Long-Term Monitoring Network
Optimization Evaluation
for
Wash King Laundry Superfund Site
Lake County, Michigan
\
June 2006
-------
Solid Waste and EPA 542-R-06-004
Emergency Response December 2006
(5102P) www.epa.gov
Long-Term Monitoring Network
Optimization Evaluation
for
Wash King Laundry Superfund Site
Lake County, Michigan
June 2006
-------
FINAL
LONG-TERM MONITORING NETWORK
OPTIMIZATION EVALUATION
FOR
WASH KING LAUNDRY SUPERFUND SITE
LAKE COUNTY, MICHIGAN
June 2006
-------
TABLE OF CONTENTS
Page
LIST OF ACRONYMS iv
SECTION 1 -INTRODUCTION 1-1
SECTION 2 - SITE BACKGROUND INFORMATION 2-1
2.1 Site Location and Operational History 2-1
2.2 Environmental Setting 2-2
2.2.1 Topography 2-2
2.2.2 Geology 2-3
2.2.3 Hydrogeology 2-3
2.2.4 Surface Water Hydrology 2-6
2.3 Nature and Extent of Contamination 2-6
2.3.1 Shallow Portion of the Surficial Aquifer 2-9
2.3.2 Deep Portion of the Surficial Aquifer 2-12
2.3.3 Middle Branch Pere Marquette River 2-12
SECTION 3 - LONG-TERM MONITORING PROGRAM AT WASH KING 3-1
3.1 Description of Monitoring Program 3-1
3.2 Summary of Analytical Data 3-4
SECTION 4 - QUALITATIVE LTMO EVALUATION 4-1
4.1 Method for Qualitative Evaluation of Monitoring Network 4-2
4.2 Results of Qualitative LTMO Evaluation for Groundwater 4-3
4.2.1 Extraction Wells 4-3
4.2.2 Monitoring Wells Screened in the Shallow Portion of the
Surficial Aquifer 4-9
4.2.3 Monitoring Wells Screened in the Deep Portion of the Surficial
Aquifer 4-11
4.3 Data Gaps 4-13
4.3.1 Shallow Portion of the Surficial Aquifer 4-13
4.3.2 Deep Portion of the Surficial Aquifer 4-14
4.4 Analytical Program 4-15
4.5 LTM Program Flexibility 4-15
-i-
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
TABLE OF CONTENTS (Continued)
Page
SECTION 5 - TEMPORAL STATISTICAL EVALUATION 5-1
5.1 Methodology for Temporal Trend Analysis of Contaminant Concentrations 5-1
5.2 Temporal Evaluation Results for Groundwater Wells 5-4
SECTION 6 - SPATIAL STATISTICAL EVALUATION 6-1
6.1 Geostatistical Methods for Evaluating Monitoring Networks 6-1
6.2 Spatial Evaluation of the Monitoring Network at Wash King 6-3
6.3 Spatial Statistical Evaluation Results 6-5
SECTION 7 - SUMMARY OF LONG-TERM MONITORING
OPTIMIZATION EVALUATION 7-1
7.1 Groundwater Monitoring Network Summary 7-1
SECTION 8 - REFERENCES 8-1
Appendix A - Comments and Responses on Draft Report
LIST OF TABLES
No. Title Page
2.1 Basecase Groundwater Monitoring Program 2-7
3.1 PCE Data Date Distribution 3-3
3.2 Summary of Occurrence of Groundwater Contaminants of Concern 3-5
3.3 Most Recent Groundwater COC Concentrations 3-6
4.1 Monitoring Network Optimization Decision Logic 4-2
4.2 Monitoring Frequency Decision Logic 4-3
4.3 Qualitative Evaluation of Groundwater Monitoring Network 4-4
5.1 Temporal Trend Analysis of Groundwater Monitoring Results 5-7
6.1 Best-Fit Semivariogram Model Parameters 6-4
6.2 Results of Geostatistical Evaluation Ranking of Wells by Relative Value
of PCE in Shallow Zone Wells 6-7
6.3 Results of Geostatistical Evaluation Ranking of Wells by Relative Value
of PCE in Deep Zone Wells 6-8
7.1 Summary of Long Term Monitoring Optimization Evaluation of Wash
King Groundwater Monitoring Program 7-2
7.2 Summary of Revised and Basecase Monitoring Programs 7-9
-11-
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
TABLE OF CONTENTS (Continued)
LIST OF FIGURES
No. Title Page
2.1 Potentiometric Surface for Shallow Zone 2-4
2.2 Potentiometric Surface for Deep Zone 2-5
2.3 Relative Well Screen Intervals 2-8
2.4 Most Recent Total Chlorinated Ethene Concentrations in the Shallow
Zone 2-10
2.5 Most Recent Total Chlorinated Ethene Concentrations in the Deep Zone 2-11
3.1 Monitoring and Extraction Wells 3-2
4.1 Qualitative Evalution Sampling Frequency Recommendations for
Shallow Zone Wells 4-7
4.2 Qualitative Evalution Sampling Frequency Recommendations for Deep
Zone Wells 4-8
4.3 OWE PCE Concentrations over Time 4-10
5.1 PCE Concentrations Through Time at Well MW-212D 5-2
5.2 Conceptual Representation of Temporal Trends and Temporal Variations
in Concentrations 5-3
5.3 Conceptual Representation of Continued Monitoring at Location Where
No Temporal Trend in Concentrations is Present 5-5
5.4 Temporal Trend Decision Rationale Flowchart 5-6
5.5 Shallow Zone Temporal Trend Results for PCE in Groundwater 5-9
5.6 Deep Zone Temporal Trend Results for PCE in Groundwater 5-10
6.1 Idealized Semivariogram Model 6-3
6.2 Impact of Missing Wells on Predicted Standard Error 6-6
6.3 Geostatistical Evaluation Results Showing Relative Value of Spatial
Information of PCE Distribution in Shallow Wells 6-9
6.4 Geostatistical Evaluation Results Showing Relative Value of Spatial
Information of PCE Distribution in Deep Wells 6-10
7.1 Combined Evalution Recommendations for Shallow Zone Wells 7-5
7.2 Combined Evalution Recommendations for Deep Zone Wells 7-6
7.3 Combined Evalution Summary Decision Logic 7-7
-111-
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
LIST OF ACRONYMS
am si
bgs
CERCLA
COC
DCE
ESRI
ft/day
ft/ft
ft/yr
GIS
OWE
LTM
LTMO
MCL
MDEQ
MDNR
mg/L
MNO
NAPL
ND
PCE
RAO
RI
ROD
TCE
USEPA
UST
VOCs
microgram(s) per liter
above mean sea level
below ground surface
Comprehensive Environmental Response, Compensation, and
Liability Act
contaminant of concern
dichloroethene
Environmental Systems Research Institute, Inc.
foot per day
foot per foot
feet per year
geographical information system
groundwater extraction
long-term monitoring
long-term monitoring optimization
maximum contaminant level
Michigan Department of Environmental Quality
Michigan Department of Natural Resources
milligrams per liter
monitoring network optimization
non aqueous-phase liquid
not detected
tetrachloroethene
remedial action objective
remedial investigation
record of decision
trichloroethene
United States Environmental Protection Agency
underground storage tank
volatile organic compounds
-IV-
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
SECTION 1
INTRODUCTION
Groundwater monitoring programs have two primary objectives (U.S. Environmental
Protection Agency [USEPA], 1994; Gibbons, 1994):
1. Evaluate long-term temporal trends in contaminant concentrations at one or
more points within or outside the remediation zone as a means of monitoring
the performance of the remedial measure (temporal objective) and
2. Evaluate the extent to which contaminant migration is occurring, particularly if
a potential exposure point for a susceptible receptor exists (spatial objective}.
The relative success of any remediation system and its components (including the
monitoring network) must be judged based on the degree to which it achieves the stated
objectives of the system. Designing an effective groundwater monitoring program
involves locating monitoring points and developing a site-specific strategy for
groundwater sampling and analysis to maximize the amount of relevant information that
can be obtained while minimizing incremental costs. Relevant information is that
required to effectively address the temporal and spatial objectives of monitoring. The
effectiveness of a monitoring network in achieving these two primary objectives can be
evaluated quantitatively using statistical techniques. In addition, there may be other
important considerations associated with a particular monitoring network that are most
appropriately addressed through a qualitative assessment of the network. The qualitative
evaluation may consider such factors as hydrostratigraphy, locations of potential receptor
exposure points with respect to a dissolved contaminant plume, and the direction(s) and
rate(s) of contaminant migration.
This report presents a description and evaluation of the groundwater monitoring
program associated with the Wash King Superfund Site located in Pleasant Plains
Township, Lake County, Michigan. This report does not address the larger issue of
remedial process optimization for this site. A monitoring network consisting of 44
groundwater monitoring wells and five groundwater extraction wells was evaluated to
identify potential opportunities to streamline monitoring activities while still maintaining
an effective monitoring program. A three-tiered approach, consisting of a qualitative
evaluation, a statistical evaluation of temporal trends in contaminant concentrations, and
a spatial statistical analysis assessed the degree to which the monitoring network
addresses the objectives of the monitoring program, as well as other important
considerations. The qualitative evaluation addressed all 49 monitoring and extraction
wells. The temporal evaluation addressed those wells with adequate historical analytical
data (>4 sampling events) to conduct a trend analysis, and the spatial statistical
1-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
evaluations included separate evaluations for those wells screened in the shallow and
deep aquifers. The results of the three evaluations were combined and used to assess the
optimal frequency of monitoring and the spatial distribution of the components of the
monitoring network. The results of the analysis were then used to develop
recommendations for optimizing the monitoring program at Wash King.
1-2
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
SECTION 2
SITE BACKGROUND INFORMATION
The location, operational history, environmental setting (i.e., geology, hydrogeology,
and surface water hydrology), and remediation history of Wash King are briefly
summarized in the following subsections. This information was derived primarily from
published and unpublished information received from the Michigan Department of
Environmental Quality (MDEQ) and the Record of Decision (ROD) prepared for the site
in 1993 (USEPA, 1993).
2.1 SITE LOCATION AND OPERATIONAL HISTORY
The Wash King Laundry site is located south of the city of Baldwin in Pleasant Plains
Township, Lake County, Michigan. The site is bordered on the east by a line
approximately 300 feet east of highway M-37, on the south by Star Lake Road (76th
Street), on the west by the C&O Railroad, and on the north by the Middle Branch Pere
Marquette River. At the time that the ROD was published in 1993, the Pere Marquette
Subdivision Plat, which comprises the site, included 123 residential lots, most of which
were not used on a year-round basis. Housing in the area consisted primarily of mobile
homes, trailers, and cottages. Numerous commercially developed lots existed along
Highway M-37. Current land use conditions are not known.
The former Wash King Laundry was granted permission to discharge soapy laundry
wastewater to four nearby unlined seepage lagoons in 1962. The lagoons were located
approximately 500 feet west of the laundry building in a wooded area. Dry cleaning
services later supplemented laundry operations, and spent dry cleaning solvent
(tetrachloroethene [PCE]) was discharged to the lagoons in the 1970s. All dry cleaning
operations ceased in 1978, but detergent laundry operations continued, with lagoon
discharge of the wash water, until 1991, when the owner filed for bankruptcy.
An underground storage tank (UST) was located approximately 20 feet south of the
former Laundromat in September 1999. The tank contained approximately 170 gallons
of fluid, believed to be mostly old boiler fuel and water; however, the possibility of
solvent contamination could not be ruled out. The fluid was pumped into drums and
disposed of off-site.
A chronological summary of investigative and remedial activities performed at the site
is provided below.
August 1973: Laundry detergent wastes and PCE were first detected in the
groundwater at the site via sampling of nearby water wells.
2-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
1976: Further groundwater contamination was discovered.
1977: Additional investigations were performed by the Michigan Department of
Natural Resources (MDNR).
1979-1980: Additional investigations performed to clearly show that the laundry
facility was the source of the PCE contamination (MDNR, 1980).
1982: The extent of groundwater contamination was further documented via
additional investigative activities.
1988: A remedial investigation (RI) was initiated to define the nature and extent of
contamination at the site and characterize the potential threats to public heath and the
environment.
1992: A baseline risk assessment and feasibility study were completed.
1993: A ROD was issued presenting the selected remedial actions for the site. The
selected remedy for groundwater consists of extraction and ex-situ treatment of
groundwater, deed restrictions, and long-term monitoring (LTM). The ROD states that
treated water would be discharged to the Middle Branch Pere Marquette River; however,
treated water is actually discharged to the seepage lagoons. The selected remedy for
contaminated sediments/soils within the lagoon consisted of excavation and off-site
disposal. In addition, a soil vapor extraction system was installed to remediate volatile
organic compounds (VOCs) in vadose zone soils; the dates of operation of this system
and its current status are not known. Available data indicate that the original six-well
groundwater extraction system (EW-1 through EW-6) was installed in the second half of
2000, and analytical results for the six original wells date back to April 2001. Extraction
well EW-5A was installed in December 2000. Both EW-5 and EW-5A are reportedly
pumped, but EW-5 is relatively low-yielding. EW-3 is reportedly not pumped. The
extraction system reportedly pumps at a total combined rate of approximately 250 gallons
per minute.
There is no current use of groundwater in the surficial aquifer by area residents or
businesses. All area water supply wells are screened in the lower sandy aquifer below the
clay aquitard; site-related contamination has not been detected in this lower aquifer.
2.2 ENVIRONMENTAL SETTING
2.2.1 Topography
The site is generally flat except for a steep embankment leading down to the Middle
Branch Pere Marquette River on the north side of the site. The ground surface elevation
at the majority of site wells ranges between 812 and 818 feet above mean sea level
(amsl). In contrast, the ground surface elevation near the river at wells MW-102S/D and
MW-202 is approximately 802 feet amsl.
2-2
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
2.2.2 Geology
The site soils are generally composed of fine- to medium-grained sands with some
clay and/or silt lenses to a depth of approximately 75 to 100 feet below ground surface
(bgs). The ROD (USEPA, 1993) indicates that these are predominantly glacial outwash
deposits. These deposits are underlain by a thicker clay layer that subdivides the shallow
sandy aquifer from a deeper, predominantly sandy aquifer that extends to a depth of
approximately 350 feet bgs. The lateral extent and continuity of the clay layer that forms
the base of the surficial aquifer is not well defined. The reported thickness of the clay
layer is variable, ranging from 20 to 56 feet.
2.2.3 Hydrogeology
Figures 2.1 and 2.2 depict the potentiometric surfaces for the shallow and deep
portions of the upper, unconfmed aquifer, respectively based on water level data collected
on April 6, 2006, while the groundwater extraction (GWE) system was operational.
Groundwater in the upper, unconfmed aquifer generally flows to the north-northeast,
discharging into the Middle Branch Pere Marquette River. It is anticipated that localized
cones of depression resulting from groundwater extraction are centered around the GWE
wells. These cones of depression are not evident in all cases on Figures 2.1 and 2.2, most
likely because the water level data set and contour interval used are not sufficiently
detailed. On May 11, 2001, the depth to groundwater in the surficial aquifer ranged from
approximately 17 to 30 feet bgs with an average depth of approximately 26 feet bgs. In
March 2005, the depth to groundwater in this aquifer ranged from approximately 13 to 31
feet bgs with an average depth of approximately 25 feet bgs. The average depths to
groundwater in December 2002, October 2003, and August 2004 were about 25 feet, 27
feet, and 25 feet bgs, respectively, indicating that seasonal fluctuations in the water table
are relatively minor.
The RI report states that the estimated average groundwater flow velocity in the upper
aquifer is 185 feet per year (ft/yr). Based on groundwater elevation data shown on Figure
2.1, the hydraulic gradient in the shallow portion of the surficial aquifer ranges from
approximately 0.003 to 0.008 foot per foot (ft/ft) (average 0.0055 ft/ft). The hydraulic
gradient in the deep portion of the surficial aquifer (Figure 2.2) had a similar range (0.004
to 0.007 ft/ft, average 0.0055 ft/ft). These gradients are the same order of magnitude as
that calculated from April 1989 data during the RI (0.004 ft/ft). Using the groundwater
flow velocity of 185 ft/yr (0.5 foot per day [ft/day]) presented in the ROD (USEPA,
1993), an estimated effective porosity for a predominantly sandy aquifer of 0.25, and the
average hydraulic gradient of 0.0055 ft/ft, the hydraulic conductivity of the shallow and
deep portions of the surficial aquifer is calculated to be approximately 23 ft/day.
Hydraulic conductivity values derived from slug tests performed in 11 monitoring wells
during the RI ranged from 0.9 to 340 ft/day, with an average value of 43 ft/day.
Comparison of water level elevations measured on 11 May 2001 in 10 monitoring well
pairs, each consisting of a shallow and deep well, indicate that both upward and
downward vertical hydraulic gradients were present on that date. The well pairs used in
the vertical gradient calculations included MW-3S/D, MW-8S/D, MW-101S/D, MW-
204S/D, MW-205S/D, MW-206S/D, MW-207S/D, MW-211S/D, MW-212S/D, and
MW-213S/D. Four of the 10 vertical gradients were calculated to be upwardly directed
2-3
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
-------
-------
(MW-204S/D, MW-205S/D, MW-206S/D, MW8S/D), with magnitudes ranging from
0.0003 ft/ft to 0.01 ft/ft (average 0.005 ft/ft). The remaining six gradients were
downwardly directed, with magnitudes ranging from 0.002 ft/ft to 0.027 ft/ft (average
0.015 ft/ft). It is likely that upward vertical gradients become more prevalent near the
Middle Branch Pere Marquette River given that groundwater in the surficial aquifer
reportedly discharges into the river. The four well pairs that exhibited upward vertical
gradients on 11 May 2001 are clustered nearer the river than the six pairs that exhibited
downward gradients, which were more widely distributed in the southern and central
portions of the site. These observations are consistent with data presented in the RI
report, which indicated a slight downward gradient, averaging 0.0016 ft/ft, in the
southern portion of the site, and a relatively strong upward gradient, averaging 0.022 ft/ft,
in the northern portion of the site near the Pere Marquette River.
