Page 2-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra from the waste boundary or at the property boundary, whichever is closer to the waste. Construction details for site monitoring wells are provided in Table 2-1. At MSS, monitoring wells are designated with either an S, D, or BR identifier. These correspond to the flow layer the well is screened in. S refers to the shallow flow layer (alluvium and saprolite), D refers to the deep flow layer (saprolite and weathered rock), and BR refer to the bedrock flow layer (sound, relatively unfractured rock). The following sections discuss groundwater monitoring activities prior to CSA activities through current CAMA-related monitoring activities. Groundwater monitoring results are presented in Section 10.0. 2.6.1 Ash Basin Voluntary Groundwater Monitoring Monitoring wells MW-6S, MW-6D, MW-7S, MW-7D, MW-8S, MW-8D, MW-9S, and MW-9D were installed by Duke Energy in 2006 as part of a voluntary monitoring system. Duke Energy implemented voluntary groundwater monitoring around the MSS ash basin from November 2007 until February 2010. During that period, the voluntary groundwater monitoring wells were sampled nine times and the analytical results were submitted to NCDENR DWR. 2.6.2 Ash Basin NPDES Groundwater Monitoring Groundwater monitoring as required by the MSS NPDES Permit NC0004987 began in February 2011. The NPDES Permit NC0004987 (effective October 01, 2016), lists the groundwater monitoring wells to be sampled, the parameters and constituents to be measured and analyzed, and the requirements for sampling frequency and reporting results (Table 2-2). Locations for the compliance groundwater monitoring wells were approved by the former NCDENR DWR Aquifer Protection Section (APS). The compliance groundwater monitoring system for the MSS ash basin consists of the following wells: MW-4, MW-4D, MW-10S, MW-10D, MW-11S, MW-11D, MW-12S, MW-12D, MW-13S, MW-13D, MW-14S, and MW-14D. All compliance monitoring wells are sampled three times per year (in February, June, and October). Analytical results are submitted to the NCDEQ DWR before the last day of the month that follows the month in which sampling of compliance monitoring wells was conducted. The compliance groundwater monitoring is performed in addition to the normal NPDES monitoring of the discharge flows from the ash basin. Compliance groundwater monitoring continued in 2017 in accordance with the NPDES Permit NC0004987, as the requirements for groundwater monitoring in Page 2-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra the renewed NPDES permit dated October 1, 2016, remained consistent with previous permits. NPDES groundwater monitoring in and around the ash basin will continue in accordance with the NPDES permit until the permit conditions provide otherwise. 2.6.3 Ash Basin CAMA Groundwater Monitoring A total of 83 groundwater monitoring wells were installed at MSS from March to July 2015 as part of the initial CSA groundwater assessment program. One comprehensive round of sampling and analysis was conducted prior to, and reported in, the September 2015 CSA report. A detailed discussion of well installation and sampling activities is provided in Section 2.7. Nine additional rounds of groundwater sampling of the CSA wells have occurred since submittal of the 2015 CSA report. Twenty (20) of the wells installed as part of the CSA in 2015 are screened within ash basin pore water, one well is screened within PV structural fill pore water, and one is screened within ash in Dry Ash Landfill (Phase II). The Dry Ash Landfill (Phase II) well (AL-3S) is screened across the ash/soil contact, and is currently dry. The remaining 61 wells installed in 2015 — located outside or beneath the ash basin boundary, dry ash landfill (Phases I and II), and PV structural fill — represent groundwater. 2.6.4 Landfill Groundwater Monitoring Duke Energy conducts routine solid waste landfill compliance groundwater monitoring during February and August each year for the dry ash landfills (Phase I and II). Leachate monitoring is conducted for the Industrial Landfill #1 (Permit No. 1812) during February and August each year. Additionally, groundwater monitoring is performed in March and September each year for the FGD landfill. These monitoring programs are conducted in accordance with NCDEQ (Divison of Waste Management-DWM) Solid Waste Section, and are not detailed herein. Various wells associated with these landfills have also been sampled for CAMA purposes since 2015. 2.7 Summary of Assessment Activities From 1988 to 2015, several environmental incidents (i.e., releases) occurred at the site that have initiated notifications to NCDEQ or required a subsurface investigation. The historical incidents have generally consisted of releases that had potential to impact soil Page 2-9 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra and groundwater at the Site, in waters of the U.S., or within a containment structure. A summary of the historical on -Site environmental incidents is provided in Table 2-3. Duke Energy was notified in a letter dated November 9, 2011, from NCDENR DWM that exceedances of 2L Standards were reported in samples collected from compliance groundwater monitoring wells at the dry ash landfill (Phases I and II) and the FGD landfill. A description of the dry ash landfill is summarized in Section 3.2.2, and a description of the FGD landfill is summarized in in Section 3.2.3. NCDENR requested that Duke Energy submit groundwater assessment work plans for each of these landfills to the DWM Solid Waste Section. The requested work plans were submitted by Duke Energy on February 9, 2012, and approved by DWM on February 20, 2012 (FGD landfill), and March 23, 2012 (dry ash landfill). The assessment of groundwater exceedances associated with the FGD landfill was submitted to NCDENR on July 12, 2012. The assessment associated with the dry ash landfill was submitted to NCDENR on December 21, 2012. Approval letters for the groundwater assessment reports for the FGD landfill and dry ash landfill were received from the DWM Solid Waste Section on May 2, 2016. In a letter dated March 16, 2012, the former NCDENR DWR APS requested that Duke Energy begin additional assessment activities at stations where measured and modeled concentrations of groundwater constituents exceeded the 2L Standards at the compliance boundary. Duke Energy submitted Proposed Groundwater Assessment Work Plan, Duke Energy Carolinas, LLC, Marshall Steam Station Ash Basin, NPDES Permit NC 0004987 (dated March 15, 2013) to address that request by NCDENR as it pertained to MSS. Efforts of this assessment were resolved prior to implementation of CAMA. CSA Activities Groundwater monitoring wells were installed at multiple locations and depths across the Site to monitor the horizontal and vertical constituent distribution. Monitoring wells were installed at the ash basin, dry ash landfills (Phase I and II), PV structural fill, locations inside and beyond the compliance area, and background areas. Groundwater monitoring well locations are shown on Figure 2-4. Background monitoring wells included two pre-existing compliance groundwater monitoring wells (MW-4 and MW-4D) and six wells (BG-1S/D, BG-2S/BR, and BG-3S/D) installed as part of the CSA effort. The background monitoring wells were installed northeast of the MSS ash basin and are separated from the source area by a series of topographic divides. Page 2-10 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Additional Assessment 2016 Additional monitoring wells were installed in 2016 to refine and expand understanding of groundwater flow direction and extent of exceedances at MSS. Several bedrock wells (AB-10BR, AB-12BR, AL-1BR, AL-3BR, and AL-4BR) were installed for vertical delineation of constituent migration. Four wells (GWA-10S/D, GWA-11S/D) were installed to provide refined assessment of potential constituent migration east of the ash basin. Seven additional background well locations (GWA-12S/D/BR, GWA-13S/D, BG- 1BR, and BG-3BR) were installed to provide information regarding constituent concentrations to assist in an understanding of site background concentrations. Additional Assessment 2017 Additional monitoring wells were installed in 2017 for CAMA purposes in response to NCDEQ requests. These wells were installed within the footprint of the ash basin and Dry Ash Landfill, as well as downgradient areas east of the ash basin. AB-1BRL, AL- 2BRLL, AL-4BRL, AB-6BRL were installed for additional vertical delineation of potential groundwater impacts. AB-6BRA, BG-1BRA, GWA-13DA and GWA-2DA were installed as replacement wells where potential grout contamination was present in initial CSA monitoring wells. GWA-14S/D were installed upgradient of the ash basin as additional background wells to refine the understanding of groundwater flow across the Site. GWA-11BR and GWA-15S were installed to further refine understanding of potential groundwater impacts in the downgradient areas east of the ash basin. Boring logs and well construction records for all of these recently installed wells are included in Appendix F. Analytical results are included in Appendix B, Table 1. Well Abandonments Several wells installed for the initial CSA were determined to be impacted from possible grout contamination associated with well construction. GWA-21), GWA-13D, BG-1BR, AB-6BR were properly abandoned by a NC -licensed driller and replaced. Historical monitoring wells associated with the FGD landfill (MS-1 through MS-7) have also been abandoned since they were determined to not be a part of the routine groundwater monitoring network in accordance with the solid waste landfill permit requirements. 2.8 Summary of Initial Abatement, Source Removal, or other Corrective Action No imminent hazard to human health or the environment has been identified; therefore, initial abatement and emergency response actions have not been required. In conjunction with decommissioning activities and in accordance with CAMA requirements, Duke Energy plans to permanently close the MSS ash basin in accordance with the pending risk ranking. Abatement activities conducted in preparation for ash Page 2-11 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra basin closure include the design and construction of new lined retention basin systems to handle certain stormwater and wastewater at the Site. In addition, Duke Energy is in the process of converting to dry handling of bottom ash (anticipated by late 2018 or early 2019). Proposed corrective action will be outlined in the updated CAP. Page 2-12 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 3.0 SOURCE CHARACTERISTICS Potential sources evaluated in this assessment are contained wholly or partially within the ash waste boundary as depicted on Figure 2-4. For the MSS, source areas include the ash management areas as described in Section 2.3. 3.1 Coal Combustion and Ash Handling System Coal ash is produced from the combustion of coal. The coal is dried, pulverized, and conveyed to the burner area of a boiler. The smaller particles produced by coal combustion, referred to as fly ash, are carried upward in the flue gas and are captured by an air pollution control device, such as an electrostatic precipitator. The larger particles of ash that fall to the bottom of the boiler are referred to as bottom ash. Coal-fired power production began at MSS in 1965. At the Site, there is a single ash basin located northwest of the Plant that contains ash generated from the Plant's historic coal combustion. The ash basin is approximately 394 acres in size, is constructed with an earthen dike, and contains approximately 16,700,000 tons of CCR (Duke Energy, updated August 11, 2017). The ash basin dam is an earthen embankment armored with rip rap on the basin side and on the downslope base of the dam. The perimeter of the basin is mostly unaltered and well -vegetated with the exception of the ash basin dam and a small shoreline section on the east (near the forebay) that are armored with rip rap. The ash basin dam and dam access road are raised about 10 feet higher than the ash basin water level. About one-half of the ash basin is covered with standing water. All coal ash from MSS was managed in the ash basin from approximately 1965 until 1984. Fly ash from the electrostatic precipitators was collected in hoppers. Bottom ash and boiler slag was collected in the bottom of the boilers. After collection, both fly ash and bottom ash/boiler slag were sluiced to the ash basin using conveyance water withdrawn from Lake Norman. Since 1984, fly ash has mainly been disposed of in the on -site dry ash landfills and the sluicing of bottom ash to the ash basin has continued. During operations of the coal-fired units, the sluice lines discharged the water/slurry and other flows to the southwest portion of the ash basin. Inflows to the ash basin fluctuate due to variability in station operations and weather. Refer to Figure 2-4 for a depiction of these features. Duke Energy is authorized to discharge wastewater from MSS to receiving waters designated as the Catawba River (Lake Norman) in accordance with NPDES Permit NC0004987 dated October 1, 2016. The NPDES permit authorizes the following discharges in accordance with effluent limitations, monitoring requirements, and other Page 3-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra conditions set forth in the permit: Outfall 001 discharges once -through cooling water and intake screen backwash. Outfall 002 discharges treated wastewater (consisting of metal cleaning wastes, coal pile runoff, ash transport water, domestic wastewater, storm water, low volume wastes, and FGD wet scrubber wastewater) from the ash settling basin. Additional permitted discharges include two yard sump overflows (Outfall 002A and 002B) and two internal outfalls (Internal Outfall 003 and 004). Two groundwater seeps (Seep Outfalls 101 and 102) are also permitted to discharge to the Catawba River (Lake Norman). 3.2 General Physical and Chemical Properties of Ash Coal ash consists of fly ash and bottom ash produced from the combustion of coal. The physical and chemical properties of coal ash are determined by reactions that occur during the combustion of the coal and subsequent cooling of the flue gas. Physical Properties Approximately 70 percent to 80 percent of the ash produced during coal combustion is fly ash (Young, S.C., 1993). Typically 65 percent to 90 percent of fly ash has particle sizes that are less than 0.010 millimeter (mm). In general, fly ash has a grain size distribution similar to that of silt. The remaining 20 percent to 30 percent of ash produced is considered bottom ash. Bottom ash consists of angular particles with a porous surface and is normally gray to black in color. Bottom ash particle diameters can vary from approximately 38 mm to 0.05 mm. In general, bottom ash has a grain size distribution similar to that of fine gravel to medium sand (EPRI, Coal ash disposal manual: Third edition, 1995). Site -specific physical properties of ash analysis from the MSS are presented in Table 3-1. Based on published literature not specific to MSS, the specific gravity of fly ash ranges from 2.1 to 2.9, and the specific gravity of bottom ash typically ranges from 2.3 to 3.0. The permeability of fly ash and bottom ash vary based on material density, but would be within the range of a soil with a similar gradation and density (EPRI, Coal ash disposal manual: Third edition, 1995). Chemical Properties The specific mineralogy of coal ash varies based on many factors, including the chemical composition of the coal, which is directly related to the geographic region where the coal was mined; the type of boiler where the combustion occurs (i.e., thermodynamics of the boiler); and air pollution control technologies employed. The overall chemical composition of coal ash resembles that of siliceous rocks from which it was derived, particularly shale. Oxides of silicon, aluminum, iron, and calcium make up more than 90 percent of most siliceous rocks, soils, fly ash, and bottom ash. Page 3-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Other major and minor elements (sulfur, sodium, potassium, magnesium, titanium) make up an additional 8 percent, while trace constituents account for less than 1 percent. The following constituents are considered to be trace elements: arsenic, barium, cadmium, chromium, lead, mercury, selenium, copper, manganese, nickel, lead, vanadium, and zinc (Pugh, Whetstone, & Redwine, September 2010). The majority of fly ash particles are glassy spheres mainly composed of amorphous or glassy aluminosilicates, crystalline matter, and carbon. Figure 3-1 presents a photograph of ash collected from the ash basin at Duke Energy's Cliffside Steam Station (considered representative of the ash at the MSS) showing a mix of fly ash and bottom ash at 10 µm and 20 µm magnifications. The glassy spheres can be observed in the photograph. The glassy spheres are generally immune to dissolution. During the later stages of the combustion process and as the combustion gases are cooling after exiting the boiler, molecules from the combustion process condense on the surface of the glassy spheres. These surface condensates consist of soluble salts [e.g., calcium (Ca 2+) and sulfate (SO2-)], metals [e.g., copper (Cu) and zinc (Zn)], and other minor elements [e.g., boron (B), selenium (Se), and arsenic (As)] (EPRI, 1995). The major elemental composition of fly ash (approximately 95 percent by weight) is composed of mineral oxides of silicon, aluminum, iron, and calcium. Oxides of magnesium, potassium, titanium and sulfur comprise approximately 4 percent of fly ash by weight (EPRI, 1995). Trace elemental composition of fly ash typically is approximately 1 percent by weight and may include antimony, arsenic, barium, boron, cadmium, chromium, copper, manganese, mercury, nickel, lead, selenium, silver, thallium, zinc, and other elements. For comparison, Figure 3-2 shows the elemental composition of fly ash and bottom ash compared with typical values for shale and volcanic ash. Table 3-2 shows the bulk composition of fly ash and bottom ash compared with typical values for soil and rock. In addition to these constituents, fly ash may contain unburned carbon. Bituminous coal ash typically yields slightly acidic to alkaline solutions (pH of 5 to 10) on contact with water. The geochemical factors controlling the reactions associated with leaching of ash are complex. Factors such as the chemical speciation of the constituent, solution pH, solution -to -solid ratio, and other factors control the chemical concentration of the resultant solution. Constituents that are held on the glassy surfaces of fly ash such as boron, arsenic, and selenium may initially leach more readily than other constituents. As noted in Table 3-2, aluminum, silicon, calcium, and iron represent the larger fractions of fly ash by weight. Calcium and iron may limit the release of arsenic by forming calcium -arsenic precipitates. Formation of iron hydroxide compounds may also sequester arsenic and retard or prevent release of arsenic to the environment. Similar Page 3-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra processes and reactions may affect other constituents of concern; however, certain constituents such as boron and sulfate will likely remain highly mobile. In addition to the variability that might be seen in the mineralogical composition of the ash, based on different coal types, different age of ash in the basin, etc., it is anticipated that the chemical environment of the ash basin varies over time, distance, and depth. Pugh, Whetstone, & Redwine (September, 2010) reports that 64 samples of coal combustion products (including fly ash, bottom ash, and flue gas desulfurization residue) from 50 different power plants were subjected to USEPA Method 1311 Toxicity Characteristic Leaching Procedure (TCLP) leaching and no TCLP result exceeded the TCLP hazardous waste limit (Pugh, Whetstone, & Redwine, September 2010). Figure 3- 3 provides the results of that testing. 3.3 Site -Specific Coal Ash Data Source characterization was performed in the following areas of the MSS to identify the physical and chemical properties of source material: '67 Ash Basin 167 Dry Ash Landfill Phase I '67 Dry Ash Landfill Phase II '67 PV Structural Fill The source characterization involved development of selected physical properties of ash, identifying the constituents found in ash, measuring concentrations of constituents present in the ash pore water and free-standing water within the ash basin, and performing laboratory analyses to estimate constituent concentrations resulting from the leaching process. The physical and chemical properties evaluated as part of this characterization will be used to better understand impacts to soil and groundwater from the source area and will also be utilized as part of groundwater model development in the CAP. Source characterization was performed through the completion of soil borings, installation of monitoring wells, and associated solid matrix and aqueous sample collection and analysis. Sampling locations used for source characterization are shown on Figure 2-4. '67 Ash samples were collected for chemical analyses from the following borings advanced within the ash basin boundary: AB-31), AB-41), AB-51), AB-6D/BR, AB - Page 3-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 7D, AB-81), AB-10D, AB-12D, AB-13D, AB-14D, AB-15D, AB-17D, AB-18D, AB- 20D, AB-21S, SB-2, SB-3, SB-7, SB-10, SB-11, SB-13, and SB-14. '67 No soil borings were advanced within the footprint of the dry ash landfill (Phase I) for the CSA. '67 Six ash samples were collected for chemical analyses from three borings (AL-21), AL-31), and AL-41)) advanced within the dry ash landfill boundary (Phase II). 17 Thirteen (13) ash samples were collected for chemical analyses from eight borings (AB-20D, SB-2, SB-3, SB-4, SB-5, SB-7, SB-8, and SB-9) advanced within the PV structural fill boundary. Note that a portion of the PV structural fill is situated on top of the ash basin. Thus, shallow ash samples (e.g. 3 feet to 5 feet below ground surface [bgs] interval collected from borings AB-20D and SB-3) represent ash within the PV structural fill. All ash samples collected from borings SB-2 and SB-7 also represent ash within the PV structural fill. Other ash samples collected from deeper intervals likely represent ash beneath the structural fill and within the ash basin. Due to a laboratory error, ash samples collected from boring SB-9 were not analyzed for total inorganics. Those results are presented in Table 3-3. In addition to total inorganic testing of ash samples, 13 ash samples collected from borings within the ash basin, dry ash landfill (Phase II), and the PV structural fill were analyzed for leachable inorganics using Synthetic Precipitation Leaching Procedure (SPLP). The purpose of the SPLP testing is to evaluate the leaching potential of constituents that may result in impacts to groundwater above 2L Standards or Interim Maximum Allowable Concentrations (IMACs). The ash samples collected from the ash basin (seven samples), the dry ash landfill (Phase II) (two samples), and the PV structural fill (four samples) for SPLP testing were collected from the deeper ash sample in the boring (i.e., approximately 2 feet to 3 feet above the ash/soil interface where field conditions allowed). Those results are presented in Table 3-4. Five free water samples (SW-1, SW-2, SW-3, SW-4, and SW-5) are samples from open water within the ash basin. Chemical speciation samples were collected at two of the five ash basin surface water sample locations (Appendix B, Table 2). Physical Properties of Ash Physical properties (grain size, effective porosity, and moisture content) terminations were performed on 10 samples from the ash basin (Table 3-1). Physical properties were measured using ASTM methods. The ash samples analyzed from the MSS are Page 3-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra dominated by silt -fractions, followed by fine sands and then clays in a smaller percentage. Bottom ash is generally characterized as a loose, poorly graded (fine- to coarse -grained) sand. Fly ash is generally characterized as a moderately dense silty fine sand or silt. Compared to soil, fly ash exhibits a lower specific gravity. Moisture content of ash samples analyzed from MSS averaged 42 percent, with one sample as high as 86.7 percent and the lowest at 19.8 percent. Estimated effective porosities, or specific yields, averaged 15 percent, with the highest at 32 percent and the lowest at 7.5 percent for two (2) of the 10 samples. The thickness of ash in the areas of investigation varied. Ash was observed up to 85 feet thick within the ash basin, 111 feet within the dry ash landfill (Phase II), and 71 feet within the PV structural fill. The contact between ash and underlying soil was distinct in each boring as physical intrusion of ash into the underlying soils appeared to have been negligible. The presence of saturated ash varies as well. Saturated ash thickness ranges from less than one foot in an upper arm of the basin (AB-7) up to 52 feet near the central portion of the basin (AB-3). On average, saturated ash thickness across the basin is approximately 25 feet. Chemical Properties of Ash 57 samples of ash were collected and analyzed for total constituent concentrations and total organic carbon (TOC) (Table 3-3). Concentrations of antimony, arsenic, boron, and selenium in ash samples collected within the ash basin were reported as being greater than North Carolina Preliminary Soil Remediation Goals (PSRG) for Protection of Groundwater (POG) and Site -specific provisional background threshold values (PBTV) for soils. In previous reports, barium, cobalt, iron, and manganese were also reported to exceed a PSRG; however, updated background concentrations are greater for those constituents than for any reported values from the ash samples. In addition to total inorganic testing of ash samples, 13 ash samples collected from borings within the ash basin, dry ash landfill (Phase II), and the PV structural fill were analyzed for leachable inorganics using SPLP (Table 3-4). The SPLP was designed to more closely approximate leaching from a material by rainwater. The SPLP is not intended to mimic complete leaching processes, and results are not necessarily indicative of resultant concentrations in groundwater. SPLP analytical results are compared with 2L and/or IMAC; however, those results do not represent groundwater samples and are presented here for comparative purposes only. SPLP results generally exceed the 2L or IMAC standard in one or more samples for several COIs; cobalt and iron tend to leach at lower concentrations from ash than soil, while manganese appears to leach at similar concentrations. Page 3-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Chemistry of Ash Pore Water Ash pore water refers to water samples collected from wells installed within the waste boundary and screened in the ash layer. Twenty (20) ash pore water monitoring wells (AB-3S, AB-4S, AB-4SL, AB-5S, AB-6S, AB-7S, AB-8S, AB-10S, AB-10SL, AB-12S, AB- 12SL, AB-13S, AB-14S, AB-15S, AB-15SL, AB-17S, AB-18S, AB-20S, AB-21S, and AL-03S) were installed within the ash basin and screened within the ash layer. Since installation of the wells in early 2015, the majority have been sampled nine times including the third quarter of 2017 as part of the CAMA monitoring program (Appendix B, Table 1). Chemical speciation samples were also collected from five ash pore water monitoring wells (AB-12S, AB-12SL, AB-15SL, AB-4S, and AB-4SL) within the ash basin (Appendix B, Table 1). Ash pore water analytical results are compared with 2L and/or IMAC standards for reference purposes only. The ash basin is a permitted wastewater system; therefore, comparison of ash pore water within the wastewater treatment residuals (ash) to 2B or 2L/IMAC standards is for source contribution information rather than compliance monitoring purposes. '67 Antimony, arsenic, barium, beryllium, boron, cadmium, chloride, chromium, cobalt, hexavalent chromium3, iron, manganese, mercury, nickel, pH, selenium, sulfate, thallium, TDS, vanadium, and zinc have been detected above the corresponding 2L or IMAC in one or more ash pore water samples (Appendix B, Table 1). �7 Ten (10) of the above constituents (barium, beryllium, cadmium, chloride, chromium, hexavalent chromium3, mercury, nickel, selenium, and zinc) are intermittently detected in exceedance of either 2L or IMAC in ash pore water. �7 Barium, beryllium, chloride, nickel, and zinc have only been detected in exceedance of the respective 2L or IMAC in one or two ash pore water wells. 47 Hexavalent chromium was detected in exceedance of the 2L for total chromium (10 µg/L) in one sample from AB-21S and mercury was detected in exceedance of the 2L in one sample from AB-4S. Concentrations of constituents detected consistently above 2L or IMAC in ash pore water have been relatively stable. Piper diagrams are used to graphically depict the general chemical character of water by plotting the relative concentrations of major ions 3 Hexavalent chromium results are compared to the 2L standard for total chromium (10 µg/L) for comparative purposes only. Page 3-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra on a series of tri-linear diagrams. These points are then projected onto a central diamond plot to show the overall chemical character of the water (Piper, 1944). Using these diagrams to understand both source area water characteristics (ash pore water) and natural water characteristics (background water) provides the basis for useful evaluation of downgradient water. Evaluation of the ash pore water Piper diagram (Figure 3-4) results in the following observations: 167 Ash pore water is characterized by two water types, calcium -bicarbonate in samples collected from the northernmost arm of the ash basin and calcium - sulfate type in samples collected closer to the central area of the ash basin. 167 Samples collected near the central area of the basin are similar to findings in a EPRI study of 40 ash leachate water samples collected from 20 different coal ash landfills and impoundments which characterized bituminous coal ash leachate as calcium -magnesium -sulfate water type and subbituminous coal ash leachate as sodium -calcium -sulfate water type (EPRI, 2012). 17 Location AB-12 demonstrates a shift from calcium -sulfate type water to calcium - bicarbonate type water with depth. While boron concentrations remain above 2L at AB-12SL, concentrations of chloride and sulfate are significantly reduced and less than the 2L standard. Comparisons of background and downgradient water types are discussed in more detail in Section 10. Radium and uranium have been monitored at five of the ash pore water wells; only one data set is available for three of those five wells to date. Total uranium concentrations exceeded the corresponding 2L value for every sample from AB-10SL (Appendix B, Table 1). Free water samples collected from within the waste boundary are compared with 2B (Class B, WS-IV) standards for reference purposes only. Concentrations of total arsenic, chloride, nickel, selenium, sulfate, and TDS are greater than these 2B values. Dissolved fractions of beryllium, cadmium, copper, lead, nickel, and zinc are also reported above the respective 2B values (Appendix B, Table 2). Most of the ash pore water within the ash basin exhibits neutral or slightly acidic pH, which is consistent with values observed in background, sidegradient and downgradient wells across the Site. Page 3-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 4.0 RECEPTOR INFORMATION Section §130A-309.201(13) of the CAMA defines receptor as "any human, plant, animal, or structure which is, or has the potential to be, affected by the release or migration of contaminants. Any well, constructed for the purpose of monitoring groundwater and contaminant concentrations shall not be considered a receptor." In accordance with the NORR CSA guidance, receptors cited in this section refer to public and private water supply wells (including irrigation wells and unused wells) and surface water features. Additional receptors (described in Section 12.0) were evaluated as part of the risk assessment related to the CSA effort. The NORR CSA receptor survey guidance requirements include listing and depicting water supply wells, public or private, including irrigation wells, and unused wells (other than those that have been properly abandoned in accordance with 15A NCAC 2C .0100) within a minimum of 1,500 feet of the known extent of contamination. In NCDEQ's June 2015 response to Duke Energy's proposed adjustments to the CSA guidelines, NCDEQ DWR acknowledged the difficulty in determining the known extent of contamination at this time. DWR stated that it expected all drinking water wells located 2,640 feet (0.5 mile) downgradient from the established compliance boundary to be documented in the CSA reports as specified in the CAMA requirements. The approach to the receptor survey in this CSA includes listing and depicting all water supply wells (public or private, including irrigation wells and unused wells) within a 0.5-mile radius of the ash basin compliance boundary (Appendix D). The NORR CSA guidance requires that subsurface utilities be mapped within 1,500 feet of the known extent of contamination in order to evaluate the potential for preferential pathways. Identification of piping adjacent to the ash basin was conducted by Stantec in 2014 and 2015, and utilities adjacent to the Site were also included on a 2015 topographic map by WSP USA, Inc. and incorporated in Figure 4-1. The Stantec and WSP files are included in Appendix E. 4.1 Summary of Receptor Survey Activities Surveys to identify potential receptors — including public and private water supply wells — and surface water features within a 0.5-mile radius of the MSS ash basin compliance boundary have been reported to NCDEQ: ,67 Marshall Steam Station Ash Basin Drinking Water Supply Well and Receptor Survey (HDR, September 30, 2014a) Page 4-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra ,67 Marshall Steam Station Ash Basin Supplement to Drinking Water Supply Well and Receptor Survey (HDR, November 6, 2014b) 07 Comprehensive Site Assessment Supplement 2 (HDR, August 4, 2016b) These reports are included as Appendix D. The purpose of the receptor survey was to identify the potential exposure locations that are critical and should be considered for inclusion in the groundwater transport modeling and screening level risk assessment activities. The supplementary information was obtained from responses to water supply well survey questionnaires mailed to owners of property within a 0.5-mile (2,640-foot) radius of the MSS ash basin compliance boundary. The questionnaires requested information about the presence of water supply wells and well usage. The survey activities included contacting agencies and/or reviewing the following records to identify public and private water supply sources, confirm the location of wells, and/or identify any wellhead protection areas located within a 0.5-mile radius of the MSS ash basin compliance boundary: 41' NCDENR Public Water Supply Sections most current Public Water Supply Water Sources Geographic Information Systems (GIS) point data set ,610 NCDENR DWR Source Water Assessment Program online database for public water supply sources 07 Environmental Data Resources local/regional water agency records review 101 Catawba County Environmental Health Department 101 City of Hickory Public Utilities Department 17 USGS National Hydrography Dataset In addition, a field reconnaissance was performed on March 31, 2014, to identify public and private water supply wells (including irrigation wells and unused wells) and surface water features located within a 0.5-mile radius of the MSS ash basin compliance boundary. A windshield survey was conducted from public roadways to identify water meters, fire hydrants, valves, and any potential well heads/well houses, and Duke Energy on -Site personnel identified water supply wells located on Duke Energy property. During the week of October 8, 2014, 262 water supply well survey questionnaires were mailed to property owners within a 0.5-mile radius of the MSS ash basin compliance Page 4-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra boundary requesting information on the presence of water supply wells and well usage for the properties. The mailing list was compiled from a query of the parcel addresses included in the Catawba County GIS database utilizing the 0.5-mile offset. 