Volume 1 S A JV 2024c 2023 Annual Report Shepleys Hill Landfill.pdf

FINAL

United States Army Corps of Engineers

New England District

2023 Annual Operations, Maintenance, and Monitoring

Report Shepley’s Hill Landfill

Former Fort Devens Army Installation

Devens, Massachusetts

Contract No. W912WJ-19-D-0014

Contract Delivery Order No. W912WJ-23-F-0055

July 2024

FINAL 2023 Annual Operations, Maintenance, and Monitoring Report

Shepley’s Hill Landfill, Former Fort Devens Army Installation i

2023 Annual Operations, Maintenance, and Monitoring Report

Shepley’s Hill Landfill

Former Fort Devens Army Installation

Devens, Massachusetts

July 2024

Prepared By: Prepared For:

SERES-Arcadis 8(a) JV 2, LLC United States Army Corps of Engineers, 669 Marina Drive, Suite B-7 New England District

Charleston, South Carolina 29492

Tel 843 216 8531

CERTIFICATION

I hereby certify that the enclosed Report, shown and marked in this submittal, is that proposed to be incorporated with Contract Number W912WJ-19-D-0014. This document was prepared in accordance with the United States Army Corps of Engineers (USACE) Scope of Work and is hereby submitted for

Government approval.

Reviewed By:

Andy Vitolins, PG

Project Manager

Heather Levesque, PMP

Deputy Project Manager

Received By:

Penelope Reddy

USACE Project Manager

NOTICE

The United States Department of Defense, Department of Army, funded wholly or in part the preparation of this document and work described herein under Contract No.

W912WJ-19-D-0014. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

ii

Contents Acronyms and Abbreviations ................................................................................................................................. v

1 Introduction

1.1 Site Location

1.2 Site Background

2 Groundwater and Hydraulic Monitoring

2.1 Hydraulic Monitoring

2.2 Groundwater Sampling

2.2.1 2023 Data Summary

2.2.2 Arsenic Concentration Trends

2.2.3 Groundwater Monitoring Field Parameters

2.2.4 Quality Assurance/Quality Control

2.3 Conclusions

3 Landfill Maintenance and Monitoring

3.1 Landfill Inspection

3.2 Landfill Gas Monitoring

4 Arsenic Treatment Plant Operations, Maintenance, and Monitoring

4.1 System Description

4.1.1 System Efficiency and Routine Maintenance

4.1.2 Non-Routine System Maintenance, Repairs, and Upgrades

4.2 Arsenic Treatment Plant Monitoring

4.2.1 Influent

4.2.2 Effluent

4.3 Arsenic Treatment Plant Optimization and Recommendations

5 Land Use Controls

5.1 Institutional Controls

5.2 Affirmative Measures

6 References iii

Tables Table 1 ROD-Specified Contaminants of Concern and Cleanup Levels (in-text)

Table 2 LTMMP Sampling Program

Table 3 LTMMP Hydraulic Monitoring Program

Table 4 Hydraulic Monitoring Data for Spring and Fall 2023

Table 5 Historical Vertical Hydraulic Gradients

Table 6 Spring 2023 LTM Analytical Results and Field Parameters

Table 7 Fall 2023 LTM Analytical Results and Field Parameters

Table 8 2023 Mann-Kendall Trend Analysis Results (in-text)

Figures

Figure 1 Site Location

Figure 2 Site Plan

Figure 3 North Impact Area

Figure 4 Area of Land Use Controls

Figure 5 Long-Term Monitoring Well Network

Figure 6 Groundwater Elevation Contour Map – June 12, 2023

Figure 7 Groundwater Elevation Contour Map – November 20, 2023

Figure 8 Dissolved Arsenic in Groundwater – Spring 2023 North Impact Area and Shepley’s Hill

Landfill

Figure 9 Dissolved Arsenic in Groundwater – Fall 2023

Figure 10 Arsenic Concentration Trends

Figure 11 Landfill Gas Probes and Vents iv

Appendices Appendix A Groundwater Monitoring Field Forms

Appendix B Laboratory Analytical Reports

Appendix C Summary of Quality Control Exceedances and Data Validation Reports

Appendix D Site-Wide Hydraulic Monitoring Results from 2019 through 2023

Appendix E Historical Groundwater Monitoring Well Analytical Results

Appendix F ProUCL Outputs

Appendix G Summary of Landfill and ATP Operations, Maintenance, and Monitoring

Table G-1 2023 Gas Monitoring Results

Table G-2 Historical Gas Monitoring Results

Table G-3 Landfill Gas Probe Construction Details

Table G-4 2023 ATP Flow Rate Summary

Table G-5 2023 ATP Monthly Operations Summary

Table G-6 Historical ATP Monthly Discharge Totals

Table G-7 2023 ATP Filter Bed Roll-Off Disposal Totals

Table G-8 2023 ATP Routine and Non-Routine System Maintenance, Repairs, and Upgrades

Table G-9 Historical ATP As/Fe/Mn Influent Concentrations

Table G-10 ATP Influent VOC and Dissolved Gases Concentrations – 3 November 2023

Table G-11 Historical ATP Monthly Effluent Sampling Results (Arsenic)

Table G-12 Historical ATP Quarterly Effluent Sampling Results

Table G-13 ATP Effluent Sampling Results – 3 November 2023

Appendix H LUCIP Documentation

Appendix I Response to Comments v

Acronyms and Abbreviations µg/L microgram per liter

AOC area of contamination

Army United States Army

ATP arsenic treatment plant bgs below ground surface

CIP clean-in-place

CL cleanup level

COC contaminant of concern

DO dissolved oxygen

DoD Department of Defense

DOE Department of Energy

Eurofins Eurofins TestAmerica Laboratories

EW extraction well

FBRO filter bed roll-off

Fort Devens former Fort Devens Army Installation ft foot ft/ft foot per foot gpm gallon per minute

