EPA/600/R-94/119
October 1994
EVALUATION OF SAMPLING AND FIELD-FILTRATION METHODS FOR
THE ANALYSIS OF TRACE METALS IN GROUND WATER
by
Karl F. Pohlmann
Gary A. Icopini
Richard D. Me Arthur
Water Resources Center
Desert Research Institute
Las Vegas, Nevada 89119
Charlita G. Rosal
Analytical Sciences Division
Environmental Monitoring Systems Laboratory
Las Vegas, NV 89193-3478
Cooperative Agreement No. CR815774-01
Project Officer
Charlita G. Rosal
Analytical Sciences Division
Environmental Monitoring Systems Laboratory
Las Vegas, NV 89193-3478
This study was conducted in cooperation with
Water Resources Center
Desert Research Institute
Las Vegas, NV 89119
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89193-3478
Printed on Recycled Paper
-------
NOTICE
The U.S. Environmental Protection Agency (EPA), through its Office of Research and Development (ORD), partially
funded and collaborated in the research described here. It has been peer reviewed by the Agency and approved as an
EPA publication. Mention of trade names or commercial products does not constitute endorsement or recommendation
for use.
-------
ABSTRACT
Selected ground-water sampling and filtering methods were evaluated to determine their effects on field
parameters and trace metal concentrations in samples collected under several types of field conditions. The study
focused on sampling in conventional standpipe monitoring wells under conditions where traditional approaches to
sampling may produce turbid samples, which often leads to the decision to filter suspended particles from the sample
before laboratory chemical analysis. However, filtration may also remove colloidal particles that are known to be
mobile under certain ground-water conditions and may be important to the transport of hydrophobic organic
contaminants and trace metals. The specific sampling and filtration variables investigated in this study were (1)
filtration with 0.45-um pore size filters or 5.0-u,m pore size filters versus no filtration; (2) sampling device, specifically
bladder pump, submersible-centrifugal pump, and bailer; and (3) sampling pump discharge rate during purging and
sample collection using a "low" rate of 300 mL/min and a "moderate" rate of 1000 mL/min. Three field sites were
visited: an active municipal solid waste landfill in Wisconsin, a closed solid waste landfill in Washington, and a site
contaminated by industrial waste in Nevada. Three wells at the Wisconsin and Washington sites and two wells at the
Nevada site were included in the evaluation. Filtration with 5.0-um filters was conducted only at one well at each site.
Bailers caused more disturbance of the sampling zone than the three pumping methods as evidenced by
measurements of field parameters and concentrations of particles, major ions, and trace metals. Bailers also produced
higher concentrations of particles of the size range potentially important to colloidal transport of contaminants (e.g.,
between 0.001 and 10.0 urn). Little variation was observed in the analytical determinations between the pumps but
some variation existed in the field indicator parameters, primarily temperature, dissolved oxygen, and turbidity. Under
low—yield conditions, the moderate-rate pumps produced dissolved oxygen and turbidity levels that were greatly
elevated over those produced by the low-rate pump. The effects of field filtration were most evident for the bailer,
which often produced trace metal concentrations in unfiltered samples that were orders-of-magnitude higher than in
0.45-um-filtered samples. The largest differences occurred at the most turbid wells and in samples containing the
highest particle concentrations, apparently reflecting the entrainment of normally non-mobile particles and associated
matrix metals in the bailed samples. Similar effects were observed in some samples collected by pumps from the most
turbid wells, particularly the low yield well. For most pump sampling, however, differences in concentrations between
0.45—u,m—filtered and unfiltered samples were not significant and particle concentrations were significantly lower
than those produced by the bailer. Overall, trace metal concentrations.in, 0.45-um-filtered samples were generally
independent of sampling method, suggesting that these constituents were present as dissolved species and not
associated with particles or were associated with particles smaller than 0.4 um. At wells where 5.0-um filtration was
conducted, physical and hydrochemical conditions resulted in minimal differences between trace metal
concentrations in the 5.0-um-filtered, 0.45-um-filtered, and unfiltered samples.
in
-------
CONTENTS
ABSTRACT iii
FIGURES v
TABLES vi
ACKNOWLEDGEMENTS vii
1. INTRODUCTION 1
Background 1
Objectives 2
2. CONCLUSIONS 4
3. RECOMMENDATIONS 6
4. MATERIALS AND METHODS 7
5. RESULTS AND DISCUSSION 13
Impacts on Field Parameters 13
Impacts on Particle Size Distribution and Concentration 20
Impacts of Filtration on Metals Concentrations 23
Impacts of Sampling Device on Metals Concentrations 34
Impacts on Concentrations of Major Ions 35
Practical Considerations 36
REFERENCES 38
APPENDICES
A. Summary of Indicator Parameter Measurements and Purging Data 41
Table A—1. Values of Field Parameters During Purging 41
Table A-2. Volumes and Times at Which Indicator Parameters Reached
Equilibrium Values 55
B. Summary of Particle Size Analysis 56
TableB-1. Summary of Particle Size Analysis 56
C. Summary of Analytical Results 59
Table C-l. Trace Metals Analytical Result 59
Table C-2. Gross Chemistry Analytical Results 61
D. Results of Statistical Analysis 70
Table D-l. Results of Statistical Analysis 70
IV
-------
FIGURES
1. Impacts of dewatering filter pack on turbidity and dissolved oxygen at WASH-1 14
2. Impacts of variations in discharge rate of CP1 and CP2 on equilibration of
field parameters at WISC-4 15
3. Trends of field parameters in high-yield well NEV-2 16
4. Equilibrium values of field parameters 17
5. Purged volumes required to reach equilibrium values of field parameters 18
6. Results of particle analyses: (a) total particle concentrations; (b) total particle concentrations,
expanded scale to show results of pumped samples; (c) concentrations of particles less than
0.4 [Am in size 22
7. Plots of turbidity (NTU) versus total particle concentration (mg/L) for (a) samples containing
less than 50 mg/1 particles, and (b) samples containing more than 50 mg/L particles
(with the exception of bailer at WISC-4) 24
8. Iron concentrations 27
9. Manganese concentrations 28
10. Barium concentrations for Wisconsin and Nevada sites and nickel concentrations for
Washington site 29
11. Chromium concentrations 30
12. Arsenic concentrations 31
-------
TABLES
I. Summary of monitoring well characteristics and hydrogeochemical conditions
as determined by bailed samples 8
2. Descriptions of the ground-water sampling devices 9
3. Criteria for stabilization of indicator parameters during purging 10
4. Analytical methods and quality control data 12
5. Filtration ratios for metals concentrations 26
VI
-------
ACKNOWLEDGEMENTS
The authors wish to acknowledge Andrew Teplitzky and James Brown of the U.S. EPA Office of Solid Waste for their
support and contributions to the project scope. Jane Denne of the U.S. EPA Environmental Monitoring Systems
Laboratory is acknowledged for her assistance in developing and carrying out the project. The collaboration of Robert
Puls of the U.S. EPA R.S. Kerr Environmental Research Laboratory, Jack Connelly and Hank Kuehling of the
Wisconsin Department of Natural Resources, Nadine Romero of the Washington Department of Ecology, Bryan
Haelsig of the U.S. Navy CLEAN Program, and Bernie Zavala and Marcia Knadle of the U.S. EPA Region X office
was invaluable. Finally, Michael Whitbeck, James Heidker, Roger Jacobson, John Hess, and Jenny Chapman of the
Desert Research Institute, Water Resources Center, provided technical assistance and advice.
VII
-------
-------
SECTION 1
INTRODUCTION
The U.S. Environmental Protection Agency (U.S. EPA) is studying how sampling techniques impact
contaminant concentrations in ground-water samples. This study was undertaken to investigate how concentrations
of trace metals were affected by selected methods of sample collection and field filtration. The study focused on
sampling in conventional standpipe monitoring wells under conditions where traditional approaches to sampling may
produce turbid samples, as might occur when sampling a monitoring well with an intake located in a water-bearing
zone containing significant amounts of fine-grained materials. Although proper choice and careful implementation
of sampling methodology is always important for the acquisition of quality ground-water samples, conditions of high
turbidity and corresponding high particle concentrations make this especially significant.
BACKGROUND
Historically, groundTwater contaminants were considered to be partitioned between two phases, a mobile phase
composed of dissolved (aqueous) solutes in water transported by natural ground-water flow and a solid phase
composed of the matrix materials of the water-bearing zone. This solid phase included the immobile formation itself
and particles derived from the formation that may have concentrated around monitoring wells as a result of disturbance
during well drilling and construction (Gillham et al., 1983). These large particles, often 10 \im and larger and often
referred to as "suspended solids" or sediments, are generally immobile under natural ground-water flow conditions
(with the exception of karst systems) and are usually settled out by gravity because ground-water velocities are
insufficient to entrain them (Yao et al., 1971). The action of purging and sampling a monitoring well may, however,
provide sufficient energy to suspend large particles that have accumulated in the sampling zone and inside the well
bore and incorporate the particles in ground-water samples. Particles composed of clay minerals, various metal oxides,
and humic material, may adsorb metal ions to their surfaces due to their high cation-exchange capacities. This is
especially true of clay minerals that may contain metals as part of their crystal structure (Drever, 1988). Inclusion of
metals associated with these normally immobile particles may bias analytical determinations, leading to elevated, and
if suspended particle concentrations are very high, improbable concentrations of mobile contaminants.
Investigations of contaminant metals in ground water have generally focused on dissolved species because they
were considered more likely to be transported under natural hydraulic gradients through ground-water systems, and
because inclusion of particles in samples might incorporate matrix metals leading to biased determinations of
contaminant metals concentrations (U.S. EPA, 1986). As a result, ground-water samples are commonly filtered in the
field to remove these suspended particles. Filtration has been considered particularly necessary under turbid
conditions where high particle (sediment) loadings might lead to significant analytical bias through inclusion of large
quantities of matrix metals in the analysis. Alternatively, the presence of particles in samples might also bias analytical
determinations through removal of metal ions from solution during shipment and storage as a result of interactions
with particle surfaces.
Unfortunately, indiscriminant use of field filtration ignores the presence of particles in ground water that may
exist between the extremes of solutes and sediments. These particles, referred to as colloids, are generally considered
to be in the size range of 0.001 to 5.0 um (Mills et al., 1991). Like larger particles, colloids are commonly composed
-------
of clay minerals, metal oxides, and humic material, and therefore, present likely sites for sorption of hydrophobic
organic and inorganic constituents (McCarthy and Zachara, 1989; Puls 1990). But like dissolved species, the small
sizes of colloids facilitate their mobility in certain ground-water systems, and also provide them with high ratios of
surface area to mass which increases their relative sorptive capabilities (McDowell-Boyer et al., 1986). Association
with colloids has been shown to provide an important mechanism for transport of hydrophobic contaminants in ground
waters including radionuclides (Champ et al., 1982; Buddemeier and Hunt, 1988: Penrose et al., 1990) and metals
(TiHekeratne et al., 1986; Gschwend and Reynolds, 1987; Magaritz et al., 1990).
The demonstrated potential for contaminant transport in association with colloids has important implications
for the practice of field filtration because the boundary between paniculate and dissolved has been operationally
defined at 0.45 u,m (U.S. EPA, 1979). This boundary presumes that the component retained on a 0.45 urn filter
represents suspended solids, while the component that passes through the filter represents dissolved metals. However,
filtration of ground-water samples may remove an unknown fraction of metals important to transport and lead to
erroneous conclusions about contaminant mass and extent (Puls and Barcelona, 1989). This is more likely where
significant concentrations of metals are associated with mobile colloids larger than the filter pore size.
Field filtration has other significant drawbacks. For example, spatial variation of ambient physical and
gcochemical conditions in the ground-water zone may cause a related distribution of aqueous and solid species of
metals. Because field filtration is designed to allow passage primarily of dissolved species, those species present as
mobile solids larger than the filter pore size at a particular well location may be removed during filtration. Further,
the act of collecting a ground-water sample may cause metals to change from one species or phase to another, which
could allow filtration to affect the concentrations present in the sample. As an example, exposure of samples to oxygen
during sampling may cause oxidation of dissolved ferrous iron (Fe2+) to ferric iron (Fe3+), producing a ferric hydroxide
precipitate (Stumm and Morgan, 1981), which if removed during filtration, could bias iron determinations. Finally,
factors associated with the filtration process itself, such as filter type, filter diameter, filtration method, and sample
volume, have been shown to affect trace metal concentrations in filtered samples, leading to uncertainty in the results
(Horowitz et al., 1992).
The issues of colloidally-transported metals and field filtration become even more important when sampling
produces turbid samples. This may occur when bailers or submersible pumps operated at moderate or higher discharge
rates (greater than 1000 mL/min) are used in wells completed in formations containing fine-grained sediments or in
inadequately designed, constructed, or developed wells. These sampling methods may entrain sediments and normally
immobile colloids, thereby introducing bias to the analyses (Puls et al., 1991; Backhus et al., 1993). Under these
conditions, filtering the samples to remove suspended material may also remove colloids and the metals associated
with them. Collecting pumped samples at flow rates that approach natural ground-water advective flow velocities may
reduce entrainment of normally immobile species, thereby alleviating the need to filter samples. This approach to
sampling has been advocated by several researchers with maximum suggested pumping rates of 100 to 300 mL/min
(Ryan and Gschwend, 1990; Puls et al., 1990; Kearl et al., 1992; Backhus et al, 1993).
Collection of ground-water samples for analysis of metals concentrations is required under several U.S.
environmental regulations, including CERCLA (Superfund), RCRA Subtitle C (Hazardous Waste), and RCRA
Subtitle D (Solid Waste). As a result, the debate regarding filtration of ground-water metals samples impacts a wide
range of sampling programs and a large number of sites, suggesting the heed for further study into the issue.
OBJECTIVES
' This study assessed, under several types of field conditions, the impacts of selected aspects of ground-water
sampling on field determinations of unstable parameters and analytical determinations of trace metal 'chemistry. The
study focused on selected sampling methods that are applicable in conventional monitoring wells for ground-water
sampling at solid waste landfills and hazardous waste sites. Disturbance of samples and the sampling zone (well intake,
-------
filter pack, and formation adjacent to the monitoring well intake) by the sample collection method was considered
a critical factor affecting the accuracy of field and laboratory determinations of ground-water hydrogeochemistry. In
particular, the interdependence of chemical concentrations and particle content was of interest because the presence
of particles, and possible removal of a fraction of particles by filtration, might affect concentrations of certain trace
metals. Therefore, sampling methods that minimize disturbance of the sampling zone were compared to more
widely-used methods that are known to agitate the sampling zone during purging and sampling.
For this study, the term "particles" refers to analytically-determined solid material larger than 0.03 u,m and
includes both mobile colloidal material and normally immobile .sediment mobilized by the sampling process. In
addition, the term "trace metals" refers to those metals and metalloids that are often included in sampling programs
at solid waste landfills and hazardous waste sites. This study focused on arsenic (As), barium (Ba), cadmium (Cd),
chromium (Cr), iron (Fe), lead (Pb), manganese (Mn), and nickel (Ni). In addition to being a result of contamination
from a waste disposal site, trace metals such as these may originate from natural components of the formation matrix,
or as a consequence of monitoring well drilling, construction, and sampling activities. To address the issue of
ground—water transport of trace metals from a waste disposal site, artifacts resulting from these other factors must be
minimized.
The specific objectives of the study were to provide a survey of the impacts of the following aspects of
ground-water sampling:
1. Impacts of sample collection method on determinations of field parameters. •
2. Impacts of filtration with 0.45-um pore size filters or 5.0-[A.m pore size filters versus no filtration on trace
metal concentrations.
3. Impacts of sampling device, specifically bailer, bladder pump, submersible-centrifugal pump at a "low"
discharge rate of 300mL/min, and submersible-centrifugal pump at a "moderate" discharge rate of
lOOOmL/min, on trace metal concentrations.
4. Impacts of sampling device on particle size distribution and total concentration.
Impacts of these variables on the concentrations of major ions were also investigated, though not the major focus
of this study. Also, although practical aspects of the sampling methods were considered, the evaluations were based
primarily on the hydrogeochemical results.
To address the objectives, three field sites were visited: an active municipal solid waste landfill in Wisconsin,
a closed solid waste landfill in Washington, and a site contaminated by industrial waste in Nevada. Three wells at the
Wisconsin and Washington sites and two wells at the Nevada site were included in the evaluation. Although certainly
not representative of geologic and hydrogeochemical conditions at all solid waste landfills and hazardous waste sites,
these sites provided typical field conditions where traditional approaches to ground-water sampling produce turbid
samples.
-------
SECTION 2
CONCLUSIONS
1. Field determinations of the unstable parameters dissolved oxygen (DO) and turbidity were the most sensitive to
disturbance of the sampling zone, with values produced by bailing often orders-of-magnitude higher than those
produced by the pumps. Under low—yield conditions, the moderate-rate pumps also elevated DO and turbidity
levels. In most other cases, however, pump type and discharge rate did not produce significant differences in these
parameters. Elevated DO concentrations resulted from (a) aeration of the sampling zone and sample during
collection by bailer and subsequent DO measurement, (b) aeration of the discharge tube when pump discharge was
stopped because of operational problems with the submersible centrifugal pumps, and (c) aeration of the sampling
zone during moderate-rate purging and sampling in low-yield wells. Elevated turbidity values .were caused by (a)
the action of purging and sampling with bailers, (b) surging discharge of submersible centrifugal pumps under
certain hydraulic conditions, and (c) disturbance of the sampling zone during moderate-rate purging and sampling
in low—yield wells. Variations in indicator parameters electrical conductivity (EC) and pH were insignificant
between the four sampling methods, suggesting they were less related to disturbance of the sampling zone than
DO and turbidity. Temperature showed little variation between the bladder pump and bailer but was highly elevated
by the operation of the submersible centrifugal pump at low discharge rates. At wells not impacted by low-yield
conditions or operational problems, the pumps provided equilibrium values of most indicator parameters at purged
volumes of 12 to 32 L and within 10 to 15 percent of equilibrium values at purged volumes of 6 to 16 L. However,
due to the nature of their operation, the bailers often did not produce equilibrium DO or turbidity conditions and
the submersible centrifugal pumps often did not produce equilibrium temperature conditions. Likewise, under
low-yield conditions, moderate-rate purging and sampling did not produce equilibrium DO or turbidity
conditions. Although purging and sampling at low pump speeds reduced purged volumes from those typically
experienced with bailers, the time required to complete purging was often considerably longer.
2. In-line filtration of pumped samples (by either 0.45-um or 5.0-jxm filter cartridges) did not significantly impact
the concentrations of trace metals or major ions in the majority of cases. Filtered concentrations were generally
not significantly lower than unfiltered concentrations, suggesting that relatively representative samples of the
potentially mobile load were obtained in the unfiltered samples. The most obvious exceptions occurred at "turbid"
or low-yield wells where moderate-rate pumping produced samples with the highest total particle concentrations
of pumped samples. As a result of the elevated particle concentrations, trace metals that comprised these particles
(such as oxides or hydroxides of iron), or that were associated with these particles (many aqueous species), were
elevated in unfiltered samples, and were likely unrepresentative of mobile species. In contrast to the pumped
samples, most bailed samples exhibited very large differences between unfiltered and filtered trace metals
concentrations. In unfiltered samples, the high concentrations of particles suspended and entrained by bailing
significantly elevated the concentrations of those trace metals associated with particles. In many cases, however,
filtered bailed samples exhibited trace metal concentrations that were roughly equivalent to those produced in
samples acquired using the pumps. Because bailers produced higher concentrations of potentially mobile particles,
e.g., those in the 0.03 to 5.0 urn ranges, it may be concluded that the trace metals detected in this study were either
not associated with colloidal transport or were associated with colloids smaller than 0.45 pirn in size. The former
seems more likely because bailing generally produced higher concentrations of sub-0.45-[xm particles but not
higher concentrations of trace metals in 0.45-u,m-filtered samples.
-------
3. Sample collection with the bailer generally caused significant differences in trace metal concentrations with
respect to the pumps only for unfiltered samples. Entrainment of high concentrations of normally immobile
particles by bailing resulted in concentrations of trace metals that were often many times higher than in unfiltered
pumped samples. Concentration differences were less for metals species that were primarily dissolved and not
associated with particles. Samples collected by the bailer and immediately filtered exhibited trace metal
concentrations that were roughly equivalent to those produced by the pumps and in-line filtration. Few consistent
or significant differences in trace metals or particle concentrations were observed between the bladder pump and
centrifugal pump or the "low" discharge rate of 300 mL/min and the "moderate" discharge rate of 1000 mL/min
with the centrifugal pump. It appears that potential differences in metals concentrations between these rates were
masked by the overall low concentrations observed and the variability associated with experimental and analytical
error. The only exceptions to this response occurred at a highly turbid, low-yield well where moderate-rate
pumping entrained higher quantities of normally immobile particles, and associated metals, than low-rate
pumping.
4. Disturbance of the sampling zone by bailing resulted in total particle concentrations that were significantly higher
than those produced by the tested pumping methods. Total particle concentrations at wells not impacted by
low-yield conditions or pump operational problems ranged from 2.8 to 41.1 mg/L in pumped samples and from
40.3 to 818 mg/L in bailed samples. Under the most turbid conditions, the bailer produced a concentration of 6970
mg/L while the highest pumped value was 1300 mg/L. The bailer also generally resulted in higher concentrations
of potentially mobile particles, suggesting that concentrations of colloidally-associated contaminants could be
biased when bailing disturbs the sampling zone and elevates turbidity. Regardless of collection method, samples
with over 30 mg/L total particle concentration contained over 50 percent of their particle mass as particles larger
than 5.0 [xm and over 95 percent as particles larger than 0.45 urn.
5. The relationship of turbidity to particle concentration and its sensitivity to the purging process, relative to other
indicator parameters, suggests that turbidity may be a useful indicator of relative particle concentrations between
wells and of stabilization of particle concentrations during monitoring well purging. If mobile particles are thought
to be important to transport of contaminants in ground water, use of field parameters such as pH, temperature, or
EC as criteria for determining adequate sampling conditions may result in underpurging.
-------
SECTIONS
RECOMMENDATIONS
1. This study included only a limited number of wells at three sites and therefore does not represent the wide variety
of geologic and hydrogeochemical conditions likely to be present at all solid waste or hazardous waste landfills.
As a result, more information is required from a variety of sites regarding the presence of colloidal particles and
the importance of these particles in the transport of trace metals and other contaminants in ground water. A better
understanding of colloidal transport processes in ground-water environments could be gained from research fo-
cused on describing hydrogeochemical conditions and colloid size distribution, composition, movement, and
association with trace metals at a variety of solid waste and hazardous waste sites. A "survey" of existing sites
could be conducted to investigate hydrogeochemical colloid-related conditions. Furthermore, controlled field ex-
periments using colloidal tracers could be conducted to better understand transport processes.
2. Given that most of the trace metals detected in this study appeared to be of natural origin (the only exception being
several metals at the Nevada site), it would be beneficial for resolution of questions regarding sampling and filtra-
tion to apply many of the same experimental techniques to several sites containing known metals contamination.
Sites where there is suspicion of metals being transported in ground water via association with colloidal particles
would be of particular importance.