2.2.4 Surface Water Hydrology
Site-specific information regarding the hydrology of the Middle Branch Pere
Marquette River was not available. However, it is assumed that the reach of the river
adjacent to the site is gaining as a result of groundwater discharge from the surficial
aquifer.
2.3 NATURE AND EXTENT OF CONTAMINATION
The primary contaminants of concern (COCs) at Wash King are PCE and
trichloroethene (TCE) given their elevated concentrations in Wash King groundwater
relative to cleanup goals, their potential to have significant negative impacts on potential
receptors, or both. The cleanup goals presented in the ROD (USEPA, 1993) are 307
Type B Cleanup Criteria (0.7 micrograms per liter {|ig/L} for PCE and 3 |ig/L for TCE).
For discussion purposes, the surficial aquifer, which extends to a depth of approximately
75 to 100 feet bgs, was subdivided into shallow and deep portions relating to the intervals
monitored by the shallow and deep monitoring and extraction wells installed at the site.
The Wash King monitoring and extraction wells are listed in Table 2.1, along with
their screen intervals, which are depicted graphically on Figure 2.3 As indicated on this
figure, the portions of the surficial aquifer (in a vertical sense) that are monitored by the
shallow and deep wells are not consistent across the site. It is assumed that previously-
collected groundwater quality data (i.e., from vertical profiling activities and monitoring
well sampling) were used to determine optimal screen intervals. Most "shallow" wells
are screened above an elevation of 760 feet amsl, corresponding to an approximate depth
of 42 to 60 feet bgs (the ground surface elevation across the site generally ranges from
802 to 820 feet amsl). The ground surface elevation at the majority of site wells ranges
between 812 and 818 feet amsl. Given groundwater elevations that generally range from
785 to 790 feet bgs (using March 2005 data), the shallow wells are generally monitoring
groundwater within the uppermost 25 to 30 feet of the saturated zone. In contrast,
screens for "shallow" wells MW-205S, MW-212S, MW-301S, MW-204S, and MW-
206D are relatively deep, and monitor a similar depth interval to that monitored by some
"deep" wells; thus, these wells are included in the deep zone for purposes of the LTMO
analysis.
2-6
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
TABLE 2.1
BASECASE GROUNDWATER MONITORING PROGRAM
LONG-TERM MONITORING OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Well Name
EW-1
EW-2
EW-4
EW-5
EW-6
MW-101D
MW-101S
MW-205D
MW-205S
MW-207D
MW-207S
MW-208
MW-209
MW-210
MW-212D
MW-212S
MW-213D
MW-213S
MW-301D
MW-301S
MW-302
MW-303
MW-304D
MW-304I
MW-304S
MW-305D
MW-305I
MW-305S
MW-3D
MW-3S
Zone
Extraction
Extraction
Extraction
Extraction
Extraction
Deep
Shallow
Deep
Deep
Deep
Shallow
Shallow
Shallow
Shallow
Deep
Deep
Deep
Shallow
Deep
Deep
Deep
Deep
Deep
Deep
Shallow
Deep
Deep
Shallow
Deep
Shallow
Current Sampling
Frequency
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Semi-annual
Earliest Sampling
Data Available
4/26/01
4/26/01
4/26/01
4/26/01
4/26/01
10/22/97
10/22/97
8/7/01
8/7/01
10/17/00
10/17/00
10/19/00
8/6/01
8/6/01
8/7/01
2/27/02
10/17/00
10/17/00
9/4/02
9/4/02
9/5/02
9/5/02
9/4/02
9/5/02
9/5/02
9/3/02
9/3/02
9/3/02
10/22/97
10/22/97
Most Recent
Data Used
6/13/05
6/13/05
6/13/05
6/13/05
6/13/05
8/30/05
8/30/05
8/29/05
8/29/05
8/30/05
8/30/05
8/29/05
8/29/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/24/05
3/31/05
8/29/05
8/30/05
8/30/05
8/30/05
8/30/05
8/29/05
8/29/05
8/29/05
8/29/05
8/29/05
Screen Interval
50-70
37-42
65-90
58.5-78.5
77-97
89-94
43-48
85-90
70-75
65-70
35-40
30-35
30-35
30-35
85-90
65-70
60-65
45-50
80-90
65-75
75-95
77-97
97-102
78-83
24-29
105-110
62-67
24-29
65-75
25-35
Wells Not (Currently Sampled
MW-103
MW-104
MW-105
MW-201
MW-202
MW-204D
MW-204S
MW-206D
MW-206S
MW-215
MW-2D
MW-4
MW-8D
MW-8S
IVTW-7S
MW-102D
MW-102S
Deep
Deep
Deep
Shallow
Shallow
Deep
Deep
Deep
Deep
Shallow
Deep
Deep
Deep
Shallow
Shallow-
Deep
Shallow
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
Not Sampled
10/22/97
10/22/97
10/19/00
10/16/00
9/4/02
10/17/00
10/17/00
10/22/03
12/3/02
8/8/01
10/22/97
10/22/97
10/22/97
10/22/97
10/19/00
10/22/97
10/22/97
10/23/03
10/23/03
10/22/03
10/22/03
10/22/03
10/22/03
10/22/03
10/22/03
10/22/03
10/23/03
10/23/03
10/23/03
8/7/01
8/7/01
10/19/00
5/13/99
5/12/99
80.3-85.3
63.5-70.5
68.5-73.5
30-35
30-35
85-90
55-60
85-90
65-70
50-55
65-75
65-75
65-75
25-35
25-35
91.1-96.1
17-22
[Deep
JWell has "S" designation, but classified as "deep" based on screen interval for LTMO
S:\ES\WPVPROIECTS\744461 - USAGE LTMO\3.xls
2-7
-------
Approximate Base of Shallow Zone
FIGURE 2.3
RELATIVE WELL
SCREEN ELEVATIONS
Extraction well
Well with "D" designation
Well with "S" designation
Well with "I" designation
Well with no designation
LONG-TERM MONITORING OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
2-8
-------
The vertical interval of the surficial aquifer monitored by the majority of "deep" (D)
and "intermediate" (I) wells ranges from 755 amsl to 718 amsl (Figure 2.3). Given
typical ground surface elevations of 812 to 818 feet amsl, the majority of "deep" and
"intermediate" wells are screened within the 57- to 100-foot bgs depth interval; this
corresponds to an interval extending from about 30 to 70 feet below the water table,
based on March 2005 groundwater elevations.
Extraction well EW-2 is screened in the shallow portion of the surficial aquifer, while
EW-4, EW-5A, and EW-6 are screened in the deep portion. The EW-1 screen has an
intermediate location, spaning both the base of the shallow interval and the top of the
deep interval (Figure 2.3). EW-5 is reportedly still pumped, but is relatively low-yielding
compared to EW-5 A.
2.3.1 Shallow Portion of the Surficial Aquifer
Figures 2.4 and 2.5 show the most recent COC results and associated chlorinated
ethene plume for the shallow and deep zones, respectively. The VOC present at the
highest concentrations in the shallow portion of the surficial aquifer at the Wash King site
is PCE. It is interesting to note that the VOC plume appears to be primarily sourced in
the immediate vicinity of the former laundry building rather than in the vicinity of the
former seepage lagoons located approximately 500 feet west-southwest of the former
laundry. The highest PCE concentrations detected in surficial aquifer groundwater
(21,000 micrograms per liter [|ig/L] in August 2005) were detected at well MW-101S,
located immediately downgradient of the former laundry (Figure 2.4). Groundwater from
nearby shallow extraction well EW-2 contained a PCE concentration of 2,500 |ig/L in
June 2005. In contrast, the shallow well nearest the former seepage lagoons (MW-215)
contained a PCE concentration of only 5 |ig/L in October 2003 (the most recent data
available for that well). It should be noted that MW-215 is screened near the base of the
shallow zone. The elevated PCE concentrations near the former laundry indicate the
presence of a significant, continuing PCE source in this area. Cohen and Mercer (1993)
state that, typically, dissolved contaminant concentrations greater than 1 percent of the
aqueous solubility of the compound are highly suggestive of the presence of non
aqueous-phase liquid (NAPL). The aqueous solubility of PCE is 150 milligrams per liter
(mg/L); therefore, dissolved PCE concentrations exceeding approximately 1.5 mg/L
(1,500 |ig/L) may indicate the presence of NAPL.
The PCE plume in the shallow portion of the surficial aquifer, shown on Figure 2.4,
extends to the north to near the Middle Branch Pere Marquette River, as evidenced by the
detection of this compound at a concentration of 3.6 |ig/L in downgradient well MW305S
in August 2005. This well is located about 200 feet from the river; therefore, it is not
known whether the shallow portion of the plume actually extends to and discharges into
the river. VOCs have historically not been detected in groundwater samples from
shallow well MW-102S, located north of the river. It should be noted that most of the
"shallow" wells installed to define the VOC plume north of the former laundry (MW-
205S, MW-202S, MW-206S, MW-204S, and MW-301S) are screened across relatively
deep intervals, as discussed in Section 2.3 and depicted on Figure 2.3. Therefore, VOC
concentrations within the uppermost 30 feet of the saturated zone north of the former
laundry building are not well defined.
2-9
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
•o
c
CD ^
0? o O O O O
CL
-------
LU
Q t
LU 5
«;
DC
c
1 — ^z
U?
f\l
FIGURE;
_i
£
o
\-
i
FRECENl
CO
O
^^
2
LU
"2.
LU
I
h-
LU
RINATED
O
x
o
_l_
h-
z
CO
2
NTRATIO
LU
O
"2.
O
o
LU
z
o
M
Q.
LU
LU
Q
i!
i= %
0*
U Q
CD ~
10NITORIN
fll IMHDV «;
^ _
s e
a: 2
uj g
\i. n.
zZ
^. <
95
— > X_
D,
\^r?r^^_*^V
^^a
(D OT
!_ CD c
(D TO O
P Q. O
£ (D Q)
O (D TO
•=: O
_ (D =C, (D
(0 C CT't
-J *-» .4-J .C 1— iJ
cr CD ro > a) u
^^^^-D^
"S .£ S ^ 2 =0
g o c w S |
.= ^ o ro c °
_i u u -a — s=
r-~ *—
O in
LT3
O € O O O O
90/EZ/9
-------
The presence of TCE and (occasionally) cis-l,2-dichloroethene (DCE) in shallow
wells indicates that some reductive dechlorination of PCE to TCE is occurring.
However, TCE and especially DCE concentrations are generally relatively low compared
to concentrations of the parent solvent (PCE), indicating that geochemical conditions in
the shallow zone are not well-suited for reductive dechlorination to occur. Groundwater
in the shallow portion of the surficial aquifer is likely somewhat aerobic and oxidizing,
although geochemical data are not available to confirm this supposition. Both PCE and
TCE are generally resistant to biodegradation under these conditions; in contrast, DCE
may be aerobically degraded.
2.3.2 Deep Portion of the Surficial Aquifer
Maximum PCE concentrations detected in the deep portion of the surficial aquifer are
substantially lower than in the shallow portion (maximum of 360 |ig/L in well MW-205S
as of August 2005, Figure 2.4), indicating that, if significant NAPL is present at the site,
it is restricted to more shallow depths. As described in Section 2.3, despite having an "S"
designation, MW-205S is screened substantially below the depth interval in which most
other "shallow" wells are screened and is therefore considered to be a deep well for
purposes of this LTMO evaluation. This well is located approximately 850 feet north of
the former Wash King laundry. PCE concentrations detected in MW-205S, MW-205D,
MW-303, and EW-6 all indicate that the "hotspot" of the deep portion of the VOC plume
is located in the vicinity of these wells, approximately 500 feet from the Middle Branch
Pere Marquette River. There are no deep wells installed between the MW-205S/205D
well pair and the river; therefore, VOC concentrations in the deep portion of the surficial
aquifer between this well pair and the river are not being characterized.
Similar to the shallow portion of the aquifer, data for MW-102D, located on the north
side of the river, suggest that the VOC plume in the deep portion of the aquifer does not
extend beneath the river in this area. It should be noted however, that MW-102D is
screened at a relatively deep interval (706 to 711 feet amsl, Figure 2.3), and a distance of
approximately 69 feet separates the bottom of the MW-102S screen and the top of the
MW-102D screen. Therefore, the data from well MW-102D do not definitively
demonstrate that the deep portion of the plume does not underflow the river.
As with the shallow portion of the surficial aquifer, the relatively low magnitude of
reductive dechlorination daughter product concentrations (TCE and DCE) at most wells
relative to PCE concentrations indicates that geochemical conditions in the deep zone are
not conducive to the widespread and sustained occurrence of reductive dechlorination.
However, the presence of daughter products indicates that some reductive dechlorination
of PCE and TCE is occurring, and that this process is more pronounced in localized areas
(i.e., MW-303, MW-206D, MW-2D).
2.3.3 Middle Branch Pere Marquette River
According to the ROD (USEPA, 1993), sampling of surface water and sediment at
three locations did not indicate levels of site-related contamination that would pose a risk
to human health or the environment.
2-12
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
SECTION 3
LONG-TERM MONITORING PROGRAM AT WASH KING
The existing groundwater monitoring program at Wash King was examined to
identify potential opportunities for streamlining monitoring activities while still
maintaining an effective monitoring program. The monitoring program at Wash King is
reviewed in the following subsections.
3.1 DESCRIPTION OF MONITORING PROGRAM
The Wash King monitoring program examined in this long-term monitoring
optimization (LTMO) consists of 49 groundwater wells, including 5 extraction wells, 25
active (i.e., currently sampled) monitoring wells, and 19 inactive monitoring wells. The
extraction wells are currently sampled quarterly, and the active monitoring wells are
sampled semiannually. The wells included in this analysis are listed in Table 2.1 and
shown on Figure 3.1 classified by their well type and sampling status (e.g., extraction
well, currently sampled well, well not currently sampled). Table 3.1 displays the number
of groundwater samples collected for VOC analysis from each well from 1997 to 2005.
As shown in Table 3.1, limited analytical results exist for the period prior to the start-up
of the extraction system in 2001. In addition, only one round of sampling was conducted
in 2004 due to a transition in site management.
The objectives of the groundwater monitoring program at Wash King are not
specified in the information reviewed for this LTMO evaluation. However, it is assumed
that the objectives are consistent with the primary spatial and temporal objectives of
groundwater monitoring programs outlined in Section 1 as summarized below:
• Evaluate groundwater at the Wash King Site for compliance with cleanup goals;
• Evaluate the effectiveness of natural attenuation processes, the groundwater
extraction system, and source reduction/removal activities at decreasing VOC
levels in groundwater; and
• Evaluate plume dynamics (i.e., is the plume increasing, stable, or decreasing in
extent both laterally and vertically).
Likely additional objectives for the groundwater monitoring program are 1) to ensure
that the remedy is protective of potential receptors, including the Middle Branch Pere
Marquette River and area residents and businesses (via vapor intrusion into occupied
3-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
-------
TABLE 3.1
PCE DATA DATE DISTRIBUTION
LONG-TERM MONITORING OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Well Name
EW-1
EW-2
EW-4
EW-5
EW-6
MW-101D
MW-101S
MW-205D
MW-205S
MW-207D
MW-207S
MW-208
MW-209
MW-210
MW-212D
MW-212S
MW-213D
MW-213S
MW-301D
MW-301S
MW-302
MW-303
MW-304D
MW-304I
MW-304S
MW-305D
MW-305I
MW-305S
MW-3D
MW-3S
Wells Not Currently Sam
MW-103
MW-104
MW-105
MW-201
MW-202
MW-204D
MW-204S
MW-206D
MW-206S
MW-211S
MW-211D
MW-215
MW-2D
MW-4
MW-8D
MW-8S
MW-7S
MW-102D
MW-102S
Number of PCE Samples per Year
1997
1
1
1
1
1998
2
3
2
2
1999
1
1
2000
1
1
1
1
1
1
1
1
2001
18
17
18
18
18
2
2
2
2
2
2
2
2
2
2
2
2
2003
8
8
8
8
8
4
3
3
3
3
3
3
3
1
2
2
3
4
3
3
3
3
3
3
3
3
3
3
3
3
2004
4
3
4
4
4
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
2005
2
2
2
2
2
3
3
2
2
2
2
2
2
2
2
2
2
2
2
1
2
2
2
2
1
2
1
2
2
2
pled
1
1
2
2
1
1
1
1
1
2
2
1
1
1
1
1
1
1
1
1
Not sampled
Not sampled
1
1
1
1
1
1
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
Note: duplicate samples not counted
Only one sampling round conducted in 2004
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\3.xls
3-3
-------
structures); and 2) to provide data for five-year reviews of remedy implementation as
required by the Comprehensive Environmental Response, Compensation, and Liability
Act (CERCLA).
3.2 SUMMARY OF ANALYTICAL DATA
The monitoring program for this plume was evaluated using results for sampling
events performed from October 1997 through August 2005. Hardcopy data were
received from the MDEQ, and were subsequently entered into an electronic database to
facilitate performance of this LTMO evaluation. The database was processed to remove
duplicate data by retaining the maximum result for each duplicate sample pair. As
discussed in Section 2.3, the COCs identified for Wash King include PCE and TCE. Cis-
1,2-DCE is also included in the analysis because it has been detected in site groundwater.