IC institutional control

KGS KOMAN Government Solutions, LLC

LEL lower explosive limit

LGP landfill gas point

LTM long-term monitoring

LTMMP Long-Term Monitoring and Maintenance Plan

LTMMP Addendum Addendum to the Long-Term Monitoring and Maintenance Plan Update

LTMMP Update Long-Term Monitoring and Maintenance Plan Update

LUC land use control

LUCIP Land Use Control Implementation Plan

MassDEP Massachusetts Department of Environmental Protection

MassDevelopment Massachusetts Development and Finance Agency vi

MCL maximum contaminant level

MF microfiltration mg/L milligram per liter

MS matrix spike

MSD matrix spike duplicate mV millivolt

NAVD88 North American Vertical Datum of 1988

NIA North Impact Area

O&M operation and maintenance

ORP oxidation-reduction potential

PLC programmable login controller ppm part per million

QAPP Quality Assurance Project Plan

QC quality control

Quality Systems Manual Department of Defense (DoD) and Department of Energy (DOE) Consolidated

Quality Systems Manual for Environmental Laboratories, Version 5.4 redox oxidation-reduction

RI remedial investigation

ROD Record of Decision

S-A JV SERES-Arcadis 8(a) Joint Venture 2, LLC

SHL Shepley's Hill Landfill

Sovereign Sovereign Consulting, Inc.

USACE United States Army Corps of Engineers

USEPA United States Environmental Protection Agency

VOC volatile organic compound

1 Introduction SERES-Arcadis 8(a) Joint Venture 2, LLC (S-A JV) prepared this 2023 Annual Operations, Maintenance, and

Monitoring Report for the United States Army Corps of Engineering (USACE) – New England District, under

Contract Number W912WJ-19-D-0014. This report summarizes the long-term monitoring (LTM) program and operation and maintenance (O&M) activities conducted during 2023 at the Shepley’s Hill Landfill (SHL) site, located at the former Fort Devens Army Installation (Fort Devens) in Devens, Massachusetts (Figure 1). LTM and

O&M activities were conducted in accordance with the Long-Term Monitoring and Maintenance Plan Update

(LTMMP Update; Sovereign Consulting, Inc. [Sovereign] 2015), the 2018 Addendum to the Long-Term Monitoring and Maintenance Plan Update (LTMMP Addendum; KOMAN Government Solutions, LLC [KGS] 2018), and the

Quality Assurance Project Plan (QAPP) for the Annual Long-Term Monitoring and Maintenance Program (S-A JV

2020).

1.1 Site Location

Fort Devens is located approximately 35 miles northwest of the city of Boston, within the towns of Ayer and

Shirley in Middlesex County and Harvard and Lancaster in Worcester County. Fort Devens was established in

1917 for military training and logistical support during World War I and operated as a permanent base from 1931 until closure by the Base Realignment and Closure Committee in 1996.

SHL encompasses approximately 84 acres in the northeast corner of the Main Post of Fort Devens (Figure 2).

The landfill is bordered to the east by Plow Shop Pond and land formerly containing a railroad roundhouse, to the west by Shepley’s Hill, to the south by recent commercial development, and to the north by wooded and residential areas. Plow Show Pond to the east drains to Nonacoicus Brook to the north, which flows north/northwest and discharges to the Nashua River. Nonacoicus Brook is located north of the landfill in the North

Impact Area (NIA; Figure 3).

1.2 Site Background

The United States Army (Army) reportedly began operating SHL by the early 1940s; however, evidence from test pits within the landfill suggests earlier usage, possibly as early as the mid-19th century. SHL contains a variety of waste materials (more than 1.5 million cubic yards) that extend to a depth of approximately 40 feet (ft) below ground surface in the central portion of the landfill. Approximately 160,000 cubic yards (11% of the total waste mass) appear to have been emplaced below the water table in a former wetland, and peat deposits have been identified below the waste (Harding ESE 2003; Sovereign 2011).

The Massachusetts Department of Environmental Protection (MassDEP) approved the landfill closure plan for

SHL in 1985. The landfill was closed in five phases between 1987 and 1993 in accordance with 310 Code of

Massachusetts Regulations 19.000. Closure consisted of capping the landfill, covering the cap with soil and vegetation, installing gas vents and groundwater monitoring wells, and constructing drainage swales. The

MassDEP issued a Landfill Capping Compliance Letter approving the closure in February 1996 (Harding ESE

2003; Sovereign 2011).

After closure of SHL, the Army completed remedial investigations (RIs) under the Comprehensive Environmental

Response, Compensation, and Liability Act to characterize soil, sediment, surface water, and groundwater. RIs confirmed the presence of various contaminants at the site, the most predominant of which is arsenic. A subsequent feasibility study and Record of Decision (ROD; USACE 1995) resulted in a remedial action, which included long-term O&M of the landfill cap, groundwater monitoring, and a contingent remedial component requiring the design and construction of a groundwater extraction and treatment system if the landfill cap did not result in a reduction of groundwater contaminant concentrations (primarily arsenic) below risk-based cleanup goals by 2008. The remedial action objectives for SHL are to:

Protect potential residential receptors from exposure to impacted groundwater migrating from the landfill that contains chemicals in excess of United States Environmental Protection Agency (USEPA) maximum contaminant levels (MCLs).

Prevent impacted groundwater from contributing to the contamination of Plow Shop Pond sediments in excess of human health and ecological risk-based concentrations.

Table 1, below, lists the contaminants of concern (COCs) identified in the ROD and their target cleanup levels

(CLs). Of the COCs listed below, only arsenic, iron, and manganese are currently sampled as part of the LTM program, in accordance with the LTMMP Update and LTMMP Addendum (Sovereign 2015; KGS 2018, respectively).

Table 1 ROD-Specified Contaminants of Concern and Cleanup Levels

COC Cleanup Level (µg/L) Selection Basis

Arsenic 10 MCLa

Manganese 1,715 Backgroundb

Iron 9,100 Background

Chromium 100 MCL

1,2-Dichlorobenzene 600 MCL

1,4-Dichlorobenzene 5 MCL

1,2-Dichloroethane 5 MCL

Lead 15 Action level

Nickel 100 MCL

Sodium 20,000 Health advisory

Aluminum 6,870 Background

Notes:

a In accordance with the 2005 Five-Year Review Report, data are screened against the updated standard of 10 µg/L (USACE 2005) rather than the original standard of 50 µg/L from the ROD (USACE 1995).

b Revised from ROD (USACE 1995) CL based on background evaluation.