3. Hydrogeochemical conditions, which are different at every point in the ground-water zone and at every individual
well site, contribute to the variability in speciation of trace metals, the presence of mobile colloidal particles, and
associations of metals to particles. As a result, it is important to understand the basic framework that these condi-
tions produce on a site-specific, and well-specific, basis. Therefore, even "routine" interpretation of ground-water
quality data should include careful analysis of physiochemical conditions. For example, redox conditions in part
of a ground-water system may promote the formation of iron oxide colloids suitable for sorption and transport
of trace metals, while in another part of the system these conditions may not be present. Filtration of samples col-
lected from the system where colloids are present may remove an important fraction of the mobile contaminant
load, while filtration of samples from that part of the system without colloids is unlikely to affect trace metals
concentrations. Collection and analysis of redox data would be critical to interpreting the analytical results.
-------
SECTION 4
MATERIALS AND METHODS
Three field sites were visited during the course of the study: an active municipal solid waste landfill in
Wisconsin, a closed solid waste landfill in Washington, and a site in Nevada that is contaminated by industrial waste.
Three wells at each of the Wisconsin and Washington sites and two wells at the Nevada site were included in the
evaluation. One of the wells sampled at the Wisconsin site (WISC-1) was not included in the evaluation for reasons
discussed below. The wells were chosen with assistance from officials of the local state regulatory agencies, and in
the case of the Washington site, EPA Region X. Relatively shallow (less than 25 m deep), 5.08-cm-diameter
monitoring wells were utilized to minimize purged volumes and times, simplify equipment requirements, and reduce
the time required for equipment decontamination procedures. Wells that demonstrated high turbidity levels (over 100
Nephelometric Turbidity Units or NTU) and evidence of impact by metals transported in ground water were targeted
for study. A summary of the well characteristics and hydrochemical conditions represented by bailed samples is
presented in Table 1.
During the planning stages of the study, it was proposed that prior to our visit to each landfill, the landfill
operators would collect ground-water samples as part of their usual monitoring program according to their standard
operating procedures. Discussions with officials of state regulatory agencies and the EPA Office of Solid Waste
revealed that these procedures most often consisted of sample collection with bailers and field filtration with
0.45-u,m-membrane filters. Splits of these samples would then be submitted to the DRI laboratory for analysis of the
same constituents to be analyzed during the pump experiments. The following week, we would conduct a suite of
experiments in the same wells utilizing the submersible pumps. This approach was followed at the Wisconsin site but
modified for the Nevada and Washington sites where routine sampling programs were not active. At these latter two
sites, we collected samples using bailers and filtration techniques similar to those used at the Wisconsin site. Bailed
samples were collected a week prior to the pump experiments at the Nevada site and subsequent to the completion
of all the pump experiments at the Washington site.
Four methods of collecting samples from conventional standpipe monitoring wells were evaluated using three
types of sampling devices and three pump discharge rates. The methods utilized at eight of the nine wells were: (1)
bailer, (2) centrifugal pump at 300 mL/min (denoted in this report as CP1), (3) bladder pump at 500 mL/min at the
Wisconsin site or 1000 mL/min at the other sites (denoted as BP), and (4) centrifugal pump at 1000 mL/min (denoted
as CP2). Experiments at the ninth well (WISC-1) included only the bailer and the bladder pump at a discharge rate
of 500 mL/min. Discharge rates were measured at ground surface and were controlled by the pump speed rather than
by flow restrictors or valves. These discharge rates were used for both purging and sampling. Descriptions of the
sampling devices are given in Table 2.
All devices were used in a portable mode. The centrifugal pumps were powered by either a 230 volt or 120 volt
converter and a generator of at least 3000 watts, while the bladder pump was powered by compressed nitrogen gas
using a pneumatic controller. Fill and discharge cycles of the bladder pump were approximately four seconds in
duration and nitrogen delivery pressure was set between 20 and 35 pounds per square inch, depending on desired
discharge rate and lift required at each well. Due to the pulsed discharge of the bladder pump, entrance velocities at
-------
CO
I
CO
O'i.
!
CO T3
? CM
t«- .2 "V _,
CO *O .Jj O
5 CM risi
CO
CM CM.
CO CM
in
cri
E5 co 2
co co -i- •
CM
CO*
co
CD
CM
* 2
CO T-
m
co
O
Q.
CM
in
CM s
2 ,
.2 = =?
l§-cS
SQ=._I
co
CO
CO
••*
o
t
CO
Q
0.0
111
Q.
CO
co
•a
_CD
'S
m
S
CD
P
I
S
•a
CO
IT
CM
CM
CM
CM
en
ri
5"
1-
o
co
CD
i.
a
DO (mg/L)
-
3 3
Q.TD
col
E o>
O O)
W c
f°-
II
.• CO
&$
1^1
O CO = —I
fillS!
-------
the pump inlet were twice the referenced surface discharge rate. Discharge tubing for the pumps was 1.27-cm I.D.
PTFE-lined polyethylene. The same piece of tubing was used for both pumps and was cut to the length required for
the deepest sampling zone at each site. New tubing was used for each site.
TABLE 2. DESCRIPTIONS OF THE GROUND-WATER SAMPLING DEVICES
I.D.
CP1
BP
CP2
Type
Submersible
centrifugal
pump
Bladder pump
Submersible
centrifugal
pump
Brand
Grundfos
Redi-Flo 2
QED Well
Wizard
Grundfos
Redi-Flo 2
Materials
Stainless steel body,
PTFE-lined PE discharge
tubing
PTFE body, PTFE-lined PE
discharge tubing
Stainless steel body,
PTFE-lined PE discharge
tubing
Discharge Rate
Low
(approx. 300 mL/min)
Moderate
(approx. 1 L/min, 500 mL/min
at Wise.)
Moderate
(approx. 1 L/min)
Bailer (Nev. and Bailer, dual
Wash. Sites) check-valve
Bailer Bailer, dual
(Wise. Site) check-valve
Monoflex
Monofiex
PVC body, disposable PP
haul line
PVC body disposable PP
haul line
Volume: 460 mL
Volume: 3.5 L
The suite of pumping experiments in each well included the following primary elements. The low-flow-rate
experiments were conducted first to minimize disturbance of the sampling zone. Upon completion of purging at the
low rate, a full suite of samples was collected for analysis of particle size distribution, dissolved solids, organic carbon,
and 0.45-u.m-filtered and unfiltered metals and major ions. An additional set of samples at one well at each site were
5.0-u.m-filtered to investigate the effects of removing from the sample only the suspended particles larger than 5.0 (xm.
Following sample collection, the pump and tubing were removed from the well, decontaminated, and installed in the
next well. After completion of the low-rate experiments in all wells, the moderate-rate experiments were conducted.
These experiments followed the same procedures as the low-rate experiments, with the exception of the pumping rate.
Experiments in individual wells usually were separated by at least 24 hours.
Many of the specific sampling procedures followed in this study, as well as the general approach to acquisition
of ground-water samples, are described in Desert Research Institute (1991). Procedures that differ from those
presented in that document, as well as a brief overview of all procedures, are described here. Each well site was
prepared prior to sampling by positioning the sampling support vehicle near the well head, spreading a plastic ground
sheet around the well, and opening the well cover. The static water level in the well was measured with a flat tape water
level probe.
If possible, the sampling device intake was positioned in the screened interval within 0.6 m of the top of the
screen, however, low-yield conditions at several wells dictated that the device intake be set to greater depths (up to
1.5 m) to maintain adequate flow into the well and device. Once established, this depth was used for all devices.
Measurements of field parameters were made during purging to evaluate the effectiveness of the purging method for
minimizing disturbance of the sampling zone and removing stagnant water, as well as to provide an indication of when
well purging was complete. Purging was considered complete and sample collection initiated when measurements of
these parameters reached "stable values" over approximately one well-screen volume. Due to the nature of their design
and operation, the bailers were often incapable of producing stable values of certain indicator parameters, particularly
DO and turbidity. As a result, purging by bailer was considered complete when the other indicator parameters
-------
stabilized or when 3 to 5 well volumes had been purged. At the Wisconsin site, the volumes specified in the site
sampling plan were used for purging with the bailers (approximately 4 casing volumes).
Use of the well-screen volume as the unit of measurement does not suggest that all stagnant water in the well
bore, well screen, or sampling zone was replaced by fresh ground water during purging. On the contrary, purging at
low to moderate rates in high-yield wells probably results in a certain degree of mixing of ground water with stagnant
water in the well bore (see for example, Unwin and Maltby, 1988; Robbins and Martin-Haydon, 1991). As a result,
the actual volume of the well screen is generally not directly related to the volume required to purge the well but does
provide a useful benchmark for comparison of purged volumes between wells. Stabilization criteria were based on
accuracy and precision data on instrumentation as supplied by the manufacturers, hydrogeochemical conditions at
each well, past experience, and guidelines suggested by Gibs and Imbrigiotta (1990), and are given in Table 3.
TABLE 3. CRITERIA FOR STABILIZATION OF INDICATOR PARAMETERS DURING PURGING
Field Parameter
Stabilization Criterion
Dissolved Oxygen
Electrical Conductivity
PH
Temperature
Turbidity
0.10mg/L
3% Full Scale Range
0.10 pH unit
0.2 °C
1.00NTU
When purging was considered complete, the pump discharge line leading to the flow-through cells was
disconnected, the pump speed was readjusted to obtain the desired discharge rate, and the discharge was directed into
sample bottles for the unfiltered samples or through high-capacity in-line filter cartridges for the filtered samples.
Bailed samples from the WISC wells were transferred to a polycarbonate holding vessel (Geotech Environmental
Equipment, Inc.) fitted with a 102-mm-diameter, 0.45 um membrane filter, while samples bailed from the NEV and
WASH wells were transferred to a holding vessel composed of polycarbonate, polypropylene, and polyvinylchloride
(QED, Inc.), and fitted with the appropriate pore-size in-line filter cartridge. In both cases, compressed nitrogen was
used to pressurize the holding vessel and force the sample through the filter. The 0.45-um in-line filter cartridges
(Geotech Environmental Equipment, Inc.) utilized Versapor® (acrylic copolymer on a nylon support) membranes and
a polypropylene body, the 5.0-um in-line filter cartridges (QED, Inc.) utilized nylon membranes and a polypropylene
body, and the 0.45-[im membrane filters were composed of polycarbonate. In-line filters were flushed with
approximately 500 mL of sample water to remove membrane preservatives and wetting agents before the filter
discharge was directed into sample bottles. Including the volumes that passed through the filters for the major ion,
dissolved solids, and organic carbon samples, approximately 3.5 L of water had passed through the filter membranes
before the metals samples were collected. Filter clogging during filtration of highly turbid samples was experienced
only with the 102-mm membrane filters (bailed samples at WISC Wells), requiring replacement of the filter during
the filtration process.
Turbidity was measured with a direct-reading Nephelometric meter (HF scientific, inc.) utilizing a flow-through
cuvette for pump discharge and a standard 28-mm cuvette for bailer discharge. The instrument provides a linear
display of turbidity in NTUs. Calibration was accomplished prior to visiting each site using standard Formazin
solutions, while standardization with a 0.02-NTU reference standard was conducted prior to each sampling event. DO,
temperature, pH, and EC of pump discharge were measured using a flow-through cell and meter system (QED, Inc.).
Measurements of turbidy and pH were more difficult with the bladder pump due to the pulsed discharge. Measurements
10
-------
of bailer discharge were made using the same probes and meter but a beaker was used in place of the flow-through
cell. Calibrations of DO and pH were conducted immediately prior to the initiation of each sampling event and pH
was checked after sample collection was complete. A 100 percent humidity calibration was used for DO and a
three-point calibration for pH using 4.01, 7.00, and 10.01 buffers. Conductivity was calibrated at the beginning and
end of each sampling day using a two- or three-point calibration of standards that bracketed the conductivity values
of the .ground water to be sampled. Oxidation-reduction conditions (Eh) were measured off-line using a silver/silver
chloride reference electrode and a platinum working electrode. Eh calibration was carried out prior to each sampling
event using a Zobell reference solution.
Samples for analysis of major ion chemistry, organic carbon, and dissolved solids were collected in duplicate
in 1-L high density polyethylene (HDPE) bottles, while samples for analysis of metals were collected in duplicate in
250-mL Nalgene® HDPE bottles. Metal samples were preserved with a sufficient volume of nitric acid to reduce
sample pH below 2.5. Samples for analysis of particle size distribution were collected in 4-L HDPE Cubitainer®
containers. All bottles were pre-cleaned to meet or exceed U.S. EPA Contract Laboratory Program analyte
specifications and detection limits and rinsed with a small quantity 'of sample water before sample collection.
Precautions were taken to minimize disturbance of the samples and contact with air during sample collection. Upon
collection, samples were sealed, labelled, and packed in ice chests with packing foam and ice for overnight delivery
to the analytical lab.
All laboratory analyses were conducted at the DRI Water Analysis Laboratory in Reno, Nevada. Determinations
of sodium, potassium, calcium, and magnesium were made by direct aspiration atomic absorption (U.S. EPA, 1979);
alkalinity by automated electrometric titration (U.S.G.S., 1979); sulfate by ion chromatography (U.S. EPA, 1984);
nitrate by automated cadmium reduction colorimetry (U.S. EPA, 1979); chloride by automated ferricyanide AAl
colorimetry (U.S. EPA, 1979); and silica by automated molybdate blue colorimetry (U.S.G.S., 1979). Barium,
chromium, iron, and nickel were analyzed by direct aspiration inductively coupled plasma (U.S. EPA, 1979);
cadmium, manganese, and lead by direct aspiration atomic absorption (U.S. EPA, 1979); and arsenic by hydride
generation atomic absorption (U.S.G.S., 1985). Analytical precision and bias were evaluated using the procedures
outlined in "Standard Methods for the Examination of Water and Wastewater" (1992). Within every analysis set, a
minimum of 10 percent laboratory-spiked samples, 10 percent laboratory duplicate samples, and 5 percent EPA
reference samples and/or natural water control samples were analyzed. Precision data for each analysis are presented
in Table 4. The quality control limits consist of a warning limit of 2 times the standard deviation and a control limit
of 3 times the standard deviation. Samples exceeding the control limit were reanalyzed.. No evidence of analytical bias
was detected and all results fell within the established quality control limits.
Estimates of particle grain size distribution were determined gravimetrically by serial ultrafiltration using EPA
Method 160.2 (U.S. EPA, 1984) and microfilters of 5.0 um, 0.4 jim, 0.1 um, and 0.03 urn pore size. The analytical
detection limit for this method was 0.1 mg/L.
The results were examined by multivariate and univariate analyses of variance using the BMDP (BMDP
Statistical Software, Inc., 1988) and Minitab (Minitab, Inc., 1989).software packages. The study was treated as a
complete randomized block design, with wells as blocks and the various device-filtration method combinations as
treatments. The replicate observations of each treatment allowed computation of experimental error as well as
sampling error.
The distribution of species in equilibrium under the hydrogeochemical conditions observed at each well were
estimated using the geochemical modeling program PHREEQE (Parkhurst et al., 1980). Thermodynamic data were
those provided with the program and those compiled by Drever (1982) and Woods and Garrels (1987).
11
-------
TABLE 4. ANALYTICAL METHODS AND QUALITY CONTROL DATA
Constituent
pH
EC
HCO3
C1
SO4
Na
K
Ca
Mg
Si
NO3-N
TDS
TOC
As
Ba
Cd
Cr
Fa
Pb
Mn
Nt
Method
EPA 150.1
EPA 120.1
USGS I-2030-78
EPA 325.1
EPA 300.0
EPA 273.1
EPA 258.1
EPA 21 5.1
EPA 242.1
USGS I-2700-78
EPA 353.2
EPA 160.1
EPA 41 5.1
USGS I-3062-85
EPA 200.7
EPA 213.1
EPA 200.7
EPA 200.7
EPA 239.1
EPA 243.1
EPA 200.7
Method Detection
Limit (mg/L)
0.02
2.2
1.6
0.2
0.1
0.01
0.01
0.03
0.01
0.05
0.003
2.0
0.2
0.001
0.002
0.002
0.01
0.01
0.02
0.01
001
Control Average
(mg/L)
8.47
224
128
90.2
10.0
10.0
9.4
70.8
18.5
71.6
1.4
168
26.1
0.21
1.0
0.026
0.15
0.078
0.11
0.34
0.10
Percent
Deviation
0.4
0.7
0.5
1.7
0.2
1.6
1.5
1.8
1.6
1.8
2.0
2.6
4.7
6.8
1.6
7.5
2.1
6.1
10.2
1.2
6.8
12
-------
SECTION 5
RESULTS AND DISCUSSION
Field hydrogeochemical conditions and particle concentrations, both site-specific and device-specific, provide
the framework for interpretation of the metals data. For this reason, field parameters, particle size distributions, and
total particle concentrations will be presented first, followed by the trace metal and major ion results.
IMPACTS ON FIELD PARAMETERS
Three important factors that influence the accuracy of field measurements of unstable parameters during
sampling from conventional standpipe monitoring wells are sampling method measurement techniques, and
hydraulics of the well. Evaluation of the impacts introduced by selected sampling methods was one of the primary
objectives of this study and will be discussed below. Impacts related to measurement techniques were considered low
because a single individual conducted all the field measurements and all procedures followed established protocol.
In contrast, well hydraulics had an impact on values of field parameters in some of the sampling events that
masked all other factors. In particular, when the discharge rate of the sampling device exceeded the yield of the well,
drawdown occurring in the well caused one or both of the following responses. First, increased hydraulic gradients
disturbed the sampling zone and mobilized large volumes of particles into the well, thereby elevating turbidity values.
DO levels also increased as dewatering of the filter pack created a larger air-water interface for transfer of oxygen.
These factors led to highly biased equilibrium values and/or greatly increased purged volumes, as demonstrated at
WASH-1 where discharge rates of approximately 1.0 L/min caused dewatering of the filter pack and concomitant
increases in turbidity and DO (Figure 1). Note that the lower discharge rate, which was probably closer to the yield
of the well, resulted in lower variability in these parameters.
Second, performance of the centrifugal pumps decreased as drawdown and therefore total head on the pump
system increased, sometimes leading to a complete loss of discharge. Resumption of pump operation was often
accompanied by a large surge and large changes in several indicator parameters, particularly turbidity and DO. Like
dewatering the filter pack, this effect resulted in biased equilibrium values and/or increased purged volumes. In
addition, operation of the centrifugal pump at low discharge rates caused noticeable increases in discharge
temperature. This response was most pronounced at WISC-4 where the yield was sufficient to prevent dewatering of
the filter pack and screen but not to prevent drawdown in the well. As a result, CP1 sampling at this well was
complicated by variations in discharge rate which impaired the ability to reach equilibrium values of turbidity, DO,
and temperature, and greatly increased purged volumes (Figure 2).
The yield of well WISC-1 was low enough that more than 24 hours were required after purging for the well to
refill. In addition, the limited volume of water available prevented measurement of all field parameters and collection
of the full suite of samples. For these reasons, and because the well hydraulics greatly differed from other wells in the
study, only limited sampling was conducted there and the results were not included in the statistical analysis.
High-yield wells, such as NEV-2 (Figure 3), produced less variable results than the wells discussed above, with
the exceptions of pump discharge temperature and several field parameters from the bailer.
13
-------
10 20 30 40 50
Volume Purged (L)
60
Figure 1. Impacts of dewatering filter pack on turbidity and dissolved oxygen at WASH-1.
Note the lower and stable values of dissolved oxygen and turbidity at the discharge
rate of 300 mL/min (CP1). Significant dewatering of the filter pack occured at 22 L
forBPandCP2.
DO concentrations were consistently higher in bailed samples than pumped samples (Figure 4 and Appendix
A). Bailing produced DO concentrations of approximately 3.0 to 5.0 mg/L, 10 to 20 times higher than final pumped
concentrations in the higher-yield wells, and showed minimal decline during purging. In contrast, the pump methods
typically produced maximum values at the initiation of pumping (between approximately 25 and 50 times higher than
final values in higher-yield wells) followed by an exponential decline to equilibrium values. The DO response
observed in the pump results is consistent with the progressive removal of stagnant DO-charged water in the well,
pump body, and discharge tubing.
As these results demonstrate, the concentration of DO and its variability during the well purging process is often
highly sensitive to sample collection method. Sampling devices that excessively agitate the sample and/or the
sampling zone may lead to aeration of the sample and elevated DO concentrations. For example, bailers subject their
samples and the water column within the well to considerable agitation and aeration as the bailer is lowered and raised
from the sampling zone. Sample aeration may also occur when the samples are transferred to a beaker for indicator
measurements and during the time measurements are conducted. Additionally, oxygen (and other gases) may diffuse
through the flexible tubing used for submersible pump discharge (Holm et al., 1988), although this effect appears to
occur on a much smaller scale. Enrichment of DO during the sampling process not only produces samples that are
unrepresentative with respect to oxygen but may result in oxidation and subsequent precipitation of reduced species,
such as ferrous iron, and coprecipitation of other metals species. Alteration of dissolved constituents in this way during
sampling may lead to erroneous conclusions about their concentrations or speciation.
14
-------
150
T~ • T««» , . .
I 1 1 I I I I I I f I I I I I I I I I I I I I I I I I I I I I I
6.0
4.0
D)
o n
Q 2.0
0.0
I til 11 I l
CP1
BP
CP2
0 10 20 30 40 50 60
Volume Purged (L)
70
Figure 2. Impacts of variations in discharge rate of CP1 and CP2 on equilibration of
field parameters at WISC-4. In contrast, note relatively smooth response of BP.
Equilibrium DO concentrations appeared to depend on hydraulic conditions of the well and not pump type or
discharge rate. In the moderate- to high-yield wells, the pumping methods all produced DO concentrations of 0.4 mg/L
or less. In lower-yield wells (WISC-4, WASH-1), DO concentrations were much higher as drawdown introduced more
oxygenated water to the pump intake.
Although volumes required to reach equilibrium DO concentrations varied between the pumps, there were no
consistent relationships. On the other hand, the pumping methods attained DO equilibrium before the bailer, which
reached equilibrium at only one well. The nature of the bailing process causes considerable variability in DO
concentrations between subsequent bails, thereby often preventing equilibrium DO conditions. At six wells, pumping
required removal of 4.4 to 23.6 L to reach equilibrium DO concentrations (Figure 5 and Appendix A), corresponding
to less than 3.8 screen volumes for five of the wells and 10 screen volumes for NEV-2, which had a screen volume
of only 1.2 L. The higher purged volumes at WISC-4 were related to operational problems with CP1 and CP2, and
at WASH-1 the higher volumes were due to low well yield. For the pumps, the purged volumes required to reach DO
equilibrium were higher than for all other indicator parameters except turbidity and CP1/CP2 temperature. These
results indicate that DO is sensitive to the purging process and further suggest that DO may be an important indicator
of the volume required to remove stagnant water from the sampling system.
15
-------
50
T3
n i n i i i i i n i i i i i i i i ri i i i i i i ri i i i IT
Bailer final turbidity = 448 NTU
22
21
20
19
7.5
£7.0
I I I I I I I I 1 I I I I I I I I II Ll L I I t I I
3 C |, I lr I I I 1 _t Itlltlllltltll |._.|_. JL.f__P. '.-'- *--!—
27
Q.
23
21
10 20 30
Volume Purged (L)
Figure 3. Trends of field parameters in high-yield well NEV-2.