Table 3.2 presents summaries of the occurrence of detected chlorinated ethenes based on
all historical data collected from Wash King for monitoring and extraction wells. For
all data groupings, PCE is the primary COC, with the highest percentage of detections
and 307 Type B criterion exceedances, followed by TCE. PCE has exceeded the 0.7-
Hg/L standard in 54% of samples and at 23 of the 42 monitoring wells. Although DCE
has been detected, concentrations have never exceeded the 70-|ig/L USEPA MCL (a
Type B criterion for cis-l,2-DCE is not specified in the ROD); therefore, this compound
is not considered to be a significant COC in site groundwater.
Table 3.3 and Figures 2.4 and 2.5 present the most recent concentrations of PCE,
TCE, and cis-l,2-DCE for the groundwater monitoring and extraction wells screened in
the shallow and deep zones, respectively. Wells depicted on Figures 2.4 and 2.5 are
classified based on their most recent PCE values (e.g., PCE concentrations greater than
1,000 |ig/L are identified at EW-2 and MW-101S by their red color) and most recent
sampling event (currently sampled wells are circled by blue and inactive wells by pink).
Samples from 15 of the 30 currently sampled monitoring and extraction wells (50%) have
at least one COC that exceeds a 307 Type B criterion and/or MCL based on the most
recent data available for each well (all 15 have PCE exceedances and 8 have TCE
exceedances). Six of the 17 wells that have available groundwater quality data and are
not currently sampled had at least one COC that exceeded a cleanup goal during their
most recent sampling event; all six had a PCE exceedance, and four also had a TCE
exceedance.
3-4
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
O £ U
** "S hJ 5
!i"l
«« x «
0 * s
*" S .S
"I — 1
ill
'*" .=
M 's -8
u P "3
1 S 1
S u K
z £
•d
|i
VI
SM _g w
1 'E ""S 1
•2 « .5 •§
S "E. s 8
y g 5 S
££ * w
o
M 5
-2 «
S "S
^
0-
^
-3-
fi *
S 1
1 1
H |
O5
L.
-^
Q>
2
M
1
e
JExtractio
"
1^1
">
o
s=
o
o
o
o
o
o
<• 1
0
y
1
^
00
o
I 4
w
o
CM
"
m
^
CO
so
f-
V=
<-l
CN
i
O
00
o
I 4
W
O
H
^_
">
o
o
o
v=
ON
m
i
O
O
l 4
w
o
—
U
.a
o
1
CN
m
-
O
so
p
v=
^
<->
<• 1
0
m
i
O
S
(I ;
w
o
CM
' — '
^>
-
CO
so
O
V=
<-l
0
m
1
o
S
(I ;
w
o
H
ON
3
O
O
o
v=
r-
CN
1
O
S
(I ;
w
o
o
o
bD
60
I
_O
o
O
o
60
fl
g
fl
"d
N
CS
"d
O
g
hJ
w
^
w
P
p;
M
O
§
I
S^
-------
TABLE 3.3
MOST RECENT GROUNDWATER COC CONCENTRATIONS
LONG-TERM MONITORING OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Well Name
EW-1
EW-2
EW-4
EW-5
EW-6
MW-101D
MW-101S
MW-207D
MW-207S
MW-210
MW-212D
MW-212S
MW-213D
MW-213S
MW-303
MW-304D
MW-304I
MW-304S
MW-205D
MW-205S
MW-208
MW-209
MW-302
MW-305D
MW-305I
MW-305S
MW-3D
MW-3S
MW-301D
MW-301S
Most Recent
Sampling Event
6/13/05
6/13/05
6/13/05
6/13/05
6/13/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/30/05
8/29/05
8/29/05
8/29/05
8/29/05
8/29/05
8/29/05
8/29/05
8/29/05
8/29/05
8/29/05
8/24/05
3/31/05
PCE
Standard=0.7|ig/L
26
2500
4.3
71
270
11
21000
4.3
1.5
ND
32
20
24
83
34
ND
ND
ND
140
360
ND
ND
ND
ND
ND
3.6
ND
ND
ND
1.7
TCE
Standard=3ug/L
5.1
ND
ND
21
ND
5.8
360
ND
ND
ND
9.5
ND
5.2
ND
86
ND
ND
ND
14
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
cis-DCE
MCL=70ug/L
ND
ND
ND
7.4
33
ND
ND
ND
ND
ND
ND
ND
ND
ND
1.3
ND
ND
ND
3.8
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Wells Not Currently Sampled
MW-103
MW-104
MW-215
MW-2D
MW-4
MW-105
MW-201
MW-202
MW-204D
MW-204S
MW-206D
MW-206S
MW-8D
MW-8S
MW-7S
MW-102D
MW-102S
10/23/03
10/23/03
10/23/03
10/23/03
10/23/03
10/22/03
10/22/03
10/22/03
10/22/03
10/22/03
10/22/03
10/22/03
8/7/01
8/7/01
10/19/00
5/13/99
5/12/99
ND
33
5
28
81
ND
ND
ND
ND
ND
4.9
3.7
ND
ND
ND
ND
ND
1.7
4.7
ND
44
ND
ND
ND
ND
ND
ND
13
9.2
ND
ND
ND
ND
ND
2.6
ND
ND
1.9
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND = analyte not detected
b/Results in ug/L
Exceedances highlighted in yellow
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\3.xls
3-6
-------
SECTION 4
QUALITATIVE LTMO EVALUATION
An effective groundwater monitoring program will provide information regarding
contaminant plume migration and changes in chemical concentrations through time at
appropriate locations, enabling decision-makers to verify that contaminants are not
endangering potential receptors, and that remediation is occurring at rates sufficient to
achieve remedial action objectives (RAOs) within a reasonable time frame. The design
of the monitoring program should therefore include consideration of existing receptor
exposure pathways as well as exposure pathways arising from potential future use of the
groundwater.
Performance monitoring wells located within and downgradient from a contaminated
area provide a means of evaluating the effectiveness of a groundwater remedy relative to
performance criteria. LTM of these wells also provides information about migration of
the contamination and temporal trends in chemical concentrations. Groundwater
monitoring wells located downgradient from the leading edge of a contaminated area
(i.e., sentry wells) are used to evaluate possible changes in the extent of the plume and, if
warranted, to trigger a contingency response action if contaminants are detected.
Primary factors to consider when developing a groundwater monitoring program
include at a minimum:
• Aquifer heterogeneity,
• Types of contaminants,
• Distance to potential receptor exposure points,
• Groundwater seepage velocity and flow direction(s),
• Potential surface-water impacts, and
• The effects of the remediation system.
These factors will influence the locations and spacing of monitoring points and the
sampling frequency. Typically, the greater the seepage velocity and the shorter the
distance to receptor exposure points, the more frequently groundwater sampling should
be conducted.
4-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
One of the most important purposes of LTM is to confirm that the contaminant plume
is behaving as predicted. Graphical and statistical tests can be used to evaluate plume
stability. If a groundwater remediation system or strategy is effective, then over the long
term, groundwater-monitoring data should demonstrate a clear and meaningful
decreasing trend in concentrations at appropriate monitoring points. The groundwater
monitoring program at Wash King was evaluated to identify potential opportunities for
streamlining monitoring activities while still maintaining an effective performance and
compliance monitoring program.
4.1 METHOD FOR QUALITATIVE EVALUATION OF MONITORING
NETWORK
The qualitative LTMO evaluation included 47 groundwater monitoring and extraction
wells with historical data located in the Wash King area. These sampling points, their
associated depth zones and basecase monitoring frequencies, and the earliest and most
recent sampling data used in the LTMO analysis are listed in Table 2.1; their locations
are depicted on Figure 3.1.
Multiple factors were considered in developing recommendations for continuation or
cessation of groundwater monitoring at each well. In some cases, a recommendation was
made to continue monitoring a particular well, but at a reduced frequency. A
recommendation to discontinue monitoring at a particular well based on the information
reviewed does not necessarily constitute a recommendation to physically abandon the
well. A change in site conditions might warrant resumption of monitoring at some time
in the future at wells that are not currently recommended for continued sampling.
Typical factors considered in developing recommendations to retain a well in, or remove
a well from, an LTM program are summarized in Table 4.1. Typical factors considered
in developing recommendations for monitoring frequency are summarized in Table 4.2.
TABLE 4.1
MONITORING NETWORK OPTIMIZATION DECISION LOGIC
LONG-TERM MONITORING OPTIMIZATION
WASH KING SUPERFUND SITE
Reasons for Retaining a Well in
Monitoring Network
Well is needed to further characterize the
site or monitor changes in contaminant
concentrations through time
Well is important for defining the lateral or
vertical extent of contaminants.
Well is needed to monitor water quality at a
compliance or receptor exposure point
(e.g. , water supply well)
Well is important for defining background water
quality
Reasons for Removing a Well from
Monitoring Network
Well provides spatially redundant
information with a neighboring well (e.g.,
same constituents, and/or short distance
between wells)
Well has been dry for more than two yearsa/
Contaminant concentrations are
consistently below laboratory detection
limits or cleanup goals
Well is completed in same water-bearing zone as
nearby well(s)
a/ Periodic water-level monitoring should be performed in dry wells to confirm that the upper boundary of the saturated
zone remains below the well screen. If the well becomes re-wetted, then its inclusion in the monitoring program
should be evaluated.
4-2
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
4.2 RESULTS OF QUALITATIVE LTMO EVALUATION FOR
GROUNDWATER
The results of the qualitative evaluation of monitoring wells at the Wash King site are
described in this subsection. The evaluation included the 47 groundwater monitoring
wells listed in Table 2.1. The qualitative LTMO evaluation for groundwater considered
historical analytical results for the three primary COCs (PCE, TCE, cis-l,2-DCE) and
whether continued monitoring of each well was desirable in light of the Wash King
groundwater monitoring goals listed in Section 3.1. In addition, potential data gaps were
considered.
TABLE 4.2
MONITORING FREQUENCY DECISION LOGIC
LONG-TERM MONITORING OPTIMIZATION
WASH KING SUPERFUND SITE
Reasons for Increasing
Sampling Frequency
Groundwater velocity is high
Change in contaminant concentration would
significantly alter a decision or course of action
Well is necessary to monitor source area or
operating remedial system
Cannot predict if concentrations will change
significantly over time, or recent significant
increasing trend in contaminant concentrations at a
monitoring location resulting in concentrations
approaching or exceeding a cleanup goal, possibly
indicating plume expansion
Reasons for Decreasing
Sampling Frequency
Groundwater velocity is low
Change in contaminant concentration would not
significantly alter a decision or course of action
Well is distal from source area and remedial system
Concentrations are not expected to change
significantly over time, or contaminant levels have
been below groundwater cleanup objectives for
some prescribed period of time
Table 4.3 includes recommendations for retaining or removing each well, the
recommended sampling frequency, and the rationale for the recommendations. The
qualitative analysis results are depicted on Figures 4.1 and 4.2 for the shallow and deep
zone wells, respectively, and are summarized by well type and aquifer zone in the
following subsections.
4.2.1 Extraction Wells
Five GWE wells were considered during the qualitative evaluation, including EW-1,
2, 4, 5, and 6. Sampling results for EW-3 were not available and this well was not
included in the evaluation. EW-3 is reportedly not located in an optimum location and is
not always pumped. A sixth GWE well, EW-5A, exists and reportedly is the primary
operating well of the EW-5/EW-5A pair. Both wells are operational, but EW-5
reportedly is relatively low-yielding. However, sampling results received from the
MDEQ indicate that they are for EW-5; sample IDs containing "EW-5A" were not
4-3
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
O U P
£s^
Qualitative Analysis
Rationale
o
II
C3
^ q^
-I d
c S
'2 S
.-§ |
1
"S
o
•d
™
H
.u M ??
"c C S
Sag
& & s,
<3 1 8
CO ta
'S
_u
_o
2
^i
M
1
2
_
"3
^
t ug/L variation over last 6
>-,
"d
o
Data can be used to monitor mass removal rates over time, optimize pumping strategy, and evaluate remedial progess;
sampling events— asymptotic behavior justifies lower sampling frequency
"3
g
.3
£H
S
X
r—t*
W
"£
3
d
_0
o
X
rVi
i
^
w
^ely stable trend over last 4
c3
"s
Data can be used to monitor mass removal rates over time, optimize pumping strategy, and evaluate remedial progess;
sampling events justifies lower sampling frequency
"3
|
S
'd
VI
X
r—i'
o
-e
3
d
^0
"o
X
W
i
^
w
13
d in ROD; minimal mass remov
-8
• ^H
rl
Only 1 . 1 ug/L variation over last 6 sampling events, and COCs < MCLs but PCE exceeds 307 Type B cleanup criterio]
occurring; evaluate need for continued pumping of this well; recent stable trend justifies lower sampling frequency
13
|
.3
£H
S
VI
X
r—i'
O
^
3
d
^0
"o
cS
"K
W
i
^
w
o
lating in narrow range (140 to 21
P
o
p
Data can be used to monitor mass removal rates over time, optimize pumping strategy, and evaluate remedial progess;
ug/L) over 22 of last 23 sampling events justifying lower sampling frequency. Recent jump in TCE < 10 ug/L.
"3
P
S
'd
VI
X
r—i'
o
-e
3
d
^0
"o
X
w
1
^
w
consistent decreasing trend ovei
|
*"w
Data can be used to monitor mass removal rates over time, optimize pumping strategy, and evaluate remedial progess;
last 8 events justifies lower sampling frequency
13
3
.3
£H
S
VI
X
r—i'
O
t;
3
d
^0
o
"S
w
1
^
w
bO
d
1
"S
1
o
PH
O
d
Decreasing PCE and TCE concentrations in source area well; active source remediation not occurring; sampled 8 times
not necessary unless more active source area remedy is implemented
3
3
H
^<
X
rt
p
.3
VI
p.
&
Q
o
'
^
S
i PCE concentrations over 7
nge in site conditions occurs
Tl -§
d ^
w ~y
Retain due to source area location and greatly elevated PCE levels. Annual frequency justified by lack of significant ti
events since Dec 02 and lack of active source area remediation; more frequent sampling not necessary unless significal
(more active source area remedy is implemented, etc.)
1
J~j
^
X
"3
g
§
'|
Vl
^
^^
Vl
Vl
_— <
o
I
E>
s
Annual frequency justified by
itire sampling history; more
ledy implemented, etc.). Large
M u <3
i-H H ^
i > tl
Retain due to relatively elevated PCE/TCE levels in plume core area near downgradient extent of monitoring well netv
apparent decreasing trend in PCE concentrations over 7 events since Dec 02 and lack of trend in TCE concentrations o
frequent sampling not necessary unless significant change in site conditions occurs (pumping strategy changes, differe]
distance to river (-600 ft) relative to groundwater velocity (185 ft/yr) indicates that annual frequency is protective
1
J~j
^
X
"3
g
§
'|
Vl
p.
&
Q
o
'
c>
S
<•}
tantial decreasing PCE trend sim
ant change in site conditions
ed groundwater velocity (185
1*5 1
p 'Q ^H
Vl £Q +-•
Retain due to relatively elevated PCE levels in plume core area near downgradient extent of monitoring well network.
Sept 02, most likely due to GWE system, justifying annual frequency. More frequent sampling not necessary unless si
occurs (pumping strategy changes, different remedy implemented, etc.). Large distance to river (-600 ft) relative to e:
ft/yr) indicates that annual frequency is protective
1
J~j
^
X
"3
g
§
'|
Vl
p.
&
Vl
I/-)
o
I
E>
s
arent increasing trend, and well
/ concentrations and lack of
PH >
PH O
CS -^
Q 2
PCE > 307 Type B criterion but < MCL since Sept 02; maximum PCE level over 10 events since Aug 01 =5.3 ug/L; n
not in primary plume flowpath from source area; retain due to cleanup goal exceedance but at very low frequency due
temporal trend
ra
H
SH
PQ
X
"3
g
§
'|
Vl
p.
&
Q
o
'
c>
S
60
PH
CS
"o
O
1
d
b"
QJ>
^H
cS
o
o
PCE > 307 Type B criterion but < MCL since Aug 01; very stable trend; well not in primary plume flowpath from soui
exceedance but at very low frequency due to low concentrations and lack of temporal trend
ra
H
SH
'PQ
X
"3
g
§
1
Vl
^
^^
Vl
Vl
^~~.
o
I
E>
s
o
cS
d
-S
1
-d
d
o
&
f
^H
g
d
All non-detect for chlorinated ethenes over 1 1 events since Aug 0 1 ; substantially cross-gradient of VOC plume and no
concern
o
"d
"o
X
w
X
"3
g
§
'|
Vl
^
^^
Vl
X
o
I
E>
s
^
o
^
o
^H
^
d
^H
O
i
•d
d
p
o
bt
f
^H
_o
o
g
All non-detect for chlorinated ethenes over 10 events since Aug 01; substantially cross-gradient of VOC plume and no
concern
o
•d
"o
X
w
X
"3
p
g
cS
'3
S
Vl
^
^^
Vl
ON
O
'
^
S
oring groundwater in area of
'd
Q
B
All non-detect for chlorinated ethenes over 9 events since Aug 0 1 ; substantially cross-gradient of VOC plume and not :
concern
o
•d
P
"o
X
w
X
"3
p
cS
'a
s
Vl
$
^^
Vl
o
I
^
s
led monitoring facilitates
.S
d
o
PCE and TCE > MCL and 307 Type B criterion, but exhibit decreasing trends since Dec 02 (PCE) or Nov 01 (TCE); c
evaluation of remedial progress, but decreasing trends justify lower frequency
1
S
J~j
^
X
"3
p
g
cS
'3
S
Vl
p.