4.2 Summary of Receptor Survey Findings As part of the 2015 CSA report, the previously completed Receptor Survey activities were updated based on the CSA Guidelines. The update included contacting agencies and/or reviewing records to identify public and private water supply sources (Section 4.1) and reviewing any questionnaires that were received after the submittal of the November 2014 supplement to the September 2014 receptor survey (e.g. questionnaires received after October 31, 2014). A summary of the receptor survey findings is provided below. The identified water supply wells are shown on the USGS map on Figure 4-2 and on an aerial photograph on Figure 4-3. Available property information for the identified water supply wells is provided in Table 4-1. 17 Based on the known groundwater flow direction, no water supply wells are located downgradient of the MSS ash basin. 101 All private water supply wells located within a 0.5-mile radius of the MSS ash basin compliance boundary are located to the north, west, and south of the facility. 167 Four public water supply wells were identified within a 0.5-mile radius of the MSS ash basin compliance boundary. 167 No wellhead protection areas were identified within a 0.5-mile radius of the ash basin compliance boundary. Catawba County owns the water system serving the area around MSS but does not operate it. The City of Hickory, through contract with Catawba County, provides operations, maintenance, and management of the system, and anyone connected to the system becomes a customer of the City of Hickory. As required by G.S. 130A-309.211(cl) of House Bill 630 (HB630), Duke Energy evaluated the feasibility and costs of providing a permanent replacement water supply to eligible households. Households were eligible if any portion of a parcel of land crossed the 0.5- mile compliance line described in HB630 and if the household currently was using well water or bottled water (under Duke Energy's bottled water program) as the drinking water source. Undeveloped parcels were identified but were not considered "eligible" because groundwater wells are not currently utilized as a drinking source. A Potable Page 4-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Water Programmatic Evaluation (Dewberry, October 2016) (Appendix D) was conducted. That evaluation included a survey of eligible households, a preliminary engineering evaluation, a cost estimate, and a schedule. The evaluation report also included a listing of, and relevant information about, households and properties within the survey area, as well as maps depicting property locations, including those properties for which a replacement water supply will be provided. 4.2.1 Water Supply Lines The City of Hickory already has a municipal line located along Sherrills Ford Road; therefore the majority of the households would require only service lines. As discussed in Dewberry's November 2016 report and the accompanying Duke Energy cover letter, connecting water supply lines to properties in the vicinity of MSS would require water line extensions along Mollys Backbone Road (approximately 175 feet), Steamplant Road (approximately 500 feet), Gregory Road (approximately 650 feet), Marshall Road (approximately 1700 feet), and Greenwood Road to Clement Circle (approximately 1800 feet) (Appendix D). 4.2.2 Public Water Supply Wells Four public water supply (PWS) wells were identified within a 0.5-mile radius of the ash basin compliance boundary; however, one of those wells (PWS: NC0118497) is not currently in use. Water supply wells classified as transient, non -community are located at the Midway Restaurant and Marina (PWS: NC0118622) and The Old Country Church (PWS: NC0118736). The Catawba County Environmental Health Department had records for one public water supply well (PWS ID: 0118676), owned by Duke Energy. 4.2.3 Private Water Supply Wells Based on the findings required under G.S. 130A-309.211(cl) of House Bill 630,125 eligible households, business and schools/churches near MSS are considered eligible for the option of a connection to a public water supply or the installation of a water treatment system. Two (2) industrial locations (located on Steamplant Road) are considered eligible for installation of a water treatment system. 4.3 Private Water Well Sampling NCDEQ coordinated sampling of private water supply wells identified within a half - mile radius of the ash basin compliance boundary from February to October in 2015. Several private wells were sampled twice during this time, however the most recent sample available is referred to for evaluation purposes. Ten (10) additional water supply wells near MSS were sampled between September 2016 and February 2017 by Duke Energy. As of January 2018, water samples have been collected from 48 of the Page 4-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra wells identified by property owners. Tabulated results, provided by Duke Energy, for the NCDEQ and Duke Energy sampling efforts, along with exceedances of 2L/IMAC standards and other reporting limits, are included in Appendix B, Table 1. The locations of the sampled wells are included on Figure 4-3. Evaluation of water supply well analytical results using Piper diagrams are not included due to charge balance differences beyond acceptable limits (10%). Greater values of charge balance differences may indicate a missing analyte, laboratory error, or elevated detection limit. A review of the analytical data for the 48 private water supply wells sampled indicated several constituents were detected above 2L or IMAC standards, including pH (33 wells), iron (five wells), manganese (two wells), zinc (one well), TDS (one well), and vanadium (39 wells). Background concentration values for bedrock are used for comparative purposes, as most private supply wells are completed as open -holes within bedrock. Concentrations of constituents exceeded the respective bedrock PBTV for a number of private water supply wells including: 07 Arsenic (2 wells) 161P Beryllium (1 well) ,61P Cadmium (6 wells) ,61P Chloride (13 wells) ,0 Iron (1 well) 07 Manganese (4 wells) y Molybdenum (1 well) 07 Nickel (1 well) y Sulfate (1 well) 167 Strontium (6 wells) y TDS (2 wells) None of the 39 detections of vanadium were greater than the PBTV. Concentrations of aluminum, calcium, hexavalent chromium, copper, lead, sodium, and zinc were detected higher than the respective PBTV, however these constituents are not identified as constituents of interest at MSS (Section 10.3.4). Considerations for evaluating concentrations greater than PBTVs in private water wells include: Page 4-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra First, the PBTVs have been developed using groundwater data from a set of ten (10) background well cluster locations, all located at MSS. The geochemical data from these wells may not be representative across the broader area encompassed by the water supply wells surrounding the Site. Second, well construction may influence analytical results. For example, galvanized pipe could yield high zinc concentrations, and brass components in well pumps and valves are a source of lead. Information concerning well construction and piping materials is important to have before attributing detections of ash -related constituents solely to the geochemistry of the groundwater. Third, there is very limited information available about the wells (date of installation, drilling method, well depth, casing length, pump set depth, etc.). Many private water supply wells in this part of the Piedmont are open -hole bedrock wells. A shallow surface casing is installed, and then the well is drilled to a depth that may be as shallow as 40 or 50 feet or as deep as several hundred feet. When a groundwater sample is collected, it is unknown from what part of the bedrock aquifer the groundwater is drawn. A review of the limited available well construction data for the supply wells, based on questionnaire responses from property owners, indicates a few wells may be installed to relatively shallow depth (less than 100 feet) which indicates they may be installed within residuum or transition zone/partially weathered rock (PWR) material. Groundwater geochemistry in residuum, PWR, and fractured bedrock aquifers is variable. Boring logs for these private wells are not available, so lithologic information at each location is unknown. A fourth reason for considering the apparent exceedances of PBTVs in groundwater is that, as previously described, private water wells in bedrock are typically installed as open -hole wells. Care must be taken when comparing geochemical data from these wells to background concentrations derived from carefully drilled and installed groundwater monitoring wells with machine -slotted well screens, proper filter pack installation, proper well development, and specific sample collection procedures employed. Based on the bedrock groundwater flow direction at MSS (discussed in Section 6.3), private water supply wells are located hydraulically upgradient of the ash basin. Information evaluated as part of the CSA indicates that the identified water supply wells are not impacted by ash basin operations; constituent concentrations detected in water supply wells are deemed to not be sourced from the ash basin. Page 4-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Comparison of analytical results with the 2L and IMAC standards and statistically derived PBTVs yields the following observations: ,67 Boron has not been detected in any water supply wells sampled greater than the PBTV of 50 µg/L. 07 Levels of pH, iron, manganese, TDS, and vanadium are consistent with observed background concentrations, although the 2L or IMAC standard has been exceeded. 167 Both the 2L standard and PBTV are exceeded for iron (MR-4), manganese (MR-12 and MR-16), TDS (MR-37), and zinc (MR-2). 4.4 Surface Water Receptors The surface water receptors for the MSS include Lake Norman (Catawba River) and the unnamed tributary to Lake Norman east of the ash basin. Lake Norman is used as a water supply for Lincoln County and the Town of Mooresville. The Town of Mooresville pumps raw (untreated) water from Lake Norman and treats the water at one of two treatment plants, both located on Charlotte Highway (U.S. Highway 21). The two plants have a combined treatment capacity of 18 million gallons per day. Lincoln County treats raw water from Lake Norman at the Lincoln County Water Treatment Facility located at 7674 Tree Farm Lane near Denver, NC. These surface water intakes are shown on Figure 4-4. Page 4-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 5.0 REGIONAL GEOLOGY AND HYDROGEOLOGY North Carolina is divided into three physiographic provinces: the Atlantic Coastal Plain, Piedmont, and Blue Ridge (Fenneman, 1938). The MSS site is located in the Piedmont province. The Piedmont province is bounded to the east and southeast by the Atlantic Coastal Plain and to the west by the escarpment of the Blue Ridge Mountains, with a width ranging from 150 miles to 225 miles in the Carolinas (LeGrand, 2004). A discussion of regional geology and hydrogeology relevant to the MSS site is provided below. 5.1 Regional Geology The topography of the Piedmont region is characterized by low, rounded hills and long, rolling, northeast -southwest trending ridges (Heath, 1984). Stream valley to ridge relief in most areas ranges from 75 feet to 200 feet. Along the Coastal Plain boundary, the Piedmont region rises from an elevation of 300 feet above mean sea level to an elevation of 1,500 feet at the base of the Blue Ridge Mountains (LeGrand, 2004). The MSS site is underlain by the Charlotte and Kings Mountain terranes, two of a number of tectonostratigraphic terranes in the southern and central Appalachians that have been defined. The Site is located within the western portion of the larger Carolina superterrane (Figure 5-1) (Horton, Jr., Drake, Jr., & Rankin, 1989; Hibbard, Stoddard, Secor, & Dennis, 2002; Hatcher, Bream, & and Merschat, 2007). On the northwest side, the Charlotte/Kings Mountain terranes are in contact with the Inner Piedmont zone along the Central Piedmont suture along its northwest boundary. The Kings Mountain terrane is distinguished by its abundance of metasedimentary and metavolcanic rocks at lower metamorphic grade than the metaigneous rocks of higher metamorphic grade in the Charlotte terrane (Butler J. , 1991; Butler & Secor, 1991; Hatcher, Bream, & and Merschat, 2007). The Charlotte terrane is differentiated from the Carolina terrane to the southeast by its higher metamorphic grade, and portions of the boundary may be tectonic (Secor, Balinsky, & Colquhoun, 1998; Dennis, Shervais, & Secor, 2000). The Charlotte terrane is dominated by a complex sequence of plutonic rocks that intrude a suite of metaigneous rocks (amphibolite metamorphic grade) including mafic gneisses, amphibolites, metagabbros, and metavolcanic rocks with lesser amounts of granitic gneiss and ultramafic rocks with minor metasedimentary rocks. The general structure of the terrane is primarily a function of plutonic contacts. The Kings Mountain terrane has a distinctive metasedimentary sequence with interlayered quartzite, metaconglomerate, marble, and schists derived from both sedimentary and volcanic protoliths (Keith & Sterrett, 1931; Kesler, 1944; King, 1955; Page 5-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Horton & Butler, 1977). Rocks of the terrane are intensely deformed with a tectonic style distinct from the adjacent terranes (Horton and Butler 1991; Schaeffer 1981). A regional geologic map for the area near the MSS is presented in Figure 5-2. 5.2 Regional Hydrogeology The groundwater system in the Piedmont province, in most cases, is comprised of two interconnected layers, or mediums: 1) residual soil/saprolite and weathered fractured rock (regolith) overlying 2) fractured crystalline bedrock (Heath R. , 1980; Harned & Daniel, 1992) (Figure 5-3). The regolith layer is a thoroughly weathered and structureless residual soil that occurs near the ground surface with the degree of weathering decreasing with depth. The residual soil grades into saprolite, a coarser grained material that retains the structure of the parent bedrock. Beneath the saprolite, partially weathered/fractured bedrock occurs with depth until sound bedrock is encountered. This mantle of residual soil, saprolite, and weathered/fractured rock is a hydrogeologic unit that covers and crosses various types of rock (LeGrand, 1988). This layer serves as the principal storage reservoir and provides an intergranular medium through which the recharge and discharge of water from the underlying fractured rock occurs. Within the fractured crystalline bedrock layer, the fractures control both the hydraulic conductivity and storage capacity of the rock mass. A transition zone (TZ) at the base of the regolith has been interpreted to be present in many areas of the Piedmont. Harned and Daniel (1992) described the zone as consisting of partially weathered/fractured bedrock and lesser amounts of saprolite that grades into bedrock, and they described the zone as "being the most permeable part of the system, even slightly more permeable than the soil zone." Harned and Daniel (1992) suggested the zone may serve as a conduit of rapid flow and transmission of contaminated water. Most of the information supporting the existence of the TZ until recently was qualitative and based on observations made during the drilling of boreholes and water - wells, although some quantitative data is available for the Piedmont region (Stewart, 1964; Stewart, Callahan, & Carter, 1964; Nutter & Otton, 1969; Harned & Daniel, 1992). Schaeffer (April 2, 2009; 2014a), using a database of 669 horizontal conductivity measurements in boreholes at six locations in the Carolina Piedmont, statistically showed that a TZ of higher hydraulic conductivity exists in the Piedmont groundwater system when considered within Harned and Daniel's (1992) conceptual framework depicting two types of bedrock. The TZ is comprised of partially weathered rock, open, steeply dipping fractures, and low -angle stress relief fractures, either singly or in various combinations below refusal - auger, roller cone, or casing advancer (Schaeffer, 2011; Schaeffer, 2014b). It has less Page 5-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra advanced weathering relative to the regolith, and generally, the weathering has not progressed to the development of clay minerals that would decrease the permeability of secondary porosity. The characteristics of the TZ vary widely based on the interaction of rock type, degree of weathering, degree of systematic fracturing, presence of stress - relief fracturing, and the general characteristics of the bedrock (foliated/layered, massive, or in between). The TZ is not a continuous layer between the regolith and bedrock; it thins and thickens within short distances and is absent in places (Schaeffer, 2011; Schaeffer, 2014b). The absence, thinning, and thickening of the TZ are related to the characteristics of the underlying bedrock (Schaeffer, 2014b). The TZ varies due to different rock types and associated rock structure. Harned and Daniel (1992) further divided the bedrock into two conceptual models: 1) foliated/layered (metasedimentary and metavolcanic sequences) and 2) massive/plutonic (plutonic and metaplutonic complexes) structures. Strongly foliated/layered rocks are thought to have a well -developed TZ due to the breakup and weathering along the foliation planes or layering, resulting in numerous rock fragments (Harped & Daniel, 1992). More massive rocks are thought to develop an indistinct TZ because they do not contain foliation/layering and tend to weather along relatively widely spaced fractures (Harned & Daniel, 1992). Schaeffer (2014a) proved Harned and Daniel's (1992) hypothesis that foliated/layered bedrock would have a better developed TZ than plutonic/massive bedrock. The foliated/layered bedrock TZ has a statistically significant higher hydraulic conductivity than that of the massive/plutonic bedrock TZ (Schaeffer, 2014a). LeGrand's (1988; 1989) conceptual model of the groundwater setting in the Piedmont incorporates Harned and Daniel's (1992) two -medium system into an entity that is useful for the description of groundwater conditions. That entity is the surface drainage basin that contains a perennial stream (LeGrand, 1988). Each basin is similar to adjacent basins, and the conditions are generally repetitive from basin to basin. Within a basin, movement of groundwater is generally restricted to the area extending from the drainage divides to a perennial stream (Slope -Aquifer System; Figure 5-4; LeGrand 1988; 1989; 2004). Rarely does groundwater move beneath a perennial stream to another more distant stream or across drainage divides (LeGrand, 1989). The crests of the water table underneath topographic drainage divides represent natural groundwater divides within the slope -aquifer system. The concave topographic areas between the topographic divides may be considered as flow compartments that extend down slope toward Lake Norman. Page 5-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Therefore, in most cases in the Piedmont, the groundwater system is a two -medium system restricted to the local drainage basin. The groundwater occurs in a system composed of two interconnected layers: residual soil/saprolite and weathered rock overlying fractured crystalline rock separated by the TZ. Typically, the residual soil/saprolite is partially saturated and the water table fluctuates within it. Water movement is generally through the weathered/fractured and fractured bedrock. The near -surface fractured crystalline rocks can form extensive aquifers. The character of such aquifers results from the combined effects of the rock type, fracture system, topography, and weathering. Topography exerts an influence on both weathering and the opening of fractures, while the weathering of the crystalline rock modifies both transmissive and storage characteristics. The igneous and metamorphic bedrock in the Piedmont consist of interlocking crystals, and primary porosity is very low, generally less than 3 percent. Secondary porosity of crystalline bedrock due to weathering and fractures ranges from 1 percent to 10 percent (Freeze & Cherry, 1979); but porosity values of 1 percent to 3 percent are more typical. (Daniel C. I., 1990) reported that the porosity of the regolith ranges from 35 percent to 55 percent near land surface but decreases with depth as the degree of weathering decreases (Daniel III & Sharpless, 1983). In natural areas, groundwater flow paths in the Piedmont are almost invariably restricted to the zone underlying the topographic slope extending from a topographic divide to an adjacent stream. Under natural conditions, the general direction of groundwater flow can be approximated from the surface topography (LeGrand, 2004). Groundwater recharge in the Piedmont is derived from infiltration of local precipitation. Groundwater recharge occurs in areas of higher topography (i.e., hilltops), and groundwater discharge occurs in lowland areas bordering surface water bodies, marshes, and floodplains (LeGrand, 2004). Mean annual recharge in the Piedmont ranges from 4.0 to 9.7 inches per year (Cunningham & Daniel, 2001). Page 5-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 6.0 SITE GEOLOGY AND HYDROGEOLOGY Information from previous site investigations and from the soil borings and monitoring wells installed for the CSA and additional assessment were used to characterize site geology and hydrogeology. 6.1 Site Geology The MSS site is underlain by the Charlotte and Kings Mountain terranes, two of a number of tectonostratigraphic terranes that have been defined in the southern and central Appalachians. The Site is located within the western portion of the larger Carolina superterrane (furthered described in Section 5.0). The Charlotte terrane consists of an igneous complex of Neoproterozoic to Paleozoic ages (Hibbard, Stoddard, Secor, & Dennis, 2002) that range from intermediate to mafic in composition (Butler & Secor, 1991). The MSS site is located at the northern extent of the Kings Mountain terrane, which underlies the western portion of the Site. The eastern portion of the Site is underlain by rocks of the Charlotte terrane (Figure 6-1). Metamorphic grade near MSS in the Kings Mountain terrane is primarily middle greenschist to lower amphibolite grade, with the northern portion of the terrane up to upper amphibolite grade (sillimanite grade) (Butler J. , 1991). At the northern boundary of the terrane near MSS, the primary rock units underlying the western portion of the Site are the Battleground Formation (Zbs) and the High Shoals Granite (IPhs). Units mapped by Goldsmith et al. (1988) underlying the eastern portion of the MSS site are alaskitic granite described as a fine-grained light colored muscovitebiotite granite (DOga) and a fine-grained biotite gneiss of granodioritic composition of probable volcanic origin (bgf)• The fractured bedrock is overlain by a mantle of unconsolidated material known as regolith. The regolith includes residual soil and saprolite zones and, where present, alluvial deposits. Saprolite, the product of chemical weathering of the underlying bedrock, is typically composed of clay and coarser granular material and reflects the texture and structure of the rock from which it was formed. The weathering products of granitic rocks are quartz -rich and sandy textured. Rocks poor in quartz and rich in feldspar and ferro-magnesium minerals form a more clayey saprolite. The subsurface encountered at the Site is generally composed of regolith (including residual soils, fill and reworked soils, alluvium, and saprolite), TZ, and bedrock. Each zone was not encountered at every boring location. Subsurface conditions varied with topography, parent rock, and Site infrastructure. Page 6-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 6.1.1 Soil Classification �7 Ash - Ash was encountered in borings advanced within the ash basin, Dry Ash Landfill (Phase II), and structural fill, as well as in some borings advanced through the ash basin perimeter and dikes. Ash was generally described as dark yellow brown to very dark gray, non -plastic, loose to very loose, and dry to wet. �7 Fill - Fill material generally consisted of reworked sandy silts, clays, and sands that were borrowed from areas of the Site and redistributed to other areas. Fill was classified in the boring logs as silty sand, gravel with clay and sand, sand with silt and gravel, and silt. Fill is used primarily in the construction of dikes as cover for ash storage areas, and as bottom liner for ash storage areas. �7 Alluvium - Alluvium encountered in borings AB-91), AB-11D, GWA-3D, and AB-20D was classified as sand, sand with silt, and gravel with sand, wet, medium dense. '67 Residuum (Residual soils) — Residuum is the in -place weathered soil that consists primarily of silt, sand with silt, clay with sand, sandy silt with gravel, clay, sandy clay, and sandy clay with gravel at the MSS site. Residuum varied in thickness and was relatively thin compared to the thickness of saprolite. 17 Saprolite — Saprolite is soil developed by in -place weathering of rock that retains remnant bedrock structure. Saprolite at the MSS site is classified primarily as sand with silt, silty sand, sand with silt and gravel, sand, clayey sand, clayey sand with gravel, and sand with gravel. Saprolite is primarily tens of feet thick, but in some cases, it is more than 80 feet thick. 6.1.2 Rock Lithology Figure 5-2 shows the Geologic Map of the Charlotte 1° x 2° Quadrangle, North Carolina and South Carolina (Goldsmith, Milton, & Horton, Jr., 1988), describing four map units underlying the MSS site as biotite gneiss (bgf), quartz sericite schist (Battleground Formation - Zbs), the High Shoals Granite (IPhs), and a quartz -alkali feldspar rich granite (alaskitic (light-colored) granite (DOga). Descriptions of lithology of rock core collected during the CSA and previous site investigations include: 47 Biotite gneiss with some schistose texture, medium- to coarse -grained Page 6-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Meta -granite Granite �7 Meta -quartz diorite �7 Biotite gneiss, fine- to medium -grained �7 Biotite schist '67 Quartz-sericite schist The High Shoals Granite is primarily logged as biotite gneiss, medium- to coarse - grained, schistose in places, and generally with well -developed banding in places consistent with the description of the unit by (Goldsmith, Milton, & Horton, Jr., 1988) as being a "very light gray, coarse -grained, porphyritic, gneissoid biotite granite or granitic gneiss." Meta -granite and granite were also logged in the area underlain by the High Shoals Granite. Alaskitic granite (DOga; as mapped by Goldsmith et al. 1988) does not underlie the MSS site. The primary rock identified from rock core in the Goldsmith et al. (1988) mapped unit is meta -quartz diorite. The mapped alaskitic granite is changed to a meta -quartz diorite unit (mqd; widespread in the Charlotte terrane), and the contacts of the unit have been reinterpreted based on the borehole data. The northern portion of the Battleground Formation (Zbs; Figure 5-2) is underlain by fine- to medium -grained biotite gneiss and biotite schist; not quartz-sericite schist as mapped by Goldsmith et al. (1988). They are similar to the biotite gneiss encountered in the area identified by Goldsmith et al. (1988) as biotite gneiss with a volcanic protolith (bgf; Figure 5-2). Quartz-sericite schist was encountered in previous boreholes at MSS, with one of the boreholes in the Battleground Formation, but encountered in the biotite gneiss unit at other locations on the MSS site. The primary lithology encountered in the CSA boreholes and previous boreholes is shown on the site geologic map (Figure 6-1), and the contacts have been reinterpreted to reflect the lithologies encountered during the CSA and previous investigations, and the implied/interpreted structural relationships. 6.1.3 Structural Geology The Charlotte and Kings Mountain terranes have been subject to multiple deformations due to tectonic stress before and during the intrusion of the meta - Page 6-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra quartz diorite and High Shoals Granite (Horton, 1981; Schaeffer, 1981; Butler & Secor, 1991). The Kings Mountain terrane has undergone polyphase folding; the two earliest folding events were isoclinal to tight, resulting in two subparallel, axial planar foliations (Schaeffer, 1981). Most of the rock encountered in the boreholes exhibits some degree of foliation/schistosity related to these fold events and is the dominant structure with respect to the bedrock underlying MSS. The revised geologic map discussed in the following sections suggests that the contact between the Zbs and bgf units is folded with a north-northeast trending fold axis parallel to the contact of the High Shoals Granite with these units. The meta -quartz diorite (mqd) was probably intruded during the late stages of the second fold event near the peak of regional metamorphism, a relationship that has been documented as existing to the south in the Kings Mountain terrane (Horton, 1981; Schaeffer, 1981). Data from the rock core show a number of joint dip angles that cannot be properly defined as joint sets since there is no orientation information. For the purpose of this discussion, the joints are assessed based on dip angle alone. The most prevalent dip angles are sets that range from 30 degrees to 40 degrees and from 0 degrees to 10 degrees. These two sets are predominant based on the number of joints noted on the boring logs. Less predominant joint sets dipping 15 degrees to 25 degrees and 60 degrees to 65 degrees are also noted on the boring logs. A steeply dipping set ranging from 80 degrees to 90 degrees is not noted as frequently, but since a vertical boring is less likely to intercept sub -vertical joints, it is probable that the 80- to 90-degree dipping set is at least as prevalent as the aforementioned sets. Iron and manganese staining is noted on all of the sets, so it is reasonable to assess that the joint sets are pathways for groundwater flow. The degree of openness of any of the joints is difficult to assess from rock core since the core is often broken at a joint and no longer retains its actual openness. However, some of the logs describe some joints as tight to open. 6.1.4 Soil and Rock Mineralogy and Chemistry Soil mineralogy and chemistry analytical results are shown in Table 6-1 (mineralogy) and Table 6-2 (elemental composition, percentages of oxides). The analytical methods used with solid and aqueous phase samples are presented in Table 6-3 and Table 6-4. Completed laboratory analyses of the mineralogy and chemical composition of TZ materials are presented in Table 6-5 (mineralogy) and Table 6-6 (elemental composition, percentages of oxides). Completed rock Page 6-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra chemistry results are presented in Table 6-7 (elemental composition, percentages of oxides). The petrographic analysis of six rock samples (thin -sections) is presented in Table 6-8. All mineralogy reports are presented in Appendix C. The dominant minerals in the soils are quartz, feldspar (both alkali and plagioclase feldspars), kaolinite, muscovite/illite, and biotite. Three samples (MS- 02, MS-06, and AB-6BR) had tremolite that reportedly accounted for 15.3 percent to 32.2 percent of the sample weight. Other minerals identified include vermiculite, hydroxyapatite, hematite, ilmenite, magnetite, mullite, and amorphous materials (that contain smectites, amorphous, mica, and/or amorphous iron oxide/hydroxide). The major oxides in the soils are SiO2 (51.15% - 68.78), Al2O3 (11.49% - 25.51%), and Fe2O3 (2.67%-10.04%). MnO ranges from 0.02% to 0.11%. The dominant minerals in the TZ are quartz, feldspar (both alkali and plagioclase feldspars), biotite, and amphibolite. The major oxides in the TZ are SiO2 (52.92% - 57.81%), Al2O3 (16.53%-19.15%), and Fe2O3 (6.51%-10.51%). MnO ranges from 0.09 percent to 0.15 percent. The dominant minerals in the six rock samples analyzed are feldspar (both alkali and plagioclase feldspars), quartz, and biotite. Two samples contained approximately 20 percent hornblende, with one of those containing 10 percent chlorite. Other minerals identified include pyroxene (one sample), epidote, muscovite, tremolite (one sample), apatite, allanite, sphene, and pyrite. Trace amounts of zircon were also detected in each sample. The major oxides in the rock samples are SiO2 (50.83% - 62.09%), Al2O3 (10.93% - 20.79%), and Fe2O3 (4.36% - 8.63 %). MnO ranges from 0.06 percent to 0.11 percent in the rock samples. 6.1.5 Geologic Mapping Geologic mapping was conducted in June 2015 to map any available outcrops at the Site and within a 2-mile radius of the Site using a Brunton compass to attempt to characterize rock types, the orientation (strike and dip) of structure such as foliation, joint sets, folds, and shears/shear zones. Only three outcrops were located that presented the opportunity to map rock structure. These measurements do provide evidence of the overall orientation of foliation in the vicinity of the MSS site. The measurements of foliation from these outcrops were: N50°E; 76°SE, N60°E; 50°SE, and N65°E; 36°SE. Page 6-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra As described above in Section 6.1.2, borehole lithologies were overlaid on the Geologic Map of the Charlotte 1° x 2° Quadrangle, North Carolina and South Carolina (Goldsmith, Milton, & Horton, Jr., 1988). Field mapping and the use of the borehole data resulted in changes in geologic units (alaskitic granite [DOga] changed to meta -quartz diorite [mqd]), locations of contacts, and contact relationships. The contact between the Zbs and bgf unit was moved south. The contact of the meta -quartz diorite into the surrounding rocks has been interpreted based on the borehole data. Figure 6-1 presents the revised geologic map of the MSS site. 6.1.6 Effects of Geologic Structure on Groundwater Flow The most important potential effects of structural geology on groundwater flow in the western and northern portion of the Site is the well -developed foliation in the IPhs, Zbs, and bgf units and the likely interconnected joint sets discussed in Section 6.1.3. The meta -quartz diorite is less foliated than the other units, and it is not known whether mafic dikes are present within the unit at the Site, as has been noted in this unit in other areas of the Charlotte terrane (Gilbert, Brown, & Schaeffer, 1982). The unit is jointed, and those joints will likely have the greatest impact on groundwater flow. Since it is difficult to define joint sets based on dip angle alone, it is also difficult to define which joints are most relevant with respect to groundwater flow. 6.2 Site Hydrogeology According to LeGrand (2004), the soil/saprolite regolith and the underlying fractured bedrock represent a composite water -table aquifer system. The regolith provides the majority of water storage in the Piedmont province, with porosities that range from 35 percent to 55 percent (Daniel & Dahlen, 2002). Calculated porosities specific to the Site (32.1% to 52.5%) are consistent with this range. Two major factors that influence the behavior of groundwater in the vicinity of the Site include the thickness (or occurrence) of saprolite/regolith and the hydraulic properties of underlying bedrock. Saprolite thickness varies across the Site but is generally thickest in upgradient areas (approximately 80 feet at GWA-14D and MW-11D) and thins in downgradient areas near Lake Norman (no saprolite logged at GWA-15S). Based on the site investigation, the groundwater system in natural materials (soil, soil/saprolite, and bedrock) at the MSS site is consistent with the regolith-fractured rock system and is an unconfined, connected aquifer system as discussed in Section 5.2. Regolith is underlain by a TZ of weathered rock that transitions to competent bedrock. The groundwater system at the MSS site is divided into three flow layers referred to in Page 6-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra this report as the shallow, deep (TZ), and bedrock layers, so as to distinguish unique characteristics of the connected aquifer system. 