16
-------
5 0 i
40-
^30-
O 20-
Q
1 0-
0 0-
stabilization criterion =+/-0.
r^~
$
* *
10m
*
r?n
™N N V.
^^
ZZXXXXyR
X
N
s
*
* c~^
ra-r-
H
x
^
tr-i — i —
^
S
*
r,
mm
mm
Z
•M
=
mm
mam
*
X
X
X
x
X
X
X
X
X
6.0
&-\^. ,,.
*
[S
X
J^
^
r^
\
\
\
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2 WASH-1 WASH-2 WASH-3
30.0-
stabilization criterion =+/-0.2 C
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2 WASH-1 WASH-2 WASH-3
9 0-
8 0
X
Q.
7.0
6.0-
5.0-
stabilization criterion = +/-0.10
\
mm
MM
HUB
mm
mm
mm
MM
^
x
k_
z
z
z
z
N
t
X
7~
MM
mm
HMM
mm
mm
MM
MM
~w.
^
\
®&
M
i
&
x
y
mm
— •
\
^
Z
/
mm
mm
mm
mm
MM
Mi
mm
mm
Mm
1
i
\
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2 WASH-1
300 i
225-
u150
75-
0-
stabilization criterion =+/-3% full scale range
^ (EC x 100) fECxIOO^
z
z
z
;/
mm
mm
MX
•
x
X
x
kXXXXXXXX
M
-;
\
x
X
X
X
(ECX10, ^1
?
z
y
mm
i
vs
' •••
-^i
X
X
\
'/
/
mm
mm
mm
mm*
rxxxxxxxxxi
I
i
1
—
WASH-2
I
h
V'
zzxxx
WASH-3
(ECxIO
?
MM
MM
WlSC-2 WISC-3 WISC-4 NEV-1 NEV-2 WASH-1
stabilization criterion =+/-1 .0 NTU * * * 800^
20.0-
1- 150-
^100-
(550-
0 0-
z
^
X
X
X
170
N
«i
s«
"«
X
x
X
X
19
71
XXXXX
z
z
*
s
X
X
>2000
X
X
X
x
67.2 44f
I71_
X
^
ZZXXXXXy
160
*
^
=
MM
^
MM
MB
^
zxxxxxx/
'/
/
Mm
MM
•M
MM
il
S
(ECxIO)
WASH-2
*
1160
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2 WASH-1
I
mm
t
I
^s
/////
\
\
5$
MMl
MM
MMH
^p
X
WASH-3
735
*
IZSfl
^
^
WASH-2 WASH-3
or parameter not measured R PI HI d
brium when samples collected ^ m
03 O 03 O
» -o ti TJ
0 M
N = device not tested
* parameter not at equil
Figure 4. Equilibrium values of field parameters (note that parameters of bailed samples at
the WISC wells were measured only at time of sample collection).
17
-------
/o.u-
"T CA A
*> oe A
0 0
Dissolved Oxygen
*
SkHJLJ
*
•i:
^M
>
] N
! N
S
V
N
\
S
s
p \/&~\
s138
s
s
s
>
s
s
V
'H
3
*
s
\
s,
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2
7-0 *
o
Temperature
*
* * Y'J
'/% P HJlp 2
)
—
!
„
|_ I
••
* N
S
V
! N
!, S
S
i ^
I S
-— ^"^ — **
P j>—'^\
s 138
S
V
s
N
S
s
N
S
5?
mat
mm
1
\
\
°'° WISC-2 WISC-3 WISC-4 NEV-1 NEV-2
/O.U-
*
I
*
i
mm
mm
*
i
* ^c
v N m
^ p:3is r/=
WASH-1 WASH-2 WASH-3
*
/
*
MM
MM*
3
^
*
* P *
nJHaJl
*
i
|
WASH-1 WASH-2 WASH-3
PH
•
r.enp _Mnp ^P __ pi
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2
5U.U-
0
0.0-
75.0
IP
ii
^^J?i^
WASH-1 WASH-2 WASH-3
Electrical Conductivity
^
[^Hlp HM IP r7§~lp FLJ
^
^ K
L
I
1
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2
*
Turbidity
*
"1
P
^
'••
ryp—
I
J^'i
3
E3 CPI
B BP
EiH CP2
C3 Bailer
0)
not measured
lies collected at preestablished vo
n samples collected
WASH-1 WASH-2 WASH-3 f | •=
S138
S
s
s
*
a
3*
° 2 '5
TJ 3 CT
to y> to
Oj W W
n *
Figure 5. Purged volumes required to reach equilibrium values of field parameters.
18
-------
As with DO, turbidity exhibited a strong dependence on sampling method. The highest turbidity values were
obtained with the bailers, while the lowest turbidities were obtained with the pumps. In fact, the highest bailed turbidity
values obtained were as much as two orders-of-magnitude greater than samples pumped from the same well.
Equilibration of turbidity, like DO and Eh, is often related to sample collection method. Agitation of the sampling zone
by motion of the sampling device (as with a bailer) often suspends and entrains particles in the sample discharge
leading to elevated sample turbidity. Additionally, discharge rates that exceed the yield of the water-bearing zone may
mobilize normally immobile particles, leading to elevated turbidity and biased analyses of trace metals and other ions.
Even with very careful operation, the repeated motion of the 4.1-cm-diameter bailer inside 5.1-cm-diameter casing
and screen caused a surging action that mobilized significantly more particles than were mobilized with the pumps.
In addition, bailer-produced turbidity values showed considerable variability between measurements and often did
not reach equilibrium values, indicating the bailing process was significantly disturbing the sampling zone.
With the exception of those wells where steady discharge could not be maintained (discussed below), turbidity
values during pumping generally followed a trend of maximum values at the initiation of pumping followed by an
exponential decline to equilibrium values. The initial maximum may result from particles formed or collected within
the well casing between sampling events as described by Backus et al. (1993), disturbance of the sampling zone during
emplacement of the portable pumps as observed by Kearle et al. (1992), or particles mobilized and then settled out
after previous sampling by bailers.
Evaluation of the turbidity data from the pumps was complicated by operational problems experienced with the
centrifugal pump, the hydraulic responses of the lower-yield wells, and the order in which the experiments were
conducted. In wells where a steady discharge was difficult to maintain with the centrifugal pump (particularly WISC-4
and WASH-2), variation in discharge rate occurred during the adjustment of the pump speed. These variations were
occasionally high enough to mobilize and entrain particles that had previously been undisturbed, thereby elevating
turbidity and increasing the volume required to reach equilibrium. In the case of WISC-4, operational problems
prevented equilibrium from being reached for turbidity, DO, and temperature. Hydraulic conditions at low-yield well
. WASH-1 presented a different problem but with similar results. The low yield of this well caused significant drawdown
and dewatering of the filter pack when purging at rates greater than at least 900 mL/min (the lower rate caused
drawdown but did not dewater the filter pack to the pump intake). As the filter pack was dewatered, the increased
hydraulic gradient and agitated conditions in the water-bearing zone appeared to have mobilized normally undisturbed
particles leading to increased turbidity values. Finally, the order in which the experiments were conducted in each well
may also have effect time to equilibration and final turbidity values. As previously discussed, sampling events using
low discharge rates were conducted prior to the moderate rate events in seven of the eight wells and bailer samples
were collected prior to pumped samples in four wells. In addition, all the wells had been extensively sampled with
bailers prior to this study. It is likely that a significant fraction of particles present in the well, filter pack, and adjacent
water-bearing zone at the initiation of the low rate pump experiments were an artifact of earlier bailing events. As a
result, the first pump experiment may have acted as further "development" of the sampling zone, causing higher initial
turbidity values and longer times to equilibration during the first pump experiment.
Omitting the pump experiments where turbidity was highly dependent on well hydraulics and/or pump
performance (WISC-4 and WASH-1), equilibration was attained after pumping 0.7 to 29.8 L (corresponding to
approximately 2 to 7.7 screen volumes). Variation was evident in final turbidity values and times to equilibration for
the pumps but no distinct relationships between device and these parameters existed. The differences were all within
+/-5.0NTU.
Values of pH showed little variation between pump methods with most values falling within the range of +/-
0.2 pH units for a given well. Bailed pH values were also within this range but were usually higher than pumped values,
possibly reflecting degassing of CO2 from the samples during collection and pH measurement. In addition, pH reached
equilibrium at lower purged volumes than all the other parameters, independent of sampling method. Although pH
is an important indicator of the speciation of trace metals in ground water, the relatively uniform values across devices
19
-------
at individual wells do not alone suggest that similar metals species might be present. Other factors, such as redox
conditions and particle concentration, may play more important roles.
Impacts of sampling on ambient ground-water temperature were due to: (1) equilibration between the sampling
device temperature and ground-water temperature subsequent to installation of the device in the well, (2) removal of
stagnant water from the well of a temperature different from ambient ground water, (3) air temperature changes during
sampling, and (4) heat generated by operation of CP1 and CP2. The first two factors were resolved by the purging
process, the third was minimized but not eliminated by limiting the length of tubing used with the pumps, and the fourth
was related to the operational design of the centrifugal pumps. Equilibrium temperatures of BP and bailer discharge
typically were similar for the same wells at the Nevada and Washington sites, following a smooth trend from stagnant
conditions. Bailer temperatures were more variable at the Wisconsin site because sampling procedures and
measurement method were slightly different there. In contrast, equilibrium temperatures of the centrifugal pumps,
CP1 in particular, were as much as 5.2 °C higher than BP values for the same well. CP1 and CP2 discharge
temperatures also showed considerable variation in response to changing pump discharge rates during purging. This
effect of elevated temperatures appears to have been caused by heat produced from the pump motor operation and was
exacerbated at lower discharge rates when less water passed over the motor housing.
For the BP and bailer, volumes purged to reach equilibrium temperatures were generally lower than for turbidity
and DO. The elevated temperatures associated with CP1 and CP2 caused purged volumes to be approximately equal
or greater than for the other indicator parameters, with the exception of turbidity. In most cases, temperatures of CP1
and CP2 did not reach equilibrium.
EC is generally considered to be independent of sampling method (when appropriate materials are utilized in
the construction of the device) because there is little opportunity for the device to impact sample quality. With few
exceptions, equilibrium conductivity values produced by the four methods were within +/- 5 percent of full scale range.
Furthermore, purged volumes required to reach equilibrium were lower than for all other parameters except pH,
indicating that under these conditions EC may not be indicative of purging completion.
Measurements of Eh should reflect the intensity of oxidizing or reducing conditions in the ground-water
environment but are often significantly altered by sample collection method. In situ ground water is generally
considered to be at a relatively stable redox state, but when samples are exposed to air, the redox system may be
overwhelmed by reactions involving oxygen (Hem, 1985). Eh is an important control on the distribution of species
in ground water and changes in Eh during sampling may alter the dominant species present. Sampling methods that
result in excessive exposure to air may increase Eh and cause shifts from dissolved species to solids (precipitates) that
may be removed during filtration. Additionally, elevated Eh values may lead to incorrect conclusions about the
distribution of species in the ground water. In this study, variability in Eh measurements generally masked any distinct
trends that might have been related to the effects of sampling method, although the bailer often produced Eh values
that were slightly higher than those obtained with the pumps. By contrast, the lower extent of disturbance of the
sampling zone assumed to be associated with CP1 did not lead to lower, or presumably more representative, values
of Eh. Volume purged to reach equilibrium Eh values was not available because the parameter was not used as a
purging criterion and for that reason was not measured continuously.
IMPACTS ON PARTICLE SIZE DISTRIBUTION AND CONCENTRATION
Particle concentration and size distribution were analyzed in the laboratory by serial ultrafiltration. This method
provides a reasonable approximation of particle size distribution (Kingston and Whitbeck, 1990), but is dependent
on sample handling and analytical holding times which may skew the data toward larger particle sizes through
precipitation, nucleation, and aggregation processes. For this reason, the analytical error of ± 1.0 mg must be kept in
mind when interpreting the particle size results. The total concentrations and size distributions of particles entrained
in samples were expected to be related to hydrogeologic conditions and well construction details at each well site and
20
-------
certain sampling variables, particularly sampling device type and discharge rate. Total particle concentrations in
samples pumped from the higher-yield wells ranged from values below 2.8 mg/L at WISC-2 and WISC-3 up to values
between 8.9 and41.1 mg/LatWASH-2and WASH-3 (Figure 6 and Appendix B). Concentrations atNEV-1 andNEV-2
were between 2.7 and 13.3 mg/L. Because essentially the same sampling techniques were utilized at each site, the
observed range in particle concentrations between the sites may reflect differences in methods of well installation,
well development, sampling history, and hydrogeologic and hydrogeochemical conditions in the water-bearing zone.
Sampling device had an important effect on total particle concentration and particle size distribution at most
wells. Samples bailed from the higher-yield wells exhibited particle concentrations ranging from 40.3 mg/L at
WASH-3 to 818 mg/L at WASH-2. Even greater concentrations were obtained when bailing from the lower-yield
wells; 845 mg/L at WASH-1 and 6970 mg/L at WISC-4. At six of the wells, the bailers produced total particle
concentrations that ranged from 6.5 to 17,000 times greater than those produced by the pumps in the same well. Of
the two remaining wells, WASH-3 particle concentrations showed little variation between the devices (a range of 39.1
to 41.1 mg/L) and WASH-2 results were heavily influenced by well hydraulics and pump operation causing CP1 to
produce the highest concentrations.
In addition to generally higher total concentrations, particle size distributions of bailed samples were highly
skewed toward particles greater than 5.0 um in diameter. The bailed samples usually contained over 90 percent of
particle mass in the fraction retained by the 5.0-um filters, while the pumped samples contained a more uniform
distribution of particle sizes. The surging action of the bailer as it passed through the water column clearly mobilized
more and larger particles than most of the tested pumping methods which generally minimized disturbance of the
sampling zone. However, pumped samples from WISC-4 and WASH-1 that contained relatively high particle
concentrations also showed size distributions that were skewed toward the larger particles. Samples exceeding 30
mg/L total particle concentration, regardless of sampling device, contained at least 50 percent by weight of particulate
matter over 5.0 um in size and 95 percent by weight of particulate matter over 0.4 um in size.
High concentrations of particles larger than 5.0 um are more likely related to geologic conditions, drilling
methods, and well installation procedures than ground-water contaminant transport and may bias analytical
determinations of mobile trace metals if not removed from samples. On the other hand, particles smaller than 5.0 um
are potentially mobile in many ground-water environments and may be associated with contaminant transport. The
5.0-um filters removed 75 to nearly 100 percent of the particles in bailed samples, corresponding to 30 to 6960 mg/L
of solids. From 0 to 70 percent of the particles were removed from the pumped samples, although this corresponded
to less than 2.2 mg/L from samples containing less than 30 mg/L total particle concentration, or up to approximately
30 mg/L from samples containing up to 50 mg/L total particles. In addition to the significant particle mass present
as particles larger than 5.0 urn, the bailer also produced concentrations of particles between 0.03 and 5.0 urn (referred
to here as sub-5.0 um particles) that were up to 17 times higher than the mean concentration produced by the pumps
at the same well. As an example,; the pumps at WISC-2 produced a mean concentration of sub-5.0-um particles of
0.5 mg/L, while the bailer produced 8.7 mg/L. This type of relationship was less evident at the WASH wells where
a much greater range in particle size distribution was observed.
The presence of measurable concentrations of sub-5.0-um particles further suggests that removal of a fraction
of them by 0.45 um filtration might bias analytical determinations of mobile contaminant loads. In many samples from
the WASH wells, concentrations of 0.4 to 5.0-um particles were higher than concentrations of sub-0.4-um particles,
indicating that a majority of the particle load would be removed by 0.45-um filtration. At the WISC wells, the fraction
of 0.4 to 5.0-um particles was roughly the same as the fraction of sub-0.4-um particles, indicating that 0.45-um
filtration could conceivably remove half of the potentially mobile particles. At the NEV wells, a larger fraction of
sub-5.0-um particles were smaller than 0.4 um so filtration might have less impact on mobile particle concentrations
at this site.
21
-------
(a)
1500-
IZOO-
9QO-
600-
300-
n.
re N PS ra
s
^
\
s
s
s
s
s
rx,
>
s
s
JZ
__
'u.
's
s
^
v
s
\
V
,,,
WISC-2 WISC-3
WISC-4 NEV-1
Total particle concentrations
(b)
S
S
s
s
s
\
^•M mm
WISC-2 WISC-3 WISC-4 NEV-1 NEV-2 WASH-1 WASH-2 WASH-3
Total particle concentrations, expanded scale
(c)
oq Tl;
T^ C\i
6.0-,
5i 5.0-
<=* 4.0-
^5 3.0-
C—
O '•""
nn.
^
Pi
_^-r-H K
bLO
\
S
S
s
s
PLB
^
_RdN 17
mm*
mm
•Ml
z
1
•M
M«
i^-vi-r-H-
O=B§I
WISC-2 WISC-3
Concentrations of particles < 0.40 um in size
P7I CP1 H BP 151 CP2 K3 Bailer
N = device not tested B = below detection
Figure 6. Results of particle analyses: (a) total particle concentrations; (b) total particle
concentrations, expanded scale to show results of pumped samples; (c) concentrations
of particles less than 0.4 urn in size.
22
-------
Differences in concentrations of sub-0.4-um particles between bailers and pumps were less than for sub-5.0-um
and total particles but the bailers still produced higher sub-0.4 urn particle concentrations at seven of the eight wells.
Bailed samples contained sub-0.4-um particle concentrations from 0.4 to 6.7 mg/L higher than the mean values of
the pumped samples from the same well (Figure 6), suggesting that determinations of colloidally-associated
contaminants might be biased.
Although distinct differences in both size distribution and total particle concentration were evident between the
pumped samples (taken as a group) and the bailed samples, there were no consistent correlations between particle
content and pump type or pump discharge rate. Under the conditions of these experiments, collecting samples at a
discharge rate of 1000 mL/min did not consistently mobilize more or larger particles than collecting samples at 300
mL/min. Likewise, individual pumps did not produce consistent particle size distributions or concentration
relationships. Interpump variability of these factors appeared to be caused by external effects such as order of the
experiments, disturbance of the sampling zone during pump installation, well hydraulics, and other sampling and
analytical errors.
A general relationship between turbidity and total particle concentration was observed, particularly for the
higher particle concentrations (Figure 7). As a rule, the methods that produced samples with the highest particle
concentrations also produced the highest levels of light scatter, and therefore turbidity, in the field. The apparent
non-significant linear regression (although there was no lack of fit) observed for samples with particle concentrations
less than 50 mg/L may be a result of deviations from the best-fit line caused by several factors. First, the distinct
ground-water color at both Nevada wells might have reduced apparent turbidity values by reducing the amount of
scattered light measured by the turbidimeter. Turbidity measurements at the NEV wells were consistently lower than
at the WISC wells, despite higher particle concentrations. Second, the first pump experiment in each well almost
always produced the highest turbidity of the three pumps in that well, while the particle concentrations were virtually
the same as in the other pump experiments. This may be an artifact of the bailing process which occurred before the
pump experiments in the WISC and NEV wells or natural colloid accumulation (particularly in the WASH wells where
bailing was conducted after the pumped samples were collected). These factors might have caused accumulation of
large numbers of particles less than 0.03 um in diameter which were not detected in the particle size analysis but
contributed to light scattering in the turbidity measurements. Finally, oxidation of dissolved ferrous iron in particle
samples from the Washington wells may have elevated total particle concentrations in these samples, particularly
WASH-3 which showed total iron concentrations of 17 to 21 mg/L. This process would not affect the in-line turbidity
measurements and therefore would lead to deviations from the best-fit line.
It is important to keep in mind that the overall turbidity/particle relationship presented here includes the effects
of four sampling methods at eight wells at three study sites. Despite this, the general relationship of turbidity to particle
concentration strongly suggests that turbidity provides a useful indicator of the relative presence of particles in ground
water. Furthermore, utilizing a single sampling device or sampling at an individual well would likely reduce or
eliminate much of the variability described above. As a result, the relationships of turbidity to particle concentration
and the sensitivity of turbidity to the purging process, relative to other indicator parameters, suggest that turbidity may
be a useful indicator of the stabilization of particle concentrations during monitoring well purging. If mobile particles
are thought to be important to the transport of contaminants in ground water, use of indicator parameters such as pH,
temperature, or EC to determine representative sampling conditions may result in underpurging, and consequently, the
collection of inaccurate samples.
IMPACTS OF FILTRATION ON METALS CONCENTRATIONS
Of the seven metals targeted for study, lead and cadmium were not detected in samples from any of the wells
and concentrations of the other five were generally below 50 ug/L. Iron was an exception in that it was detected in
concentrations up to 71 mg/L. The low concentrations of most of the targeted metals suggests that analytical variability
may be an important contributor to the overall variability observed in the metals data. For example, at the detection
23
-------
a.)
b.)
.£•
;o
1
2U
15
10
5
1
°(
_ 1st degree fit, r=0.49
_ * •
'•^VM ' B • ' ' ' ' '
3 10 20 30 40
-
-
I
5
1200
800
1 400
=5
|
i • i 1
1 st degree fit, r=0.87
500 1000
Total Particle Concentration (mg/L)
1500
Figure 7. Plots of turbidity (NTU) versus total particle concentration (mg/L) for (a) samples
containing less than 50 mg/L particles, and (b) samples containing more than 50 mg/L
particles (with the exception of bailer at WISC-4).
level, analytical error may be ± 100% of the concentration (0.01 mg/L for all detected metals except arsenic). This
variability must be kept in mind when interpreting the very low concentrations or comparing similar concentrations
in the data set. Analytical results are contained in Appendix C. Estimates of the speciation of detected trace metals
were made using the geochemical modeling program PHREEQE.
Of the observed metals, iron was the most likely to form colloids under the hydrochemical conditions present
at the study sites, primarily as iron hydroxides. However, significant concentrations of dissolved ferrous iron may also
have been present under the lower redox and pH conditions of the Washington site. If ferrous iron concentrations were
high, oxidation and precipitation might occur in samples where oxygenation occurred during sample collection. If a
significant fraction of this precipitate was removed during filtration, iron concentrations might be biased. Although
manganese may form oxides in much the same way as iron, it is considerably more stable in aqueous form under oxic
conditions than iron and is commonly present as the aqueous phases Mn2+ or MnHCOs+. Like other cations, aqueous
manganese species are likely to sorb to particle surfaces and therefore might be impacted by filtration. Likewise, the
aqueous form of nickel, Ni2+, might react with particles although nickel carbonate precipitate (NiCOs) might also be
present, depending on pH and redox conditions. Nickel was targeted for analysis only at the Washington site. Barium,
chromium, and arsenic were all expected to be primarily in the aqueous phase. Barium, as aqueous Ba2+, is readily
sorbed to metal oxides or hydroxides and so it might be expected to be impacted by filtration where particles are of
this composition. Chromium was likely to be present as chromate (CrO42~) under conditions in which it was detected
and also may be associated with particles or colloids. Finally, arsenic was likely to be present primarily as arsenate
(HaAsCV) at pH below 7 or arsenite (HAsO42-) at pH above 7. The high potential for arsenate sorption on colloidal
material was demonstrated by Puls and Powell (1992).