&
Q
(N
'
E>
S
on of remedial progress but
'"§
p
13
PCE > MCL and 307 Type B criterion, but exhibits decreasing trend since Sept 02; continued monitoring facilitates ev
decreasing trend justifies lower frequency
1
J~j
^
X
"3
p
cS
'a
s
Vl
p.
&
Vl
0s)
(N
'
^
S
it time in Aug 05 after 1 1 non-
decreasing PCE trend justifies
•- -d"
d
O L^
PCE > MCL and 307 Type B criterion, but exhibits decreasing trend since Feb 02; TCE > MCL and Type B criterion f
detect results; continued monitoring facilitates evaluation of remedial progress for PCE and potential increasing TCE t:
lower frequency
3
3
H
^<
X
"3
p
.3
Vl
P.
&
Q
r^J
'
E>
S
-------
O U P
§5^
^e Analysis
Qualitatr
i
v
Rationale
o
H
C3
^ "2
^ «
sr. g
fi M
'2 S
.-§ I
1
15
o
•d
™
H
W
i_a W) fr
S g a
2 i, 1
J" g« S
5 C ^^
1 C5 ^
'S
^
_u
o
2
"^,
M
1
^H
"3
dial progress
p
rl
jcreasing trend since Jun 03; continued monitoring facilitates evaluation of ren
rpe B criterion, but exhibits di
H
r-
o
i
j
A
W
13
p
H
H
"^
X
p
§
• ^
£3
C/2
g:
J2
13
.c!
C/2
C/2
2
o
O CH
P
60 "d
K- i
307 Type B criterion) and decreasing;
:ent of plume in this area, and EW-4 contains very similar PCE concentrations;
jectives; EW-4 data indicate remedial progress in this area
me adjacent to EW-4; PCE cc
:h) monitors cross-gradient ex
nation to meet monitoring ob
At W. edge of VOC plui
(screened at similar depi
not provide useful infon
•s
_3
"p
X
w
X
g
a
• ^
£3
OJ
P.
&
VI
o
m
i
erf
205D; continued
ogress in this are;
1 i-H
K~! 0
ept 02), otherwise all non-detect despite presence of VOCs in nearby well MW
ion about plume; monitoring of MW-205D will allow assessment of remedial ]
W" M
£5
0 12
O £
^ p
° o
i-H IT— (
Only 1 historic detectior
monitoring of this well c
i
"p
X
w
X
13
p
§
• ^
£3
VI
p.
1
o
CO
i
O
•& -d
2 « g
-£j flj j.^
§£83
ml
§ ri 'i
•d y g a
s s § e
-fa A .S _g
*-r! LiJ • — < MCL and 307 Type B
1 of EW-6, so continued mon:
Aug 05 (86 ug/L) after 7 trac
.evels justifies continued semi
are sampling of EW-6 indicati
Co-located with EW-6; '.
magnitude similar to tha
criterion for first time in
abrupt increase in TCE 1
conditions, and/or if fun
13
P
erf
'H
S
VI
X
13
§
• ^
£3
VI
p.
&
o
CO
i
o
undwater in area
P
ince Sept 02; substantially cross-gradient of VOC plume and not monitoring gi
litors easterly extent of plume in this area
nated ethenes over 8 events s:
ir MW-3S/3D adequately moi
All non-detect for chlori
concern; nearby well pa:
•3
P
"3
X
w
X
13
p
§
• ^
£3
C/2
p.
a
Q
o
c^i
i
o
undwater in area
P
ince Sept 02; substantially cross-gradient of VOC plume and not monitoring gi
litors easterly extent of plume in this area
nated ethenes over 8 events s:
ir MW-3S/3D adequately moi
All non-detect for chlori
concern; nearby well pa:
•3
P
"3
X
w
X
13
3
a
cS
•g
a
VI
p.
a
_
o
f^l
i
o
undwater in area
P
ince Sept 02; substantially cross-gradient of VOC plume and not monitoring gi
litors easterly extent of plume in this area
nated ethenes over 7 events s:
ir MW-3S/3D adequately moi
All non-detect for chlori
concern; nearby well pa:
•3
P
"3
X
w
X
13
p
§
• ^
£3
VI
g
,3
VI
VI
o
f)
i
ca
cs
t/3
,o
1 If
a -§
•d P
8 cr
CS CS
•d >
b
O
f/3
13
&
PH
P
a
ca
_p
"p
V
ig/L; resume sampling to evaluate remedial progress, then dicontinue if VOCs
)3, when PCE detected at 33 V,
Not sampled since Oct C
consecutive events
"s
a
^
"^
X
•d
p
"En
a
cS
VI
*
p.
1
o
'— '
i
. r,
6
p
p
ter in area of con
ro
^
)Cs all non-detect during 3 historic sampling events; does not monitor groundv
' plume over time
ient (east) of VOC plume; CC
1 to monitor easterly extent of
Substantially cross-grad
MW-3S/3D better suitec
•3
j3
"p
X
W
X
-d
"En
a
ca
VI
*
p.
1
o
'— '
i
cS
ts
•d
\
1"
p
p
rs westerly extent of plume over time; cross-gradient location justifies lower fr
_o
1
P ^
p S
•7 a
||
13 JQ
^^ o
S v
P O
Only sampled once in O
continue to indicate VO'
13
S
CH
P
pq
X
•d
"En
a
(S
^_J
*
g
,3
VI
o
-------
O U P
§5^
I
•Sj
OJ
Qualitativi
C
Rationale
u
11
||
£ s
sr. e
c S
'2 S
5 8
"3 ^
I
1
"S
o
•d
™
H
UJ
«- gf u4
s s 1
S S" 5.
s a g
_) « £1
05 £
'S
u
'M
o
2
•d
HrH*
K
1
Z
^
"3
^
M
"S "3
SH ^
PH 3
CS g
•d S
g -^
^ V
|8
"w >
'fl "S
° 2
11
o • ^
a 3
rH" o
o 3
'-3 .a
0 fl
o o
' — i O
"3 3
^
§>§
^ <~
2 '^
o b
||
^H rH
!* C^
o ^
CD -^
"S o
0} P.
1 S
M ni
Lonitors we
.e in this an
B §
§ a
'x O
£2 t
fl p.
> 1-H
" P.
g 0
i g
fl ^H
'3
3
SH
'PQ
X
•d
o
t
cS
GO
"S
p
O
Q
GO
O
^J
g
s
M
a
o
^3
o
S3
'-H
o
o
1
E/T
CD
1-H
bO
o
PH
•-•H
• ,-H
id
g
a
o
"3
S
_o
bO
S3
1
nd TCE > MCL); resume
criterion a
PQ
PH
H
O
A +2
WH
O >
g u
•B >
P • i-H
n Oct 03 (TCE
5 over 2 conseci
o o
|j
VI V
"3
g
J~j
^
X
•d
1
cS
VI
"o
p
o
Q
Q
o
^j
g
S
^
o
p
o
o
•d
fl
o
&
o
PH
,— 1
c3
^d
r^
g
(^
CD
jS
_ ^
c3
£
1
"En
g
i
CD
c
iterion and TCE > MCL);
o
PQ
PH
id PCE > 307
ive events
<-H -^J
cS p
, o
H 1
O g
C- o
o
0 °
o $•
cS o
§|
(N g.
.a P.
led twice
5 < cleanu
|o
1
3
J~j
"^C
X
•d
o
t
cS
GO
"S
p
O
Q
GO
O
^J
g
s
cS
cS
1
8
IH-H
o
1
O
1
ni
d to define w<
X
fl
.2
Q
Q
o
'1
1
"d
rH
1
&
"d
"o
X
W
X
-d
1
cS
"S
p
O
Q
Q
ON)
^J
g
S
1
p
&
f^
60
"En
a
"3
E 1
d bO
I 2
P i
O Co1
O 1>
£ a
'W EH
S W
^ W
W CD
|1
.2 3
c3 ^
1 1
£> cS
43 O
cd m
10 historic groundwater qu
aan cleanup goals, otherwi
" ^j
is M
ra Qj)
i-H t/3
18
DC plume but
(biennial if V
> |
l~; f**
^ 5
CW £H
fe ^fl
;ross-gradient (A
s of initial samp
ars to be c
on result;
o "d
\Jt ' 03
<,, n
•^
a
SH
'PQ
X
•d
o
t
cS
GO
"S
^
_o
^
GO
GO
| |
ON)
^J
g
S
bfl
.fl
"En
fll
jontinui
M
•d
fl
43
"M
rH
O
QJ
.>
consecui
o
g
nover
o
levels belowType B criteri
iterion); if'
d progress
LJ '•o
0 ^
PQ |
PH S
H1 3
t~- 3
T 1
A OJ
"3 -2
"> '"
o J
r-H <-£-)
W "3
O fl
§ §
0 0
tl
S 'S
r3 0
"3
g
^
"^C
X
•d
1
cS
GO
"S
g:
rtf
GO
2
CN)
^J
g
S
g
'g
S3
-4j
o
bo
S3
1
c3
3
CH
Q
• 22
' O
g
^
^"^
1
>
1?
I a
3 J1
> bO
'+3 cS
fl P-
o w
cleanup goals over two co
ient of historic source area
vels below
downgradi
<~-~< o
bO rt
p cS
s ^
^T QJ
^ si
II 0
w "S
O £
W o
cO ^3
O ,— i
o ^
art
H
13 ^
p O
1 1
cS o
h- 1 O
1
J~j
"^c
X
•d
o
t
cS
"S
p
O
Q
Q
^j
g
S
g
1
o
o
M
1
"S
bo
rH
PH
a
cw
p
PI
£
o
0
cw
^d
1
cw
I
"3
p
cS
goals over two consecutiv
;r laundry
ll
"o o
if levels belo
s downgradie
ir 2
^d b£
P° 2
P
00 ;3
i Oct 03 (PCE =
evaluate remed
• fl o
•d ^
u .2
tl
c3 ^H
-M S
03 P
K^ c3
1
3
^
"^C
X
•d
o
f
GO
"S
p
O
Q
T^-
^j
g
S
8
!3
o
o
o
c3
S
C3
_fl
1-H
Qj
1
•6
1
3
'3
not mo
o
I
8
O
1-H
non-detect
1
i
"P.
0
o
f 1 t
o
4j
cS
fl
1
O
•d
3
CD
S
^ ^
PH
O
^jt
J
o
-^
CO1
1
ontinue monitoring at low
o
"S
1
o
S3
8
o
"3
ON
C\
E?
S
o
;s from Oct 97 t
of remedy
led 4 time
jtiveness i
1 o
GO &
• S
3
S-H
'PQ
X
•d
o
t
cS
VI
"o
£
J2
c3
GO
GO
O
^J
g
s
— £
"3
^
^
o
z
•d
o
o
s
s
o
o
«
a
S3
o
o
o
p.
cS
•d
a
o
o
-fcl
•d
S3
p.
tfl
•S
"S
PQ
"3
g
J~j
^
X
>
o
GO
i— i
GO
a
o
o
o
p.
cS
•d
a
o
o
Js
•d
S3
p.
fl
tfl
•S
"S
PQ
"3
g
^
^
X
>
o
GO
o
rtf
r<1
GO
i
GO
•d
rH
ional on results from S-2 t
o
o
1
>
o
o
p.
cS
•d
a
o
1
•d
S3
cS
p.
fl
tfl
•s
"S
PQ
"3
p
<^
^
^
'H-J
_§
"o
CM
X
>
o
c3
GO
^t
GO
i
GO
•d
rH
ional on results from S-2 ;
o
o
1
-ogress over t
p.
cS
•d
a
S3
o
•d
S3
cS
p
p.
fl
tfl
•s
"S
PQ
"3
p
<^
^
^
'H-J
_§
"S
CM
X
>
o
Ctf
GO
1^1
GO
a
fe
o
o
o
p.
cS
•d
a
o
o
Js
•d
S3
p.
fl
tfl
•3
"S
PQ
3
g
^
"^
X
%
o
GO
^D
GO
a
-ogress over t
p.
cS
•d
a
o
o
•d
S3
cS
1
p.
fl
1
"S
PQ
3
3
^
"^C
X
>
o
r^
GO
t-J-
GO
a
>
o
o
p.
^
•d
a
o
1
•d
S3
cS
p.
fl
tfl
•3
"S
PQ
s
fl
5
"^
X
p
CD
Q
i
Q
a
-ogress over t
p.
cS
•d
a
S3
o
•d
S3
cS
p
p.
g
tfl
•s
"S
PQ
i
fl
^
"^C
X
p
CD
Q
i
Q
ional on results from D-2
o
o
1
S3
o
o
o
p.
cS
•d
a
o
o
Js
•d
S3
p.
g
tfl
•S
"S
PQ
p
<^
^
cd
'H-J
_§
"S
CM
X
p
CD
Q
cn
I
Q
a
>
o
o
p.
^
•d
a
o
1
•d
S3
cS
p.
g
tfl
•s
"S
PQ
s
fl
^
"^C
X
p
CD
Q
^j-
Q
a
-ogress over t
p.
cS
•d
a
S3
o
•d
S3
cS
p
p.
g
tfl
•s
"S
PQ
i
fl
^
"^C
X
o
Q
"JO
i
Q
a
o
o
&
o
p.
cS
•d
a
o
o
Js
•d
S3
03
P.
g
tfl
•S
"S
PQ
3
g
^
"^C
X
p
CD
Q
^?
Q
-------
L- O) C
CD CO o
E Q- o
E CD a
O CD CO
90/9/9 ueo pxw-S||9M6u!XHSEA/VSIO/6u!XHseM/3OVSn/OI/\lll/P3W3J/S3/:s
-------
Ob
O
II £
sgg^l
V 11
LJJ
3i§»g°
^ > S LU ^ O
W LLJ !l 77 HJ z
*sgl§s
SP.Sagfe
Cffc O
2uj
^m
a:
LJJ
H
6
z
O
a:
LJJ
0.
D
W
a:
Q
z
D
<
O
CD
o w
"S -
'-> fc 5D
— CD ^ a3 "CD
03 C O)>^ g
o <23.~ -D
*- £ ._ O) C
§®|| I o
O" CD 03 > O) T3
^llEll
„. ^- o -C 1— ^
CD O X (/i CD ^
r~ -— ^ ' "*-l s
.E .c o cc c °
I o O "O — ^
t
0
ill
g 0 «
i_ .__
0)
O)
^ ^ (/) < DQ
— Q_
ro a)
a 2
X
LU
I
^
T3
0
T3
0
E
I
0
a:
CD
0
"5
Q_
• O O 0 0
-------
received. It is assumed that EW-5 and EW-5A are spatially co-located and screened at
approximately the same depths.
Graphs of PCE concentrations versus time for the extraction wells are shown on
Figure 4.3. These wells are currently sampled quarterly. As shown on Figure 4.3, PCE
concentrations in wells EW-1, 2, 4, and 5 exhibit relatively stable trends. The PCE
concentrations in EW-1 and EW-4 have not varied by more than 4 |ig/L and 1.6 |ig/L,
respectively, over the most recent six sampling events. The PCE concentration at EW-5
fluctuated between 130 and 210 |ig/L over 22 sampling events from February 2002 to
March 2005. The most recent four data points for EW-2 suggest that the PCE
concentration may be stabilizing, and data for EW-6 exhibit a steady decreasing trend.
The recent stable trends in four of the five GWE wells (described above) indicate that
the sampling frequency at these wells can be reduced with little loss of important
information. Continued frequent monitoring at locations where no temporal trend in
contaminant concentrations is present serves merely to confirm the results of previous
monitoring activities at that location; results of continued monitoring through time are
likely to fall within the historic range of concentrations that have already been detected.
Therefore, reduction in the monitoring frequency for EW-1, 2, 4, and 5 to semiannual is
recommended. Historic data for EW-6 indicate that either the deceasing trend will
continue in the future or PCE concentrations will begin to stabilize. Given the lack of
significant fluctuations in contaminant concentrations in this well over the past two years,
and the likelihood that future contaminant concentrations will fall within a relatively
narrow and predictable range, semiannual sampling of this well is also deemed to be
appropriate. Semiannual sampling of these wells should allow for adequate definition of
mass removal rates and remedial progress over time. However, if future concentrations
deviate significantly from the expected trends, then resumption of quarterly sampling
should be considered.
Mass removal rates at EW-4 are currently extremely low given the low magnitude of
VOC concentrations in the extracted water (PCE at 3 to 5 |ig/L over the most recent six
sampling events). However, the PCE concentrations do exceed the 307 Type B Cleanup
criterion of 0.7 |ig/L; therefore, continued sampling of this well at a lower frequency is
recommended to track future progress on achieving cleanup goals in this area.
4.2.2 Monitoring Wells Screened in the Shallow Portion of the Surficial Aquifer
A total of 16 monitoring wells screened in the shallow portion of the surficial aquifer
were evaluated qualitatively. The shallow portion of the surficial aquifer is defined for
purposes of this LTMO evaluation to extend vertically downward to an elevation of
approximately 760 feet amsl, roughly corresponding to the uppermost 25 to 30 feet of the
saturated zone. Nine of the 16 wells are currently sampled on a semiannual basis, six of
the remaining seven wells are not sampled on a regular basis, and there are no available
sampling results for one well (MW-211S). As described in Section 2.3, five wells
containing an "S" designation (MW-204S, 205S, 206S, 212S, and 301S) appear to
actually be screened at elevations more similar to some of the "deep" wells (Figure 2.3);
therefore, they were considered to be deep wells for the purposes of the LTMO
evaluation.
4-9
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
FIGURE 4.3
GWE PCE CONCENTRATIONS OVER TIME
LONG-TERM MONITORING OPTIMIZATION
WASH KING SUPERFUND SITE
90 - -
80 - -
70 - -
*
%ft f
u 'i
*
EW-1. PCE (ug/L)
t
a
l\A
* ^\
«*1
ttA/»
Dcc-00 Juri-01 Dec-01 Jun-02 Dcc-02
400
350
300
250
200
150
100
50
0
DC
2000
1500
1000
500
0
Ck
A
t / x-
S-. X
Jun-03 Dcc-03 Jun-04 Dcc-04 Jun-05 DC..