6.2.1 Hydrostratigraphic Layer Development The hydrostratigraphic classification system of Schaeffer (Schaeffer, 2014a) used to evaluate natural system hydrostratigraphic layer properties. The classification system is based on Standard Penetration Testing values and the rock core recovery (REC) and rock quality designation (RQD) collected during the drilling and logging of the boreholes (borehole/well logs included in Appendix F). The Schaeffer classification system uses the terms M1 and M2 to classify saprolite material with the M2 designation indicating greater competency. A TZ of weathered, fractured rock is delineated between overlying saprolite and underlying bedrock based on REC and RQD. The bedrock zone is classified as having REC of greater than 85 percent and RQD of greater than 50 percent. For discussion purposes, hydrostratigraphic units are recognized in the text and supporting documents as follows: '67 Shallow Unit — Alluvium/Saprolite (S wells) '67 Deep Unit — Saprolite and weathered rock (D wells) '67 Bedrock Unit — Sound rock, relatively unfractured (BR wells) The shallow zone generally corresponds to the M1 unit, and the deep zone incorporates the M2 and weathered, fractured rock layers. Bedrock is identified per the REC and RQD criteria. The designations ash, fill, saprolite, TZ, and bedrock are used on the generalized geologic cross -sections presented in Figures 6-2 to 6-4 showing site geology and groundwater flow directions. 6.2.2 Hydrostratigraphic Layer Properties Ash Pore Water Assessment results indicate the ash thickness within the basin ranges from a few feet in thickness up to 85 feet. The majority of ash located within the ash basin is saturated with saturated ash thickness ranging from less than one foot up to 52 feet. Ash contained within dry ash landfill (Phase II) is observed to a depth of 111 feet bgs with depth to water measured at approximately 112 feet bgs in AL-2S (shallow flow system). Ash in the PV structural fill was encountered at depths up to 71 feet bgs with depth to water measured at approximately 61 feet bgs in AB- 20S (shallow flow system). Page 6-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Shallow Flow Layer The shallow flow layer consists of regolith (soil/saprolite) material. Thickness of regolith is directly related to topography, type of parent rock, and geologic history. Topographic highs tend to exhibit thinner soil-saprolite zones, while topographic lows typically contain thicker soil-saprolite zones. Wells within the shallow flow layer that are installed within shallow wells contain an "S" designation. Deep Flow Layer The deep flow layer (TZ) consists of a relatively transmissive zone of partially weathered bedrock. Observations of core recovered from this zone included rock fragments, unconsolidated material, and highly oxidized bedrock material. Deep flow layer wells are labeled with a "D" designation. Bedrock Flow Layer The fractured bedrock unit occurs within competent bedrock. Bedrock in the immediate vicinity of the ash basin consists of several varieties of gneiss, granite, and schist as described in Section 6.1.2. The majority of water producing fracture zones was found within 50 feet of the top of competent rock. Water -bearing fractures encountered are only mildly productive (providing water to wells). Bedrock wells are labeled with a 'BR" designation. 6.3 Groundwater Flow Direction Based on the CSA site investigation, groundwater within the shallow, deep, and bedrock flow layers at the Site generally flows from the northwest to the southeast toward Lake Norman. Voluntary, compliance, and groundwater assessment monitoring wells were gauged for depth to water within a 24-hour period during comprehensive groundwater elevation measurement events on March 10, 2017, and September 22, 2017, to provide seasonal water level elevation data pertaining to the Site. March is considered the wet season, and September is considered the dry season. Depth -to -water measurements were subtracted from surveyed top -of -well -casing elevations to produce groundwater elevations in shallow, deep, and bedrock monitoring wells (Table 6-9). Groundwater flow direction was estimated by contouring those groundwater elevations. The shallow, deep, and bedrock water -level maps for March 2017 and September 2017 are included as Figures 6-5 through 6-10. Groundwater flow at the MSS site follows the local slope aquifer system (Figure 5-4), as described by LeGrand (2004). In general, groundwater within the shallow wells (S), Page 6-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra wells in the TZ (D), and wells in fractured bedrock (BR) flows southeasterly from the ash basin toward Lake Norman. Minor, localized groundwater divides exist adjacent to topographic divides associated with the natural ridges separating historic draws, which are now coves of Lake Norman. 6.4 Hydraulic Gradient Horizontal hydraulic gradients were derived for September 2017 water -level measurements in the shallow, TZ, and fractured bedrock wells by calculating the difference in hydraulic head over the length of the flow path between two wells with similar well construction (e.g., wells within the same water -bearing unit). The following equation was used to calculate horizontal hydraulic gradient: i=dh/dl Where i is the hydraulic gradient; dh is the difference between two hydraulic heads (measured in feet); and dl is the flow path length between the two wells (measured in feet). Applying this equation to wells installed during the CSA activities yields the following average horizontal hydraulic gradients (measured in feet/foot): '67 Shallow wells 0.017 feet/foot '67 Deep wells: 0.014 feet/foot '67 Bedrock wells: 0.010 feet/foot Generally horizontal gradients in the ash pore water within the basin range from 0.003 feet/foot to 0.012 feet/foot. Horizontal gradients in the shallow flow zone beneath and downgradient of the ash basin range from 0.007 feet/foot to 0.016 feet/foot. Horizontal gradients in deeper flow zones range from 0.005 feet/foot to 0.022 feet/foot. A summary of horizontal hydraulic gradient calculations is presented in Table 6-10. Vertical hydraulic gradients were calculated by taking the difference in groundwater elevation in a deep and shallow well pair over the difference in total well depth of the deep and shallow well pair. A positive output indicates downward flow, and a negative output indicates upward flow. Vertical gradient calculations for 42 shallow and deep well pair locations and 17 deep to bedrock well pair locations were used to calculate vertical hydraulic gradient across the Site. Applying that calculation to wells installed during the CSA activities yields the following average vertical hydraulic gradients (measured in feet/foot): Page 6-9 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Shallow to Deep wells:-0.0022 feet/foot Deep to Bedrock wells: 0.012 feet/foot Based on review of the results, vertical gradients were mixed across the Site, with slightly more than half of locations showing downward gradient values (approximately 52 percent). The greatest downward vertical gradients (positive values) are observed upgradient and beyond the ash basin footprint at the GWA-12 location (GWA-12D/BR; 0.45 foot/foot) and the BG-3 location (BG-3D/BR; 0.101 foot/foot). Upward vertical gradients (negative values) are observed on the eastern boundary of the ash basin near Lake Norman and its tributaries. Additionally, upward vertical gradients are predominant in the center of the historic channel of Holdsclaw Creek beneath the ash basin. Specifically, upward vertical gradients are observed at AB-15 location (AB-15D/BR; -0.081 foot/foot), AB-12 location (AB-12D/BR; -0.073 foot/foot), and the AB-9 location (AB-9D/BR; -0.028 foot/foot). Analytical results from the bedrock wells at these locations verify that constituent migration to the bedrock flow system is limited to specific areas identified in Section 11.1.1. Vertical gradient calculations are summarized in Table 6-11 and shown on Figure 6-11. 6.5 Facility Soil Data Soil samples were collected during CSA monitoring well installations. Comparison of soil analytical results with background is discussed below based on the area of the Site. 6.5.1 Soil Beneath Ash Basin The contact between the ash and underlying soils in the ash basin borings was visually distinct. There was no visible evidence of substantial migration of ash into underlying soils or mixing of ash with those soils. Of the groundwater COIs, values for arsenic, barium, chromium, iron, molybdenum, nickel, selenium, strontium, sulfate, and vanadium exceeded either the PBTV or PSRG POG, whichever is higher, in at least one soil sample (Appendix B, Table 4). Of those constituents, only strontium exceeds the PBTV on a consistent basis in soil beneath the ash basin. A POG has not been established for strontium. One vanadium concentration exceeded the PBTV by 5 mg/kg in AB-15D. SPLP was used to determine the ability of simulated rainwater to leach site - specific constituents out of the soil to groundwater. The 2L/IMAC standards are used for reference only of SPLP data. SPLP test results do not represent Page 6-10 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra groundwater; therefore, comparison with 2L/IMAC standards is for source contribution information rather than compliance monitoring purposes. For comparative purposes only, SPLP results are compared to 2L/IMAC standards herein. The SPLP analyses revealed that barium, boron, chromium, cobalt, iron, lead, manganese, thallium, and vanadium leach from soils beneath the ash basin at concentrations that exceed the 2L/IMAC standards. However, cobalt, iron, manganese, and vanadium appear to be ubiquitous across MSS in soils regardless of location (e.g., beneath ash, upgradient, and downgradient) and tend to leach in concentrations that are often greater than the 2L/IMAC in the leachate even for soils not beneath the ash basin. Additionally, the only exceedances of barium, lead, and thallium concentrations were in one location beneath the ash basin (AB-15D). Therefore, boron and chromium appear to be the only constituents detected in soil leachate at concentrations greater than 2L that may be attributable to the source area. 6.5.2 Soil Beyond Waste Boundary and Within Compliance Boundary Detected concentrations of arsenic, barium, chromium, molybdenum, selenium, strontium, sulfate, and vanadium exceeded either the PBTV or PSRG POG, whichever is higher, in at least one soil sample (Appendix B, Table 4). Of those constituents, only selenium and strontium concentrations exceeded a PBTV or PSRG POG value at more than one sampling location. Aside from strontium, detections of COIs in soil were sporadic and inconsistent and did not indicate a source of soil impact beyond the ash basin waste boundary. SPLP results for soils beyond the waste boundary indicate that iron, manganese, and vanadium readily leach from natural soils. Leaching of chromium, cobalt, lead and thallium from natural soils is inconsistently observed. 6.5.3 Comparison of PWR and Bedrock Results to Background Four samples were collected from the TZ or bedrock and analyzed as soil samples. All four samples [AB-15BR (104.5), AL-2BR (204), AB-09BR (79), AB- 05BR (105)] were collected from intensely fractured and stained/mineralized biotite gneiss. One concentration of arsenic (5.9 mg/kg) at AB-9BR (79) slightly exceeded the PSRG POG of 5.8 mg/kg. No other TZ or bedrock samples exceeded both PSRG POG and PBTVs for arsenic. Strontium concentrations exceeded the soil PBTV in all four samples. A PSRG POG has not been established for strontium. Page 6-11 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 6.6 Hydraulic Conductivity Hydraulic conductivity (slug) tests were completed in the monitoring wells for the original CSA (HDR, September 8, 2015a). Slug tests were performed to meet the requirements of the NCDENR Memorandum titled, Performance and Analysis of Aquifer Slug Tests and Pumping Tests Policy (May 31, 2007). Water level change during the slug tests was recorded by a data logger. The slug test was performed for no less than 10 minutes, or until such time as the water level in the test well recovered 95 percent of its original pretest level, whichever occurred first. Slug tests were terminated after 60 minutes even if the 95 percent pretest level was not achieved. Slug test field data was analyzed using the Aqtesolv (or similar) software and the Bouwer and Rice method. Additionally, in -situ hydraulic conductivities were calculated using slug test results reported in CSA Supplement 2 (HDR, August 4, 2016b). Those previously reported horizontal and vertical groundwater conductivity results are presented in Table 6-12 and 6-13. Historic slug test results are included in Table 6-14. Additional evaluation was conducted to determine groundwater velocity by grouping hydraulic conductivity (slug) test data into the respective hydrostratigraphic units and calculating the geometric mean, maximum, and minimum. Hydrostratigraphic layers are defined in Section 11.1. Hydraulic conductivity values for wells screened in saprolite have a geometric mean of 5.50 x 10-4 centimeters per second (cm/sec). Hydraulic conductivity values for wells screened in the TZ have a geometric mean of 4.23 x 10-4 cm/sec. The range of horizontal conductivities in the TZ spanned from 2.73 x 10-7 cm/sec to 1.22 x 10-2 cm/sec. These measurements reflect the variable nature of the TZ, where hydrologic properties are heavily influenced by the formation of clays and other byproducts of weathering. Similarly, bedrock heterogeneities are apparent across the Site with a range of hydraulic conductivity results for bedrock wells from 2.48 x 10-5 cm/sec to 1.35 x 10-z cm/sec. The geometric mean of hydraulic conductivities measured in bedrock wells is 2.15 x 10-4 cm/sec. The hydraulic conductivity measurements in bedrock wells are generalized representations of the localized bedrock fractures in specific areas of a well cluster. In -situ horizontal hydraulic conductivity values for each hydrostratigraphic unit established are in Table 6-15. 6.7 Groundwater Velocity To calculate the velocity with which water moves through a porous media, the specific discharge, or Darcy flux, is divided by the effective porosity, ne. The result is the average linear velocity or seepage velocity of groundwater between two points. Groundwater flow velocities for the shallow and deep flow zones were calculated using Page 6-12 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Darcy's Law, which describes the flow rate or flux of fluid through a porous media by the following formula: vs = Ki/ne v., = seepage velocity, K = horizontal hydraulic conductivity, i = the horizontal hydraulic gradient; and ne = effective porosity. Seepage velocities for groundwater were calculated using horizontal hydraulic gradients established by grouping hydraulic conductivity (slug) test data into the respective hydrostratigraphic units and calculating the geometric mean, maximum, and minimum. Horizontal hydraulic conductivity values for each hydrostratigraphic unit are presented in Tables 6-14 and 6-15, and effective porosity values are presented in Tables 6-16 and 6-17. Hydrogeologic porosity reports are provided in Appendix C. Hydrostratigraphic layers are defined in Section 11.1. Average groundwater seepage velocity results are summarized in Table 6-10. At MSS, groundwater movement in the bedrock flow zone is primarily due to secondary porosity represented by fractures in the bedrock. Primary (matrix) porosity is negligible; therefore, it is not technically appropriate to calculate groundwater velocity using effective porosity values and the method presented above. Bedrock fractures encountered at MSS tend to be isolated with low interconnectivity. Further, hydraulic conductivity values measure the fractures immediately adjacent to a well screen, not across the distance between two bedrock wells. Groundwater flow in bedrock fractures is anisotropic and difficult to predict, and velocities change as groundwater moves between factures of varying orientations, gradients, pressure, and size. For those reasons, bedrock groundwater velocities calculated using the seepage velocity equation are not representative of actual site conditions and were not calculated. For additional information on the movement of groundwater around and downgradient of the Ash Basin over time, refer to discussion concerning groundwater flow and transport modeling (Section 13.0). 6.8 Contaminant Velocity Migration, retardation, and attenuation of COIs in the subsurface is a factor of both physical and chemical properties of the media in which the groundwater passes. Contaminant velocity depends on factors such as the rate of groundwater flow, the effective porosity of the aquifer material, and the soil -water partitioning coefficient, or Kd term. Soil samples were collected and analyzed for grain size, total porosity, soil Page 6-13 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra sorption (Kd), and anions/cations to provide data necessary for completion of the three- dimensional groundwater model discussed in Section 13. Ash sluiced to, and accumulated within, the basin is the primary source of impacts to groundwater. Gradients measured within the ash basin support the interpretation that ash pore water mixes with shallow/surficial groundwater and migrates downward into the TZ and bedrock flow zones. Continued vertical migration of impacted groundwater is also evidenced by detected constituent concentrations. Boron is relatively mobile in groundwater and is associated with low Kd values. This is primarily because boron is mostly inert, has limited potential for sorption, and lacks an affinity to form complexes with other ions. In general, the low Kd measured for boron allows the constituent to move at a similar velocity to groundwater. The higher Kd values measured for the remaining metals, like thallium and cobalt, agree with the limited migration of these constituents. Constituents like cobalt and thallium have much higher Kd values, and they will move at a much slower velocity than groundwater as it sorbs onto surrounding soil. Groundwater migrates under diffuse flow conditions in the shallow and deep aquifer in the direction of the prevailing gradient. It should be noted that the fractured bedrock flow system is highly heterogeneous in nature and high permeability zones with a geomean in -situ horizontal conductivity of 0.000215 cm/sec observed, but these hydraulic conductivity measurements measure the fractures immediately adjacent to a well screen, not across the distance between two bedrock wells, and cannot be applied across the entire Site. Geochemical mechanisms controlling the migration of constituents are discussed further in Section 13. Groundwater modeling to be performed for the updated CAP will include a discussion of contaminant velocities for the modeled constituents. 6.9 Slug Test and Aquifer Test Results As previously discussed, hydraulic conductivity values for the various hydrogeologic zones in which the wells are screened varied Site -wide as determined by the slug test method (Table 6-15). Slug test field and analytical methods and results are presented in Appendix C. In -situ horizontal (open hole) and vertical (flush bottom) permeability tests, either falling or constant head as appropriate for field conditions, were performed in each of the hydrostratigraphic units. The flush bottom test involves advancing the borehole through the overburden with a casing advancer until the test interval is reached. The cutting tool is removed from the casing and the casing is filled with water to the top and Page 6-14 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra the water level drop in the casing is measured over 60 minutes. In the open hole test, after the top of the test interval is reached, the cutting tool but not the casing, is advanced an additional number of feet (5 feet in the majority of tests) and water level drop in the casing is measured over 60 minutes. The constant head test is similar except the water level is kept at a constant level in the casing and the water flow -in is measured over 60 minutes. The constant head test was used only when the water level in the borehole was dropping too quickly back to the static water level such that the time interval was insufficient to calculate the hydraulic conductivity. The results from the field permeability testing are summarized in Table 6-18 and the worksheets are provided in Appendix C. Packer tests (shut-in and pressure tests) were conducted during the intial CSA field effort. The shut-in test is performed by isolating the zone between the packers (in effect, a piezometer) and measuring the resulting water level over time until the water level is stable. The shut-in test provides an estimate of the vertical gradient during the test interval. The pressure test involves forcing water under pressure into rock through the walls of the borehole, providing a means of determining the apparent horizontal hydraulic conductivity of the bedrock. Each interval is tested at three pressures with three steps of 20 minutes up and two steps of 5 minutes back down. The pressure test results are summarized in Table 6-18, and the shut-in and packer tests worksheets are provided in Appendix C. Shelby tube samples were collected at 11 locations and were used for vertical hydraulic conductivity tests, each conducted on media from four distinct zones: soil/saprolite, saprolite/weathered rock, ash, and fill (Table 6-19). The vertical conductivities were calculated to be, on average, relatively similar to or slightly less than the horizontal results. These data indicate relatively similar vertical and horizontal components of flow in general across the Site. 6.10 Fracture Trace Study Results Fracture trace analysis is a remote sensing technique used to identify lineaments on topographic maps and aerial photography that may correlate to locations of bedrock fractures exposed at the earth's surface. Although fracture trace analysis is a useful tool to identify potential fracture locations, and hence potential preferential pathways for infiltration and flow of groundwater near a site, results are not definitive. Lineaments identified as part of fracture trace analysis may or may not correspond to actual locations of fractures exposed at the surface, and if fractures are present, it cannot be determined from fracture trace analysis whether these are open or healed. Healed Page 6-15 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra fractures intruded by diabase are common in the vicinity of the Site based on observations from the advancement of boreholes. Strongly linear features at the earth's surface are commonly formed by weathering along steeply dipping to vertical fractures in bedrock. Morphological features such as narrow, sharp -crested ridges, narrow linear valleys, linear escarpments, and linear segments of streams otherwise characterized by dendritic patterns are examples. Linear variations in vegetative cover are also sometimes indicative of the presence of exposed fractures, though in many cases these result from unrelated human activity or other geological considerations (e.g., change in lithology). Straight (as opposed to curvilinear) features are commonly associated with the presence of steeply dipping fractures. Curvilinear features in some cases are associated with exposed moderately dipping fractures, but these can also be a result of preferential weathering along lithologic contacts, or along foliation planes or other geologic structure. As part of this study, only strongly linear features were considered, as they are far more commonly indicative of steeply dipping or vertical fractures. The effectiveness of fracture trace analysis in the eastern United States, including in the Piedmont, is commonly hampered by the presence of dense vegetative cover, and oftentimes extensive land -surface modification owing to present and past human activity. Aerial photography interpretation is most affected, as identification of small- scale features is rendered difficult or impossible in developed areas. 6.10.1 Methods Prior to aerial photography and topographic map interpretation, available geologic maps for the area were consulted to identify lithologies and geologic structure in the area that can control fracture occurrence and orientation. Topographic map interpretation was performed over an area of approximately 60 square miles, and aerial photography interpretation was performed over an area of approximately 20 square miles. Topographic map interpretation involved examination of the Lake Norman North, N.C., Troutman, N.C., Denver, N.C., and Catawba, N.C. 1:24,000-scale USGS 7.5-minute topographic quadrangles. Digital copies of the quadrangles were obtained and viewed on a monitor at up to 7x magnification. Lineaments identified were plotted directly on the digital images. Photography provided for review included 1"=600' scale, 9-inch by 9-inch black - and white (grayscale) contact prints dated April 17, 2014. Stereo coverage was Page 6-16 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra complete across the review area. The photography was examined using a Lietz Sokkia MS-27 mirror stereoscope with magnifying binocular eyepiece. Lineaments identified on the photographs were marked on hard copies of scanned images (600dpi resolution), and subsequently compiled onto a photomosaic base. 6.10.2 Results Lineaments identified from topographic maps are shown and lineament trends indicated by a rose diagram are included on Figure 6-12. A total of 55 topographic lineaments were identified across the study area, mainly north, east, and southeast of the Site in areas underlain by fine-grained biotite gneiss. Lineaments are less well -developed in areas underlain by the High Shoals Granite in the northwest part of the study area, and by biotite meta-granodiorite in the southwest corner of the study area. The lineaments trend primarily toward the northeast and north-northeast, subparallel to the strike of foliation and fold axes in rocks near the Site (Goldsmith, Milton, & Horton, Jr., 1988). The north- northeast trending lineaments have a strong preferred orientation at approximately NYE due to a well -developed axial planar cleavage/foliation in the Kings Mountain terrane rocks near the Site. Northwest trending lineaments are also relatively common. These likely reflect trends of regional joint sets that commonly overprint earlier structures in the Piedmont, and in some areas, are associated with Triassic diabase dike intrusion. Lineaments identified from aerial photography are shown and lineament trends indicated by a rose diagram are included on Figure 6-13. A total of 36 lineaments were identified, primarily in the form of linear morphological lows (linear stream valleys, ravines, and gullies), linear morphological features in upland areas, and light colored linear outcrops that appear to be quartzo-feldspathic material intruded into the igneous units (or possibly the quartz-sericite schist of the Battleground Formation [Zbs]). Lineaments were identified mainly south of the Site, in areas underlain by the biotite gneiss unit. Few lineaments were discernable in areas north of the Site, in part due to the presence of dense vegetative cover. Lineament trends identified from examination of aerial photography are generally consistent with those identified from topographic map interpretation. Prevalent orientations are toward the north-northeast and northeast in conformance with bedrock structure as discussed for the topographic map lineaments. The well-defined NYE orientation for north-northeast trending Page 6-17 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra lineaments identified on topographic maps was also apparent for smaller -scale features identified in aerial photography. Page 6-18 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 7.0 SOIL SAMPLING RESULTS Soil samples were collected and tested in accordance with GAP (Section 7.1.1), and the analytical methods for testing soils are summarized in Table 6-3. Soil test data are included in Appendix C. Soil borings were conducted in upgradient and downgradient areas of the ash basin to collect soil samples from the unsaturated zone and the zone of saturation for those areas (Figure 2-4). Mineralogical analysis of soil samples indicates clay minerals (kaolinite, smectite, and illite) and amorphous iron oxide/hydroxide comprise the bulk portion of Site soils, along with quartz, feldspars, and amphiboles (Table 6-1). 7.1 Background Soil Data Three upgradient boring locations (BG-1S/D, BG-2BR, and BG-3D) were originally installed in 2015 for use as background wells and soil borings. Additional samples collected in 2015 from borings for installation of monitoring wells GWAABR, GWA-4D, GWA-51), and MW-14BR were also initially included in the background data set. Four additional soil samples were collected in March 2017 from borings advanced to install additional monitoring wells GWA-2DA and GWA-14S. The GWA-2DA boring is located adjacent to existing upgradient well GWA-2D. GWA-14S is located off -Site, upgradient, and to the northwest of the ash basin system. A background soil dataset based on this data was provided to NCDEQ on May 26, 2017, for consideration of background soil concentrations. Additionally, the revised Statistical Methods for Developing Reference Background Concentrations for Groundwater and Soil at Coal Ash Facilities (statistical methods document) (HDR and SynTerra, 2017) was provided to NCDEQ as a basis for determination. On July 7, 2017, NCDEQ provided a response letter for each Duke Energy coal ash facility that identified soil and groundwater data appropriate for inclusion in the statistical analysis used to determine PBTVs for both media. PBTVs were calculated using the approved data set identified in that July 7, 2017, NCDEQ letter to Duke Energy. In a letter dated October 11, 2017, NCDEQ provided concurrence/non-concurrence with provided PBTVs. PBTVs were accepted for all constituents except: antimony, beryllium, cadmium, magnesium, potassium, selenium, and thallium. Those soil PBTVs are presented in Table 7-1. The soil PBTVs are subject to change based on further refinement of, and addition to, the background soil data set as described on the following page. NCDEQ requested that Duke Energy collect a minimum of 10 valid background samples, rather than the previously planned eight samples) prior to the determination Page 7-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra of PBTVs for each constituent. Soil samples that meet the following criteria are considered valid for use in statistical determinations of PBTVs: 07 The sample was collected from a location that is not impacted by coal combustion residuals or coal -associated materials. 07 The sample was collected from a location that is not impacted by other potential anthropogenic sources of constituents. 07 The sample was collected from the unsaturated zone, greater than 1 foot above the seasonal high water table elevation. NCDEQ determined samples collected from several locations/depths did not meet NCDEQ Inactive Hazardous Site Branch (IHSB) Guidance requirements; therefore, they were deemed not appropriate for use in determining PBTVs. The background soil dataset also included laboratory reporting limits for antimony and thallium greater than the NCDEQ IHSB PSRG Protection of Groundwater values (dated October 2016). NCDEQ requested the values for antimony, selenium, and thallium be reported as less than the PSRG POG values. To address those requirements, 14 additional soil samples were collected from three background locations on August 16, 2017. These sample results were not included in the data set for approved PBTVs provided by NCDEQ on October 11, 2017. Boring logs associated with the additional soil samples are included in Appendix F. The updated background dataset was screened for outliers prior to statistical determinations. The updated background data set and soil PBTVs are provided in Appendix H. 7.2 Secondary Sources For soil samples beneath the ash basin, arsenic, barium, chromium, iron, molybdenum, nickel, selenium, strontium, sulfate, and vanadium concentrations in at least one soil sample exceeded both the calculated soil PBTV and the PSRG POG value (Table 7-2). Of these constituents, arsenic and nickel are not found greater than 2L in downgradient groundwater. No other COIs were detected in soil beneath the ash basin at concentrations greater than both a PBTV and PSRG POG value. These exceedances are subject to change based on further refinement of, and addition to, the soil PBTVs as additional data may be added to the background data set. The updated values are subject to NCDEQ approval prior to implementation. Analysis of soil analytical data presented in Appendix B, Table 4 and Table 7-2 shows that only in a limited extent have COIs from the source mobilized and sorbed onto soils Page 7-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra beneath the ash basin. Arsenic, chromium, nickel, selenium, and strontium were detected at concentrations greater than the respective PBTV or PSRG POG in a consistent manner beneath or downgradient of the ash basin. Notably, only one location AB-15D (48-49) resulted in concentrations of iron and vanadium at levels greater than the PSRG POG and PBTV. Sampling results indicated greater concentrations of arsenic, chromium, nickel, selenium, and strontium in soil beneath the ash basin than in other areas where sampling occurred. Figure 7-1 shows soil exceedances in relation to the ash basin. Saturation and other factors may also affect constituent occurrence in the samples. Page 7-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 8.0 SEDIMENT RESULTS Sediment samples were collected from two locations beyond the perimeter of the ash basin (Figure 2-4) during the 2015 CSA field effort. Sediment sampling procedures and variances are provided in Appendix G, and analytical results are presented in Appendix B, Table 5. 8.1 Sediment/Surface Soil Associated with AOWs One sediment sample was collected coincidentally with Areas of Wetness (AOW) S-1 (previously identified as surface water sample SW-06) to represent a downgradient location east of the ash basin. The second sample was collected coincidentally with AOW S-2. The "sediment" that was collected was actually surface soil over which water at the AOW was flowing or seeping. Sediment samples were collected on July 18 and July 20, 2015. The sediment sample results were compared with North Carolina PSRGs for POG (Appendix B, Table 5). Sediment sample locations are shown on Figure 2-4. A description of sample locations S-1 and S-2 and the results of sediment analysis are provided below: S-1(SW-06): A steady flow emerges from several springs within the channel. The flow appears to follow natural topography that trends away from a ridge separating the ash basin from the drainage area. Sediment was collected from the channel. Chromium and selenium exceeded both the PBTVs for soil and the POG values. Cobalt, iron, manganese, and vanadium concentrations exceeded the POG but were less than the respective PBTVs. 