Because most samples contained suspended particles and colloids ranging from 0.03 um to over 5.0 urn in size,
it was expected that filtration would impact concentrations of those metals that either existed in the solid phase as
24
-------
particles greater than 0.4 urn, generally oxides of iron, or were associated with particles in this size range, which might
include oxides, clays, minerals other than clays, or organics. Under these conditions, the greatest differences in
concentration were expected to occur in samples collected by methods that produced the highest particle
concentrations. These concentration differences were represented by their "filtration ratio," the ratio of unfiltered
concentration to filtered concentration (Table 5). A filtration ratio of 1.00 indicates no difference between filtered and
unfiltered concentrations, while a filtration ratio of zero (0) indicates that both filtered and unfiltered concentrations
were below analytical detection levels. Filtration ratios denoted as undefined (und) indicate that an unfiltered
concentration was detected but that the corresponding filtered concentration was below the analytical detection level.
Filtration ratios found to be undefined are important because they indicate that all trace metals detected in unfiltered
samples were removed during the filtration process.
An exploratory statistical analysis of the metals data set was conducted to identify major factors contributing
to the variability observed in the concentrations. Iron, manganese, and arsenic were chosen for this analysis because
they were detected by most devices in most wells. Concentrations reported as below the analytical detection level were
set at the detection level for that particular analyte. In addition, a log transformation of the data set was made due to
the skewed distribution and the large range in data values (two to three orders of magnitude). Well WISC-3 was not
included in this analysis because only three of the four sampling devices were tested there.
The multivariate ANOVA indicated that the effect of sampling device was marginally significant and that the
effects of filtration method and device-filtration interactions were not significant at the 0.05 level. These results reflect
the highly variable hydrogeochemical conditions at each well and the fact that individual metals and metals species
responded to filtration and device differently, depending on these unique conditions.
Univariate ANOVAs were subsequently conducted to investigate the response of individual metal ions to the
experimental factors. These analyses incorporated the same seven wells as the multivariate ANOVA, unfiltered and
0.45-um-filtered samples, all four sampling devices, and the metals iron, manganese, and arsenic. Univariate
ANOVAs were also conducted for the barium data from WISC-2, WISC-4, NEV-1, and NEV-2 (barium was not
analyzed in samples from the Washington site) and the nickel data from WASH-1, WASH-2, and WASH-3 (nickel was
not analyzed in samples from the other two sites). The results showed a significant filtration effect for iron, manganese,
and arsenic at the 0.025 significant level suggesting that concentrations of these ions showed consistent responses to
0.45-um filtration compared to no filtration. In contrast, barium and nickel did not show significant responses to the
effects of 0.45-u.m filtration.
The effects of sampling device appeared significant for manganese and barium and marginally significant for
iron at the 0.05 significance level. However, the results were biased by large effects at one or two wells and the other
differences generally were not analytically significant, particularly for manganese. As a result, overall concentration
trends (including both 0.45-u.m-filtered and unfiltered samples) were not strong enough to indicate that the type of
sampling device was an important contributor to concentration variation. Device-filtration interactions were
significant at the 0.05 level for iron and barium, marginally significant at the 0.05 level for manganese, and not
significant for arsenic and nickel. The effects of device-filtration interactions are clearly seen at wells where,sampling
devices produced high particle concentrations (Figures 8 through 12).
The greatest impacts of filtration on metals concentrations were observed at WISC-4 and WASH-1. Although
filtration ratios were greatest for samples that contained the highest particle concentrations, virtually all samples were
affected, regardless of sampling method. For samples with particle concentrations between 200 and 6970 mg/L (Bailer
at WISC-4; BP, CP2, and Bailer at WASH-1), iron filtration ratios were 3400 for CP2 at WASH-1 and undefined for
the other methods. Because over 95 percent of the particles in these samples were larger than 0.4 urn and therefore
were removed during filtration, most of the solid iron hydroxide and any other associated iron species were also
removed.
25
-------
TABLE 5. FILTRATION RATIOS FOR METALS CONCENTRATIONS
ID
WISC-2
CP1, 0.45 nm
CP1,5.0nm
BP, 0.45 urn
BP, 5.0 urn
CP2, 0.45 urn
CP2, 5.0 urn
Bail, 0.45 urn
WISC-3
CP1, 0.45 urn
CP2, 0.45 urn
Ball, 0.45 urn
WISC-4
CP1, 0.45 urn
BP, 0.45 urn
CP2, 0.45 urn
Bail, 0.45 (im
NEV-1
CP1, 0.45 urn
CP1, 5.0 urn
BP, 0.45 jim
BP, 5.0 urn
CP2, 0.45 urn
CP2, 5.0 \un
Bail, 0.45 urn
Bail, 5.0 urn
NEV-2
CP1, 0.45 urn
BP, 0.45 urn
CP2.0.45 urn
Bail, 0.45 urn
WASH-1
CP1, 0.45 urn
BP, 0.45 urn
CP2, 0.45 urn
Bail, 0.45 urn
WASH-2
CP1, 0.45 urn
CP1, 5.0 nm
BP, 0.45 urn
CP2, 0.45 urn
Bail, 0.45 urn
WASH-3
CP1, 0.45 urn
CP1,5.0nm
BP, 0.45 urn
BP, 5.0 urn
CP2, 0.45 urn
CP2,5.0nm
Bail, 0.45 urn
Bail, 5.0 urn
Cr
0
0
0
0
0
0
0
0
0
0
0
0
0
und
0
0
0
0
und
1.00
0
0
0
0
und
und
und
0
und
und
0
0
0
0
und
0
0
0
0
0
0
0
0
Fe
5.00
1.25
2.33
3.50
0
5.50
340
5.00
2.50
81.1
100
1.50
11.4
und
1.00
1.00
1.00
1.00
1.25
1.00
34.5
19.7
1.20
1.00
1.00
350
81.0
und
3400
und
1.29
1.04
1.16
1.21
16.4
1.01
1.00
1.03
1.03
1.16
1.05
1.01
1.02
Mn
und
und
0
0
0
0
11.0
und
0
7.50
2.00
0.80
1.25
12.4
1.01
0.99
1.00
1.00
1.00
1.00
1.07
1.08
0.97
0.98
1.01
1.34
1.88
3.17
15.0
12.9
1.00
1.00
1.00
1.00
4.75
1.01
1.01
1.03
1.04
1.13
1.05
1.01
1.02
As
0
0
0
0
0
0
0
0
0
0
0
0
0
und
0.93
1.00
1.00
1.09
1.07
1.07
0.94
1.07
1.00
1.00
1.09
1.09
1.40
1.33
3.00
3.33
0
0
0.
0
2.00
0.88
0.88
0.88
0.88
1.00
1.00
1.20
1.00
Ba
1.00
1.00
1.00
1.00
1.00
1.00
3.08
0.83
0.90
4.38
1.10
0.99
1.03
3.64
1.08
1.04
1.00
1.00
1.04
1.04
1.88
1.81
1.00
1.04
1.09
15.6
-
-
-
-
-
-
-
—
-
-
-
- '
-
-
-
-
—
Ni
-
-
—
—
—
-
-
-
-
—
-
-
-
—
-
-
-
-
-
—
-
—
•
-
-
—
4.00
2.00
18.0
und
1.00
und
0
0
und
1.33
1.00
1.00
1.00
1.00
1.00
1.00
0.80
26
-------
Concentration (mg/L)
O 0 O O O C
o o '-^ '-*• ro h
o 01 o 01 o c
(3.4)
y~l
/
/
J
t
/
/
/
/
/
/
••
/
/
/
/
/
/
/
/
/
/
/
/
/
— 'fit
CP1 BP CP2 Bail
0.20
0.15
0.10
0.05
noo
(6.5)
/
/
/
3 nt
/
/
3
/
/
/
/
/
/
/
/
/
/
/
/
/
mm
CP1 BP CP2 Bail
0.20
0.15
0.10
0.05
nnn
f
/
/
/
/
/
/
/
/
/
/
/
/
t
(2.0) (0.5C
n
n 1
y
/
/
/
/
/
/
/
/
/
/
/
/
Mi
«M
m
/
/
/
/
/
/
/
/
/
/
/
/
/
f
(71.0
*
CP1 BP CP2 Bail
NEV-1
-i 0 20
D)
£j
c 015
0
2
c 0.10
-------
0.20
0.15
0.10
0.05
0.00
CP1 BP CP2 Bail
0.25
0.20
0.15
0.10
0.05
nnn
WISC-3
n * nt * *
/
>
/
/
/
/
/
3
CP1 BP CP2 Bail
0.25
0.20
0.15
0.10
0.05
nnn
WISC-4
ife
/
/
•HI
MM
/
/
i
/
y
/
/
/
/
(1.35
MM
Ml
Ml
CP1 BP CP2 Bail
NEV-1
C.AI
1 F
•\ o
0,8
0.4
0.0
/-
~/Z
/:
^ ••
S,\ .
/-
CP1 BP
/ —
/-
-
I
/
= ;_
•;
CP2 Bail
1.6
1.2
0.8
0.4
n n
/
/
/
/
/
/
MI
••
M.
Ml
/
/
/
/
/
~
•H
••
HN
/
y
<
Ml
=
7\
',
a
E
M
V\ Unfiltered
@ 0.45-um filtered
H 5.0-um filtered
nt Device not tested
* Below detection level
CP1 BP CP2 Bail
U.kib
0.20
.10
0.10
0.05
n.n
(1.21)
-
j
/
/
y
/
/
/
/
^
.•
Ml
MM
/
/
/
/
/
/
/
/
/
i
/
1
/
/
/
/
/
/
/
MM
Mi
Ml
Ml
/
/
/
(0.90
~
«•
U.iiO
0.20
0.15
0.10
0.0
ranra ra
^
^
^
/
1
o.u
4.0
3.0
2.0
1.0
0.0
/
/
/
/
/
/
/
/
! !
Ml
M>
MM
MH
P]
/
/
/
/
Mi
MH
7
/
/
/
Mi
Ml
•I
MM
M*
Ml
7
/
/
/
HH
M
CP1 BP CP2 Bail
CP1 BP CP2 Bail
Figure 9. Manganese concentrations
28
-------
Barium Concentration (mg/L)
0.08
0.06
0.04
0.02
0.00
/
/
/
/
/
/
f-lnt
CP1 BP CP2 Bail
0.08
0.06
0.04
0.02
nnn
C^B nt PS
7
/
/
/
/
/
/
|
CP1
BP CP2 Bail
0.18
0.16
0.14
0.12
01
7
/
/
/
/
/
/
/
4
mm
MM
Ml
MM
•1
/
/
,/
/
/
/
/
/
MM
Hi
••
MM
M
Mi
MM
,_^
P"
/
/
/
/
/
/
/
/
s
mm
Ml
«w
••
.•
mm
mm
/
/
/
/
/
/
/
'/
/
/
/
/
/
/
(0
Ml
Ml
Ml
Ml
Ml
mm
Ml
M
NH
MM
Ml
Ml
.690
CP1 BP CP2 Bail
c
o
1
CD
e
'I
m
0.10
0.08
0.06
0.04
0.02
0.0
NEV-1
CP1 BP CP2 Bail
Nickel Concentration (mg/L)
o o o o c
° b o b b -
o K> f* a> co c.
(0.175)
^
/
3
/
/
5
£
/
^
^
r
/
^
^
CP1 BP CP2 Bail
0.10
OOfl
0 06
0 04
0.02
0 0
NtV-ii
—
/
/
/
M>
MM
CP1
/i
^i
BP
/
/
j
^i
1
1
/
/
/
/,
/
/
/
V
y
y
/
/
0.375
=i
I
1
CP2 Bail
010 WASH~2
0 08
0 06
0 04
0 02
no
m ** »
p
/
/
',
*
Unfiltered
B 0.45-um filtered
Hfl 5.0-um filtered
nt Device not tested
* Below detection level
CP1 BP CP2 Bail
0 08
0 06
n n4
0.02
nn
-
^
/
/
M
MM
.-•
•M
•^
/
$
/
^
-
••
MO
III
"*
-
^
^
/
/
/
/
/
t
mm
mm
mm
mm
-
/
/
/
/
Mi
•M
iH
mm
'ij
V
^~
CP1 BP CP2 Bail
Figure 10. Barium concentrations for Wisconsin and Nevada sites, nickel concentrations for
Washington site (barium not analyzed at Washington site, nickel not analyzed at
Wisconsin and Nevada sites).
29
-------
Concentration (mg/L)
p p p p p
§o -* —*- to
01 0 01 0
WISC-2
*** *** *** ** nf
CP1 BP CP2 Bail
0.20
0.15
0.10
0.05
0.00
WISC-3
** nt ** **
CP1 BP CP2 Bail
0.20
WISC-4
0.15
0.10
0.05
0.0
CP1 BP CP2 Bail
0.20
NEV-1
0.15
0.10
0.05
0.0
..n ... cm Jl
0.20
0.15
0.10
0.05
nn
NEV-2
/
** ** rv* / *
CP1 BP CP2 Bail
CP1 BP CP2 Bail
\A Unfiltered
H 0.45-um filtered
H 5.0-um filtered
nt Device not tested
* Below detection level
0.20
0.15
0.10
n fm
0.0
WASH-1
;
* « *
/
/
/
/
/
/
Y\
* > *
CP1 BP CP2 Bail
0.0
WAOH-i;
** ** ** YA *
n 9fi
n 1^
Oin
n (1*=;
0.0
vvrton— o
CP1 BP CP2 Bail
CP1 BP CP2 Bail
Figure 11. Chromium concentrations
30
-------
WISC-2
5" 0.04
en "•*"
§ 0.03
I
1= 0.02
8
O 001
0 00
0 20
•gj 016
__^
o 012
1
m 0 OR
0
g
o n fi4
0 0
0 05
^004
"&
^T 003
0
1
•g 0.02
0)
o
r9 001
0.0
CP1 BP CP2 Bail
NEV-1
* A '•*
/"! y Z
p-S-i /Z9 yE'
/I" ySl /Z| /Z<
/=• Oil /=! >=J
> Z ' / ~ * y Z 1 / ^ t'
' ** ( > "*• it ^ x ™* '^
_/ U r S ij| ^ ~ ;j / Z : ,-
CP1 BP CP2 Bail
WASH-1
/
ft '
fe KH xi Z /Z ^IS
^E /= >~ /""
CP1 BP CP2 Bail
WASH-2
** Pla
U.Ut)
OO4.
n n^
n r\o
n m
n A,<
n n?
Om
0.0
** ** **[/)*
CP1 BP CP2 Bail
0 Unfiltered
S 0.45-um filtered
H 5.0-um filtered
nt Device not tested
WASH-3
/i^ <=S ^si PSI
CP1 BP CP2 Bail
CP1 BP CP2 Bail
Figure 12. Arsenic concentrations
31
-------
Similar relationships were observed for other metals at these wells, although to a much lesser degree. For
example, the manganese filtration ratio for samples bailed from WISC-4 was 12.4, wliile the largest ratio for pumped
samples was 2.00 (CP1). At WASH-1, the manganese filtration ratio ranged from 1.88 (CP1) to 15.0 (CP2) and
unfiltered manganese concentrations correlated to the total particle concentrations. At WISC-4, barium had a filtration
ratio of 3.64 for bailed samples, while pumps produced ratios of 0.99 to 1.10. Arsenic was detected only in the
unfiltcred bailed sample at WISC-4. Samples with the highest particle concentrations at WASH-1 showed arsenic
filtration ratios of around 3.0ft while samples with low particle concentrations had arsenic filtration ratios less than
1.40 (CP1 and BP). Chromium was detected only in unfiltered samples that contained very high particle
concentrations. These results suggest that significant fractions of manganese, barium, arsenic, and chromium species
were associated with particles that were mobilized during the more vigorous sampling events and that filtration either
removed the metals species entirely or removed significant quantities with the particles. Note, however, that sampling
methods that caused minimal disturbance to the sampling zone in these turbid wells also generally produced either
filtration ratios near 1.00 or resulted in nondetection of the metal.
Despite the relationships evident for samples with high total particle concentrations, samples with lower total
particle concentrations did not necessarily produce filtration ratios near 1.00. For example, total particle
concentrations at WISC-2 and WISC-3 ranged between 0.7 and 2.8 mg/L in the pumped samples and 122 and 180 mg/L
in the bailed samples. Iron filtration ratios ranged from 2.33 to 5.00 (with one undefined value) for the 0.45-um filtered
pumped samples and from 81.1 to 340, for the bailed samples. Similarly, 5.0-um filtration ratios for the pumps ranged
from 1.25 to 5.50. Because 14 to 57 percent of the particles in the pumped samples were smaller than 0.4 um, it appears
that an important fraction of the iron may be in colloidal form or associated with colloids that passed through the
0.45-um filters to become part of the analysis. Although the absolute differences in iron concentration evident at
WISC-2 and WISC-3 are significant, they actually represent only relatively small differences in concentration,
particularly for the pumped samples, where concentrations differed by as little as 30 u.g/L. Variation in iron results
at these low concentrations are likely the result of artifacts of sample handling and analysis which are commonly
higher for iron analyses.
The relationship between relatively high particle concentrations and measurable differences between filtered
and unfiltered concentrations of the other metals was also evident at WISC-2 and WISC-3. For example, manganese
detected in the bailed samples exhibited filtration ratios of 11.0 and 7.5, respectively. Likewise, barium, which was
detected in pumped samples with filtration ratios of 0.83 to 1.00, exhibited filtration ratios of 3.08 and 4.38,
respectively, for the bailed samples. Arsenic and chromium were not detected in any samples from these wells.
Samples from WASH-2, which contained particle concentrations between 8.9 mg/L (CP1) and 818 mg/L
(bailer), exhibited much lower iron filtration ratios. The iron filtration ratio was approximately 16 for bailed samples
and ranged from 1.04 to 1.29 for pumped samples. Lower particle concentrations in the pumped samples were
accompanied by a generally higher fraction of sub-0.4-um particles, approximately 75 percent for the BP and CP2.
It appears that the lower total particle concentrations and greater proportion of particles passing through the 0.4-um
filters (for the pumped samples only) contributed to less pronounced concentration differences between filtration
method than observed at the WISC wells and WASH-1. The pattern of response of the other metals was similar to that
observed at the wells discussed previously in that filtration ratios were the greatest, or undefined, for the bailed
samples, while the pumped samples had filtration ratios of 1.00 or did not contain measurable concentrations.
Iron concentrations in unfiltered and filtered samples pumped from NEV-1 and NEV-2 were virtually
indistinguishable (filtration ratios between 1.00 and 1.25, inclusive; 5.0 um filtration ratios of 1.00), although
concentrations of sub-0.4-um particles in the pumped samples varied from 27 to 90 percent (0.8 to 11.8 mg/L) of total
particles. This suggests that the solid Fe(OH)3 may be associated with particles of a fairly constant fraction of sub-0.4
Um particles, or that the iron is actually present in dissolved form. In contrast to the pumped samples, iron filtration
ratios of bailed samples were much higher: 34.5 and 350 for the 0.45-um filters and 19.7 for the 5.0-um filter. As was
observed for most other bailed samples, over 85 percent of particles in the samples bailed from NEV-1 and NEV-2
32
-------
were larger than 0.4 um and were removed from the samples by filtration, thereby removing the majority of the iron
oxide particles mobilized or formed during the bailing process. Because the 5.0-um filters allowed slightly more
particles to pass into the NEV-1 samples than the 0.45-um filters, the 5.0-um-filtered samples exhibited slightly higher
concentrations. The 0.45-um and 5.0-um filtration ratios for manganese, barium, and arsenic in pumped samples (and
arsenic for bailed samples) were all within the range of 0.93 to 1.09, reflecting the predominance of the aqueous phases
or association with particles smaller than 0.4 um. In contrast, bailed samples showed manganese and barium filtration
ratios of 1.07 to 15.6, reflecting association of these metals with the greater concentrations of particles mobilized
during the bailing process and were then removed during filtration.
Iron filtration ratios at WASH-3 were also very low, between 1.00 and 1.16, but unlike all other wells in the study
the ratio was also low for the bailed samples. Also, unlike all the other wells in the study, total particle concentrations
for all devices were similar, ranging from 39.1 to 41.1 mg/L. Chemical equilibrium modeling suggests that an
important fraction of the iron at this well may be dissolved ferrous iron which would be relatively unaffected by
filtration. Furthermore, analysis of the grain-size distribution of samples from WASH-3 revealed that all methods
produced less than 5 percent of total particle concentration in the 0.03- to 0.45-um size range. As a result, either the
iron is present primarily as a dissolved species or, if the iron is colloidal, the particles are smaller than 0.03 um. The
higher DO concentrations of the bailed samples with respect to the pumped samples did not appear to impact filtration
results, suggesting that little ferrous iron was available to precipitate or that the precipitate was formed of particles
smaller than 0.45 um.
Filtration ratios for manganese, nickel, and arsenic at WASH-3 ranged from 0.80 to 1.33 for all devices while
chromium was not detected. Equilibrium modeling suggested that these metals were present primarily as aqueous
species and because filtration had little impact on their concentrations, there appears to be little association with
sub-0.4 um particles. The importance of understanding the speciation of metals with regard to sample filtration is
clearly illustrated for the case of nickel at WASH-1 and WASH-3 (Figure 10). At WASH-3, where equilibrium
calculations suggest that the aqueous phase Ni2+ and possibly the aqueous phase nickel bicarbonate (NiHCC>3+) are
thought to predominate and total particle concentrations were roughly equal for all sampling methods, neither 0.45-um
nor 5.0-um filtration had a significant effect on nickel concentrations. By contrast, at WASH-1 the solid phase nickel
carbonate is the predominant nickel species indicating that nickel carbonate may have precipitated out of solution and
accumulated in sediments near the well. The action of sample collection may have mobilized and incorporated
significant concentrations of this species in the unfiltered samples, particularly in those samples with high particle
concentrations. Filtration of the samples removed the bulk of the nickel carbonate and the remaining concentration
represents nickel carbonate particles smaller than 0.4 um in size or dissolved. What cannot be determined from these
data, however, is whether an important fraction of the nickel solid phase is mobile within the ground-water system.
As discussed above, results of univariate ANOVAs for individual metals in the unfiltered and 0.45-um filtered
samples (taken as a group) suggested that 0.45-um filtration was a significant effect for certain metals, namely iron,
manganese, and arsenic. However, results from individual wells clearly show that bailed samples, which contained
the highest particle concentrations, were the most important contributor to this filtration effect. Univariate ANOVAs
for these same three metals collected with the pumps alone indicated that significant differences existed only for iron.
In other words, no statistically significant differences existed in concentrations of manganese, arsenic, nickel, and
barium between unfiltered and 0.45-um filtered samples collected at the same well by the bladder pump and the
centrifugal pump at the two rates. It appears that the significant differences in iron concentrations result from the close
association of iron with particles entrained in samples, a major fraction of which were removed by 0.45-um filtration
(particularly in the most turbid samples). When the ANOVA was expanded to include just bailed samples that were
0.45-um filtered, iron remained the only metal with concentrations significantly different between device-filter
combinations. This occurrence suggests that for most of the metals analyzed (and for which there were sufficient data),
0.45-um filtration resulted in statistically indistinguishable concentrations in samples from different devices in the
same well.