µg/L = microgram per liter

Groundwater Extraction and Treatment

In March 2006, the Army installed two extraction wells (EW-01 and EW-04) near the downgradient edge of the landfill (the “toe” of the landfill) to capture groundwater flowing from beneath the landfill and constructed a groundwater extraction and treatment system. The system was designed to remove dissolved arsenic through oxidation using chlorine dioxide co-precipitation with iron and manganese, and microfiltration. System operation started with an initial pumping rate of 25 gallons per minute (gpm) in 2006, and then gradually increased the combined extraction rate to approximately 50 gpm by 2014.

In 2023, the system operated similarly to the previous years of operation for the first eight months of the year extracting approximately 50.3 gpm on average with an up-time of approximately 89%. On August 25, 2023, the system was shut down for significant upgrades. The purpose of the upgrades was to install an alternative oxidation system using potassium permanganate in place of the existing chlorine dioxide generation and dosing system to improve the performance and overall safety of the plant. The upgrades included the removal of all chlorine dioxide generation and dosing equipment and the installation potassium permanganate injection and clarification systems. This change was accompanied by a set of changes to mixing vessels, pumps, settling tanks, etc. These changes are described in more detail in Section 4. Due to these changes, the operation of the system was curtailed for the last four months of 2023 while the upgraded system was brought on-line.

Between September 5 and September 11, 2023, a new additional extraction well (EW-03) was installed southeast of the treatment building and near the landfill access gate. Once the treatment system came back on-line and following prove-out testing, groundwater extraction was performed from all three wells; EW-01, -04, and -03. The target for the upgraded system operation is to maintain the overall 50 gpm flow rate by reducing pumping EW-04 to 10 gpm and extract water from EW-03 at a rate of 15 gpm.

Plow Shop Pond and Red Cove

In 2011, the Army investigated Red Cove/Plow Shop Pond (known as Area of Contamination [AOC] 72) located east of SHL. The AOC 72 RI (AMEC 2011) concluded that the SHL remedy in place in 2011 (landfill capping and groundwater extraction) did not eliminate groundwater flow and arsenic migration from SHL to Red Cove/Plow

Shop Pond, and that groundwater discharge was contributing arsenic to sediment that could accumulate to levels that posed unacceptable risk to receptors. As a result, in August and September 2012, the Army installed a barrier wall between SHL and AOC 72 as a non-time critical removal action. Documentation of the barrier wall installation was provided in a Removal Action Completion Report (Sovereign 2013a).

In addition to LTM and O&M activities, land use controls (LUCs) were established for the NIA in 2013 (Sovereign

2013b), and the Army subsequently issued an LUC Implementation Plan (LUCIP) for SHL in 2014 (Sovereign

2014). The LUC area is shown on Figure 4 and discussed in Section 5.

2 Groundwater and Hydraulic Monitoring Groundwater monitoring was conducted at SHL in 2023 in accordance with the LTMMP Update, LTMMP

Addendum, additional work requested by USEPA, and QAPP (Sovereign 2015; KGS 2018; S-A JV 2020, respectively). Sampling events for 2023 occurred in June and October/November. Results of sampling are presented in the following sections.

The LTM program includes wells in the Barrier Wall Area, Landfill Area, Nearfield Area, NIA, and Upgradient

Area. These sampling and hydraulic monitoring locations are summarized in Tables 2 and 3, respectively. A site plan showing the LTM well network is provided on Figure 5. Groundwater monitoring field forms, laboratory analytical reports, and data validation reports are provided in Appendix A, Appendix B, and Appendix C, respectively.

2.1 Hydraulic Monitoring

Groundwater elevations at SHL monitoring wells and piezometers, including piezometers used to monitor performance of the barrier wall, were gauged on June 12 and November 20, 2023, while the ATP was online and groundwater was being pumped from extraction wells (EW-01 and EW-04 during the spring event and EW-01 and

EW-04 during the fall event). Hydraulic monitoring data are provided in Table 4. During each monitoring event, an electronic water level meter was used to measure depth to water with an accuracy of ±0.01 foot from the marked reference point at the top of the inside casing of each monitoring well/piezometer. LTM network wells were inspected, and new well locks were installed during the 2023 monitoring events where replacements were needed.

A total of 62 monitoring wells were measured during the spring 2023 monitoring event. In fall 2023, 154 monitoring wells and one staff gauge was measured; three of the four stream gauges included in Table 4 could not be accessed or viewed, as they are located in wetland areas and were partially obscured by dirt and mud.

Groundwater elevation contour maps for June 12 and November 20, 2023 (Figures 6 and 7, respectively) were developed using survey data from the 2017 and 2018 monitoring well survey reports, which use the

Massachusetts State Plane Coordinate System of North American Datum of 1983 for horizontal control and the

North American Vertical Datum of 1988 (NAVD88) for vertical control. As described in Table 4 (last column) each groundwater contour map includes contour values for shallow overburden wells only; deep paired overburden well and bedrock well data were excluded, but several were used to assess vertical gradients. Similar to previous monitoring events, the water levels measured on November 20, 2023, illustrate that groundwater flows generally from the southwest to the north toward Nonacoicus Brook, with a deflection of groundwater flow to the north by the barrier wall in the area west of the wall. Groundwater elevations in overburden well locations were observed to be lower on June 12 than November 20 by an average of 0.90 ft (Table 4). Historical groundwater elevations from the last 5 years (2019 through 2023) are included in Appendix D; groundwater elevations in overburden well locations were observed to be lower in fall 2022 compared to November 20, 2023, by an average of 1.91 ft. The following groundwater hydraulic gradients were calculated in 2023:

The gradient across the southern portion of the Landfill Area was calculated as 0.012 foot per foot (ft/ft) on

November 20, 2023, using wells SHL-15 (Upgradient Area) and PZ-12-10 (Barrier Wall Area).