41' SS 2: This AOW formed within a depression of a displaced tree root ball. Orange aerobic bacteria/floc/iron-oxidation was observed on substrate. Flow from the culvert area is directed toward S-2 (sign) location. Sediment from the channel indicated selenium exceeded both the PBTV for soil and the POG value. Chromium, cobalt, iron, manganese, and vanadium concentrations exceeded the POG but were less than the respective PBTVs. Page 8-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 9.0 SURFACE WATER RESULTS The purpose of surface water characterization is to evaluate whether storage of ash has resulted in impacts to surface waters in the vicinity of the ash basin. Surface water parameters and laboratory methods used for analysis are presented in Table 6-4. Surface water sample results for total and dissolved fractions of constituents are presented in the Comprehensive Table (Appendix B, Table 2). Surface water sample locations are shown on Figure 2-4. MSS is located adjacent to Lake Norman, which is a part of the Catawba River system. Water samples discussed within the following sections include four distinct types: 1) ash basin wastewater and wastewater conveyance (effluent channels), 2) AOWs, 3) industrial storm water, and 4) named surface waters. For this CSA, it is pertinent that a comparison with NCDENR Title 15A, Subchapter 02B. Surface Water and Wetland Standards (2B) standards includes only sample results from named surface waters. AOWs, wastewater and wastewater conveyances (effluent channels), and industrial storm water are evaluated and regulated in accordance with the NPDES Program administered by NCDEQ DWR. That evaluation/regulation process, conducted parallel to the CSA, is ongoing and subject to change. Representative, previously identified AOW locations were sampled as part of the 2015 CSA. Samples associated with AOWs were collected for water -quality analysis from the following locations: y S-1(SW-06): This AOW is a natural channel approximately 2 feet wide and 4 inches deep with slow flow and pockets of standing water. Orange aerobic bacteria/floc/iron-oxidation sheen is visible on the substrate; the source of the channel was observed at two head cuts. '610 SS2: This AOW formed within a depression of a displaced tree root ball. Orange aerobic bacteria/floc/iron-oxidation was observed on the substrate. Flow from the culvert area is toward the S-02 (sign) location. Other water samples used for the 2015 CSA characterization are as follows: y As reported in the 2015 CSA, NCDENR collected two surface -water samples (MSW001 and MSW002) from the downgradient side of the ash basin dam. MSWO01 was collected near Outfall SWO01 and MSWO02 at Outfall 002. Page 9-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 0 Free-standing water within the ash basin was collected from locations SW-01 through SW-05. Sampling at these locations continues on a quarterly basis. y S-1 (formerly SW-06) represents channelized flow from the area to the east of the ash basin. This location was added as a seep to the NPDES permit (effective October 2016), and has been sampled on a monthly basis in accordance with the permit. -0 As reported in the December 7, 2015, Corrective Action Plan Part 1 (CAP Part 1), SW-07 and SW-08 represent channelized flow from upgradient areas north and west of the ash basin. Sampling at these locations continues on a quarterly basis. SW-09 through SW-12 are recently added, at DEQ request, surface water sample locations within Lake Norman east of the ash basin. Since the 2015 CSA and CAP evaluations, AOW locations S-1 and S-2 have been included in NPDES permit NC0004987, with an effective date of October 1, 2016. As described above, discussion of sample results from the former AOW locations will not be included in this CSA due to the approved permit outfall status. 9.1 Comparison of Exceedances to 2B Standards Surface water data represents a one-time, single sample; therefore, compliance with either the acute or chronic 2B standard may not be determined based on 15A NCAC 02B .0211 (11)(e). Sample results from representative background surface water and Lake Norman sample locations are compared with 2B (Class B WS-IV) values for comparative purposes only. A review of surface water sample data indicates: y Throughout two years of monitoring, dissolved oxygen has been measured less than the 2B standard one or more times at SW-08, SW-09, SW-10, SW-11, and SW- 12. Generally measurement less than the allowable limit of the 2B standard appear to be one time, isolated events with the exception of SW-11 where four of eight measurements are below the 2B. 01 A single detection of copper (D) at a concentration greater than the 2B standard has been measured at each of these locations: SW-07, SW-10, and SW-11. 0 A single detection of TDS at a concentration greater than the 2B standard has been measured at SW-08 and SW-10. y SW-10 is located adjacent to Outfall 001. Intermittent detection of the following constituents at concentrations greater than the 2B standard have been measured at this location — dissolved oxygen, chloride, TDS, arsenic, selenium, cadmium Page 9-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra (D), copper (D), and lead (D). Only dissolved oxygen has been measured outside the 2B standard range since September, 2016. 9.2 Discussion of Results for Constituents without Established 213 Standard A 2B value has not been established for a number of constituents. A summary of results for select constituents without associated 2B values follows. -0 Antimony is not detected at concentrations greater than the reporting limit of 0.5 µg/L in any surface water sample. Estimated values, presented for samples collected from Lake Norman, range from 0.11 µg/L to 0.75 µg/L. y Boron is not detected at upgradient surface water locations. However, boron is detected in most surface water samples collected from Lake Norman. Recent results from September 2017 indicate a range of concentrations from 56 µg/L at SW-11 to 2,560 µg/L at SW-10. Cobalt is detected in both upgradient and Lake Norman surface water samples. Concentrations range from 0.21 µg/L to 3.5 µg/L at upgradient locations and from non -detect to 3.5 µg/L at Lake Norman locations. Iron is generally detected in higher concentrations at upgradient locations with concentrations ranging from 353 µg/L to 4100 µg/L. Iron concentrations at Lake Norman surface water sample locations range from 103 µg/L to 541 µg/L. 0 Manganese is detected in most samples. Concentrations at upgradient locations range from 13.1 µg/L to 270 µg/L, and at Lake Norman locations, concentrations range from 37.6 µg/L to 828 µg/L. -61 Molybdenum generally occurs at levels below the reporting limit of 0.5 µg/L. Molybdenum has been detected at concentrations up to 8.1 µg/L at SW-10. y Strontium concentrations are generally consistent at all surface water sampling locations, with a maximum value of 629 µg/L reported at SW-10. -61, Thallium is generally reported at estimated levels below the reporting limit of 0.1 µg/L, with a maximum value of 0.12 µg/L reported at upgradient location SW-08. 01 Vanadium is generally detected in higher concentrations at upgradient locations, with concentrations ranging from 0.31 µg/L to 3.2 µg/L. Vanadium concentrations at Lake Norman surface water sample locations range from 0.39 µg/L to 2.5 µg/L. Page 9-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 9.3 Discussion of Surface Water Results Surface water and AOWs were sampled in order to evaluate potential influence from the ash management areas. Observation of boron concentrations is useful as an indicator of potential CCR impact. Generally, surface water samples collected from Lake Norman demonstrate compliance with 2B standards, with the occasional exception of dissolved oxygen, dissolved copper, and TDS. Additional parameters including chloride, arsenic, selenium, cadmium (D), and lead (D) have been measured above the 2B standard at a single location. Most of these exceptions occur at SW-10, which is located adjacent to Outfall 001. Groundwater monitoring wells located downgradient of the ash basin and adjacent to the Lake Norman shoreline include: MW-7S/D, MW-8S/D, MW-9S/D, MW-10S/D, and GWA 15S. Surface water results for boron are within a similar range of observed boron results in groundwater samples, although boron results from SW-12 show concentrations that are substantially less than those of adjacent well GWA-15S. Levels of TDS are generally lower in surface water samples than levels of TDS from adjacent groundwater monitoring wells. Piper Diagrams Piper diagrams are used to graphically depict the general chemical character of water by plotting the relative concentrations of major ions on a series of tri-linear diagrams. These points are then projected onto a central diamond plot to show the overall chemical character of the water (Piper, 1944). Using these diagrams to understand both source area water characteristics (ash pore water) and natural water characteristics (background water) provides the basis for useful evaluation of downgradient water. Evaluation of the background groundwater Piper diagram (Figure 9-1) results in the following observations: y Free-standing water collected from within the basin (SW-01) is characterized as calcium sulfate type water and plots within the same area as pore water in the central portion of the ash basin. Surface water collected from the unnamed tributary adjacent to the Dry Ash Landfill Phase I (S-1/SW-6) and near the downgradient shore adjacent to the ash basin (SW-11, and SW-12) is characterized as a mixture between calcium bicarbonate and calcium sulfate type water. Page 9-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Sample location SW-10, located near Outfall 001, is characterized as calcium sulfate type water. Background surface water samples are characterized as calcium bicarbonate water and plot within the same area on the diagram as background groundwater. Page 9-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 10.0 GROUNDWATER SAMPLING RESULTS This section provides a summary of groundwater monitoring and an overview analysis of the most recent data available (3Q 2017 or most recent data available). A more detailed discussion of analytical data results and trends is included is Section 11. Comprehensive tables with all media analytical results are provided in Appendix B. As directed by NCDEQ, the data with turbidity greater than 10 Nephelometric Turbidity Units (NTUs) and pH greater than 8.5 that may be a result of grout intrusion, as well as data that may be auto -correlated because it was collected within 60 days of a previous sampling event are excluded for statistics and other evaluation methods. The most recent data available are shown on the pertinent figures. The most recent data available and reported herein are from September 2017 (3Q 2017). One comprehensive round of sampling and analysis was conducted prior to, and reported in, the August 2015 CSA. In addition, the following sampling and analysis events were conducted: '67 Comprehensive Round (1) - July 2015 (reported in September 2015 CAP Part 1) '67 Comprehensive Round (2) - September/October 2015 (reported in March 2016 CAP Part 2 report) 167 Limited Round (3) - November 2015 (background wells only, reported in the CSA Supplement 1 as part of the February 2016 CAP Part 2) 01 Limited Round (4) - December 2015 (background wells only, reported in the CSA Supplement 1 as part of the February 2016 CAP Part 2) 01 Comprehensive Round (5) - March/April 2016 (reported in August 2016 CSA Supplement 2) 01 Comprehensive Round (6) - June 2016 01 Comprehensive Round (7) - September 2016 01 Comprehensive Round (8) - December 2016 01 Comprehensive Round (9) - March 2017 -61, Comprehensive Round (10) - June 2017 01 Comprehensive Round (11) - September 2017 Groundwater sampling methods and the rationale for sampling locations were in general accordance with the procedures described in the Work Plan (HDR, 2014c) and Page 10-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra are included in Appendix G. Variances from the proposed well installation locations, methods, quantities, and well designations are presented in Appendix G. As described in the approved Work Plan, both unfiltered and filtered (0.45 um filter) samples were collected for analyses of constituents whose results may be biased by the presence of turbidity. Unless otherwise noted, concentration results discussed are for the unfiltered samples and represent total concentrations. 10.1 Background Groundwater Concentrations Locations for background monitoring wells installed in 2015 for the initial CSA field effort were chosen based on the information available, including the previously installed NPDES monitoring well network, horizontal distance from the waste boundary, and the relative topographic and groundwater elevation difference compared to the ash basin surface water. After the background wells were installed and a sufficient number of samples were collected, statistical analysis was used to confirm the analytical results represented background conditions. The following monitoring wells have been approved by NCDEQ as background monitoring wells (Zimmerman to Draovitch, July 7, 2017; Appendix A). Background monitoring wells are depicted on Figure 2-4. 167 BG-1S/D 167 BG-2S/BR 167 BG-3S/3D/3BR 01 GWA-4S/D 167 GWA-5S/D 161, GWA-6S/D 161, GWA-8S/D 161, GWA-12S/BR '61P MS-10 167 MW-4 and MW-4D Monitoring well locations BG-1BR and GWA-12D were not approved for use pending replacement and reevaluation as potential background locations. Page 10-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 10.1.1 Background Dataset Statistical Analysis The revised background groundwater datasets and statistically determined PBTVs are presented below. The current background monitoring well network consists of wells installed within three flow zones — shallow, deep, and bedrock. NCDEQ requested that the updated background groundwater dataset exclude data associated with one or more of the following conditions: 167 Sample pH is greater than or equal to 8.5 standard units unless the regional DEQ office has determined an alternate background threshold pH for the Site. Sample turbidity is greater than or equal to 10 Nephelometric Turbidity Units (NTUs). y Result is a statistical outlier identified for background sample data presented to NCDEQ on May 26, 2017. 01 Sample collection occurred less than a minimum 60 days between sampling events. ,67 Non -detected results are greater than 2L/IMAC. Statistical determinations of PBTVs were performed in accordance with the revised Statistical Methods for Developing Reference Background Concentrations for Groundwater and Soil at Coal Ash Facilities (statistical methods document) (HDR and SynTerra, 2017). The background datasets for each flow system used to statistically determine naturally occurring concentrations of inorganic constituents in groundwater are provided in Appendix H. The following sections summarize the refined background datasets along with the results of the statistical evaluations for determining PBTVs. Shallow Flow System Ten (10) wells — BG-1S, BG-2S, BG-3S, GWA-4S, GWA-5S, GWA-6S, GWA-8S, GWA-12S, MS-10, and MW-4 — are used to monitor background groundwater quality within the shallow flow zone. The shallow background groundwater dataset meets the minimum requirement of 10 samples for all constituents. PBTVs were calculated for constituents monitored within the shallow flow zone using formal Upper Tolerance Limit (UTL) statistics and are included in Table 10-1. Page 10-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Deep Flow System Six (6) wells — BG-11), BG-31), GWA-41), GWA-51), GWA-61), and GWA-81) — are used to monitor background groundwater quality within the deep flow zone. The background groundwater dataset meets the minimum requirement of 10 samples for all constituents. PBTVs were calculated for constituents monitored within the deep flow system using formal UTL statistics (Table 10-1). Bedrock Five wells (5) — BG-1BRA, BG-2BR, BG-3BR, GWA-12BR, and MW-41) — are used to monitor background groundwater quality within the fractured bedrock. BG-1BRA was installed in 2017 as a replacement to BG-1BR (grout contamination). BG-1BRA has been sampled three times in 2017, and results indicate viable background data with the exception of the first sample taken in March 2017, which contained elevated turbidity. The background groundwater dataset meets the minimum requirement of 10 samples for all constituents. PBTVs were calculated for constituents monitored within the deep flow system using formal UTL statistics (Table 10-1). Summary The calculated groundwater PBTVs were less than the applicable 2L/IMAC standard for every constituent within each of the three flow units except: 47 Cobalt: PBTVs of 3.7 µg/L (shallow); 2.4 µg/L (deep); 3.8 (bedrock) versus IMAC of 1 µg/L. 47 Iron: PBTVs of 818 µg/L (shallow); 337 µg/L (deep); 676 µg/L (bedrock) versus 2L of 300 µg/L. ,67 Manganese: PBTVs of 82 µg/L (shallow);187 µg/L (deep); 310 µg/L (bedrock) versus 2L of 50 µg/L. 07 Vanadium: PBTVs of 6.88 µg/L (shallow); 4.37 µg/L (deep); 22.9 µg/L (bedrock) versus IMAC of 0.3 µg/L. Groundwater PBTVs were calculated for the following constituents that do not have a 2L standard, IMAC standard, or Federal Maximum Contaminant Level (MCL) established: alkalinity, bicarbonate, calcium, carbonate, magnesium, methane, potassium, sodium, strontium, sulfide, and TOC. Evaluation of background threshold values will continue, and the values will be adjusted over time as additional background data become available (Appendix H). Page 10-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 10.1.2 Piper Diagrams — Background Wells Piper diagrams are used to graphically depict the general chemical character of water by plotting the relative concentrations of major ions on a series of tri-linear diagrams. These points are then projected onto a central diamond plot to show the overall chemical character of the water (Piper, 1944). Using these diagrams to understand both source area water characteristics (ash pore water) and natural water characteristics (background water) provides the basis for useful evaluation of downgradient water. Evaluation of the background groundwater Piper diagram (Figure 10-1) results in the following observations: y Generally, background groundwater at MSS is classified as calcium - bicarbonate type water. 101 Analytical results indicate groundwater at GWA-12S is sodium dominant, potentially indicating a different water type of unknown origin. ,67 Analytical results indicate groundwater at BG-01S plots close to calcium sulfate type water. BG-01S is an approved background location and therefore the water type is not considered reflective of influence from basin operations. The USGS has conducted a regionally significant hydrogeologic evaluation of groundwater flow and groundwater quality in the vicinity of MSS. The five-year study at the Langtree Peninsula Research Station (on a nearby shore of Lake Norman) provides additional information about background water quality conditions in the region (USGS, 2008). Observations from that study indicate the majority of samples collected were calcium -bicarbonate water type, consistent with background wells located at the MSS. 10.2 Downgradient Groundwater Concentrations The following is a summary of groundwater analytical data for areas around the MSS ash basin. The comprehensive groundwater analytical data table is included as Appendix B, Table 1. 10.2.1 Monitoring Wells Beneath Ash Basin Nine monitoring wells are used to evaluate shallow groundwater immediately below the ash basin. Five of those wells are located within the waste boundary of the active ash basin while the remaining four are adjacent to the waste boundary and directly below or downgradient of other source areas such as the Industrial Landfill No. 1 or the Dry Ash Landfill (Phase II). Page 10-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Eighteen (18) monitoring wells within the waste boundary and three wells beneath the Dry Ash Landfill (Phase II) are used to evaluate deep groundwater below the ash basin. Seven bedrock wells within the waste boundary and five bedrock wells beneath the Dry Ash Landfill (Phase II) are used to evaluate the bedrock flow system beneath the ash basin. Deep bedrock wells AL-02BRL and AL-02BRLL, located within the Dry Ash Landfill (Phase II), are installed for evaluating an area of apparent downward vertical migration. 10.2.2 Monitoring Wells Downgradient of Ash Basin The area downgradient from the ash basin is limited in geographic extent. Generally, the area between the ash basin dam and Lake Norman and the area between the Dry Ash Landfill (Phase I) and the unnamed tributary east of the basin are considered downgradient. Thirty-one (31) monitoring wells installed to support the ongoing CAMA CSA effort are used to monitor those areas. Downgradient area wells are screened within the shallow (12), deep (13), and bedrock (6) flow systems. 10.2.3 Piper Diagrams - Wells Beneath and Downgradient from the Ash Basin Groundwater analytical results for Q3 2017 have been plotted on a Piper diagram to allow comparison with background and ash pore water characteristics. Cation —anion charge balance calculations are an indicator of data quality. Samples with a charge balance of greater than 10% are not included in the diagrams. Cation - anion charge balance results are provided in Table 10-2. Generalized areas of the ash pore water and background groundwater have been included on the figure for comparative purposes. Evaluation of the Piper diagram for wells beneath the ash basin (Figure 10-2) results in the following observations: 167 In general the majority of groundwater samples collected from beneath the ash basin indicate a similar water type to ash pore water. �7 Samples collected from eight locations (AB-71), AB-9BR, AB-111), AB-131), AB-171), AB-181), AB-20D, and AB-21D) indicate a similar water type to background water. �7 Locations that plot within the area similar to background water characteristics have boron concentrations of <50 µg/L. Page 10-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Evaluation of the Piper diagram for wells located downgradient of the ash basin (Figure 10-3) results in the following observations: ,67 Groundwater at downgradient areas of the MSS ash basin - described as the ash basin dam and area between the Dry Ash Landfill (Phase I) and the unnamed tributary - are similar to characteristics of ash pore water near the central portion of the ash basin. 01 Samples collected from GWA-11BR and MW-10D indicate a similar water type to background water. 167 Boron concentrations at GWA-11BR and MW-10D are <50 µg/L. 10.2.4 Radiological Laboratory Testing Radionuclides may exist dissolved in water from natural sources (e.g. soil or rock). The USEPA regulates various radionuclides in drinking water. Radium- 226, radium-228, total radium, uranium-238, and total uranium were analyzed in 46 wells as part of the CAMA sampling event in September of 2017. Results for radiological laboratory testing are presented in Appendix B, Table 1. Radium and uranium isotopes were detected at levels greater than the USEPA MCLs at the following seven (7) locations: ,67 Background - BG-2BR ,67 Beneath the Ash Basin - AB-12D/BR 167 Downgradient - AL-01D/BR, AB-01BR, MW-07D 10.3 Site -Specific Exceedances (Groundwater COIs) Site -specific groundwater COIs were developed by evaluating groundwater sampling results with respect to 2L/IMAC and PBTVs, and additional regulatory input/requirements. In addition, the distribution of constituents in relation to the ash basin, co -occurrence with CCR indicator constituents such as boron, and likely migration directions based on groundwater flow direction are also considered in determination of groundwater COIs. The approach to determining those constituents, which should be considered COIs for the purpose of evaluating a Site remedy, is discussed in the following section. 10.3.1 Background Threshold Values (PBTVs) As presented in 15A NCAC 02L .0202 (b)(3) - "Where naturally occurring substances exceed the established standard, the standard shall be the naturally occurring concentration as determined by the Director" - the following report provided to NCDEQ: Statistical Methods for Developing Reference Background Page 10-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Concentrations for Groundwater and Soil at Coal Ash Facilities (HDR and SynTerra, 2017) was provided to NCDEQ. NCDEQ (July 7, 2017) addressed each Duke Energy coal ash facility and identified soil and groundwater data appropriate for inclusion in the statistical analysis to determine BTVs and PBTVs. A revised and updated technical memorandum that summarized background groundwater datasets and statistically determined PBTVs for MSS was submitted to NCDEQ on September 5, 2017. NCDEQ approved or provided alternative PBTVs for consideration for MSS in a letter dated October 11, 2017. Subsequent to the October 11, 2017 letter additional background wells (BG-1S/D/BR, BG-2S/BR, and BG-3S/D/BR) were included in the statistical evaluation resulting in minor changes to PBTVs as described in Appendix H. 10.3.2 Applicable Standards As part of CSA activities at the Site, sampling and analysis for inorganic constituents has been conducted for coal ash, ponded water in the ash basin, ash pore water, AOW, surface water, sediment, soil, and groundwater downgradient/sidegradient of the ash basin and in background areas. Based on comparison of those sampling results from the multiple media to background values and applicable regulatory values, potential lists of COIs were developed in the 2015 CSA, CAPs, and CSA Supplement. For the purpose of developing the groundwater COIs, constituent exceedances in downgradient groundwater of PBTVs and 2L or IMAC are considered a primary focus. Additionally, NCDEQ requested that hexavalent chromium be included as a COI at each CAMA-related site due to public interest and receptor wells. An evaluation of data at MSS indicates hexavalent chromium has only been detected in one valid sample out of 867 analysis at a level greater than the 2L standard for total chromium (10 µg/L); therefore, hexavalent chromium is not considered as a COI. Molybdenum and strontium do not have 2L or IMACs established; however, these constituents are considered potential COIs with regards to CCR and are evaluated as potential COIs for the Site at the request of NCDEQ. The following constituents do not have a 2L standard, IMAC, or Federal MCL established: alkalinity, aluminum, bicarbonate, calcium, carbonate, magnesium, methane, potassium, sodium, sulfide, and TOC. Results from these constituents are useful in comparing water conditions throughout the Site. For example calcium is listed as a constituent for detection monitoring in Appendix III to 40 CFR Part 257. Although these constituents will be used to compare and understand groundwater quality conditions at the Site, because there are no Page 10-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra associated 2L, IMACs, or MCLs, these constituents are not evaluated as potential COIs for the Site. 10.3.3 Additional Requirements NCDEQ requested that consideration of constituents be included in the CSA for the constituents in 40 CFR 257, Appendix III detection monitoring and 40 CFR 257, Appendix IV assessment monitoring (CCR Rule) (USEPA, April 2015). Detection monitoring constituents in 40 CFR 257 Appendix III are: ,0 Boron 10 Calcium 101 Chloride 10 Fluoride (limited historical data at this Site, not on assessment constituent list) y pH 0 Sulfate 10 TDS Constituents for assessment monitoring listed in 40 CFR 257 Appendix IV include: 10 Antimony ,61P Arsenic 161P Barium ,61P Beryllium 161P Cadmium ,61P Chromium ,61P Cobalt ,67 Fluoride (limited historical data at this Site, not on assessment constituent list) ,67 Lead ,67 Lithium (not analyzed) 01 Mercury Page 10-9 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 47 Molybdenum 10 Selenium 167 Thallium 10 Radium 226 and 228 combined Aluminum, copper, iron, manganese, and sulfide were originally included in the Appendix IV constituents in the draft rule; USEPA removed these constituents in the final rule. Therefore, these constituents are not included in the listing above; however, they are included as part of the current Interim Monitoring Plan (IMP; Section 15.3). NCDEQ requested that vanadium be included as a COI. 10.3.4 Marshall COIs Based on site -specific conditions, observations, and findings, the following list of COIs has been developed for MSS: 01 Antimony 47 Arsenic 167 Barium 47 Beryllium ,67 Boron ,67 Cadmium 167 Chloride ,67 Chromium (total) 161P Cobalt ,61P Iron -61, Manganese 161P Molybdenum 167 Nickel 167 Selenium ,67 Strontium 167 Sulfate ,67 TDS Page 10-10 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra ,610 Thallium ,610 Vanadium Table 10-1 lists the COIs and other constituents at MSS along with the established PBTVs and associated 2L/IMACs. COIs detected at concentrations greater than the PBTVs and associated 2L/IMACs (if applicable) for each flow unit beyond the compliance boundary (or within bedrock monitoring wells within or beyond the compliance boundary) are as follows: 101 Shallow - antimony, boron, chromium, cobalt, iron, manganese, molybdenum, selenium, strontium, sulfate, TDS, thallium y Deep - boron, cobalt, iron, manganese, strontium, sulfate, TDS, vanadium ,67 Bedrock - antimony, barium, boron, chromium, cobalt, iron, manganese, molybdenum, strontium, TDS A constituent was not associated with a flow layer in the lists above if concentrations detected greater than the PBTVs and associated 2L/IMACs were exclusively observed in upgradient, or background, locations. This section provides a summary of groundwater monitoring and an overview analysis of the most recent data available (3Q 2017 or most recent data available). A more detailed discussion of analytical data results and trends is included in Section 11. A comprehensive table with all media analytical results is provided in Appendix B. As directed by NCDEQ, the data with turbidity greater than 10 NTUs and pH greater than 8.5 that may be a result of grout intrusion, as well as data that may be auto -correlated because it was collected within 60 days of a previous sampling event, are excluded for statistics and other evaluation methods. The most recent data available are shown on the pertinent maps. Groundwater sampling methods and the rationale for sampling locations were in general accordance with the procedures described in the Work Plan (HDR, 2014c) and are included in Appendix G. Variances from the proposed well installation locations, methods, quantities, and well designations are presented in Appendix G. As described in the approved Work Plan, both unfiltered and filtered (0.45 um filter) samples were collected for analyses of constituents whose results may be biased by the presence of turbidity. Unless otherwise noted, concentration results discussed are for the unfiltered samples and represent total concentrations. Page 10-11 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 11.0 HYDROGEOLOGICAL INVESTIGATION Results from the hydrogeological assessment of MSS, summarized in this section, are primary components of the Site Conceptual Model (SCM). 11.1 Plume Physical and Chemical Characterization Plume physical and chemical characterization for each groundwater COI is detailed below. The horizontal extent and vertical extent of constituent concentrations are presented on isoconcentration maps and cross -sections. These descriptions and depictions are based primarily on the most recent groundwater dataset available (September 2017 or most recent data available). 