Groundwater Monitoring
Network Optimization
Frontier Hard Chrome Superfund Site,
Vancouver, Washington
-------
Solid Waste and EPA 542-R-07-021
Emergency Response December 2007
(5203P) www.epa.gov
Groundwater Monitoring
Network Optimization
Frontier Hard Chrome Superfund Site,
Vancouver, Washington
-------
GROUNDWATER MONITORING NETWORK OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
EXECUTIVE SUM MARY
The following report reviews and provides recommendations for instituting a long-term
groundwater monitoring network for Frontier Hard Chrome Superfund Site in Vancouver,
Washington (FHC Site). The FHC Site consists of a former chrome plating facility in the
floodplain of the Colombia River. Shallow groundwater in the FHC area has been
impacted by residual hexavalent chromium from chrome-plating operations conducted
between 1958 and 1983. Affected groundwater migrated downgradient from the source
under the influence of industrial groundwater pumping south of the FHC site.
Extensive site remediation activities were completed at the FHC Site in 2003. The area
around FHC is currently undergoing rapid urban redevelopment to residential and
commercial property use. The primary goal of developing an optimized groundwater
monitoring strategy at the FHC Site is to create a dataset that fully supports site
management decisions while minimizing time and expense associated with collecting
and interpreting data. The long-term groundwater monitoring network for the FHC Site
should be designed to support site management decisions while accommodating on-
going redevelopment.
In the following report, the current FHC groundwater monitoring network has been
evaluated using a formal qualitative approach as well as statistical tools found in the
Monitoring and Remediation Optimization System software (MAROS).
Recommendations are made for groundwater sampling frequency and location based on
current hydrogeologic conditions and long-term monitoring (LTM) goals for the system.
The following report evaluates the monitoring system using analytical and hydrogeologic
data collected after installation of the remedy to the present, a time-frame between
October 2003 and June 2007. The following report outlines recommendations based on
a formal evaluation, but final determination of sampling locations and frequencies are to
be decided by the overseeing regulatory agencies.
Current Site Conditions
The broad area of shallow groundwater contamination associated with chrome plating
operations at FHC was discovered in the 1980's and investigated and delineated
through the 1990's. The Record of Decision (ROD) (USEPA, 2001) for groundwater at
FHC produced in 2001, detailed an in-situ chemical reduction of mobile hexavalent
chromium (Cr(VI)) as the final remedy. The regulatory screening level for total chromium
for the Site was determined to be 50 ug/L, based on the State of Washington
Department of Ecology Model Toxics Control Act (MTCA) Standard A value.
As a result of aggressive remedial treatments and cessation of industrial pumping, total
chromium concentrations across the site have dropped below the regulatory screening
level. It should be noted that for the past 3 years, total chromium levels in groundwater
at FHC have consistently been measured below the clean-up level of 50 ug/L. The FHC
groundwater plume, that is the extent of groundwater affected above the regulatory
Frontier Hard Chrome Site i Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
screening level, has largely disappeared. However, for the purpose of the following
analysis, the term "plume" is used to describe the historic extent of groundwater affected
by chromium originating from the FHC site. In this document, the term 'plume' describes
all chromium concentrations at any detectible level within the current FHC Site
groundwater monitoring network. Analytical results for total chromium were used in the
analysis of the groundwater network as a conservative surrogate for assessing the
concentration of soluble hexavalent chromium.
Site Groundwater Monitoring Goals and Objectives
Primary monitoring goals for the FHC Site groundwater include defining the extent and
magnitude of residual contamination and evaluating the efficacy of the chosen remedy.
The specific groundwater monitoring objective for FHC is to "ensure dilution and
dispersion of affected groundwater" until site groundwater meets state cleanup
standards (USEPA, 2001). Shallow groundwater in the FHC area is protected by
institutional controls prohibiting construction of water-supply wells in groundwater that
may be affected by industrial contaminants. Monitoring data will provide support for
institutional controls by delineating the extent of affected groundwater. Data from the
network will provide evidence of concentration stability and indicate if constituents begin
to remobilize. Analytical data collected from the network will document continued
efficacy of the remedy and attenuation of chemical constituents confirming that the
remedy is achieving site clean-up goals.
Project Goals and Objectives
The goal of the long-term monitoring optimization (LTMO) process is to review the
current groundwater monitoring program and provide recommendations for improving
the efficiency and accuracy of the network in supporting site monitoring objectives.
Specifically, the LTMO process provides information on the site characterization, stability
of constituent concentrations, sufficiency and redundancy of monitoring locations and
the appropriate frequency of network sampling. Tasks involved in the LTMO process
include:
• Evaluate well locations and screened intervals within the context of the
hydrogeologic regime to determine if the site is well characterized;
• Evaluate overall 'plume' stability through trend and moment analysis;
• Evaluate individual well concentration trends over time for target constituents of
concern (COCs);
• Develop sampling location recommendations based on an analysis of spatial
uncertainty;
• Develop sampling frequency recommendations based on qualitative and
quantitative statistical analysis results;
• Evaluate individual well analytical data for statistical sufficiency and identify
locations that have achieved clean-up goals.
The end product of the LTMO process at the FHC Site is a recommendation for specific
sampling locations and frequencies that best address site monitoring goals and
objectives while providing sufficient flexibility for site redevelopment.
Frontier Hard Chrome Site ii Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
Results
Statistical and qualitative evaluations of FHC Site analytical data have been conducted
and the following general conclusions have been drawn based on the results of these
analyses:
• After a qualitative evaluation of well locations, screened intervals and
hydrogeologic characteristics, affected groundwater at the FHC Site is delineated
to the relevant regulatory standards established for the site (Washington State
Department of Ecology MTCA A Standards, 50 ug/L for total chromium).
Groundwater areas where concentrations occasionally exceed regulatory
standards are bounded by wells where results are below the standard. No major
data gaps in site characterization were found.
• The historic area of affected groundwater evaluated shows overall stable to
decreasing concentration trends for total chromium. None of the well data
reviewed show increasing concentration trends. Many "no trend' findings result
from intermittent detections, data outliers or apparently cyclical variation in
concentrations, especially in Zone B wells.
• Moment trend analysis indicated that total dissolved mass measured within the
monitoring network is decreasing over time. The center of mass in Zone B is
retreating toward the source.
• Results from the spatial redundancy analysis indicate that several wells could be
removed from the program, as they do not provide unique information. Wells
identified as redundant are listed in Table 5.
• No areas of high concentration uncertainty were found; therefore no new
monitoring locations are recommended.
• The sampling frequency analysis recommended a reduced sampling frequency
for the majority of wells. Annual to biennial sampling frequencies were
recommended by the MAROS algorithm based on the rate of change and trend
of well concentrations.
• Many locations evaluated were statistically below the screening level for
chromium using both the student's T-test with a power analysis and the
sequential T-test. Approximately two-thirds of monitoring locations have achieved
the cleanup goals with 80% or greater statistical power, given the current
dataset.
Frontier Hard Chrome Site iii Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
Recommendations
The following general recommendations are made based on the findings summarized
above and those described in Section 4 below.
• Several areas of spatial redundancy were identified. 10 wells are recommended
for exclusion from the monitoring program.
• No new monitoring locations are recommended.
• Reduce the frequency of monitoring to annual sampling.
• Monitoring data show fairly high variance. In most cases, variance in the data
can be explained by site characteristics and geochemical processes. Continue
monitoring concentration trends for both total chromium and hexavalent
chromium and potentiometric water levels to determine how the hydraulic
influence of the Columbia River may be contributing to underlying variance in the
data.
• The majority of the analysis above was completed before several wells in the
network were damaged as a result of site redevelopment. Some wells may need
to be replaced or rehabilitated in order to achieve stated site monitoring
objectives. The recommendation that no new monitoring locations are needed
does not imply that monitoring wells damaged or destroyed during site
redevelopment do not need to be replaced. New wells may be required, but their
placement near 'old' locations identified as important is recommended.
• Continue development and updating of the comprehensive site database.
Results for both total chromium and hexavalent chromium concentrations should
be added to the database. Validated analytical data for all wells in the area
should be added to database within a reasonable time after sampling. Each well
should have a complete record of historic sampling events.
• Survey location coordinates and elevations for all wells. Share data with all
stakeholders. A common set of coordinates should be used by planners,
regulators, and construction and development companies.
Frontier Hard Chrome Site i Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
1.0 INTRODUCTION
The Frontier Hard Chrome Superfund Site (FCH Site) is a National Priorities Listed
(NPL) site administered under the Comprehensive Environmental Response,
Compensation and Liability Act (Superfund). The site is located in Vancouver,
Washington in Clark, County near the Columbia River (see Figure 1). The FHC site is
currently administered by the Washington Department of Ecology (Ecology) with support
from the US Environmental Protection Agency (EPA) Region 10. The original FHC
property is a 1/2-acre historic chrome-plating facility, built and operated between 1958
and 1983. The Site has traditionally been organized into soil and groundwater operable
units (OU). Only the groundwater OU will be considered in this report.
Groundwater monitoring plays a critical role in long-term restoration of the FHC Site. The
purpose of the following LTMO evaluation is to review the current groundwater
monitoring network and provide recommendations for improving the efficiency and
accuracy of the network for supporting site management decisions during and after site
redevelopment.
At the FHC Site, monitoring goals define why and how data collected from the site will be
used. The primary groundwater monitoring goal for the site is to "ensure dilution and
dispersion of affected groundwater", with monitoring to continue until "all remaining
groundwater meets state standards for groundwater cleanup" (USEPA, 2001).
Monitoring data from the site network are used to support institutional controls, by
identifying areas of affected groundwater and to document continued attenuation of site
constituents.
In order to recommend an optimized network that addresses the stated monitoring
objectives, spatial and analytical data from the site were analyzed using a series of
quantitative and qualitative tools. Tasks performed during LTMO analyses include:
• Evaluate well locations and screened intervals within the context of the
hydrogeologic regime to determine if the site is well characterized;
• Evaluate overall 'plume stability' through concentration trend and moment
analysis;
• Evaluate individual well concentration trends over time for target constituents of
concern (total chromium);
• Develop sampling location recommendations based on an analysis of spatial
uncertainty;
• Develop sampling frequency recommendations based on both qualitative and
quantitative statistical analysis results;
• Evaluate individual well analytical data for statistical sufficiency and identify
locations that have achieved clean-up goals.
A discussion of site background and regulatory context for the FHC Site is provided
below. Section 2 of the report details the analytical and statistical approach taken during
the LTMO evaluation. A detailed discussion of results is provided in Section 3. Summary
conclusions and recommendations are presented in Section 4.
Frontier Hard Chrome Site 1 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
1.1 Site Background and Regulatory History
The FHC Site is located in a former industrial area in the city of Vancouver in
southwestern Washington near the Columbia River. The site is located within the greater
Portland, Oregon/Vancouver, Washington metropolitan area. Because of Vancouver's
location along the Columbia River and proximity to the Pacific Ocean, the region has
historically been the home to several shipyards and supporting industrial activity.
As the regional economy has changed in recent years, the Vancouver shipyards have
been redeveloped into residential and commercial property to support rapid increases in
population. The area to the south of the FHC Site has been redeveloped, and the
industrial water supply wells that contributed to the spread of chromium-affected
groundwater to the southwest have been removed from service. The FHC Site is
scheduled for redevelopment into commercial properties in the near future.
The FHC Site is located in a floodplain, approximately one-half mile north of the
Columbia River. One-quarter mile north of the site, a steep rise in elevation marks an
area of residential land use. In the mid-1950's, much of the floodplain, including the FHC
Site, was filled with hydraulic dredge material and construction rubble. East of the FHC
Site, a topographic depression exists at the original level of the floodplain where the City
of Vancouver operates two groundwater well fields to provide public water supply. The
Pioneer Plating Company operated a chrome plating facility on the one-half acre FHC
site from 1958 through 1970. Chrome plating operations continued under Frontier Hard
Chrome management until 1983.
During much of its operational history, liquid wastes from chrome-plating operations
were discharged directly to the public sanitary sewer system. By 1975, the City of
Vancouver determined that chromium in wastewater was impacting the operation of its
secondary waste water treatment systems. FHC was directed to find an alternate
disposal method for liquid wastes. In 1976, FHC received a permit to discharge
untreated wastes to a drywell behind the facility. The permit included a schedule for the
installation of a treatment system for chromium-affected waste water; however, no
treatment systems were installed between 1976 and 1981.
By 1982, Ecology found FHC in violation of state waste disposal regulations. During the
same time period, chromium contamination was discovered in an industrial water supply
well southwest of the site, near the Columbia River. A broad area of shallow
groundwater contamination associated with chrome plating operations at FHC was
discovered. In December 1982, the FHC Site was proposed for inclusion on the NPL
under the CERCLA. In 1983, FHC closed all operations and the site was officially placed
on the NPL. Under a cooperative agreement with EPA, Ecology began the Remedial
Investigation and Feasibility Study (RI/FS) process. Records of Decision (ROD's) for the
site have been published in 1987 (for the soil OU) and 1988 (for the groundwater OU)
(USEPA, 1987 and 1988).
The 1987 ROD for soil called for excavation, stabilization and replacement of affected
soils with concentrations over 550 mg/Kg total chromium. Subsequently, the proposed
method of soil stabilization as a means of preventing leaching of chromium was found to
Frontier Hard Chrome Site 2 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
be ineffective. The 1988 ROD for groundwater recommended extraction and treatment
of groundwater from areas where concentrations of total chromium exceeded 50,000
ug/L. However, groundwater monitoring indicated that the area of affected groundwater
was shrinking after the downgradient industrial supply wells were removed from service.
The combination of changing site conditions and the development of new cost-effective
technologies motivated the EPA to reevaluate the proposed remedies for FHC.
An amended ROD was completed in 2001 (USEPA, 2001) detailing the final remedial
action planned for the site. The selected groundwater remedy included treatment of
mobile hexavalent chromium (Cr(VI)) through in-situ reduction to relatively insoluble
trivalent chromium (Cr(lll)). An In-situ Redox Manipulation (ISRM) technology was
chosen as the groundwater OU remedy (see Figure 1 for approximate location of
groundwater and soil ISRM treatment areas). An area downgradient from the source
was injected with reducing agents, resulting in the reduction of naturally occurring iron in
the subsurface. The area of reduced iron forms an in-situ permeable reactive barrier,
reducing soluble Cr(VI) in groundwater to Cr(lll). The purpose of the reactive barrier was
to 1) provide containment and prevent downgradient transport of affected groundwater,
2) reduce mass of Cr(VI) in high concentration areas; and 3) provide long-term
protection against future leaching of Cr(VI) (USEPA, 2001).
An ISRM technology was also chosen for the soil OU. The area of the former chrome-
plating tank and main building of FHC was treated with reducing agents, applied directly
to the soil. Aggressive treatment of the source area was anticipated to prevent further
Cr(VI) inputs to site groundwater.
Remedial activities for soil and groundwater were completed in September 2003.
Regular monitoring of site groundwater was included in the ROD to "ensure dilution and
dispersion of affected groundwater", with monitoring to continue until "all remaining
groundwater meets state standards for groundwater cleanup" (USEPA, 2001). The
groundwater cleanup standard for the FHC site has been established at 50 ug/L. Site
groundwater has been monitored quarterly between 2003 and 2007.
Analytical data for total dissolved chromium have been collected and used in the
following report, as this chemical analysis reflects concentrations of the more toxic and
soluble oxidation state of Cr(VI). Chromium solubility and mobility are strongly influenced
by redox reactions, chemical speciation, adsorption/desorption phenomena, and
precipitation/dissolution reactions. The reduced form of chromium (Cr(lll)) is significantly
less soluble in water than Cr(VI). Areas of the FHC site shallow subsurface have been
chemically treated with reducing agents, converting Cr(VI) to Cr(lll). Groundwater
samples at certain monitoring well locations are under low reducing conditions due to
the continued presence of reducing agents.
During the process of groundwater sampling some water samples may appear clear
(indicating Cr in the dissolved phase), and subsequently form a precipitate when
exposed to the atmosphere. When groundwater samples are removed from the
subsurface, Cr (III) compounds can precipitate as amorphous hydroxides. When sample
turbidity exceeds 10 Nephelometric Turbidity Units (NTUs), samples are filtered
removing the Cr(lll) species, but for samples with relatively low turbidity, the samples are
Frontier Hard Chrome Site 3 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
not filtered even though they may contain suspended Cr(lll). The data that are derived
after adjusting for the interfering precipitation are below clean-up standards for the site.
However, the redox changes introduced during sampling may introduce a higher level of
variance in samples collected in the region of the ISRM remedy.
1.2 Geology and Hydrogeology
The FHC Site is underlain by several geologic units, with the upper two being of interest
for this report. The top unit consists of hydraulic fill and construction debris used to
elevate the adjacent floodplain in the 1940's and1950's. Fill materials are largely silt and
sand and heterogeneous, poorly-compacted construction waste. Fill extends
approximately 12 to 20 feet below ground surface (ft bgs) across the site. The fill unit is
generally unsaturated, but localized areas of perched groundwater may be present.
(USEPA, 2001)
Underlying the fill is an alluvial unit, consisting of a clayey silt subunit and a sand-and-
ground unit. Groundwater in the alluvial unit is hydraulically connected to the Columbia
River. The clayey silt is heterogeneous in character and is 3 to 7 feet thick, thinning to
the north of the site. The clayey silt unit separates the lower sand-and-ground unit from
the fill. The sand-and-ground unit consists of poorly sorted sandy gravels, silty sandy
gravels and sandy silts with scattered large cobbles. Deposits in this unit resulted from
overbank deposition during flooding of the Columbia River and from channel deposition
that resulted in more particle sorting than the overbank deposits. The alluvial unit is
approximately 70 feet in thickness and is highly heterogeneous and anisotropic.
During initial site characterization, the alluvial unit was considered to have three layers.
Upper and lower permeable zones (Zones A and B) separated by an aquitard were
described in the RI/FS (issued in 1987). Zone A was described as a sand and gravel
layer beginning about 20 ft bgs and extending to about 35 ft bgs. A confining "lower
aquitard" below Zone A is described in the 1988 ROD (USEPA, 1988) and was the basis
for separating groundwater in the alluvial unit into A and B zones. Currently, this silt zone
is seen as semi-continuous fine-grained unit of dense sandy silt to silty sand. The layer
is now thought to be semi-confining and not a significant hydraulic barrier within the
alluvial aquifer.
Zone B, or the deeper alluvial unit, is also made up of sands and gravel, but with higher
permeability than Zone A. The lower alluvial unit extends from approximately 35 ft bgs
down to 80 to 100 ft bgs. Groundwater velocity in this zone is about 2.25 ft/d to the
south-southwest. There is no distinct vertical gradient between A and B Zones. Wells in
the FHC network are designated as either A or B Zone wells based on the depth of the
screened interval. During the LTMO analysis, the zone designations were used to
separate the data into two analysis groups to evaluate groundwater in zones based on
permeability. This is done with the understanding that Zones A and B are most likely
hydraulically connected.
Groundwater flow in the region of the FHC site is generally to the south/southwest as the
potentiometric surface data indicate a shallow slope to the south. Historically,
groundwater flow direction has been influenced by pumping at downgradient industrial
Frontier Hard Chrome Site 4 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
water supply wells, but when these wells were deactivated, groundwater flow returned to
a generally southerly flow direction. The average hydraulic gradient is 0.00015 ft/ft and
groundwater velocity is between 0.5 and 5 ft/d. Recharge to site groundwater occurs
from local infiltration of precipitation and from the recharge from another alluvial aquifer
north of the site near the topographic rise. Downgradient from the Site, groundwater
discharges to the Columbia River and area potentiometric surfaces are influenced by
Columbia River stage. Groundwater parameters used in the LTMO analysis are listed in
Table 2.
Frontier Hard Chrome Site 5 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
2.0 ANALYTICAL APPROACH
Evaluation of the groundwater monitoring network in the vicinity of the FHC Site
consisted of both quantitative and qualitative methods. A quantitative statistical
evaluation of the site was conducted using tools in the MAROS software. The qualitative
evaluation reviewed hydrogeologic conditions, well construction and placement. Both
quantitative statistical and qualitative evaluations were combined using a 'lines of
evidence' approach to recommend a final groundwater monitoring strategy to support
site monitoring objectives.