The gradient across the northern portion of the Landfill Area was calculated as 0.009 ft/ft on November 20, 2023, using wells SHM-10-15 (Landfill Area) and SHL-23 (Nearfield Area).

The gradient across the Nearfield Area ranged from 0.002 ft/ft (June 12, 2023) using wells EPA-PZ-2012-1A and SHP-2016-2A (Nearfield Area) to 0.005 ft/ft (November 20, 2023) using wells SHM-10-06A and

SHM-10-16 (Nearfield Area).

The gradient across the NIA ranged from 0.003 ft/ft (June 12, 2023) using wells SHM-05-40X and SHM-13-08

(NIA) to 0.004 ft/ft (November 20, 2023) using wells SHM-05-40X and SHM-13-03 (NIA), respectively.

The gradient across the Barrier Wall Area was calculated as 0.088 ft/ft on June 12, 2023, and 0.101 ft/ft on

November 20, 2023, using wells PZ-12-05 and PZ-12-06 (Barrier Wall Area).

In general, the horizontal hydraulic gradient is steepest across the Landfill Area and across the barrier wall

(typically two to three times greater in 2023) compared to other areas, and the gradient across the NIA is comparable to the gradient of the Nearfield Area.

The four sets of overburden and bedrock well pairs gauged during the November 20, 2023, monitoring event were assessed with respect to the presence of vertical gradients, as shown in Table 5. The vertical gradient observations are summarized below:

A positive vertical gradient (downward flow component) was observed in the Upgradient Area from overburden to bedrock. This is consistent with recent historical observations. The Upgradient Area cluster had a vertical gradient equal to that of the horizontal gradient in the southern part of the landfill (Table 5).

A negative vertical gradient (upward flow component) was observed in the Barrier Wall Area near Red Cove.

Gradients have fluctuated between positive and negative over the last six fall gauging events (-0.006 to

0.002), most likely due to the relative degree of groundwater discharge from the bedrock towards Plow Shop

Pond, compared to the relative degree of influence from the water levels in Plow Shop Pond at the time of gauging.

A negative vertical gradient (upward flow component) was observed in the Nearfield Area. The gradient in this area has fluctuated between positive/downward flow component in previous fall gauging events. These fluctuations are likely due to changes in precipitation from year to year and/or flow rates at the extraction wells at the time of gauging. At overburden triplet monitoring well SHM-05-41A/B/C, groundwater flow had both an upward and downward component from 64 ft bgs toward the 44 ft bgs and 93 ft bgs intervals (Table 5).

In the North Impact Area, the vertical gradient was two-times greater than the horizontal gradient in the landfill indicating primarily vertical flow (Table 5).

2.2 Groundwater Sampling

A total of 62 and 121 well locations were sampled in spring and fall 2023, respectively. All locations were sampled in accordance with the LTMMP Update, LTMMP Addendum, and QAPP (Sovereign 2015; KGS 2018; S-A JV

2020, respectively).

Samples were collected in accordance with the procedures outlined in the Low Stress (low flow) Purging and

Sampling Procedure for the Collection of Groundwater Samples from Monitoring Wells Revision 3 and 4

(USEPA 2017). Dissolved metals and dissolved organic carbon samples were field-filtered using a

0.45-micrometer polyethersulphone filter. Groundwater was purged using a bladder pump or peristaltic pump, depending on the measured water level. Groundwater was purged through a calibrated YSI multi-meter and a turbidity meter to monitor groundwater chemistry parameters, and samples were collected upon achievement of field parameter stability. Calibration records are included in Appendix A.

Groundwater samples were submitted to Eurofins TestAmerica Laboratories (Eurofins) of Savannah, Georgia for the following analyses: dissolved metals (arsenic, iron, and manganese), chloride/sulfate, total alkalinity, and dissolved organic carbon. In accordance with the QAPP (S-A JV 2020), quality assurance / quality control (QC) procedures included the collection of field duplicate samples, matrix spike (MS)/matrix spike duplicate (MSD) samples, and equipment blank samples.

2.2.1 2023 Data Summary

The spring and fall 2023 sampling results are summarized in Tables 6 and 7, respectively. A summary of historical groundwater sampling results is provided in Appendix E. Figures 8 and 9 present dissolved arsenic concentration results for the spring and fall 2023 sampling events, respectively. The following is a summary of dissolved metals results from the sampling events conducted in 2023:

Spring 2023 Sampling Event

Dissolved arsenic concentrations in groundwater exceeded the CL of 10 µg/L at 41 of the 62 sampled locations (66%, compared to 63% in 2022), with the highest concentration of 3,100 µg/L detected in extraction well EW-04 (Nearfield Area).

Dissolved iron concentrations in groundwater exceeded the CL of 9,100 µg/L at 30 of the 62 sampled locations (48%, compared to 44% in 2022), with the highest concentration of 74,000 µg/L detected in well

SHL-11 (Barrier Wall Area, east of Landfill Area, north of the barrier wall).

Dissolved manganese concentrations in groundwater exceeded the CL of 1,715 µg/L at 20 of the 62 sampled locations (32%, equal to that in 2022), with the highest concentration of 10,000 µg/L detected in well

EPA-PZ-2012-5B (Nearfield Area).

Fall 2023 Sampling Event

Dissolved arsenic concentrations in groundwater exceeded the CL of 10 µg/L at 67 of the 121 sampled locations (55%, compared to 58% in 2022), with the highest concentration of 7,400 µg/L detected in well

SHM-10-15 (Landfill Area).

Dissolved iron concentrations in groundwater exceeded the CL of 9,100 µg/L at 52 of the 121 sampled locations (43%, equal to that in 2021), with the highest concentration of 79,000 µg/L detected in well

PZ-12-02 (Barrier Wall, east of Landfill Area, west of the barrier wall).