33
-------
IMPACTS OF SAMPLING DEVICE ON METALS CONCENTRATIONS
Relationships of trace metal concentrations to sampling devices were heavily dependent on the phase in which
the species was present and the potential for association with particles of that species. For example, unfiltered
concentrations of the predominantly aqueous arsenic showed very little variation between devices except in samples
that contained extremely high particle concentrations (primarily bailed samples). Filtration had little effect on
concentrations except in the most turbid samples, so the filtered samples also showed very little variation between
devices. In the turbid samples, primarily at WISC-4 and WASH-1, concentrations in unfiltered samples were slightly
higher than in filtered samples but filtration removed the fraction associated with the larger particles and produced
essentially equal concentrations for all devices at individual wells. Likewise, where nickel was present primarily in
aqueous form (WASH-3), all devices produced similar concentrations in both filtered and unfiltered samples. For these
constituents, filtration generally did not impact relative concentrations between devices with the exception of those
devices that produced samples with very high particle concentrations.
Manganese and barium, both of which were also primarily present as aqueous species, showed a closer relation
between device type and unfiltered concentrations than arsenic and nickel. For manganese and barium, the bailers
usually produced the highest unfiltered concentrations, up to 15 times higher than the unfiltered pumped samples.
Concentrations in the unfiltered pumped samples were essentially equal except in the more turbid samples, where
slightly more variation was observed, and at WASH-3, where CP1 produced manganese concentrations that were
slightly higher than the other devices. In most cases, the filtered concentrations of all four devices were essentially
equal to each other and equal to the unfiltered pumped samples. Exceptions occurred at WASH-1 where the high
particle concentrations produced by most devices were accompanied by high unfiltered manganese concentrations
relative to filtered concentrations. The higher unfiltered concentrations most often observed in bailed samples suggest
a direct relationship to the higher particle concentrations produced by these devices and, as a result, are unlikely to
represent the potential for mobility in the ground-water environment. As with aqueous arsenic and nickel, however,
filtration removed the fraction of these constituents associated with particles larger than 0.45 urn, resulting in
relatively little variation in unfiltered concentrations between devices.
Aqueous chromate was usually present in unfiltered bailed samples but rarely detected in unfiltered samples
from other devices. The response to filtration was similar to the metals discussed above but to a greater extreme, In
this case, filtration removed all chromium detected in the unfiltered samples for all devices and wells (with the
exception of low concentrations detected at NEV-1), indicating that virtually all of the chromate was associated with
particles larger than 0.45 um. The result was that all four sampling methods produced essentially "equal"
concentrations of chromium in filtered samples, that is, below detection.
As discussed previously, iron generally exhibited the highest variability in both filtered and unfiltered samples,
reflecting complicated relationships to redox conditions and particle concentrations. Despite this variability,
unfiltered iron concentrations were highest in samples collected by the device that produced the most turbid samples,
which was the bailer in six of the eight wells. In the less-turbid wells, unfiltered bailed samples contained iron
concentrations up to 300 times higher than pumped samples, while in the most turbid wells concentrations in unfiltered
bailed samples were sometimes over 1000 times higher than in pumped samples. As noted previously, filtration of
pumped samples from the less turbid wells resulted in slightly lower concentrations than in unfiltered samples but in
most cases not significantly because concentrations were near analytical detection levels. In turbid samples, filtration
significantly reduced concentrations in pumped samples as a result of removal of iron particles. Filtration of bailed
samples resulted in iron concentrations that usually fell within the range present in the pumped samples, although all
iron was removed in the most turbid samples. Variability in both unfiltered and filtered pumped samples caused by
particle concentrations and size distribution was sufficient to mask any significant response pattern related to pump
type, although CP1 showed slightly higher concentrations than the other pumps at NEV-1, NEV-2, and WASH-3. These
higher concentrations appear to be unrelated to particle concentrations and may result from this pump being the first
34
-------
to sample each well at the NEV and WASH sites. Therefore, as with the other metals, any differences in concentrations
produced by different pumps and discharge rates were considered insignificant.
Impacts of filtration on metals present primarily as solid species can also be seen for nickel at WASH-1. Nickel
concentrations in the unfiltered samples show a relationship to particle concentrations that suggests the presence of
nickel carbonate precipitated out in the sediments of the sampling zone. A significant fraction of nickel was then
removed during filtration and the fraction of these particles smaller than 0.45 urn was relatively uniform for the pumps
as evidenced by the filtered concentrations. These concentrations were low for the pumps (from below detection up
to 0.05 mg/L) but were all below detection for the bailer.
Univariate ANOVAs of the unfiltered concentrations of iron, manganese, and arsenic with the four sampling
devices at seven wells confirmed the overall trends discussed above. In particular, significant differences were found
to exist in the unfiltered concentrations produced by the different sampling devices for all three of these metals. Those
samples containing the highest particle concentrations also had unfiltered metals concentrations that were
significantly higher than the other samples. In most cases, the bailer produced samples with the highest particle
concentrations and, therefore, produced significantly higher metals concentrations in unfiltered samples. In contrast,
no significant differences existed between concentrations for these three metals in the 0.45-um-filtered samples,
suggesting that filtration removed nonrepresentative particle loads from the turbid samples and caused samples from
all four devices to be essentially indistinguishable.
IMPACTS ON CONCENTRATIONS OF MAJOR IONS
Methods of sampling and filtration had little impact on the concentrations of major anions with most differences
less than five percent, even for those samples with very high particle concentrations (Appendix C). Relative
differences in concentration greater than five percent usually occurred near analytical detection levels where errors
are considered to be highest. Univariate ANOVAs for individual anions showed no significant effect of device,
filtration method, or device-filtration interactions, indicating that, taken as a group, anion concentrations were not
significantly impacted by these factors. Furthermore, associations between anions and particles appeared minimal
because filtration of samples with high particle concentrations did not cause large differences in anion concentrations.
An exception to this overall trend was observed at WISC-2 where 0.45-um filtration caused analytically
significant differences in anion concentrations. In most samples from this well, unfiltered concentrations were usually
over 10 percent higher than 0.45-um-filtered concentrations. Differences exhibited by the divalent anion carbonate
-) and sulfate (SO42-) were larger than for the monovalent anion chloride (Cl~), bicarbonate (HCOf), and nitrate
), demonstrating the greater potential for reactions of these divalent anions with particle surfaces. The
differences in concentrations observed at WISC-2, though small, appear to be related to removal by filtration of an
important fraction of positively-charged particles with which certain anionic species were associated. This
relationship between anions and particles was not observed at the other wells.
Concentrations of major cations were more affected by particle concentrations than major anions due to their
positive charge and the generally negative surface charge of many particles in the size range investigated in this study.
Univariate ANOVAs for individual cations indicated that device and filtration effects and interactions were all
significant, although not all effects were significant for all cations. As expected, samples containing less than 50 mg/L
particles (primarily pumped samples) showed small differences between filtered and unfiltered samples and between
devices. In these less turbid samples, differences were always less than 10 percent and most were less than 5 percent.
As a result, the pumps generally showed little variation in cation concentrations, both between devices and between
filtration methods. Bailers seldom produced samples with less than 50 mg/L particles.
Alternatively, samples containing in excess of 50 mg/L particles showed significant differences in cation
concentrations between filtered and unfiltered samples. Significant differences were also evident between unfiltered
35
-------
cation concentrations in these samples and those in unfiltered samples from devices that produced low particle
concentrations. The monovalent major cations potassium (K+) and sodium (Na+) showed the least variation, often less
than 5 percent while the divalent major cations calcium (Ca2+) and magnesium (Mg2+) showed the most variation.
Filtration caused unfiltered magnesium concentrations to be up to 3 times higher than 0.45-u.m-filtered concentrations,
and unfiltered calcium concentrations to be up to 4 times higher than 0.45-um-filtered concentrations. These
differences were most likely related to the association of these positively-charged ions with high concentrations of
particles having primarily negative surface charge and the effect of removing significant concentrations of these
particles by sample filtration.
Under conditions where all analytes are truly dissolved, the sum of positive and negative charges in filtered and
unfiltered samples (from a single sampling event) should balance and therefore differences in concentrations of cations
between filtered and unfiltered samples should not exceed differences in anions. As a result, there should only be
minimal differences in the overall variability of anion and cation concentrations across filtration methods. The fact
that cations exhibited higher variability than anions in this study is primarily related to significant particle
concentrations in many of the unfiltered samples. These particles were a likely source of cations, particularly calcium
and magnesium, released into solution during sample digestion and not naturally balanced by dissolved anions.
Furthermore, because most of the particles in the turbid samples were larger than 5.0 urn and therefore unlikely to be
mobile in the ground-water environment, these higher cation concentrations are clearly unrepresentative of mobile
constituents.
Univariate ANOVAs for organic carbon indicated no significant effect for filtration but a significant effect for
device, with the bailer significantly different from the other devices. Organic carbon concentrations were generally
highest in the bailed samples because bailing entrained more particles, and in some wells (particularly WISC-4 and
WASH-2) organic carbon was an important component of the particle load. The trend observed at some wells of
slightly higher organic carbon concentrations in 0.45-|im-filtered samples than in the associated unfiltered samples
represents traces of glycerol (used to maintain structural integrity) and organic membrane wetting agents that were
washed off the filters into the sample container. This effect was not observed for the 5.0-um filters.
Laboratory measurements of pH, although not representative of absolute field conditions, indicate that bailed
samples were subjected to greater degassing than pumped samples. Bailed samples exhibited overall significantly
higher pH values relative to pumped samples based on univariate ANOVAs. Filtration did not have a significant effect
on laboratory pH measurements.
PRACTICAL CONSIDERATIONS
Several practical considerations should be evaluated when designing and implementing a program for sampling
trace metals in ground water. First, turbidity measurements offer a useful indicator of particle concentrations in ground
water, thereby providing important information about the effectiveness of monitoring well purging and the relative
concentration of particles at different wells. Turbidity is relatively easy to measure in the field and the turbidimeter
calibration is generally highly stable. In addition, the use of a flow-through cell allows nearly continuous monitoring
during pumping and greatly reduces variability in the measurements.
Sample collection at low to moderate discharge rates (up to 1000 mL/min) requires a sampling device capable
of providing stable discharge at these rates over relatively long periods of time. In this study, the bladder pump met
these criteria with few adjustments needed to ensure continuous discharge at the desired rate. The bladder pump also
performed well under the variable head conditions encountered at two low-yield wells. In contrast, the submersible
centrifugal pump was not generally capable of meeting these criteria, especially under low-yield conditions. The
inability to maintain constant discharge was apparently related to head conditions during purging and sampling; as
drawdown increased with pumping, total head on the pump system increased and pump discharge was reduced. At
low discharge rates, it appeared that very little increase in total head was required to cause a total loss of discharge.
36
-------
This situation was addressed by frequent measurements of the discharge rate and adjustments of the pump speed as
necessary. However, this approach required the field personnel's near-constant attention and prevented them from
carrying out other tasks related to the sampling event. If this type of device is under consideration for
low-discharge-rate sampling, hydraulic conditions should be carefully evaluated to determine whether a low-yield
well will be encountered. Finally, although bailers are theoretically capable of producing low "discharge" rates, they
are very difficult and impractical to operate in this manner. Even with very gentle use, the surging action generated
by a bailer as it is raised and lowered through the water column is more than enough to suspend and entrain normally
immobile particles.
Although purging and sampling at low pump speeds may reduce purging volumes and produce less disturbed
samples, the time required to complete purging may be considerably longer. In this study, purging times for the pumps
were up to four times higher than for the bailer (Table A-l) although it must be pointed out that equipment problems
with the centrifugal pump significantly increased purging times at several wells. Also, several purging criteria were
not met when purging with the bailer due to agitation of the sampling zone by the action of the bailer in the well bore.
In several of the wells, purging times for the pumps were lower or no more than 50% higher than for the bailer. Purging
times associated with low-rate pumping might be minimized through the use of dedicated equipment that would
eliminate the initial disturbance of the sampling zone and water column observed during the emplacement of a
sampling device in the well (Puls et al., 1991; Kearle et al., 1992). Dedication of sampling pumps to individual wells
may also allow simultaneous purging of several wells, as suggested by Backhus (1993), significantly reducing the
overall time required to purge a network of wells. However, dedicated systems require considerable up front capital
costs that may limit their widespread use.
On the other hand, dedication of sampling equipment may also save time and other resources in many aspects
of the sampling process. For example, the time and effort required for decontamination of portable sampling
equipment is completely eliminated, as are the time to collect samples and the cost of analyzing equipment rinsate
blanks. Also, the quantity of equipment needing transport from well to well is reduced. Although not tested in this
study, positive displacement pumps other than bladder pumps may be equally capable of providing representative
samples of trace metals and other constituents and are often available with more portable power sources. For example,
submersible gear pumps and progressing cavity pumps are both powered by 12-volt D.C. which can easily be supplied
by a deep-cycle marine battery. This type of power system is considerably more portable than the generator necessary
for the submersible centrifugal pump or the nitrogen cylinder or generator necessary for the bladder pump. Peristaltic
pumps may also be useful under conditions where depth to water is less than approximately 6 meters from ground
surface, although degassing may be a problem.
37
-------
REFERENCES
1. Backhus, D.A., J.N. Ryan, D.M. Groher, J.K. MacFarlane, and P.M. Gschwend. Sampling Colloids and
Colloid-Associated Contaminants in Ground Water. Ground Water, 31 (3): 466-479,1993.
2. Buddmeier, R.W., and J.R. Hunt. Transport of Colloidal Contaminants in Ground Water: Radionuclide
Migration at the Nevada Test Site. Applied Geochemistry, 3:535-548,1988.
3. BMDP Statistical Software, Inc., BMDP Statistical Software Manual, Volume 2, Method PV4. W.J. Dixon et
al., eds. University of California Press, 1988.
4. Champ, D.R., W.F. Merritt, and J.L. Young. Potential for the Rapid Transport of Plutonium in Groundwater as
Demonstrated by Core Column Studies. In: Scientific Basis for Radioactive Waste Management, Volume 5,
W. Lutze ed. Elsevier Science Publication Co., 1982. pp. 745-754.
5. Drever, J.I. The Geochemistry of Natural Waters. Prentice Hall, Inc., Englewood Cliffs, New Jersey. 1982.
388 pp.
6. Desert Research Institute. Quality Assurance Plan for Saturated Zone Sampling and Monitoring
Methodologies for Hazardous Waste Sites. CR-815774-01. Desert Research Institute, 1991.
7. Gibs, J., and T.E. Imbrigiotta. Well Purging Criteria for Sampling Purgeable Organic Compounds.
Groundwater, 28(1): 68-78,1990.
8. Gillhaus, R.W., M.J.L. Robin, J.F. Barker, and J.A. Cherry. Groundwater Monitoring and Sample Bias,
Publication No 4367, American Petroleum Institute, 1983,206 pp.
9. Gschwend, P.M., and M.D. Reynolds. Monodisperse Ferrous Phosphate Colloids in an Anoxic Groundwater
Plume. Journal of Contaminant Hydrology, 1:309-327,1987.
10. Hem, J.D. Study and Interpretation of the Chemical Characteristics of Natural Water. Water Supply Paper
2254, U.S. Geological Survey, Alexandria, Virginia, 1985.263 pp.
11. Holm, T.R., O.K. George, and M.J. Barcelona. Oxygen Transfer Through Flexible Tubing and its Effects on
Ground Water Sampling Results, Ground Water Monitoring Review, 8(3): 83-89,1988.
12. Horowitz, A.J., K.A. Elrick, and M.R. Colberg. The Effect of Membrane Filtration Artifacts on Dissolved
Trace Element Concentrations. Water Resources, 26(6): 753-763,1992.
13. Kearle, P.M., N.E. Korte, and T.A. Cronk. Suggested Modifications to Ground Water Sampling Procedures
Based on Observations from the Colloidal Boroscope. Groundwater Monitoring Review, 12(2): 155-161,
1992.
14. Kingston, W.L., and M. Whitbeck. Characterization of Colloids Found in Various Groundwater Environments
in Central and Southern Nevada. DRI-45083, Desert Research Institute, Water Resources Center, Reno,
Nevada, 1991,136 pp.
38
-------
15. Magaritz, M., AJ. Amiel, D. Ronen, and M.C. Wells. Distribution of Metals in a Polluted Aquifer: A
Comparison of Aquifer Suspended Material to Five Sediments of the Adjacent Environment. Journal of
Contaminant Hydrology, 5:333-347, 1990.
16. McDowell-Boyer, L.M., J.R. Hunt, and N. Sitar. Particle Transport Through Porous Media. Water Resources
Res., 22(13): 1901-1921, 1986.
17. McCarthy, J.R, and J.M. Azchara, Subsurface Transport of Contaminants. Environmental Science
Technology, 23(5): 496-502, 1989.
18. Mills, W.B., S. Liu, and F.K. Fong. Literature Review and Model (COMET) for Colloid/Metals Transport in
Porous Media. Ground Water, 29(2): 199-208, 1991.
19. Parkhurst, D.L., D.C. Throstenson, and L.N. Plummer. PHREEQE - A Computer Program for Geochemical
Calculations, Water Resources Investigations Paper 80-96, U.S. Geological Survey, Alexandria, Virginia,
1980. 21 pp.
20. Penrose, W.R., W.L. Polzer, E.H. Essington, D.M. Nelson, and K.A. Orlandini. Mobility of Plutonium and
Americium Through a Shallow Aquifer in a Semiarid Region. Environmental Science Technology. 24:
288-234, 1990.
21. Puls, R.W. Colloidal Considerations in Groundwater Sampling and Contaminant Transport Predictions.
Nuclear Safety, 31(1): 58-65, 1990.
22. Puls, R.W., and MJ. Barcelona. Groundwater Sampling for Metals Analysis. EPA /540/4-89/001, U.S.
Environmental Protection Agency, Ada, Oklahoma, 1989, 6 pp.
23. Puls, R.W., J.H. Eychauer, and R.M. Powell. Colloidal-Facilitated Transport of Inorganic Contaminants in
Ground Water: Part I. Sampling Considerations. EPA/600/M-90/023, U.S. Environmental Protection Agency,
Ada, Oklahoma, 1991. 12 pp.
24. Puls, R.W., R.M. Powell, D.A. Clark, and CJ. Paul. Facilitated Transport of Inorganic Contaminants in
Ground Water: Part II. Colloidal Transport. EPA/600/M-91/040, U.S. Environmental Protection Agency, Ada,
Oklahoma, 1991. 12pp.
25. Robbins, G.A., and J.M. Martin-Hayden. Mass Balance Evaluation of Monitoring Well Purging, Part I,
Theoretical Models and Implications for Representative Sampling. Journal of Contaminant Hydrology,
8:203-224, 1991.
26. Ryan, J.N., and P.M. Gschwend. Colloid Mobilization in Two Atlantic Coastal Plain Aquifers: Field Studies.
Water Resources Res., 26(2): 307-322, 1990.
27. Standard Methods for the Examination of Water and Wastewater: American Public Health Association,
American Works Assocation, Water Pollution Control Federation. New York, N.Y.: American Public Health
Association, ISthed., 1992.
28. Stumm, W. Chemical Interaction in Particle Separation, Environmental Science Technology, (11): 1066-1070,
1977.
29. Stumm, W, and J.J. Morgan. Aquatic Chemistry (2nd ed.). Wiley Interscience, New York, 1981. 780 pp.
30. Tillekeratne, S., T. Miwa, and A. Mizuike. Determinations of Traces of Heavy Metals in Positively Charged
Inorganic Colloids in Fresh Waters. Microchimica Acta, 1985(3): 289-296, 1985.
39
-------
31. Unwin, J., and V. Maltby. Investigations of Techniques for Purging Ground-Water Monitoring Wells and
Sampling Ground Water for Volatile Organic Compunds. Ground-Water Contamination: Field Methods,
ASTM STP 963, eds. A.G." Collins and A.I. Johnson, pp. 240-252,1988.
32. U.S.G.S. Methods for the Determination of Inorganic Substances in Water and Fluvial Sediments, M.W.,
Skougstad, M.J. Fishmann, L.C. Freidman, D.E. Erdmann, and S.S. Duncan, eds., Tech. of Water-Resourced
Invest, of the U.S. Geologieal Survey, Book 5, Chapter Al, 1979.
33. U.S. EPA. Methods for Chemical Analysis of Water and Wastes. EPA-600/4-79/020, U.S. Environmental
Protection Agency, Cincinnati, Ohio, 1979.
34. U.S. EPA. RCRA Ground Water Monitoring Technical Enforcement Guidance Document. OS WER-9950.1,
U.S. Environmental Protection Agency, 208 pp.
35. U.S. EPA. The Determination of Inorganic Anions in Water Ion Chromatography. EPA-600/4-84/017, U.S.
Environmental Protection Agency, Cincinnati, Ohio, 1984.
36. Woods, T.L., and R.M. Garrels. Thermodynamic Values at Low Temperature for Natural Inorganic Materials:
An Uncritical Summary, Oxford University Press, New York, 1987, 272 pp.
37. Yao, K., M.T. Habibian, and C.R. O'Melia. Water and Waste Water Filtration: Concepts and Applications,
Environmental Sciences Technology, 5(11): 1105-1112,1971.
40
-------
APPENDIX A
Summary of Indicator Parameter Measurements and Purging Data
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
WISC-2
CP1 (Q=370 mL/min) 2.45
2.75
3.95
5.15
6.35
7.55
9.55
11.55
13.55
15.55
17.55
19.25
20.95
22.65
24.35
BP (Q=500 mL/min) 4.8
5.4
6
6.6
7.2
8.2
9.2
10.2
11.2
12.2
13.2
15.2
17.7
20.2
22.7
25.2
CP2(Q=940mL/min) 3
5
6.8
8.7
10.5
12.4
14.2
16
17.9
19.7
21.6
23.4
Bail 65.3
Time
(min)
5
8
11
14
17
20
25
30
35
40
45
50
55
60
65
6
8
10
12
14
16
18
20
22
24
26
30
35
40
45
50
3
5
7
9
11
13
15
17
19
21
23
25
33
DO
(mg/L)
3
1.6
1.3
0.9
0.8
0.7
0.6
0.5
0.5
0.5
0.4
0.4
0.4
0.3
0.3
4.8
1.1
0.8
0.7
0.7
0.7
0.6
0.5
0.5
0.4
0.4
0.3
0.3
0.3
0.3
0.2
1.5
0.6
0.5
0.4
0.4
0.3
0.3
0.2
0.2
0.2
0.2
0.2
3.9
T
(°C)
10.1
10.2
10.1
10.6
11.8
12.3
12.1
12.3
12.5
12.7
12.9
13.2
13.2
13.2
13.6
10
9.8
9.7
9.6
9.6
9.6
9.6
9.5
9.4
9.4
9.4
9.4
9.4
9.4
9.4
9.4
10.7
11
11.4
11.3
11.4
11.4
11.5
11.6
11.9
11.8
11.5
11.4
12.2
PH
7.98
8.42
8.7
8.82
8.86
8.88
8.89
8.91
8.92
8.91
8.92
8.92
8.91
8.91
8.91
7.32
8.35
8.04
8.91
9.02
9.07
9.04
9.1
9.1
9.1
9.1
9.1
9.1
9.1
9.09
9.1
8.17
8.76
8.93
8.96
8.98
9
9
9
9
9
9.01
9.01
8.20
EC
(US)
259
259
228
221
212
213
209
208
209
209
204
204
204
205
205
289
282
260
235
228
223
220
218
218
220
218
217
217
215
214
214
243
215
211
210
208
207
207
207
202
201
203
203
212
Turb.