2.1 MAROS Method
The MAROS 2.2 software was used to evaluate the LTM network at the FHC Site.
MAROS is a collection of tools in one software package that is used in an explanatory,
non-linear but linked fashion to statistically evaluate groundwater monitoring programs.
The tool includes models, statistics, heuristic rules, and empirical relationships to assist
in optimizing a groundwater monitoring network system. Results generated from the
software tool can be used to develop lines of evidence, which, in combination with
professional judgment, can be used to inform regulatory decisions for safe and
economical long-term monitoring of affected groundwater. A summary description of
each tool used in the analysis is provided in Appendix A of this report. For a detailed
description of the structure of the software and further utilities, refer to the MAROS 2.2
Manual (AFCEE, 2003; http://www.gsi-net.com/software/MAROS V2 2Manual.pdf) and
Azizetal., 2003.
In MAROS 2.2, two levels of analysis are used for optimizing long-term monitoring plans:
1) an overview statistical evaluation based on temporal trend analysis resulting in 'plume
stability' information; and 2) a more detailed statistical optimization based on spatial and
temporal redundancy reduction methods (see Appendix A or the MAROS Users Manual
(AFCEE, 2003)).
2.1.1 COC Choice
MAROS includes a short module that provides recommendations on prioritizing COCs
for the entire network based on toxicity, prevalence, and mobility of the compounds
dissolved in groundwater. However, the priority constituent at the FHC site is total
dissolved chromium, analyzed as a surrogate for Cr(VI). Volatile organic compounds
(VOCs) are present in small amounts in site groundwater from off-site sources, but these
compounds are not risk-drivers for site management. The COC choice module was not
used for the FHC site.
2.1.2 Plume Stability
Within MAROS, time-series concentration data are analyzed to develop a conclusion
about 'plume stability'. For the MAROS analysis, a plume is defined as the extent of
groundwater within the monitoring network affected by any concentration of the target
contaminant over time. Practically, the 'plume' area is defined as the maximum extent of
affected groundwater over the time-frame of the investigation. The definition of 'plume'
Frontier Hard Chrome Site 6 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
used in this document is distinct from the regulatory definition in that concentrations do
not need to exceed the regulatory screening limit in order to be considered part of the
'plume'.
For the purpose of this analysis, a groundwater plume is said to be stable when
constituent concentrations at individual monitoring locations as well as moments
estimated from the entire network are not changing rapidly. If a plume is found to be
stable, in many cases, the number of locations and monitoring frequency can be
reduced without loss of information.
Individual well concentrations are evaluated using both Mann-Kendall and Linear
Regression trend tools. The Mann-Kendall nonparametric evaluation is considered one
of the best methods to evaluate concentration trend as it does not assume the data fit a
particular distribution (Gilbert, 1987). Individual well concentration trends were
calculated for chromium for the time period 2003 to 2007. Individual well Mann-Kendall
trends were also used in the sampling frequency analysis, where trends determined for
the 2006 to 2007 interval were compared with trends calculated using the entire dataset
for each well. During the final 'lines of evidence' evaluation, individual well concentration
trends are considered along with summary statistics such as percent detection and
historic maximum concentration to make recommendations for the final sampling
network.
Moment analysis algorithms in MAROS are simple approximations of complex
calculations and are meant to estimate the total dissolved mass (zeroth moment), center
of mass (first moment) and spread of mass (second moment) within the monitoring
network and the trend for each of these estimates over time. Trends for the first moment
indicate the relative amount of mass upgradient vs. downgradient and the change in the
distance of the center of mass from the source over time. Trends in the second moment
indicate relative dispersivity by evaluating the spread of mass about the center of mass
over time.
2.1.3 Well Redundancy and Sufficiency
Spatial analysis modules in MAROS recommend elimination of sampling locations that
have little impact on the historical characterization of contaminant concentrations while
identifying areas within the monitoring network where additional data are needed. For
details on the redundancy and sufficiency analyses, see Appendix A or the MAROS
Users Manual (AFCEE, 2003).
Sample locations are evaluated in MAROS for their importance in providing information
to define concentrations within the area of affected groundwater. Wells identified as
providing information redundant with surrounding wells are recommended for elimination
from the program. (Note: 'elimination' from the program does not necessarily mean
plugging and abandoning the well. See Section 2.3 below.)
Well sufficiency is evaluated in MAROS using the same spatial analysis method as that
for redundancy. Areas identified as having unacceptably high levels of concentration
uncertainty are recommended for additional monitoring locations.
Frontier Hard Chrome Site ^ Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
The well redundancy and sufficiency analyses use the Delaunay method and are
designed to select the minimum number of sampling locations based on the relative
importance of information supplied at each sampling location in the monitoring network.
The importance of each sampling location is assessed by calculating a slope factor (SF)
and concentration and area ratios (CR and AR respectively). Sampling locations with a
high SF provide unique information and are retained in the network. Locations with low
SF are considered for removal. Areas ringed by wells with high SF's may be candidates
for new well locations. SF's were calculated for all wells at the FHC Site and the results
were used to determine the importance of each well in the network for defining
chromium concentrations.
The results from the Delaunay method and the method for determining new sampling
locations are derived solely from the spatial configuration of the monitoring network and
the spatial pattern of the contaminant plume based on a two-dimensional assumption.
No parameters such as the hydrogeologic conditions are considered in the analysis.
Therefore, professional judgment and regulatory considerations must be used to confirm
final decisions.
2.1.4 Sampling Frequency
MAROS uses a Modified Cost Effective Sampling (MCES) method to optimize sampling
frequency for each location based on the magnitude, direction, and uncertainty of its
concentration trends. The MCES method is based on the Cost Effective Sampling (CES)
method developed by Ridley et al. (1995). The MCES method estimates a conservative
lowest-frequency sampling schedule for a given groundwater monitoring location that still
provides needed information for regulatory and remedial decision-making.
The MAROS algorithm recommended a preliminary location sampling frequency (PLSF)
for each monitoring location at the FHC Site based on a combination of recent (2006-
2007) and long-term (2003-2007) trends and the magnitude and rate of concentration
change. The PLSF has been reviewed qualitatively and a final optimal sampling
frequency has been recommended consistent with monitoring objectives and regulatory
requirements.
2.1.5 Data Sufficiency
The MAROS Data Sufficiency module employs simple statistical methods to evaluate
whether analytical data are adequate both in quantity and in quality for revealing
changes in constituent concentrations. Statistical tests for the MAROS module were
taken from the USEPA Methods for Evaluating the Attainment of Cleanup Standards
Volume 2: Groundwater statistical guidance document (USEPA, 1992).
Two types of statistical analyses have been performed on analytical samples from each
individual well. First, hypothesis testing using a Sequential T-test has been performed to
determine if groundwater concentration is statistically below the screening level for total
chromium (screening levels were set to Ecology MTCA Standard A of 50 ug/L). The
Sequential T-test indicates if the well has a sufficient number of samples at low enough
Frontier Hard Chrome Site 8 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
concentrations to be categorized as having "attained" groundwater cleanup goals
(confidently below the screening level concentration). The statistical standard set by the
Sequential T-test is quite high, and if the well data indicate the groundwater
concentration has 'attained' cleanup, then there is high confidence that the groundwater
is statistically below the regulatory limit. If measured concentrations are high or there are
an insufficient number of data points, then the well is recommended for further sampling.
A Student's T-test followed by statistical power analysis was also performed in the Data
Sufficiency module to assess the reliability of the hypothesis test and to suggest the
number of additional samples that may be required to reach statistical significance. The
power analysis uses the number of samples (n), the variance of the samples, the
minimum detectible difference and the significance (a) of the test to determine if the well
is below the screening level with very high confidence. The power analysis provides a
higher level of certainty that the well is not affected above risk-based levels. Locations
that pass the power test are considered "statistically clean".
The Data Sufficiency module is designed to evaluate 6 years of sampling data. While
quarterly sampling for the past 3 years has provided a sufficient number of events to
evaluate the data using most techniques, 3 more years of sampling is necessary before
wells at the FHC Site can be confidently evaluated using this module. The analysis was
conducted with the current dataset and results are reported, but the results should be
considered preliminary, at this point.
At the FHC Site, locations that monitor groundwater areas "statistically below screening
levels" or "statistically clean" may be considered for reduced sampling frequency or
elimination from the program. Statistically 'clean' ring locations can be retained in the
program to help bound the areas of affected groundwater, set institutional control
boundaries or function as surrogate point of compliance locations.
2.2 Data Input, Consolidation and Site Assumptions
Groundwater analytical data from the FHC Site were supplied by Region 10 EPA and
from the Frontier Hard Chrome Event 11 Long-Term Monitoring Report (Weston, 2007).
Site data were supplemented with information from historic site reports including the
RODs. Groundwater monitoring locations included in the evaluation are listed in Table
1, with additional aquifer and site details provided in Table 2.
Chemical analytical data collected between October 2003 and June 2007 and well
information data were organized in a database, from which summary statistics were
calculated. In all, 33 sample locations were considered in the network evaluation for the
FHC Site. Wells are described in Table 1, and well locations are illustrated on Figure 1.
Groundwater monitoring data collected prior to 2003 are available for a subset of FHC
wells; however, the installation of the remedy changed the nature and distribution of
dissolved constituents as well as groundwater geochemistry. Therefore, data collected
before 2003 are not comparable with those collected after installation of the remedy. In
order to provide reasonable consistency in statistical comparisons, analyses have been
limited to the 2003 - 2007 time-frame. Individual well trend evaluations were performed
Frontier Hard Chrome Site 9 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
for data collected between 2003 and 2007. The data represent a roughly 3 year record
for many wells, and provide an indication of long-term, post-remediation trends in site
constituent concentrations. Spatial analyses and recent sampling frequency analyses
were conducted for data collected 2006-2007. Duplicate samples in the dataset were
averaged to develop one analytical result for each quarter. No other data consolidation
was performed.
It should be noted that only total chromium concentrations in groundwater were used for
this evaluation. Analytical data for total chromium concentrations were collected as a
conservative surrogate for Cr(VI), which is the soluble form of the metal. Groundwater
samples at certain monitoring well locations were under low reducing conditions in the
subsurface due to the injection of the reductant. During the process of groundwater
sampling, the clear water samples would form a precipitate when exposed to the
atmosphere and filtering of the sample was necessary because turbidity was greater the
10 NTUs. Use of total chromium results should be considered conservative as the
method will over-predict soluble chromium concentrations by including residual
suspended Cr(lll) in the result. Using total chromium analysis at all sites should improve
consistency in evaluating groundwater under a variety of subsurface redox conditions.
2.3 Qualitative Evaluation
Multiple factors should be considered in developing recommendations for monitoring at
sites undergoing long-term groundwater restoration. The LTMO process for the FHC Site
includes developing a 'lines of evidence' approach, combining statistical analyses with
qualitative review to recommend an improved monitoring network. Results from the
statistical analyses in combination with a qualitative review were used to determine
continuation or cessation of monitoring at each well location along with a proposed
frequency of monitoring for those locations retained in the network.
The primary consideration in developing any monitoring network is to ensure that
information collected efficiently supports site management decisions. Site information
needs are reflected in the monitoring objectives for the network. For this reason, any
proposed changes to the network are reviewed to be consistent with and supportive of
the stated monitoring objectives. The qualitative review process begins with evaluating
each monitoring location for the role it plays supporting site monitoring objectives. For
example, a location may provide vertical or horizontal delineation of affected
groundwater or may provide information on decay rates in the source area. Each well in
the FHC Site network was evaluated for its contribution to site monitoring objectives.
Qualitatively, redundant locations are those where multiple wells address the same
monitoring objective in approximately the same location.
A recommendation to eliminate chemical analytical monitoring at a particular location
based on the data reviewed does not necessarily constitute a recommendation to
physically abandon (plug) 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. In some cases, stakeholders may pursue a comprehensive
monitoring event for all historic wells every five to ten years to provide a broad view of
plume changes over time. In general, continuation of water level measurements in all
Frontier Hard Chrome Site 10 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
site wells is recommended. Data on hydraulic gradients and potentiometric surfaces are
often relatively inexpensive to collect and can be used to support model development
and resource planning. However, when site redevelopment is an issue, optimization of
the network can be used to identify redundant locations that can be plugged without loss
of information.
Qualitative evaluation for sampling frequency recommendations includes consideration
of factors such as the rate of change of concentrations, the groundwater flow velocity,
and the type and frequency of decisions that must be made about the site. Additionally,
consideration is given to the concentration at a particular location relative to the
regulatory screening level, the length of the monitoring history and the location relative to
potential receptors.
A summary of the lines of evidence used to develop a final monitoring network
recommendation is presented below.
Key Point: Several lines of evidence were used to develop recommendations for the monitoring
network.
Lines of Evidence
• Individual well trend
• Plume-Wide Trends
• Well Redundancy and Sufficiency
• Sampling Frequency
• Data Sufficiency
• Qualitative Evaluation
Method
• Mann-Kendall (Linear regression)
• Moment Analysis: Total dissolved mass,
center of mass and distribution of mass
trends.
• Delaunay triangulation and slope factor
calculation, along with area ratios and
concentration ratios.
• Modified Cost Effective Sampling
• Sequential T-Test, Student's T-Test and
Power Analysis
• Hydrogeologic factors, monitoring
objectives, stakeholder concerns and all
statistical results to develop final
recommendation.
Frontier Hard Chrome Site
Vancouver, Washington
11
Groundwater Monitoring
Network Optimization
-------
3.0 SITE RESULTS
Data from 33 monitoring wells at depths corresponding to Zones A and B were included
in the quantitative network analysis for the FHC Site. Summary statistics for the wells
(including percent detections and maximum concentrations) are shown on Table 3.
Qualitative considerations are discussed alongside statistical interpretations below.
3.1 Plume Stability
3.1.1 Concentration Trends
Individual well chromium concentration trends using the Mann-Kendall method are
summarized in the table below. Trends were evaluated for data collected between 2003
and 2007. Detailed results of the trend evaluations performed are summarized on Table
3. Results of the individual well Mann-Kendall trends are also illustrated on Figure 2 and
Figures 7 and 8. Detailed Mann-Kendall reports for each well in the network are located
in Appendix B.
Alluvial
Aquifer Zone
Zone A
ZoneB
All Wells
Total
Wells
16
17
33
Number and Percentage of Wells for Each Trend Category
Non Detect
0
0
0
PD, D
5(31%)
7(41%)
12 (36%)
S
7 (44%)
2(12%)
9 (27%)
I, PI
0
0
0
No Trend
4 (25%)
8 (47%)
12(36%)
Note: Number and percentage of total wells in each category shown. Decreasing trend (D), Probably Decreasing trend
(PD), Stable (S), Probably Increasing trend (PI), and Increasing trend (I).
All wells had sufficient analytical data to evaluate trends. Because chromium is present
naturally at low levels in the aquifer, all wells groundwater analyzed showed detectable
quantities. None of the sampling locations showed increasing or probably increasing
trends for total chromium. The site cleanup standard for chromium is 50 ug/L. Overall,
two-thirds of well datasets showed stable to decreasing concentration trends.
Several wells with historic high concentrations near the ISRM zone indicate no trend or
high variance in the data. Overall most of the measured concentrations at these
locations are quite low, but occasional spikes in concentration are seen (see MW-12A
and MW-15A and B in Appendix B). Data variability may be a result of artifacts due to
changing redox environment during sampling and subsequent filtering of samples (see
discussion above). Many of these wells monitor former "hot spots" or areas with possibly
high residual chromium concentrations. Greater variability in the analytical data was
found in data from Zone B wells, as indicated by the relatively large number of No Trend
(NT) results.
3.1.1.2 Moments
Moment analysis was used to estimate the total dissolved mass (Zeroth Moment), center
of mass (First Moment) and distribution of mass (Second Moment) for total chromium in
Frontier Hard Chrome Site
Vancouver, Washington
12
Groundwater Monitoring
Network Optimization
-------
the Zones A and B. The values were determined using the current well configuration.
The Mann-Kendall trends of the moments were determined for data between 2004 and
2007 (4th quarter 2003 data did not include all sampling locations). Estimates of the
zeroth and first moments for the FHC Site are shown in Table 4. Moment trends are
summarized in the table below, and first moments over time are illustrated on Figure 2.
Total mass values are rough estimates of mass in the dissolved phase, assuming a
constant porosity and uniform saturated thickness across the site. The mass estimates
are best interpreted as metrics for determining the trend of dissolved mass within the
network. For both A and B zone groundwater, mass estimates decreased strongly
between 2004 and 2007. Mass estimates are greater for Zone B as the saturated
thickness is greater.
First moments, indicating the trend in center of mass, show No Trend in Zone A.
Concentrations measured in Zone A wells are low, and minor fluctuations in
concentration are seen in concentration vs. time graphs (Appendix B). The alluvial
aquifer is influenced by stages of the Columbia River, and the fluctuations in both first
moments and concentrations in Zone A may result from hydraulic influence of the river.
First moments for Zone B indicate that the center of mass is regressing toward the
source, indicating decreasing concentrations in the tail area relative to the source.
Moment
Type
Zeroth
First
Second
Moment Analysis
Zone A
Decreasing'. Total mass of chromium
showed a strongly decreasing trend
2004-2007
No Trend. The movement of center of
mass in Zone A shows no trend over
time.
No Trend in both X and Y directions
Comment
Zone B
Decreasing'. Total mass of chromium
showed a strongly decreasing trend 2004-
2007.
Decreasing'. The center of mass in Zone B
is moving closer to the source, supporting
the conclusion of a shrinking plume.
Probably Increasing in the X direction
(direction of groundwater flow).
Increasing in the Y direction (orthogonal to
groundwater flow).
Second moments indicate the pattern of dilution and dispersion of mass as it moves
from the center of the network to the edges. No clear trend in second moments was
found for Zone A. Zone B second moments indicate relatively more mass is moving to
the edges relative to the center. Increasing second moments support the conclusion that
total chromium in Zone B is dispersing in both the X (direction of groundwater flow) and
Y (orthogonal to groundwater flow) directions.
3.2 Well Redundancy and Sufficiency
The spatial redundancy analysis was performed using data collected between 2006 and
2007. Spatial redundancy results include slope factor (SF) and area (AR) and
concentration ratio (CR) calculations to rank the importance of the well in the network.
Summary results for the redundancy analysis as well as a summary of the data
Frontier Hard Chrome Site
Vancouver, Washington
13
Groundwater Monitoring
Network Optimization
-------
sufficiency results (identifying wells where groundwater is statistically below the
screening level) are presented on Table 5.
Of the 16 wells screened in the Zone A depth range, eight were identified as possibly
redundant based on SF and AR and CR results. Interior well locations with SF below 0.3
were considered for removal, while hull wells were retained if SF was above 0.1. The
average SF for each well is shown on Table 5. Wells identified by the MAROS algorithm
as redundant include B85-4, MW-14A, MW-15A, MW-16A, W85-6A, W85-7A, W-92-16A
and W98-20A.
Seventeen Zone B locations were evaluated and six were identified as redundant based
on the criteria described above. Wells B87-8, MW-11B, MW-12B and C, MW-13C and
W85-6B were identified as not providing unique information to characterize the affected
or potentially affected groundwater.
The decision to remove a well from routine monitoring is based on a combination of
spatial statistical analyses and qualitative review of the function of the well in supporting
site monitoring objectives. The spatial statistics for Zone A and B wells were considered
along with other lines of evidence including whether the well monitors groundwater
below the screening level, trend results, detection frequencies and overall monitoring
objectives before a final recommendation was made.
In the case of FHC, location of wells should be compatible with site redevelopment while
still meeting the objectives of the program. Proposed plans for site redevelopment were
received from stakeholders, and well locations were reviewed to try to accommodate
proposed development (see Figures 7 and 8). In the case of nested well locations
(locations where multiple wells monitor several depths), if one well was very important
for monitoring one depth profile the other well is recommended for retention in the
program as well. Final recommendations for wells to retain in the monitoring program are
summarized below and shown on Table 5.