Dissolved manganese concentrations in groundwater exceeded the CL of 1,715 µg/L at 44 of the

121 sampled locations (36%, compared to 34% in 2022), with the highest concentration of 10,000 µg/L detected in well SHM-13-02 (Northern Impact Area).

2.2.2 Arsenic Concentration Trends

Arsenic concentration trends over time were performed at 61 wells specified in the LTMMP Update (Sovereign

2015). Criteria for performing the statistical analyses include using wells that have a minimum of four rounds of data available and have dissolved arsenic concentrations greater than or equal to 10 µg/L. Of the 61 wells specified in the LTMMP Update, a total of 23 wells (37%) had arsenic concentrations less than the CL in 2023, and a total of 38 wells (62%) exceeded the CL in 2023.

Arsenic data for the subset of 38 wells exceeding the CL in 2023 were evaluated using the Mann-Kendall trend test available in the ProUCL statistical software for environmental applications. The wells were evaluated to identify whether there are statistically significant trends for arsenic concentrations over time at each location at a

95% confidence level.

The evaluation identified that of the 38 wells statistically tested, 23 wells (61%) exhibited statistically significant decreasing concentrations, 13 wells (34%) exhibited no statistically significant trend, and two wells (5%;

SHM-10-07 and SHM-10-11) exhibited increasing concentrations at the selected 95% confidence level.

Monitoring wells SHM-10-07 and SHM-10-11 are in the east-central portion and southern portion of the Landfill

Area, respectively. Results are shown in Table 8, below. Figure 10 shows the arsenic concentration trend results and identifies locations included in the LTMMP Update (Sovereign 2015) for trend evaluation that exhibited arsenic concentrations less than the CL in 2023. ProUCL outputs for each well are included in Appendix F.

Table 8 2023 Mann-Kendall Trend Analysis Results

Arsenic Concentration

Below the CL of 10 µg/L in

Statistically Significant

Evidence of a

Decreasing Trend

Insufficient Evidence of a Statistically

Significant Trend

Statistically Significant

Evidence of an

Increasing Trend

EPA-PZ-2012-1A EPA-PZ-2012-1B EPA-PZ-2012-3A SHM-10-07

EPA-PZ-2012-2A EPA-PZ-2012-3B EPA-PZ-2012-5B SHM-10-11

EPA-PZ-2012-2B EPA-PZ-2012-4B EPA-PZ-2012-7B

EPA-PZ-2012-4A EPA-PZ-2012-6B SHL-5

EPA-PZ-2012-5A N5-P1 SHL-9

EPA-PZ-2012-6A SHL-22 SHM-10-06A

EPA-PZ-2012-7A SHM-05-40X SHM-10-12

SHL-8S SHM-05-41A SHM-10-14

SHL-8D SHM-05-41B SHM-10-15

SHL-23 SHM-05-41C SHM-10-16

SHM-05-42A SHM-05-42B SHM-13-05

SHM-10-02 SHM-10-06 SHM-93-22C

SHM-10-03 SHM-10-13 SHM-13-03a

SHM-10-04 SHM-13-04

SHM-10-05A SHM-13-06

SHM-10-08 SHM-13-07

SHM-10-10 SHM-13-08

SHM-13-01 SHM-93-22B

SHM-13-02 SHM-96-5Bb

SHM-13-14S SHM-96-5C

SHM-13-14D SHM-99-31C

SHM-13-15 SHM-99-32X

SHP-99-29X

Notes:

a. SHM-13-03 was incorrectly listed twice in Data Quality Objective Step 5 of the LTMMP Update as SHP-13-03.

b. SHM-96-5B was incorrectly listed in Data Quality Objective Step 5 of the LTMMP Update as SHL-96-5B.

2.2.3 Groundwater Monitoring Field Parameters

Water quality parameters (including pH, specific conductivity, dissolved oxygen [DO], temperature, and oxidation-reduction potential [ORP]) were collected during each of the 2023 sampling events. ORP and DO are particularly significant field parameters at SHL, as they can be used as indicators of oxidation-reduction (redox) conditions in the aquifer (ITRC 2022). Under the reducing conditions present at much of the site, naturally occurring arsenic associated with iron and manganese oxyhydroxide minerals is released into solution due to reductive dissolution.

For example, ferric iron present in the oxides is reduced to ferrous iron, which is more soluble. Similarly, solid manganese (Mn4+) minerals are reduced to more soluble manganese (Mn2+). These dissolution processes also release arsenic associated with the minerals; and the iron, manganese, and arsenic can then be detected in dissolved metals analyses. Microbial reduction of iron and manganese (frequently coupled to organic matter oxidation) is most favorable in the absence of oxygen, and dissolved oxygen can reoxidize iron and manganese;

therefore, low DO is expected in zones with elevated iron and manganese. The ORP is a measure of the presence of redox-sensitive species in solution (including iron and manganese). Therefore, higher concentrations of iron and manganese would be expected in locations where low/negative ORP values and low DO concentrations are measured (Gannett Fleming, Inc. 2011). There are two sources of carbon reducing conditions at SHL. Historically, the peat and wetlands underlying the landfill likely provided reducing conditions and mobilized arsenic through reductive dissolution of iron hydroxyoxides on which arsenic was entrained (KGS

2016). In addition, the presence of the cap deters the percolation of oxygenated waters enhancing the reducing conditions in the aquifer.

Metals concentrations and associated field ORP measurements from the spring and fall 2023 groundwater sampling events are shown in Tables 6 and 7. The summary of correlations presented below indicates that dissolved metals in groundwater at SHL are more likely to exceed the CL when the redox state in groundwater is reducing and that the redox condition of the aquifer influences whether groundwater meets or exceeds the CLs for arsenic, iron, and manganese.