(NTU)
31.5
28.8
19.5
15.7
13.5
12.2
9.5
8
7.5
6.5
6.4
6
5.8
5.5
5
6.2
4.9
4.5
4.2
3.5
2.8
2.3
3
2.4
1.8
1.8
1.6
1.3
1.3
1.3
1.2
2.5
2.45
1.93
1.51
1.87
1.42
2.25
3.06
1.82
1.32
1.92
1.66
170
continued
41
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Voi
(L)
WISC-3
CP1 (Q=320 mL/min) 0.38
1.14
1.90
2.80
3.06
3.58
4.62
5.18
5.94
6.74
7.74
8.74
9.34
9.94
10.9
11.6
12.2
12.9
14.4
15.7
16.3
16.9
17.5
18.1
18.7
20.2
20.9
CP2 (Q=900 mL/min) 4.0
5.0
7.0
9.0
11.0
,14.3
16.3
18.1
19.9
23.5
25.3
27.1
28.9
30.7
32.5
34.3
36.1
37.9
39.7
Bail 56.8
Time
(min)
1
3
5
6
7
9
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
51
53
4
5
7
9
11
15
17
19
21
25
27
29
31
33
35
37
39
41
43
28
DO
(mg/L)
3.7
1.8
1.4
1.1
1.1
1.1
0.7
0.7
0.6
0.5
0.5
0.4
0.4
0.4
0.3
0.3
0.3
0.4
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.2
3.8
3.1
0.7
0.4
0.4
0.3
0.3
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
-
T
8.0
8.5
8.5
10.2
10.1
10.1
10.2
11.0
11.8
11.9
11.9
11.9
11.9
12.0
12.0
12.0
11.5
11.5
11.3
11.4
11.8
11.9
11.3
11.5
11.5
11.4
10.8
10.2
10.0
10.2
10.9
10.9
11.1
11.8
11.6
11.4
11.3
11.3
11.4
11.3
11.3
11.4
11.4
11.5
11.4
11.5
9.8
pH
7.63
8.21
8.53
8.93
8.94
8.96
8.98
8.98
8.98
8.98
8.98
8.98
8.98
8.98
8.98
8.98
8.99
8.99
8.99
8.98
8.99
8.99
8.99
8.99
8.99
8.99
9.00
9.56
9.22
8.98
8.96
8.98
8.99
8.99
9.00
9.01
9.00
9.00
9.00
9.00
9.00
9.00
9.00
9.00
9.00
9.00
9.00
EC
311
248
240
209
209
208
209
207
209
202
202
201
200
201
200
200
201
201
199
199
194
193
194
201
202
202
198
218
213
215
212
209
207
205
209
203
205
205
203
202
202
201
200
202
201
198
240
Turb.
(NTU)
3.28
2.71
2.51
5.11
4.88
5.31
5.56
5.41
3.92
2.91
2.40
1.83
1.89
1.51
1.31
1.11
2.61
1.73
1.31
1.21
0.53
0.51
0.34
0.72
2.98
0.51
0.11
16.28
10.38
35.5
43.2
20.7
10.5
10.2
9.2
6.4
4.7
4.4
4.2
5.7
4.7
6.1
5.2
4.7
3.7
5.1
195
continued
42
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
WISC-4
CP1 (Q-330 mL/min) 1.5
2.4
3.0
3.6
4.5
6.9
8.8
10.2
11.8
13.4
15.0
16.6
18.2
19.8
21.4
23.0
24.6
26.2
28.13
29.12
BP(Q=510 mL/min) 1.48
2.96
4.44
5.58
6.60
7.62
8.64
9.66
10.68
12.21
14.76
17.31
19.86
22.41
24.96
27.51
30.06
32.61
35.16
37.71
CP2 (Q=1 000 mL/min) 4.00
5.00
6.00
6.94
8.82
10.70
12.58
14.46
Time
(min)
5
8
10
12
15
20
25
30
35
40
45
50
55
60
65
70
75
80
86
89
2
4
6
8
10
12
14
16
18
21
26
31
36
41
46
51
56
61
66
71
4
5
6
7
9
11
13
15
DO
(mg/L)
4.5
4.5
5.1
5.4
5.1
4.6
4.2
4.8
4.3
4.1
3.7
3.4
3.2
2.7
1.3
1.1
1.0
0.9
1.2
1.2
5.8
3.6
2.9
2.3
2.0
1.8
1.6
1.3
1.3
1.1
0.9
0.9
0.7
0.7
0.7
0.7
0.6
0.7
0.6
0.6
3.8
2.8
2.3
2.0
1.8
1.4
1.1
0.9
T
(°C)
13.2
13.8
14.0
13.9
13.4
13.5
13.9
14.2
14.8
15.4
15.6
15.6
15.2
14.9
15.4
15.0
14.3
13.6
12.9
13.0
10.0
10.9
11.1
11.0
11.0
10.8
10.7
10.5
10.5
10.5
10.5
10.4
10.5
10.5
10.5
10.5
10.4
10.4
10.4
10.4
10.5
11.1
11.4
11.8
11.8
11.6
11.5
11.5
PH
6.30
4.66
5.37
7.04
7.15
7.15
7.14
7.07
7.07
7.09
7.06
7.08
7.08
7.07
7.08
7.09
7.09
7.09
7.09
7.10
7.08
7.16
7.16
7.16
7.16
7.16
7.17
7.18
7.18
7.18
7.19
7.20
7.19
7.19
7.19
7.19
7.19
7.20
7.19
7.20
6.89
6.92
6.95
6.99
7.01
7.04
7.06
7.08
EC
(>iS)
258
171
782
1248
1409
1428
1406
1435
1430
1459
1469
1459
1449
1449
1459
1430
1435
1448
1422
1428
1351
1418
1439
1446
1446
1446
1484
1470
1477
1470
1484
1484
1519
1526
1512
1498
1505
1498
1512
1498
1428
1432
1480
1458
1458
1514
1507
1514
Turb.
(NTU)
0.12
0.15
3.70
3.12
3.80
11.1
23.5
27.3
18.3
23.8
52
41.9
39.7
36.0
22.1
12.0
8.0
7.3
10.5
21.7
14.5
41.5
36.4
32.1
50.5
70.3
52.5
43.0
28.2
19.7
11.2
7.8
6.23
5.46
4.55
4.12
3.30
2.68
2.43
2.23
9.66
24.4
26.4
21.3
58.3
47.0
27.2
16.8
continued
43
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
CP2 (Q=1 000 mL/min) 1 6.34
J8.22
20.10
21.98
23.86
25.46
26.46
29.98
32.06
32.90
33.80
35.60
39.20
41.00
43.2
45.4
47.6
54.6
57.8
62.4
64.8
Bail 41.6
NEV-1
CP1 (Q=310 mL/min) 1.50
3.30
3.94
4.58
5.22
5.86
6.50
7.16
7.82
8.50
9.18
9.78
10.38
10.98
11.58
12.18
12.78
13.36
13.94
14.52
BP (Q=990 mL/min) 0.74
1.48
2.22
2.96
5.36
6.46
Time
(min)
17
19
21
23
25
27
29
33
36
37
38
40 •
44
46
48
50
52
56
58
61
63
21
5
11
13
15
,17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
1
2
3
4
6
7
DO
(mg/L)
0.8
0.7
0.6
0.6
0.5
0.5
0.9
0.4
0.3
2.8
0.8
0.6
0.4
0.4
0.3
0.3
0.3
0.5
0.5
0.3
0.3
4.6
0.80
0.70
0.70
0.70
0.70
0.70
0.60
0.60
0.50
0.50
0.60
0.50
0.40
0.40
0.50
0.40
0.40
0.40
0.40
0.40
2.90
1.70
1.10
0.70
0.40
0.40
T
11.5
11.4
11.4
11.3
11.3
11.1
11.9
11.8
11.4
11.1
12.0
11.8
11.4
11.2
12.0
11.8
11.6
11.6
12.0
11.5
11.4
14.9
24.50
25.60
25.90
26.10
26.10
26.20
26.20
26.20
26.30
26.40
26.40
26.40
26.60
26.60
26.50
26.50
26.50
26.50
26.40
26.40
22.10
23.10
23.60
24.00
24.20
24.20
PH
7.09
7.10
7.11
7.11
7.12
7.12
7.12
7.11
7.12
7.07
7.06
7.10
7.12
7.12
7.12
.7.13
7.13
7.11
7.12
7.13
7.13
7.23
6.82
6.85
6.88
6.89
6.89
6.90
6.91
6.91
6.91
6.91
6.91
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.86
6.86
6.88
6.89
6.90
6.90
EC
1507
1493
1527
1521
1514
1507
1498
1478
1521
1555
1528
1491
1526
1527
1517
1498
1527
1508
1517
1514
1500
1490
23.20
23.20
22.90
22.90
23.00
23.00
23.00
23.00
23.00
22.70
22.60
22.60
22.70
22.70
22.60
22.60
22.60
22.60
22.60
22.60
22.70
23.10
23.00
23.10
23.10
23.10
Turb.
(NTU)
12.3
8.30
5.8
. 4.6
3.5
3.1
3.8
49.2
138.3
96.6
58.5
52.6
50.2
35.3
18.0
12.2
11.3
17.1
10.5
7.9
5.9
>2000
0.29
0.13
0.13
0.11
0.10
0.10
0.09
0.10
0.08
0.08
0.07
0.07
0.07
0.06
0.07
0.06
0.06
0.05
0.06
0.06
0.30
0.18
0.86
0.46
0.14
0.13
44
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
BP (Q=990 mL/min) 7.56
9.54
11.40
13.26
15.12
16.98
18.96
20.94
22.84
24.82
26.80
28.74
30.68
32.62
CP2 (Q=990 mL/min) 0.90
2.70
4.50
6.44
8.38
10.32
12.26
14.20
16.14
18.10
20.06
22.02
24.00
25.98
27.96
29.94
Bail 4.60
23.00
35.00
46.00
58.00
69.00
80.00
92.00
103.00
115.00
126.00
138.00
NEV-2
CP1 (Q=310 mL/min) 0.60
0.98
1.36
1.88
2.42
3.66
Time
(min)
8
10
12
14
16
18
20
22
24
26
28
30
32
34
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
10
20
28
34
42
60
69
74
78
89
107
117
3
5
7
9
11
13
DO
(mg/L)
0.30
0.30
0.30
0.30
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
1.70
0.70
0.40
0.30
0.30
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
3.00
2.30
2.30
2.90
2.70
2.30
2.50
2.50
2.40
2.30
2.50
2.80
2.10
1.40
1.00
0.80
0.60
0.50
T
(°C)
24.30
24.30
24.40
24.50
24.50
24.50
24.50
24.60
24.60
24.50
24.50
24.50
24.60
24.60
23.00
24.40
25.30
25.50
25.60
25.70
25.80
25.80
25.80
25.80
25.80
25.80
25.90
25.90
25.90
25.90
22.40
23.40
24.10
24.00
24.10
23.80
24.10
24.00
23.90
24.20
24.10
24.30
24.20
24.30
24.50
25.50
26.70
26.90
pH
6.90
6.90
6.90
6.90
6.90
6.90
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.84
6.80
6.82
6.86
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
7.05
7.02
6.99
7.04
7.07
7.18
7.03
7.08
7.10
7.11
7.13
7.07
7.04
7.01
6.97
6.93
6.93
6.93
EC
(US)
23.10
23.10
23.00
23.10
23.10
23.00
23.00
23.00
23.10
23.00
23.00
22.90
22.90
22.90
22.70
23.10
22.90
23.10
23.10
23.10
22.70
22.60
22.60
22.60
22.60
22.70
22.70
22.70
22.60
22.60
22.90
22.70
22.60
22.60
22.60
22.50
22.50
22.50
22.50
22.60
22.60
22.60
20.80
20.90
21.00
20.90
21.20
20.90
Turb.
(NTU)
0.14
0.14
0.12
0.10
0.08
0.08
0.07
0.08
0.06
0.05
0.05
0.05
0.05
0.05
0.60
0.37
0.24
0.23
0.16
0.13
0.13
0.13
0.13
0.10
0.09
0.07
0.08
0.08
0.07
0.06
45.00
96.50
82.80
118.30
110.20
113.50
104.20
104.70
89.20
66.00
80.70
67.20
17.60
11.50
9.30
9.20
5.05
3.88
continued
45
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
CP1 (Q=310mL/min) 4.28
4.90
5.52
6.14
6.76
7.38
8.00
8.62
9.24
9.86
10.50
11.14
11.77
12.40
13.03
13.65
14.27
BP (Q=970 mL/min) 0.95
1.90
2.85
3.80
4.75
5.70
6.67
7.64
8.62
9.59
10.56
11.53
12.50
13.47
14.44
15.41
16.38
18.32
20.26
22.20
24.14
26.12
28.06
30.00
CP2 (Q=990 mL/min) 3.00
4.20
5.40
6.60
7.80
8.90
10.00
Time
(min)
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
19
21
23
25
27
29
31
2
3
4
5
6
7
8
DO
(mg/L)
0.40
0.40
0.40
0.40
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.20
0.20
0.20
2.60
2.00
1.50
1.00
0.90
0.70
0.60
0.50
0.50
0.40
0.40
0.30
0.30
0.30
0.30
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
1.00
0.50
0.40
0.40
0.30
0.30
0.30
T
(°C)
27.10
27.10
27.20
27.20
27.30
27.30
27.30
27.40
27.50
27.50
27.50
27.50
27.40
27.40
27.50
27.50
27.50
22.60
23.30
23.70
23.90
24.00
24.10
24.20
24.20
23.50
24.00
24.20
24.20
24.20
24.30
24.30
24.20
24.20
24.20
24.30
24.20
24.20
24.30
24.30
24.30
24.00
25.10
25.40
25.70
25.80
25.90
26.00
PH
6.93
6.93
6.93
6.93
6.93
6.93
6.93
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.92
6.94
6.86
6.87
6.88
6.88
6.88
6.88
6.88
6.89
6.89
6.89
6.89
6.89
6.89
6.90
6.89
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.90
6.77
6.80
6.86
6.87
6.87
6.88
6.88
EC
(US)
21.00
21.00
21.00
21.00
21.10
21.10
21.10
21.10
21.20
21.20
21.20
21.20
21.20
21.20
21.20
21.20
21.20
20.10
21.10
21.10
21.20
21.20
21.20
21.20
21.20
21.30
21.10
21.20
21.10
21.10
21.10
21.20
21.10
21.10
21.20
21.20
21.00
21.00
21.10
21.00
21.10
21.30
21.50
21.60
21.70
21.30
21.40
21.40
Turb.
(NTU)
3.56
3.33
3.12
2.96
2.77
2.94
2.70
2.69
2.62
2.37
2.28
2.16
1.90
1.87
1.68
1.50
1.63
6.90
5.33
4.62
3.68
2.30
2.08
1.47
1.29
1.00
0.97
0.87
0.80
0.75
0.69
0.70
0.64
0.65
0.63
0.51
0.50
0.46
0.44
0.55
0.43
11.55
4.84
2.72
1.63
1.08
0.61
0.56
46
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
CP2 (Q=990 mL/min) 1 0.99
12.97
14.93
16.89
18.85
20.81
22.79
24.77
26.75
28.73
30.71
Bail 2.76
6.44
12.88
19.32
25.76
32.20
38.64
WASH-1
CP1 (Q=300 mL/min) 0.14
0.28
0.42
0.70
1.30
1.90
2.38
3.26
3.88
4.50
5.10
5.70
6.38
7.06
7.66
8.26
8.86
9.46
10.06
10.66
11.26
11.86
12.46
13.06
13.46
13.86
15.36
15.96
16.56
Time
(min)
9
11
13
15
17
19
21
23
25
27
29
5
9
15
19
32
40
55
1
2
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
54
56
58
DO
(mg/L)
0.30
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.10
0.10
0.10
3.80
2.20
2.50
3.40
3.50
3.30
3.30
2.8
1.9
1.3
1.0
1.0
0.6
0.5
0.4
0.4
0.4
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.3
0.3
0.2
0.2
T
(°C)
26.10
26.20
26.10
26.20
26.10
26.20
26.20
26.20
26.20
26.20
26.20
21.60
22.80
22.70
23.30
22.90
22.90
22.30
7.0
7.1
7.0
6.8
5.9
7.8
9.4
12.6
13.7
13.2
12.8
13.2
13.5
14.0
14.2
13.7
13.7
13.9
13.8
13.8
13.6
13.4
13.2
13.1
13.0
12.8
15.4
15.8
15.1
PH
6.88
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.89
6.88
7.15
7.14
7.10
7.14
7.13
7.17
7.13
8.77
8.35
8.08
7.90
7.74
7.56
7.44
7.38
7.39
7.42
7.43
7.43
7.44
7.45
7.46
7.48
7.48
7.49
7.51
7.52
7.54
7.54
7.55
7.55
7.55
7.56
7.56
7.56
7.57
EC
(US)
21.50
21.60
21.50
21.60
21.60
21.60
21.70
21.70
21.70
21.70
21.70
21.40
21.10
21.50
21.30
21.20
21.30
21.60
808
810
810
823
805
781
775
767
811
816
809
814
816
809
809
824
804
809
814
818
838
829
824
819
824
822
824
822
829
Turb.
(NTU)
0.46
0.50
0.41
0.36
0.36
0.34
0.33
0.30
0.30
0.26
0.28
46.00
138.60
185.00
162.00
304.00
380.00
448.00
128.8
122.9
109.4
89.6
75.8
52.9
38.8
36.5
24.9
21.8
18.3
13.26
11.85
10.79
10.27
10.72
11.09
12.24
12.22
11.50
9.48
8.36
7.78
6.49
5.75
4.67
16.4
5.80
5.52
continued
•47
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
WeU ID/Device Vol
(L)
CP1 (Q=300 mL/min) 16.86
17.46
16.34
18.94
19.54
20.14
20.74
21.20
21.36
21.96
22.78
23.58
24.14
24.46
24.82
BP (Q=1 00 0 mL/min) 0.60
1.40
2.00
2.80
3.76
5.69
6.64
7.62
8.60
9.58
10.56
11.54
12.52
13.50
14.48
15.46
16.44
17.42
18.41
19.40
21.38
23.38
25.38
27.38
29.38
33.38
38.38
39.38
41.38
43.38
49.38
55.60
56.60
Time
(min)
59
61
63
65
67
69
71
73
75
77
79
81
83
85
87
2
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
25
27
29
31
33
37
42
43
45
47
53
70
71
DO
(mg/L)
0.4
0.4
0.5
0.3
0.3
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.3
0.4
0.5
7.1
4.5
4.2
4.1
3.7
3.1
2.8
2.8
3.1
2.9
3.1
4.1
4.3
4.8
4.8
4.2
4.2
3.9
3.8
3.4
3.4
3.2
3.4
3.7
4.2
5.3
5.7
5.2
5.7
5.6
5.3
5.1
5.0
T
(°C)
13.9
13.2
13.4
15.4
15.2
14.0
13.2
13.7
13.7
13.4
12.9
15.4
15.9
15.3
15.4
7.5
8.3
8.7
9.6
9.8
10.0
10.1
10.2
10.2
,10.2
10.2
10.3
10.3
10.3
10.3
10.3
10.4
10.4
10.3
10.3
10.4
10.4
10.4
10.4
10.4
10.4
10.4
10.4
10.4
10.4
10.4
9.8
10.2
PH
7.58
7.58
7.57
7.56
7.56
7.57
7.57
7.57
7.57
7.57
7.57
7.56
7.57
7.57
7.57
7.28
7.28
7.32
7.41
7.42
7.35
7.35
7.34
7.35
7.35
7.38
7.42
7.45
7.49
7.52
7.55
7.57
7.58
7.57
7.58
7.57
7.56
7.58
7.56
7.55
7.54
7.56
7.57
7.59
7.59
7.61
7.52
7.54
EC
(US)
828
' 827
807
824
826
821
819
827
835
827
811
824
827
829
817
554
710
747
781
773
771
777
775
787
787
795
801
803
810
812
819
827
831
827
838
850
861
883
886
892
894
866
864
855
855
852
840
844
Turb.
(NTU)
8.02
15.35
16.73
10.54
9.79
10.32
10.62
10.50
9.41
9.21
9.25
6.28
5.70
7.24
11.23
178.9
169.2
109.8
87.2
86.3
83.6
89.7
77.2
70.3
63.9
79.3
74.0
66.3
68.1
75.4
78.3
83.3
93.1
94.3
101.3
92.2
94.1
123.3
164.2
220
330
370
160
continued
48
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Wei! ID/Device Vol
(L)
CP2 (Q=950 mL/min) 1.00
2.00
3.00
4.00
5.00
5.96
6.94
7.92
8.92
9.92
10.08
11.10
12.12
13.14
14.16
15.18
16.20
17.22
18.24
20.28
22.32
24.40
26.88
28.88
32.88
34.88
37.76
40.64
43.34
45.14
46.94
50.50
Bail 1 .38
3.68
8.28
16.56
24.84
33.12
WASH-2
CP1 (Q=280 mL/min) 1.02
1.36
1.70
2.04
2.72
3.32
3.92
4.52
Time
(min)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
21
23
27
29
31
35
37
39
42
45
47
49
53
2
8
18
37
52
61
3
4
5
6
8
10
12
14
DO
(mg/L)
3.0
2.8
1.9
1.6
1.5
1.5
1.4
1.5
1.9
2.3
2.2
1.5
1.1
0.6
0.5
0.4
0.3
0.3
0.3
0.4
0.4
0.9
2.3
3.0
2.7
1.9
1.1
0.7
0.6
0.7
0.8
1.1
4.2
4.9
5.2
5.3
5.2
6.0
10.3
4.5
3.1
2.2
1.5
1.0
0.8
0.6
T
(°C)
7.6
8.2
9.6
10.9
11.2
11.5
11.8
12.3
12.3
12.6
12.8
12.7
12.7
12.8
12.8
12.8
12.9
12.9
12.8
12.9
13.1
11.5
13.2
13.4
13.7
13.7
13.6
13.4
13.4
13.3
13.4
13.5
12.1
10.8
10.7
10.6
10.5
10.6
2.7
3.6
3.7
3.7
3.6
3.1
4.5
7.1
pH
8.09
7.80
7.62
7.55
7.55
7.57
7.60
7.64
7.66
7.71
7.76
7.83
7.85
7.85
7.85
7.85
7.83
7.82
7.83
7.79
7.77
7.86
7.86
7.88
7.82
7.78
7.76
7.77
7.77
7.76
7.75
7.74
7.10
7.47
7.68
7.76
7.76
7.71
9.77
9.11
8.69
8.14
7.30
6.72
6.49
6.32
EC
(US)
784
787
789
776
785
793
784
797
801
813
803
824
829
816
824
822
844
851
864
883
918
832
832
838
880
922
922
918
915
911
911
910
757
772
820
834
872
886
448
363
296
244
203
147
129
123
Turb.