Wells
Retained
Wells
Excluded
Final Network Recommendation
Zone A
685-4, MW-16A, W85-6A, W92-16A
MW-12A, MW-17A, W97-18A,
MW-15A, W97-19A, W98-21A,
W99-5A,
MW-11A, MW-13A, MW-14A, W85-7A,
W98-20A
Zone B
685-3, 687-8, MW-12B, MW-12C, MW-15B
MW-16B, W85-6B, W92-16B, W97-18B,
W97-19B, W98-21B, W99-R5B
MW-11B, MW-13B, MW-13C, MW-14B,
W85-7B
The graphical well sufficiency analyses for Zones A and B are illustrated in Figures 3
and 4, respectively. MAROS uses the Delaunay triangulation and SF calculations to
identify areas with high concentration uncertainties. Graphical results illustrate polygons
created by the triangulation method and indicate areas of high uncertainty with an "L" or
an "E" in the center of the triangle. For both Zones A and B, no areas of high
Frontier Hard Chrome Site
Vancouver, Washington
14
Groundwater Monitoring
Network Optimization
-------
concentration uncertainty were found; all areas show an "S" (for small uncertainty) or an
"M" (medium). Overall, the networks have very low spatial uncertainty. Some areas of
moderate spatial uncertainty were identified near the source "hot spots", but these areas
do not require additional well locations. No new well locations are recommended for the
monitoring networks.
Site data excluding the wells recommended for elimination were re-run in the MAROS
data sufficiency module to determine if eliminating wells from the program would
increase concentration uncertainty. Figures 5 and 6 illustrate the concentration
uncertainty found after elimination of redundant locations in Zones A and B. No increase
in statistical uncertainty was found when wells listed above were eliminated, supporting
the redundancy of locations recommended for exclusion from the program.
3.3 Sampling Frequency
Table 6 summarizes the results of the MAROS preliminary sampling frequency
recommendation. The MCES method evaluates overall and recent temporal trends, and
recommends an optimized sampling frequency based on the rate of concentration
change. As with the redundancy analysis, a qualitative review of all data is conducted
before recommending a final sampling frequency.
The rate of change of chromium concentrations for FHC wells is very low. The majority
of well locations have decreasing to stable concentration trends for the period analyzed.
For the recent data, many wells show stable trends, indicating that the rate of
concentration reduction at most locations has slowed. Many wells show some fluctuation
in the data that may be consistent with hydraulic influence from the Columbia River or
redox conditions during sampling.
The majority of wells in both Zones A and B have preliminary recommendations for
annual to biennial (every two years) sampling. The current sampling frequency is
quarterly. Quarterly monitoring has already provided a statistically significant dataset
(sufficient number of sample points to perform statistical analyses). After a qualitative
review, annual sampling frequency is recommended for all wells remaining in the
network during long-term groundwater monitoring. Annual sampling is consistent with
the very low rate of change seen over the past 3 years, and relatively low groundwater
flow velocities and limited number of site management decisions to be made.
One well, MW-12A, had a PLSF recommendation for quarterly sampling, based on the
'no trend' concentration trend result and the presence of one outlying sample result.
With the exception of one possible data outlier, the well shows a fairly low overall rate of
change, so the MW-12A is also recommended for annual sampling.
The table below summarizes the current monitoring frequency and the recommended
sampling frequency after the lines of evidence evaluation.
Frontier Hard Chrome Site 15 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
Monitoring Wells
Total Samples (average
per year)
Total Wells
Well Sampling frequency Analysis Source OU
Sampling
Frequency
Quarterly
Semi-annual
Annual
Biennial
Current Sampling
Frequency
33
0
0
0
99
33
Sampling Frequency
Recommendation
0
0
23
0
23
23
3.4 Data Sufficiency
The Data Sufficiency module was used to identify wells monitoring groundwater that has
statistically achieved site cleanup goals with >80% statistical power and those that have
attained cleanup using the Sequential T-test method (even more stringent). Statistical
power increases with the number of samples taken and with reduction in both the
concentration and detection limits for the dataset. For the FHC data set, the data were
assumed to be log-normally distributed and the statistics were performed using this
assumption. The groundwater cleanup goal for the FHC Site is 50 ug/L and the majority
of detection limits are 0.5 ug/L for most samples.
The Data Sufficiency tools are normally run on datasets with greater than 6 years of
data, but quarterly data for the past 3 years provides enough data to perform a
preliminary analysis. Preliminary results for all sampling locations are reported in Table
5. Achievement of "clean" status was considered along with other lines of evidence when
considering elimination of wells from the program and for reduction in sample frequency.
Data sufficiency should be revisited when 3 more years of data have been collected.
Results of the data sufficiency indicate that the majority of wells in the network are at or
approaching cleanup goals and have a sufficient number of sample events to provide
confidence in the statistical outcome (although the number of sample years since source
remediation is insufficient).
The data support the conclusion that the ISRM groundwater treatment in combination
with removal of the industrial supply wells has reduced site-wide concentrations. The
groundwater network indicates groundwater is approaching and may have achieved
cleanup goals and that a reduction in monitoring effort may be appropriate at this time.
The table below summarizes the results of the Data Sufficiency analysis. Identification of
specific wells that have achieved cleanup can be found on Table 5.
Frontier Hard Chrome Site
Vancouver, Washington
16
Groundwater Monitoring
Network Optimization
-------
Groundwater
Zone
A
B
Total
Total Wells
16
17
33
Data Sufficiency Results
Wells Statistically
Below MTCA with >80%
Power
15(94%)
12(71%)
27 (82%)
Wells Statistically "Attained"
Clean-up Goals
4 (25%)
1 (5%)
5(15%)
MTCA = Washington State Model Toxics Control Act Standard A = 50|jg/L
Frontier Hard Chrome Site
Vancouver, Washington
17
Groundwater Monitoring
Network Optimization
-------
4.0 CONCLUSIONS AND RECOMMENDATIONS
4.1 General Conclusions
The primary goal of developing an optimized monitoring strategy at the FHC Site is to
create a dataset that fully supports site management decisions while minimizing time
and expense associated with collecting and interpreting the data. A summary of the final
recommended monitoring network is presented in Table 7. The recommended network
reduces monitoring effort and cost by reducing both the frequency of groundwater
sampling and the number of locations sampled.
Tasks identified in the Section 1 were performed for each of the groundwater zones. A
summary of general results and recommendations resulting from each task is presented
below:
• Evaluate well locations and screened intervals within the context of the
hydrogeologic regime to determine if the site is well characterized.
Result: Part of the network optimization process is to identify possible gaps in site
characterization that may require additional sampling locations or site investigation.
Based on well locations, screened intervals and hydrogeologic characteristics,
affected groundwater at the FHC Site is delineated to the specified screening
levels (MTCAs Standard A, 50ug/L). Groundwater areas where concentrations
historically exceed screening levels (hot spots) are bounded by wells where results
are below MTCAs. Monitoring locations in the tail of the network have average
concentrations below the screening levels for both Zone A and Zone B. A "hot
spot" was identified in Zone A near location MW-12A, while the "hot spot" in Zone
B is shifted to the south near wells MW-15B and B87-8. All wells in the network
have a sufficiently large data set to perform statistical calculations. No major data
gaps were identified during the qualitative evaluation.
Recommendation: LTMO is appropriate for the site at this time. No additional
fundamental site investigation is recommended at this time. In order to comply with
stated monitoring objectives, future groundwater monitoring should include historic
"hot spot" wells as well as regulatory compliance points.
• Evaluate overall plume stability through trend and moment analysis.
Result: Total chromium concentrations evaluated are largely decreasing to stable,
even though some concentration trends (for both individual wells and plume
moments) show no trend. Many 'no trend' findings result from occasional outliers in
the dataset (see MW-14B) or from wells where the concentration fluctuates at very
low to non-detect concentrations (see W97-18B, W97-19B). Another source of
data variance includes possible influence of Columbia River stages on the aquifer
and conservative sampling artifacts resulting from monitoring total chromium from
a highly reduced geochemical regime. Overall, total dissolved mass estimates
(zeroth moment) within the monitoring networks are strongly decreasing. Center of
Frontier Hard Chrome Site 1 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
mass estimates show some variation, but are consistent with shrinking extent of
affected groundwater. The distribution of mass within the Zone B network indicates
that dilution and dispersion of dissolved chromium is ongoing.
Recommendation: Reduced monitoring effort is appropriate for stable or shrinking
plumes. Monitoring frequency can be reduced where groundwater concentrations
are not changing rapidly. After an initial steep drop in concentrations (2003-2004),
groundwater concentrations are not changing rapidly at the FHC Site and
concentrations within the network appear to have stabilized at a low level, largely
below the 50ug/L screening level. This finding is consistent with reduced
monitoring effort.
• Evaluate individual well concentration trends over time for target constituents of
concern (total chromium);
Result: For 33 wells evaluated at the FHC Site, approximately two-thirds of
locations showed stable to decreasing concentration trends (63%). No increasing
or probably increasing trends were calculated. No statistically significant trend was
found at roughly one-third of locations.
Recommendation: Individual well trend evaluations at the FHC Site provide
support for the conclusion that total chromium concentrations are largely stable.
Monitoring frequency can be reduced for locations where concentrations are not
changing rapidly or are decreasing below screening levels. Some variation in
concentrations is seen at "hot spot" locations, where occasional spikes in
concentration have been recorded. "Hot spot" locations should be monitored
periodically to develop a longer-term dataset (>6 years).
In the future, both dissolved and total chromium analytical data should be collected
at the appropriate locations. The dissolved chromium concentrations in
groundwater should be used instead of or along side total chromium for the
evaluations in order to reduce variance in the data introduced through sampling
artifacts and variable redox conditions. The appropriate locations to use the
dissolved data in the evaluations include monitoring well locations where the
groundwater samples have greater then 10 NTUs turbidity readings.
Wells in the tail area of the network (south and west of W85-6A/B) are largely
stable to decreasing with very low concentrations; these locations should be
monitored in the future as delineation or compliance points to confirm the absence
of affected groundwater in this area.
• Develop sampling location recommendations based on an analysis of spatial
uncertainty;
Result: The spatial redundancy analysis indicated that several wells could be
removed from the routine monitoring program, as they do not provide unique
information.
Frontier Hard Chrome Site 2 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
The spatial analysis did not identify any areas of high spatial uncertainty.
Recommendation: 10 wells are recommended for exclusion from routine
monitoring. The wells include locations near the remedy and some downgradient
locations. A spatial analysis was conducted for the reduced network and no
increase in spatial concentration uncertainty was found for data between 2006 and
2007. The resulting network of 23 locations should provide adequate information to
monitor "dilution and dispersion" of dissolved chromium until all areas achieve
cleanup goals with statistical confidence.
No new monitoring locations are recommended.
Develop sampling frequency recommendations based on both qualitative and
quantitative statistical analysis results;
Result: The sampling frequency analysis recommended a dramatically reduced
sampling frequency for the majority of wells. Annual to biennial sampling frequencies
were recommended by the algorithm based on the rate of change and trend of well
concentrations.
Recommendation: Reduce the frequency of monitoring. An annual sampling
frequency was recommended for future monitoring. While quarterly sampling has
been effective to characterize the success of the remedy, long-term data over a
period of years are required to achieve the stated monitoring objectives. These long-
term objectives are not achieved by frequent sampling events, but rather by sampling
a consistent set of wells at a frequency comparable to the rate of change of
concentrations. The recommendation is to collect annual data for approximately six
more years, and re-evaluate the plume for statistical attainment of site cleanup
objectives.
Evaluate individual well analytical data for statistical sufficiency and identify locations
that have achieved clean-up goals.
Result: 82% of wells are statistically below cleanup standards with greater than 80%
power. 15% of locations have achieved cleanup using the Sequential T-test, a very
rigorous statistical test.
Recommendation: Data sufficiency should be revisited when 3 more years of data
have been collected. Preliminary results indicate that remedial actions and
management decisions at the FHC site have resulted in a reduction in groundwater
concentrations with groundwater concentrations achieving or close to cleanup
objectives. The high number of sampling locations currently achieving cleanup
objectives is consistent with a reduced monitoring effort. All locations recommended
for removal from routine monitoring have achieved the cleanup goal based on the
Student's T-test and power analysis.
Frontier Hard Chrome Site 3 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
Additional Recommendations:
• The majority of the analysis above was completed before several wells in the
network were damaged as a result of site redevelopment. Some wells may need to
be replaced or rehabilitated in order to achieve stated site monitoring objectives.
The general recommendations for the network are to: 1) monitor "hot spots" in Zones
A and B, and 2) monitor sufficient delineation points down and cross-gradient to
confirm contaminant containment.
The recommendation that no new monitoring locations are needed does not imply
that monitoring wells damaged or destroyed during site redevelopment do not need
to be replaced. New wells may be required, but their placement near 'old' locations
identified as important is recommended.
• Monitoring data at the FHC Site show some variance relative to concentrations
(resulting in no trend). In most cases, variance in the data can be explained by site
characteristics and geochemical processes. Continue monitoring for concentration
trends and potentiometric water levels to determine how the hydraulic influence of
the Columbia River may be contributing to underlying variance in the data.
Additionally, area redevelopment may cause changes in recharge patterns (new
paved areas, installation of permeable paving), which may be reflected in aquifer
characteristics and concentration trends.
• Collect analytical data on total chromium as well as dissolved (Cr(VI)) chromium.
Monitor turbidity in groundwater samples to ensure that only dissolved chromium is
being measured in the sample. Flag samples that have been filtered.
• Continue development and updating of the comprehensive site database including
both total and dissolved chromium analytical results. Validated analytical data for all
wells in the area should be added to database within a reasonable time after
sampling. Each well should have a complete record of historic sampling events.
• Survey location coordinates and elevations for all wells. Make data available to all
stakeholders.
Frontier Hard Chrome Site 4 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
5.0 CITED REFERENCES
AFCEE. (2003). Monitoring and Remediation Optimization System (MAROS) 2.2
Software Users Guide. Air Force Center for Environmental Excellence.
http://www.gsi-net.com/software/MAROS V2 1Manual.pdf
AFCEE. (1997). Air Force Center for Environmental Excellence, AFCEE Long-Term
Monitoring Optimization Guide, http://www.afcee.brooks.af.mil.
Aziz, J. A., C. J. Newell, M. Ling, H. S. Rifai and J. R. Gonzales (2003). "MAROS: A
Decision Support System for Optimizing Monitoring Plans." Ground Water 41(3):
355-367.
Gilbert, R. O. (1987). Statistical Methods for Environmental Pollution Monitoring. New
York. Van Norstrand Reinhold.
Ridley, M.N., Johnson, V. M and Tuckfield, R. C. (1995). Cost-Effective Sampling of
Ground Water Monitoring Wells. HAZMACON. San Jose, California.
USEPA. (2001) Frontier Hard Chrome Superfund Site Amended Record of Decision.
USEPA Region X. August, 2001.
USEPA (1988) Frontier Hard Chrome Superfund Site Record of Decision: Groundwater
Operable Unit. USEPA Region X. July 1988.
USEPA (1987) Frontier Hard Chrome Superfund Site Record of Decision: Soil Operable
Unit. USEPA Region X. December, 1987.
USEPA (1992). Methods for Evaluating the Attainment of Cleanup Standards: Volume 2
Ground Water. Washington, D.C., United States Environmental Protection
Agency Office of Policy Planning and Evaluation.
Weston (2007). Frontier Hard Chrome Event 11 Long-Term Monitoring Report (June
2007 Results). Weston Solutions, Inc. August, 2007.
Frontier Hard Chrome Site 5 Groundwater Monitoring
Vancouver, Washington Network Optimization
-------
GROUNDWATER MONITORING NETWORK OPTIMIZATION
FRONTIER HARD CHROME
Vancouver, Washington
TABLES
Table 1 Monitoring Well Network Summary
Table 2 Aquifer Input Parameters
Table 3 Well Trend Summary Results: 2003-2007
Table 4 Moment Estimates and Trends
Table 5 Well Redundancy and Cleanup Status Summary Results
Table 6 MCES Sampling Frequency Analysis Results
Table 7 Final Recommended Groundwater Monitoring Network Frontier Hard Chrome
-------
Issued: 21-DEC-2007
Page 1 of 2
TABLE 1
MONITORING WELL NETWORK SUMMARY
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
Well Name
Hydrologic
Zone
Screened
Interval [ft
bgs]
Source or
Tail (for
MAROS)
Minimum
Sample Date
Maximum
Sample Date
Number of
Samples
(2003-2007)
Current
Sampling
Frequency
Well Description
Zone A
B85-4
RA-MW-11A
RA-MW-12A
RA-MW-13A
RA-MW-14A
RA-MW-15A
RA-MW-16A
RA-MW-17A
W85-6A
W85-7A
W92-16A
W97-18A
W97-19A
W98-20A
W98-21A
W99-R5A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
21.5-26.5
22.9-27.6
23.2-27.9
22.5-27.1
20.3-25.1
22.1-26.6
22.2-26.7
21.7-26.2
17-27
16.5-26.5
24-34
22.5-27.5
20-25
22-27
21-26
22-32
S
S
S
T
T
S
T
T
T
T
T
T
T
T
T
T
2/5/2004
10/16/2003
10/17/2003
10/15/2003
10/15/2003
10/15/2003
10/14/2003
10/14/2003
2/9/2004
2/6/2004
2/5/2004
2/5/2004
2/6/2004
2/7/2004
2/9/2004
2/7/2004
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
11
12
12
12
12
12
12
12
9
11
11
11
11
11
11
11
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Historic high concentrations, monitors central area downgradient
of plume.
Monitors northern edge of plume, near area of historic high, area
of remedy barrier, paired with Zone B well.
Monitors area of historic highest concentrations and permeable
reactive barrier, nested with Zone B wells.
Very low concentration area near permeable reactive barrier,
nested with Zone B wells.
Monitors area of historic high concentrations and eastern
permeable reactive barrier, nested with Zone B well.
Monitors center plume between remedy and B85-4, low Cr in
Zone A, paired with high Cr well in Zone B.
Monitors center plume between remedy and B85-4, low Cr in
Zone A, paired with high Cr well in Zone B.
Northeastern edge in remedy zone, not paired with Zone B well.
Downgradient, center of plume, paired with Zone B well.
Downgradient, center-west of plume, paired with Zone B well.
Delineates western edge of plume near-downgradient of remedy,
low detection frequency, paired with Zone B well.
Delineates far eastern edge of plume, paired with Zone B well.
Delineates far western downgradient edge of plume, paired wit
Zone B well.
Monitors downgradient, center of plume, not paired with Zone B
well.
Monitors downgradient, eastern edge of plume, paired with Zone
Swell.
Monitors farthest downgradient tail, historic edge of plume, paired
with Zone B well.
-------
Issued: 21-DEC-2007
Page 2 of 2
TABLE 1
MONITORING WELL NETWORK SUMMARY
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
Well Name
Hydrologic
Zone
Screened
Interval [ft
bgs]
Source or
Tail (for
MAROS)
Minimum
Sample Date
Maximum
Sample Date
Number of
Samples
(2003-2007)
Current
Sampling
Frequency
Well Description
Zone B
B85-3
B87-8
RA-MW-11B
RA-MW-12B
RA-MW-12C
RA-MW-13B
RA-MW-13C
RA-MW-14B
RA-MW-15B
RA-MW-16B
W85-6B
W85-7B
W92-16B
W97-18B
W97-19B
W98-21 B
W99-R5B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
24-29
24.5-29.5
28.3-32.9
28.3-32.8
34.5-39
27.3-31.9
34.6-39.5
25.5-30.1
27.7-32.5
27.9-32.5
44-49
44-49
35-45
39.5-44.5
40-45
39-44
44-49
T
S
S
T
T
T
T
T
S
S
T
T
S
T
T
T
T
2/5/2004
2/4/2004
10/16/2003
10/17/2003
10/17/2003
10/16/2003
2/3/2004
10/15/2003
10/15/2003
10/14/2003
2/9/2004
2/6/2004
10/14/2003
2/6/2004
2/6/2004
2/9/2004
2/7/2004
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
6/5/2007
11
11
12
12
12
12
11
12
12
12
9
11
12
11
11
11
11
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Monitors upgradient northwest corner, near former FHC building,
west of reducing zone.
Monitors high concentration area immediately downgradieent of
remedy, not paired with Zone A well.
Monitors northern edge of plume, near area of historic high
groundwater concentrations, paired wit A Zone well.
Monitors northern edge of upper Zone B, near area of historic
high groundwater concentrations, paired with Zone A well.
Monitors northern edge of lower Zone B, near area of historic high
groundwater concentrations, paired with Zone A well.
Monitors upper Zone B near reactive barrier, area of low
concentrations.