-05
EW4, PCE (ug/L)
t
ill
M
*9w
-00 Jun-01
f
v*v^*«il i
4 _^
Jan-02 Jul-02 Fcb-03 Scp-03 Mar-04 Oct-04 Apr-05 Ho
EW6,
A*
r% \
L*i
t-OO Apr-01 Nov-01 May-02 DC
PCE (1101)
j\
•^\ _-^_
^-^--^
-05
c-02 Jun-03 Jan-04 Aug-04 Feb-05 Sep-05
EW2, PCE (iifl,'L)
1500 -
1000 -
500 -
t.
* 1 ^^ *N»
JNJVN /
U*4 * W
Dcc-00 Jul-01 Fcb-02 Aug-02
1800
1600
1400
1200
1000
800
600
400
200
0
De
\ \
IVfaf-03 Stp-03 Apr-04 Oct-04 May-05 Dec
-05
EW5, PCE (ug.l)
|
I
1
|| Jl
^J^ ^ +IP ^^^^f^fii — J|__^— - + — »— _(^-»^_^
c-00 Jun-01 Dec-01 Jyfi-02 Ch
c-02 Jun-03 Dec-03 Jun-04 Dec-04 Jun-05 Dee
-05
Four of the nine shallow wells that are currently sampled are recommended for
deletion from the LTM program because they are cross-gradient and distant from the
VOC plume; these wells include MW-208, 209, 210, and 304S (Table 4.3 and Figure
4.1). Shallow wells MW-7S and 8S are also recommended for deletion for the same
reason. Chlorinated ethenes have never been detected in these wells, and continued
sampling of these wells does not provide any information about the remaining areas of
concern.
If collection of background groundwater geochemical data is desired in the future to
support a natural attenuation evaluation, then all or a subset of the cross-gradient wells
could be sampled for that purpose. Similarly, if data regarding upgradient groundwater
quality were desired in the future (e.g., if there is reason to believe that an upgradient
source of chlorinated ethenes is present that is contributing to the Wash King plume),
then MW-7S could be sampled for that purpose.
4-10
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
Future sampling of 10 existing shallow wells on an annual to biennial (every other
year) basis is recommended, as detailed in Table 4.3 and on Figure 4.1. Five of these
wells (MW-101S, 207S, 213S, 305S, and 3S) are currently sampled semiannually as part
of ongoing LTM activities. The remaining five wells (MW-201, 202, 21 IS, 215, and
102S) are not currently sampled and are recommended for future inclusion in the
monitoring program.
MW-3S, MW-201, MW-211S, and MW-202 are recommended for future sampling to
assist in the definition of the lateral plume boundaries over time in the shallow zone. The
recommended sampling frequency for MW-3S, MW-201, and MW-211S is biennial
given their inferred cross-gradient location relative to the VOC plume and the apparently
stable to diminishing extent of the plume footprint over the last several years; this
frequency should be conditional on new data continuing to indicate VOC concentrations
less than cleanup goals for wells that have not been sampled recently. However, if
hydraulic conditions change (e.g., change in on-site pumping conditions or initiation of
off-site groundwater extraction in the vicinity), or the sampling results indicate plume
expansion, then more frequent sampling should be considered. Annual sampling of MW-
202 is recommended given its location at the downgradient end of the plume near the
river.
Five of the remaining six shallow wells (MW-101S, 207S, 213S, 305S, and 215) are
recommended for continued sampling because their plume-interior locations facilitate
assessment of remedial progress over time. MW-305S is the furthest downgradient well
containing a detectable concentration of a COC, and continued monitoring of this well
provides some indication of plume stability. MW-215 was last sampled in October 2003
and contained a PCE concentration that exceeded the 307 Type B cleanup level. If
current VOC concentrations in this well are below cleanup levels over at least two
consecutive monitoring events then removal of this well from the LTM program could be
considered. An annual to biennial sampling frequency for these five plume-interior wells
is recommended (annual for all wells except MW-207S). A reduction in sampling
frequency relative to the current semiannual program is supported by the lack of
increasing trends in these wells, the generally stable to diminishing nature of the plume,
and the perceived low degree of risk posed by the plume to potential receptors.
Shallow well MW-102S functions as a downgradient sentry well and provides useful
information regarding the northern extent of the plume. Chlorinated ethenes have never
been detected in this well, and there is no reason to believe that this will change given
that groundwater in the shallow zone likely discharges to the Pere Marquette River.
Therefore, a relatively low (biennial) sampling frequency is recommended.
4.2.3 Monitoring Wells Screened in the Deep Portion of the Surficial Aquifer
A total of 28 monitoring wells screened in the deep portion of the surficial aquifer
were evaluated qualitatively. The deep portion of the surficial aquifer is defined for
purposes of this LTMO evaluation to extend from roughly 25 to 30 feet below the water
table to the clay aquitard that is believed to separate the surficial aquifer from the
underlying sandy drinking water aquifer (i.e., from roughly 30 to at least 60 or 65 feet
below the water table). Sixteen of the 28 wells are currently sampled on a semiannual
basis, 11 of the remaining 12 wells are not sampled on a regular basis, and there are no
4-11
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
available sampling results for one well (MW-211D) . The five wells containing an "S"
(shallow) designation (MW-204S, 205S, 206S, 212S, and 301 S) that are actually
screened at deeper elevations than the other "shallow" wells, and the two wells
containing an "I" (intermediate) designation (MW-304I and 3051) are lumped into the
"deep" zone for purposes of the LTMO evaluation (see Figure 2.3 for depiction of screen
intervals).
Two of the 16 deep wells that are currently sampled are recommended for deletion
from the LTM program because they are cross-gradient and distant from the VOC plume;
these wells include MW-304I and 304D (Table 4.3 and Figure 4.2). Chlorinated ethenes
have never been detected in these wells, and continued sampling of these wells does not
provide any useful information about the remaining areas of concern.
Deep zone wells MW-301S and 301D are located adjacent to GWE well EW-4, near
the inferred western edge of the VOC plume. MW301D has been non-detect for COCs
throughout its monitoring history (since September 2002), and there is no reason to
believe that this will change unless hydraulic conditions in the surficial aquifer change.
Monitoring of nearby well MW204D, screened in a similar depth interval, would allow
assessment of the western plume boundary over time in this area. The screen interval of
MW-301S corresponds to the uppermost 10 feet of the screen interval of EW-4, and PCE
concentrations in both wells are similar. Continued monitoring of EW-4 should indicate
future remedial progress in this area, eliminating the need to continue sampling MW-
301S.
Well MW302 is spatially redundant with well pair MW-205S/205D; the latter wells
contain elevated concentrations of COCs, while MW-302 (which has a relatively long
screen interval) has been non-detect for COCs over seven monitoring events since
December 2002 and is not providing useful information. Therefore, continued
monitoring of MW-205S/205D and removal of MW302 from the monitoring program is
recommended.
MW-211D is spatially redundant with MW-204D and they are screened at similar
depths as shown on Figure 2.3. Sampling of both of these wells is not required to define
the western plume boundary in this area; continued sampling of MW-204D is
recommended.
Future sampling of 20 existing deep zone wells on an annual to biennial basis is
recommended, as detailed in Table 4.3. Eleven of these 20 wells (MW-101D, 205D,
205S, 207D, 212D, 212S, 213D, 303, 305D, 3051, and 3D) are currently sampled
semiannually as part of ongoing LTM activities. The remaining nine wells (MW-103,
104, 204D, 204S, 206D, 206S, 2D, 4, and 102D) are not currently sampled and are
recommended for future inclusion in the monitoring program.
MW-3D, 103, 204S, 204D, 3051, and 305D are recommended for future sampling to
assist in the definition of the lateral plume boundaries over time in the deep zone. The
recommended sampling frequency for these wells is annual (MW-305I/D only) to
biennial given their inferred cross-gradient location relative to the VOC plume and the
apparently stable to diminishing extent of the plume footprint over the last several years.
However, if hydraulic conditions change (e.g., change in on-site pumping conditions or
4-12
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
initiation of off-site groundwater extraction in the vicinity), or if new sampling results for
wells not sampled recently indicate plume expansion, then more frequent sampling
should be considered.
Thirteen of the remaining 14 deep wells (MW-101D, 205D, 205S, 207D, 212D, 212S,
213D, 303, 104, 206D, 206S, 2D, and 4) are recommended for continued sampling
because their plume-interior locations facilitate assessment of remedial progress over
time. The most recent sampling event performed in each of these wells has indicated the
presence of at least one COC at concentrations that exceed 307 Type B cleanup goals.
Available data indicate that five of these 13 wells (MW-104, 206D, 206S, 2D, and 4)
were last sampled in October 2003; those sampling results indicate PCE concentrations
ranging from 3.7 to 81 |ig/L (compared to the Type B cleanup criterion of 0.7 |ig/L) and
TCE concentrations ranging from non-detect to 44 |ig/L (compared to a Type B cleanup
criterion of 3 |ig/L). If current VOC concentrations in these five wells are below cleanup
levels over at least two consecutive monitoring events, then removal of these wells from
the LTM program could be considered. An annual sampling frequency for 11 of the 13
plume-interior wells (all except for MW-303 and MW-207D, see Table 4.3 for details) is
recommended. A reduction in sampling frequency relative to the current semiannual
program is supported by the lack of increasing trends in these wells and the fact that there
are no nearby receptors (the distance from the most downgradient of these well to the
Pere Marquette River is approximately 500 feet, versus an estimated average
groundwater flow velocity of 185 feet per year). The plume in the deep zone appears to
be generally stable to diminishing under the influence of the GWE system and natural
attenuation processes.
The last deep well recommended for continued sampling (MW-102D) is located on
the north (downgradient) side of the river, and was non-detect for COCs during four
sampling events performed from October 1997 to May 1999. Additional sampling of this
well at a low (biennial) frequency is recommended given that it serves as a sentry well to
verify the lack of plume underflow beneath the river in the deep zone of the surficial
aquifer.
4.3 DATA GAPS
Specific data gaps in the groundwater monitoring network were assessed during
performance of the qualitative evaluation, as summarized in the following subsections. It
is our understanding that additional groundwater quality data (apart from the monitoring
and extraction well data assessed for this LTMO evaluation) were collected during
previous site characterization activities. For example, the database contains vertical
profiling VOC data for "GP"-series sampling points (perhaps referring to Geoprobe
points). However, the locations of these additional samples were not provided to
Parsons. Therefore, the validity of the recommendations for additional well installations
provided below should be weighed in light of all available data.
4.3.1 Shallow Portion of the Surficial Aquifer
COC concentrations in the shallow portion of the surficial aquifer are not well defined
downgradient of the source area, as shown on Figures 2.4 and 4.1. This is partly because
five of the wells designated as being "shallow" are screened in what appears to be the
4-13
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
upper portion of the "deep" zone, as shown on Figure 2.3. Therefore, there are no plume
interior wells installed between EW-2 (located near the source area) and MW-305S
(located near the downgradient plume toe). As a result, the current extent and magnitude
of the plume in the shallow zone downgradient of the source area is not well defined, and
remedial progress in this area cannot be assessed (assuming that there is significant
groundwater contamination present in this area—the presence of PCE concentrations
exceeding the 307 Type B criterion in MW-305S suggests that there is). The only GWE
well screened entirely in the shallow zone is EW-2; EW-1 is screened in the lower
portion of the shallow zone and the upper portion of the deep zone (Figure 2.3). As a
result, much of the shallow zone between EW-2 and the river may not be impacted by the
GWE system, depending on the capture zones of the deeper extraction wells..
Installation of up to approximately seven new shallow zone monitoring wells (S-l
through S-7) at the locations shown on Figure 4.1 would better define the extent and
magnitude of the shallow zone plume and facilitate assessment of progress in achieving
cleanup goals in the shallow zone across the site. This information would permit more
accurate assessment of the schedule and cost to complete site remediation. The screen
intervals of these wells should be based on the results of historical COC data for
monitoring wells as well as vertical profiling data collected in the 1990s during site
characterization activities. In addition, the MDEQ has indicated that MW-202 has
reportedly been destroyed; this well should be replaced with a new well at the same
location.
Installation of wells S-4 and S-5 should be conditional on sampling results from S-2
and S-3. If S-2 and S-3 appear to bound the higher-concentration portion of the dissolved
contaminant plume, then installation of S-4 and/or S-5 may not be necessary. If installed,
it is assumed that these wells would be sampled annually for at least three years, followed
by re-evaluation of the monitoring frequency based on results obtained.
4.3.2 Deep Portion of the Surficial Aquifer
Groundwater quality in the deep portion of the surficial aquifer is well characterized
along an east-west-trending band located approximately 500 feet from the Pere Marquette
River. Several wells screened in the deep zone have been installed in the general vicinity
of GWE wells EW-4, 5, and 6 to define dissolved chlorinated ethene concentrations and
the lateral plume boundaries in this area (Figure 4.2). The next best-characterized area in
terms of groundwater quality is located near the southern edge of the plume, where four
monitoring wells and one GWE well have been installed. There is a large area between
the two areas described above that is relatively devoid of deep zone wells, and
installation of up to three deep monitoring wells in this area (D-l, D-2, and D-3), as
shown on Figure 4.2, would allow better definition of the extent and magnitude of the
plume to monitor remedial progress over time. Installation of D-3 could be optional and
dependent on sampling results from D-2.
Installation of a fourth new well (D-4) directly downgradient of the elevated VOC
concentrations detected at the MW-305S/306D well pair is recommended to better define
the downgradient extent of elevated VOC concentrations and the proximity of the plume
toe to the river. The degree to which the plume is discharging to the river is not known at
this time. Installation of two additional new deep zone wells (D-5 and D-6) is also
4-14
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
recommended to pair the recommended new shallow wells S-6 and S-7 and better define
groundwater quality near the river. If installed, it is assumed that these wells would be
sampled annually for at least three years, followed by re-evaluation of the monitoring
frequency based on results obtained.
4.4 ANALYTICAL PROGRAM
Groundwater samples are analyzed by the MDEQ Environmental Laboratory for a
total of 66 VOCs using method 8260. In addition, analysis for metals is routinely
performed, although it is not clear if all samples are analyzed for metals during every
sampling event. Generation of significant metals contamination at a laundry/dry-cleaner
site would not typically be expected, and the degree to which detected metal
concentrations may be representative of background conditions in the aquifer is not well
understood. The following recommendations pertaining to the groundwater analytical
program should be considered:
1) Discuss optimizing the target VOC list to a short-list of key compounds (i.e., the
chlorinated ethenes PCE, TCE, DCE, and vinyl chloride) with the analytical laboratory.
Potential advantages include lower laboratory analytical costs and lower data
management/validation/reporting costs.
2) Compare detected metal concentrations to background concentrations from
upgradient and cross-gradient wells to assess the degree to which they are representative
of background conditions. If adequate background wells do not exist, consider installing
a small number of background wells to allow this question to be definitively addressed.
3) Confirm that wells with elevated metal concentrations do not have metal screens
that could be contributing to the elevated concentrations detected in groundwater
samples.
3) If background comparisons indicate that detected metal concentrations are site-
related as opposed to representative of background conditions (and if metal screens are
not an issue), then optimize the metals analysis program to a short-list of key analytes and
sampling points using the principles and procedures outlined in this report.
4) If sufficient dissolved oxygen and oxidation-reduction potential data are not
available, then obtain field data for these parameters in the source area to support
evaluation of appropriate source remediation approaches (such as in situ chemical
oxidation).
4.5 LTM PROGRAM FLEXIBILITY
The LTM program recommendations summarized in Table 4.3 are based on available
data regarding current (and assumed future) site conditions. Changing site conditions
(e.g., periods of drought or excessive rainfall or changes in hydraulic stresses such as
number and location of pumping wells or pumping rates) could affect contaminant fate
and transport. Therefore, the LTM program should be reviewed if hydraulic conditions
change significantly, and revised as necessary to adequately track changes in the
magnitude and extent of COCs in environmental media over time.
4-15
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
SECTION 5
TEMPORAL STATISTICAL EVALUATION
Chemical concentrations measured at different points in time (temporal data) can be
examined graphically or using statistical tests, to evaluate dissolved-contaminant plume
stability. If removal of chemical mass is occurring in the subsurface as a consequence of
attenuation processes or operation of a remediation system, mass removal will be
apparent as a decrease in chemical concentrations through time at a particular sampling
location, as a decrease in chemical concentrations with increasing distance from chemical
source areas, and/or as a change in the suite of chemicals detected through time or with
increasing migration distance.
5.1 METHODOLOGY FOR TEMPORAL TREND ANALYSIS OF
CONTAMINANT CONCENTRATIONS
Temporal chemical-concentration data can be evaluated for trends by plotting
contaminant concentrations through time for individual monitoring wells (e.g., Figure
5.1), or by plotting contaminant concentrations versus downgradient distance from the
contaminant source for several wells along the groundwater flowpath over several
monitoring events. Plotting temporal concentration data is recommended for any analysis
of plume stability (Wiedemeier and Haas, 2000); however, visual identification of trends
in plotted data may be a subjective process, particularly if (as is likely) the concentration
data do not exhibit a uniform trend, but are variable through time (Figure 5.2).
The possibility of arriving at incorrect conclusions regarding the fate and transport of
dissolved contaminants on the basis of visual examination of temporal concentration data
can be reduced by examining temporal trends in chemical concentrations using various
statistical procedures, including regression analyses and the Mann-Kendall test for trends.