(NTU)
101.9
85.9
67.3
41.8
33.2
28.6
28.1
42.5
77.9
67.5
77.5
133.5
148.0
141.2
143.6
138.6
116.6
115.9
141.8
440
400
450
800
39.8
277.0
390.0
700.0
769.0
1156.0
22.7
21.3
20.5
17.75
16.89
12.69
11.48
9.54
continued
49
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
CP1 (Q=280 mL/min) 5.12
5.72
6.32
6.92
7.52
8.12
8.60
8.84
9.30
9.76
10.56
10.80
11.20
11.84
12.48
13.12
13.90
14.42
14.94
15.46
15.98
16.58
17.18
17.78
18.38
18.98
19.58
20.78
22.38
23.54
24.12
24.70
25.28
26.16
26.72
27.28
27.84
28.40
28.96
29.52
30.08
30.64
31.20
BP (Q=1 000 mL/min) 1 .88
2.82
3.82
4.82
5.82
Time
(min)
16
18
20
22
24
26
28
30
32
34
38
40
44
50
52
54
56
58
60
62
64 .
66
68
70
72
74
76
80
84
88
90
92
94
96
98
100
102
104
106
108
110
112
114
2
3
4
5
6
DO
(mg/L)
0.6
0.5
0.4
0.4
0.3
0.3
0.3
0.3
0.2
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2 •
0.2
0.2
0.2
0.2
0.2
4.9
1.3
0.8
0.5
0.5
T
7.4
7.3
7.3
7.6
7.8
7.5
7.4
6.7
6.7
6.9
7.0
6.9
5.5
6.2
7.4
8.0
8.1
10.0
10.0
8.5
8.3
8.4
8.5
8.2
7.3
6.9
6.5
9.3
9.1
8.3
7.8
7.5
7.1
7.5
9.3
9.4
8.6
7.1
7.5
9.5
9.2
9.2
9.2
6.7
7.8
8.2
8.4
8.5
PH
6.24
6.20
6.19
6.18
6.18
6.17
6.18
6.18
6.18
6.18
6.18
6.18
6.19
6.20
6.20
6.20
6.21
6.21
6.21
6.21
6.22
6.22
6.22
6.22
6.22
6.23
6.24
6.21
6.22
6.22
6.23
6.23
6.23
6.23
6.23
6.22
6.23
6.23
6.24
6.23
6.22
6.22
6.22
6.65
6.40
6.30
6.27
6.25
EC
ftiS)
121
126
124
125
123
125
124
125
116
111
111
112
114
112
111
109
110
109
109
113
111
111
112
112
113
111 ,
113
111
111
111
110
113
112
112
110
111
112
112
112
111
111
111
111
104
109
110
110
110
Turb.
(NTU)
9.34
8.88
8.83
8.49
8.29
8.05
7.73
7.92
8.50
9.18
8.57
7.92
8.02
58.2
51.7
40.2
55.8
39.2
40.2
22.2
23.2
22.5
21.4
20.4
17.4
17.6
16.2
16.5
13.0
11.5
10.8
10.7
10.6
10.8
13.7
10.9
10.3
10.3
9.8
11.0
10.3
9.3
9.0
42.3
31.2
29.8
28.2
25.2
50
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
BP(Q=1000mL/min) 6.82
7.82
8.82
10.82
12.82
14.82
16.82
18.82
20.82
22.82
24.82
26.82
28.82
29.82
31.82
33.82
37.82
39.82
41.82
43.82
CP2 (Q=1 040 mL/min) 1 .00
2.00
3.00
4.00
5.20
6.40
7.58
8.76
9.80
10.84
11.88
12.92
13.96
15.02
16.08
17.16
18.18
19.20
20.22
21.24
22.28
23.32
24.36
25.40
26.44
27.48
28.52
29.56
Time
(min)
7
8
9
11
13
15
17
19
21
23
25
27
29
30
32
34
38
40
42
44
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
DO
(mg/L)
0.4
0.4
0.3
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
4.8
2.4
1.2
0.8
0.5
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
T
(°C)
8.6
8.7
8.7
8.6
8.7
8.8
8.7
8.7
8.8
8.9
8.9
8.8
8.7
8.7
8.6
8.7
8.6
8.6
8.6
8.6
5.5
8.1
8.8
9.4
9.9
10.1
10.0
10.0
10.3
10.3
10.2
10.3
10.3
10.4
10.5
10.5
10.4
10.4
10.3
10.3
10.3
10.3
10.2
10.4
10.3
10.3
10.2
10.0
pH
6.24
6.24
6.24
6.24
6.24
6.24
6.22
6.18
6.18
6.21
6.23
6.22
6.21
6.20
6.20
6.20
6.19
6.19
6.13
6.14
8.18
6.91
6.57
6.50
6.45
6.42
6.42
6.42
6.42
6.42
6.42
6.43
6.43
6.43
6.42
6.43
6.43
6.42
6.43
6.43
6.43
6.43
6.43
6.43
6.43
6.43
6.43
6.43
EC
(US)
110
110
111
110
111
108
110
110
107
108
108
107
110
110
110
110
110
110
110
110
94.4
113
113
113
110
111
110
110
110
111
110
110
111
110
111
111
110
111
111
111
110
111
111
111
111
111
110
110
Turb.
(NTU)
23.2
20.5
19.7
16.7
15.3
13.7
12.3
12.2
11.1
10.8
10.3
10.8
9.7
9.3
9.3
9.2
8.8
8.9
8.5
8.4
67.3
28.2
21.3
17.3
14.5
11.48
10.49
10.22
9.53
9.21
8.96
8.23
8.63
7.73
8.32
7.65
7.82
7.82
7.44
7.35
7.16
7.20
7.03
7.06
6.86
6.81
6.72
6.64
continued
51
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well IDTDevice
CP2(Q=1 040 mL/min)
Bail
WASH-3
CP1 (Q=300 mL/min)
Vol
(L)
30.60
32.68
33.72
34.76
36.76
37.80
6.44
12.88
19.32
25.76
32.20
1.80
2.60
3.80
4.28
4.88
5.64
6.36
6.96
7.26
8.16
8.76
9.36
9.96
10.56
11.16
11.76
12.36
12.96
13.56
14.16
14.76
15.36
15.96
16.60
17.20
17.80
18.40
19.00
19.60
20.20
20.80
21.40
22.00
22.60
23.20
23.96
Time
(min)
29
31
32
33
35
36
6
12
17
24
32
. 3
5
7
9
11
13
15
17
18
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
51
53
55
57
59
63
65
67
69
71
73
. 75
DO
(mg/L)
0.1
0.1
0.1
0.1
0.1
0.1
3.3
2.6
2.8
3.1
3.5
4.2
1.8
0.8
0.8
0.7
0.4
0.4
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
T
9.9
10.1
10.3
10.3
9.9
9.8
9.2
9.5
9.5
9.4
9.4
8.1
9.2
10.0
10.0
9.8
9.3
9.8
9.9
9.8
9.7
9.8
10.0
10.0
10.2
10.3
10.3
10.0
10.0
10.3
10.4
10.2
10.2
10.3
10.4
10.1
10.1
10.4
10.2
10.2
9.9
9.9
9.7
9.4
9.1
9.2
9.5
PH
6.43
6.43
6.43
6.43
6.43
6.43
7.15
6.63
6.51
6.47
6.45
6.62
, 6.55
6.55
6.55
6.54
6.53
6.53
6.53
6.52
6.52
6.51
6.50
6.50
6.49
6.48
6.48
6.47
6.47
6.47
6.46
6.45
6.45
6.45
6.44
6.44
6.44
6.44
6.43
6.43
. 6.42
6.42
6.42
6.42
6.42
6.42
6.42
EC
110
110
111
111
110
109
120
112
113
112
113
1017
1105
1079
1054
1047
998
978
969
941
952
924
921
910
901
894
886
875
864
864
861
840
836
834
831
820
819
808
796
792
773
768
792
776
768
776
784
Turb.
(NTU)
6.65
6.60
6.50
.6.49
6.39
6.28
414.0
695.0
735.0
720.0
735.0
33.5
24.3
7.3
8.08
9.03
7.40
10.28
10.22
8.50
9.20
9.06
9.58
8.93
7.62
7.32
7.08
6.97
6.51
6.20
5.96
5.55
5.28
5.06
4.92
4.57
4.38
4.25
4.02
3.88
3.82
4.43
3.89
3.85
3.62
3.52
3.68
continued
52
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device Vol
(L)
CP1 (Q=300 mL/min) 24.68
25.28
25.88
26.48
27.12
27.72
28.32
BP (Q=980 mL/min) 2.20
3.04
3.94
5.80
6.76
7.74
8.72
9.70
10.68
11.64
13.64
15.64
17.64
19.64
21.64
23.64
25.64
27.60
31.52
33.48
35.44
37.36
39.28
41.20
43.12
CP2 (Q=1 1 00 mL/min) 0.860
1.72
2.58
3.44
4.44
5.44
6.44
7.44
8.44
9.60
10.76
11.92
13.12
15.52
16.72
19.12
21.28
Time
(min)
77
79
81
83
85
87
89
2
3
4
6
7
8
9
10
11
12
14
16
18
20
22
24
26
28
32
34
36
38
40
42
44
1
2
3
4
5
6
7
8
9
10
11
12
13
15
16
18
20
DO
(mg/L)
0.2
0.2
0.2
0.2
0.2
0.2
0.2
7.7
3.4
2.3
1.4
1.2
1.0
1.0
0.8
0.8
0.7
0.6
0.6
0.5
0.5
0.5
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.3
0.3
5.8
2.4
1.1
0.6
0.4
0.3
0.3
0.3
0.2
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
T
(°C)
9.7
9.6
9.5 .
9.4
9.3
9.1
9.2
9.4
10.4
10.4
10.3
10.2
10.2
10.2
10.1
10.1
10.0
10.0
10.0
9.8
9.9
9.8
9.8
9.8
9.8
9.8
9.8
9.8
9.8
9.8
9.7
9.7
7.0
8.7
9.4
9.9
10.5
11.0
11.2
11.3
11.2
11.0
11.2
11.1
10.9
10.9
11.0
10.9
11.0
pH
6.41
6.41
6.41
6.41
6.41
6.41
6.41
6.78
6.62
6.53
6.42
6.37
6.33
6.32
6.30
6.27
6.23
6.18
6.18
6.15
6.13
6.11
6.09
6.07
6.11
6.12
6.11
6.07
6.07
6.10
6.11
6.10
7.21
6.52
6.51
6.53
6.54
6.53
6.53
6.53
6.50
6.50
6.49
6.48
6.48
6.48
6.47
6.47
6.46
EC
(fiS)
775
765
769
759
765
759
759
664
969
938
915
869
847
815
808
780
759
729
719
694
686
682
673
665
661
654
649
666
650
642
640
654
864
873
978
955
932
886
874
855
839
814
791
770
659
751
744
743
739
Turb.
(NTU)
3.45
3.28
3.32
3.28
2.98
2.89
3.08
10.66
10.28
8.28
4.96
4.35
4.18
4.16
4.14
3.96
3.88
3.72
3.53
3.48
3.38
3.40
3.41
3.26
3.22
4.42
3.28
3.46
3.18
3.53
2.92
3.31
42.2
16.5
6.92
5.73
8.3
5.8
5.8
5.7
5.1
5.2
4.9
5.5
5.8
4.9
5.7
5.7
5.6
continued
53
-------
TABLE A-1. VALUES OF FIELD PARAMETERS DURING PURGING
Well ID/Device
CP2(Q=1100mL/min)
Bail
Vol
(L)
23.44
25.76
28.08
30.40
32.72
34.88
37.04
3.22
6.90
13.80
20.70
27.60
Time
(min)
22
24
26
28
30
32
34
4
8
18
23
31
DO
(mg/L)
0.3
0.3
0.3
0.3
0.3
0.3
0.3
2.7
2.6
2.7
3.4
3.3
T
(°C)
10.9
10.6
10.6
10.7
10.8
10.9
10.8
10.3
10.1
9.8
9.6
9.5
pH
6.46
6.46
6.45
6.45
6.45
6.44
6.44
6.81
6.99
7.08
6.99
7.04
EC
(US)
729
729
724
724
703
702
714
861
824
722
692
687
Turb.
(NTU)
4.7
5.7
4.9
4.6
4.9
4.6
4.2
48.3
32.2
21.0
18.3
10.3
54
-------
TABLE A-2. VOLUMES AND TIMES AT WHICH INDICATOR PARAMETERS REACHED EQUILIBRIUM VALUES
Well ID
WISC-2
WISC-3
WISC^
NEV-1
NEV-2
WASH-1
WASH-2
WASH-3
Device
CP1
BP
CP2
Bailer
CP1
CP2
Bailer
CP1
BP
CP2
Bailer
CP1
BP
CP2
Bailer
CP1
BP
CP2
Bailer
CP1
BP
CP2
Bailer
CP1
BP
CP2
Bailer
CP1
BP
CP2
Bailer
Volume (L)
19.3a
15.2
17.9a
65.3b
16.9
27.1
56.8b
29. 1C
32.6
64.8C
41 .6b
11.6
13.3
12.3
138°
9.2
11.5
13.0
38.6d
24.8C
56.6C
50.5C
33.1d
9.30d
29.8
22.3a
32.2d
19.0a
23.6
23.4a
27.6d
Screen Volumes
2.1
1.6
1.9
7.0
1.8
2.9
6.1
4.7
5.3
10.5
6.7
0.2
0.3
0.2
2.8
7.7
9.6
10.8
32.2
4.0
9.1
8.1
5.3
1.5
4.8
3.6
5.2
3.1
3.8
3.8
4.5
Time (min)
50
30
19
33
43
29
28
89
61
63
21
37
14
13
117
31
12
11
55
87
61
53
61
32
30
21
32
57
24
22
31
a Temperature did not reach equilibrium.
b Pre-determined purge volume.
0 Dissolved oxygen, temperature, and turbididity did not reach equilibrium. Purged volume shown is volume at which
samples were collected.
d Dissolved oxygen and turbidity did not reach equilibrium, purged volume shown is volume at which samples were
collected.
55
-------
APPENDIX B
Summary of Particle Size Analysis
TABLE B-1. SUMMARY OF PARTICLE SIZE ANALYSIS
Concentration (mg/L)
ID
WISC-1
BP
WISC-2
CP1
BP
CP2
Bail
WISC-3
CP1
CP2
Bail
WISC-4
CP1
BP
CP2
Ball
NEV-1
CP1
BP
CP2
Bail
NEV-2
CP1
BP
CP2
Bail
WASH-1
CP1
BP
CP2
Ball
WASH-2
CP1
BP
CP2
Bail
WASH-3
CP1
BP
CP2
Bail
>5.0
(im
16.2
0.5
0
0.6
113.6
0.4
0.1
169
131
2.2
0
6956
2.2
0
0.9
72.4
1.8
1.9
0.2
630
29.5
105.5
1300
841.9
2
2.6
2.2
807.3
17.9
36.9
38.1
30.3
5.0-0.4 0.4-0.1 0.
nm urn
5.9
0.1
0.5
0.2
7.6
0.8
0.5
6.4
4.4
0.8
0.4
8
2.2
0.6
0.6
2.1
1
0
0.9
3.1
0.6
94.1
2.9
2.5
6.4
6.6 ,
1.1
8.9
20.1
2.4
3
8.3
1.2
0.1
0.2
0.1
0.6
0
0
1.8
0.5
0
0
1.2
0.6
1.5
0.8
2
0.8
0
0
1.1
0.1
0.1
0
0.4
0.5
1.5
3.2
0.5
1.1
0.6
0
1.1
.1-0.03
0.0
0
0
0.2
0.5
1.6
0.4
2.6
0.2
0.1
0
0
1
2.2
11
10.4
0.4
0.8
1.6
0.5
0
0.4
0.4
0.4
0
24.3
4.9
0.9
0
0.5
0
0.6
Total °'
23.3
0.7
0.7
1.1
122.3
2.8
1
179.8
136.1
3.1
0.4
6965.2
6
4.3
13.3
86.9
4
2.7
2.7
634.7
30.2
200.1
1303.3
845.2
8.9
35
11.4
817.6
39.1
40.4
41.1-
40.3
Weight Fraction
4-0.03 c 0 0.4-0.03
[im >o.U[im ^m
1.2
0.1
0.2
0.3
1.1
1.6
0.4
4.4
0.7
0.1
0
1.2
1.6
3.7
11.8
12.4
1.2
0.8
1.6
1.6
0.1
0.5
0.4
0.8
0.5
25.8
8.1
1.4
1.1
1.1
0
1.7
0.70
0.71
0.00
0.55
0.93
0.14
0.10 '
0.94
0.96
0.71
0.00
1.00
0.37
0.00
0.07
0.83
0.45
0.70
0.07
0.99
0.98
0.53
1.00
1.00
0.22
0.07
0.19
0.99
0.46
0.91
0.93 ,
0.75
0.05
0.14
0.29
0.27
0.01
0.57
0.40
0.02
0.01
0.03
0.00
0.00
0.27
0.86
0.89
0.14
0.30
0.30
0.59
0.00
0.00
0.00
0.00
0.00
0.94
0.26
0.29
1.00
0.97
0.97
1.00
0.96
>0.4
urn
0.95
0.86
0.71
0.73
0.99
0.43
0.60
0.98
0.99
0.97
1.00
1.00
0.73
0.14
0.11
0.86
0.70
0.70
0.41
1.00
1.00
1.00
1.00
1.00
0.94
0.26
0.29
1.00
0.97
0.97
1.00
0.96
56
-------
APPENDIX C
Summary of Analytical Results
TABLE C-1. TRACE METALS ANALYTICAL RESULTS (CONCENTRATIONS EXPRESSED AS MG/L)
Device/Filtration Type
WISC-1
BP, unfilt.
BP, unfilt.
BP, 0.45 urn
BP, 0.45 urn
Bail, 0.45 urn
Bail, 0.45 um
WISC-2
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1 , 0.45 um
CP1, 5.0 um
CP1,5.0um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WISC-3
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Cd
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
Cr
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Fe
2.8
0.93
0.20
0.27
6.1
6.2
0.05
0.05
<0.01
0.01
0.04
0.03
0.07
0.07
0.03
0.03
0.02
<0.01
0.11
0.11
<0.01
<0.01
0.02
0.03
3.5
3.3
0.01
<0.01
0.04
0.05
0.02
0.01
0.05
0.04
0.01
0.03
6.47
Pb
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Mn
5.0
4.9
4.9
4.9
4.88
4.88
0.01
0.02
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.11
0.10
0.02
0.01
<0.01
0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.14
As .
<0.002
<0.002
<0.002
<0.002
0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
Ba
0.137
0.134
0.137
0.138
0.136
0.137
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.010
0.044
0.036
0.013
0.012
0.010
0.010
0.011
0.012
0.010
0.009
0.010
0.010
0.054
continued
57
-------
TABLE 0-1. TRACE METALS ANALYTICAL RESULTS (CONCENTRATIONS EXPRESSED AS MG/L)
Device/Filtration Type
Bail, unfilt.
Bail, 0.45 urn
Bail, 0.45 urn
WISC-4
CP1, unfilt.
CP1, unfilt
CP1, 0.45 urn
CP1, 0.45 urn
BP, 0.45 urn
BP, 0.45 um
BP, unfilt.
BP, unfilt.
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Bail, unfilt.
Ball, 0.45 um
Bail, 0.45 um
NEV-1
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
CP1,5.0um
CP1,5.0um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Ball, unfilt.
Bail, 0.45 um
Cd
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
Cr
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.09
0.13
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.02
0.02
<0.02
<0.02
0.02
<0.02
<0.02
<0.02
<0.02
Fe
6.52
0.04
0.13
2.0
2.0
0.02
0.02
0.03
0.04
0.06
0.07
0.59
0.56
0.04
0.06 '
62.0
80.0
<0.01
<0.01
0.11
0.10
0.08
0.14
.0.11
0.10
0.04
0.04
0.04
0.04
0.04
0.04
0.04
0.06
0.06
0.03
0.05
0.05
1.28
1.47
0.04
Pb
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Mn
0.16
0.02
0.02
0.06
0.06
0.03
0.03
0.07
0.04
0.04
0.04
0.05
0.05
0.04
0.04
1.15
1.56
0.11
0.11
0.73
0.74
0.73
0.73
0.75
0.75
0.73
0.72
0.73
0.73
0.73
0.73
0.74
0.74
0.74
0.74
0.74
0.74
0.79
0.79
0.73
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
0.008
0.008
<0.002
<0.002
0.14
0.12
0.14
0.14
0.12
0.14
0.12
0.13
0.12
0.12
0.11
0.11
0.16
0.16
0.14
0.15
0.15
0.15
0.14
0.15
0.15
0.060
0.012
0.014
0.175
0.175
0.159
0.159
0.163
0.168
0.164
0.164
0.179
0.171
0.170
0.170
0.819
0.566
0.189
0.192
0.027
0.026
0.025
0.026
0.026
0.026
0.027
0.027
0.027
0.027
0.027
0.027
0.027
0.027
0.026
0.026
0.026
0.026
0.043
0.050
0.025
continued
58
-------
TABLE C-1. TRACE METALS ANALYTICAL RESULTS (CONCENTRATIONS EXPRESSED AS MG/L)
Device/Filtration Type
Bail, 0.45 urn
Bail, 5.0 urn
Bail, 5.0 um
NEV-2
CP1,unfilt.
CP1,unfilt.
CP1,0.45um
CP1,0.45um
BP, unfilt.
BP, unfilt..
BP, 0.45 um
BP, 0.45 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WASH-1
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WASH-2
CP1 , unfilt.
CP1 , unfilt.