Monitors lower Zone B near reactive barrier, area of low
concentrations.
Monitors eastern edge of Zone B near remedy, paired with Zone A
well.
Monitors immediately downgradient of permeable barrier, 'hot
spot' in Zone B.
Monitors center plume between remedy and downgradient area,
low Cr in Zone A, paired with high Cr in Zone B.
Downgradient, center of plume, paired with Zone A well.
Downgradient center of plume, paired with Zone A well.
Delineates western edge of plume near-downgradient of remedy,
variable concentrations, paired with Zone A well.
Delineates far eastern edge of plume, paired with Zone A well.
Delineates far western downgradient edge of plume, paired wit
Zone A well.
Monitors downgradient, eastern edge of plume, paired with Zone
A well.
Downgradient tail edge of plume, paired with Zone A well.
Notes:
1. Wells listed are in current monitoring program. Data from USEPA Region 10, Sept. 2007. Well locations illustrated on Figi
2. Groundwater zones are based on the depth of the well screened interval. Zone A is in the upper alluvial aquifer; Zone B
3. Number of samples is the number of quarters the well has been sampled 2003-2007.
lure 1.
is in the more transmissive lower depth of the alluvial aquifer.
-------
Issued 21-DEC-2007
Page 1 of 1
TABLE 2
AQUIFER INPUT PARAMETERS
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
Parameter
Current Plume5 Length
Maximum Plume5 Length
Plume5 Width
SeepageVelocity (ft/yr) 3
Zone A
Zone B
Distance to Receptors (Columbia River)
GWFIuctuations
SourceTreatment
Contaminant Type
NAPLPresent
Priority Constituent
Chromium (total)
Parameter
Groundwater flow direction
Porosity
Source Location near Well
Source X-Coordinate
Source Y-Coordinate
Coordinate System
Saturated Thickness
Zone A
Zone B
Value
1000
2500
1000
182.5
821.25
3000
Yes
Permeable reactive
barrier/chemical reductant
Metals
No
Cleanup Goals
50
Value
s/sw
0.3
North of RA-MW-1 1
1091615.515
112599.082
NAD 83 SP Washington South
15
50
Units
ft
ft
ft
ft/yr
ft/yr
ft
~
—
-
~
ug/L
225 degrees
-
~
ft
ft
ft
ft
Notes:
1. Aquifer data from ROD and ROD Amendment (USEAPA, 1988; USEPA, 2001).
2. Source coordinates estimated to center of historic FHC building.
3. * = a wide range of transmissivites are present in the aquifer, and groundwater velocity
calculations result in a range, with values shown being the best estimate.
4. Cleanup objectives are Model Toxics Control Act Method A promulgated by the
Washington State Department of Ecology for property with unrestricted use.
5. 'Plume' as used in this report descripes the extent of groundwater affected by
source-associated chromium at any concentration; rather than groundwater above
the regulatory screening limit.
-------
Issued 21-DEC-2007
Page 1 of 1
TABLE 3
WELL TREND SUMMARY RESULTS: 2003-2007
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
WellName
Number of
Samples
Number of
Detects
Percent
Detection
Maximum
Result [ug/L]
Max Result
Above
Standard?
Average
Result [ug/L]
Average
Result Above
Standard?
Mann-
Kendall
Trend
Linear
Regression
Trend
Overall
Trend Result
Zone A Wells
B85-4
RA-MW-1 1A
RA-MW-12A
RA-MW-1 3A
RA-MW-1 4A
RA-MW-1 5A
RA-MW-1 6A
RA-MW-1 7A
W85-6A
W85-7A
W92-16A
W97-18A
W97-19A
W98-20A
W98-21A
W99-R5A
Zone B Wells
B85-3
B87-8
RA-MW-1 1B
RA-MW-1 2B
RA-MW-1 2C
RA-MW-1 3B
RA-MW-1 3C
RA-MW-1 4B
RA-MW-1 SB
RA-MW-1 6B
W85-6B
W85-7B
W92-16B
W97-18B
W97-19B
W98-21B
W99-R5B
11
12
12
12
12
12
12
12
9
11
11
11
11
11
11
11
11
11
12
12
12
12
11
12
12
12
9
11
12
11
11
11
11
10
10
12
10
9
11
11
11
8
9
7
6
10
10
10
4
8
11
11
10
12
6
10
9
12
12
8
4
12
8
9
10
10
91%
83%
100%
83%
75%
92%
92%
92%
89%
82%
64%
55%
91%
91%
91%
36%
73%
100%
92%
83%
100%
50%
91%
75%
100%
100%
89%
36%
100%
73%
82%
91%
91%
37.7
50.1
5260
4.4
5.4
37
9.2
10.2
14.3
3.6
6.3
0.6
7.9
5.1
7.1
4.1
6.3
241
69.2
26
12.2
7.1
7.3
7
192
225
13
18
225
1.3
12.5
6.6
9.9
No
Yes
Yes
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
Yes
No
No
No
No
No
Yes
Yes
No
No
Yes
No
No
No
No
8.3
9.7
682.0
1.4
1.8
6.2
3.3
5.1
4.6
1.8
1.6
0.5
2.7
2.1
2.6
0.8
2.84
61.8
9.9
5.3
4.0
1.4
2.6
1.5
78.1
46.4
4.5
3.7
46
0.881
3.41
2.77
3.99
No
No
Yes
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
No
No
No
No
No
No
Yes
No
No
No
No
No
No
No
No
D
D
NT
S
S
NT
D
S
S
S
NT
S
PD
S
PD
NT
NT
NT
D
D
S
NT
S
NT
NT
NT
D
D
NT
NT
D
D
D
D
D
NT
S
NT
NT
D
S
S
S
PD
D
D
D
D
NT
NT
NT
D
NT
S
NT
S
NT
NT
NT
D
D
NT
NT
D
D
D
D
D
NT
S
S
NT
D
S
S
S
S
PD
D
PD
D
NT
NT
NT
D
S
S
NT
S
NT
NT
NT
D
D
NT
NT
D
D
D
A/ores
1 . Trends were evaluated for data collected between 2003 and 2007.
2. Number of Samples is the number of quarterly samples for the compound at this location.
4. Screening level Standard from Washington Department of Ecology = 50ug/L. Values above the Standard indicated irgold •
5. D = Decreasing; PD = Probably Decreasing; S = Stable; PI = Probably Increasing; I = Increasing; N/A = Insufficient Data to determine trend;
^TManW-ReWfilMBraWiaiT^^e nRfirfr!SSsS'
-------
Issued 21-DEC-2007
Page 1 of 1
TABLE 4
MOMENT ESTIMATES AND TRENDS
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
Zone
Effective Sample
Event Date
Number of
wells in
network
Dissolved Cr Mass
Estimate [Kg]
Zone A
Zone B
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
16
16
16
16
15
15
16
16
16
16
16
Trend
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
17
17
17
17
16
16
17
17
17
17
17
Trend
0.32
0.63
0.59
0.26
0.37
0.25
0.18
0.26
0.08
0.25
0.25
D
5.66
3.34
2.70
2.76
1.91
1.41
1.16
1.74
0.58
1.68
1.44
D
Distance of Center of
Mass from Source [ft]
830
1192
957
1038
753
931
962
912
1058
984
1010
NT
901
1092
1026
817
709
878
831
792
673
737
881
D
Notes:
1. Input parameters for the moment analysis are listed in Table 2.
2. Moments are based on all wells sampled during the quarter including the effective date indicated.
3. Number of wells is the total number of locations sampled for the plume during the year indicated.
4. Estimated mass is the total dissolved mass of total chromium within the network indicated.
5. Trends are Mann Kendall trends on the moments, S=Stable, D = Decreasing, NT = No Trend.
6. First moments are illustrated on Figure 2.
-------
Issued: 21-DEC-2007
Page 1 of 1
TABLE 5
WELL REDUNDANCY AND CLEAN-UP STATUS SUMMARY RESULTS
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
WellName
Mann-Kendall
Trend
2003-2007
Maximum
Concentration
Cr(ug/L)
Cr Average Slope
Factor
Statistically Below
Screening Level
>80% Power
Sequential T-Test
Result
MAROS
Statistically
Redundant
Recommendation After
Qualitative Review
Zone A
B85-4
RA-MW-11A
RA-MW-12A
RA-MW-13A
RA-MW-14A
RA-MW-15A
RA-MW-16A
RA-MW-17A
W85-6A
W85-7A
W92-16A
W97-18A
W97-19A
W98-20A
W98-21A
W99-R5A
Zone B
B85-3
B87-8
RA-MW-11B
RA-MW-12B
RA-MW-12C
RA-MW-13B
RA-MW-13C
RA-MW-14B
RA-MW-15B
RA-MW-16B
W85-6B
W85-7B
W92-16B
W97-18B
W97-19B
W98-21B
W99-R5B
D
D
NT
S
S
NT
D
S
S
S
NT
S
PD
S
PD
NT
NT
NT
D
D
S
NT
S
NT
NT
NT
D
D
NT
NT
D
D
D
37.70
50.10
5260.00
4.40
5.40
37.00
9.20
10.20
14.30
3.60
6.30
0.60
7.90
5.10
7.10
4.10
6.30
241 .00
69.20
26.00
12.20
7.10
7.30
6.50
192.00
225.00
12.90
17.70
225.00
1.30
12.50
6.60
9.90
0.12
0.43
0.58
0.40
0.09
0.12
0.08
0.15
0.20
0.07
0.27
0.41
0.10
0.09
0.04
0.32
0.26
0.17
0.17
0.17
0.22
0.42
0.23
0.46
0.39
0.35
0.21
0.52
0.33
0.45
0.23
0.08
0.05
YES
YES
WO
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
YES
WO
YES
YES
YES
YES
YES
YES
WO
WO
-
YES
WO
YES
YES
YES
YES
Attained
Attained
Attained
Attained
Attained
Yes
No
No
No
Yes
Yes
Yes
No
Yes
Yes
Yes
No
No
Yes
No
No
No
Yes
Yes
Yes
Yes
No
Yes
No
No
No
Yes
No
No
No
No
No
No
Retain
Exclude
Retain
Exclude
Exclude
Retain
Retain
Retain
Retain
Exclude
Retain
Retain
Retain
Exclude
Retain
Retain
Retain
Retain
Exclude
Retain
Retain
Exclude
Exclude
Exclude
Retain
Retain
Retain
Exclude
Retain
Retain
Retain
Retain
Retain
Wotes:
1.
Slope factors close to 1 show the concentrations cannot be estimated from the nearest neighbors, and the well is important in the network.
2. Slope factors were calculated using data between January 2006 and June 2007.
3. Locations with slope factors below 0.3 and area ratios below 0.8 were considered for elimination.
4. Wells statistically below the cleanup level (50 ug/L) by Student's-T Test and >80% Power indicated. "Attained" indicates wells statistically
and the final recommendation reflects both
statistical findings and regulatory and site specific factors.
-------
Issued: 21-DEC-2007
Page 1 of 1
TABLE 6
MCES SAMPLING FREQUENCY ANALYSIS RESULTS
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
Well Name
Recent
Concentration
Rate of Change
[mg/yr]
Recent MK
Trend (2006
2007)
Frequency
Based on
Recent Data
(2006-2007)
Overall
Concentration
Rate of Change
[mg/yr]
Overall MK
Trend
(2003 - 2007)
Frequency
Based on
Overall Data
(2003 - 2007)
MAROS
Recommended
Frequency
Current
Sampling
Frequency
Final
Recommended
Frequency
Zone A Wells
B85-4
RA-MW-11A
RA-MW-12A
RA-MW-13A
RA-MW-14A
RA-MW-15A
RA-MW-16A
RA-MW-17A
W85-6A
W85-7A
W92-16A
W97-18A
W97-19A
W98-20A
W98-21A
W99-R5A
-4.00E-06
1.31E-06
-1.27E-03
-1.42E-06
-1.84E-07
-3.69E-06
-2.06E-06
-7.68E-06
1.71E-06
5.67E-07
-2.46E-07
-1 .49E-07
1 .84E-06
1 .66E-06
-1 .20E-07
-3.32E-07
S
S
NT
S
S
S
S
S
S
S
NT
PD
NT
NT
S
S
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
-1.48E-05
-1.75E-05
5.99E-04
-7.25E-07
-3.91 E-07
1.17E-06
-2.56E-06
-1.45E-06
-4.82E-06
-2.96E-07
-2.35E-06
-4.89E-09
-3.38E-06
-2.23E-06
-2.70E-06
-1.12E-06
D
D
NT
S
S
NT
D
S
S
S
NT
S
PD
S
PD
NT
Annual
Annual
Quarterly
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Biennial
Biennial
Quarterly
Biennial
Biennial
Biennial
Biennial
Biennial
Biennial
Biennial
Biennial
Annual
Biennial
Biennial
Biennial
Biennial
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Zone B Wells
B85-3
B87-8
RA-MW-11B
RA-MW-12B
RA-MW-12C
RA-MW-13B
RA-MW-13C
RA-MW-14B
RA-MW-15B
RA-MW-16B
W85-6B
W85-7B
W92-16B
W97-18B
W97-19B
W98-21B
W99-R5B
-7.10E-06
9.07E-05
-1 .35E-05
-9.87E-06
-8.58E-07
-5.49E-07
-8.67E-07
-2.35E-07
-2.64E-04
3.04E-05
-7.12E-06
O.OOE+00
2.55E-05
3.45E-07
1.74E-06
-2.01 E-06
-2.71E-06
S
S
NT
S
NT
S
S
S
S
NT
S
S
NT
NT
NT
S
S
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
5.25E-07
-4.39E-05
-2.08E-05
-6.41 E-06
-5.10E-07
-7.81 E-07
-5.17E-07
-8.40E-07
4.74E-06
1.01E-05
-5.15E-06
-1.13E-05
8.99E-06
2.61 E-07
-5.83E-06
-3.20E-06
-5.94E-06
NT
NT
D
D
S
NT
S
NT
NT
NT
D
D
NT
NT
D
D
D
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Biennial
Annual
Biennial
Biennial
Biennial
Biennial
Biennial
Biennial
Annual
Annual
Biennial
Biennial
Annual
Biennial
Biennial
Biennial
Biennial
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Quarterly
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Annual
Notes:
1. Concentration rate of change is from linear regression calculations. 'Recent' concentration rate of change and MK trends are calculated from data collected 2006 - 2007.
2. MK trend = Mann Kendall trend. D = Decreasing, PD = Probably Decreasing, S = Stable, NT = No Trend.
3. Recent data frequency is the estimated sample frequency based on the recent trend.
4. Overall rate of change and MK trend are for the full data set (2003-2007) for each well. The overall result is the estimated sample frequncy based on the full data record.
6. MAROS Recommended Frequency is the final frequency from the MAROS calculations based on both recent and overall trends.
7. Current frequency is the approximate sampling frequency currently implemented.
8. The final recommended sampling frequency is based on a combination of qualitative and statistical evaluations.
-------
Issued: 21-DEC-2007
Page 1 of 1
TABLE 7
FINAL RECOMMENDED MONITORING NETWORK FRONTIER HARD CHROME
LONG-TERM MONITORING OPTIMIZATION
FRONTIER HARD CHROME SUPERFUND SITE
VANCOUVER, WASHINGTON
WellName
Zone A Wells
B85-4
RA-MW-11A
RA-MW-12A
RA-MW-13A
RA-MW-14A
RA-MW-15A
RA-MW-16A
RA-MW-17A
W85-6A
W85-7A
W92-16A
W97-18A
W97-19A
W98-20A
W98-21A
W99-R5A
Zone B Wells
B85-3
B87-8
RA-MW-11B
RA-MW-12B
RA-MW-12C
RA-MW-13B
RA-MW-13C
RA-MW-14B
RA-MW-15B
RA-MW-16B
W85-6B
W85-7B
W92-16B
W97-18B
W97-19B
W98-21B
W99-R5B
Total Chromium
Percent
Detection
Mann
Kendall
Trend
Statistically
Below
Standard?
MAROS
Redundancy
Determination
Recommendation
After Qualitative
Review
Final
Recommended
Frequency
91%
83%
100%
83%
75%
92%
92%
92%
89%
82%
64%
55%
91%
91%
91%
36%
73%
100%
92%
83%
100%
50%
91%
75%
100%
100%
89%
36%
100%
73%
82%
91%
91%
D
D
NT
S
S
NT
D
S
S
S
NT
S
PD
S
PD
NT
NT
NT
D
D
S
NT
S
NT
NT
NT
D
D
NT
NT
D
D
D
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
Retain
Exclude
Retain
Exclude
Exclude
Retain
Retain
Retain
Retain
Exclude
Retain
Retain
Retain
Exclude
Retain
Retain
Annual
Exclude
Annual
Exclude
Exclude
Annual
Annual
Annual
Annual
Exclude
Annual
Annual
Annual
Exclude
Annual
Annual
Retain
Retain
Exclude
Retain
Retain
Exclude
Exclude
Exclude
Retain
Retain
Retain
Exclude
Retain
Retain
Retain
Retain
Retain
Annual
Annual
Exclude
Annual
Annual
Exclude
Exclude
Exclude
Annual
Annual
Annual
Exclude
Annual
Annual
Annual
Annual
Annual
Notes:
1. Mann Kendall Trends: D = Decreasing; PD = Probably Decreasing; S = Stable; PI = Probably Increasing; I = Increasing;
NT = No Trend; ND = well has all non-detect.
2. Mann-Kendall trends 2003 - 2007 are shown.
3. Statistically below standard based Student's T-Test with >80% statistical power for data between 2003-2007.
Cleanup standard is Washington Ecology MTCA A = 50ug/L Total Chromium.
4. MAROS redundancy indicates well has low SF and high AR and CR.
5. Final Recommendation based on statistical as well as qualitative evaluation.
-------
FIGURES
GROUNDWATER MONITORING NETWORK OPTIMIZATION
FRONTIER HARD CHROME
Vancouver, Washington
Figure 1 Frontier Hard Chrome Groundwater Monitoring Network
Figure 2 Frontier Hard Chrome Concentration Trend and First Moment Results
Figure 3 Zone A Chromium Concentration Uncertainty
Figure 4 Zone B Chromium Concentration Uncertainty
Figure 5 Zone A Optimized Network Concentration Uncertainty
Figure 6 Zone B Optimized Network Concentration Uncertainty
Figure 7 Frontier Hard Chrome Zone A Source Area Summary
Figure 8 Frontier Hard Chrome Zone B Source Area Summary
-------
Average [Cr] [mg/L]
A ND-0.001
A 0.001 - 0.01
A 0.01 - 0.05
A 0.05-0.5
A >0.5
Legend
- Roads
Cassidy Mfg.
Soil Treatment
Area (Approximate)
ISRM Groundwater
Treatment Zone
(Approximate)
Notes:
1. Aerial map from 1994 shows historic site features
and road locations. FHC buildings and industrial
area to the south have been demolished.
2. The Soil Remedy and ISRM Groundwater Remedy
are approximate areas based on USEPAmaps.
3. Groundwater monitoring locations for each Zone
are indicated. Average Cr concentrations 2003 - 2007
at each location are indicated by color.
4. Total Cr MTCA Standard A cleanup level = 0.05 mg/L.
Large Map Scale (ft)
0 150 300
FRONTIER HARD CHROME
GROUNDWATE
MONITORING NETWORK
Vancouver, Washington
Coord. Sys.
NAD 83 SP Wash. S. FT.
Drawn By:
MV
Chk'd By:
MV
AppVd By:
MV
Issued:
21-DEC-2007
Revised:
Map ID:
FIGURE 1
-------
Zone A Well Locations
Legend
Mann Kendall Trend Cr
^ Decreasing
• Probably Decreasing
O Stable
• Probably Increasing
• Increasing
Non Detect (2003-2007)
No Trend
Insufficient Data
First Moment
(Effective Date Shown)
Soil Remedy Area
ISRMTreatment
Cassidy Mfg.
Roads
Notes:
1. Aerial map from 1994 shows historic site features
and road locations. FHC buildings and industrial
area to the south have been demolished.
2. Mann-Kendall trends we re determined for data
collected between 2003-2007.
3. First moments were determined using quarterly data.
An effective date of the quarterly sampling event is
indicated.
4. Total CrMTCA Standard A cleanup level = 0.05 mg/L.
Scale (ft)
^•=
0 150 300
FRONTIER HARD CHROME
CONCENTRATION TREND AND
FIRST MOMENT RESULTS
Vancouver, Washington
Coord. Sys.