11.1.1 Plume Physical Characterization The groundwater plume is defined as locations (in three-dimensional space) where groundwater quality is impacted by the ash basin. COI (defined in Section 10.0) background concentrations are used to help refine the extent and degree to which areas are impacted by groundwater from the ash basin. The comprehensive groundwater data table (Appendix B, Table 1) and an understanding of groundwater flow dynamics and direction (Section 6.3, Figures 6-5 to 6-10) were used to define the horizontal and vertical extent of the plume. Not all constituents with PBTV exceedances are attributed to the ash basin, in many instances because the calculated background value does not fully encompass the range of potential regional variability in groundwater concentrations. Naturally occurring groundwater contains varying concentrations of alkalinity, aluminum, bicarbonate, cadmium, carbonate, copper, lead, magnesium, methane, nickel, potassium, sodium, TOC, and zinc. Inconsistent and low -concentration exceedances of those constituents in the groundwater data do not necessarily demonstrate horizontal or vertical distribution in groundwater that indicates impact from the ash basin. Horizontal Extent - Isoconcentration Maps The horizontal extent of the plume in each flow unit is interpreted in concentration isoconcentration maps (Figures 11-1 to 11-63). These maps use valid groundwater analytical data to spatially and visually define areas where groundwater concentrations are greater than the respective constituent PBTV and 2L/IMAC standard, if applicable. The thin vertical extent of unconsolidated material beneath the ash basin precludes the installation of monitoring wells within the shallow flow system at many locations within the ash basin waste boundary. Therefore ash pore water concentrations, while not representative of Page 11-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra shallow groundwater, were used to infer potential concentrations in the underlying shallow flow system for generating isoconcentration figures. The farthest downgradient edge (leading edge) of the plume is observed at the MW-14 location. Generally, constituents at this location occur at concentrations greater than the 2L or PBTV while analytical results from monitoring well clusters located farther downgradient (GWA-7 and MW-10) are less than the 2L standard or PBTVs. However, concentrations at MW-14 tend to decrease with depth. For example, boron concentrations decrease from 2,340 µg/L at MW-14S to 2,140 µg/L at MW-14D to 53.7 µg/L at MW-14BR. This trend is also observed in downgradient areas of the ash basin towards the unnamed tributary of Lake Norman. Boron is consistently less than or slightly over the detection limit in GWA-11BR, but is routinely detected at concentrations greater than the 2L beyond the ash basin compliance boundary in GWA-11S/D and GWA-15S. Boron is considered an ideal constituent to evaluate the overall geometry of impacted groundwater because it is highly mobile in groundwater and relatively unaffected by changing geochemical conditions. Review of Figures 11-13 to 11- 15, which depict the horizontal extent of boron in groundwater, are useful for defining the general plume geometry of impacted groundwater. Concentrations above background conditions in the shallow flow unit are generally limited to the perimeter of the ash basin and the downgradient area adjacent to Lake Norman, with the exception of the area surrounding MW-10S. In the vicinity of the unnamed tributary east of the ash basin, boron remains non -detect at the GWA-7 cluster. However, migration of impacted groundwater within the shallow and deep flow layers toward the unnamed tributary is apparent with consistent boron detections in MW-14S/D. The plume geometry of impacted groundwater remains consistent with depth but encompasses a smaller area with each successive flow unit. As described in Section 6.0, there is no hydrogeologic confining unit at MSS; therefore, under these unconfined conditions, groundwater moves across each unit. General plume geometry indicates migration of impacted groundwater downgradient and east of the ash basin toward the unnamed tributary of Lake Norman and Lake Norman proper. A lobe -like geometry of the plume observed in certain COIs (e.g. boron, iron, and manganese) due to relatively lower concentrations at the MW-1 location. The lower concentration of constituents at MW-1 may result from shallow groundwater flow that is somewhat isolated from deeper groundwater flow influenced by the basin. Migration of CCR- Page 11-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra impacted groundwater is not apparent within MW-10S/D, located in the eastern peninsula immediately adjacent to Lake Norman. Vertical Extent - Cross -Sections The vertical extent of impacted groundwater is depicted on the generalized cross -sectional views of the Site (Figures 11-64 to 11-120). These cross -sections display vertical delineation of constituent concentrations along the three transects across the ash basin. Transects span from northwest to southeast (A -A'), southwest to southeast (B-B'), and northeast to southwest (C-C'). These transects capture upgradient, source area, and downgradient wells relative to the ash basin. COIs have been contoured in the cross -sectional depictions. Constituent isocontours reflect values above PBTVs and the 2L/IMAC standard, as applicable. The shallow flow zone was not screened at many locations beneath the ash basin. Thickness of the shallow flow zone varies across the Site, however for Site assessment purposes, the impacts in this zone are sufficiently understood as the TZ is interpreted to be the primary flow zone with a predominantly horizontal component of flow that would impact receptors. The well screens in the CAMA wells accurately monitor groundwater conditions and impact to the deep groundwater flow zone. The boron cross -sections (Figures 11-76 to 11-78), used as a conservative estimate of vertical migration, display impacts beneath the ash basin and downgradient areas in the shallow and deep flow zones. Other COI concentrations also exceed 2L/IMAC and PBTV values in these two flow zones beneath and downgradient of the ash basin. Generally, limited vertical migration of constituents into deep bedrock occurs across the Site. Based on the boron cross -sections, vertical migration of impacted waters into deep bedrock is not apparent, however additional analysis of data has identified areas of potential vertical migration into deep bedrock as discussed below. Though not captured on cross -sectional views, vertical migration of COI concentrations into underlying bedrock is apparent beneath the Dry Ash Landfill (Phase II). A deep bedrock well (AL-2BRLL) for vertical delineation beneath the Dry Ash Landfill (Phase II) was recently installed. Boron was detected greater than 9,000 µg/L in this well for the first and only sample collected to date. Vertical migration in this area appears relatively isolated to beneath the Dry Ash Page 11-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Landfill (Phase II), as several adjacent bedrock wells within the ash basin (AB- 12BR and AB-10BR) do not indicate boron concentrations of this magnitude. The A -A' and C-C' contoured cross -sections display limited vertical migration downgradient of the Dry Ash Landfill (Phase II). Strontium and TDS (and sulfate for A -A') are the only constituents that are consistently detected at concentrations above the PBTVs in bedrock wells along the transects. These observations may indicate a set of isolated fractures at depth beneath the Dry Ash Landfill (Phase II) in which dilution, or influence from meteoric and non - impacted waters, is not occurring. It appears source area water is migrating to these fractures and may be experiencing retarded flow with lateral distance as the fractures dissipate or are subject to reduced hydraulic conductivity. Another area of apparent vertical migration that is not captured on the cross - sections is at the AB-1 location (AB -OR), immediately east of the ash basin. Several COIs in deep and bedrock groundwater (boron, chloride, iron, manganese, strontium, and TDS) display increasing trends over time at this location. Two potential scenarios could explain the observation of increasing trends: 07 A compromised grout column at AB-1BR could allow overlying concentrations from the shallow and deep flow system to migrate downward, influencing groundwater concentrations measured within the vicinity of the well installation. y Relatively slow groundwater migration through the bedrock flow system could result in the observation of increasing groundwater concentrations once the leading edge of the plume has passed AB-1BR and source concentrations continue to migrate toward the downgradient discharge zone. As an example of retarded groundwater migration through bedrock, the Site - wide geometric mean hydraulic conductivity of 0.609 feet per day would result in a transport time through bedrock (for mobile constituents with a negligible Kd) of approximately 17 years from the central area of the basin near the AL-2 cluster to the AB-1 cluster. Continued monitoring of the recently installed AB-1BRL (deeper bedrock) will allow continued evaluation of these two scenarios to better understand constituent concentrations and trends in the area of AB-1. These concentrations Page 11-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra are presumed to migrate beneath, and eventually discharge to, the Catawba River (Lake Norman). In general across the Site, vertical hydraulic gradients (Figure 6-11) remain relatively neutral. Therefore, groundwater analytical results are the primary means of interpretation for vertical constituent migration. In summary, the horizontal and vertical extent of the plume has been defined. Monitoring wells across the Site are appropriately placed and screened to the correct elevations to adequately assess the conditions across the Site. Monitoring wells installed for other regulatory programs¢ have added additional information about the orientation and extent of the downgradient plume and have helped refine an understanding of the vertical and horizontal distribution of the plume. Separate flow transects from cross-section transects, planned for use in the geochemical model, have been selected and are shown on Figure 11-121. Transects used for geochemical modeling are planned to focus on specific areas of interest to optimize geochemical evaluation. Details of the geochemical model are further discussed in Section 13. A description of each planned transect and data inventory for each transect is detailed below (asterisks (*) shown denote wells screened within ash). Data inventory associated with the geochemical flow transects is summarized on Table 11-1. Western Transect Wells that comprise the western transect for geochemical modeling beginning with the upgradient wells and continuing down the flow path are as follows: AB- 5S*/D/BR 4 AB-3S/D 4 AB-2S/D 4 MW-9S/D. In the southwest portion of the ash basin, east of the FGD Residue Landfill, groundwater flows east from the ponded fingers of the Ash Basin to Lake Norman. Boron concentration decreases with distance along the flow transect in the shallow flow zone and increases with distance along the flow transect in the 4 Pursuant to the CCR rule, owners and operators of CCR units must install the required groundwater monitoring system; develop the required groundwater sampling and analysis program to include selection of the statistical procedures to be used for evaluating groundwater monitoring data; and begin detection monitoring, which requires owners and operators to have a minimum of eight samples for each well and begin evaluating groundwater monitoring data for statistically significant increases over background levels for the constituents listed in Appendix III of 40 C.F.R. Part 257. The data is posted to Duke Energy's publicly accessible Internet site as required under the CCR rule. The most recent data available has been considered in this assessment for refinement of constituent distribution. Page 11-5 P:\Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra deep flow zone. Observations at the two bedrock wells on the western portion of the ash basin (AB-5BR and GWA-1BR) are both below detection limits. This flow transect consists of one well screened within the ash pore water, three wells screened within the shallow flow zone, four wells screened within the deep flow zone, and one well screened within the bedrock flow zone. In the shallow and deep flow zones, most CCR constituents have a decreasing trend between the AB-5 location and the AB-2 location, and a slight increase in concentration at the MW-9 well pair. Of the nine wells along the centerline of this flow transect, all have at least seven valid sampling events. There are 18 wells located perpendicular to the centerline of flow for the Western Transect: two in ash pore wells, seven shallow wells, eight deep wells, and one bedrock well. Of these 18 wells, 16 have at least six viable sampling events each (Table 11-1). Central Transect Wells that comprise the central transect for geochemical modeling beginning with the upgradient wells and continuing down the flow path are as follows: AB- 12S*/SL*/D/BR 4 AB-9S/D/BR 4 AB-1S/D/BR/BRL. This flow transect consist of two wells screened within ash pore water, two wells screened within the shallow flow zone, three wells screened within the deep flow zone, and four wells screened within the bedrock flow zone. Boron concentration along the central transect decreases from source area, AB-12 well cluster to the downgradient, AB-1 well cluster in both the shallow and deep flow zones, and increases along the transect for the bedrock flow zone. The concentrations of the centrally located AB-9 well cluster are below detection limits for all flow zones. Note, the shallow flow zone at the AB-12 location is represented by ash pore water rather than a well screened within the shallow flow zone beneath the ash. While the ash pore well at AB-12 can be incorporated into the geochemical model, a well screened within the shallow flow zone beneath the ash would allow for characterization of the ash basin impact on groundwater beneath the basin. The concentration of boron in the deep and bedrock wells at AB-1 are comparable, indicating interaction between these flow zones. Of the 11 wells along the centerline of this flow transect, MW-12BR and AB-1BRL are the only wells with less than six valid sampling events. MW-12BR has three Page 11-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra valid sampling events between March 2017 and September 2017, and AB-1BRL has only been sampled once. There are 17 wells located perpendicular to the proposed centerline of flow for the Central Transect: four ash pore wells, five shallow wells, seven deep wells, and one bedrock well. Of those 17 wells, 15 have at least seven valid sampling events (Table 11-1). Eastern Transect Wells that comprise the eastern transect for geochemical modeling beginning with the upgradient wells and continuing down the flow path are as follows: CCR-13S/D 4 GWA-11S/D/BR 4 GWA-15S. This eastern transect consists of three shallow wells, two deep wells, and one bedrock well. Along the southeastern portion of the ash basin waste boundary, groundwater flows east through the Dry Ash Landfill (Phase I) toward an unnamed tributary to Lake Norman. Trends in the boron concentration along the eastern flow transect decrease in the shallow flow zone, increase in the deep flow zone, and concentrations in the one bedrock well, GWA-11BR, are below the detection limit. Of the six wells along the centerline of the Eastern Transect, two have less than seven valid sampling events: GWA-15S with three valid sampling events between May and September 2017 and GWA-11BR with two valid sampling events in April and June 2017. There are 15 wells located perpendicular to the centerline of flow for the eastern transect: seven shallow wells, six deep wells, and two bedrock wells. With the exception of AL-1BR (four valid sampling events), all 14 wells have at least seven valid sampling events as of September 2017 (Table 11-1). 11.1.2 Plume Chemical Characterization Plume chemical characterization is detailed below for each COI. Data evaluations are primarily based on the September 2017 groundwater sampling event. The range of detected concentrations is presented with the number of detections for the sampling event. Lab -qualified data are not considered for the summary below. Descriptions of the COIs identified for MSS are also provided. PBTVs and 2L/IMAC standards are included in Table 10-1. Ash pore water (source) concentrations are discussed in Section 3.0 and omitted from the data set. Samples that have turbidity greater than 10 NTUs or pH greater than 8.5 Page 11-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra (indicative of grout contamination in the well) have also been removed from the data set. Antimony Detected Range: 0.6 µg/L - 0.89 µg/L; 3 detections/108 samples (3/108) ,0 One concentration from an upgradient, deep flow zone well west of the ash basin (GWA-13DA) exceeded both the PBTV and IMAC. 167 Four deep flow zone groundwater samples and one bedrock sample (AL- 2BR) exceeded the PBTV of 0.5 µg/L. '61' Concentrations from four ash pore water wells exceeded the IMAC. One concentration from one downgradient bedrock well east of the ash basin (GWA-11BR) exceeded both the PBTV and IMAC. Antimony is a silvery -white, brittle metal. Small amounts of antimony are naturally present in rocks, sediments, soils, and water. Since no 2L value has been established for this constituent by NCDEQ, concentrations of antimony are compared with IMAC standards. Antimony can occur in pyrite and sulfides associated with pyrite in coal (Finkelman, 1995). Antimony is a minor trace element in the crust that occurs in concentrations of 0.26 milligrams per kilogram in felsic (light colored, silica rich) rocks to 2.0 mg/kg in clays and shales (Parker, 1967), Table 19). Arsenic Detected Range: 0.10 µg/L -15.1 µg/L; 46 Detections/108 Total Samples (46/108) Concentrations in one sample (AB-16S; shallow flow zone beneath ash basin) exceeded the 2L. y Concentrations from four downgradient wells in the shallow flow zone exceeded the PBTV. 1610 Concentrations from six wells in the deep flow zone and two wells in the bedrock flow zone beneath the ash basin exceeded the PBTV. 07 Vertical migration of arsenic at concentrations above the PBTV is not apparent across the MSS site. Arsenic is a trace element in the crust, with estimated concentrations ranging from less than one mg/kg in mafic igneous rocks to 13 mg/kg in clay -rich rocks (Parker, 1967). The USEPA estimates that the amount of natural arsenic released Page 11-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra into the air as dust from the soil is approximately equal to the amount of arsenic released by all human activities (EPRI, 2008). Arsenic occurs in multiple valence states (As5+, As3+, and As3-). Arsenic in coal occurs primarily in pyrite (iron sulfide, with arsenic replacing iron in the crystal structure) (Finkelman, 1995). Arsenic condenses on fly ash as arsenate (As5+) (Goodarzi, Huggins, & Sanei, 2008). Leaching tests on ash indicate that trace quantities of up to 50 percent of the arsenic present can be leached. In addition to the solubility of the source, the concentration of calcium and presence of oxides appear to limit the mobility of arsenic (Izquierdo & Querol, 2012). The primary attenuation mechanisms for arsenic with aquifer solids are precipitation of metal arsenates or arsenites, precipitation of arsenic sulfides, co -precipitation with common soil/sediment minerals such as iron oxides and iron sulfides, and adsorption to iron oxyhydroxides, iron sulfides, or other mineral surfaces. The long-term stability of arsenic immobilized onto aquifer solids will depend on the groundwater chemistry over time, specifically pH, redox potential (Eh), and the concentration of other anions competing for sorption sites (EPA, 2007). Barium Detected Range: 10.2 µg/L -1,010 µg/L; 108 Detections/108 Total Samples (108/108) 07 One downgradient sample (AL-11); deep flow zone) exceeded the 2L. '67 Concentrations from two downgradient wells in the shallow flow zone exceeded the PBTV. '67 A slightly increasing trend above the PBTV is observed at downgradient well MW-1 (shallow flow zone). 07 Three exceedances of the PBTV in the shallow flow zone may indicate downgradient migration east of the ash basin toward the unnamed tributary to Lake Norman. 07 Concentrations in 13 wells in the deep flow zone beneath the ash basin or downgradient exceeded the PBTV. Barium is a naturally occurring component of minerals that are found in small but widely distributed amounts in the Earth's crust (Kunesh, 1978); (Miner, 1969). Two forms of barium, barium sulfate (barite) and barium carbonate (witherite), are often found in nature as ore deposits. Barium enters the environment naturally through the weathering of rocks and minerals. Anthropogenic releases are associated primarily with industrial processes. Page 11-9 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Barium is sometimes found naturally in drinking water and food. However, because the dominant naturally occurring barium compounds (barium sulfate and barium carbonate) have a low to moderate solubility in water under most conditions, the amount of barium found in drinking water is typically small. Barium compounds such as barium acetate, barium chloride, barium hydroxide, barium nitrate, and barium sulfide dissolve more easily in water than barium sulfate and barium carbonate, but because they are not commonly found in nature, the latter two compounds do not usually occur in drinking water unless the water is contaminated by barium compounds that are released from waste sites (EPRI, 2008b). Barium is naturally released into the air by soils as they erode and is released into the soil and water by eroding rocks. Barium released into the air by human activities comes mainly from barium mines, metal production facilities, and industrial boilers that burn coal and oil (EPRI, 2008b). The leachability of barium has been found to be relatively independent of pH but is controlled instead by the presence of calcium, with which it competes for sulfate (Fruchter, Rai, & Zachara, 1990). In an overview of leachability studies found in the International Journal of Coal Geology, the mobility of barium typically ranged from 0.02 percent to 2 percent (Izquierdo & Querol, 2012). Regional metamorphic grade greenschist to upper amphibolite in the King's Mountain Belt of the Piedmont contains deposits of barium sulfate (barite). Barium is especially common as concretions and vein fillings in limestone and dolostone, which are not common geologic facies in North Carolina; however, at various times in the past century and a half, the Carolinas have been major producers of barite (USEPA, 2017a) . In a statistical summary of groundwater quality in North Carolina, the Superfund Research Program at the University of North Carolina (UNC) analyzed 1,898 private well water samples in Gaston and Mecklenburg Counties. The samples were tested by the North Carolina State Laboratory of Public Health from 1998-2012. This study found an average barium concentration of 50 µg/L. No samples exceeded the 2,000 µg/L Primary Maximum Contaminant Level for barium (NCDHHS, 2010a). Page 11-10 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Beryllium Detected Range: 0.10 µg/L — 5.3 µg/L; 22 Detections/108 Total Samples (22/108) 17 One concentration from one shallow, downgradient well east of the ash basin (AL -IS) exceeded the IMAC. �? The majority of detections above the PBTV exist in ash pore water and the shallow flow zone in the southeastern portion of the ash basin and limited downgradient wells east of the ash basin. 07 Sporadic detections in exceedance of the PBTV are present in the deep flow zone primarily beneath the ash basin. 07 Vertical migration below the deep flow zone is not apparent at MSS. Beryllium is a hard, gray metal that is very lightweight. In nature, it combines with other elements to form beryllium compounds. Small amounts of those compounds are naturally present in soils, rocks, and water. Emeralds and aquamarines are gem -quality examples of a mineral (beryl) that is a beryllium compound. Beryllium combines with other metals to form mixtures called alloys. Beryllium and its alloys are used to construct lightweight aircraft, missiles, and satellite components. Beryllium is also used in nuclear reactors and weapons, X-ray transmission windows, heat shields for spacecraft, rocket fuel, aircraft brakes, bicycle frames, precision mirrors, ceramics, and electrical switches (EPRI, 2008b). Most of the beryllium occurring in North Carolina is along the south and southwest sides of the Blue Ridge Mountains. The most notable mines include the Biggerstaff, Branchand, and Poteat mines in Mitchell County; the Old Black mine in Avery County; and the Ray mine in Yancey County. Beryllium forms golden or pale -green well -formed prismatic crystals ranging in size from a fraction of an inch to about 3 inches in diameter. It is generally found near the cores of bodies of pegmatites of moderate size that contain considerable amounts of perthitic microcline. Production has been negligible, and no regular production appears possible. Green beryllium (aquamarine and emerald) was mined commercially many years ago at the Grassy Creek emerald mine and the Grindstaff emerald mine on Crabtree Mountain in Mitchell County. The Ray mine in Yancey County has also produced some golden beryllium and aquamarine (Brobst, 1962). Beryllium -containing minerals are also common in Page 11-11 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra granites and pegmatites throughout the Piedmont; however, those minerals are less common there than in the Blue Ridge Mountains Province (Brobst, 1962). Beryllium is concentrated in silicate minerals relative to sulfides and in feldspar minerals relative to ferromagnesium minerals. The greatest known naturally occurring concentrations of beryllium are found in certain pegmatite bodies. Beryllium is not likely to be found in natural water above trace levels due to the insolubility of oxides and hydroxides at the normal pH range (Brobst, 1962). Boron Detected Range: 53.7 µg/L — 29,500 µg/L; 38 Detections/108 Total Samples (38/108) 47 Concentrations in 24 samples across all flow zones exceeded the 2L. y The highest concentrations at MSS were observed beneath the Ash Landfill (Phase II) in all three flow zones. 07 Concentrations greater than 2L and the PBTVs in all three flow zones indicate plume migration toward the unnamed tributary to Lake Norman east of the ash basin. 167 Increasing concentrations are observed at AB-1BR (downgradient bedrock). Boron is a trace element in the crust, with estimated concentrations ranging from as little as 1 mg/kg in mafic igneous rocks to hundreds of milligrams per kilogram in clay rich rocks (Parker, 1967). It occurs only in the trivalent form (B3+) and is concentrated in sedimentary rocks (Urey & Mem, 1953). This observation indicates that a mechanism exists to concentrate boron in minerals because the oceans could dissolve all of the boron estimated to be present in the crust (Fleet, 1965). Fleet (1965) presents both biogenic and mineralogical processes to account for the preferential concentration of boron in the crust. Boron is a micronutrient (Goldberg, Reactions of boron with soils, 1997) that is concentrated in plant tissue, including the plants from which coal formed. While boron is relatively abundant on the Earth's surface, boron and boron compounds are relatively rare in all geological provinces of North Carolina. Natural sources of boron in the environment include volatilization from seawater, geothermal vents, and weathering of clay -rich sedimentary rocks. Total contributions from anthropogenic sources are less than contributions from natural sources. Anthropogenic sources of boron include agriculture, refuse, Page 11-12 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra coal- and oil -burning power plants, byproducts of glass manufacturing, and sewage and sludge disposal (EPRI, 2005). Because boron is associated with the carbon (fuel) in coal, it tends to volatilize during combustion and subsequently condense onto fly ash as a soluble borate salt (Dudas, 1981). Boron leaches readily (up to 50 percent of total present) and rapidly from fly ash (Cox, Lundquist, Przyjazny, & Schmulbach, 1978). Boron is considered a marker COI for coal ash because boron is rarely associated with other types of industrial waste products. Boron is the primary component of a few minerals, including tourmaline, a rare gem mineral that forms under high temperature and pressure (Hurlbut, 1971). The remaining common boron minerals, including borax that was mined in the Mojave desert, in Boron, California, form from the evaporation of seawater in deposits known as evaporites. For this reason, boron mobilized into the environment will remain in solution until incorporation into plant tissue or adsorption by a mineral. Fleet (1965) describes sorption of boron by clays as a two-step process. Boron in solution is likely to be in the form of the borate ion (B(OH)4-). The initial sorption occurs onto a charged surface. Observations that boron does not tend to desorb from clays indicates that it migrates rapidly into the crystal structure, most likely in substitution for aluminum. Goldberg et al. (1996) determined that boron sorption sites on clays appear to be specific to boron. For this reason, there is no need to correct for competition for sorption sites by other anions in transport models (Goldberg, Forster, Lesch, & Heick, 1996). Goldberg (1997) lists aluminum and iron oxides, magnesium hydroxide, clay minerals, calcium carbonate (limestone), and organic matter as important sorption surfaces in soils (Goldberg, 1997). Boron sorption on oxides is diminished by competition from numerous anions. Boron solubility in groundwater is controlled by adsorption reactions rather than by mineral solubility. Goldberg concludes that chemical models can effectively replicate boron adsorption data over changing conditions of boron concentration, pH, and ionic strength. Cadmium Reported Range: 0.099 µg/L - 0.79 µg/L;16 Detections/108 Total Samples (16/108) 0, No exceedances of 2L in valid groundwater samples. Page 11-13 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra ,67 Wells along the downgradient flow transect show consistent concentrations at or slightly greater than the PBTV. ,67 Sporadic detections above the PBTV in deep groundwater are present beneath the Ash Landfill (Phase II) and southeastern portion of the ash basin. 07 The majority of detections above the PBTV are in ash pore water. Cadmium is generally characterized as a soft, ductile, silver -white or bluish - white metal, and is listed as 64th in relative abundance among the naturally occurring elements. Cadmium is found principally in association with zinc sulfide based ores and, to a lesser degree, as an impurity in lead and copper ores. It is also found in sedimentary rocks at higher levels than in igneous or metamorphic rocks, with the exception of the nonferrous metallic ores of zinc, lead, and copper (WHO, 2011). Cadmium often co-occurs with zinc minerals like sphalerite, and cadmium can substitute in the sphalerite crystal structure during weathering (USGS, 1985). Cadmium is found throughout the environment from natural sources and processes such as the erosion and abrasion of rocks and soils and from singular events such as forest fires and volcanic eruptions (USGS, 1985). Cadmium occurs sporadically in the auriferous parts of the Charlotte and Carolina Slate belts in North Carolina. Cadmium is widespread in the Carolina Slate Belt, but it is found in the Charlotte Belt only near its southeastern boundary. A cluster of cadmium sites marks the mineralized district in the northeast corner of the Carolina Slate Belt, where cadmium was found in all zinc - rich samples. The solubility of cadmium in water is influenced to a large degree by its acidity; suspended or sediment -bound cadmium may dissolve when there is an increase in acidity. In natural waters, cadmium is found mainly in bottom sediments and suspended particles (WHO, 2011). Contamination of drinking water may occur as a result of the presence of cadmium as an impurity in the zinc of galvanized pipes or cadmium -containing solders in fittings, water heaters, water coolers, and taps. Levels of cadmium could be higher in areas supplied with soft water of low pH, as that would tend to be more corrosive in plumbing systems containing cadmium. Cadmium is used in battery production, dye and pigment manufacturing, coatings and Page 11-14 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra plating, plastic production (as a stabilizing agent), nonferrous alloys, and photovoltaic devices (WHO, 2011). In a statistical summary of groundwater quality in North Carolina, the Superfund Research Program at UNC analyzed private well water samples tested by the North Carolina State Laboratory of Public Health from 1998-2010. The average cadmium concentrations were 0.5 µg/L in Stokes County and 0.6 µg/L in Rockingham County. Chloride Reported Range: 1.0 µg/L - 282 µg/L;106 Detections/108 Total Samples (106/108) Concentration from one downgradient sample immediately east of the ash basin (AL-11); deep flow zone) exceeded the 2L. Concentrations from four ash pore water wells exceeded the 2L. 0 The majority of downgradient wells contain concentrations greater than the PBTV in the shallow and deep flow zones. 07 Increasing concentrations are observed at AB-1BR (downgradient bedrock). Chloride is a major ion that occurs widely as a salt of sodium (NaCI), potassium (KCI), and calcium (CaC12). Oceans typically contain about 19,000 mg/L of chloride (Feth, 1981). Elevated levels of chloride may occur in groundwater as a result of sea water intrusion, or erosion of halite (U.S. Geological Survey, 2009). The USEPA has not established an MCL for chloride because it is not known to have adverse effects on human health. A Secondary Maximum Contaminant Level (SMCL) of 250 mg/L has been established for chloride because of taste and corrosive considerations. The taste threshold for chloride depends on the associated cation. A study by Lockhart (1955) found that people detected a salty taste in water at 210, 310, and 222 mg/L from the respective sodium, potassium, and calcium salts. Chloride concentrations above 250 mg/L in drinking water may cause corrosion in water distribution systems (McConnell & Lewis, 1972). Chromium Detected Range: 0.53µg/L - 52.1 µg/L; 62 Detections/108 Total Samples (62/108) -61, Concentrations in one downgradient sample (MW-7S; shallow flow zone) and one sample adjacent to the FGD Residue Landfill (MS-15) exceeded the 2L. The shallow groundwater PBTV is greater than the 2L. Page 11-15 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra O7 Detections above the PBTV are sporadic in ash pore water and the shallow flow zone. ,67 Detections above the PBTV in the deep flow zone are consistent with plume geometry east of the ash basin. ,67 Vertical migration of chromium is not apparent at MSS. Chromium is a blue -white metal found naturally occurring in combination with other substances. It occurs in rocks, soils, plants, and volcanic dust and gases (ASTM, 2014). Background concentrations of chromium in groundwater generally vary according to the media in which they occur. Chromium tends to occur in higher concentrations in felsic igneous rocks (such as granite and metagranite) and ultramafic igneous rocks; however, it is not a major component of the igneous or metamorphic rocks found in the North Carolina Piedmont or the Blue Ridge (Chapman, Cravotta, III, Szabo, & Lindsey, 2013). In a statistical summary of groundwater quality in North Carolina, the Superfund Research Program at UNC analyzed 1,898 private well water samples in Gaston and Mecklenburg counties. The samples were tested by the North Carolina State Laboratory of Public Health from 1998 to 2012. The average chromium concentrations were 5.1 µg/L in Gaston County and 5.2 µg/L in Mecklenburg County. Cobalt Detected Range: 0.1 µg/L - 55.8 µg/L; 74 Detections/108 Total Samples (74/108) y Concentrations in 29 samples across the shallow, deep, and bedrock flow zones exceeded the IMAC. 07 Concentrations greater than the IMAC and PBTV in shallow and deep groundwater are consistent with plume geometry, indicating constituent migration in downgradient areas east of the ash basin prior to discharging to Lake Norman. 101 Vertical migration of cobalt into bedrock groundwater is not apparent. ,67 Increasing concentrations are observed at AB-11) (deep flow zone downgradient). Cobalt is a naturally occurring, silvery grey solid at room temperature and the 33rd most abundant element on earth. Cobalt occurs in the 0, +2, and +3 oxidation states, with +2 being the most common form. In the +2 oxidation state, cobalt Page 11-16 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra forms hydroxide and sulfide solid phases and adsorbs primarily iron and manganese oxides. The factors that affect the speciation and mobility of cobalt in water, sediments, and soil include organic ligands such as humic acids, anions, pH, and redox potential. Cobalt mobilizes in groundwater in acidic, oxidizing environments with relative ease, but does not undergo extensive migration since it combines with the hydroxides of iron and manganese as well as silty minerals (Kim, Gibb, & Howe, 2006). As a cationic species, Coe+ sorption will increase with increasing pH. This is a manifestation of the attraction of Coe+ to mineral surfaces as the surface transitions from a net positive charge to a net negative charge with increasing pH. Cobalt can form a wide range of mineral phases including CoS2 (cattierite), CoSe (freboldite), CoAs2 (safflorite), and CoFe2O4 (cobalt ferrite). Relatively high concentrations of sulfide, selenium, and arsenic are required to form cattierite, freboldite, and safflorite and in general these are not expected, except under some anomalously high concentrations of Co and Se. Cobalt ferrite is predicted to precipitate under conditions with high ferric iron concentrations. Geologically cobalt is normally associated with copper or nickel. A 1984 study to determine the mineral potential of North and South Carolina, cobalt was found to be, "particularly widespread in the Carolina Slate Belt,"and area which covers the lower Piedmont of both North and South Carolina (Griffitts, Whitlow, Siems, Duttweiler, & Hoffman, 1984). Cobalt occurs in clay minerals and substitutes into the pyrite crystal structure. There is also evidence that it is organically bound in coal (Finkelman, 1995). (Izquierdo & Querol, 2012) report limited leaching of cobalt from coal, attributing this observation to incorporation into iron oxide minerals. The concentration of cobalt in surface and groundwater in the United States is generally low — between 1 and 10 parts of cobalt in 1 billion parts of water (parts per billion; ppb) in populated areas. The concentration may be hundreds or thousands times higher in areas that are rich in cobalt -bearing minerals or in areas near mining or smelting operations. In most drinking water, cobalt levels are less than 1 to 2 ppb (USGS, 1973). Cobalt concentration is compared with the IMAC standard since a 2L standard has not been established for this constituent by NCDEQ. Iron Detected Range: 19 µg/L - 55,300 µg/L; 84 Detections/108 Total Samples (84/108) 101 Concentrations in 35 samples across the shallow, deep, and bedrock flow zones exceeded the 2L. Page 11-17 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra ,67 Exceedances of the PBTV are widespread in shallow and deep groundwater across the majority of the ash basin footprint. ,67 Exceedances of the PBTV in downgradient areas east of the ash basin are sporadic in the shallow and deep flow zones. ,67 Exceedances of the PBTV in downgradient areas east of the ash basin are consistent in shallow and deep groundwater at the MW-8 and AB-1 locations. 