Spring 2023 Sampling Event

Of the 62 groundwater samples collected and analyzed for arsenic during the spring 2023 event, 41 samples had concentrations exceeding the CL for arsenic and 21 samples had concentrations that either were not detected above the limit of detection or were detected below the CL:

Of the 21 samples that met the CL for arsenic, DO and ORP measurements indicate that 11 samples

(52%) were collected from locations where the aquifer was oxidizing (ORP greater than 50 millivolts [mV] and DO greater than 1 milligram per liter [mg/L]), 5 samples (24%) were collected from locations screened in groundwater that was moderately reducing (where either ORP was greater than 50 mV or DO was greater than 1 mg/L), and 4 samples (19%) were collected at locations with groundwater that was reducing (ORP less than 50 mV and DO less than 1 mg/L).

In comparison, of the 41 samples that exceeded the CL for arsenic, samples were generally collected from locations with more strongly reducing geochemical conditions: no samples were collected from oxidizing conditions, six samples (15%) from moderately reducing conditions, and 35 samples (85%) where the aquifer was reducing.

Of the samples that met the CL for arsenic, the average ORP value was 80.9 mV (median of 90.3 mV) and the average DO concentration was 3.63 mg/L (median of 2.88 mg/L).

In comparison, samples that exceeded the CL for arsenic had a lower ORP and DO concentration: an average ORP of -67.3 mV (median of -71.3 mV) and an average DO concentration of 1.00 mg/L (median of 0.73 mg/L).

Locations where iron and/or manganese concentrations exceeded the CL generally overlapped with locations that exceeded the CL for arsenic, indicating reducing conditions in groundwater has affected all three of these naturally occurring metals:

Of the 30 samples that exceeded the CL for iron in spring 2023, 29 samples (97%) also exceeded the CL for arsenic.

Of the 20 samples that exceeded the CL for manganese in spring 2023, 19 samples (95%) also exceeded the CL for arsenic.

Fall 2023 Sampling Event

Of the 121 groundwater samples collected and analyzed for arsenic during the fall 2023 sampling event, 67 samples had concentrations exceeding the CL for arsenic and 54 samples had concentrations that either were not detected above the limit of detection or were detected at concentrations equal to or below the CL:

Of the 54 samples that met the CL for arsenic, DO and ORP measurements indicate that 21 samples

(39%) were collected from locations where the aquifer was oxidizing, 21 samples (39%) were collected from locations screened in groundwater that was moderately reducing, and 12 samples (22%) were collected at locations with groundwater that was reducing.

In comparison, of the 67 samples that exceeded the CL for arsenic, samples were collected from locations with more strongly reducing geochemical conditions: one sample (2%) from groundwater under oxidizing conditions, 15 samples (22%) from moderately reducing conditions, and 51 samples (76%) where the aquifer was reducing.

Of the samples that met the CL for arsenic, the average ORP value was 73.6 mV (median of 86.3 mV) and the average DO concentration was 2.64 mg/L (median of 1.28 mg/L).

In comparison, samples that exceeded the CL for arsenic had a lower ORP and DO concentration: an average ORP of -53.8 mV (median of -55.0 mV) and an average DO concentration of 0.69 mg/L (median of 0.49 mg/L).

Locations where iron and/or manganese concentrations exceeded the CL generally overlapped with locations that exceeded the CL for arsenic, indicating reducing conditions in groundwater has affected all three of these naturally occurring metals:

Of the 52 samples that exceeded the CL for iron in fall 2023, 51 samples (98%) also exceeded the CL for arsenic.

Of the 44 samples that exceeded the CL for manganese fall 2023, 34 samples (77%) also exceeded the

CL for arsenic.

2.2.4 Quality Assurance/Quality Control

Eurofins was the primary contract laboratory used for the spring and fall 2023 LTM sampling events. Eurofins is compliant with the United States Department of Defense (DoD) and Department of Defense (DOE) Consolidated

Quality Systems Manual for Environmental Laboratories, Version 5.4 (Quality Systems Manual; DoD and DOE

2021) under the DoD Environmental Laboratory Accreditation Program and holds current National Environmental

Laboratory Accreditation Program accreditation for applicable analytical methods.

As part of the 2023 groundwater sampling events, quality assurance/QC samples were collected to confirm that the data quality objectives were met for the laboratory analysis. Field duplicate samples were collected during sampling events to evaluate field precision at the rate of 10% (1 per 10 samples). MS/MSD pairs were also submitted at a rate of 5% (1 per 20 samples) to evaluate matrix effects on analytical precision and accuracy.

Analytical results from the 2023 sampling events were reviewed and evaluated for data acceptability by

Laboratory Data Consultants, Inc. of Carlsbad, California in accordance with the USEPA Region 1 Environmental

Data Review Program Guidance (USEPA 2018), the QAPP (S-A JV 2020), and the DoD General Data Validation

Guidelines (DoD 2019). The method requirements from the Quality Systems Manual (DoD and DOE 2021) and the USEPA Test Methods for Evaluating Solid Waste, Physical/Chemical Methods, SW-846 (USEPA 2014) were also used as supplemental information.

Laboratory Data Consultants, Inc. of Carlsbad, California reviewed and updated the database with appropriate data qualifiers as needed. Laboratory analyses were confirmed to have been performed in general compliance with the precision, accuracy, representativeness, completeness, comparability, and sensitivity requirements listed in the QAPP (S-A JV 2020). Sample results that were qualified as estimated (UJ/J) due to QC exceedances are usable with caution; no results were rejected. Analytical results reported for the 2023 LTM samples were deemed valid and usable for intended purposes. The data validation reports and a summary of quality control exceptions noted for the SHL 2023 LTM sampling data are presented in Appendix C.

2.3 Conclusions

The results of the spring monitoring event and the more comprehensive fall 2023 site-wide monitoring event continue to illustrate a general groundwater flow direction from the west/southwest to the north toward

Nonacoicus Brook, with a deflection of groundwater flow to the north in the area west of the barrier wall near Red

Cove at Plow Shop Pond. Concentrations of arsenic, iron, and manganese in groundwater continue to exceed

CLs in various wells at SHL and in the NIA (Tables 6 and 7). Locations where dissolved metals (arsenic, iron, and manganese) concentrations exceed the CL are generally under more reducing geochemical conditions than locations where dissolved metals concentrations meet the CL.