NAD 83 SP Wash. S. FT.
Drawn By:
MV
Chk'd By:
MV
AppVd By:
Issued:
21-DEC-2007
Revised:
Map ID:
FIGURE 2
-------
NORT
1 1 2600.0 -
1 1 2400.0 -
1 1 2200.0 -
1 1 2000.0 -
111800.0-
111600.0-
111400.0-
111200.0-
111000.0-
1 1 0800.0 -
1089
H
Figures RA-MW-HA RA-MW-12A RA_MW_17A
Zone A Chromium \,. \ / RA-MW-ISA
W92-16A ^mk_Jli/
Concentration Uncertainty J^S^*T£-— RA-MW-MA
x' / | / B85-4 //
/S I S / /
/' / i / ;//
/ / s \ i / /I
/ * / i / / / /
/ \\/ / /
/ _— ( inmr n Jr vny°5 BA * /
/ ^- ••— WSS-^A— —— — —— ^ WWO-1 DA y I
xx ^^*^^ / xx ^ / /
W97-1 9A X, ^- ^ 1 xX l^/
/-- S ' X^ 3 / /
/ -----..^ 1 xxX I / /
. ' — Jr* V /
/ X W98-20A ^^ /
7 xX ^^^^ /
/ ^X ^^^ /
/ S ,' ^
/ ^x s ^s ^^_____ Very low spatial uncertainty across the plume
/ /' ^'^-— — -" "
/ xx ^^ "
' x ^
/ /'',/'''
^'
New Location
Analysis for
CHROMIUM, TOTAL
Existing
Locations
Potential areas for
new locations are
indicated by triangles
with a high SF level.
Estimated SF Level:
S - Small
M - Moderate
L - Large
E - Extremely large
High SF-> high
estimation error ->
possible need for
new locations
Low SF -> low
estimation error ->
no need for new
locations
Back to
Access
V J
1 ' EAST
500.0 1090000.0 1090500.0 1091000.0 1091500.0 1092000.0 1092500.0
-------
1 1 9ftnn n New Location
i i zouu.u ~i~
1 1 2600.0 -
1 1 2400.0 -
1 1 2200.0 -
1 1 2000.0 -
111800.0-
111600.0-
111400.0-
111200.0-
111000.0-
1 1 0800.0 -
Figure 4
Zone B Chromium
B85-3
Concentration Uncertainty RAMWHR
' RA-MW-11B . RA-MW-12B/C
2006-2007 v. X/*K /
><< ' \ V\/ RA-MW-13B/C
W92-16B x ^^LA OW. /
~^^>Cy/ •Alsl&jMtr--- " — RA-MW-14B
' ' ^^^^%T^^^Bt~---__
XX / M / | X^^NS^"S^:^ RA-MW-16B
/ / / II B87-8 ^S. ~~/f W97-18B
X / / || //'\v
/ / * II / ' ^\.
/ .' ^ || / / RA-MW-15B
S ^ / / M M Si
/ / ' m \ \ / ,
/ "'/ /
/ '/ J
* _/ 1^ *
X -,-"'*" W85-7B ^ ' /
xx --" ^ s \
™»-^^- s \ > /
/ — — x I /
, — — V I /
/ "~"~—3^ W98-21B
/ ^-^
/ ^^^
' S ^"'
f ^
/ ^''
/ ^'
/ ^'*'
/ **
/ ^''
/ ^^'"
/ ^
/ ^
/ ^^
11111
Analysis for
CHROMIUM, TOTAL
Existing
•
Locations
Potential areas for
new locations are
indicated by triangles
with a high SF level.
Estimated SF Level:
S - Small
M - Moderate
L - Large
E - Extremely large
High SF-> high
estimation error ->
possible need for
new locations
Low SF -> low
estimation error ->
no need for new
locations
Back to
Access
V J
EAST
1089500.0 1090000.0 1090500.0 1091000.0 1091500.0 1092000.0 1092500.0
-------
! ! ofiNoo o1^ New Location
1 1 ZDUU.U -j~~~^^
1 1 2400.0 -
1 1 2200.0 -
1 1 2000.0 -
111800.0-
111600.0-
111400.0-
111200.0-
111000.0-
1 1 0800.0 -
Figure 5 RA MW I?A
3 KA MVV ^A RA-MW-17A
Zone A
Optimized Network W92-16^ ^^fs RA_MW_16A
Concentration Uncertainty x' j V'TSTV^'NS
2006-2007 ' I/ S^1 s^^N
RA,BW-15A < Tl -^
x , . ~~~^P W97-18A
x 1 / B85-4 .7
1 S / "
/ 1 / / /
X 1 / S / /
' / / /
XX / ' '
.' 1 / / /
xX I / / /
7 S ' ' /X /
X i / / /
/ I'/ s/
X — -"f W85-6A /
X _ — — """""" I /
^c\a929A S 1 /
' "~~"~~~"~ — — I/
/ ~""~—jJ( W98-21A
/ ^^
' ^
/ ^^
/ ^^^
/ S s'
1 ^'
/ ^s'
t s'
/ ^'
, S
f ^
/ /''''
1 ^
1^'
11111
Analysis for
CHROMIUM, TOTAL
Existing
•
Locations
Potential areas for
new locations are
indicated by triangles
with a high SF level.
Estimated SF Level:
S - Small
M - Moderate
L - Large
E- Extremely large
High SF-> high
estimation error ->
possible need for
new locations
Low SF-> low
estimation error ->
no need for new
locations
s -\
Back to
Access
V J
EAST
1089500.0 1090000.0 1090500.0 1091000.0 1091500.0 1092000.0 1092500.0
-------
H9Rnnn™ New Location
I I ZOUU.U -j~~~^^
1 1 2600.0 -
1 1 2400.0 -
1 1 2200.0 -
1 1 2000.0 -
111800.0-
111600.0-
111400.0-
111200.0-
111000.0-
1 1 0800.0 -
Figure 6
Zone B
Optimized Network B85 3
Concentration Uncertainty ^L RA MW 123/0
2006-2007 W92-16B / 1 \^V RA-MW-15B
•x* ^^3T — —— ff^i RA MW 16C
' xx | ^— _?^^^
X XX | M I B87-8 ~~~ — — *p W97-18B
xX xX I | /I
/ / I'M //
xX xxX 1 ' / 1
/ X 1 i X /
x'V s .' x^ /
X x 'I X /
X XX ll XX /
X x 1 / M'
Xx ____^W85-6B /
^y — — \ '
W97-19B X'' _____ — — """" . /
^~ — S \ ,
/ ~~~~~~~~~~~^ \ '
/ ~~~-yil W98-21B
/ ^'"'
/ S ^"^
' ^-^
/ ^-^^
/ ^-"
/ ^
/ ^"'
1 ^'
/ ^
• ^
/ ^'"
/ ^, -^
iTW99-R5B
11111
Analysis for
CHROMIUM, TOTAL
Existing
Locations
Potential areas for
new locations are
indicated by triangles
with a high SF level.
Estimated SF Level:
S - Small
M - Moderate
L - Large
E- Extremely large
High SF-> high
estimation error ->
possible need for
new locations
Low SF-> low
estimation error ->
no need for new
locations
Back to
Access
V J
EAST
1089500.0 1090000.0 1090500.0 1091000.0 1091500.0 1092000.0 1092500.0
-------
0 If
Average Cr Concentration 2003-2007
Legend
Average Cr Concentration [mg/L]
A ND-0.001
A 0.001 - 0.01
A 0.01 - 0.05
A 0.05-0.5
A >0.5
Cr Screening Level = 0.05 mg/L
Mann Kendall Trend Cr
Decreasing • Non Detect (2003-2007)
Probably Decreasing •
Stable •
Probably Increasing
No Trend
Insufficient Data
Increasing
Soil Remedy Area
Proposed Development
Recommended Sampling Frequency
D Annual Sampling
Eliminate from
routine monitoring
Notes:
1. Average Total Cr concentrations calculated
for data 2003-2007.
2. Mann Kendall trends were determined
for Total Cr 2003-2007.
3. All built structures are proposed
for the property redevelopment.
Drawingsfordevelopment received
from developer October, 2007.
FRONTIER HARD CHROME
ZONE A SOURCE AREA
SUMMARY
Vancouver, Washington
Coord. Sys.
NAD 83 SP Wash. S. FT.
Drawn By:
MV
Chk'd By:
MV
AppVd By:
MV
Issued:
21-DEC-2007
Revised:
Map ID:
FIGURE 7
-------
T"* L-L
^ i-r^—j_i4_| ra -i ™-» - ,• cj_
•——IlJ ^ I—IfM-MMMeB .
- .,!_ I - T X|
RA-MW-15B p=-—
Average Cr Concentration 2003-2007
Legend
Average Cr Concentration [mg/L]
A ND-0.001
A 0.001 - 0.01
A 0.01 - 0.05
A 0.05-0.5
A >0.5
Cr Screening Level = 0.05 mg/L
Mann Kendall Trend Cr
Decreasing • Non Detect (2003-2007)
Probably Decreasing •
Stable •
Probably Increasing
No Trend
Insufficient Data
Increasing
Soil Remedy Area
Proposed Development
Recommended Sampling Frequency
D Annual Sampling
Eliminate from
routine monitoring
Notes:
1. Average Total Cr concentrations calculated
for data 2003-2007
2. Mann Kendall trends were determined
for Total Cr 2003-2007.
3. All built structures are proposed
for the property redevelopment.
Drawingsfordevelopment received
from developer October, 2007.
FRONTIER HARD CHROM
ZONE B SOURCE ARE
SUMMARY
Vancouver, Washington
Coord. Sys.
NAD 83 SP Wash. S. FT.
Drawn By:
MV
Chk'd By:
MV
AppVd By:
Issued:
21-DEC-2007
Revised:
Map ID:
FIGURE 8
-------
GROUNDWATER MONITORING NETWORK OPTIMIZATION
FRONTIER HARD CHROME
Vancouver, Washington
APPENDIX A:
MAROS 2.2 Methodology
-------
APPENDIX A
MAROS 2.2 METHODOLOGY
Contents
1.0 MAROS Conceptual Model 1
2.0 Data Management 2
3.0 Site Details 2
4.0 Constituent Selection 3
5.0 Data Consolidation 3
6.0 Overview Statistics: Plume Trend Analysis 3
6.1 Mann-Kendall Analysis 4
6.2 Linear Regression Analysis 4
6.3 Overall Plume Analysis 5
6.4 Moment Analysis 6
7.0 Detailed Statistics: Optimization Analysis 8
7.1 Well Redundancy Analysis- Delaunay Method 8
7.2 Well Sufficiency Analysis - Delaunay Method 9
7.3 Sampling Frequency - Modified CES Method 10
7.4 Data Sufficiency- Power Analysis 11
Cited References
Tables
Table 1 Mann-Kendall Analysis Decision Matrix
Table 2 Linear Regression Analysis Decision Matrix
Figures
Figure 1 MAROS Decision Support Tool Flow Chart
Figure 2 MAROS Overview Statistics Trend Analysis Methodology
Figure 3 Decision Matrix for Determining Provisional Frequency
-------
MAROS METHODOLOGY
MAROS is a collection of tools in one software package that is used in an explanatory,
non-linear but linked fashion. The tool includes models, statistics, heuristic rules, and
empirical relationships to assist the user in optimizing a groundwater monitoring network
system. The final optimized network maintains adequate delineation while providing
information on plume dynamics over time. Results generated from the software tool can
be used to develop lines of evidence, which, in combination with expert opinion, can be
used to inform regulatory decisions for safe and economical long-term monitoring of
groundwater plumes. For a detailed description of the structure of the software and
further utilities, refer to the MAROS 2.2 Manual (AFCEE, 2003; http://www.gsi-
net.com/software/MAROS V2 1Manual.pdf) and Aziz et al., 2003.
1.0 MAROS Conceptual Model
In MAROS 2.2, two levels of analysis are used for optimizing long-term monitoring plans:
1) an overview statistical evaluation with interpretive trend analysis based on temporal
trend analysis and plume stability information; and 2) a more detailed statistical
optimization based on spatial and temporal redundancy reduction methods (see Figures
A.1 and A.2 for further details). In general, the MAROS method applies to 2-D aquifers
that have relatively simple site hydrogeology. However, for a multi-aquifer (3-D) system,
the user has the option to apply the statistical analysis layer-by-layer.
The overview statistics or interpretive trend analysis assesses the general monitoring
system category by considering individual well concentration trends, overall plume
stability, hydrogeologic factors (e.g., seepage velocity, and current plume length), and
the location of potential receptors (e.g., property boundaries or drinking water wells). The
method relies on temporal trend analysis to assess plume stability, which is then used to
determine the general monitoring system category. Since the monitoring system
category is evaluated for both source and tail regions of the plume, the site wells are
divided into two different zones: the source zone and the tail zone.
Source zone monitoring wells could include areas with non-aqueous phase liquids
(NAPLs), contaminated vadose zone soils, and areas where aqueous-phase releases
have been introduced into ground water. The source zone generally contains locations
with historical high ground water concentrations of the COCs. The tail zone is usually the
area downgradient of the contaminant source zone. Although this classification is a
simplification of the plume conceptual model, this broadness makes the user aware on
an individual well basis that the concentration trend results can have a different
interpretation depending on the well location in and around the plume. The location and
type of the individual wells allows further interpretation of the trend results, depending on
what type of well is being analyzed (e.g., remediation well, leading plume edge well, or
monitoring well). General recommendations for the monitoring network frequency and
density are suggested based on heuristic rules applied to the source and tail trend
results.
The detailed statistics level of analysis or sampling optimization consists of well
redundancy and well sufficiency analyses using the Delaunay method, a sampling
frequency analysis using the Modified Cost Effective Sampling (MCES) method and a
Appendix A 7 MAROS 2.2 Methodology
-------
data sufficiency analysis including statistical power analysis. The well redundancy
analysis is designed to minimize monitoring locations and the Modified CES method is
designed to minimize the frequency of sampling. The data sufficiency analysis uses
simple statistical methods to assess the sampling record to determine if groundwater
concentrations are statistically below target levels and if the current monitoring network
and record is sufficient in terms of evaluating concentrations at downgradient locations.
2.0 Data Management
In MAROS, ground water monitoring data can be imported from simple database-format
Microsoft® Excel spreadsheets, Microsoft Access tables, previously created MAROS
database archive files, or entered manually. Monitoring data interpretation in MAROS is
based on historical analytical data from a consistent set of wells over a series of
sampling events. The analytical data is composed of the well name, coordinate location,
constituent, result, detection limit and associated data qualifiers. Statistical validity of the
concentration trend analysis requires constraints on the minimum data input of at least
four wells (ASTM 1998) in which COCs have been detected. Individual sampling
locations need to include data from at least six most-recent sampling events. To ensure
a meaningful comparison of COC concentrations over time and space, both data quality
and data quantity need to be considered. Prior to statistical analysis, the user can
consolidate irregularly sampled data or smooth data that might result from seasonal
fluctuations or a change in site conditions. Because MAROS is a terminal analytical tool
designed for long-term planning, impacts of seasonal variation in the water unit are
treated on a broad scale, as they relate to multi-year trends.
Imported ground water monitoring data and the site-specific information entered in Site
Details can be archived and exported as MAROS archive files. These archive files can
be appended as new monitoring data becomes available, resulting in a dynamic long-
term monitoring database that reflects the changing conditions at the site (i.e.
biodegradation, compliance attainment, completion of remediation phase, etc.). For
wells with a limited monitoring history, addition of information as it becomes available
can change the frequency or identity of wells in the network.
3.0 Site Details
Information needed for the MAROS analysis includes site-specific parameters such as
seepage velocity and current plume length and width. Information on the location of
potential receptors relative to the source and tail regions of the plume is entered at this
point. Part of the trend analysis methodology applied in MAROS focuses on where the
monitoring well is located, therefore the user needs to divide site wells into two different
zones: the source zone or the tail zone. Although this classification is a simplification of
the well function, this broadness makes the user aware on an individual well basis that
the concentration trend results can have a different interpretation depending on the well
location in and around the plume. It is up to the user to make further interpretation of the
trend results, depending on what type of well is being analyzed (e.g., remediation well,
leading plume edge well, or monitoring well). The Site Details section of MAROS
contains a preliminary map of well locations to confirm well coordinates.
Appendix A 2 MAROS 2.2 Methodology
-------
4.0 Constituent Selection
A database with multiple COCs can be entered into the MAROS software. MAROS
allows the analysis of up to 5 COCs concurrently and users can pick COCs from a list of
compounds existing in the monitoring data. MAROS runs separate optimizations for
each compound. For sites with a single source, the suggested strategy is to choose one
to three priority COCs for the optimization. If, for example, the site contains multiple
chlorinated volatile organic compounds (VOCs), the standard sample chemical analysis
will evaluate all VOCs, so the sample locations and frequency should based on the
concentration trends of the most prevalent, toxic or mobile compounds. If different
chemical classes are present, such as metals and chlorinated VOCs, choose and
evaluate the priority constituent in each chemical class.
MAROS includes a short module that provides recommendations on prioritizing COCs
based on toxicity, prevalence, and mobility of the compound. The toxicity ranking is
determined by examining a representative concentration for each compound for the
entire site. The representative concentration is then compared to the screening level
(PRG or MCL) for that compound and the COCs are ranked according to the
representative concentrations percent exceedence of the screening level. The
evaluation of prevalence is performed by determining a representative concentration for
each well location and evaluating the total exceedences (values above screening levels)
compared to the total number of wells. Compounds found over screening levels are
ranked for mobility based on Kd (sorption partition coefficient). The MAROS COC
assessment provides the relative ranking of each COC, but the user must choose which
COCs are included in the analysis.
5.0 Data Consolidation
Typically, raw data from long-term monitoring have been measured irregularly in time or
contain many non-detects, trace level results, and duplicates. Therefore, before the data
can be further analyzed, raw data are filtered, consolidated, transformed, and possibly
smoothed to allow for a consistent dataset meeting the minimum data requirements for
statistical analysis mentioned previously.
MAROS allows users to specify the period of interest in which data will be consolidated
(i.e., monthly, bi-monthly, quarterly, semi-annual, yearly, or a biennial basis). In
computing the representative value when consolidating, one of four statistics can be
used: median, geometric mean, mean, and maximum. Non-detects can be transformed
to one half the reporting or method detection limit (DL), the DL, or a fraction of the DL.
Trace level results can be represented by their actual values, one half of the DL, the DL,
or a fraction of their actual values. Duplicates are reduced in MAROS by one of three
ways: assigning the average, maximum, or first value. The reduced data for each COC
and each well can be viewed as a time series in a graphical form on a linear or semi-log
plot generated by the software.
6.0 Overview Statistics: Plume Trend Analysis
Within the MAROS software there are historical data analyses that support a conclusion
about plume stability (e.g., increasing plume, etc.) through statistical trend analysis of
Appendix A 3 MAROS 2.2 Methodology
-------
historical monitoring data. Plume stability results are assessed from time-series
concentration data with the application of three statistical tools: Mann-Kendall Trend
analysis, linear regression trend analysis and moment analysis. The two trend methods
are used to estimate the concentration trend for each well and each COC based on a
statistical trend analysis of concentrations versus time at each well. These trend
analyses are then consolidated to give the user a general plume stability estimate and
general monitoring frequency and density recommendations (see Figures A.1 through
A.3 for further step-by-step details). Both qualitative and quantitative plume information
can be gained by these evaluations of monitoring network historical data trends both
spatially and temporally. The MAROS Overview Statistics are the foundation the user
needs to make informed optimization decisions at the site. The Overview Statistics are
designed to allow site personnel to develop a better understanding of the plume
behavior over time and understand how the individual well concentration trends are
spatially distributed within the plume. This step allows the user to gain information that
will support a more informed decision to be made in the next level or detailed statistics
optimization analysis.