0 Concentrations at AB-1BR (downgradient bedrock) are increasing over time. Other areas of vertical migration into bedrock are limited to AB-15BR within the ash basin footprint. Iron is a naturally occurring element that may be present in groundwater from the erosion of natural deposits (NCDHHS, 2010b). An (April, 2015) study by NCDEQ DWR (Summary of North Carolina Surface Water Quality Standards 2007-2014) found that while concentrations vary regionally, "iron occurs naturally at significant concentrations in the groundwaters of NC," with a statewide average concentration of 1,320 µg/L. Iron is estimated to be the fourth most abundant element in the Earth's crust at approximately 5 percent by weight (Parker, 1967). Only oxygen (46.60 percent by weight), silicon (27.72 percent), and aluminum (8.13 percent) occur in higher concentrations. Iron occurs in divalent (ferrous, Fee+), trivalent (ferric, Fe3+), hexavalent (Fe 6+), and Fee- oxidation states. Iron is a common mineral -forming element, occurring primarily in mafic (dark colored) minerals that include micas, pyrite (iron disulfide), and hematite (iron oxide), as well as in reddish -colored clay minerals. Clay minerals and pyrite are common impurities in coal. Under combustion conditions in a coal-fired boiler, clay minerals would be dehydrated to mullite or gibbsite, possibly liberating iron, and pyrite would oxidize to hematite or magnesioferrite. Research summarized by Izquierdo and Querol (2012) indicates that iron leaching from coal ash is on the order of one percent of the total iron present due to the low pH required to solubilize iron minerals. Despite the low apparent mobilization percentage, iron is often one of the COIs detected in the highest concentrations in ash pore water. The extent to which iron dissolves in water depends on the amount of oxygen present in the water, and to a lesser extent, its degree of acidity (Stumm & Morgan, 1996). Fe(III) is the dominant oxidation state at high DO concentrations (greater than 1-2 mg/L). As the DO is Page 11-18 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra depleted, Fe(III) is reduced to the more mobile Fe(II) species; pH values also play an important role in iron speciation and mobility. At low pH values, both Fe(II) and Fe(III) are stable and mobile in solution; as pH increases from -2--8, the dissolved concentrations of Fe(III) decreases as Fe(III) forms hydroxide complexes and ultimately precipitates iron bearing mineral phases. Fe (II) is soluble at pH less than 2 at typical surface conditions (25°C and one atmosphere total pressure) (Schmidt, 1962). For this reason, dissolved iron in surficial waters is typically oxidized to Fe(III), resulting in formation of ferric iron oxide flocculation that exhibits a characteristic reddish tint. Manganese Detected Range: 5 µg/L - 5,410 µg/L; 88 Detections/108 Total Samples (88/108) '61' Concentrations in 45 samples across all flow zones exceeded the 2L. ,610 Manganese exceeds the PBTV and 2L in shallow and deep groundwater beneath the ash basin and in limited downgradient areas. ,67 Vertical migration of manganese in concentrations greater than the PBTV for bedrock groundwater is limited to areas beneath the Dry Ash Landfill (Phase II), ash basin, and AB-1BR (increasing concentrations over time). Manganese is a naturally occurring silvery -gray transition metal that resembles iron but is more brittle and is not magnetic. It is found in combination with iron, oxygen, sulfur, or chlorine to form manganese compounds. High manganese concentrations are associated with silty soils, and sedimentary, unconsolidated, or weathered lithologic unit and low concentrations are associated with non - weathered igneous bedrock and soils with high hydraulic conductivity (Gillespie, 2013) (Polizzotto, et al., 2015). Manganese is most readily released to the groundwater through the weathering of mafic or siliceous rocks (Gillespie, 2013). When manganese -bearing minerals in saprolite, such as biotite, are exposed to acidic weathering, the metal can be liberated from the host mineral and released to groundwater. It then migrates through pre-existing fractures during the movement of groundwater through bedrock. If this aqueous -phase manganese is exposed to higher pH in the groundwater system, it will precipitate out of solution. This results in preferential pathways becoming "coated" in manganese oxides and introduces a concentrated source of manganese into groundwater (Gillespie, 2013). Manganese (II) in suspension of silt or clay is commonly oxidized by microorganisms present in soil, leading to the precipitation of manganese minerals (ATSDR, 2012). Roughly 40 percent to 50 percent of North Carolina wells have manganese concentrations higher than Page 11-19 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra the state drinking water standard (Gillespie, 2013). Concentrations are spatially variable throughout the state, ranging from less than 0.01 mg/L to more than 2 mg/L. This range of values reflects naturally derived concentrations of the constituent and is largely dependent on the bedrock's mineralogy and extent of weathering (Gillespie, 2013). Manganese is estimated to be the 12f most abundant element in the crust at 0.1 weight percentage (Parker, 1967). Manganese exhibits geochemical properties similar to iron with Mn'+ Mn6+ Mn4+ Mn3+ Mn2+ and Mn1- oxidation states. Manganese substitutes for iron in many minerals. Similar to iron, manganese leaching from coal ash is limited to less than 10 percent of the total manganese present due to the low pH required to solubilize manganese minerals (Izquierdo & Querol, 2012). Despite the low apparent mobilization percentage, manganese is detected in relatively high concentrations in ash pore water. A 2016 study associated with the NC Division of Water Resources and the NC Geological Survey used laboratory, spectroscopic, and geospatial analyses to demonstrate how natural pedogenetic (soil -forming) and hydrogeochemical processes couple to export manganese from the near -surface to fractured bedrock aquifers with the Piedmont. The study determined that Mn-oxides accumulate near the water table within the saprolite and hydrologic gradients provide a driving force for downward mobility to the fractured -bedrock (Gillespie, 2013). Molybdenum Detected Range: 0.5 µg/L - 9.9 µg/L; 33 Detections/108 Total Samples (33/108) 47 Concentrations of molybdenum exceed the PBTV in ash pore water and shallow groundwater primarily beneath the ash basin, and in limited downgradient wells. 07 One concentration in the deep flow zone (GWA-1D) was greater than the PBTV. 07 Molybdenum is detected in bedrock groundwater at concentrations greater than the PBTV beneath the central portion of the ash basin as well as beneath the Dry Ash Landfill (Phase II). Molybdenum is a trace element that exists predominantly as Mo(IV) and Mo(VI). As a free metal, it is silvery gray, although it does not occur in this form in nature. It is mined for use in alloys. Molybdenum commonly forms oxyanions in groundwater that are affected by redox and pH (Ayotte, Gronbert, & Apodaca, Page 11-20 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 2011). Molybdenum has been observed to leach less from coal cleaning rejects in acidic than neutral conditions, unlike many other metals (Jones & Ruppert, 2017). Molybdenum has been shown to become more mobile in procedures that use deionized water as a leachant, which may be similar to actual disposal conditions unlike many other coal ash elements that are more mobile when subjected to weak acid (Jones & Ruppert, 2017). Nickel Detected Range: 0.5 µg/L — 48 µg/L; 82 Detections/108 Total Samples (82/108) 47 No concentrations in groundwater greater than the 2L of 100 µg/L. 47 Concentrations of nickel exceed the PBTV primarily in pore water, shallow, and deep groundwater beneath the central portion of the ash basin. 167 Vertical migration of nickel is not apparent at MSS. One concentration in bedrock (AB-513R) exceeded the PBTV of 9.9 µg/L. Nickel, like chromium and cobalt, is a base metal that exhibits geochemical properties similar to iron and manganese. It occurs as a divalent and trivalent ion. Finkelman (1995) reported evidence that nickel occurs in association with organic matter in coal. Goodarzi et al. (2008) found nickel in the oxide form in both coal and ash. Kim and Kazonich (2004) report association of nickel with the silicate (likely clay) fraction in coal. Solubility of nickel from ash is pH sensitive, with mobilization greatest under highly acidic conditions (Izquierdo & Querol, 2012). Selenium Detected Range: 0.54 µg/L —109 µg/L; 25 Detections/108 Total Samples (25/108) 41' Concentrations in five samples from shallow and deep groundwater beneath the Dry Ash Landfill (Phase II) exceeded the 2L. y One concentration from GWA-11S exceeded the 2L and may represent downgradient migration east of the ash basin toward the unnamed tributary to Lake Norman. 47 Concentrations of selenium are greater than the PBTV in ash pore water and shallow groundwater primarily beneath the ash basin and in shallow wells east of the ash basin. Page 11-21 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra ,67 Concentrations in deep and bedrock groundwater greater than the PBTVs are limited to the Dry Ash Landfill (Phase II) and downgradient areas east of the ash basin toward unnamed tributary. Selenium is a semi -metallic gray metal that commonly occurs naturally in rocks and soil. It is common to find trace amounts of selenium in food, drinking water, and air -borne dust. Over geologic time, selenium has been introduced to the Earth's surface and atmosphere through volcanic emissions and igneous extrusions. Weathering and transport partition the element into residual soils, where it is available for plant uptake, or to the aqueous environment, where it may remain dissolved, enter the aquatic food chain, or redeposit within a sedimentary rock such as shale (EPRI, 2008). Groundwater containing selenium is typically the result of either natural processes or industrial operations. Naturally, selenium's presence in groundwater is from leaching out of selenium -bearing rocks. Most soils contain between 0.1 and 2 mg/kg, but soils originating from the Upper Cretaceous marine sedimentary rocks (shale) show elevated selenium concentrations ranging from 0.6 to 103 mg/kg (EPA, 2007). Anthropogenically, selenium is released as a function of the discharge from petroleum and metal refineries and from ore mining and processing facilities. Ore mining may enhance the natural erosive process by loosening soil, creating concentrations in erodible tailings piles, and exposing selenium -bearing rock to runoff (Martens, 2002) (USEPA, 2017c). Selenium exists as oxyanionic species under oxidizing conditions as selenite (Se032-) or selenate(SeO42-). Similar to arsenic's behavior, both selenite and selenate sorb to mineral surfaces (primarily iron oxides)( (EPA, 2007) (Strawn, Doner, Zavarin, & McHugo, 2002) (Zawislandski & Zavarin, 1996). Thus, oxidation of Se(IV) to Se(VI) could influence the mobility of selenium in the subsurface. However, both are subject to competition from other oxoanions such as phosphate and sulfate. Selenium is also readily reduced to zero valent Se° which is relatively stable under mildly reducing conditions. The formation of reduced Se( -II) species can have an impact on other metal ion concentrations such as through the precipitation of CoSe(s). In a statistical summary of groundwater quality in North Carolina, the Superfund Research Program at UNC analyzed 399 private well water samples in Stokes and Rockingham counties from 1998-2010. The values ranged from 2.5 Page 11-22 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra to 26 µg/L, and no samples exceeded the 50 µg/L primary MCL for selenium (NCDHHS, 2010). The mean concentration in both counties was 2.7 µg/L. Strontium Detected Range: 6.6 µg/L—10,200 µg/L;107 Detections/108 Total Samples (107/108) y The majority of concentrations detected in shallow, deep, and bedrock flow zones beneath the ash basin exceeded the respective PBTVs. y Downgradient concentrations in exceedance of the PBTVs in shallow, deep, and bedrock groundwater indicate constituent migration towards the unnamed tributary to Lake Norman. Strontium is a soft silver -yellow alkaline earth metal. It is highly chemically reactive and forms a dark oxide layer when it interacts with air. It is chemically similar to Ca and replaces Ca or K in igneous rocks in minor amounts. Strontium is generally present in low concentrations in surface waters but may exist in higher concentrations in some groundwater (Hem, 1985). Strontium is present as a minor coal and coal ash constituent. Strontium has been observed to leach from coal cleaning rejects more in neutral conditions than acidic, unlike many other metals (Jones & Ruppert, 2017). It has been shown to behave conservatively in surface waters downstream of coal plants (Ruhl, et al., 2012). Su/fate Detected Range: 0.12 µg/L —1,480 µg/L; 87 Detections/108 Total Samples (87/108) y Concentrations in four samples from either the shallow or deep flow zones beneath the Ash Landfill (Phase II) exceeded the 2L. 07 One exceedance of the 2L at MW-14S, and exceedance of PBTV in MW- 14D (200 mg/L), may indicate downgradient migration towards the unnamed tributary to Lake Norman. ,610 Exceedances of 2L are consistent in ash pore water. 01 Exceedances of the PBTVs in deep and bedrock groundwater are consistent beneath the central and eastern portions of the ash basin. 07 Concentrations are greater than the PBTV in downgradient, deep groundwater wells east of the ash basin. Page 11-23 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Sulfate is a naturally occurring substance found in minerals, soil, and rocks. It is present in ambient air, groundwater, plants, and food. Primary natural sources of sulfate include atmospheric deposition, sulfate mineral dissolution, and sulfide mineral oxidation. The principal commercial use of sulfate is in the chemical industry. Sulfate is discharged into water in industrial wastes and through atmospheric deposition (USEPA, 2003). Anthropogenic sources include coal mines, power plants, phosphate refineries, and metallurgical refineries. Sulfate has a SMCL, and no enforceable maximum concentration set by the USEPA. Ingestion of water with high concentrations of sulfate may be associated with diarrhea, particularly in susceptible populations, such as infants and transients (USEPA, 2012). However, adults generally become accustomed to high sulfate concentrations after a few days. It is estimated that about 3 percent of the public drinking water systems in the United States may have sulfate concentrations of 250 mg/L or greater (Miao, Brusseau, Carroll, & others, 2012). Sulfate is on the list of enforced regulated contaminants that may cause cosmetic effects or aesthetic effects in drinking water (USEPA, 2017a). TDS Detected Range: 26 mg/L -1,950 mg/L; 104 Detections/108 Total Samples (104/108) '41' Concentrations in nine samples from either the shallow or deep flow zones beneath the ash basin exceeded the 2L. '67 Four samples from either the shallow or deep flow zones in downgradient wells east of the ash basin exceeded the 2L. 07 Increasing concentrations are observed at AB-1BR, indicating vertical migration in this downgradient area east of the ash basin. 410 Exceedances of 2L are consistent in ash pore water. 47 Concentrations greater than the PBTVs are consistent in all flow zones beneath the majority of the ash basin footprint and downgradient areas east of the ash basin. Groundwater contains a wide variety of dissolved inorganic constituents as a result of chemical and biochemical interactions between the groundwater and the elements in the soil and rock through which it passes. TDS mainly consist of dissolved cation and anion particles (e.g., calcium, chlorides, nitrate, phosphorus, iron, sulfur, and others) in varying concentrations. TDS is measured by weighing Page 11-24 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra the solid residue left after evaporating a measured volume of sample to dryness (Freeze & Cherry, 1979). TDS is therefore a measure of the total amount of dissolved ions in the water, but does not identify specific constituents or explain the nature of ion relationships. The ions listed below are referred to as the major ions as they make up more than 90 percent of the TDS in groundwater. '67 Sodium (Na+) 411 Magnesium (Mg2+) '410 Calcium (Caz+) 1610 Chloride (Cl-) '67 Bicarbonate (HCO3-) 167 Sulfate (SO4 2-) In the CCR Rule, the USEPA listed TDS as an indicator constituent (along with boron, calcium, chloride, fluoride, pH, and sulfate). USEPA defines indicator constituents as those that are present in CCR and would rapidly move through the surface layer, relative to other constituents, and thus provide an early detection of whether contaminants are migrating from the CCR unit (USEPA, April 2015). Thallium Reported Rane: 0.12 µg/L - 0.48 µg/L; 8 Detections/108 Total Samples (8/108) 410 Concentrations in two shallow groundwater samples (AB-1S, MW-7S) exceeded the IMAC. ,67 Exceedances of the IMAC are predominantly in ash pore water. ,67 One concentration in deep groundwater below the ash basin (AB-51)) exceeded the PBTV. ,67 Vertical and horizontal migration of thallium is not apparent at MSS. Pure thallium is a soft, bluish white metal that is widely distributed in trace amounts in the Earth's crust. In its pure form, it is odorless and tasteless. It can be found in pure form or mixed with other metals in the form of alloys. It can also be found combined with other substances such as bromine, chlorine, fluorine, and iodine to form salts (Institute, 2008c). Page 11-25 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Traces of thallium naturally exist in rock and soil. As rock and soil is eroded, small amounts of thallium end up in groundwater. In a USGS study of trace metals in soils, the variation in thallium concentrations in A (i.e., surface) and C (i.e., substratum) soil horizons was estimated across the United States. The overall thallium concentrations range from less than 0.1 mg/kg to 8.8 mg/kg. Thallium is compared to an IMAC (0.2 µg/L) since no 2L standard has been established for this constituent. In a study by the Georgia Environmental Protection Division of the Blue Ridge Mountain and Piedmont aquifers, 120 sites were sampled for various constituents. Thallium was not detected at any of these sites (Method Reporting Limit (MRL) =1 µg/L) (Donahue & Kibler, 2007). Vanadium Detected Range: 0.374 µg/L — 25.6 µg/L; 80 Detections/108 Total Samples (80/108) 07 One sample of deep groundwater beneath the ash basin (AB-7D) exceeded both the PBTV and IMAC. 07 Concentrations from all locations in each flow zone across the MSS site, with very few exceptions, are reported in exceedance of the IMAC. PBTVs for all flow zones are greater than the IMAC, and may still not be representative of regional groundwater values. Vanadium is estimated to be the 22nd most abundant element in the crust at 0.011 weight percent (Parker, 1967). Vanadium occurs in four oxidation states: V+s, V+41 V+3, and V+2. It is a common trace element in both clay minerals and plant material. The National Uranium Resource Evaluation (NURE) program was initiated by the Atomic Energy Commission in 1973 with a primary goal of identifying uranium resources in the United States (Smith, 2016). The Hydrogeochemical and Stream Sediment Reconnaissance program (initiated in 1975) was one component of NURE. Planned systematic sampling of the entire United States began in 1976 under the responsibility of four Department of Energy national laboratories. Samples were collected from 5,178 wells across North Carolina. Of these, the concentration of vanadium was equal to or higher that the former IMAC of 0.3 µg/L in 1,388 well samples (27 percent). Page 11-26 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 11.2 Pending Investigation(s) Supplemental data collection to support groundwater modeling and long-term monitoring is anticipated to support the CAP process. Additional metal oxy-hydroxide phases of iron (HFO) and aluminum (HAO) data are anticipated to support geochemical modeling conducted as part of the CAP. A Solid Phase -Groundwater Interactions Data Map was completed to indicate where solid and aqueous phase data exist and where additional samples should be collected to address data needs for the updated geochemical model (Figure 11-121). Data distribution and results of information related to soil/groundwater/COI interaction is presented on Figure 11-121. Additional HFO samples are proposed to be collected for the following locations within the screened interval (* indicates an ash pore water well): 101 Western Transect- AB-5S*, AB-51), AB-SBR, AB-3S, AB-31), AB-21), MW-9S, and MW-9D ,67 Central Transect- AB-12S*, AB-12D, AB-12BR, AB-9D, AB-9BR, AB -ID, and AB- 1BR 47 Eastern Transect- CCR-13S, CCR-13D, GWA-11S, GWA-11D, GWA-11BR, and GWA-15S In accordance with 15A NCAC 02L.0106(k)(5) and (1)(6), the CAP may include an evaluation of whether groundwater migrating downgradient of the ash basin may have contaminant concentrations that would result in violations of standards for surface water. One means of accomplishing this objective is to collect surface water samples to document existing conditions. Another method will be to conduct groundwater to surface water modeling. It is anticipated that documenting current conditions through the collection of the additional surface water samples in NPDES permitted seeps and Lake Norman will be coordinated with NCDEQ guidance. Additional wells screened in the shallow flow zone beneath the ash basin, where a significant thickness of regolith was not previously screened, may be warranted to improve groundwater modeling of source conditions and to monitor effectiveness of the selected corrective action remedy. Pending review of data from recently installed deep bedrock wells, further assessment may be to refine vertical extent of groundwater impacts. Results from the hydrogeological assessment of MSS, summarized in this section, are primary components of the Site Conceptual Model (SCM). Increasing concentration trends apparent in AB-1BR will also be further evaluated to assess the possibility of preferential downward migration due to well construction rather than increasing Page 11-27 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra concentations from the source area. Recently installed AB-1BRL will support this evaluation with additional data for vertical delineation at this location. Page 11-28 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 12.0 RISK ASSESSMENT A baseline human health and ecological risk assessment was performed as a component of CAP Part 2 for MSS (Appendix I, HDR 2016a). The 2016 risk assessment characterized potential risks to humans and wildlife exposed to coal ash constituents present in environmental media for the purpose in aiding corrective action decisions. Implementation of corrective action is intended to achieve future Site conditions protective of human health and the environment, as required by CAMA. This update to the risk assessment evaluates groundwater and surface water results collected since the 2016 risk assessment in order to confirm or update risk conclusions (Section 7 of Appendix I) in support of remedial action. Data used in the 2016 risk assessment included groundwater, surface water, sediments, AOW water and AOW soil collected from February 2011 through October 2015 (Section 3 of Appendix I). This risk assessment update uses sample data presented in Attachment A of the 2016 risk assessment (HDR 2016a) along with groundwater and surface water data presented in Appendix B of this report. AOW locations are outside the scope of this risk assessment because AOWs, wastewater, and wastewater conveyances (effluent channels) are evaluated and governed wholly separate in accordance with the NPDES Program administered by NCDEQ DWR. This process is on -going in a parallel effort to the CSA and subject to change. No new sediment or soil samples, other than for background evaluations, have been collected that are applicable to the risk assessment; therefore, risk estimates associated with sediment and soil have not changed. As part of the 2016 risk assessment, human health and ecological conceptual site models (CSMs) were developed to guide identification of exposure pathways, exposure routes, and potential receptors for evaluation in the risk assessment. Section 2 of Appendix I provides a detailed description of the exposure pathways, exposure routes, and potential receptors considered in the 2016 assessment. The CSMs (Figures 2-3 and 2-4 of Appendix I) describe the sources and potential migration pathways through which groundwater beneath the ash basin may have transported coal ash -derived constituents to other environmental media (receiving media) and, in turn, to potential human and ecological receptors. Exposure scenarios and exposure areas were presented in detail in Sections 2 and 5 of the 2016 CAP Part 2 risk assessment (Appendix I). Page 12-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra This risk assessment update included the following: ,67 Identification of maximum constituent concentrations for groundwater and surface water; 101 Inclusion of new groundwater and surface water data to derive overall average constituent concentrations for exposure areas; 167 Comparison of new maximum constituent concentrations to the risk assessment human health and ecological screening values (ESV); 07 Comparison of new maximum constituent concentrations to human health Risk - Based Concentrations (RBC); and 07 Incorporation of new maximum constituent concentrations into wildlife Average Daily Dose (ADD) calculations for comparison to ecological Toxicity Reference Values (TRVs). Evaluation of new groundwater and surface water data and the influence on the 2016 risk assessment conclusions are summarized below by exposure areas (Figure 12-1) at MSS. 12.1 Human Health Screening Summary On -Site Groundwater Groundwater samples were collected from 74 locations: MW-6D through MW-14D, GWA-1S through GWA-9BR, AB-1S through AB-21D, AL-1S through AL-41) and OB-1. These wells were evaluated because they represent the potential worker exposure area as determined in the 2016 risk assessment. Sampling locations BG-1S through BG-31) and MW-4 are considered background locations and are not included in the risk assessment. The 2016 risk assessment found no evidence of unacceptable risks under the commercial/industrial worker or construction worker scenarios exposed to on -Site groundwater (Section 5 of Appendix I). Groundwater analytical results are included in Appendix B, Table 1. Groundwater data collected since the 2016 risk assessment were compared to USEPA's human health regional screening levels (RSLs) and RBCs from the 2016 risk assessment. Inclusion of new data resulted in maximum concentrations greater than the data presented in the 2016 assessment for boron, aluminum, antimony, arsenic, barium, cadmium, hexavalent chromium, total chromium, cobalt, copper, iron, mercury, molybdenum, nickel, selenium, thallium, vanadium, and zinc (Appendix B, Table 1). The new maximum concentrations were less than the respective screening value or RBC for all constituents Page 12-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra measured; therefore, no evidence of unacceptable risks to workers potentially exposed to on -site groundwater was identified. Surface Water Surface water samples used in the 2016 risk assessment were collected from six locations at the site: SW-6 (S-1) and SW-9 north of the active ash basin, SW-10 and SW- 11 along the shore of Lake Norman, and SW-7 and SW-8 in the northwestern area of the site adjacent to the PV Structural Fill (Figure 12-1). Sample locations SW-1 through SW- 5 were collected from within the active ash basin and were not include in the risk assessment. For purposes of this risk assessment update, SW-6 was excluded as a surface water location because it was re -identified as a permitted seep (S-1), SW-10 was excluded because the location is adjacent to NPDES outfall MSSW002, and SW-7 and SW-8 were excluded because the locations are surface water drainage features upgradient of the ash basin. Therefore, surface water exposure is evaluated using off - site Lake Norman samples SW-9, SW-11, and SW-12. Surface water data is included in Appendix B, Table 2 of this report. The 2016 risk assessment concluded that on -site surface water posed no potential unacceptable risks under the worker and trespasser exposure scenarios (Section 5 of Appendix I). Surface water data collected since the 2016 risk assessment from SW-9, SW-11, and SW-12 were compared to USEPA's human health RSLs and RBCs established for the most conservative exposure scenario/receptor from the 2016 risk assessment. New maximum concentrations of aluminum (418 µg/L), hexavalent chromium (0.9 µg/L), cobalt (8.7 µg/L), iron (541 µg/L), and manganese (511 µg/L) exceeded respective RSLs but did not exceed respective RBCs under the worker and trespasser exposure scenarios. As detailed in Sections 5 and 7 of Appendix I, the 2016 risk assessment estimated potential risks to recreational and subsistence fishermen due to modeled cobalt concentrations in fish tissue. The 2016 Hazard Quotient (HQ) was 7.2 for the recreational fisher and 210 for the subsistence fisher. Using the new maximum cobalt concentration in surface water of 8.7 µg/L, the revised HIs are 2.6 and 79, respectively. The use of highly conservative assumptions regarding transfer of cobalt from surface water to fish tissue and assumed fish ingestion rates likely combine to overestimate risks under the fisher exposure scenarios. The average concentration of cobalt detected in Lake Norman (1.29 µg/L) is less than the recreational fisher RBC of 3.4 µg/L. Lake Norman has both a statewide and county fish consumption advisory issued by the Occupational and Environmental Epidemiology Branch, N.C. Division of Public Health (http://epi.publichealth.nc.gov/oee/fish/advisories.html). In addition, the area of Lake Page 12-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Norman in question is restricted from boat access, so subsistence fishing is unrealistic, and recreational fishing is limited. Cobalt occurring in Lake Norman is not anticipated to pose unacceptable risks to humans under the recreational and subsistence fishing exposure scenarios. 12.2 Ecological Screening Summary The 2016 risk assessment included evaluation of potential ecological risk by exposure area (Section 6 of Appendix I). Two ecological exposure areas (Figure 12-1) were established based on available data and include the following: 07 Ecological Exposure Area 1,located north, east, and south of the active ash basin along the shore of Lake Norman and an unnamed tributary east of basin; 47 Ecological Exposure Area 2, located in the northwestern area of the site adjacent to the PV Structural Fill. The 2016 ecological risk assessment resulted in potential risks to the piscivorous birds exposed to selenium (HQ =1) and vanadium (HQ =11) in Exposure Area 1. The risk estimates were primarily attributable to estimated tissue concentrations modeled from surface water and sediment concentrations. No unacceptable risks were reported for Exposure Area 2. The 2016 ecological risk assessment included both surface water and AOW analytical data (Section 3 of Appendix I) combined for the aquatic exposure scenarios. As stated previously, AOW locations are outside the scope of this risk assessment update. For this updated risk assessment, surface water samples representing Exposure Area 1 include SW-9, SW-11, and SW-12. Exposure Area 2 is excluded from this risk assessment update based on its location upgradient of the ash basin. Surface Water- Exposure Area I Results from SW-9, SW-11, and SW-12 surface water samples are included in Appendix B, Table 2. New maximum concentrations of aluminum (418 µg/L), cadmium (0.4 µg/L), copper (7.4 µg/L), and manganese (511 µg/L) were detected and exceeded respective ESVs used in the 2016 risk assessment. No other constituents had new maximum concentrations that exceeded ESVs. The detected concentrations did not result in ADD estimates that exceeded TRVs; therefore, there is no evidence of potential unacceptable risks to wildlife exposed to surface water from the SW-9, SW-11, and SW- 12 sample locations. Page 12-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 12.3 Private Well Receptor Assessment Update An independent study was conducted that evaluated three private well groundwater datasets: y Data collected by NCDEQ within a half -mile of the Duke Energy facility properties where ash basins are located y Data collected by NCDEQ from areas identified as background locations ,67 Data collected from locations within 2 to 10 miles of each Duke Energy facility in North Carolina that manages or has managed coal ash on -site The objectives of the study were to provide a review of private well data collected by NCDEQ with respect to regulatory and risk -based screening levels and to review background concentrations of constituents in groundwater in the vicinity of Duke Energy facilities. The independent study concluded that the presence of hexavalent chromium and vanadium in private wells could likely be attributed to background conditions and did not indicate impact from constituents derived from coal ash (Haley & Aldrich, 2015). Results from off -Site water supply well samples collected in 2015 were re-evaluated by comparing constituents without published 2L standards to North Carolina Department of Health and Human Services (NCDHHS) Screening Levels, federal drinking water MCLs, and tap water RSLs. Additionally, data was compared to the bedrock PBTVs recently developed for the site. Parameters that exceed 2L or IMAC standards at one or more locations include pH, TDS, cobalt, iron, lead, manganese, vanadium and zinc. Cobalt, iron, manganese, vanadium and zinc concentrations were below EPA's RSL. All but two detected concentrations of lead were below EPA's RSL. The concentrations of parameters pH, cobalt, vanadium, and manganese are comparable to the bedrock PBTVs. Aluminum, hexavalent chromium, magnesium, and alkalinity do not have 2L or IMAC standards; however, all results are less than bedrock PBTVs. Molybdenum does not have a 2L or IMAC standard; however, all but one result was less than the PBTV. Molybdenum concentrations are all less than the NCDHHS screening level and EPA's RSL. Based on the bedrock groundwater flow direction at the Site (as presented in Figures 6-7 and 6-10 and discussed in Section 10), water supply wells near the Site are located upgradient and not influenced by groundwater migration from the ash basins. 