The Mann-Kendall nonparametric test (Gibbons, 1994) is well-suited for evaluation of
environmental data because the sample size can be small (as few as four data points), no
assumptions are made regarding the underlying statistical distribution of the data, and the
test can be adapted to account for seasonal variations in the data. The Mann-Kendall test
statistic can be calculated at a specified level of confidence to evaluate whether a
statistically significant temporal trend is exhibited by contaminant concentrations
detected through time in samples from an individual well. A negative slope (indicating
decreasing contaminant concentrations through time) or a positive slope (increasing
concentrations through time) provides statistical confirmation of temporal trends that may
have been identified visually from plotted data (Figure 5.2). In this analysis, a 90%
confidence level is used to define a statistically significant trend.
5-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
FIGURE 5.1
PCE CONCENTRATIONS THROUGH TIME
AT WELL MW-212D
LONG-TERM MONITORING NETWORK OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
300
250
200
o
o
U
LU
O
Q.
150
100
50
0
Dec-00 Jun-01 Dec-01 Jun-02 Dec-02 Jun-03 Dec-03 Jun-04 Dec-04 Jun-05 Dec-05
The relative value of information obtained from periodic monitoring at a particular
monitoring well can be evaluated by considering the location of the well with respect to
the dissolved contaminant plume and potential receptor exposure points, and the presence
or absence of temporal trends in contaminant concentrations in samples collected from
the well. The degree to which the amount and quality of information that can be obtained
at a particular monitoring point serves the two primary (i.e., temporal and spatial)
objectives of monitoring (Section 1) must be considered in this evaluation. For example,
the continued non-detection of a target contaminant in groundwater at a particular
monitoring location provides no information about temporal trends in contaminant
concentrations at that location, or about the extent to which contaminant migration is
occurring, unless the monitoring location lies along a groundwater flowpath between a
contaminant source and a potential receptor exposure point (e.g., downgradient of a
known contaminant plume). Therefore, a monitoring well having a history of
contaminant concentrations below detection limits may be providing little or no useful
information, depending on its location.
5-2
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
Decreasing Trend
Increasing Trend
No Trend
Confidence Factor
HIGH
Confidence Factor
LOW
Variation
LOW
Variation
HIGH
FIGURE 5.2
CONCEPTUAL REPRESENTATION OF
TEMPORAL TRENDS AND TEMPORAL
VARIATIONS IN CONCENTRATIONS
Long-Term Monitoring Optimization
Wash King Laundry Superfund Site
draw\739732\diffusion\williamsA.cdr pg1 nap 4/3/02
5-3
-------
A trend of increasing contaminant concentrations in groundwater at a location between
a contaminant source and a potential receptor exposure point may represent information
critical in evaluating whether contaminants are migrating to the exposure point, thereby
completing an exposure pathway. Identification of a trend of decreasing contaminant
concentrations at the same location may be useful in evaluating decreases in the areal
extent of dissolved contaminants, but does not represent information that is critical to the
protection of a potential receptor. Similarly, a trend of decreasing contaminant
concentrations in groundwater near a contaminant source may represent important
information regarding the progress of remediation near, and downgradient from, the
source. By contrast, the absence of a statistically significant (as defined by the Mann-
Kendall test with a 90% confidence level) temporal trend in contaminant concentrations
at a particular location within or downgradient from a plume indicates that virtually no
additional information can be obtained by frequent monitoring of groundwater at that
location, in that the results of continued monitoring through time are likely to fall within
the historic range of concentrations that have already been detected (Figure 5.3).
Continued monitoring at locations where no temporal trend in contaminant
concentrations is present serves merely to confirm the results of previous monitoring
activities at that location.
The temporal trends and relative location of wells can be weighed to determine if a
well should be retained, excluded, or continued in the program with reduced sampling.
Figure 5.4 presents a flowchart demonstrating the method for using trend results to draw
these conclusions.
5.2 TEMPORAL EVALUATION RESULTS FOR GROUNDWATER WELLS
The analytical data for groundwater samples collected from the 47 groundwater
monitoring and extraction wells in the Wash King LTM program from April 2001
through August 2005 were examined for temporal trends using the Mann-Kendall test.
Note that only recent (i.e. post-extraction well start up) analytical results were used in this
analysis. The temporal analysis is intended to measure trends obtained from a "steady
state" system, so using results obtained before the extraction system was implemented
could lead to spurious trends. The objective of the evaluation was to identify those wells
having increasing or decreasing concentration trends for each COC, and to consider the
quality of information represented by the existence or absence of concentration trends in
terms of the location of each monitoring point. Increasing or decreasing trends are those
identified as having positive or negative slopes, respectively, by the Mann-Kendall trend
analysis with a confidence level of 90%.
Summary results of Mann-Kendall temporal trend analyses for COCs in groundwater
samples from Wash King are presented in Table 5.1. Table 5.1 also contains the relative
location designation assigned to each well and the number of results used in the analysis.
Trends for the three COCs (PCE, TCE, and DCE) were evaluated to assess the value of
temporal information provided by each well. As implemented, the algorithm used to
evaluate concentration trends assigned a value of "ND" (not detected) to those wells with
sampling results that were consistently below analytical detection limits through time,
rather than assigning a surrogate value corresponding to the detection limit - a procedure
that could generate potentially misleading and anomalous "trends" in concentrations.
5-4
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
c
o
+*
re
+->
c
0)
u
c
o
O
Likely Future
Results
(A
=
CD 2
c =
a g
0£ g
o
O
Historic Results
Time
FIGURE 5.3
CONCEPTUAL REPRESENTATION
OF CONTINUED MONITORING AT
LOCATION WHERE NO TEMPORAL
TREND IN CONCENTRATIONS
IS PRESENT
Long-Term Monitoring Optimization
Wash King Laundry Superfund Site
draw\739732\diffusion\williamsA.cdr pg2 nap 4/3/02
5-5
-------
FIGURE 5.4
TEMPORAL TREND DECISION RATIONALE FLOWCHART
LONG-TERM MONITORING NETWORK OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Recent
Concentrations
« MCLs?
Increasing
Trend?
Exclude/Reduce
Frequency
Exclude/Reduce
Frequency
Recent
Concentrations
«MCLs?
Decreasing
Trend?
Dovwigradient v Y
Sentry Well? '
Excude/Reduce
Dovwigradient
Sentry Well?
No—*/ High Variation? >—Yes-W Retain
Exclude/Reduce
Downgradient
Sentry Well?
Exclude/Reduce
Frequency
5-6
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
CO
s
i
w
a!
O
5
§
|
o
K
W
H
<<
^
ll
s
!*
||B
iis
K g o
s si
fe 9 1
O § J
a sg
CC C* H
>H H W
S ^rs=
•< ° 2
iar !T "fl
^3^
OJ
1
Retain
Exclude/
Reduce
tt
I
•1
U
g
s
O
—
Number of
Results
.2
Relative Loca
aulic Unit
1
Well Name
1
variation TCE downg
so
ri
OJ
1
X
g
No Trend
S
g
Q
~
c
(U
T3
ra
D)
o
Q
0
i
tt
1
> standard.
I
&
PCE downg
M
1
X
Decreasing
Decreasing
,
c
(U
T3
ra
D)
O
Q
0
t3
ti
W
LU
J DCE recent NDs.
ear standard. TCE an
(D
'•3
fch
PCE down
easing
1
X
Decreasing
Decreasing
?
m
S
b
(U
Q
~
c
(U
T3
ra
D)
o
0
t3
tl
LJJ
OJ
•3
s
00
o
§
w
M
1
X
Decreasing
i
D)
C
'(/)
0)
o
(U
Q
«
c
(U
T3
ra
D)
o
Q
0
i
tt
LU
vngradient.
ow variation TCE do\
1
g
ra
0)
Q
X
Decreasing
I
D)
C
(U
Q
9
c
(U
T3
ra
D)
o
Q
0
t3
ti
W
1
LU
1
1
a
1
1
1
VI
W
•o
easing
(D
Q
X
Q
Z
Probably Decreasing
D)
C
(U
Q
0
c
(U
T3
ra
D)
1
Q
&
(D
Q
Q
:
&
g
^
X
Q
Z
!
T3
C
(U
1-
o
0)
0
o
1
33
w
i
g DCE « standard.
,g
S3
0>
•3
a
M
O
O
E"1
ion PCE an(
1
X
Decreasing
No Trend
T3
c
(U
1-
o
-
c
(U
T3
ra
D)
o
Q
&
(D
Q
Q
1
1
g
tion PCE do
1
•a
3
X
1
Probably Decreasing
T3
c
(U
1-
O
-
c
(U
T3
ra
D)
o
Q
_0
13
VI
1
lowngradient.
1
i
ion PCE < s
1
o
J
X
Q
Z
Q
Z
T3
C
(U
H
o
0
(U
T3
ra
D)
o
Q
o.
Q
Q
2
iowngradient.
1
3
ion PCE < s
_ri
ri
X
Q
Z
Q
Z
T3
C
(U
1-
o
0
c
(U
T3
ra
D)
o
Q
_0
13
VI
2
b
1
S3
1
(D
•o
o
X
Q
Z
Q
Z
Z
-
c
(U
T3
ra
S
o
O
_0
13
VI
1
°
I
%
O
1
•o
o
X
Q
Z
Q
Z
Z
0
•E
(U
T3
O
O
_0
VI
2
b
1
ffi
O
|
•o
o
X
Q
Z
Q
Z
Z
0)
•E
(U
T3
O
O
_0
13
VI
0
C\l
2
ards downgradient.
creasing TCE > stand
w"
^
ion PCE am
1
o
J
X
T3
C
0)
1-
o
Z
Decreasing
T3
c
(U
1-
o
0
c
(U
T3
ra
D)
o
(D
Q
Q
C\l
C\l
2
i recent NDs.
P
S
1
05
M
0
W
X
Decreasing
Decreasing
D)
c
gj
o
(U
Q
-
c
(U
T3
ra
D)
o
Q
_0
13
VI
W
C\l
C\l
ing event. Decreasing PCE > standard.
G
rt
1
O
G
•0
S
from ND to
&
£
X
Q
Z
Probably Increasing
D)
C
(U
Q
C\l
c
(U
T3
ra
D)
o
Q
&
(D
Q
Q
CO
C\l
owngradient.
1
Jj
ion PCE > s
1
o
J
X
Q
Z
Q
Z
T3
C
(U
h-
o
OM
(U
T3
ra
D)
O
Q
_0
13
VI
w
CO
C\l
2
b
1
a
S3
o
S3
|
•o
o
X
Q
Z
Q
Z
N
•E
(U
T3
O
O
&
(D
Q
Q
i
of TCE < standard in 12/02.
ngradient. One detect
-!•
1
V
ri
OJ
1
X
Q
Z
!
ecreasing
Q
5
o
CL
(D
c
(U
T3
ra
D)
O
Q
_0
13
VI
(n
i
g or ND downgradien
creasin
Q
standard. One DCE dection «standard in 8/05
in most recent sampli
i
t/3
from ND >:
&
£
X
Probably Increasing
1
D)
C
(U
Q
00
c
(U
T3
ra
D)
o
Q
&
(D
Q
2
b
1
S3
1
(D
•o
o
X
Q
Z
Q
Z
Z
00
•E
(U
T3
O
O
&
Q
Q
1
b
1
oa
1
1
(D
•o
o
X
Q
Z
Q
Z
Z
00
c
(U
T3
ra
I5
o
0
IH
•3
£
1
o
down grad
|
•o
o
X
Q
Z
Q
Z
Z
(D
•E
(U
T3
ra
D)
O
Q
_0
13
VI
W
2
1
|
•o
o
X
Q
Z
Q
Z
Z
00
•p
(U
C/)
Downgradient
o.
(D
Q
Q
2
wa
1
(D
•o
o
X
Q
Z
Q
Z
Z
-
£T
(U
c/)
Downgradient
•3
,3
1
id DCE ND.
it < standards. TCE ar
.§
S
ion PCE do
1
X
Q
Z
Q
Z
T3
C
(U
1-
o
00
c
(U
T3
ra
D)
o
Q
_0
13
VI
w
i
;ent NDs PCE & TCE.
id down gradient. Rec
o
c
0
1
ra
0)
Q
X
Q
Z
1
D)
C
0)
o
(U
Q
-
c
(U
T3
ra
I5
o
0
0>
Q
pi
CO
°
I
S3
o
S3
|
•o
o
X
Q
Z
Q
Z
Z
0
•E
(U
T3
O
O
_0
13
VI
1
TI
Mot Currently Samp
U)
1
in
han 4 measuremen
0)
mendation
ecom
o
<4meas
<4meas
tn
V
CO
c
(U
T3
ra
D)
o
Q
o.
Q
|
in
han 4 measuremen
-------
The color-coding of Table 5.1 entries denotes the presence or absence of temporal
trends, and allows those monitoring points having nondetectable concentrations,
decreasing or increasing concentrations, or no discernible trend in concentrations to be
readily identified. Those trend that have confidence levels between 90 and 95% are
indicated by the "probably increasing" and "probability decreasing" classifications
("increasing" and "decreasing" classifications correspond to confidence levels of over
95%). Trend results with bold borders indicate those analytical results that contained
over 50% non-detections, historically. Although these chemical trends are not deserving
of the "ND" classification and resulting decision process, decisions made based on these
trends should take the high number of non-detects into consideration.
Fourteen of the seventeen monitoring wells that are not currently sampled had fewer
than four analytical results for each of the COCs could not be analyzed using the Mann-
Kendall trend analysis, and have a "<4Meas" designation. Figures 5.5 and 5.6 display the
Mann-Kendall results for PCE thematically by well, along with the relative plume
location for the shallow and deep zone wells, respectively.
The basis for the decision to exclude, reduce the sampling frequency, or retain a well
in the monitoring program based on the value of its temporal information is described in
the "Rationale" column of Table 5.1, and a flow chart of the decision logic applied to the
temporal trend analysis results is presented on Figure 5.4. Trend results for PCE were
given more weight than those from the other COCs, given its relatively higher impact;
however, the most conservative trend was used in all cases (e.g., if TCE trend resulted in
a recommendation to retain a well, that well would be recommended for retention.)
Wells that have decreasing trends in a source area in which concentrations are above
standards (e.g., MW-101D) are valuable because they provide information on the
effectiveness of the remedial actions performed to date and are thus recommended for
retention in the monitoring system; conversely, wells located downgradient area that have
either decreasing concentrations (e.g., MW-212S, EW-4) will provide limited valuable
temporal information in the future and are recommended for exclusion or reduced
sampling. Downgradient wells with increasing concentrations (e.g., EW-2, EW-4 and
MW-213D) are valuable because they identify areas where the plume may be expanding
or where remedial systems are not effective. Wells with stable (low variation), 'no trend'
results (e.g., MW-205D and MW-305S) were recommended for exclusion or monitoring
reduction because continued frequent sampling would not likely yield new information,
while wells with highly variable COC concentrations (e.g., wells MW-101S and MW-
303) were recommended for retention.
Table 5.1 summarizes recommendations to retain 10 and exclude or reduce the
frequency for 26 of the 36 wells analyzed in the temporal evaluation (not including the
non-sentry wells with fewer than four measurements). The recommendations provided in
Table 5.1 are based on the evaluation of temporal statistical results only, and must be
used in conjunction with the results of the qualitative and spatial evaluations to generate
final recommendations regarding retention of monitoring points in the LTM program, and
the frequency of monitoring at particular locations in Wash King.
5-8
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
o
w
I
2"/
I «/
\S
U\
\0 \
-Jn-^1 _
1 1
d^
Q a
#
Q
DD
CO*
h uj
Ujffi
z IE
S8g
HI
oc
^
o
^ ^
z LJJ
Ob
tFW
^°
§ui
opi
o
o:
o
< < r^
CO
UJ
2^
§ £
mp
^ D
£^
o£
O D
Z W
is
1<
^o
a: z
UJ ^
Ow
in
o
in
oc
<
a
0*000
-------
- O) ,_
CD CO o
"0-0
- CD O)
O CD CO
-------
SECTION 6
SPATIAL STATISTICAL EVALUATION
Spatial statistical techniques also can be applied to the design and evaluation of
groundwater monitoring programs to assess the quality of information generated during
monitoring and to evaluate monitoring networks. Geostatistics, or the theory of
regionalized variables (Clark, 1987; Rock, 1988; American Society of Civil Engineers
Task Committee on Geostatistical Techniques in Hydrology, 1990a and 1990b), is
concerned with variables having values dependent on location, and which are continuous
in space but vary in a manner too complex for simple mathematical description.
Geostatistics is based on the premise that the differences in values of a spatial variable
depend only on the distances between sampling locations, and the relative orientations of
sampling locations - that is, the values of a variable (e.g., chemical concentration)
measured at two locations that are spatially close together - will be more similar than
values of that variable measured at two locations that are far apart.
6.1 GEOSTATISTICAL METHODS FOR EVALUATING MONITORING
NETWORKS
Ideally, application of geostatistical methods to the results of the groundwater
monitoring program at Wash King could be used to estimate COC concentrations at
every point within the dissolved contaminant plume, and also could be used to generate
estimates of the "error," or uncertainty, associated with each estimated concentration
value. Thus, the monitoring program could be optimized by using available information
to identify those areas having the greatest uncertainty associated with the estimated
plume extent and configuration. Conversely, sampling points could be successively
eliminated from simulations, and the resulting uncertainty examined, to evaluate if
significant loss of information (represented by increasing error or uncertainty in
estimated chemical concentrations) occurs as the number of sampling locations is
reduced. Repeated application of geostatistical estimating techniques, using tentatively
identified sampling locations, then could be used to generate a sampling program that
would provide an acceptable level of uncertainty regarding the distribution of COCs with
the minimum possible number of samples collected. Furthermore, application of
geostatistical methods can provide unbiased representations of the distribution of COCs
at different locations in the subsurface, enabling the extent of COCs to be evaluated more
precisely. The application of geostatistics at hazardous, toxic, and radioactive waste sites
is discussed in greater depth in USAGE Engineer Technical Letter 1110-1-175 (1997).