6.1 Mann-Kendall Analysis
The Mann-Kendall test is a statistical procedure that is well suited for analyzing trends in
data over time. The Mann-Kendall test can be viewed as a non-parametric test for zero
slope of the first-order regression of time-ordered concentration data versus time. One
advantage of the Mann-Kendall test is that it does not require any assumptions as to the
statistical distribution of the data (e.g. normal, lognormal, etc.) and can be used with data
sets which include irregular sampling intervals and missing data. The Mann-Kendall test
is designed for analyzing a single groundwater constituent, multiple constituents are
analyzed separately. The Mann-Kendall S statistic measures the trend in the data:
positive values indicate an increase in concentrations over time and negative values
indicate a decrease in concentrations over time. The strength of the trend is proportional
to the magnitude of the Mann-Kendall statistic (i.e., a large value indicates a strong
trend). The confidence in the trend is determined by consulting the S statistic and the
sample size, n, in a Kendall probability table such as the one reported in Hollander and
Wolfe (1973).
The concentration trend is determined for each well and each COC based on results of
the S statistic, the confidence in the trend, and the Coefficient of Variation (COV). The
decision matrix for this evaluation is shown in Table 3. A Mann-Kendall statistic that is
greater than 0 combined with a confidence of greater than 95% is categorized as an
Increasing trend while a Mann-Kendall statistic of less than 0 with a confidence between
90% and 95% is defined as a probably Increasing trend, and so on.
Depending on statistical indicators, the concentration trend is classified into six
categories:
• Decreasing (D),
• Probably Decreasing (PD),
• Stable (S),
• No Trend (NT),
• Probably Increasing (PI)
• Increasing (I).
Appendix A 4 MAROS 2.2 Methodology
-------
These trend estimates are then analyzed to identify the source and tail region overall
stability category (see Figure 2 for further details).
6.2 Linear Regression Analysis
Linear Regression is a parametric statistical procedure that is typically used for
analyzing trends in data over time. Using this type of analysis, a higher degree of
scatter simply corresponds to a wider confidence interval about the average log-slope.
Assuming the sign (i.e., positive or negative) of the estimated log-slope is correct, a level
of confidence that the slope is not zero can be easily determined. Thus, despite a poor
goodness of fit, the overall trend in the data may still be ascertained, where low levels of
confidence correspond to "Stable" or "No Trend" conditions (depending on the degree of
scatter) and higher levels of confidence indicate the stronger likelihood of a trend. The
linear regression analysis is based on the first-order linear regression of the log-
transformed concentration data versus time. The slope obtained from this log-
transformed regression, the confidence level for this log-slope, and the COV of the
untransformed data are used to determine the concentration trend. The decision matrix
for this evaluation is shown in Table 4.
To estimate the confidence in the log-slope, the standard error of the log-slope is
calculated. The coefficient of variation, defined as the standard deviation divided by the
average, is used as a secondary measure of scatter to distinguish between "Stable" or
"No Trend" conditions for negative slopes. The Linear Regression Analysis is designed
for analyzing a single groundwater constituent; multiple constituents are analyzed
separately, (up to five COCs simultaneously). For this evaluation, a decision matrix
developed by Groundwater Services, Inc. is also used to determine the "Concentration
Trend" category (plume stability) for each well.
Depending on statistical indicators, the concentration trend is classified into six
categories:
Decreasing (D),
Probably Decreasing (PD),
Stable (S),
No Trend (NT),
Probably Increasing (PI)
Increasing (I).
The resulting confidence in the trend, together with the log-slope and the COV of the
untransformed data, are used in the linear regression analysis decision matrix to
determine the concentration trend. For example, a positive log-slope with a confidence
of less than 90% is categorized as having No Trend whereas a negative log-slope is
considered Stable if the COV is less than 1 and categorized as No Trend if the COV is
greater than 1.
6.3 Overall Plume Analysis
General recommendations for the monitoring network frequency and density are
suggested based on heuristic rules applied to the source and tail trend results.
Appendix A 5 MAROS 2.2 Methodology
-------
Individual well trend results are consolidated and weighted by the MAROS according to
user input, and the direction and strength of contaminant concentration trends in the
source zone and tail zone for each COC are determined. Based on
i) the consolidated trend analysis,
ii) hydrogeologic factors (e.g., seepage velocity), and
iii) location of potential receptors (e.g., wells, discharge points, or property
boundaries),
the software suggests a general optimization plan for the current monitoring system in
order to efficiently but effectively monitor groundwater in the future. A flow chart utilizing
the trend analysis results and other site-specific parameters to form a general sampling
frequency and well density recommendation is outlined in Figure 2. For example, a
generic plan for a shrinking petroleum hydrocarbon plume (BTEX) in a slow
hydrogeologic environment (silt) with no nearby receptors would entail minimal, low
frequency sampling of just a few indicators. On the other hand, the generic plan for a
chlorinated solvent plume in a fast hydrogeologic environment that is expanding but has
very erratic concentrations over time would entail more extensive, higher frequency
sampling. The generic plan is based on a heuristically derived algorithm for assessing
future sampling duration, location and density that takes into consideration plume
stability. For a detailed description of the heuristic rules used in the MAROS software,
refer to the MAROS 2.2Manual (AFCEE, 2003).
6.4 Moment Analysis
An analysis of moments can help resolve plume trends, where the zeroth moment shows
change in dissolved mass vs. time, the first moment shows the center of mass location
vs. time, and the second moment shows the spread of the plume vs. time. Moment
calculations can predict how the plume will change in the future if further statistical
analysis is applied to the moments to identify a trend (in this case, Mann Kendall Trend
Analysis is applied). The trend analysis of moments can be summarized as:
• Zeroth Moment: An estimate of the total mass of the constituent for each sample
event
• First Moment: An estimate of the center of mass for each sample event
• Second Moment: An estimate of the spread of the plume around the center of
mass
The role of moment analysis in MAROS is to provide a relative estimate of plume
stability and condition within the context of results from other MAROS modules. The
Moment analysis algorithms in MAROS are simple approximations of complex
calculations and are meant to estimate changes in total mass, center of mass and
spread of mass for complex well networks. The Moment Analysis module is sensitive to
the number and arrangement of wells in each sampling event, so, changes in the
number and identity of wells during monitoring events, and the parameters chosen for
data consolidation can cause changes in the estimated moments.
Plume stability may vary by constituent, therefore the MAROS Moment analysis can be
used to evaluate multiple COCs simultaneously which can be used to provide a quick
way of comparing individual plume parameters to determine the size and movement of
constituents relative to one another. Moment analysis in the MAROS software can also
Appendix A 6 MAROS 2.2 Methodology
-------
be used to assist the user in evaluating the impact on plume delineation in future
sampling events by removing identified "redundant" wells from a long-term monitoring
program (this analysis was not performed as part of this study, for more details on this
application of moment analysis refer to the MAROS Users Manual (AFCEE, 2003)).
The zeroth moment is the sum of concentrations for all monitoring wells and is a mass
estimate. The zeroth moment calculation can show high variability over time, largely due
to the fluctuating concentrations at the most contaminated wells as well as varying
monitoring well network. Plume analysis and delineation based exclusively on
concentration can exhibit fluctuating temporal and spatial values. The mass estimate is
also sensitive to the extent of the site monitoring well network over time. The zeroth
moment trend over time is determined by using the Mann-Kendall Trend Methodology.
The zeroth Moment trend test allows the user to understand how the plume mass has
changed over time. Results for the trend include: Increasing, probably Increasing, no
trend, stable, probably decreasing, decreasing or not applicable (N/A) (Insufficient Data).
When considering the results of the zeroth moment trend, the following factors should be
considered which could effect the calculation and interpretation of the plume mass over
time: 1) Change in the spatial distribution of the wells sampled historically 2) Different
wells sampled within the well network over time (addition and subtraction of well within
the network). 3) Adequate versus inadequate delineation of the plume over time
The first moment estimates the center of mass, coordinates (Xc and Yc) for each
sample event and COC. The changing center of mass locations indicate the movement
of the center of mass over time. Whereas, the distance from the original source location
to the center of mass locations indicate the movement of the center of mass over time
relative to the original source. Calculation of the first moment normalizes the spread by
the concentration indicating the center of mass. The first moment trend of the distance to
the center of mass over time shows movement of the plume in relation to the original
source location over time. Analysis of the movement of mass should be viewed as it
relates to 1) the original source location of contamination 2) the direction of groundwater
flow and/or 3) source removal or remediation. Spatial and temporal trends in the center
of mass can indicate spreading or shrinking or transient movement based on season
variation in rainfall or other hydraulic considerations. No appreciable movement or a
neutral trend in the center of mass would indicate plume stability. However, changes in
the first moment over time do not necessarily completely characterize the changes in the
concentration distribution (and the mass) over time. Therefore, in order to fully
characterize the plume the First Moment trend should be compared to the zeroth
moment trend (mass change over time).
The second moment indicates the spread of the contaminant about the center of mass
(Sxx and Syy), or the distance of contamination from the center of mass for a particular
COC and sample event. The Second Moment represents the spread of the plume over
time in both the x and y directions. The Second Moment trend indicates the spread of
the plume about the center of mass. Analysis of the spread of the plume should be
viewed as it relates to the direction of groundwater flow. An Increasing trend in the
second moment indicates an expanding plume, whereas a declining trend in the second
moment indicates a shrinking plume. No appreciable movement or a neutral trend in the
center of mass would indicate plume stability. The second moment provides a measure
of the spread of the concentration distribution about the plume's center of mass.
Appendix A 7 MAROS 2.2 Methodology
-------
However, changes in the second moment over time do not necessarily completely
characterize the changes in the concentration distribution (and the mass) over time.
Therefore, in order to fully characterize the plume the Second Moment trend should be
compared to the zeroth moment trend (mass change over time).
7.0 Detailed Statistics: Optimization Analysis
Although the overall plume analysis shows a general recommendation regarding
sampling frequency reduction and a general sampling density, a more detailed analysis
is also available with the MAROS 2.2 software in order to allow for further reductions on
a well-by-well basis for frequency, well redundancy, well sufficiency and sampling
sufficiency. The MAROS Detailed Statistics allows for a quantitative analysis for spatial
and temporal optimization of the well network on a well-by-well basis. The results from
the Overview Statistics should be considered along with the MAROS optimization
recommendations gained from the Detailed Statistical Analysis described previously.
The MAROS Detailed Statistics results should be reassessed in view of site knowledge
and regulatory requirements as well as in consideration of the Overview Statistics
(Figure 2).
The Detailed Statistics or Sampling Optimization MAROS modules can be used to
determine the minimal number of sampling locations and the lowest frequency of
sampling that can still meet the requirements of sampling spatially and temporally for an
existing monitoring program. It also provides an analysis of the sufficiency of data for
the monitoring program.
Sampling optimization in MAROS consists of four parts:
• Well redundancy analysis using the Delaunay method
• Well sufficiency analysis using the Delaunay method
• Sampling frequency determination using the Modified CES method
• Data sufficiency analysis using statistical power analysis.
The well redundancy analysis using the Delaunay method identifies and eliminates
redundant locations from the monitoring network. The well sufficiency analysis can
determine the areas where new sampling locations might be needed. The Modified CES
method determines the optimal sampling frequency for a sampling location based on the
direction, magnitude, and uncertainty in its concentration trend. The data sufficiency
analysis examines the risk-based site cleanup status and power and expected sample
size associated with the cleanup status evaluation.
7.1 Well Redundancy Analysis - Delaunay Method
The well redundancy analysis using the Delaunay method is designed to select the
minimum number of sampling locations based on the spatial analysis of the relative
importance of each sampling location in the monitoring network. The approach allows
elimination of sampling locations that have little impact on the historical characterization
of a contaminant plume. An extended method or wells sufficiency analysis, based on
the Delaunay method, can also be used for recommending new sampling locations.
Appendix A g MAROS 2.2 Methodology
-------
Details about the Delaunay method can be found in Appendix A.2 of the MAROS Manual
(AFCEE, 2003).
Sampling Location determination uses the Delaunay triangulation method to determine
the significance of the current sampling locations relative to the overall monitoring
network. The Delaunay method calculates the network Area and Average concentration
of the plume using data from multiple monitoring wells. A slope factor (SF) is calculated
for each well to indicate the significance of this well in the system (i.e. how removing a
well changes the average concentration.)
The Sampling Location optimization process is performed in a stepwise fashion. Step
one involves assessing the significance of the well in the system, if a well has a small SF
(little significance to the network), the well may be removed from the monitoring network.
Step two involves evaluating the information loss of removing a well from the network. If
one well has a small SF, it may or may not be eliminated depending on whether the
information loss is significant. If the information loss is not significant, the well can be
eliminated from the monitoring network and the process of optimization continues with
fewer wells. However if the well information loss is significant then the optimization
terminates. This sampling optimization process allows the user to assess "redundant"
wells that will not incur significant information loss on a constituent-by-constituent basis
for individual sampling events.
7.2 Well Sufficiency Analysis - Delaunay Method
The well sufficiency analysis, using the Delaunay method, is designed to recommend
new sampling locations in areas within the existing monitoring network where there is a
high level of uncertainty in contaminant concentration. Details about the well sufficiency
analysis can be found in Appendix A.2 of the MAROS Manual (AFCEE, 2003).
In many cases, new sampling locations need to be added to the existing network to
enhance the spatial plume characterization. If the MAROS algorithm calculates a high
level of uncertainty in predicting the constituent concentration for a particular area, a new
sampling location is recommended. The Slope Factor (SF) values obtained from the
redundancy evaluation described above are used to calculate the concentration
estimation error for each triangle area formed in the Delaunay triangulation. The
estimated SF value for each area is then classified into four levels: Small, Moderate,
Large, or Extremely large (S, M, L, E) because the larger the estimated SF value, the
higher the estimation error at this area. Therefore, the triangular areas with the
estimated SF value at the Extremely large or Large level can be candidate regions for
new sampling locations.
The results from the Delaunay method and the method for determining new sampling
locations are derived solely from the spatial configuration of the monitoring network and
the spatial pattern of the contaminant plume. No parameters such as the hydrogeologic
conditions are considered in the analysis. Therefore, professional judgment and
regulatory considerations must be used to make final decisions.
7.3 Sampling Frequency Determination - Modified CES Method
Appendix A g MAROS 2.2 Methodology
-------
The Modified CES method optimizes sampling frequency for each sampling location
based on the magnitude, direction, and uncertainty of its concentration trend derived
from its recent and historical monitoring records. The Modified Cost Effective Sampling
(MCES) estimates a conservative lowest-frequency sampling schedule for a given
groundwater monitoring location that still provides needed information for regulatory and
remedial decision-making. The MCES method was developed on the basis of the Cost
Effective Sampling (CES) method developed by Ridley et al (1995). Details about the
MCES method can be found in Appendix A.9 of the MAROS Manual (AFCEE, 2003).
In order to estimate the least frequent sampling schedule for a monitoring location that
still provides enough information for regulatory and remedial decision-making, MCES
employs three steps to determine the sampling frequency. The first step involves
analyzing frequency based on recent trends. A preliminary location sampling frequency
(PLSF) is developed based on the rate of change of well concentrations calculated by
linear regression along with the Mann-Kendall trend analysis of the most recent
monitoring data (see Figure 3). The variability within the sequential sampling data is
accounted for by the Mann-Kendall analysis. The rate of change vs. trend result matrix
categorizes wells as requiring annual, semi-annual or quarterly sampling. The PLSF is
then reevaluated and adjusted based on overall trends. If the long-term history of
change is significantly greater than the recent trend, the frequency may be reduced by
one level.
The final step in the analysis involves reducing frequency based on risk, site-specific
conditions, regulatory requirements or other external issues. Since not all compounds in
the target being assessed are equally harmful, frequency is reduced by one level if
recent maximum concentration for a compound of high risk is less than 1/2 of the
Maximum Concentration Limit (MCL). The result of applying this method is a suggested
sampling frequency based on recent sampling data trends and overall sampling data
trends and expert judgment.
The final sampling frequency determined from the MCES method can be Quarterly,
Semiannual, Annual, or Biennial. Users can further reduce the sampling frequency to,
for example, once every three years, if the trend estimated from Biennial data (i.e., data
drawn once every two years from the original data) is the same as that estimated from
the original data.
7.4 Data Sufficiency Analysis - Power Analysis
The MAROS Data Sufficiency module employs simple statistical methods to evaluate
whether the collected data are adequate both in quantity and in quality for revealing
changes in constituent concentrations. The first section of the module evaluates
individual well concentrations to determine if they are statistically below a target
screening level. The second section includes a simple calculation for estimating
projected groundwater concentrations at a specified point downgradient of the plume. A
statistical Power analysis is then applied to the projected concentrations to determine if
the downgradient concentrations are statistically below the cleanup standard. If the
number of projected concentrations is below the level to provide statistical significance,
then the number of sample events required to statistically confirm concentrations below
standards is estimated from the Power analysis.
Appendix A 70 MAROS 2.2 Methodology
-------
Before testing the cleanup status for individual wells, the stability or trend of the
contaminant plume should be evaluated. Only after the plume has reached stability or is
reliably diminishing can we conduct a test to examine the cleanup status of wells.
Applying the analysis to wells in an expanding plume may cause incorrect conclusions
and is less meaningful.
Statistical power analysis is a technique for interpreting the results of statistical tests.
The Power of a statistical test is a measure of the ability of the test to detect an effect
given that the effect actually exists. The method provides additional information about a
statistical test: 1) the power of the statistical test, i.e., the probability of finding a
difference in the variable of interest when a difference truly exists; and 2) the expected
sample size of a future sampling plan given the minimum detectable difference it is
supposed to detect. For example, if the mean concentration is lower than the cleanup
goal but a statistical test cannot prove this, the power and expected sample size can tell
the reason and how many more samples are needed to result in a significant test. The
additional samples can be obtained by a longer period of sampling or an increased
sampling frequency. Details about the data sufficiency analysis can be found in
Appendix A.6 of the MAROS Manual (AFCEE, 2003).
When applying the MAROS power analysis method, a hypothetical statistical compliance
boundary (HSCB) is assigned to be a line perpendicular to the groundwater flow
direction (see figure below). Monitoring well concentrations are projected onto the
HSCB using the distance from each well to the compliance boundary along with a decay
coefficient. The projected concentrations from each well and each sampling event are
then used in the risk-based power analysis. Since there may be more than one sampling
event selected by the user, the risk-based power analysis results are given on an event-
by-event basis. This power analysis can then indicate if target are statistically achieved
at the HSCB. For instance, at a site where the historical monitoring record is short with
few wells, the HSCB would be distant; whereas, at a site with longer duration of
sampling with many wells, the HSCB would be close. Ultimately, at a site the goal would
be to have the HSCB coincide with or be within the actual compliance boundary
(typically the site property line).
Appendix A 77 MAROS 2.2 Methodology
-------
Concentrations
projected to this
•""* line
The nearest
downgradient
receptor
Groundwater flow direction
In order to perform a risk-based cleanup status evaluation for the whole site, a strategy
was developed as follows.
• Estimate concentration versus distance decay coefficient from plume centerline
wells.
• Extrapolate concentration versus distance for each well using this decay
coefficient.
• Comparing the extrapolated concentrations with the compliance concentration
using power analysis.
Results from this analysis can be Attained or Not Attained, providing a statistical
interpretation of whether the cleanup goal has been met on the site-scale from the risk-
based point of view. The results as a function of time can be used to evaluate if the
monitoring system has enough power at each step in the sampling record to indicate
certainty of compliance by the plume location and condition relative to the compliance
boundary. For example, if results are Not Attained at early sampling events but are
Attained in recent sampling events, it indicates that the recent sampling record provides
a powerful enough result to indicate compliance of the plume relative to the location of
the receptor or compliance boundary.
Appendix A
12
MAROS 2.2 Methodology
-------
CITED REFERENCES
AFCEE 2003. Monitoring and Remediation Optimization System (MAROS) 2.1 Software
Users Guide. Air Force Center for Environmental Excellence, http://www.gsi-
net.com/software/MAROS V2 1Manual.pdf
AFCEE. 1997. Air Force Center for Environmental Excellence, AFCEE Long-Term
Monitoring Optimization Guide, http://www.afcee.brooks.af.mil.
Aziz, J. A., C. J. Newell, M. Ling, H. S. Rifai and J. R. Gonzales (2003). "MAROS: A
Decision Support System for Optimizing Monitoring Plans." Ground Water 41(3): 355-
367.
Gilbert, R. O., 1987, Statistical Methods for Environmental Pollution Monitoring, Van
Nostrand Reinhold, New York, NY, ISBN 0-442-23050-8.
Hollander, M. and Wolfe, D. A. (1973). Nonparametric Statistical Methods, Wiley, New
York, NY.