12.4 Risk Assessment Update Summary An update to the 2016 human health and ecological risk assessment was conducted. There is no evidence of unacceptable risks to humans exposed to groundwater on -site at Page 12-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra MSS. The 2016 risk assessment identified potential risks under a hypothetical recreational and subsistence fisher exposure scenario and to the great blue heron (for selenium and vanadium). This risk assessment update supports that the fisher risks were overestimated based on exposure and modeled fish tissue uptake assumptions. Surface water concentrations of selenium and vanadium did not exceed ESVs, and thus do not pose risks to the piscivorous birds. Limited potential for unacceptable risks to vanadium and selenium for piscivorous birds (primarily attributable to sediment and modeled tissue concentrations) in Exposure Area 1 is consistent with the 2016 assessment. This update to the human health and ecological risk assessment supports a risk classification of "Low" for groundwater -related considerations. Page 12-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 13.0 GROUNDWATER MODELING RESULTS The Site -specific groundwater flow and transport models and the Site -specific geochemical models are currently being updated for use in the CAP. Groundwater flow, transport, and geochemical models simulate movement of COIs through the subsurface to support the evaluation and design of remedial options at the Site. The existing data inventory requested in section 4.3 of "Monitored Natural Attenuation For Inorganic Contaminants in Groundwater: Guidance for Developing Corrective Action Plans Pursuant to NCAC 15A.0106(l)" (NCDEQ-DWR, 2017) is summarized in Table 11-1. The models will provide insights into: 01 COI mobility: Geochemical processes that affect chemical precipitation, adsorption, and desorption onto solids will be simulated based on lab data and thermodynamic principles. That simulation will support the prediction and mobility in groundwater. ,67 COI movement: Simulations of the groundwater flow system will be combined with estimates of source concentrations, sorption, effective porosity, and dispersion. That combination will support the prediction of the paths, extent, and rates of constituent movement at the field scale. ,67 Scenario Screening: The flow, transport, and geochemical models will be adjusted to simulate how various ash basin closure design options and groundwater remedial technologies will affect the short-term and long-term distribution of COIs. 01 Design: Model predictions will be used to help design basin closure and groundwater corrective action strategies in order to achieve compliance with 2L. The groundwater flow model linked with the transport model will be used to establish transport predictions that best represent observed conditions at the Site particularly for the constituents, such as boron, that are negligibly affected by geochemical processes. Information from the geochemical model will provide insight into the complex processes that influence constituent mobility. Once the flow, transport, and geochemical models for the Site accurately reproduce observed Site conditions, they will be used as predictive tools to evaluate the conditions that would result from various remedial options for basin closure. The updated CAP will further discuss the purpose and scope of both the groundwater and geochemical models. The CAP will detail model development, calibration, assumptions, and limitations. The CAP will also include a detailed remedial option Page 13-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra evaluation, based on observed conditions and the results of predictive modeling. The evaluation of the potential remedial options will include comparisons of predictive model results for long-term source concentration and plume migration trends toward potential receptors. The model predictions will be used in combination with other evaluation criteria to develop the optimal approach for basin closure and groundwater remediation. The following subsections provide a brief summary of modeling efforts completed to date at MSS and present a description of the updated models to be submitted in the updated CAP. 13.1 Summary of Flow and Transport Model Results CAP Part 1 Model The initial groundwater flow and transport model was developed by HDR, Inc. in conjunction with University of North Carolina at Charlotte (UNCC) to gain an understanding of COI migration after closure of the ash basin at MSS. The initial groundwater model in the CAP Part 1 (HDR, December 7, 2015b) included a calibrated steady-state flow model of June/July 2015 conditions; a calibrated historical transient model of constituent transport to match June/July 2015 conditions; and three potential basin closure scenarios. Those basin closure simulation scenarios included: ,67 No change in site conditions (basin remains open, as is) 101 Cap -in -Place (with existing ash configuration) y Ash removal (excavation of ash) Both the initial model presented in the CAP Part 1 and the revised model presented in the CAP Part 2 used antimony, arsenic, barium, beryllium, boron, chloride, chromium, hexavalent chromium, cobalt, selenium, sulfate, thallium, and vanadium as primary modeling constituents. The remedial alternative evaluation simulations in both models were run to a total time of 250 years. Appendix C of the CAP Part 1 (HDR, December 7, 2015b) contains the report describing the development and results from this modeling. CAP Part 2 Model The revised model in the CAP Part 2 (HDR, March 3, 2016a) included a calibrated steady-state flow model of June/July 2015 conditions and a calibrated historical transient model of constituent transport to June/July 2015 conditions. In addition, CSA data was used to refine the layer assignments, hydraulic conductivity values, and recharge rates within the model. Four water supply wells, which were located within the model domain, were included in the revised model. Page 13-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Two potential ash basin closure scenarios were considered in the revised model. Those ash basin closure simulation scenarios included: ,67 No change in site conditions (basin remains open, as is) 101 Cap -in -Place (with existing ash configuration) Appendix B of the CAP Part 2 (HDR, March 3, 2016a) contains the report describing the development and results from this modeling. Updated CAP Mode/ The flow and transport model is currently being updated as a part of the updated CAP and will include: development of a calibrated steady-state flow model that includes data available through the fourth quarter of 2017; development of a historical transient model of constituent transport; and predictive simulations of basin closure plus groundwater corrective action scenarios. The updated flow and transport model will consider boron and additional COIs that are negligibly affected by geochemical processes. Predictive remedial scenarios will have simulation times that will continue until modeled COI concentrations are below the 2L standard at the compliance boundary. The distribution of recharge, location of drains, and distribution of material will be modified to represent the following different basin closure options in the revised/updated models: Complete Excavation with passive remediation Complete Excavation with active remediation Cap -in -place with passive remediation (MNA) 0 Cap -in -place with active remediation The results of these simulations will be included as part of the updated CAP submittal. 13.2 Summary of Geochemical Model Results The MSS geochemical model investigates how variations in geochemical parameters affect movement of constituents through the subsurface. The geochemical SCM will be updated as additional data and information associated with Site constituents, conditions, or processes are developed, and will be updated based on discussions and comments from NCDEQ. The geochemical modeling approach presented in the following subsections was developed using laboratory analytical procedures and Page 13-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra computer simulations to understand the geochemical conditions and controls on groundwater concentrations in order to predict how remedial action and/or natural attenuation may occur at the Site and avoid unwanted side effects. The updated geochemical model will be presented in the updated CAP. CAP Part 2 Geochemical Model The geochemical model presented in the CAP Part 2 (HDR, March 3, 2016a) included: 101 Eh -pH (Pourbaix) diagrams showing potential stable chemical phases of the aqueous electrochemical system, calibrated to encompass pH and redox potentials at the Site y Sorption model in which the aqueous and surface speciation of constituents was modeled using the USGS geochemical modeling program PHREEQC y Simulations of the anticipated geochemical speciation that would occur for each COI in the presence of adsorption to soils and in response to changes in Eh and pH ,610 Attenuation calculations in which the potential capacity of aquifer solids to sequester constituents of interest were estimated Details on the development and results of the CAP Part 2 geochemical model are presented in Appendix C of the CAP Part 2 (HDR, March 3, 2016a). Updated Geochemical Model Development The geochemical model in the updated CAP will contain refinements based on updated data and on comments and discussions with NCDEQ. The updated geochemical modeling will use hydraulically significant flow transects to investigate the geochemical controls on groundwater concentrations in and around source areas at the Site for selected constituents (Figure 11-121). The model will compare trends in the concentrations of each COI along transects with the model output to verify that the conceptual and qualitative models can predict COI behavior. Then the model will be used to evaluate the potential impacts of remediation activities. The model will relate the COI concentrations observed in groundwater along flow transects to key geochemical parameters influencing constituent mobility (i.e., Eh, pH, and saturation/solubility controls). The hydraulically significant flow transects that will be used in the geochemical model were described in Section 11.1. These flow transects were determined by considering: Page 13-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 410 Associated source areas ,0 Plume movement focusing on the centerline of the area(s) of impact (i.e., the plume(s)) 101 Geologically derived flow zones (geozones) 47 Hydraulic gradients 17 Downgradient receptors IC7 Available data The updated geochemical model will develop parameters for each aquifer or geozone by considering the bulk densities, porosities, and hydraulic gradients used in the fate and transport model. These parameters are used to constrain the sorption site concentrations in the model input and will be incorporated in the 1-D ADVECTION model to accompany the capacity simulations. The objective of these capacity simulations is to determine the mass balance on iron and aluminum sorption sites when simulating flow through a fixed region. Groundwater concentrations and initial solid phase iron and aluminum concentrations will be fixed based on Site -specific data. Thus, the model will be able to simulate the stability of the HFO and HAO phases assumed to control constituent sorption. The updated geochemical model report will include a Site -specific discussion of: ,610 The model description ,67 The purpose of the geochemical model 'CIO Modeling results compared with observed conditions -61, COI sensitivity to pH, Eh, and iron/aluminum oxide content 1610 Model limitations The updated geochemical modeling will also present multiple methods of determining constituent mobility at the Site. Aqueous speciation, surface complexation, and solubility controls will be presented in the revised report. These processes will be modeled using: y Pourbaix diagrams created with the Geochemist Workbench v10 software using Site -specific minimum and maximum constituent concentrations 07 PHREEQC's combined aqueous speciation and surface complexation model and the 1-D ADVECTION function to gain a comprehensive understanding of Page 13-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra current geochemical controls on the system and evaluate how potential changes in the geochemical system might affect constituent mobility in the future Data Inventory To appropriately characterize and model the geochemical behavior of constituents in the subsurface requires a significant amount of data collection and analysis. In addition to hydraulic controls and groundwater data, it is important to properly evaluate the physical and chemical properties of the solid phase aquifer material in contact with groundwater. In preparation for the updated geochemical model to be presented with the CAP, a data inventory was conducted to determine where additional data needs exist (Table 11-1). As part of the data inventory, a Solid Phase -Groundwater Interactions Data Map (Figure 11-121) was completed to indicate where solid and aqueous phase data exist and where additional samples should be collected to address data needs for the updated geochemical model. This data map includes the following: 47 The boron plume within the shallow aquifer (boron is used as a proxy for the general area of impact) 47 The geochemical flow transects 07 The soil water pairs (i.e., where solid phase total metals analysis are available within a screened interval of a well) 101 The currently available data and locations for HFO/HAO, Kd (Table 13-1), and hydraulic conductivity (k) values 07 Proposed locations for additional HFO/HAO samples to be used in the updated geochemical model as described in Section 11.2 The requested data inventory is summarized in Table 11-1. Page 13-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 14.0 SITE ASSESSMENT RESULTS A site conceptual model (SCM) is an interpretation of processes and characteristics associated with hydrogeologic conditions and constituent interactions at the Site. The site assessment results provide the information to evaluate distribution of constituents with regard to site -specific geological/hydrogeological properties. 14.1 Nature and Extent of Contamination The site assessment described in this CSA presents the results of investigations required by CAMA and 2L regulations. Ash sluiced to, and accumulated within, the basin is determined to be a source of impacts to groundwater. The site assessment investigated the Site hydrogeology, determined the direction of groundwater flow from the ash basin, and determined the horizontal and vertical extent of impacts to groundwater and soil sufficient to proceed with preparation of a CAP. Constituents of Interest Soil and groundwater beneath the ash management areas and downgradient areas east of the waste boundary have been influenced by ash storage at MSS. COIs identified as being associated with MSS ash management areas include antimony, arsenic, barium, beryllium, boron, cadmium, chloride, chromium, cobalt, iron, manganese, molybdenum, nickel, selenium, strontium, sulfate, TDS, thallium and vanadium. Constituent migration in groundwater occurs at variable rates depending on a number of physical and chemical conditions and properties (e.g., constituent sorption properties, redox state, pH, hydraulic conductivity, etc.). Some COIs, such as boron, readily solubilize and migrate with minimal retention. In contrast, some COIs such as arsenic readily adsorb to aquifer materials, do not readily solubilize, and thus are relatively immobile. Those constituents are detected in a more sporadic manner than boron at concentrations greater than PBTVs and 2L/IMAC. Hydrogeo/ogic Conditions Site hydrogeologic conditions were evaluated by installing and sampling groundwater monitoring wells and piezometers; conducting in -situ hydraulic tests; sampling soil for physical and chemical testing; and sampling surface water, seeps, and sediment. The groundwater flow system serves to store and provide a means for groundwater movement. The nature of groundwater flow across the Site is based on the character and configuration of the ash basin relative to man-made and natural drainage features, engineered drains, streams, and Lake Norman; hydraulic boundary conditions; and subsurface media properties. Page 14-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra The ash basin occupies former tributaries to the Catawba River (Lake Norman). The ash basin was formed when the natural drainage system for this arm of tributaries was dammed. The stream valley in which the ash basin was constructed is a distinct slope - aquifer system in which flow of groundwater into the ash basin and out of the ash basin is restricted to the local flow regime. Groundwater generally flows from northwest to southeast across the Site. Minor variances in groundwater flow direction are present based on topography and localized drainage features. Localized topographic relief results in adjacent groundwater divides associated with the natural ridges separating historic draws. Active sluicing contributes to free-standing water within the ash basin and is controlled downgradient by the ash basin dam and the NPDES outfall to Lake Norman (east side of ash basin). Four hydrostratigraphic units were identified at MSS and were evaluated during the CSA. A detailed description of each unit is provided in Section 6.2.2. 07 Ash - The ash pore water unit consists of saturated ash material. Thickness of ash varies across the Site, from a few feet to upwards of 80 feet thick. The majority of ash within the basin is saturated. ,67 Shallow - The shallow/surficial unit consists of regolith material (soil, saprolite, and alluvial material) that overlie the upper fractured bedrock. ,67 Deep - The deep (TZ) flow unit lies directly above competent bedrock and is a zone of partially weathered bedrock. 07 Bedrock - Top of competent bedrock was generally defined during the CSA investigation with REC>85 percent and RQD>50 percent. The majority of water - producing fracture zones were found within 50 feet of the top of competent bedrock. Deeper fractures encountered are generally limited water -producing. Groundwater flow directions and the overall morphology of the potentiometric surface vary little from "dry' to "wet" seasons. Hydraulic gradients remain relatively neutral throughout much of the Site, but are locally influenced by topographic relief. Downward vertical gradients are apparent in upgradient areas beyond the ash basin. These downward gradients are indicative of groundwater recharge areas. Upward vertical gradients (from bedrock to deep or shallow zones), occur on the eastern boundary of the ash basin near Lake Norman and its tributaries and within the central portion of the ash basin (main channel of historic Holdsclaw Creek). These upward gradients are indicative of groundwater discharge zones and support the limited downward migration of constituents to the bedrock flow zone beneath the ash basin. Page 14-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Groundwater head potentials are similar between different flow zones across the Site (shallow, deep, and bedrock), indicating that there is no distinct horizontal confining layer beneath the Site. While groundwater head potential and distribution of COIs indicate limited vertical migration across the Site, two exceptions are identified at AL- 2BRLL beneath the Dry Ash Landfill (Phase II) and at AB-1BR near the toe of the ash basin dam. Horizontal gradients, hydraulic conductivities, and groundwater velocities indicate that most of the transport of CCR-impacted groundwater occurs through the shallow and deep flow zones at MSS. Horizontal and Vertical Extent of Impact The groundwater plume is defined as locations (in three-dimensional space) where groundwater quality is impacted by the ash basin. Naturally occurring groundwater contains varying concentrations of a number of constituents (e.g., alkalinity, aluminum, magnesium, sodium, zinc, etc.). Inconsistent and low -concentration exceedances of these constituents in the groundwater data do not necessarily demonstrate distribution of CCR-impacted groundwater. Boron is useful as a marker constituent to evaluate the extent of CCR-related impacts in groundwater due to its natural mobility in the environment. Boron, in its most common forms, is soluble in water, and has a very low Kd value, indicating the constituent is highly mobile in groundwater. At MSS, boron is detected at concentrations greater than the PBTV and 2L standard beneath and downgradient (east) of the ash basin. Boron is not detected in background groundwater at a concentration greater than the reporting limit of 50 µg/L. Therefore, the detection of boron in groundwater provides a close approximation of the distribution of CCR-impacted groundwater. The area farthest downgradient at which boron is detected at a concentration greater than the applicable PBTVs is interpreted as the leading edge of the CCR-derived plume moving downgradient from the source area. Observations on the extent of CCR influence are as follows: 47 Based on the extent of boron concentrations above the PBTV and the 2L standard, lateral migration through the shallow and deep flow zones is apparent in areas east and downgradient of the ash basin waste boundary. ,67 Lateral groundwater movement east of the ash basin towards the unnamed tributary of Lake Norman appears predominant over vertical migration as boron is consistently less than or slightly over the detection limit in MW-14BR and GWA-11BR, but is routinely detected at concentrations greater than the 2L Page 14-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra beyond the ash basin compliance boundary in MW-14S/D, GWA-11S/D, and GWA-15S. ,C7 The unnamed tributary of Lake Norman immediately east of the ash basin serves as a groundwater discharge zone in the furthest downgradient area with CCR impacts. Concentrations of COIs consistently detected in the shallow and deep flow zones at the MW-14 location are interpreted as the leading edge of the plume migration. 167 Boron concentrations are non -detect at MW-10S/D, indicating CCR-impacted groundwater does not migrate in the peninsula located to the east along the shore of Lake Norman. ,67 The extent of vertical migration appears relatively limited to downgradient areas east of the ash basin (AB-1BR) and beneath the Dry Ash Landfill (Phase II). The leading edge of the bedrock plume is interpreted to be at/near the Lake Norman shoreline adjacent to SW-10 sample location. In summary, the shallow and deep zone flow units at MSS - beneath and downgradient of the ash basin - are impacted by CCR-derived constituents; however, these impacts do not necessarily migrate vertically in the same areas. Impact to the bedrock flow unit is confined, approximately, to the Dry Ash Landfill (Phase II) and immediately east of the waste boundary at AB-1. The limited vertical extent of the plume is represented by relatively minor groundwater concentrations in bedrock wells beneath the majority of the ash basin. 14.2 Maximum COI Concentrations The following summary of maximum COI concentrations is based on the most recent sample result available: -61 Ash Basin - arsenic, barium, beryllium, chloride, chromium, iron, manganese, TDS Beneath the Dry Ash Landfill (Phase II) - boron, molybdenum, strontium, vanadium y PV Structural Fill- antimony, cadmium, cobalt, nickel, selenium, sulfate, thallium The sample locations are shown on Figure 2-4 and results are listed in Appendix B, Table 1. COI exceedances of the 2L or IMAC standards in each hydrostratigraphic unit are depicted in Figures 11-1 through 11-63. Certain constituents present in ash pore water Page 14-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra are inhibited by geochemical mechanisms such as sorption (measured by the distribution coefficient Kd) and precipitation and may not be present in the underlying aquifer media. The COIs with low Kd values have observed migration patterns from the ash basin, and the COIs with higher Kd values have very limited migration patterns (if any). Time -series graphs showing changes in COI concentrations over time are included as Figures 14-1 through Figure 14-38. Representative flow transects from the source area (ash basin) and downgradient areas were selected to illustrate Site conditions. Additionally, the graphs are organized to facilitate trend analysis in relation to distance from the source area over time. Review of these figures provide observations consistent with analyses presented in Section 14.1. For the central transect (AB-154AB-124AB-1) through the ash basin, increasing concentrations of several COIs are observed in downgradient deep and bedrock wells at the ABA cluster (listed below). Trends within the ash basin in pore water and shallow/deep groundwater are inconsistent between constituents. In some instances (e.g. chloride at AB-12SL/D), constituent concentrations appear to be reaching a state of equilibrium between the ash pore water and underlying groundwater. For the eastern transect (MW-14GWA-114GWA-15), increasing concentrations of several COIs are observed in the shallow and deep groundwater downgradient of the waste boundary. Vertical migration of constituent concentrations is not apparent in this area. Concentration trends at all wells, surface water, and seep locations are graphically depicted in plan view and included as Figure 14-39 through Figure 14-76. COI concentrations are generally stable, often indicating slight variation which may be attributed to natural fluctuations. However, increasing concentrations of several COIs are observed in downgradient monitoring wells: ,67 AB-1D/BR: boron, chloride, cobalt, iron, manganese, strontium, TDS 07 GWA-11S/D: boron, TDS ,67 GWA-15S: boron ,67 MW-14S/D/BR: chloride ,67 MW-1: barium, chloride Page 14-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 410 MW-61): boron, chloride ,610 MW-71): boron, chloride 14.3 Contaminant Migration and Potentially Affected Receptors Contaminant Migration Groundwater COIs migrate laterally and vertically into and through surficial regolith, the regolith/bedrock TZ, upper bedrock and limited areas of deeper bedrock. Regolith at MSS is variable in thickness, ranging from less than ten (10) feet to upwards of 80 feet, and is typically saturated. Regolith is thickest in upgradient areas beyond the waste boundary, and generally contains the first occurrence of groundwater below a vadose zone of variable thickness. Ash sluiced to, and accumulated within, the ash basin is considered the primary source evaluated in this CSA. Constituent migration from source areas to downgradient receptors (Lake Norman) is mostly limited to the shallow and deep flow systems. Observation of mobile constituents, such as boron, indicates lateral contaminant migration follows the naturally occurring preferential flow formed by the former drainage tributaries to Lake Norman. In addition to migration along natural drainage features, migration through the shallow and deep flow system to the east of the ash basin towards the unnamed tributary beyond the ash basin compliance boundary is apparent in groundwater monitoring wells MW-14S/D and GWA-15S and surface water location SW-6. COI concentrations (i.e. boron) in MW-14BR are comparable to the PBTV, consistent with the conceptual model of upward vertical gradients near groundwater discharge areas (i.e. the unnamed tributary). The limited areal extent of bedrock groundwater concentrations representative of CCR- related impacts may indicate limited downgradient constituent transport through the fractured bedrock system. However, apparent downward vertical migration of constituents is observed at two well clusters, AL-2 and AB-1. Concentrations at AL- 2BRLL are similar to those in the shallow and deep flow system. A scenario that could result in these concentrations at depth would be a bedrock fracture that is connected vertically to overlying flow systems but isolated horizontally at depth. This scenario would allow downward migration of constituent concentrations while limiting attenuation processes (e.g. dispersion and dilution) by natural waters. Concentrations in deep bedrock well AB-1BRL are similar to those in the overlying well AB-113R. Currently, only one sample is available for evaluation at AB-1BRL. Sampling results from nine sampling events at AB-1BR indicate increasing concentrations of boron, chloride, iron, manganese, strontium, and TDS. Many of these constituents are Page 14-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra considered to be indicators of the leading edge of CCR-impacted water. Lake Norman is the downgradient receptor that receives groundwater discharge from the vicinity of the AB-1 cluster. Surface water sampling conducted to date indicates general compliance with 2B standards as discussed below. Recent concentrations of COIs in solid media, as well as available geochemical properties of soils, are provided on Figure 14-77. Recent concentrations of COIs in groundwater, surface water, and AOWs are provided on Figures 14-78 and 14-79. Potentially Affected Receptors The 2016 risk assessment found no evidence of unacceptable risks to humans exposed to groundwater on -site at MSS. The risk assessment identified potential risks under a hypothetical recreational and subsistence fisher exposure scenario and to the great blue heron (for selenium and vanadium). This risk assessment update supports that the fisher risks were overestimated based on exposure and modeled fish tissue uptake assumptions. Surface water concentrations of selenium and vanadium did not exceed ESVs, and thus do not pose risks to the piscivorous birds. Limited potential for unacceptable risks to vanadium and selenium for piscivorous birds (primarily attributable to sediment and modeled tissue concentrations) in Exposure Area 1 (Figure 12-1) is consistent with the 2016 assessment. Water Supply Wells Results from water supply wells did not indicate human health risks to off -Site residents potentially exposed to groundwater associated with the ash basin. 07 Based on the known groundwater flow direction, no water supply wells are located downgradient of the MSS ash basin. �7 Four public water supply wells were identified within a 0.5-mile radius of the MSS ash basin compliance boundary. NCDEQ coordinated sampling of 38 private water supply wells located upgradient of the MSS ash basin. Ten (10) additional water supply wells near MSS were sampled between September 2016 and February 2017 by Duke Energy. All supply wells are located to the north, west, and south of MSS within a 0.5-mile radius of the ash basin compliance boundary. None of the wells within the survey radius are determined to be downgradient of the ash basin. A review of the analytical data for the 48 private water supply wells sampled indicated several constituents were detected above 2L or IMAC standards, including pH (33 wells), iron (five wells), manganese (two wells), zinc (one well), TDS (one well), and vanadium (39 wells). In addition, concentrations of other Page 14-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra COIs exceeded the respective PBTVs in a number of these private water supply wells. However, these exceedances of 2L/IMAC and PBTV should be interpreted with caution for the reasons described below: ,67 PBTVs were developed using groundwater data from a set of 10 background wells or well clusters all located on the Site. The geochemical data from these wells may not be representative across the broader area encompassed by the water supply wells. 07 There is very limited information available about the sampled wells. Well construction equipment such as pipes, pumps, and fittings may influence water quality. ,67 The flow zone of the supply wells is unknown. It has been assumed that the wells are constructed within bedrock, however it is also possible that some of the wells are constructed within the overburden (bored wells). Boron is not detected greater than the PBTV in water supply wells. Additionally, values of pH, manganese, TDS, and vanadium are below the respective PBTVs. Detected concentrations of constituents within supply wells are determined to not be sourced from the ash basin. For the reasons previously stated, the presence of iron greater than the 2L and PBTV (2 wells) and zinc greater than the 2L and PBTV (1 well) do not present enough evidence to conclude influence from the Site. Three of the four public water supply wells are believed to currently be in use. These wells are located upgradient and west of the MSS along Sherrils Ford Road and south of the MSS beyond an arm of Lake Norman along E NC HWY 150. Surface Water Surface water samples were collected and analyzed from seven locations, including two upstream locations for reference. Surface water sample data generally indicates compliance with 2B standards, with the occasional exception of dissolved oxygen, dissolved copper, and TDS. Additional parameters including chloride, arsenic, selenium, cadmium (dissolved), and lead (dissolved) have been measured above the 2B standard at a single location. Most of these exceptions occur at SW-10, which is located adjacent to NPDES Outfall 001. Comparison of water type characteristics indicates that free-standing water collected from the ash basin is similar to ash pore water, upstream surface water sample locations are similar to background groundwater, and samples collected from Lake Norman Page 14-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra adjacent to the downgradient shoreline of the ash basin represent a mixture of both water types. Lake Norman is used as a water supply for the greater Charlotte area. Locally, the Town of Mooresville pumps raw (untreated) water from an intake in Lake Norman located approximately 3.8 miles upstream of MSS. The water is treated at one of two treatment plants, both located on Charlotte Highway (U.S. Highway 21). The two plants have a combined treatment capacity of 18 million gallons per day. The next closest intake is located approximately 5.6 miles downstream of MSS; Lincoln County pumps raw (untreated) water from Lake Norman at this intake. Pumped water is treated at the Lincoln County Water Treatment Facility located at 7674 Tree Farm Lane near Denver, NC. Additional surface water sampling will be completed and an evaluation of potential impacts of groundwater to surface water will be presented in the CAP. Page 14-9 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 15.0 CONCLUSIONS AND RECOMMENDATIONS The assessment described in this CAMA CSA investigated the Site hydrogeology, determined the direction of groundwater flow from the ash basin, and determined the horizontal and vertical extent of impacts to groundwater and soil sufficient to proceed with preparation of a CAP in accordance with the CAMA and 2L. The following conclusions are based on evaluation described in this CAMA CSA report: ,67 Ash sluiced to, and accumulated within, the ash basin is determined to be a source and cause of groundwater impacts at MSS. 01 The updated CSA has determined no unacceptable risks to public health and safety from exposure to groundwater, surface water, or sediment impacts related to the ash basin. The MSS ash basin is currently designated as "Intermediate" risk under CAMA, meaning that closure of the ash basin is required by 2024. A "Low" risk classification and closure via a cap -in -place scenario are considered appropriate as alternative water supplies are being provided in accordance with G.S. 130A-309.213.(d)(1) of House Bill 630. ,67 Receptors including water supply wells and surface water bodies were identified and found to be not impacted by the ash basin and generally in compliance with applicable regulatory standards. Significant exposure pathways are understood and constituent concentrations detected in water supply wells are deemed to not be from the ash basin. 167 Impacts to groundwater in all three flow zones have been identified beneath and downgradient of the ash basin at MSS. Supplemental data collection to support groundwater modeling and long-term monitoring is anticipated to support the CAP process. 10 Secondary sources have been identified in soil beneath the ash basin. Shallow soil impacts are anticipated to be addressed through basin closure and the CAP. 01 Surface water receptors downgradient of the ash basin (e.g. Lake Norman) demonstrate compliance with 2B standards, with the occasional exception of dissolved oxygen, dissolved copper, and TDS. Localized influence from NPDES permitted outfalls are likely contributing to these exceptions. Additional surface water and sediment data collection is anticipated to support the evaluation of potential MNA in the area of the groundwater plume discharge to surface water. Page 15-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra The collection of additional data to support potential MNA as a groundwater corrective action strategy and additional plume definition is anticipated. A discussion of preliminary corrective action alternatives that may be appropriate to consider during the updated CAP development are presented herein. 