Fundamental to geostatistics is the concept of semivariance [j(h)], which is a measure
of the spatial dependence between sample variables (e.g., chemical concentrations) in a
6-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
specified direction. Semivariance is defined for a constant spacing between samples (h)
by:
y(h) = — JL[g(x) - g(x + h) ]2 Equation 6-1
Where:
j(h) = semivariance calculated for all samples at a distance h from each other;
g(x) = value of the variable in sample at location x;
g(x + h) = value of the variable in sample at a distance h from sample at location x;
and
n = number of samples in which the variable has been determined.
Semivariograms (plots of y(h) versus h) are a means of depicting graphically the range
of distances over which, and the degree to which, sample values at a given point are
related to sample values at adjacent, or nearby, points, and conversely, indicate how close
together sample points must be for a value determined at one point to be useful in
predicting unknown values at other points. For h = 0, for example, a sample is being
compared with itself, so normally y(0) = 0 (the semivariance at a spacing of zero, is
zero), except where a so-called nugget effect is present (Figure 6.1), which implies that
sample values are highly variable at distances less than the sampling interval. Analytical
variability and sampling error can contribute to the nugget. As the distance between
samples increases, sample values become less and less closely related, and the
semivariance therefore increases, until a "sill" is eventually reached, where y(h) equals
the overall variance (i.e., the variance around the average value). The sill is reached at a
sample spacing called the "range of influence," beyond which sample values are not
related. Only values between points at spacings less than the range of influence can be
predicted; but within that distance, the semivariogram provides the proper weightings,
which apply to sample values separated by different distances.
When a semivariogram is calculated for a variable over an area (e.g., concentrations of
PCE in the Wash King shallow zone), an irregular spread of points across the
semivariogram plot is the usual result (Rock, 1988). One of the most subjective tasks of
geostatistical analysis is to identify a continuous, theoretical semivariogram model that
most closely follows the real data. Fitting a theoretical model to calculated semivariance
points is accomplished by trial-and-error, rather than by a formal statistical procedure
(Davis, 1986; Clark, 1987; Rock, 1988). If a "good" model fit results, then y(h) (the
semivariance) can be confidently estimated for any value of h, and not only at the
sampled points.
6-2
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
FIGURE 6.1
IDEALIZED SEMIVARIOGRAM MODEL
LONG-TERM MONITORING NETWORK OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
3500 -
-------
used in the deep well set because they are screened significantly below the other deep
wells; MW-304D and MW-305D both have associated "intermediate" wells that would
be more appropriate to include, as shown in Figure 2.3.
The commercially available geostatistical software package Geostatistical Analyst™
(an extension to the Arc View® geographic information system [GIS] software package)
(Environmental Systems Research Institute, Inc. [ESRI], 2001) was used to develop
semivariogram models depicting the spatial variation in the shallow and deep aquifers for
PCE in groundwater.
As semivariogram models were calculated for each scenario (Equation 6-1),
considerable scatter of the data was apparent during fitting of the models. Several data
transformations (including a log transformation) were attempted to obtain a
representative semivariogram model. Ultimately, the concentration data were
transformed to "rank statistics," in which, for example, the 24 wells in the deep zone
were ranked from 1 (lowest concentration) to 24 (highest concentration) according to
their most recent PCE concentrations. Tie values were assigned the median rank of the
set of ranked values; for example, if five wells had non-detected concentrations, they
would each be ranked "3", the median of the set of ranks: [1,2,3,4,5]. Transformations of
this type can be less sensitive to outliers, skewed distributions, or clustered data than
semivariograms based on raw concentration values, and thus may enable recognition and
description of the underlying spatial structure of the data in cases where ordinary data are
too "noisy."
The rank statistics were used to develop semivariograms that most accurately modeled
the spatial distribution of the data in the two scenarios. The parameters for best-fit
semivariograms for the two spatial evaluations are listed in Table 6.1.
TABLE 6.1
BEST-FIT SEMIVARIOGRAM MODEL PARAMETERS
LONG-TERM MONITORING NETWORK OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Parameter
Model
Range (ft)
Sill
Nugget
Shallow Zone
Circular
1000
17.5
0
Deep Zone
Circular
600
45
15
After the semivariogram models were developed, they were used in the kriging system
implemented by the Geostatistical Analyst™ software package (ESRI, 2001) to develop
two-dimensional kriging realizations (estimates of the spatial distribution of PCE in
groundwater at Wash King), and to calculate the associated kriging prediction standard
errors. The median kriging standard deviation was obtained from the standard errors
calculated using the entire monitoring network for each scenario (e.g., the 24 wells in the
deep zone). Next, each of the wells was sequentially removed from the network, and for
each resulting well network configuration, a kriging realization was completed using the
6-4
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
COC concentration rankings from the remaining wells. The "missing-well" monitoring
network realizations were used to calculate prediction standard errors, and the median
kriging standard deviations were obtained for each "missing-well" realization and
compared with the median kriging standard deviation for the "base-case" realization
(obtained using the complete monitoring network), as a means of evaluating the amount
of information loss (as indicated by increases in kriging error) resulting from the use of
fewer monitoring points.
Figure 6.2 illustrates an example of the spatial-evaluation procedure by showing
kriging prediction standard-error maps for three kriging realizations for the 15 shallow
zone wells. Each map shows the predicted standard error associated with a given group
of wells based on the semivariogram parameters discussed above. Lighter colors
represent areas with lower spatial uncertainty, and darker colors represent areas with
higher uncertainty; regions in the vicinity of wells (i.e., data points) have the lowest
associated uncertainty. Map A on Figure 6.2 shows the predicted standard error map for
the "base-case" realization in which all 15 wells are included. Map B shows the
realization in which well MW-207S was removed from the monitoring network, and Map
C shows the realization in which MW-201 was removed. Figure 6.2 shows that when a
well is removed from the network, the predicted standard error in the vicinity of the
missing well increases (as indicated by a darkening of the shading in the vicinity of that
well). If a "removed" (missing) well is in an area with several other wells (e.g., well
MW-207S; Map B on Figure 6.2), the predicted standard error may not increase as much
as if a well (e.g., MW-201; Map C) is removed from an area with fewer surrounding
wells.
Based on the kriging evaluation, each well received a relative value of spatial
information "test statistic" calculated from the ratio of the median "missing well" error to
median "basecase" error. If removal of a particular well from the monitoring network
caused very little change in the resulting median kriging standard deviation, the test
statistic equals one, and that well was regarded as contributing only a limited amount of
information to the LTM program. Likewise, if removal of a well from the monitoring
network produced larger increases in the kriging standard deviation (typically more than
1 percent), this was regarded as an indication that the well contributes a relatively greater
amount of information and is relatively more important to the monitoring network. At
the conclusion of the kriging realizations, each well was ranked from 1 (providing the
least information) to the number of wells included in the zone analysis (providing the
most information), based on the amount of information (as measured by changes in
median kriging standard deviation) the well contributed toward describing the spatial
distribution of PCE, as shown in Tables 6.2 and 6.3. Wells providing the least amount of
information represent possible candidates for exclusion from the monitoring network at
Wash King.
6.3 SPATIAL STATISTICAL EVALUATION RESULTS
Figures 6.3 and 6.4 and Tables 6.2 and 6.3 present the test statistics and associated
rankings of the evaluated subsets of monitoring locations for PCE in the shallow and
deep zones, respectively. The wells are ranked from least to most spatially relevant based
on the relative value of the associated recent COC information provided by each well, as
calculated based on the kriging realizations. Examination of these results indicate that
6-5
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
A) Basecase (All wells)
B) Missing well MW-207S:
relative small change in
spatial uncertainty
C) Missing well MW-201:
relative large change in
spatial uncertainty
Legend
Well missing from
kriging realization
Predicted Standard Error Map
Less spatial uncertainty
o
Greater spatial uncertainty
FIGURE 6.2
IMPACT OF MISSING WELLS
ON PREDICTED STANDARD ERROR
LONG TERM MONITORING OPTIMIZATION
PARSONSD
-------
TABLE 6.2
RESULTS OF GEOSTATISTICAL EVALUATION RANKING OF WELLS
BY RELATIVE VALUE OF PCE IN THE SHALLOW ZONE
LONG-TERM MONITORING OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Well Name ^
MW-304S
MW-207S
MW-213S
MW-208
MW-101S
MW-102S
MW-209
MW-305S
MW-210
MW-7S
MW-202
MW-215
MW-8S
MW-201
MW-3S
Kriging
Metric
1.0025
1.0050
1.0072
1.0073
1.0122
1.0160
1.0193
1.0210
1.0252
1.0303
1.0336
1.0436
1.0580
1.0702
1.1400
Kriging
Ranking b/
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Exclude
X
X
X
X
c/
~
~
~
~
~
~
Retain
~
~
~
~
~
~
~
X
X
X
X
z Well set includes 15 "shallow zone" wells designated in Table 2.1.
b/1= least relative amount of information; 15= most relative amount of information.
c/ Well in the "intermediate" range; received no recommendation for excludsion or retention.
(see Section 6.2).
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\3.xls
6-7
-------
TABLE 6.3
RESULTS OF GEOSTATISTICAL EVALUATION RANKING OF WELLS BY RELATIVE
VALUE OF PCE IN THE LOWER ZONE
LONG-TERM MONITORING OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Well Name ^
MW-301D
MW-301S
MW-206D
MW-206S
MW-303
MW-212S
MW-205D
MW-212D
MW-302
MW-101D
MW-205S
MW-204S
MW-204D
MW-213D
MW-104
MW-304I
MW-4
MW-3D
MW-207D
MW-105
MW-2D
MW-8D
MW-103
MW-305I
Kriging
Metric
0.9998
0.9998
0.9999
.0000
.0001
.0003
.0005
.0005
.0005
.0011
.0013
.0016
.0022
.0024
.0032
.0035
.0041
.0062
.0089
.0123
.0139
.0159
.0195
.0312
Kriging
Ranking b/
1
2
3
4
5
6
gC/
8
8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Exclude
X
X
X
X
X
X
X
X
X
d/
—
~
—
~
~
—
~
Retain
~
—
~
—
~
~
—
~
X
X
X
X
X
X
X
Well set includes "deep zone" wells designated in Table 2.1, excluding MW-304D, MW-305D, and MW-102D.
b/1= least relative amount of information; 24= most relative amount of information.
c/ Tie values receive the median ranking of the set.
* Well in the "intermediate" range; received no recommendation for excludsion or retention.
(see Section 6.2).
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\3.xls
6-8
-------
-------
-------
monitoring wells in close proximity to several other monitoring wells (e.g., red color
coding on Figures 6.3 and 6.4) generally provide relatively lesser amounts of information
than do wells at greater distances from other wells or wells located in areas having
limited numbers of monitoring points (e.g., blue color coding on Figures 6.3 and 6.4).
This is intuitively obvious, but the analysis allows the most valuable and least valuable
wells to be identified quantitatively. For example, Table 6.3 identifies the wells ranked 8
and below that provide the relative least amount of information, and the wells ranked at
or above 18 that provide the greatest amount of relative information regarding the
occurrence and distribution of PCE in groundwater among those wells in the lower zone.
The lowest-ranked wells are potential candidates for exclusion from the Wash King
groundwater monitoring program, and the highest-ranked wells are candidates for
retention in the monitoring program; intermediate-ranked wells receive no
recommendation for removal or retention in the monitoring program based on the spatial
analysis. The recommendations provided in Tables 6.2 and 6.3 are based on the
evaluation of spatial statistical results only, and must be used in conjunction with the
results of the qualitative and temporal evaluations to generate final recommendations
regarding retention of monitoring points in the LTM program, and the frequency of
monitoring at particular locations in Wash King. Also note that due to the limitations of
the data (e.g., 2005 results are not available for all wells) the qualitative and temporal
results are given more weight in the combined analysis.
6-11
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
SECTION 7
SUMMARY OF LONG-TERM MONITORING OPTIMIZATION
EVALUATION
Forty-nine groundwater monitoring and extraction wells at Wash King were evaluated
qualitatively using hydrogeologic, hydrologic, and contaminant information, and
quantitatively using temporal and spatial statistical techniques. As each tier of the
evaluation was performed, monitoring points that provide relatively greater amounts of
information regarding the occurrence and distribution of COCs in groundwater were
identified, and were distinguished from those monitoring points that provide relatively
lesser amounts of information. In this section, the results of the evaluations are combined
to generate a refined monitoring program that potentially could provide information
sufficient to address the primary objectives of monitoring, at reduced cost. Monitoring
points not retained in the refined monitoring network could be removed from the
monitoring program with relatively little loss of information. It should be noted that
development of an optimized long-term monitoring program for the site would benefit
from a better understanding of the hydraulic effects of the current pump-and-treat system.
In addition, further evaluation of how the persistence of the VOC plume could be reduced
(e.g., via additional source removal efforts) should be considered.
7.1 GROUNDWATER MONITORING NETWORK SUMMARY
The results of the qualitative, temporal, and spatial evaluations for the groundwater
monitoring and extraction wells are summarized in Table 7.1, along with the final
recommendations for sampling point retention or exclusion and sampling frequency.
These final recommendations are also shown on Figures 7.1 and 7.2. The results of the
evaluations were combined and summarized in accordance with the decision logic shown
on Figure 7.3 and described below.
1. Each well retained in the monitoring network on the basis of the qualitative
hydrogeologic evaluation was recommended to be retained in the refined
monitoring program.
2. Those wells recommended for exclusion from the monitoring program on the
basis of all three evaluations, or on the basis of the qualitative and temporal
evaluations (with no recommendation resulting from the spatial evaluation)
were recommended for removal from the monitoring program.
3. If a well was recommended for removal based on the qualitative evaluation and
recommended for retention based on the temporal or spatial evaluation, the final
recommendation was based on a case-by-case review of well information.
7-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
o
o
o-
o
z
o
H
hH
Z
o
Q
Z
O
o
o
z
a
w
O5
^
£
w
w
H
o
TABLE 7.1
VALUATION
W
[IZATION
^
hH
H
PH
O
o
z
o
H
hH
Z
o
H
O
z
o
-
ta
o
z
O w
H hH
S O
§ 5
H b
O a*
I MONITORING
G LAUNDRY SU]
^ Z
LONG-TEB
WASHKI
t'
CS
S
G
in
iluation
W
•a
-*^
sS
a
VI
valuation
Temporal E
a
o
'is
3
13
W
llitative
d
-*^
g
OJ
s
•_
U
—
"3
o
1
Recommended
Monitoring
Frequency
a
'3
1
Exclude
a
'3
1
o
^
_s
"0
M
H
1
-^-i
2
Exclude/
Reduce
•a
•a S° >•
a ° w
S 'c 5
s s S
S"3 ^
III
C£
B
'3
1
Exclude
gf&
a s
c w
1 «
5/5 ta
o
d
OJ
§
Z
"3
£
ve evaluation. |
Temporal statistics confirm qualitati
Semiannual
X
T3
s
>:
a
p
^
a
0
o
s
X
W
p£
w
m
•*'
:
%
:
%
:
%
:
%
^
w
0
Statistics confirm qualitative evaluat
m
X
3
?
m
X
3
?
w
X
3
=?
W
X
3
?
W
X
3
=?
m
^
S
ive evaluation. |
Temporal statistics confirm qualitiat
1
r,
m
^
S
0
Statistics confirm qualitative evaluat
Biennial
X
X
X
^
(D
m
X
1
H
cj3
^
S
m
X
S
'o
i-l
a
-4-^
O
£
p.
0
Q
m
i-H
^
S
(N
-------
o
o
o-
o
z
o
H
hH
Z
o
Q
Z
O
o
o
z
a
w
O5
^
£
w
w
H
o
TABLE 7.1
VALUATION
W
[IZATION
^
hH
H
PH
O
o
z
o
H
hH
Z
o
H
O
z
o
-
ta
o
z
O w
H hH
S O
§ 5
H b
O a*
I MONITORING
G LAUNDRY SU]
^ Z
LONG-TEB
WASHKI
£
55
S
!/5
aluation
H
"cs
'•i
a
!/5
valuation
Temporal E
Evaluation
llitative
5
a
£
u
o
es
_o
'-£
£
Recommended
Monitoring
Frequency
a
'3
1
Exclude
a
'S
1
Exclude
1
-^-i
•2
Exclude/
Reduce
•a
•a S° ^
- S o
Recommer
Monitori
Frequen
a
'3
1
Exclude
M Jj
.S a
& S
1 *
5/5 ta
o
1
at
cS
Z
S
tnendations based
g
;—
•>
S
,
a
S
1
W
g m
.§ "*'
Hd ,
s
1
1
^
s
•>
a
g
:
B
<
1
1
>>%
a
->
a
13
o
a
ta
tative overrides s
a
Exclude
X
X
T3
s
>-.
a
-.
a
<
-^j
o
^
1
o
X
W
X
T3
1
1
tt
t3
tative overrides s
s
Exclude
X
lendation
o
o
£
o
^
T3
S
>->
S
->
"^3
a
^-
JRecommended r
1
progress ovei
C3
1
aa
•3
S
Tt
S
C3
n
i
CO
1
progress ovei
C3
1
aa
!