Ridley, M.N. et al., 1995. Cost-Effective Sampling of Groundwater Monitoring Wells, the
Regents of UC/LLNL, Lawrence Livermore National Laboratory.
U.S. Environmental Protection Agency, 1992. Methods for Evaluating the Attainment of
Cleanup Standards Volume 2: Ground Water.
Weight, W. D. and J. L. Sonderegger (2001). Manual of Applied Field Hydrogeology.
New York, NY, McGraw-Hill.
Appendix A 73 MAROS 2.2 Methodology
-------
Mann-Kendall
Mann-Kendall
Statistic
S>0
S>0
S>0
S<0
S<0
S<0
S<0
TABLE 1
Analysis Decision Matrix
Confidence in the
Trend
> 95%
90 - 95%
< 90%
< 90% and COV > 1
< 90% and COV < 1
90 - 95%
> 95%
(Aziz, et. al., 2003)
Concentration Trend
Increasing
Probably Increasing
No Trend
No Trend
Stable
Probably Decreasing
Decreasing
Linear Regression
Confidence in the
Trend
< 90%
90 - 95%
> 95%
TABLE 2
Analysis Decision Matrix (Aziz, et. al., 2003)
Log-slope
Positive Negative
COV < 1 Stable
No Trend
COV > 1 No Trend
Probably Increasing Probably Decreasing
Increasing Decreasing
-------
MAROS: Decision Support Tool
MAROS is a collection of tools in one software package that is used in an explanatory, non-linear fashion. The tool
includes models, geostatistics, heuristic rules, and empirical relationships to assist the user in optimizing a
groundwater monitoring network system while maintaining adequate delineation of the plume as well as knowledge
of the plume state over time. Different users utilize the tool in different ways and interpret the results from a different
viewpoint.
T
Overview Statistics
What it is: Simple, qualitative and quantitative plume information can be gained through evaluation of monitoring
network historical data trends both spatially and temporally. The MAROS Overview Statistics are the foundation the
user needs to make informed optimization decisions at the site.
What it does: The Overview Statistics are designed to allow site personnel to develop a better understanding of the
plume behavior over time and understand how the individual well concentration trends are spatially distributed within
the plume. This step allows the user to gain information that will support a more informed decision to be made in the
next level of optimization analysis.
What are the tools: Overview Statistics includes two analytical tools:
1) Trend Analysis: includes Mann-Kendall and Linear Regression statistics for individual wells and results in
general heuristically-derived monitoring categories with a suggested sampling density and monitoring
frequency.
2) Moment Analysis: includes dissolved mass estimation (0th Moment), center of mass (1st Moment), and
plume spread (2nd Moment) over time. Trends of these moments show the user another piece of
information about the plume stability overtime.
What is the product: A first-cut blueprint for a future long-term monitoring program that is intended to be a
foundation for more detailed statistical analysis.
T
Detailed Statistics
What it is: The MAROS Detailed Statistics allows for a quantitative analysis for spatial and temporal optimization of
the well network on a well-by-well basis.
What it does: The results from the Overview Statistics should be considered along side the MAROS optimization
recommendations gained from the Detailed Statistical Analysis. The MAROS Detailed Statistics results should be
reassessed in view of site knowledge and regulatory requirements as well as the Overview Statistics.
What are the tools: Detailed Statistics includes four analytical tools:
1) Sampling Frequency Optimization: uses the Modified CES method to establish a recommended future
sampling frequency.
2) Well Redundancy Analysis: uses the Delaunay Method to evaluate if any wells within the monitoring
network are redundant and can be eliminated without any significant loss of plume information.
3) Well Sufficiency Analysis: uses the Delaunay Method to evaluate areas where new wells are
recommended within the monitoring network due to high levels of concentration uncertainty.
4) Data Sufficiency Analysis: uses Power Analysis to assess if the historical monitoring data record has
sufficient power to accurately reflect the location of the plume relative to the nearest receptor or
compliance point.
What is the product: List of wells to remove from the monitoring program, locations where monitoring wells may
need to be added, recommended frequency of sampling for each well, analysis if the overall system is statistically
powerful to monitor the plume.
Figure 1. MAROS Decision Support Tool Flow Chart
-------
Select Representative Wells in "Source" and "Plume" Zone
Source Zone i Tail Zone
Identify Site Constituents of Concern (COCs).
(Assistance provided by software.)
Analyze Lines of Evidence (LOEs)
for Plume Stability (by well and by COC)
Categorize concentrations of COC in each well as:
• Increasing (I)
• Probably Increasing (PI)
• No Trend (NT)
• Stable (S)
• Probably Decreasing (PD)
for Each Well Based On All
LOE's
SOURCE PLUME
"Lump Lines of Evidence"
Determine General Trend for Source and
Tail Zones
Increasing (I)
Probably Increasing (PI)
No Trend (NT)
Stable (S)
Probably Decreasing (PD)
Decreasing (D)
"Lump Wells" in Source and Tail Zone
Determine
LTMP
Monitoring
Category
for COC By
Source / Tail
-------
Sampling
Frequency
Q: Quarterly
S: SemiAimual
A: Annual
0>
CE3
T3
I
PI
Rate of Change (Linear Regression)
High MH Medium LM Low
Figure 3. Decision Matrix for Determining Provisional Frequency (Figure A.3.1 of the
MAROS Manual (AFCEE 2003)
-------
GROUNDWATER MONITORING NETWORK OPTIMIZATION
FRONTIER HARD CHROME
Vancouver, Washington
APPENDIX B:
MAROS Reports
Zone A
Mann-Kendall Reports
Moment Reports
Zone B
Mann-Kendall Reports
Moment Reports
-------
MAROS Mann-Kendall Statistics Summary
Project: FHC
Location: Vancouver
User Name: MV
State: Washington
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Source/ Number of
Well Tail Samples
Number of
Detects
Coefficient
of Variation
Mann-Kendall
Statistic
Confidence
in Trend
All
Samples Concentration
"ND" ? Trend
CHROMIUM, TOTAL
RA-MW-11A
RA-MW-12A
RA-MW-15A
B85-4
W85-6A
RA-MW-13A
RA-MW-14A
RA-MW-1 7A
W99-R5A
W85-7A
W92-16A
W97-18A
W97-19A
W98-20A
W98-21A
RA-MW-1 6A
S
S
S
S
T
T
T
T
T
T
T
T
T
T
T
T
12
12
12
11
9
12
12
12
11
11
11
11
11
11
11
12
10
12
11
10
8
10
9
11
4
9
7
6
10
10
10
11
1.23
2.18
1.61
1.48
0.96
0.75
0.78
0.57
1.29
0.51
1.17
0.07
0.81
0.74
0.73
0.65
-43
-18
-2
-25
-6
-9
0
-17
0
-14
-5
-9
-21
-17
-21
-26
99.9%
87.5%
52.7%
97.0%
69.4%
70.4%
47.3%
86.0%
46.9%
84.0%
61 .9%
72.9%
94.0%
89.1%
94.0%
95.7%
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
D
NT
NT
D
S
S
S
S
NT
S
NT
S
PD
S
PD
D
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A)-
Due to insufficient Data (< 4 sampling events); Source/Tail (S/T)
The Number of Samples and Number of Detects shown above are post-consolidation values.
MAROS Version 2,.2 2006, AFCEE
Thursday, September 13, 2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: B85-4
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
O)
o
I
o
o
3.5E-02 -
3.0E-02 -
2.5E-02 -
2.0E-02 •
1.5E-02-
1.0E-02-
5.0E-03 -
n np4-nn .
*
•
* * *
* • * +
Mann Kendall S Statistic:
Confidence in
Trend:
I 97.0%
Coefficient of Variation:
1.48
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
B85-4
B85-4
B85-4
B85-4
B85-4
B85-4
B85-4
B85-4
B85-4
B85-4
B85-4
Well Type
s
s
s
s
s
s
s
s
s
s
s
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
3.8E-02
8.1E-03
3.7E-03
1.1E-03
2.7E-02
5.8E-03
9.0E-04
1.5E-03
5.0E-04 ND
2.8E-03
2.4E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-11A
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
5nc n9
.UE-U^ •
4.5E-02 -
2- 4.0E-02 -
|" 3.5E-02 -
r 3.0E-02 •
o
s 2.5E-02 •
i 2.0E-02 •
| 1.5E-02-
0 1.0E-02-
5.0E-03 -
Data Table:
Well Well Ty
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
RA-MW-11A S
Date
/ /VVVV* /VVV* /,/
^^^>O^'i
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-12A
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
6npi.nn
.utruu •
_ 5.0&-00 -
£ 4.0&-00 -
o
s 3.0&00 •
§ 2.0&-00 -
o
0 1.0&-00 -
Oncu-nn
m\ICr\I\I
Data Table:
Well Well Ty
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
RA-MW-12A S
Date
/ /vvvv* //*
^*^*
•
•
•
* * •* * + *
Effective
Pe Date Constituent
10/15/2003 CHROMIUM, TOTAL
2/10/2004 CHROMIUM, TOTAL
4/5/2004 CHROMIUM, TOTAL
8/15/2004 CHROMIUM, TOTAL
5/5/2005 CHROMIUM, TOTAL
12/12/2005 CHROMIUM, TOTAL
3/8/2006 CHROMIUM, TOTAL
6/15/2006 CHROMIUM, TOTAL
9/25/2006 CHROMIUM, TOTAL
12/15/2006 CHROMIUM, TOTAL
3/30/2007 CHROMIUM, TOTAL
6/5/2007 CHROMIUM, TOTAL
/vv\/
<) <$• ^
* + *
Result (mg/L) Flag
6.4E-01
1.8E-01
1.4E-01
8.0E-02
1.1E-01
1.3E+00
8.5E-02
1.3E-01
5.3E+00
8.1E-02
7.9E-02
1.1E-01
Mann Kendall S Statistic:
I ^?8
Confidence in
Trend:
1 87.5%
Coefficient of Variation:
I 2'18
Mann Kendall
Concentration Trend:
(See Note)
I NT
Number of Number of
Samples Detects
2 2
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-13A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
5nc ni
.uc-uo •
4.5E-03 -
;;[• 4.0E-03 -
|" 3.5E-03 -
r 3.0E-03 •
o
s 2.5E-03 •
i 2.0E-03 •
| 1.5E-03-
0 1.0E-03-
5.0E-04 -
Data Table:
Well Well Ty
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
RA-MW-13A T
Date
/ /vvvv* /vt
^*^*
• • *
* * *
Effective
Pe Date Constituent
10/15/2003 CHROMIUM, TOTAL
2/10/2004 CHROMIUM, TOTAL
4/5/2004 CHROMIUM, TOTAL
8/15/2004 CHROMIUM, TOTAL
5/5/2005 CHROMIUM, TOTAL
12/12/2005 CHROMIUM, TOTAL
3/8/2006 CHROMIUM, TOTAL
6/15/2006 CHROMIUM, TOTAL
9/25/2006 CHROMIUM, TOTAL
12/15/2006 CHROMIUM, TOTAL
3/30/2007 CHROMIUM, TOTAL
6/5/2007 CHROMIUM, TOTAL
/vv\/
<) <$• ^
•
•
•
Result (mg/L) Flag
1.3E-03
4.4E-03
5.0E-04 ND
1.3E-03
5.6E-04
2.0E-03
1.9E-03
1.5E-03
6.3E-04
5.0E-04 ND
1.4E-03
1.1E-03
Mann Kendall S Statistic:
I ^9
Confidence in
Trend:
1 70.4%
Coefficient of Variation:
I °75
Mann Kendall
Concentration Trend:
(See Note)
I S
Number of Number of
Samples Detects
2 1
1 1
1 0
1 1
1 1
1 1
1 1
1 1
1 1
1 0
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-14A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
, — ,
B>
o
1
Concent
6.UE-03 -
5.0E-03 -
4.0E-03 -
3.0E-03 •
2.0E-03 -
1.0E-03-
n np4-nn .
•
•
' • •. '•
•
Mann Kendall S Statistic:
Confidence in
Trend:
I 47.3%
Coefficient of Variation:
0.78
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
RA-MW-14A
Well Type
T
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
10/15/2003
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
5.0E-04 ND
5.4E-03
5.0E-04 ND
2.0E-03
7.3E-04
3.2E-03
1.8E-03
1.8E-03
1.4E-03
5.0E-04 ND
2.2E-03
1.6E-03
I S
Number of Number of
Samples Detects
2 0
1 1
1 0
1 1
1 1
1 1
1 1
1 1
1 1
1 0
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-15A
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
A np no
4.UC-U^ •
3.5E-02 -
^ 3.0E-02 •
~ 2.5E-02 -
o
s 2.0E-02 •
| 1.5E-02-
o 1.0E-02-
O
5.0E-03 -
Data Table:
Well Well Ty
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
RA-MW-15A S
Date
/ /vvvv* /vvv* /,/
^^<^^><^'i
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-16A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
1 HP no
I .UC-U^ •
9.0E-03 -
^ 8.0E-03 -
|" 7.0E-03 -
r 6.0E-03 •
o
s 5.0E-03 •
i 4.0E-03 •
| 3.0E-03 •
0 2.0E-03 •
1.0E-03-
Data Table:
Well Well Ty
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
RA-MW-16A T
Date
/ /vvvv* /v%
^*^*
•
•
•
*
Effective
Pe Date Constituent
10/15/2003 CHROMIUM, TOTAL
2/10/2004 CHROMIUM, TOTAL
4/5/2004 CHROMIUM, TOTAL
8/15/2004 CHROMIUM, TOTAL
5/5/2005 CHROMIUM, TOTAL
12/12/2005 CHROMIUM, TOTAL
3/8/2006 CHROMIUM, TOTAL
6/15/2006 CHROMIUM, TOTAL
9/25/2006 CHROMIUM, TOTAL
12/15/2006 CHROMIUM, TOTAL
3/30/2007 CHROMIUM, TOTAL
6/5/2007 CHROMIUM, TOTAL
>*vv\/
<) <$• ^
* •
Result (mg/L) Flag
4.8E-03
9.2E-03
2.0E-03
3.5E-03
2.2E-03
4.1E-03
3.7E-03
2.8E-03
1.7E-03
5.0E-04 ND
2.9E-03
2.6E-03
Mann Kendall S Statistic:
_
Confidence in
Trend:
1 95.7%
Coefficient of Variation:
1 0.65
Mann Kendall
Concentration Trend:
(See Note)
I °
Number of Number of
Samples Detects
2 2
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 0
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-17A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
B)
o
1
c
o
o
1.2E-U2 -
1.0E-02-
8.0E-03 -
6.0E-03 •
4.0E-03 -
2.0E-03 -
n np4-nn .
•
•
•
•
* * *
•
•
*
Mann Kendall S Statistic:
Confidence in
Trend:
I 86.0%
Coefficient of Variation:
0.57
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
RA-MW-17A
Well Type
T
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
10/15/2003
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
6.3E-03
1.0E-02
2.6E-03
5.0E-03
9.2E-04
7.6E-03
8.6E-03
5.7E-03
4.0E-03
5.0E-04 ND
5.0E-03
4.9E-03
I S
Number of Number of
Samples Detects
2 2
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 0
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W85-6A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
£
_J
B)
o
1
Concent
1.4E-02-
1.2E-02-
1.0E-02-
8.0E-03 •
6.0E-03 -
4.0E-03 •
2.0E-03 -
n np4-nn .
*
*
* * * * *
Mann Kendall S Statistic:
Confidence in
Trend:
I 69.4%
Coefficient of Variation:
0.96
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W85-6A
W85-6A
W85-6A
W85-6A
W85-6A
W85-6A
W85-6A
W85-6A
W85-6A
Well Type
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
1.4E-03
1.4E-02
9.1E-03
2.9E-03
2.2E-03
4.1E-03
5.0E-04 ND
3.4E-03
3.2E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W85-7A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
_J
O)
c
O
1
1
O
O
3.5E-03 -
3.0E-03 -
2.5E-03 -
2.0E-03 •
1.5E-03-
1.0E-03-
5.0E-04 -
n np4-nn .
»
* *
^
* * •
* * *
» »
Mann Kendall S Statistic:
Confidence in
Trend:
I 84.0%
Coefficient of Variation:
0.51
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W85-7A
W85-7A
W85-7A
W85-7A
W85-7A
W85-7A
W85-7A
W85-7A
W85-7A
W85-7A
W85-7A
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
1.7E-03
5.0E-04 ND
3.6E-03
2.8E-03
1.9E-03
1.7E-03
1.5E-03
1.6E-03
5.0E-04 ND
2.7E-03
1.5E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
0
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W92-16A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
_J
B)
o
1
Concer
6.0E-03 •
5.0E-03 •
4.0E-03 -
3.0E-03 -
2.0E-03 -
1.0E-03-
n np4-nn .
•
•
•
* * * * * *
Mann Kendall S Statistic:
Confidence in
Trend:
I 61.9%
Coefficient of Variation:
1.17
Mann Kendall
Concentration Trend:
(See Note)
[ NT
Data Table:
Well
W92-16A
W92-16A
W92-16A
W92-16A
W92-16A
W92-16A
W92-16A
W92-16A
W92-16A
W92-16A
W92-16A
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L)
4.2E-03
5.0E-04
6.3E-03
7.0E-04
5.0E-04
5.0E-04
1.1E-03
2.1E-03
5.0E-04
5.6E-04
9.4E-04
Flag
ND
ND
ND
ND
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
0
1
1
0
0
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W97-18A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
Mann Kendall S Statistic:
_j
B>
o
1
Concen
6.0E-04 -
5.8E-04 -
5.6E-04 -
5.4E-04 •
5.2E-04 •
5.0E-04 -
4.8E-04 -
4.6E-04 -
A AF-flA .
*
• •
* *
* * * * * *
Confidence in
Trend:
I 72.9%
Coefficient of Variation:
0.07
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W97-18A
W97-18A
W97-18A
W97-18A
W97-18A
W97-18A
W97-18A
W97-18A
W97-18A
W97-18A
W97-18A
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L)
5.6E-04
5.0E-04
5.0E-04
5.0E-04
5.6E-04
5.3E-04
6.0E-04
5.3E-04
5.0E-04
5.0E-04
5.0E-04
Flag
ND
ND
ND
ND
ND
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
0
0
1
1
1
1
1
0
0
0
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W97-19A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
_j
1
o
1
Concen
8.0E-03 •
7.0E-03 -
6.0E-03 -
5.0E-03 •
4.0E-03 •
3.0E-03 -
2.0E-03 -
1.0E-03-
n np4-nn .
^
*
• * * *
* » »
Mann Kendall S Statistic:
Confidence in
Trend:
I 94.0%
Coefficient of Variation:
0.81
Mann Kendall
Concentration Trend:
(See Note)
[ PD
Data Table:
Well
W97-19A
W97-19A
W97-19A
W97-19A
W97-19A
W97-19A
W97-19A
W97-19A
W97-19A
W97-19A
W97-19A
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
2.2E-03
7.9E-03
5.4E-03
3.7E-03
1.4E-03
1.2E-03
1.2E-03
2.1E-03
5.0E-04 ND
2.0E-03
2.2E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W98-20A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
«° ^
O)
o
1
Concent
5.0E-03 -
4.0E-03 -
3.0E-03 •
2.0E-03 -
1.0E-03-
n np4-nn .
* *
•
•
* * *
• • •
•
Mann Kendall S Statistic:
Confidence in
Trend:
I 89.1%
Coefficient of Variation:
0.74
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W98-20A
W98-20A
W98-20A
W98-20A
W98-20A
W98-20A
W98-20A
W98-20A
W98-20A
W98-20A
W98-20A
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
2.0E-03
4.8E-03
5.1E-03
1.7E-03
1.0E-03
1.5E-03
1.0E-03
1.0E-03
5.0E-04 ND
1.7E-03
2.3E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W98-21A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
_j
B>
o
1
Concent
7.0E-03 -
6.0E-03 -
5.0E-03 -
4.0E-03 •
3.0E-03 -
2.0E-03 •
1.0E-03-
n np4-nn .