The groundwater monitoring program will continue in accordance with the LTMMP Update and LTMMP

Addendum (Sovereign 2015; KGS 2018, respectively) until a new Long-Term Monitoring and Maintenance Plan

(LTMMP) is prepared.

The 2023 scope included the following additional work requested by USEPA in addition to what is included in the

LTMMP Update and LTMMP Addendum:

Continued collection of geochemical and hydrologic data from the nested piezometers (SHP 2016-1A/B through SHP 2016-5A/B) in spring and fall.

Continued collection of geochemical and hydrologic data for USEPA-monitoring network added wells

(SHM-05-39A/B, SHM-99-31A/B).

Continued collection of water level data from piezometers SHP-2017-01 and SHP-2017-02 located near the extraction wells in spring and fall.

Continued collection of geochemical and hydrologic data from bedrock wells SHP-2016-06A/B/C and

SHP-2016-07A/B in spring and fall.

Continued collection of hydrologic data from wells SHM-10-11, N6-P1, and SHP-99-35X in spring and fall to provide more hydraulic control in the southern portion of the landfill.

Geochemical and hydrologic data was collected from 10 newly installed monitoring wells (SHP-2023-1A,B;

SHP-2023-2A,B; SHP-2023-3A,B; SHP-2023-4A,B, SHP-2023-5A,B) and one newly installed extraction well

(EW-03) in the fall as part of the effort to re-evaluate and optimize the capture of the arsenic plume

(USACE 2022, S-A JV 2023).

The Army is currently conducting an assessment to confirm the effectiveness of the barrier wall installed adjacent to Red Cove. The results of the assessment will be reported under a separate cover and will be considered as part of the LTMMP update.

3 Landfill Maintenance and Monitoring In accordance with the LTMMP Update (Sovereign 2015), SHL was inspected, and routine maintenance was performed during 2023. All landfill maintenance and monitoring activities were completed by Sovereign.

3.1 Landfill Inspection

The annual inspection identifies and corrects problems pertaining to the effectiveness of the cap system, erosional controls, gas vents, and sampling points. The landfill inspection report with supporting photographs is presented in Appendix G. A summary of the report findings is provided below:

The surface cover had no new depressions. No tree or shrub growth was observed on the landfill surface during the inspection. Small tree and shrub growth on the margins of the landfill were removed during the annual mowing event.

An annual inspection of the landfill and adjacent areas was conducted on October 19, 2023. The overall condition of the landfill, utility berm, and culverts was found to be satisfactory. Vegetation atop the utility berm was mowed during the annual mowing event between October 3 and October 5, 2023. The southern drainage swale exhibited large vegetative growth that was removed during annual mowing activities. Small growth and wetland plant life were not disturbed, as they have become a natural retardant to erosive forces. Vegetation in the northern drainage swales was also removed during annual mowing activities. No new topographic depressions or significant erosion was observed.

As noted in previous reports, gas vent GV-3 continues to be leaning at a 15 to 20 angle off-center. This includes the concrete pad the vent is set in. The vent is still firmly in the ground, is stable, and does not rock or move in any way. There is no evidence that the vent was struck by a vehicle, and the vent still appears to be functional and appears to be in otherwise good condition (see photograph in Appendix G). The remaining landfill gas points (LGPs) and gas vents (Figure 11) were in good condition.

Monitoring wells and piezometers located on the landfill cap or perimeter that are currently part of the LTM network were observed to be in good condition overall.

The landfill access road was observed to be in good condition.

Perimeter fencing was not installed historically along much of the western boundary of the landfill (wooded area along Shepley’s Hill) and along the southeastern boundary near the railroad tracks. No roads have open access to the landfill and the fence gates across roads that access the landfill are secured with chains and padlocks.

Wetland encroachment was observed in the swales located in the southern portion of the landfill. Overall, the areas of encroachment are small and confined to the swales.

The recreational field located south of the access road in the middle of the southern half of the landfill appeared to be in good condition. There was no tire rutting observed between the field and the landfill access road.

Sovereign provided the following recommendations for 2024:

Continue monitoring for settlement that creates depressions.

Continue monitoring for tree and shrub growth on the surface cover. Remove tree/shrub growth as necessary.

Continue monitoring the utility berm for negative impacts on the drainage patterns.

Continue annual inspections and mowing of the landfill cap, utility berm, drainage swales, and culverts

(including removal of tree and shrub growth).

Continue to monitor GV-3 to ensure the angle of the vent does not worsen. Continue monitoring the condition of the remaining LGPs and monitoring wells.

Continue monitoring for erosion and depressions/settlement within the landfill and repair, as necessary.

Conduct routine checks of the recreational field to confirm the integrity of the landfill cover, gas vents, and monitoring wells in this area.

Continue to monitor road condition and repair as needed.

3.2 Landfill Gas Monitoring

Sovereign completed annual sampling of landfill gas vents and LGPs (Figure 11) on October 19, 2023. Sampling activities were conducted in accordance with procedures described in the MassDEP Landfill Technical Guidance

Manual, as revised May 1997 (MassDEP 1997):

Landfill gas monitoring is conducted by first capping off the gas vents and connecting the handheld monitoring devices to the sample port on the cap for the initial readings.

Following collection of the initial readings, an adjustable flow rate sampling pump is connected to the sample port on the cap. Two vent volumes are purged from the gas vents using the sampling pump, and the post-purge monitoring readings are collected with the handheld monitoring device.

Each handheld monitoring device is equipped with an internal pump and is allowed to operate for approximately one minute before recording the monitoring data.

Landfill gas sampling was conducted with a multi-gas meter and landfill gas meter and included the following parameters: volatile organic compounds (VOCs), oxygen, hydrogen sulfide, lower explosive limit (LEL), carbon monoxide, carbon dioxide, and methane. Each parameter was measured and recorded twice (initial reading and post-purge reading). Gas monitoring results (Table G-1 and Table G-2) and gas probe construction details

(Table G-3) are provided in Appendix G. Results appear to be generally consistent or lower than historical observations. The landfill gas monitoring results are summarized below:

Field readings of barometric pressure were variable and ranged from 29.70 to 30.45 inches of mercury. The background barometric pressure ranged from 29.88 at the start of the day to 30.45 at mid-day and decreased to 29.70 at the end of the day.