15.1 Overview of Site Conditions at Specific Source Areas The ash management areas incorporated for Site assessment at MSS include the ash basin, Dry Ash Landfills (Phases I and II), and the PV structural fill. Ash sluiced to, and accumulated within, the ash basin waste boundary was determined to be a source of impacts to groundwater. CCR in the additional ash management areas have also contributed to groundwater impact. Boron is the primary marker constituent detected in groundwater at concentrations greater than background and the 2L near or beyond the compliance boundary. The interpreted extent of boron concentrations greater than the 2L standard is near or beyond the compliance boundary in the shallow and deep flow zones (Figures 11-13 and 11-14). The boron concentration is less than the 2L in the bedrock flow unit near the compliance boundary downgradient toward the unnamed tributary of Lake Norman. Apparent vertical migration of COIs into bedrock is observed near the compliance boundary east of the ash basin at AB-1BR. 15.2 Revised Site Conceptual Model Site Conceptual Models (SCMs) are developed to be a representation of what is known or suspected about a site with respect to contamination sources, release mechanisms, transport, and fate of those contaminants. SCMs are a written and/or graphic presentation of site conditions to reflect the current understanding of the site, identify data gaps, and can be updated as new information becomes available. SCMs are used to develop an understanding of the variable aspects of site conditions, such as a hydrogeologic conceptual site model to help understand the site hydrogeologic conditions affecting groundwater. SCMs can also be used in a risk assessment to understand contaminant migration and pathways to receptors. Based on conclusions within this CAMA CSA, the following is a brief description of the MSS SCM. The MSS occupies a distinct slope aquifer system that is encircled by topographic ridges resulting in a groundwater divide between the Site and surrounding upgradient properties. Lake Norman is located downgradient (east) of the ash basin and represents a groundwater discharge zone for groundwater leaving the Site. Groundwater flow and geochemical characteristics are the primary controls of constituent migration across the Site. Coal ash sluiced to, and accumulated within, the ash basin is the primary source of COIs in groundwater. Boron is a key indicator of CCR groundwater impacts. Boron, in its most common forms, is soluble in water, and has a very low Kd value, indicating the Page 15-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra constituent is highly mobile in groundwater. Boron is detected at concentrations greater than the 2L and PBTV beneath and downgradient (east) of the ash basin waste boundary, but is not detected in background groundwater. Therefore, the detection of boron in groundwater provides a close approximation of the distribution of CCR- impacted groundwater. Geochemical modeling to be performed within the CAP will evaluate how the occurrence of various constituents react to changes in the geochemical environment (Eh and pH) along selected flow transects and through time. Boron concentrations tend to increase in the downgradient direction beneath the central portion of the ash basin in the shallow and deep flow zones, but show an overall decreasing temporal trend. Elevated boron concentrations (relative to these flow zones beneath the basin) are observed within the shallow flow system adjacent to Lake Norman in AB-1S. Increasing concentrations over time are observed at this well cluster within the deep and bedrock flow zones, and in all three flow zones in the downgradient area towards the unnamed tributary of Lake Norman. Groundwater enters Lake Norman (nearest groundwater discharge area and surface water receptor) immediately downgradient from the AB-1 cluster. Surface water quality of Lake Norman adjacent to the ash basin is generally in compliance with 2B standards. An evaluation of groundwater to surface water interaction will be used to support the CAP process. Constituent concentrations observed in soils collected beneath the ash basin indicate the potential for secondary sources. Refined understanding of specific constituent behavior (such as arsenic, barium, chromium, and strontium) through geochemical modeling will be used to support the CAP process. Additional data collection regarding the shallow flow system beneath the ash basin is recommended to improve groundwater modeling of source conditions and to monitor performance of the selected corrective action remedy. Continued monitoring of recently installed deep bedrock wells will continue to provide information regarding the limited downward vertical migration cited in this report. The SCM will continue to be refined following evaluation of the groundwater models to be presented in the CAP and additional information obtained in subsequent data collection activities. 15.3 Interim Monitoring Program An Effectiveness Monitoring Program (EMP) is required by CAMA §130A-309.209 (b)(1)e. The EMP for MSS is anticipated to begin once the basin closure and groundwater CAP have been implemented. In the interim, an IMP has been developed Page 15-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra at the direction of NCDEQ. The CAP, and a proposed EMP, will be submitted at a future date. 15.3.1 IMP Implementation An IMP has been implemented in accordance with NCDEQ correspondence (NCDEQ, December 20, 2017; Appendix A) that provided approved "Revised Interim Monitoring Plans." Sampling will be conducted quarterly until approval of the CAP or as otherwise directed by NCDEQ. Groundwater samples will be collected using low -flow sampling techniques in accordance with the Low Flow Sampling Plan, Duke Energy Facilities, Ash Basin Groundwater Assessment Program, North Carolina, June 10, 2015 (Appendix G) conditionally approved by NCDEQ in a June 11, 2015 email with an attachment summarizing its approval conditions. Samples will be analyzed by a North Carolina certified laboratory for the parameters listed in Table 15-1. The table includes targeted minimum detection limits for each listed constituent. Analytical parameters and detection limits for each medium were selected so the results could be used to evaluate the effectiveness of a future remedy, conditions within the aquifer that may influence the effectiveness of the remedy, and migration of constituents related to the ash basin. Laboratory detection limits for each constituent are targeted to be at or below applicable regulatory values (i.e., 2L, IMAC, or 213). Monitoring wells and surface water locations that will be sampled and monitored as part of the IMP, as approved in NCDEQ correspondence (NCDEQ, December 20, 2017; Appendix A), are included in Table 15-2. 15.3.2 IMP Reporting Currently, data summary reports comprised of analytical results received during the previous month are submitted to NCDEQ on a monthly basis. In addition, NCDEQ directed that an annual IMP report be submitted by April 30 of the following year of data collection. The reports shall include materials that provide "an integrated, comprehensive interpretation of site conditions and plume status." The initial report was to be submitted to NCDEQ no later than April 30, 2018; however, the December 20, 2017 correspondence provides that the required date for an annual monitoring report will be extended to a date in 2018 to be determined later. 15.4 Preliminary Evaluation of Corrective Action Alternatives This preliminary evaluation of corrective action alternatives is included to provide insight into the updated CAP preparation process. The preliminary evaluation is based Page 15-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra on data available and the current understanding of regulatory requirements for the Site. It is assumed a source control measure of either an engineered capping system (cap -in - place) to minimize infiltration, or MNA, or a combination of the two, will be designed following completion of the risk classification process. Closure of the ash basin is required by 2024 under CAMA (currently ranked as Intermediate Risk). The updated risk assessment (Section 12.0) has determined there is no imminent risk to human health or the environment due to groundwater, surface water, or sediment impacts. Alternative water supplies are being provided to meet requirements of G.S. 130A-309.213.(d)(1) of House Bill 630; therefore, a "Low" risk classification will be appropriate. In addition, groundwater modeling conducted as part of the CAP Part 1 (HDR, 2015) has indicated closure by excavation compared to a Cap- in -place closure does not substantively accelerate groundwater clean-up. For basin closure reduction of infiltrating water will have the greatest positive impact on groundwater and surface water quality downgradient of the ash basin. Potential groundwater remedial strategies are to be considered as part of the CAP. 15.4.1 CAP Preparation Process The CAP preparation process is designed to identify, describe, evaluate, and select remediation alternatives with the objective of bringing groundwater quality to levels that meet applicable standards, to the extent that the objective is economically and technologically feasible, in accordance with 2L .0106 Corrective Action. Sections (h), (i), and (j) regarding CAP preparation read as follows: (h) Corrective action plans for restoration of groundwater quality, submitted pursuant to Paragraphs (c), (d), and (e) of this Rule shall include: (1) A description of the proposed corrective action and reasons for its selection; (2) Specific plans, including engineering details where applicable, for restoring groundwater quality; (3) A schedule for the implementation and operation of the proposed plan; and (4) A monitoring plan for evaluating the effectiveness of the proposed corrective action and the movement of the contaminant plume. (i) In the evaluation of corrective action plans, the Secretary shall consider the extent of any violations, the extent of any threat to human health or safety, the extent of damage or potential adverse impact to the environment, technology available to accomplish restoration, the potential for degradation of the contaminants in the Page 15-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra environment, the time and costs estimated to achieve groundwater quality restoration, and the public and economic benefits to be derived from groundwater quality restoration. (j) A corrective action plan prepared pursuant to Paragraphs (c), (d), or (e) of this Rule shall be implemented using a remedial technology demonstrated to provide the most effective means, taking into consideration geological and hydrogeological conditions at the contaminated site, for restoration of groundwater quality to the level of the standards. Corrective action plans prepared pursuant to Paragraphs (c) or (e) of this Rule may request an exception as provided in Paragraphs (k), (l), (m), (r), and (s) of this Rule. To meet these requirements and to provide a comprehensive evaluation, it is anticipated that the CAP will include: 17 Corrective action objectives and evaluation criteria ,0 Technology assessment ,0 Formulation of remedial action alternatives 17 Analysis, modeling, selection, and description of selected remedial action alternative(s) 167 Conceptual design elements, including identification of pre -design testing such as pilot studies 07 Monitoring requirements and performance metrics 167 Implementation schedule The following Site conditions significantly limit the effectiveness of a number of possible technologies: 167 The COIs in groundwater flow primarily through the TZ and upper fractured bedrock. y The formations are very heterogeneous with anisotropic flow conditions. The preliminary screening of potential groundwater corrective action included: 47 Source control by capping in place, and monitored natural attenuation, will be vital components to the CAP. Page 15-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra ,67 Groundwater migration barriers. The depth and heterogenitity of the impacted zone must be considered to determine if this technology may be feasible. ,67 In -situ chemical immobilization. This technology has been demonstrated to be effective for a number of relevant COIs. It's ability to achieve reguired boron concentrations will require evaluation. 07 Permeable reactive barrier. Similar to in -situ chemical immobilization, permeable reactive barrier technology can be evaluated for all Site COIs. 167 Groundwater extraction. Groundwater extraction could potentially be a viable choice as a key element of groundwater corrective action in combination with source control and MNA. However, further analysis is required and will be addressed in the updated CAP. Potentially viable options will be further evaluated in the CAP with updated flow and transport and geochemical modeling. 15.4.2 Summary This preliminary evaluation of corrective action alternatives is intended to provide insight into the revised CAP preparation process, as outlined in 2L. It is based on data available and the current regulatory requirements for the Site. It addresses potentially applicable technologies and remedial alternatives. Potential approaches are based on the currently available information about Site hydrogeology and COIs. In general, three hydrogeologic units or zones of groundwater flow are described for the Site: shallow/surficial zone, deep, and bedrock. The Site COIs include a list of common CCR-related constituents such as boron. If required, the potentially applicable technologies to supplement source control and MNA include groundwater extraction technologies such as conventional vertical wells, angle -drilled and horizontal wells. Migration barriers, in -situ chemical immobilization, and permeable reactive barriers are also identified as potentially applicable remedial action alternatives. In the event that extracted groundwater may require treatment prior to discharge, several water treatment technologies for the relevant COIs would be evaluated, including pH adjustment, metals precipitation, ion exchange, permeable membranes, and adsorption technologies. Page 15-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra The CAP will further evaluate basin closure options to reduce the potential impacts to human health or the environment; short- and long-term effectiveness, sustainability, implementability, and potential for attenuation of contaminants; time and cost to achieve restoration; public and economic benefits; and compliance with applicable laws and regulations. The CAP evaluation process will be used to determine which approach, or combination of approaches, will be most effective. Modeling will also be used to evaluate the various options prior to selection. Page 15-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra 16.0 REFERENCES ASTM. (2014). E1689-95: Standard Guide for Developing Conceptual Site Models for Contaminated Sites. ATSDR. (2012). Toxicological profile for Manganese. Atlanta: U.S. Department of Health and Human Services, Public Health Service. Ayotte, J. D., Gronbert, J. M., & Apodaca, L. E. (2011). Trace Elements and Radon in Groundwater Across the United States 1992-2003. U.S. Geological Survey Scientific Investigations Report 2011-5059. Brobst, D. A. (1962). Geology of the Spruce Pine District Avery, Mitchell and Yancey Counties North Carolina. United States Geological Survey Bulletin 1122-A. Butler, J. (1991). Metamorphism, in, Horton, J. W., Jr. and Zullo, V. A., eds. The Geology of the Carolinas: The University of Tennessee Press, Knoxville, p. 127-141. Butler, J., & Secor, D. (1991). The Central Piedmont, in the Geology of the Carolinas. In J. W. Horton, & V. A. Zullo (Eds.), The geology of the Carolinas: Carolina Geological Society fiftieth anniversary volume (1 ed.). Knoxville, TN: Univ. of Tennessee Press. Chapman, M. J., Cravotta, III, C. A., Szabo, Z., & Lindsey, B. D. (2013). Naturally occurring contaminants in the Piedmont and Blue Ridge crystalline -rock aquifers and Piedmont Early Mesozoic basin siliciclastic-rock aquifers, eastern United States, 1994- 2008. United States Geological Survey, Water Resources Investigations Report 00- 4286. Cox, J., Lundquist, G., Przyjazny, A., & Schmulbach, C. (1978). Leaching of boron from coal ash. Environmental Science & Technology, 12(6), 722-723. Cunningham, W. L., & Daniel, C. L. (2001). Investigation of ground -water availability and quality in Orange County, North Carolina. North Carolina: U.S. Dept. of the Interior, U.S. Geological Survey; Branch of Information Services. Daniel III, C., & Sharpless, N. B. (1983). Ground -water supply potential and procedures for well -site selection upper Cape Fear basin, Cape Fear basin study, 1981-1983. North Carolina Department of Natural Resources and Community Development and U.S. Water Resources Council in cooperation with the U.S. Geological Survey, 73 p. Daniel, C. C., & Dahlen, P. R. (2002). Preliminary hydrogeologic assessment and study plan for a regional ground -water resource investigation of the Blue Ridge and Piedmont Page 16-1 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL—Marshall_CSA—Report-2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra provinces of North Carolina. Raleigh, North Carolina: U.S. GEOLOGICAL SURVEY Water -Resources Investigations Report 02-4105. Daniel, C. I. (1990). Evaluation of Site Selection Criteria, Well Design, Monitoring Techniques, and Cost Analysis for a Ground Water Supply in Piedmont Crystalline Rocks, North Carolina. U.S. Geological Survey Water -Supply Paper, 2341-B, 35 p. Dennis, A., Shervais, J., & Secor, D. (2000). Faults bounding eclogite-bearing gneisses. Geological Society of America Abstracts with Program, v. 32 (2), A-14. Dewberry. (October 2016). Potable water programmatic evaluation. Donahue, J., & Kibler, S. (2007). Groundwater Quality iin the Peidmont/Blue Ridge Unconfined Aquifer System of Georgia. Atlanta: Georgia Department of Natural Resources. Environmental Protection Division. Dudas, M. (1981). Long-term leachability of selected elements from fly ash. Environmental Science & Technology, 15(7), 840-843. Duke Energy. (2017, August 11). Retrieved December 2017, from Duke Energy Coal Plants and Ash Management: https://www.duke-energy.com/—/media/pdfs/our- company/ash-management/duke-energy-ash-metrics.pdf?la=en EPA. (2007). Monitored Natural Attenuation of Inorganic Contaminants in Groundwater - Volume 2: Assessment for Non -Radionuclides Including Arsenic, Cadmium, Chromium, Copper, Lead, Nickel, Nitrate, Perchlorate, and Selenium. National Risk Management Research Laboratory, Office of Research and Development. Ada, Oklahoma: USEPA. EPRI. (1995). Coal ash disposal manual: Third edition. Palo Alto, CA: Electric Power Research Institute, TR-104137. EPRI. (2005). Chemical Constituents in Coal Combustion Product Leachate: Boron. Palo Alto: Electric Power Research Institute. EPRI. (2008a). Chemical Profile: Selenium. Electric Power Research Institute, Palo Alto, CA and Hydro One Networks, Inc., Toronto, Canada. EPRI. (2008b). Toxics Release Inventory. Chemical Profile: Arsenic. Palo Alto, CA: Electric Power Research Institute. EPRI. (2008c). Toxics Release Inventory. Chemical Profile: Barium. Palo Alto, CA: Electric Power Research Institute. Page 16-2 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra EPRI. (2008d). Toxics Release Inventory. Chemical Profile: Beryllium. Palo Alta, CA: Electric Power Research Institute. EPRL (2012). Groundwater Quality Signatures for Assessing Potential Impacts from Coal Combustion Product Leachate. Palo Also, CA: EPRI. Fenneman, N. (1938). Physiography of the Eastern United States. York, PA: Mc-Graw Hill Book Company, Inc. Feth, J. (1981). Chloride in natural continental water - A review: U.S. Geological Survey Water Supply Paper. 2176, 30p. Finkelman, R. (1995). Modes of occurrence of environmentally -sensitive trace elements in coal. In D. Swain, & F. Goodzarzi, Environmental Aspects of Trace Elements in Coal (pp. 24-50). Kluwer Academic Publishers. Fleet, M. (1965). Preliminary investigations into the sorption of boron by clay minerals. Clay Minerals, 6(1): 3-16. Freeze, R. A., & Cherry, J. A. (1979). Groundwater. Englewood Cliffs, NJ: Prentice -Hall. Fruchter, J. S., Rai, D., & Zachara, J. M. (1990). Identification of solubility -controlling solid phases in a large fly ash field lysimeter. Environmental Science and Technology, 24(8), 1173-1179. Gilbert, N., Brown, H., & Schaeffer, M. 1. (1982). Structure and geologic history of a part of Charlotte belt, South Carolina Piedmont. Southeastern Geology, v. 23, p. 129-145. Gillespie, E. (2013). Characterizing the Sources and Variability of Manganese in Well Water of the North Carolina Piedmont. International Conference on the Biogeochemistry of Trace Elements. Goldberg, S. (1997). Reactions of boron with soils. Plant and Soil, 193: 35-48. Goldberg, S., Forster, H., Lesch, S., & Heick, E. (1996). Influence of anion competition on boron adsorption by clays and soils. Soil Science, 161 (2): 99-103. Goldsmith, R., Milton, D. J., & Horton, Jr., J. W. (1988). Geologic map of the Charlotte 1 degree x 2 degrees quadrangle, North Carolina and South Carolina. Miscellaneous Investigations Series Map I-1251-E, scale 1:250,000. United States Geological Survey. Goodarzi, F., Huggins, F., & Sanei, H. (2008). Assessment of elements, speciation of As, Cr, Ni and emitted Hg for a Canadian power plant burning bituminous coal. International Journal of Coal Geology, 74(1): 1-12. Page 16-3 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Griffitts, W. R., Whitlow, J. W., Siems, D. F., Duttweiler, K. A., & Hoffman, J. D. (1984). Distribution of cobalt in heavy -mineral -concentrate samples from the Charlotte 1 X 2 quadrangle, North Carolina and South Carolina. U.S. Geologic Survey. Haley & Aldrich. (2015). Evaluation of NC DEQ private well data, volumes 1 and 2. Harned, D. A., & Daniel III, C. C. (1992). The transition zone between bedrock and regolith: Conduit for contamination?, p. 336-348, in Daniel, C. C., III, White, R. K., and Stone, P. A., eds., Groundwater in the Piedmont. Proceedings of a Conference on Ground Water in the Piedmont of the Eastern United States, October 16-18, 1989, Clemson University, 693p. Harned, D., & Daniel, C. (1992). The transition zone between bedrock and regolith: Conduit for contamination. In Daniel, C.C., White, R., and Stone, P., eds., Groundwater in the Piedmont, Proceedings of a Conference on Ground Water in the Piedmont of the Eastern United States, Charlotte, N.C., Oct. 16-18, 1989. Clemson, SC: Clemson University (336-348). Hatcher, R. D., Bream, B. R., & and Merschat, A. J. (2007). Tectonic map of the southern and central Appalachians: A tale of three orogens and a complete Wilson cycle. Geological Society of America Memoir: 4-D Framework of Continental Crust, 200, p. 595-632. HDR and SynTerra. (September 30, 2014a). Marshall steam station ash basin, drinking water supply well and receptor survey. HDR. HDR and SynTerra. (November 6, 2014b). Marshall seam station ash basin supplement to drinking water supply well and receptor survey. HDR. HDR and SynTerra. (2014c). Marshall Steam Station Ash Basin, Proposed Groundwater Assessment Work Plan (Rev. 1). December 30. HDR and SynTerra. (September 8, 2015a). Comprehensive Site Assessment Marshall Steam Station Ash Basin- September 8, 2015. Terrell, NC. HDR and SynTerra. (December 7, 2015b). Corrective Action Plan Part 1 Marshall Steam Station Ash Basin- December 7, 2015. Terrell, NC. HDR and SynTerra. (March 3, 2016a). Corrective Action Plan- Part 2 : Marshall Steam Station Ash Basin - March 3, 2016. Terrell, NC. HDR and SynTerra. (August 4, 2016b). Comprehensive Site Assesement Supplement 2 - Marshall Steam Station Ash Basin. Terrell, NC. Page 16-4 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL—Marshall_CSA—Report-2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra HDR and SynTerra. (2017). Statistical Methods for Developing Reference Background Concentrations for Groundwater and Soil at Coal Ash Facilities. HDR Engineering, Inc. and SynTerra Corporation. Heath, R. (1980). Basic elements of groundwater hydrology with reference to conditions in North Carolina. United States Geological Survey, Open -File Report: 80-44. Heath, R. C. (1984). Ground -Water Regions of the United States. United States Geological Survey Water -Supply Pater 2242. Hem, J. D. (1985). Study and Interpretation of the Chemical Characteristics of Natural Water. United States Geological Survey, Water -Supply Paper 2254. Hibbard, J., Stoddard, E., Secor, D., & Dennis, A. (2002). The Carolina zone: Overview of Neoproterozoic to early Paleozoic Peri-Gondwanan terranes along the eastern flank of the southern Appalachians. Earth Science Reviews, 57(3): 299-339. Horton, J. W. (1981). Geology map of the Kings Mountain belt between Gaffney, SC, and Lincolnton, NC, p. 6-18. in Horton, J.W., Jr., J. R. Butler & D. J. Milton, eds., Geological investigations of the Kings Mountain Belt and adjacent areas in the Carolinas. Carolina Geological Society Field Trip Guidebook: Columbia South Carolina, South Carolina Geological Survey, 247p. Horton, J. W., & Butler, J. R. (1977). Guide to the geology of the Kings Mountain belt in the Kings Mountain area, North Carolina and South Carolinas, in Burt, E. R. ed., Field Guides for the Geological Society of America, Southeastern Section Meeting, Winston-Salem, North Carolina. Geology and Mineral Resources Section, North Carolina Department of Natural and Economic Resources, p. 76-143. Horton, J.W., Jr. and Bultler, J.R. 1981. Geology and mining history of the Kings Mountain belt in the Carolinas - A summary and status report. In Horton, J.W., Jr, and other eds. Geological investigations of the Kings Mountain belt and adjacent areas in the Carolinas (Carolina Geological Society Field Trip Guidebook 1981): Columbia South Carolina Geological Survey, p.194-212. Horton, Jr., J. W., Drake, Jr., A. A., & Rankin, D. W. (1989). Tectonostratigraphic terranes and their Paleozoic boundaries in the central and southern Appalachians, in Dallmeyer, R.D., ed., Terranes in the Circum-Atlantic Paleozoic Orogens. Geological Society of America Special Paper 230, 213-245. Hurlbut, C. S. (1971). Dana's manual of mineralogy (18 ed.). John Wiley & Sons Inc. Institute, E. P. (2008c). Toxics Release Inventory. Chemical Profile: Thallium. Palo Alto, CA. Page 16-5 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL—Marshall_CSA—Report-2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Izquierdo, M., & Querol, X. (2012). Leaching behaviour of elements from coal combustion fly ash: An overview. International Journal of Coal Geology, 94.54-56. Jones, K. B., & Ruppert, L. F. (2017, February). Leaching of trace elements from Pittsburgh coal mill rejects compared with coal combustion products from a coal- fired power plant in Ohio, USA. United States Geological Survey Bulletin, 171, 130- 141. Keith, A., & Sterrett, D. B. (1931). Description of the Gaffney and Kings Mountain quadrangles, South Carolina -North Carolina. U. S. Geological Survey Geological Atlas, Folio 222, 13p. Kesler, T. L. (1944). Correlation of some metamorphic rocks in the central Carolina Piedmont. Geological Society of America Bulletin, v. 55, p. 755-782. Kim, A.G. and Kazonich, G. 2004. The silicate/non-silicate distribution of metals in fly ash and its effect on solubility. Fuel, v.83(17), p. 2285-2292. Kim, J. H., Gibb, H. J., & Howe, P. D. (2006). Cobalt And Inorganic Cobalt Compounds. Geneva: World Health Organization. King, P. B. (1955). A geologic section across the southern Appalachians: An outline of the geology in the segment in Tennessee, North Carolina, and South Carolina, in, Russell, F. J., ed., Guides to southeastern geology: New York. Geological Society of America, p. 332-373. Kunesh, C. (1978). Barium. (M. Grayson, & D. Eckroth, Eds.) Kirk-Othmer encyclopedia of chemical technology, Vol. 3, pp. 457-463. LeGrand, H. (1988). Region 21, Piedmont and Blue Ridge. In: J. Black, J. Rosenshein, P. Seaber, ed. Geological Society of America, 0-2, (pp. 201-207). LeGrand, H. (1989). A conceptual model of ground water settings in the Piedmont region, in groundwater in the Piedmont. In: Daniel C., White, R., Stone, P., ed. Ground Water in the Piedmont of the Eastern United States (pp. 317-327). Clemson, SC: Clemson University. LeGrand, H. (2004). A master conceptual model for hydrogeological site characterization in the Piedmont and Mountain Region of North Carolina: A guidance manual. North Carolina Department of Environment and Natural Resources, Division of Water Quality, Groundwater Section, Raleigh, NC, 55. Lockhart, E. E., Tucker, C. L., & Merritt, M. C. (1955). The Effect of water impurities on the flavor of brewed coffee. Food Res., 20:598. Page 16-6 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Martens, D. A. (2002). Selenium. In E. o. Science. New York: Marcel Dekker. McConnell, H. H., & Lewis, J. (1972). Add Salt to Taste. Environment, p. 14:38. Miao, Z., Brusseau, M. L., Carroll, K. C., & others. (2012). Environ Geochem Health. 34:539. https://doi.org/10.1007/sl0653-011-9423-1. Miner, S. (1969). Air pollution aspects of barium and its compounds. Bethesda: Litton Systems. NCDEQ-DWR. (October 2017). Monitored Natural Attenuation for Inorganic Contaminants in Groundwater: Guidance for Developing Corrective Action Plans Pursuant to NCAC 15A .0106 (I). Division of Environmental Quality, Division of Water Resources. NCDEQ-DWR. (April, 2015). Summary of North Carolina Surface Water Quality Standards 2007-2014. North Carolina Department of Environment and Natural Resources -Division of Water Resources. NCDHHS. (2010a). Concentration of Barium Detected in NC Private Well Water, Average 1998-2010 and Average 2010. Well Water & Health. University of North Carolina Superfund Research Program. NCDHHS. (2010b). Concentration of Iron Detected in NC Private Well Water, Average 1998-2010 and Average 2010. Well Water & Health. University of North Carolina Superfund Research Program. NCDHHS. (2010c). Concentration of Selenium Detected in NC Private Well Water, Average 1998-2010 and Average 2010. Well Water & Health. University of North Carolina Superfund Research Program. Nutter, L. J., & Otton, E. G. (1969). Groundwater occurrence in the Maryland Piedmont. Maryland Geological Survey Report of Investigations, No. 10, 42. Parker, R. (1967). Chapter D. composition of the Earth's crust. Geological survey paper 440-D. In Data of Geochemistry. 6th ed. Washington, D.C.: U.S. Government Printing Office, 1967. Piper, A. (1944). A graphic procedure in the geochemical interpretation of water analyses. American Geophysical Union Transactions, 25, 914-928. Polizzotto, M., Amoozegar, A., Austin, R., Bolich, R., Bradley, P., Duckworth, O., et al. (2015). Surface and Subsurface Properties Regulating Manganese Contamination of Groundwater in the North Carolina Piedmont. Water Resources Research Institute of The University of North Carolina. Page 16-7 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Pugh, J., Whetstone, G., & Redwine, J. (September 2010). Comparison of coal combustion products to other common materials. Electric Power Research Institute, TR-1020556. Ruhl, L., Vengosh, A., Dwyer, G. S., Hsu -Kim, H., Schwartz, G., Romanski, A., et al. (2012, September 30). The Impact of Coal Combustion Residue Effluent on Water Resources: A North Carolina Example. Environmental Science and Technology, 12226-12233. Schaeffer, M. F. (1981). Polyphase folding in a portion of the Kings Mountain belt, north -central South Carolina, p. 72-90, in Horton, J.W., J. R. Butler, and D. J. Milton, eds., Geological investigations of the Kings Mountain Belt and adjacent areas in the Carolinas. Carolina Geological Society Field Trip Guidebook: Columbia, South Carolina, South Carolina Geological Survey, 247p. Schaeffer, M. F. (2011). Carolina Piedmont groundwater system: What does the transition zone look like? 19th Annual David S. Snipes/Clemson Hydrogeology Symposium April 7, 2011, (pp. 43-44). Schaeffer, M. F. (2014a, April). Piedmont groundwater system, Part 1- The transition zone between regolith and bedrock: Existence. Geological Society of America, Abstracts with Program, 46 no. 3, 26-27. Schaeffer, M. F. (2014b, April). Piedmont groundwater system, Part 2 - The Transition zone between regolith and bedrock: Characteristics. Geological Society of America, Abstracts with Program, 46 no. 3, 27. Schaeffer, M. F. (April 2, 2009). Hydraulic conductivity of Carolina Piedmont soil and bedrock: Is a transition zone present between the regolith and bedrock? 7th Annual David S. Snipes/Clemson Hydrogeology Symposium, p. 32-36. Schmidt, H. (1962). Equilibrium diagrams for minerals at low temperature and pressure. The Geological Club of Harvard, 199. Secor, D. T., Balinsky, M., & Colquhoun, D. (1998). The Carolina terrane in northeastern South Carolina: history of an exotic volcanic arc. South Carolina Geology, v. 40, 1- 17. Smith, S. M. (2016). National Geochemical Database. Retrieved October 20, 2017, from USGS: https:Hpubs.usgs.gov/of/1997/ofr-97-0492/ Stewart, J. W. (1964). Infiltration and permeability of weathered crystalline rocks. Georgia Nuclear Laboratory, pp. D1-D57. Page 16-8 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra Stewart, J. W., Callahan, J. T., & Carter, R. F. (1964). Geologic and hydrologic investigation at the site of the Georgia Nuclear Laboratory. United States Geological Survey Bulletin 1133-F, pp. F1-F90. Strawn, D., Doner, H., Zavarin, M., & McHugo, S. (2002, February). Microscale investigation into the geochemistry of arsenic, selenium, and iron in soil developed in pyritic shale materials. Geoderma(108), 237-257. Stumm, W., & Morgan, J. (1996). Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters. 3rd ed. Chicago: Wiley-Interscience. U.S. Department of Agriculture. (1975). Soil Survey of Catawba County, North Carolina. Washington, DC: United States Department of Agriculture Soil Conservation Service. Urey, H., & Mem, R. (1953). On the concentration of certain elements at the earth's surface. In Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences (pp. 281-292). The Royal Society Publishing, 219(1138). USEPA. (1997). Ecological risk assessment guidance for superfund: Process for designing and conducting ecological risk assessments. Edison, NJ: U.S. Environmental Protection Agency, Environmental Response Team. EPA 540-R-97-006. USEPA. (2003). Drinking Water Advisory: Consumer Acceptability Advice and Health Effects Analysis on Sulfate. Washington: U.S. Environmental Protection Agency, EPA 822- R-03-007. USEPA. (October, 2007). Monitored natural attenuation of inorganic contaminants in ground water technical basis for assessment, volume I, October 2007. United States Environmental Protection Agency. EPA/600/R-07/139. USEPA. (2012). Sulfate in Drinking Water. Retrieved from EPA: http://water.epa.gov/drink/contaminants/unregulated/sulfate.cfm USEPA. (2013). Seconday Drinking Water Regulations: Guidance for Nuisance Chemicals. Retrieved from water.epa.gov: http://water. epa. gove/drink/contaminants/secondarystandards. cfm USEPA. (April 2015). Disposal of Coal Combustion Residual From Electric Utilities: Final Rule - April 17, 2015. Code of Federal Register, Vol. 80(No. 74). USEPA. (2017a). National recommended water quality criteria for aquatic life. Retrieved October 20, 2017, from EPA: https://www.epa.gov/wgc/national-recommended- water-quality-criteria-aquatic-life-criteria-table#altable Page 16-9 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx 2018 Comprehensive Site Assessment Update January 2018 Marshall Steam Station SynTerra USEPA. (2017b). National recommended water quality criteria for aquatic life. Retrieved October 20, 2017, from EPA: https://www.epa.gov/wgc/national-recommended- water-quality-criteria-aquatic-life-criteria-table#altable USGS. (1973). United States Mineral Resources. United States Government Printing Office. USGS. (1985). Cadmium. Ch. In Mineral Facts and Problems, 675. USGS. (2008). Hydrogeologic Setting, Ground -Water Flow, and Ground -Water Quality at the Langtree Peninsula Research Station, Iredell County, North Carolina, 2000- 2005. North Carolina Department of Environment and Natural Resources, Division of Water Quality. USGS. (2009). Chloride in Groundwater and Surface Water in Areas Underlain by the Glacial Aquifer System, Northern United States. Retrieved from pubs.usgs.gov: http://pubs.usgs.gove/sir/2009/5086/pdf/sire2009-5086.pdf WHO. (1996). Guidelines for drinking -water quality: Health criteria and other supporting information. (2nd, Ed.) Vol. 2. WHO. (2011). Cadmium in drinking -water. Background document for development of WHO guidelines for drinking -water quality. Geneva, World Health Organization, (WHOI SDEIWSH/03.04/80IRev1). Young, S.C. (1993). Physical and Hydraulic Properties of Fly Ash and Other By -Products From Coal Combustion. Palo Alto, CA. TR-101999: Electric Power Research Institute. Zawislandski, P. T., & Zavarin, M. (1996). Nature and rates of selenium transformations: A laboratory study of Kesterson Reservoir soils. Soil Science Society of America Journal, 60(3), 791-800. Page 16-10 P: \ Duke Energy Carolinas\ 18. MARSHALL \ CSA Update January 2018 \ FINAL_Marshall_CSA_Report_2018.docx