T-1
ca
-------
O
o
o-
o
z
o
H
hH
Z
o
Q
Z
O
o
o
z
a
w
O5
^
^
w
w
H
o
TABLE 7.1
VALUATION
tu
[IZATION
^
hH
H
PH
O
o
z
o
H
hH
Z
o
H
O
z
o
-
ti.
o
z
O w
H hH
S O
§ 5
H b
o s
I MONITORING
G LAUNDRY SU]
^ Z
LONG-TEB
WASHKI
£,
sS
S
S
s
0
Ti
CS
13
H
is
"S3
a.
_o
^
sS
S
13
Temporal E
Evaluation
V
'•g
._a
O-
—
CS
B
O
1
•a
icommende
Monitoring
Frequency
(2
_B
1
Exclude
I
"S
•a
"0
_B
"S
^
Exclude/
Reduce
Recommended
Monitoring
Frequency
_s
•8
£
OJ
•a
™
a a 1
i» — 3
s II
U^ |
a
tS
E
cS
Z
"3
^
-------
o
CM
CM
•a
_ CD
CD
O
c
o
^
ro
•o
c
0)
E
o
o
.E
— CD ^ CD CO
CO C O) ^ g
o JP. 3- — -a
^ £ c CD C
CO ^ O CD o O
^ ~a '-^ jz ci -^3
a~ CD co > a) o
llllll
.— S o co ^ o
I o O "O — ^
t
CD
(/) =5
= c
fl) C
5 i
> 0
0
CD •=
c -
-
DQ LU
0
T3
0
O
IT
a:
0
ro
'•£
0
"5
a.
O « O O 0 0
90/9/9 ueo pxuj-S||9AA6u!XHSEA/VSIO/6u!XHseAA/3OVSn/OI/\lll/P3LU3J/S3/:s
-------
-------
FIGURE 7.3
COMBINED EVALUTION SUMMARY DECISION LOGIC
LONG-TERM MONITORING NETWORK OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Retain Monitoring Point
(Increase Frequeno/on a Case-
by-Case Basis)
Reduce Monitoring Frequency
(Case-By-Case)
Exclude Well from Futwe
Sampling
4. If a well was recommended for retention based on the qualitative evaluation and
recommended for removal based on the temporal and spatial evaluation, the
well was recommended to be retained, but the possibility of reducing the
sampling frequency was evaluated based on a case-by-case review of well
information.
It should be noted, as stated in number four above, that the final recommended
monitoring frequencies that resulted from the combined analysis are not, in all cases, the
same as those recommended as a result of the qualitative evaluation. The justifications for
the final recommendations are provided in the "Rationale" column in Table 7.1, and fall
into the following general categories:
• Temporal and/or spatial statistical results confirm the sampling frequency
recommendations from the qualitative evaluation. For example, well MW-301S is
recommended for exclusion from the network or for sampling frequency reduction
by both the temporal and spatial statistical results; thus, the statistics confirm the
7-7
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
qualitative recommendation to exclude the well. Similarly, well MW-207S is
recommended for exclusion or reduction by the temporal and spatial statistical
results; thus the statistics confirm the relatively low (biennial) sampling frequency
recommended by the qualitative evaluation. Likewise, well MW-303 is
recommended for retention based on the temporal statistical evaluation, which
confirms maintaining the relatively higher (semiannual) sampling frequency
recommendation stemming from the qualitative evaluation.
• Increase monitoring frequency based on statistics. For example, well MW-101S
is recommended for annual sampling based on the qualitative evaluation, but for
semiannual sampling based on the summary evaluation due to the increasing TCE
trends identified in the temporal evaluation.
• Qualitative factor overrides statistics recommendations. For example, although
well MW-205S is recommended for exclusion or reduction based on the limited
value of its temporal trend information, the qualitative evaluation noted that the
concentrations are decreasing post December 2002, and thus override the high
variation no trend results driving the temporal evaluation recommendations that
considered a longer time period. Additionally, although wells MW-8D and MW-
8S are recommended for retention based on the spatial evaluation, they are
ultimately recommended for exclusion from the monitoring program because the
qualitative evaluation points out that these wells are outside of the area of interest
of the plume.
Table 7.2 presents a summary of the revised monitoring network as compared to the
basecase network (number shown in parentheses). For the Wash King groundwater
monitoring wells, the LTMO results indicate that a refined monitoring program
consisting of 35 existing wells sampled less frequently (11 wells sampled biennially, 16
sampled annually, and 8 sampled semiannually) and 10 primary and 2 optional new wells
(sampled annually) would be adequate to address the two primary objectives of
monitoring listed in Section 1. This refined network (including the two optional new
wells) would result in an average of 49.5 well-sampling events per year, compared to 50
per year under the current quarterly (EW) and semi-annual (MW) monitoring program.
A well-sampling event is defined as a single sampling of a single well. Implementing
these recommendations for optimizing the LTM monitoring program at Wash King
would increase the number of wells sampled from 30 to 47 (including the two optional
new wells), but entail essentially the same number of groundwater well-sampling events
per year. The costs of the basecase and refined monitoring programs would likely be
similar given the similar number of well-sampling events per year.
7-8
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
TABLE 7.2
SUMMARY OF REVISED AND BASECASE MONITORING PROGRAMS
LONG-TERM MONITORING OPTIMIZATION
WASH KING LAUNDRY SUPERFUND SITE
Well Type
Monitoring
Extraction
Total Wells
Monitoring Frequency3'
Exclude
14(17)
0(0)
14 (17)
Biennial
11(0)
0(0)
11(0)
Annual
28b/(0)
0(0)
28(0)
Semiannual
3(25)
5(0)
8(25)
Quarterly
0(0)
0(5)
0(5)
Total Sampling
Points
42 (25)
5(5)
47 (30)
37 Basecase sampling frequency corresponding to Table 2.1 shown in parentheses.
b/ Number includes 10 primary and 2 optional new wells.
7-9
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
SECTION 8
REFERENCES
American Society of Civil Engineers (ASCE) Task Committee on Geostatistical
Techniques in Hydrology. 1990a. Review of Geostatistics in Geohydrology - I.
Basic concepts. Journal of Hydraulic Engineering 116(5):612-632.
ASCE Task Committee on Geostatistical Techniques in Hydrology. 1990b. Review of
Geostatistics in Geohydrology - II. Applications. Journal of Hydraulic
Engineering 116(6): 63 3 -65 8.
Clark, I. 1987. Practical Geostatistics. Elsevier Applied Science, Inc., London.
Cohen, R.M. and J.W. Mercer. 1993. DNAPL Site Evaluation. CRC Press, Inc.
Environmental Systems Research Institute, Inc. 2001. ArcGIS Geostatistical Analyst
Extension to ArcGIS 8 Software. Redlands, CA.
Gibbons, R.D. 1994. Statistical Methods for Groundwater Monitoring. John Wiley &
Sons, Inc., New York, NY.
Michigan Department of Natural Resources (MDNR). 1980. Groundwater Quality
Investigation, Wash King Laundry, Baldwin, Michigan.
Rock, N.M.S. 1988. Numerical Geology. Springer-Verlag, New York, NY.
U.S. Army Corps of Engineers. 1997. Proactive Aspects of Applying Geostatistics at
Hazardous, Toxic, and Radioactive Waste Sites, Technical Lett No. 1110-1-175,
June 30.
U.S. Environmental Protection Agency (USEPA). 1993. EPA Superfund Record of
Decision: Wash King Laundry, EPA ID MID980701247, OU1, Pleasant Plains
TWP,MI. March 31.
USEPA. 1994. Methods for Monitoring Pump-and-Treat Performance. Office of
Research and Development. EPA/600/R-94/123.
Wiedemeier, T.H., and P.E. Haas. 2000. Designing Monitoring Programs to Effectively
Evaluate the Performance of Natural Attenuation. Air Force Center for
Environmental Excellence (AFCEE). August.
8-1
S:\ES\WP\PROJECTS\744461 - USAGE LTMO\2.doc
-------
APPENDIX A
COMMENTS AND RESPONSES ON DRAFT REPORT
-------
Response
Commen
'>
m
1
0
sa
to
pie
fe
ges
report.
d revi
ic
re
P
rive
eval
1=1 i
^ O 7, ^ O
o
C/3
I
P.,
-------
s
at
H
=
o
-
«
2
Q
at
•5 •£>
S3 ^
o Q
£
I
I
8 1
S3 ®
I £
3 «
^i ^^
^H ft
Hi1
Of • rt S3
Sfl
&P
S §,H
!l
H
I z
(N
•« -c
at
S
•§>
Cfi
at
o
(N
t-'
60
cc
5
II
-------
g
WD
O
-
g
c o
•
.2
o e
'5x.2
e vp
ft
O o
e
.§
Cfi
1
o
e-
2
mmen
e
o
QH -
o
CO ffi
II
ged to
be
••a a
-------
S
4)
H
Draft Long
o>
•5 •£>
S3 ^
o ^
ronmental Protection Agency Region V Review Comments
Monitoring Network Optimization Evaluation, Parsons, 1
(Continued)
• PN
^
Cfi
S
-s
+^
5
$
o
8-
S
Comment
a
_o
*-*^
o
OJ
es
a.
u
& £
S %
•3 *
I 0 %
3 ^—* J-H
tfi 5-H O
O <4H
rig
nee that it would b
tical profiling date
ng recommendati
§p g ^
g a 1
c> •S in
I concur with the data gaps identified and the recommendations for the new wells. I
am distressed by the apparent loss of locations for vertical profiling VOC data (the
GP-series). Depending on the location and results where the vertical profiling was
done, this data might be able to show some of the proposed wells unnecessary.
3 |
.° ^
co Q
r--
Is recommended well S-7 proposed with an annual sampling frequency as shown in
this table or is it a potential well as shown on Figure 4.1? There is an S-8 in table
4.3 ; it does not appear on Figure 4. 1 and text talks about up to seven new shallow
zone monitoring wells. Is there really an S-8?
rn **
c ^
• ° H
CO §
OS
T3 ^j T3 c/T +2 O
|^ | ^||| |^| |
|ll|l|||l|||l|l|
| g i § g S f| | ij^ ^ 1 -§, | -S
llNslll^^lf^l^
•a-S«>Bg-5x> .uiV^SS"1^
- M 1 s §1 ^ i ^^ u S ^S 2 ^. ^ tS
ai^Pl|.§||||lili|h
According to the report the algorithm for calculating Mann-Kendall assigned a value
of ND to non-detects rather than assuming and assigning a numerical value. While
the Mann-Kendal does not require a quantitative value, it does require a qualitative
ranking. How were the NDs used? Were all detection limits consistent within an
individual trend analysis? rf detection limits differed, how was this handled? Were
there any detections (usually qualified with a "j") below the detection limit? How
were these handled; were they deemed higher than a ND? hi other words, would a
value detected at less than the detection limit be treated as greater than the detection
limit in this algorithm?
(N
O
CO
00
1
o
% .§
3 ^ ^§
^sl^
u ^ 2 •? S ^s
^1 "H S a ^
1-9 « 1 -B £
1 j.f § fl
iifi
fe o « 3 * -^ *j
^ S|l ^§e i
S la's a^ I
a -a a ^ 1 1 -a
H a ^ x> o a 3
This is an admittedly picky comment. There is considerable information in this
table, however, it is not the easiest table to understand. The intensity of some of the
colors (blue and green particularly) obscures the printing (even if the words are made
redundant by the colors). The bolded boundaries are not as intuitively obvious as
one would like. Once I finally identified the bolded cells, they seemed obvious
thereafter, but on several separate occasions I studied the table without discerning the
differences.
oC£
c ^
• ° H
CO §
00
•n
1
^
£
"3
1
8
S3
1
I am not quite sure how the geostatistical analysis would best handle this issue, but
a
^o
CO
-------
s
0)
H
=
o
-
«
2
Q
o>
•5 •£>
ss ^
o £j
"S ^
S g
So
s2
SS ^5
S-3
^1
•l§ 1
III
PH N e
&13]
"!
g is
O o
ll
p
£ g>
13 'C
tS o
1
o
8-
S
& •-•
S §
e 5
o ^
!/5
-------
s
0)
H
=
o
-
«
2
Q
o>
•5 •£>
ss ^
o £j
^ |
c o
h
a°:
:*• sa
|-2
^ «
S|
i! l
Of • rt S3
Sfl
&l3j
"!
S "2
O o
1^
p
£ ^
•« -c
*- O
sl
S3 §
S3 5
o «
'>
VI
IS
'3
1
o
e-
S
° s g>-B «
g&^l^*
S
•§>
w
o
I
P.,
-------
fl
C
tZ
o>
H
6Jj
C
0
_
«
2
Q
o>
^ >T5
^ 0
§a
lity Review Comments
Evaluation, Parsons,
ea C
Environmental Qu
etwork Optimizatio
% ^
Department (
Monitoring ]
»v
C
«
.2f
^u
«M
o
55
s
o
'S
:s
Q
S
O
ction, Remediation and Redeveloj
^ ^ ^ £3 ^ i SH ^ *^H "S fn
05 f^ ' g .C >^ O U '— ^ ^ ^ ^_^ 1 s^*~ "^ Q
j3-S'§3S^^'C>'' "0 S H* § cP '^
23|53§c2|'Soi§|l'|'|-|
~3> SS"' ^^"^fe'" §S ^ S &
_2 D S ^ ^ *rf JJ5 o CH r^i S- ^ ^ f
rR1 "S5 53 (D 2 ^ B JB '^ ^H ^ S tS tS ^
§8 G? S 'S a"fe'SH^ § g;0^^ g.
s ^.s^Sw^^Ili'^ijsl;1!
1 •|'ii-c|'^^'|>l3S^p^5-^'o
S OS^§<''E3S>LoS ^ Q ^0 '.^ ft Q QJ _^ ^ ^ ^ -^ QH o^ Q ~£i ^ (L>
lil|l||l!li|l|!I!li^|f|it||'=|
iSf silili |!ii iff liyilil i-sin 1
iP!!|p.lii|!lI!l|.!|l!!tl!!iI|
S .y o fe^£"§ § «3 "g ^ ri a> fi-^^^t^ g3^ 22 ^ 9 -S ^ SS °
t/3 O W I^H . *^ W QH PH G r1^ ^3 -P ^3 (D G cd ^ "^ O . *^ ra &J) . ^ .S KH ^3 ^3 (D c^ F-H
^H (N
-------
Response
V
Commen
a
''
a
o
I
a.
It
oo
-------
Response
omment
u
a
o
£
!/5
&
c«
0.
u
report will be modified accordingly to
include these new recommendations.
Recent water level elevation data sets
provided to Parsons do not contain data
for all site wells (e.g., data for shallow
wells MW-8S and MW-202 are not
included). It would be helpful to prepare
potentiometric surface maps using water
level data for all site wells in order to
make them as accurate and useful as
possible.
2-^.S"
| 8 •3*|'S »|:S .
•S ^s^use a a 2
°£ g °-8 «| 2^|
11^^1,111
SMl-i-aSal!
lUi.S'glllg'
Agreed that the spatial analysis has
limitations based on the limitations on the
data and the well zone classifications and
would be a good idea to re-conduct after
more data is obtained. However, data
from 2003 was included only after review
to ensure that it was in-line with 2005
results, and the spatial evaluation results
are used only as one of three-tiers of the
lines of evidence. More value was given
to both the qualitative and temporal results
in the combined evaluation. Text will be
added to clarify the above comments in
the report.
+-. S to
*->X>^SfloSC-S^1 %
IlllSl^ltl
Ip^S^lsl
^t^li^i
|^ll5.glll«
1 &1 §«s £g^ii
IgSa^a-gall
w-HrtT3wT3^-£<->0
w 5£ •'-( '"S S"^ C3 w-*-1
l| § § i § % 1 < §
„ 3 '& & 0 ^ G '3 PH 3
£•§! §s § §!w. !
"jta c«3^3 a c^ S
e3u>cgQcgooH-;s-H
hJT3
-------
li
§ «
o w
'&we
02 c 2
^•s|
a « ^
« N e
s e o
a | u
*^ ,p ^_^
13 a
|0
1°
II
^Z
S «
«*M -P«
si
S3 •••
§ e
S 5
t: S
«
a
0>
Q
*v
«
.2f
u
«M
o
3
«
-*^
en
Response
1
S
0
U
C
_o
^
1
0.
JM
addressing the site-specific analytical
program.
0
81
tn U
O 43
|*|
, "^
4n' i-2
Q >
1 1
Q TO
§
-<->
•"^ CH QJ
y '^ ^
8 •§ >
13 o ^
CH y c/a
C^H ^ ^
•S 1 ^
H *-"
"y^ "^ ^
S cS
60 ™
g V ^
•S3 g In
(D^ ^3 ^
a
2
^H
O
e*
J
^
^
made in this report could be implemented
now, rather than waiting for data gaps to
be addressed. The LTM plan should be
dynamic and should be periodically re-
evaluated as new information is obtained.
Money saved by implementing the
recommended changes to the current
sampling program could be better spent
filling data gaps and performing a holistic
remedial process optimization evaluation
to ensure that the cleanup effort is
optimized.
Comment noted; we will contact you to
discuss the minor edits.
e
1 S
s!
-% *
rt
0 ^
5-H "^
o rt
1 &
B be
yj
!i
OS ^
^ o
ll
r, ^
^H
S '<„
» §
S 0
-^J _TO
~C^ CH
o
^ <->
•a u
^ 0 i
0 „:
5 a
N .2
•-§ a
!§
H ^
bo C
•S 1
0 u
3 ua
11
II
T3 bO
U .«
(D «
^ 1
42 &
W
|
•3
->
cy
^H
O
ua
Id
; minor
ter deta
on ™
9 a
^ W)
co S
ua •'"'
^ i-H
O g
'^ QJ
" i-H
t~~
8
T3
(N
a,
I
S
•S
J
w
o
<
£
I
P.,
&
&
------- |