VOC detections ranged from 0.0 to 1.4 parts per million (ppm) in initial readings, and ranged from 0.0 to

0.6 ppm in post-purge readings. The highest initial reading of 1.4 ppm (gas probes LGP-09-08X) decreased to

0.5 ppm post-purge.

Oxygen ranged from 6.0 to 20.9% in initial readings; and 4.9 to 20.9% post-purge.

Hydrogen sulfide remained at 0.0 ppm for all initial and post purge readings.

The LEL ranged from 0.0 to 99% (initial) and 0.0 to 99% (post-purge).

Carbon monoxide remained at 0.0 ppm for all initial and post purge readings.

Carbon dioxide detections ranged from 0.0 to 15.5% (initial) and 0.0 to 0.7% (post-purge).

Methane detections ranged from 0.0 to 16.8% in initial readings; and 0.0 to 11.7% post-purge. The highest initial readings were detected at gas probe LGP-09-08X.

4 Arsenic Treatment Plant Operations, Maintenance, and Monitoring

The SHL ATP treated and discharged 19,026,200 gallons of groundwater between January 1 and

December 31, 2023 (Table G-4 and Table G-5), bringing the cumulative treated volume total to approximately

393,329,675 gallons since system startup in 2006 (Table G-6). All ATP operations, maintenance, and monitoring in 2023 was conducted by Sovereign; data are presented in the following sections.

4.1 System Description

The ATP is designed to remove dissolved arsenic from extracted groundwater through oxidation and precipitation of dissolved iron and co-precipitation of the arsenic. Until September 2023, the ATP consisted of the following components:

Extraction system (extraction wells EW-01 and EW-04)

Chlorine dioxide generation and addition in order to oxidize iron

Coagulation via a contact tank with a direct drive batch tank mixer

Microfiltration (MF) of the water to concentrate precipitated solids

Solids removal via an inclined plate clarifier

Bag filtration and discharge of the inclined plate clarifier decant water

Polymer-aided flocculation of sludge using a filter bed roll-off (FBRO)

Treated water discharge to the Devens publicly owned treatment works.

Historical monthly discharge totals and an operational summary of the monthly online hours, flow totals, and operating status during the first eight months of 2023 are presented in Appendix G.

In September and October 2023, the ATP was upgraded and modified to replace the existing onsite 17-year old chlorine dioxide generation and dosing system with a new potassium permanganate oxidation and clarification system to improve the performance and safety of the plant. Following completion of the upgrades, the system continued to utilize oxidation and co-precipitation of iron and arsenic as the primary treatment methodology, but system operations were significantly altered from those utilized between 2006 and the first half of 2023.

The upgraded ATP now consists of the following components (the changed processes are bolded).

Extraction system (extraction wells EW-01, EW-03, and EW-04)

Potassium permanganate addition to oxidize dissolved iron

Coagulation via contact tanks with a direct drive batch tank mixers

Precipitated solids separation and removal via new Lamella Gravity Settler inclined plate clarifiers

MF of the water exiting the Lamella clarifies to “polish” the decant water and concentrate residual precipitated solids

Tertiary solids removal via an inclined plate clarifier

Bag filtration and discharge of the inclined plate clarifier decant water

Polymer-aided flocculation of sludge using a filter bed roll-off (FBRO)

Treated water discharge to the Devens publicly owned treatment works

4.1.1 System Efficiency and Routine Maintenance

Routine Operations from January 1 through August 25, 2023

During the first eight months of 2023, prior to system upgrades, and during a period of typical operation, the treatment plant was operational approximately 89% of the total available hours during the calendar year

(i.e., “uptime”), which is similar to the uptime in 2022 of 88% (Table G-4). When online before September 2023, the plant achieved an effective average extraction rate (i.e., the extraction rate calculated by dividing the volume pumped by the elapsed time during the eight months, not the actual system online time) of 50.3 gpm in 2023

(compared to 48.05 gpm in 2022). The monthly effective pumping rates for 2023 are displayed in Chart 1 below.

Non-operational time during the period through August 2023 is associated with both routine and non-routine system maintenance, repairs, and upgrades (Section 4.1.2). The ATP MF system continued to require monthly clean-in-place (CIP) procedures to prevent long-term fouling of the filter membranes (Table G-8). Extended CIP events have been conducted only in situations when the level of filter fouling is greater than normal, most likely because of a non-optimal dosage concentration of chlorine dioxide.

The FBRO was emptied routinely (every 1 to 2 weeks) in 2023 due to high sludge generation from inorganics

(primarily iron) in the influent. This is consistent with historical FBRO maintenance. A total of 203.45 tons of sludge were removed in 2023 by Global Remediation Services, Inc. (Table G-7). The sludge removal process was optimized, which resulted in zero downtime attributable to FBRO operations for the ATP in 2023.

ATP System Upgrades and Testing – September through December 2023

The system was offline from August 25 to October 31, 2023, for system upgrade activities, with minor on periods for system testing, and during this period during which time almost no groundwater was extracted and treated. In

November and December 2023, the newly upgraded system was operated intermittently during startup and shakedown in an effort to optimize the performance of the revised system. During November and December, the system extracted and treated groundwater at effective pumping rates of approximately 17 and 25 gpm, respectively, as shown in Chart 1 below, and groundwater was extracted at varying rates from EW-01, EW-03, and EW-04 as part of system testing. Also during this period, the FBRO was again emptied routinely (every

2 weeks), and due to reduced filter loading as a result of the upgrades, no CIP events were conducted as well.

Note: System offline between August 25 and October 31, 2023 for upgrades.

4.1.2 Non-Routine System Maintenance, Repairs, and Upgrades

This section details major system non-routine…