*
*
* *
* *
*
Mann Kendall S Statistic:
Confidence in
Trend:
I 94.0%
Coefficient of Variation:
0.73
Mann Kendall
Concentration Trend:
(See Note)
[ PD
Data Table:
Well
W98-21A
W98-21A
W98-21A
W98-21A
W98-21A
W98-21A
W98-21A
W98-21A
W98-21A
W98-21A
W98-21A
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
1.7E-03
7.1E-03
4.9E-03
2.1E-03
2.8E-03
1.9E-03
1.2E-03
2.5E-03
5.0E-04 ND
1.7E-03
1.9E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W99-R5A
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Median
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
_j
1
o
1
Concent
4.0E-03 -
3.5E-03 -
3.0E-03 -
2.5E-03 •
2.0E-03 •
1.5E-03-
1.0E-03-
5.0E-04 -
n np4-nn .
*
^ *** *****
Mann Kendall S Statistic:
Confidence in
Trend:
I 46.9%
Coefficient of Variation:
1.29
Mann Kendall
Concentration Trend:
(See Note)
[ NT
Data Table:
Well
W99-R5A
W99-R5A
W99-R5A
W99-R5A
W99-R5A
W99-R5A
W99-R5A
W99-R5A
W99-R5A
W99-R5A
W99-R5A
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L)
4.1E-04
4.1E-03
5.0E-04
5.0E-04
5.0E-04
7.0E-04
5.0E-04
5.5E-04
5.0E-04
5.0E-04
5.0E-04
Flag
ND
ND
ND
ND
ND
ND
ND
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
0
0
0
1
0
1
0
0
0
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/13/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Project: Frontier Hard Chrome
Location: Vancouver
User Name: MV
State: Washington
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Source/ Number of
Well Tail Samples
Number of
Detects
Coefficient
of Variation
Mann-Kendall
Statistic
Confidence
in Trend
All
Samples Concentration
"ND" ? Trend
CHROMIUM, TOTAL
RA-MW-15B
B87-8
RA-MW-11B
W92-16B
RA-MW-16B
RA-MW-12B
RA-MW-12C
RA-MW-13B
B85-3
RA-MW-14B
W99-R5B
W85-7B
W97-18B
W97-19B
W98-21B
RA-MW-13C
S
S
S
S
S
T
T
T
T
T
T
T
T
T
T
T
12
11
12
12
12
12
12
12
11
12
11
11
11
11
11
11
12
11
11
12
12
10
12
6
8
9
10
4
8
9
10
10
0.94
1.10
1.62
1.47
1.45
1.09
0.70
1.32
0.80
1.21
0.74
1.58
0.39
0.99
0.62
0.85
4
2
-48
2
8
-28
-4
-13
2
-6
-33
-34
12
-24
-34
-6
58.0%
53.0%
100.0%
52.7%
68.1%
96.9%
58.0%
79.0%
53.0%
63.1%
99.5%
99.6%
79.9%
96.4%
99.6%
64.8%
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
NT
NT
D
NT
NT
D
S
NT
NT
NT
D
D
NT
D
D
S
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A)-
Due to insufficient Data (< 4 sampling events); Source/Tail (S/T)
The Number of Samples and Number of Detects shown above are post-consolidation values.
MAROS Version 2,.2 2006, AFCEE
Thursday, September 06, 2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: B85-3
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
j"
1
o
«
>I
c
S
c
o
o
6.0E-03 •
5.0E-03 •
4.0E-03 -
3.0E-03 -
2.0E-03 -
1.0E-03-
n np4-nn .
*
* *
*
*
» ^
* * *
Mann Kendall S Statistic:
Confidence in
Trend:
I 53.0%
Coefficient of Variation:
0.80
Mann Kendall
Concentration Trend:
(See Note)
[ NT
Data Table:
Well
B85-3
B85-3
B85-3
B85-3
B85-3
B85-3
B85-3
B85-3
B85-3
B85-3
B85-3
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
5.0E-03
5.0E-04 ND
5.0E-04 ND
1.1E-03
6.3E-03
4.9E-03
5.4E-03
9.0E-04
5.0E-04 ND
2.5E-03
3.6E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
0
0
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: B87-8
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
«° *s
_J
B)
_£
c
~
k.
C
c
o
o
2.5E-01 -
2.0E-01 -
1.5E-01 •
1.0E-01 -
5.0E-02 -
n np4-nn .
•
•
•
* *
» *****
Mann Kendall S Statistic:
Confidence in
Trend:
I 53.0%
Coefficient of Variation:
1.10
Mann Kendall
Concentration Trend:
(See Note)
[ NT
Data Table:
Well
B87-8
B87-8
B87-8
B87-8
B87-8
B87-8
B87-8
B87-8
B87-8
B87-8
B87-8
Well Type
s
s
s
s
s
s
s
s
s
s
s
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
1.8E-02
2.4E-01
7.1E-02
1.9E-02
3.1E-02
5.0E-02
2.2E-02
4.6E-02
3.1E-02
2.0E-02
1.3E-01
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
1
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-11B
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
7nc n9
.UE-U^ •
6.0E-02 •
^j
1 5.0E-02 •
§ 4.0E-02 -
S 3.0E-02 -
c
01
c 2.0E-02 -
o
O
1.0E-02-
Data Table:
Well Well Ty
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
RA-MW-1 1B S
Date
/ /vvvv* /vvv* /,/
^^<^^><^'i
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-12B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
2ep n9
.OE-U^ •
2- 2.0E-02 -
E
c 1.5E-02-
o
1
•£ 1.0E-02-
c
0 5.0E-03 •
Oncu-nn
m\ICr\I\I
Data Table:
Well Well Ty
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
RA-MW-12B T
Date
/ /vvvv* /v
^* ^*
•
•
•
* * * * *
•
Effective
Pe Date Constituent
10/15/2003 CHROMIUM, TOTAL
2/10/2004 CHROMIUM, TOTAL
4/5/2004 CHROMIUM, TOTAL
8/15/2004 CHROMIUM, TOTAL
5/5/2005 CHROMIUM, TOTAL
12/12/2005 CHROMIUM, TOTAL
3/8/2006 CHROMIUM, TOTAL
6/15/2006 CHROMIUM, TOTAL
9/25/2006 CHROMIUM, TOTAL
12/15/2006 CHROMIUM, TOTAL
3/30/2007 CHROMIUM, TOTAL
6/5/2007 CHROMIUM, TOTAL
vvv\/
% <) <$• ^
* *
*
Result (mg/L) Flag
2.1E-02
7.6E-03
5.0E-04 ND
4.2E-03
4.1E-03
1.1E-02
3.3E-03
2.4E-03
2.4E-03
5.0E-04 ND
3.4E-03
3.0E-03
Mann Kendall S Statistic:
_
Confidence in
Trend:
1 96.9%
Coefficient of Variation:
1 1.09
Mann Kendall
Concentration Trend:
(See Note)
I °
Number of Number of
Samples Detects
2 2
1 1
1 0
1 1
1 1
1 1
1 1
1 1
1 1
1 0
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-12C
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
1 HP no
I .UC-U^ •
9.0E-03 -
^ 8.0E-03 -
|" 7.0E-03 -
r 6.0E-03 •
o
s 5.0E-03 •
i 4.0E-03 •
| 3.0E-03 •
0 2.0E-03 •
1.0E-03-
Data Table:
Well Well Ty
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
RA-MW-12C T
Date
/ /vvvv* /vi
^*^*
*
•
* *
«
Effective
Pe Date Constituent
10/15/2003 CHROMIUM, TOTAL
2/10/2004 CHROMIUM, TOTAL
4/5/2004 CHROMIUM, TOTAL
8/15/2004 CHROMIUM, TOTAL
5/5/2005 CHROMIUM, TOTAL
12/12/2005 CHROMIUM, TOTAL
3/8/2006 CHROMIUM, TOTAL
6/15/2006 CHROMIUM, TOTAL
9/25/2006 CHROMIUM, TOTAL
12/15/2006 CHROMIUM, TOTAL
3/30/2007 CHROMIUM, TOTAL
6/5/2007 CHROMIUM, TOTAL
e//v\/
' ^
*
•
Result (mg/L) Flag
9.1E-03
2.8E-03
2.7E-03
9.8E-04
4.4E-03
8.7E-03
2.2E-03
6.0E-04
1.5E-03
5.1E-03
5.6E-03
4.2E-03
Mann Kendall S Statistic:
I -4
Confidence in
Trend:
1 58.0%
Coefficient of Variation:
I °7°
Mann Kendall
Concentration Trend:
(See Note)
I S
Number of Number of
Samples Detects
2 2
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-13B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
8.0E-03 •
7.0E-03 -
? 6.0E-03 -
~ 5.0E-03 -
| 4.0E-03 •
g 3.0E-03 -
o
o 2.0E-03 •
O
1.0E-03-
Data Table:
Date
d^X/////////.
*
* * *
* * * * » » •
$v Mann Kendall S Statistic:
I -13
Confidence in
Trend:
1 79.0%
Coefficient of Variation:
I 1'32
Mann Kendall
Concentration Trend:
(See Note)
_
Effective Number of Number of
Well Well Type Date Constituent Result (mg/L) Flag Samples Detects
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
RA-MW-13B T
10/15/2003 CHROMIUM, TOTAL 5.0E-04 ND 2 0
2/10/2004 CHROMIUM, TOTAL 2.3E-03 1 1
4/5/2004 CHROMIUM, TOTAL 5.0E-04 ND 1 0
8/15/2004 CHROMIUM, TOTAL 1.3E-03 1 1
5/5/2005 CHROMIUM, TOTAL 7.1E-03 1 1
12/12/2005 CHROMIUM, TOTAL 1.4E-03 1 1
3/8/2006 CHROMIUM, TOTAL 5.0E-04 ND 1 0
6/15/2006 CHROMIUM, TOTAL 7.0E-04 1 1
9/25/2006 CHROMIUM, TOTAL 5.0E-04 ND 1 0
12/15/2006 CHROMIUM, TOTAL 5.0E-04 ND 1 0
3/30/2007 CHROMIUM, TOTAL 1.2E-03 1 1
6/5/2007 CHROMIUM, TOTAL 5.0E-04 ND 1 0
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-13C
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
_J
O)
• — '
c
o
1
1
o
o
7.0E-03 -
6.0E-03 -
5.0E-03 -
4.0E-03 •
3.0E-03 -
2.0E-03 •
1.0E-03-
n np4-nn .
*
*
« *
*
* . ' *
Mann Kendall S Statistic:
Confidence in
Trend:
I 64.8%
Coefficient of Variation:
0.85
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
RA-MW-13C
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
3.7E-03
1.4E-03
6.8E-04
7.3E-03
1.2E-03
1.4E-03
4.1E-03
5.5E-03
5.0E-04 ND
2.2E-03
9.1E-04
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-14B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
7.0E-03 -
6.0E-03 •
U
1 5.0E-03 •
§ 4.0E-03 -
S 3.0E-03 -
c
01
c 2.0E-03 -
o
O
1.0E-03-
Data Table:
Date
0oVv
*
*
» * * »**» *
$v Mann Kendall S Statistic:
Confidence in
Trend:
1 63.1%
Coefficient of Variation:
I 1'21
Mann Kendall
Concentration Trend:
(See Note)
_
Effective Number of Number of
Well Well Type Date Constituent Result (mg/L) Flag Samples Detects
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
RA-MW-14B T
10/15/2003 CHROMIUM, TOTAL 5.5E-04 2 1
2/10/2004 CHROMIUM, TOTAL 3.5E-03 1 1
4/5/2004 CHROMIUM, TOTAL 5.0E-04 ND 1 0
8/15/2004 CHROMIUM, TOTAL 8.1E-04 1 1
5/5/2005 CHROMIUM, TOTAL 6.5E-03 1 1
12/12/2005 CHROMIUM, TOTAL 1.5E-03 1 1
3/8/2006 CHROMIUM, TOTAL 5.0E-04 ND 1 0
6/15/2006 CHROMIUM, TOTAL 7.0E-04 1 1
9/25/2006 CHROMIUM, TOTAL 6.4E-04 1 1
12/15/2006 CHROMIUM, TOTAL 5.0E-04 ND 1 0
3/30/2007 CHROMIUM, TOTAL 1.5E-03 1 1
6/5/2007 CHROMIUM, TOTAL 6.6E-04 1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-15B
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
2ep ni
.OE-U I •
;;[• 2.0E-01 -
E
c 1.5E-01 •
o
1
•£ 1.0E-01 •
8
c
0 5.0E-02 •
Oncu-nn
.uc^uu
Data Table:
Well Well Ty
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
RA-MW-15B S
Date
o^VVVVVV* «*%
O X T^ i^ N* ^? x* 5 ^
*
•
•
•
•
•
Effective
Pe Date Constituent
10/15/2003 CHROMIUM, TOTAL
2/10/2004 CHROMIUM, TOTAL
4/5/2004 CHROMIUM, TOTAL
8/15/2004 CHROMIUM, TOTAL
5/5/2005 CHROMIUM, TOTAL
12/12/2005 CHROMIUM, TOTAL
3/8/2006 CHROMIUM, TOTAL
6/15/2006 CHROMIUM, TOTAL
9/25/2006 CHROMIUM, TOTAL
12/15/2006 CHROMIUM, TOTAL
3/30/2007 CHROMIUM, TOTAL
6/5/2007 CHROMIUM, TOTAL
>*vv\/
<) <$• ^
. * *
Result (mg/L) Flag
2.0E-02
1.4E-01
5.5E-03
2.2E-03
1.9E-01
1.1E-01
1.9E-01
1.5E-01
3.3E-02
2.1E-02
3.2E-02
4.1E-02
Mann Kendall S Statistic:
I 4
Confidence in
Trend:
1 58.0%
Coefficient of Variation:
1 0.94
Mann Kendall
Concentration Trend:
(See Note)
I NT
Number of Number of
Samples Detects
2 2
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
1 1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: RA-MW-16B
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
2ep ni
.OE-U I •
;;[• 2.0E-01 -
E
c 1.5E-01 •
o
1
•£ 1.0E-01 •
8
c
0 5.0E-02 •
Oncu-nn
.uc^uu
Data Table:
Well Well Ty
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
RA-MW-16B S
Date
/ /vvvv* /vvv* /,/
^^<^^><^'i
-------
MAROS Mann-Kendall Statistics Summary
Well: W98-21B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
c
'a
_§
o
«
>i
§
c
o
o
6.0E-03 •
5.0E-03 •
4.0E-03 -
3.0E-03 -
2.0E-03 -
1.0E-03-
n np4-nn .
^
•
*
* * *
•
*
Mann Kendall S Statistic:
Confidence in
Trend:
I 99.6%
Coefficient of Variation:
0.62
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W98-21B
W98-21B
W98-21B
W98-21B
W98-21B
W98-21B
W98-21B
W98-21B
W98-21B
W98-21B
W98-21B
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
3.6E-03
6.6E-03
4.6E-03
2.7E-03
3.2E-03
2.2E-03
1.2E-03
2.2E-03
5.0E-04 ND
1.5E-03
2.2E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W97-19B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
'a
_§
c
o
«
>i
c
S
c
o
O
1.2E-02-
1.0E-02-
8.0E-03 •
6.0E-03 •
4.0E-03 -
2.0E-03 •
n np4-nn .
*
•
* * * * *
• •
Mann Kendall S Statistic:
Confidence in
Trend:
I 96.4%
Coefficient of Variation:
0.99
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W97-19B
W97-19B
W97-19B
W97-19B
W97-19B
W97-19B
W97-19B
W97-19B
W97-19B
W97-19B
W97-19B
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
1.3E-02
5.1E-03
5.1E-03
3.4E-03
5.0E-04 ND
1.8E-03
2.1E-03
2.1E-03
5.0E-04 ND
2.0E-03
2.4E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
0
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W97-18B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
c
j"
_§
o
«
>I
c
S
c
o
o
1.2E-03-
1.0E-03-
8.0E-04 •
6.0E-04 •
4.0E-04 •
2.0E-04 •
n np4-nn .
• *
*
* 4 ^
» * * *
Mann Kendall S Statistic:
I 12
Confidence in
Trend:
I 79.9%
Coefficient of Variation:
0.39
Mann Kendall
Concentration Trend:
(See Note)
[ NT
Data Table:
Well
W97-18B
W97-18B
W97-18B
W97-18B
W97-18B
W97-18B
W97-18B
W97-18B
W97-18B
W97-18B
W97-18B
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
4.1E-04
5.0E-04 ND
1.1E-03
1.3E-03
1.0E-03
5.0E-04 ND
1.0E-03
1.3E-03
5.0E-04 ND
8.8E-04
1.2E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
0
1
1
1
0
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W92-16B
Well Type: s
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
2ep ni
.OE-U I •
;;[• 2.0E-01 -
E
c 1.5E-01 •
o
1
•£ 1.0E-01 •
8
c
0 5.0E-02 •
Oncu-nn
.uc^uu
Data Table:
Well Well Ty
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
W92-16B S
Date
/ /vvvv* /vvv* /,/
^^<^^><^'i
-------
MAROS Mann-Kendall Statistics Summary
Well: W85-7B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
O)
o
1
I
o
o
1.8E-02-
1.6E-02-
1.4E-02-
1.2E-02-
1.0E-02-
8.0E-03 •
6.0E-03 •
4.0E-03 •
2.0E-03 -
n np4-nn .
*
*
•
Mann Kendall S Statistic:
Confidence in
Trend:
I 99.6%
Coefficient of Variation:
1.58
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W85-7B
W85-7B
W85-7B
W85-7B
W85-7B
W85-7B
W85-7B
W85-7B
W85-7B
W85-7B
W85-7B
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L)
1.8E-02
1.1E-02
8.0E-03
8.4E-04
5.0E-04
5.0E-04
5.0E-04
5.0E-04
5.0E-04
5.0E-04
5.0E-04
Flag
ND
ND
ND
ND
ND
ND
ND
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
0
0
0
0
0
0
0
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W85-6B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
Mann Kendall S Statistic:
'a
_§
o
«
>i
§
c
o
o
1.2E-02-
1.0E-02-
8.0E-03 •
6.0E-03 •
4.0E-03 -
2.0E-03 •
n np4-nn .
^
» »
W W
*
•
I ~23
Confidence in
Trend:
I 99.1%
Coefficient of Variation:
0.79
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W85-6B
W85-6B
W85-6B
W85-6B
W85-6B
W85-6B
W85-6B
W85-6B
W85-6B
Well Type
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
1.3E-02
4.7E-03
5.6E-03
2.9E-03
4.8E-03
3.8E-03
5.0E-04 ND
2.9E-03
2.0E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
-------
MAROS Mann-Kendall Statistics Summary
Well: W99-R5B
Well Type: T
COC: CHROMIUM, TOTAL
Time Period: 10/15/2003 to 6/5/2007
Consolidation Period: No Time Consolidation
Consolidation Type: Geometric Mean
Duplicate Consolidation: Average
ND Values: Specified Detection Limit
J Flag Values : Actual Value
Date
«° *s
B)
o
?
c
o
o
1.0E-02-
8.0E-03 -
6.0E-03 •
4.0E-03 -
2.0E-03 -
n np4-nn .
•
•
•
» •
• . * • *
•
Mann Kendall S Statistic:
I ~33
Confidence in
Trend:
I 99.5%
Coefficient of Variation:
0.74
Mann Kendall
Concentration Trend:
(See Note)
Data Table:
Well
W99-R5B
W99-R5B
W99-R5B
W99-R5B
W99-R5B
W99-R5B
W99-R5B
W99-R5B
W99-R5B
W99-R5B
W99-R5B
Well Type
T
T
T
T
T
T
T
T
T
T
T
Effective
Date
2/10/2004
4/5/2004
8/15/2004
5/5/2005
12/12/2005
3/8/2006
6/15/2006
9/25/2006
12/15/2006
3/30/2007
6/5/2007
Constituent
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
CHROMIUM, TOTAL
Result (mg/L) Flag
7.5E-03
9.9E-03
4.8E-03
6.7E-03
4.5E-03
1.8E-03
1.4E-03
2.5E-03
5.0E-04 ND
1.9E-03
2.4E-03
Number of
Samples
1
1
1
1
1
1
1
1
1
1
1
Number of
Detects
1
1
1
1
1
1
1
1
0
1
1
Note: Increasing (I); Probably Increasing (PI); Stable (S); Probably Decreasing (PD); Decreasing (D); No Trend (NT); Not Applicable (N/A) -
Due to insufficient Data (< 4 sampling events); ND = Non-detect
MAROS Version 2.2, 2006, AFCEE
9/6/2007
Page 1 of 1
------- |