CP1,0.45um
CP1,0.45um
Cd
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
Cr
<0.02
0.04
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.02
<0.02
<0.02
0.06
0.05
<0.02
<0.02
0.05
0.04
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.19
0.19
<0.02
<0.02
0.06
0.07
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Fe
0.04
0.10
0.04
0.06
0.06
0.05
0.05
0.02
0.03
0.02
0.02
0.02
0.06
0.05
0.02
14.1
13.8
0.03
0.05
1.52
1.73
0.03
0.02
7.69
7.00
<0.01
<0.01
67.6
68.2
0.02
<0.01
42.1
42.3
<0.01
<0.01
1.38
1.40
1.05
1.10
Pb
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Mn
0.74
0.73
0.72
1.05
1.06
1.07
1.09
1.06
1.05
1.07
1.07
1.10
1.11
1.10
1.09
1.43
1.42
1.04
1.07
0.17
0.13
0.08
0.08
0.19
0.18
0.06
0.06
1.17
1.24
0.08
0.08
0.90
0.90
0.07
0.07
0.03
0.03
0.03
0.03
As
0.16
0.14
0.14
0.11
0.11
0.10
0.11
0.08
0.08
0.08
0.08
0.12
0.12
0.11
0.10
0.13
0.12
0.12
0.11
0.006
0.007
0.006
0.005
0.008
0.008
0.006
0.006
0.015
0.015
0.005
0.005
0.020
0.020
0.006
0.006
<0.002
<0.002
<0.002
<0.002
Ni
0.026
0.026
0.026
0.025
0.024
0.025
0.025
0.026
0.026
0.026
0.025
0.025
0.025
0.024
0.023
0.399
0.350
0.024
0.024
0.05
0.03
0.01
0.02
0.02
0.02
<0.0
0.01
0.17
0.18
0.01
0.02
0.09
0.09
<0.01
<0.01
0.01
<0.01
0.01
<0.01
continued
59
-------
TABLE C-1. TRACE METALS ANALYTICAL RESULTS (CONCENTRATIONS EXPRESSED AS MG/L)
Device/Filtration Type
CP1,5.0um
CP1,5.0um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WASH-3
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
CP1,5.0um
CP1,5.0um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Ball, 0.45 um
Bail, 0.45 um
Bail, 5.0 um
Bail, 5.0 um
Cd
<0.005
<0.005
<0.005
. <0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
Cr
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.04
0.04
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
Fe
1.34
1.34
1.52
1.52
1.31
1.31
1.42
1.42
1.16
1.18
19.6
17.4
1.14
1.12
19.9
20.9
20.1
20.1
20.6
20.3
19.1
18.9
18.5
18.2
18.2
18.5
19.7
19.4
16.3
17.3
18.7
18.5
18.4
19.0
18.5
18.5
18.4
18.4
Pb
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Mn
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.18
0.19
0.03
0.04
1
2.97
.3.06
2.97
2.97
3.01
2.97
2.60
2.60
2.56
2.50
2.48
2.50
2.81
2.75
2.43
2.50
2.66
2.64
2.64
2.72
2.66
2.64
2.64
2.64
As
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
0.004
0.003
0.002
0.002
0.007
0.007
0.008
0.007
0.007
0.008
0.007
0.007
0.008
0.007
0.007
0.008
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.006
0.005
0.005
0.006
0.006
Ni
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.05
0.05
<0.01
<0.01
0.04
0.03
0.03
0.03
0.04
0.04
0.06
0.04
0.04
0.05
0.05
0.06
0.05
0.05
0.06
0.05
0.05
0.06
0.04
0.04
0.04
0.04
0.05
0.05
60
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS (CONCENTRATIONS EXPRESSED AS MG/L)
Device / Filtration Type
WISC-1
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
Bail, 0.45 um
Bail, 0.45 um
WISC-2
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
CP1, 5.0 um
CP1, 5.0 um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
Device / Filtration Type
WISC-1
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
Bail, 0.45 um
Bail, 0.45 um
WISC-2
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
CP1, 5.0 um
CP1, 5.0 um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
EC
2060
1950
2040
1920
2030
2030
227
219
214
216
216
218
204
208
203
203
203
203
HCO3
1430
1410
1420
1410
1390
1390
125
125
116
111
113
113
115
110
115
117
112
112
COS
<5
<5
<5
<5
<5
<5
3.2
2.2
7.9
10.4
9.4
9.4
8.5
10.8
8.4
6.9
10.0
9.6
PH
6.87
6.90
6.81
6.84
6.88
6.96
8.49
8.43
8.79
8.80
8.78
8.74
8.76
8.82
8.78
8.78
8.77
8.77
Cl
23.2
23.7
23.8
24.0
23.6
23.8
7.5
6.4
5.5
5.2
5.7
6.2
4.0
4.3
3.4
3.4
3.3
3.3
Dissolved Si
Solids
1012
1040
1044
982
1230
1267
138
130
128
130
126
125
128
128
124
126
124
118
SO4
177
176
181
179
161
162
2.03
1.53
1.16
1.14
1.25
1.51
0.95
1.0
0.69
0.67
0.68
0.69
26.1
26.4
26.2
26.2
25.8
25.7
11.8
11.7
12.0
11.8
11.7
11.6
11.7
11.6
11.9
11.9
11.6
11.7
NO3 Na
<0.01 16.4
0.04 16.4
0.01 16.5
0.03 16.6
<0.01 16.5
<0.01 16.4
0.02 39.9
0.1 40.3
0.01 40.5
<0.01 41.1
0.01 40.7
0.01 40.5
0.02 38.9
0.02 38.9
<0.01 38.9
<0.01 38.9
<0.01 38.9
<0.01 39.3
Organic
Carbon
7.4
8.3
8.6
8.2
8.5
7.5
3.3
3.1
4.4
3.5
3.1
3.3
3.0
3.1
3.5
3.5
3.5
2.9
K
1.66
1.69
1.69
1.71
1.69
1.69
0.81
0.81
0.83
0.78
0.81
0.81
0.78
0.78
0.76
0.76
0.78
0.76
Ca
312
312
311
311
305
305
6.66
6.57
6.39
6.46
6.57
6.48
6.39
6.39
6.21
6.21
6.21
6.21
Mg
126
128
126
126
123
123
3.05
3.05
2.97
3.10
3.02
3.02
2.99
2.97
2.94
2.88
2.91
2.91
61
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
WISC-2
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 urn
CP2, 0.45 um
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WISC-3
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1, 0.45 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Device / Filtration Type
WISC-2
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WISC-3
CP1, unfilt.
CP1, unfilt.
CP1, 0.45 um
CP1, 0.45 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 urn
EC
204
205
201
201
201
201
217
217
206
206
212
216
216
216
217
215
216
217
HCO3
131
132
116
114
126
123
124
125
117
120
126
128
125
127
142
144
126
" 127
COS
<5
<5
8.0
9.0
3.2
3.5
8.8
6.6
7.7
6.1
'
8.3
7.7
8.9
8.4
<5
<5
8.8
7.9
PH
8.07
8.13
8.79
8.78
8.48
8.50
8.74
8.70
8.71
8.70
8.70
8.68
8.72
8.73
8.26
8.25
8.73
8.73
Cl
3.3
3.3
3.1
3.1
3.0
3.0
3.3
3.3
3.0
3.2
>
2.6
2.6
'2.7
2.8
2.6
2.7
2.7
2.7
Dissolved
Solids
126
128
126
121
120
124
154
156
126
127
132
128
131
128
130
128
132
130
'SO4
0.73
0.70
0.62
0.60
0.53
0.52
0.72
0.76
0.49
0.47
<0.5
<0.5
<0.5
<0.5
0.5
<0.5
<0.5
<0.5
Si
11.
11.
11.
11.
11.
11.
11.
11.
11.
11.
12,
12
13
12
Organic
Carbon
5
5
8
8
7
6
6
6
7
6
.5
.7
.2
.9
12.7
NO3
0.04
0.02
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
12
12
13
.6
.9
.0
<0.01
0.01
0.02
0.02
0.02
<0.01
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
WISC-3
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WISC-4
CP1, unfilt.
CP1, unfilt.
CP1, 0.45 urn
CP1,0.45um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Bail, unfilt.
Device / Filtration Type
WISC-3
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WISC-4
CP1, unfilt.
CP1, unfilt.
CP1 , 0.45 um
CP1,0.45um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Bail, unfilt.
EC
239
245
217
217
1430
1430
1430
1440
1420
1430
1430
1430
1420
1350
1420
1420
1450
1450
HC03
142
158
135
131
860
865
858
858
857
857
856
857
854
857
854
854
132
1327
CO3
6.7
<5
4.0
6.7
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
75
<5
pH
8.66
8.29
8.52
8.61
7.75
7.77
7.80
7.81
7.34
7.60
7.60
7.68
7.67
7.65
7.48
7.48
7.88
7.84
Cl
3.7
3.6
2.7
2.8
39.1
39.1
38.9
39.2
38.3
38.9
38.7
38.8
37.6
37.9
38.0
38.2
39.6
40.1
Dissolved Si
Solids
155
161
128
131
889
911
895
889
874
822
864
870
668
749
840
812
918
933
S04
1.0
<1.0
<1.0
<1.0
128
145
141
141
147
147
146
146
140
140
140
147
149
147
12.8
12.8
13.0
12.9
25.2
25.8
25.8
30.0
26.6
26.9
27.8
27.5
28.4
28.5
28.5
28.6
23.8
24.7
NO3 Na
<0.01 39.8
<0.01 39.6
<0.01 39.2
<0.01 39.0
0.08 67.3
0.10 68.3
0.07 66.3
0.11 66.3
0.03 69.6
0.04 68.6
0.04 68.6
0.05 68.6
0.05 68.6
0.04 68.4
0.05 68.8
0.04 68.4
0.10 69.0
0.10 69.0
Organic
Carbon
4.0
4.0
4.1
3.9
2.7
2.7
4.1
3.1
2.6
2.6
3.5
3.3
2.8
2.8
3.5
3.0
15.1
14.7
K
1.28
1.30
0.97
0.99
2.88
2.95
2.78
2.80
3.18
3.18
3.18
3.20
3.01
3.01
3.03
3.03
3.50
3.47
Ca
31.2
35.2
6.73
6.73
72.2
72.2
68.8
68.5
68.3
67.9
67.9
67.9
67.3
67.0
65.8
66.2
220
141
Mg
16.4
18.2
4.33
4.33
142
142
141
142
141
140
140
140
138
138
137
136
179
143
63
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
WISC-4
Bail, 0.45 um
Bail, 0.45 um
NEV-1
CP1,unfilt.
CPl.unfilt.
CP1,0.45um
CP1,0.45um
CP1,5.0um
CP1,5.0um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Device / Filtration Type
WISC-4
Bail, 0.45 um
Bail, 0.45 um
NEV-1
CPl.unfilt.
CPl.unfilt.
CP1,0.45um
CP1,0.45um
CP1,5.0um
CP1,5.0um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2. 0.45 um
EC
1450
1440
f
22000
22000
22000
22000
22000
22000
22000
22000
22000
22000
22000
22100
22000
22000
22000
22000
HCO3
857
857
415
414
413
415
415
413
415
414
415
416
414
415
415
415
414
414
PH
7.89
7.89
7.21
7.22
7.23
7.23
7.32
7.41
7.22
7.20
7.18
7.21
7.22
7.24
7.12
7.12
7.19
7.14
COS
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
CI
39.0
39.0
6320
6320
6300
6310
6300
6300
6220
6310
6300
6290
6320
6300
6260
6320
6310
6280
Dissolved Si
Solids
889
903
14950
14960
14960
14980
15000
14980
14880
14950
14930
14960
14890
14940
14890
14930
14910
14930
SO4
140
139
2570
2560
2550
2560
2560
2550
2540
2550
2540
2540
2550
2550
2560
2550
2570
2560
26.4
27.1
100
104
101
102
102
102
101
101
102
101
102
102
101
100
102
101
N03
0.04
0.04
<0.01
<0.01
0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Na
69.0
68.8
4500
4480
4460
4480
4480
4460
4480
4510
4500
4500
4520
4510
4550
4480
4500
4500
Organic
Carbon
2.7
2.8
6.1
6.3
7.3
8.0
7.0
5.9
6.4
7.7
6.5
6.2
6.6
6.8
6.7
7.1
8.0
7.5
K
3.07
3.07
60.8
60.6
61.1
61.4
60.8
60.8
63.8
63.8
63.8
63.8
63.8
63.8
64.1
63.8
64.1
64.1
Ca
68.9
68.9
437
434
434
437
437
437
457
457
457
461
457
457
457
457
453
453
Mg
141
140
214
212
216
216
214
215
222
218
220
222
222
222
218
218
217
218
64
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
NEV-1
CP2, 5.0 urn
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
Bail, 5.0 um
Bail, 5.0 um
NEV-2
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
EC
22000
22000
22000
22000
22000
22000
22000
22000
20600
20600
20700
20600
20800
20800
20800
20800
PH
7.22
7.24
7.60
7.61
7.63
7.57
7.72
7.81
7.40
7.41
7.52
7.59
7.26
7.31
7.23
7.27
Dissolved
Solids
14950
14930
14930
14900
14940
14900
14920
14900
14070
14080
14030
14010
14100
14100
14130
14160
Si
101
101
101
102
103
102
101
101
98.2
98.0
98.6
97.4
97.7
97.4
97.0
96.9
Organic
Carbon
6.4
6.7
7.4
8.1
8.2
8.1
7.9
7.5
5.8
5.8
6.1
6.3
5.7
5.9
5.5
5.5
Device / Filtration Type HCO3 COS Cl
SO4
N03
Na
K
Ca
Mg
NEV-1
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
Bail, 5.0 um
Bail, 5.0 um
NEV-2
CP1, unfilt.
7 CP1, unfilt.
CP1,0.45um
CP1,0.45um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
414
415
421
417
417
417
417
416
394
394
394
393
395
396
395
397
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
6310
6300
6150
6240
6260
6250
6270
6280
5750
5760
5770
5750
5800
5780
5780
5770
2530
2540
2580
2470
2580
2520
2570
2570
2640
2640
2650
2640
2650
2650
2640
2630
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.01
0.01
<0.01
<0.01
0,01
0.01
4470
4480
4440
4470
4450
4480
4490
4470
4170
4210
4190
4190
4280
4280
4250
4250
64.4
64.1
62.5
62.3
62.3
62.3
62.3
62.0
55.0
55.6
55.0
55.3
57.5
57.5
56.9
57.5
457
457
452
452
441
441
441
441
416
416
416
416
428
428
428
428
217
218
221
221
215
218
216
216
19
197
197
197
202
199
199
202
65
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
NEV-2
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Ball, unfilt.
Bait, unfilt.
Ball, 0.45 um
Ball, 0.45 um
WASH-1
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
EC
20800
20800
20700
20800
20700
20700
20700
20700
839
844
841
837
867
866
865
864
pH
7.28
7.17
7.17
7.19
7.36
7.44
7.49
7.60
8.08
8.05
8.04
8.08
8.17
8.20
8.15
8.19
Dissolved
Solids
14120
14180
14140
14100
14140
14170
14150
14150
470
. 475
468
459
466
470
461
458
Si
98.1
97.7
98.5
97.8
96.9
97.2
98.6
98.0
26.7
26.8
26.9
26.8
25.9
26.0
26.1
25.8
Organic
Carbon
6.1
6.0
5.6
5.4
5.7
5.8
7.1
7.1
2.0
1.9
2.9
2.4
2.0
2.2
2.9
3.7
Device/Filtration Type HCO3 CO3 Cl
SO4
NO3
Na
Ca
Mg
NEV-2
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Ball, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WASH-1
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
396
395
397
396
412
417
395
395
492
492
488
489
514
513
507
508
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
5800
5790
5820
5810
5660
5750
5790
5830
39.9
40.2
40.4
40.0
38.4
38.9
39.2
39.2
2650
2630
2650
2650
2690
2670
2680
2680
22.0
21.9
22.0
22.1
22.0
22.2
22.8
22.8
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.01
0.01
0.02
0.02
0.02
0.02
<0.01
<0.01
<0.01
<0.01
4300
4260
4280
4250
4200
4220
4220
4220
69.8
69.8
70.6
69.8
72.7
72.3
71.4
71.9
57.8
57.5
57.5
57.8
56.1
56.1
55.3
55.0
3.81
3.78
3.74
3.74
4.45
4.43
4.30
4.28
428
428
428
428
481
488
419
416
48.2
48.6
48.4
48.2
52.9
52.9
51.3
51.3
199
202
198
200
205
207
197
199
46.4
45.7
46.3
45.7
50.0
49.
47.6
47.3
66
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
WASH-1
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 urn
CP2, 0.45 urn
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WASH-2
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
CP1,5.0um
CP1, 5.0 um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
Device / Filtration Type
WASH-1
CP2, unfilt.
CP2, unfiit.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WASH-2
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1,0.45um
CP1,5.0um
CP1, 5.0 um
BP, unfilt.
BP, unfilt.
BP, 0.45 um
BP, 0.45 um
EC
814
834
837
837
867
877
883
863
96.7
99.7
99.7
101
100
100
100
99.0
99.3
99.0
HCO3
517
514
483
484
536
542
527
510
39.5
38.4
38.1
39.3
39.6
38.8
37.1
38.1
37.9
38.7
PH
8.13
8.15
8.12
8.15
8.17
8.14
8.16
8.14
6.86
6.79
6.65
6.52
6.51
6.46
6.60
6.58
6.57
6.52
COS Cl
<5 40.3
<5 39.3
<5 40.8
<5 40.3
<5 37.8
<5 38.2
<5 38.4
<5 38.4
<5 3.5
<5 3.7
<5 3.6
<5 3.6
<5 3.7
<5 3.8
<5 3.6
<5 3.7
<5 3.6
<5 3.6
Dissolved Si
Solids
475
466
467
471
458
474
486
483
92
88
82
85
88
86
86
87
82
82
SO4
25.0
25.5
24.7
24.5
22.8
21.9
21.4
22.2
11.82
11.91
11.86
11.93
11.91
11.9
12.1
12.1
12.0
12.0
26.0
26.1
26.5
25.9
25.5
25.7
26.4
25.9
17.5
17.6
17.6
17.4
17.3
17.3
17.2
17.4
17.3
17.3
NO3 Na
<0.01 69.4
<0.01 70.1
<0.01 69.4
<0.01 69.3
0.01 72.7
0.01 72.8
0.02 70.6
0.02 71.4
0.05 6.34
0.04 6.29
0.04 6.29
0.04 6.17
0.04 6.29
0.04 6.29
0.03 6.10
0.03 6.14
0.03 6.10
0.03 6.02
Organic
Carbon
1.9
2.0
2.7
2.6
2.4
2.4
4.1
2.8
13.2
13.3
13.7
13.5
13.0
13.2
12.7
12.3
12.9
12.9
K
5.06
5.08
4.18
4.18
4.95
4.95
4.24
4.26
0.62
0.62
0.62
0.61
0.62
0.62
0.64
0.64
0.64
0.64
Ca
74.1
74.6
52.9
52.4
66.1
65.6
49.2
49.2
6.36
6.36
6.27
6.17
6.27
6.27
6.60
6.60
6.50
6.50
Mg
52.8
52.
43.9
43.6
56.4
56.6
47.1
47.3
5.27
5.25
5.20
5.13
5.19
5.27
5.25
5.30
5.22
5.09
67
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
WASH-2
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 urn
CP2, 0.45 urn
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Ball, 0.45 um
WASH-3
CP1, unfilt.
CP1, unfilt.
CP1,0.45um
CP1, 0.45 um
CP1,5.0um
CP1,5.0um
BP, unfilt.
BP, unfilt.
Device / Filtration Type
WASH-2
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
Bail, unfilt.
Ball, unfilt.
Bail, 0.45 um
Bail, 0.45 um
WASH-3
CP1, unfilt.
CP1, unfilt.
7CP1,0.45um
CP1,0.45um
CP1, 5.0 um
CP1,5.0um
BP, unfilt.
BP. unfilt.
EC
101 •
99.0
98.4
98.6
104
104
98.5
98.5
645
657
665
650
635
648
608
608
HCO3
39.2
38.7
37.8
37.5
43.7
44.3
38.5
39.1
366
372
378
367
355
368
340
339
COS
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
PH
6.68
6.65
6.64
6.67
6.70
6.72
6.64
6.69
6.56
6.54
6.59
6.62
6.60
6.66
6.90
7.08
Cl
3.6
3.7
3.5
3.5
4.8
4.8
3.6
3.6
8.2
8.3
8.1
7.9
8.3
8.2
8.1
8.2
Dissolved
Solids
85
88
84
84
144
151
84
85
398
398
383
381
391
394
374
376
SO4
12.3
12.1
12.1
12.0
14.1
18.40
12.0
12.0
28.5
28.2
27.6
28.4
29.1
28.1
27.1
27.1
NO3
0.03
0.03
0.03
0.03
0.03
0.03
0.04
0.04
4.21
4.08
3.84
4.06
4.28
4.00
2.77
2.38
Si
17.2
17.4
17.4
17.5
17.0
17.2
17.1
17.0
41.0
41.4
42.3 ,
42.0
41.0
41.3
40.5
40.6
Na
6.10
6.10
6.18
6.18
7.48
7.24
7.24
7.40
' 19.5
19.4
19.1
19.0
18.6
18.3
16.8
16.5
Organic
Carbon
12.7
13.4
12.0
12.1
21.6
19.7
14.1
12.7
22.3
20.4
23.5
22.6
22.2
20.8
22.9
23.4
K
0.64
0.64
0.63
0.64
1.63
1.51
0.74
0.74
17.4
17.6
17.1
17.2
17.1
17.0
15.8
15.8
Ca
6.50
6.50
6.50
6.61
15.7
14.2
7.44
7.44
52.1
52.0
52.0
52.0
52.0
51.8
49.1
49.1
Mg
5.20
5.25
5.16
5.16
11.4
10.7
4.13
4.03
30.5
29.
29.6
29.7
29.3
29.6
29.2
29.0
68
-------
TABLE C-2. GROSS CHEMISTRY ANALYTICAL RESULTS CONT.
Device / Filtration Type
WASH-3
BP, 5.0 um
BP, 5.0 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
Bail, 5.0 um
Bail, 5.0 um
Device / Filtration Type
WASH-3
BP, 0.45 um
BP, 0.45 um
BP, 5.0 um
BP, 5.0 um
CP2, unfilt.
CP2, unfilt.
CP2, 0.45 um
CP2, 0.45 um
CP2, 5.0 um
CP2, 5.0 um
Bail, unfilt.
Bail, unfilt.
Bail, 0.45 um
Bail, 0.45 um
Bail, 5.0 um
Bail, 5.0 um
EC
586
596
624
623
611
624
601
607
615
618
628
623
612
605
HCO3
328
335
322
331
351
352
341
353
335
338
346
347
352
345
340
333
COS
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
<5
PH
7.00
6.80
6.88
6.98
6.98
6.92
6.95
6.93
6.85
6.89
7.05
6.91
6.96
7.10
Cl
7.8
7.8
8.1
8.3
8.1
8.1
7.8
8.0
8.2
8.3
8.2
8.3
7.9
7.8
8.4
8.2
Dissolved
Solids
361
368
385
387
370
388
375
373
373
377
366
353
373
370
SO4
28.4
27.7
28.5
27.9
27.3
27.4
27.9
27.0
28.2
27.9
27.8
27.7
27.5
27.9
28.0
28.2
NO3
2.56
2.37
3.00
2.75
3.30
3.86
3.07
3.56
3.68
2.75
1.74
2.02
2.34
2.15
2.26
2.39
Si
40.3
40.5
40.5
41.0
41.7
42.0
40.8
40.5
40.0
40.5
41.0
40.9
40.9
40.2
Na
16.5
16.4
16.4
16.4
17.5
17.3
17.0
17.2
16.9
16.9
17.2
17.1
17.0
16.8
16.8
16.8
Organic
Carbon
23.8
22.8
22.4
22.4
22.6
22.7
23.1
23.6
24.9
24.9
23.9
24.8
25.0
23.3
K
15.7
15.5
15.5
15.7
16.5
16.5
15.2
15.7
15.9
15.9
15.2
15.7
15.6
15.7
15.7
15.7
Ca
48.2
48.2
48.2
48.2
51.2
50.8
47.8
49.5
49.5
49.5
47.8
48.6
48.2
48.2
48.2
48.6
Mg
29.1
28.8
28.9
28.9
29.2
29.3
27.9
28.6
28.9
28.6
28.0
28.8
28.8
28.8
28.9
28.8
69
-------
APPENDIX D
Results of Statistical Analysis
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS
DeviceNumbering
1=CP1
2 = BP
3 = CP2
4 = Bailer
Filtration Method Numbering
1 = Unfiltered
2 = 0.45 urn filtered
Well Numbering
1 = WISC-2 6 = WASH-1
3 = WISC-4 7 = WASH-2
4 = NEV-1 9 = WASH-3
5 = NEV-2
#3 Ba 4 Wells, 4 Devices, 2 Filtration Methods
Table of Means (log transformed)
Device Ba
1
2
3
4
Filter
1
2
Analysis of Variance Table
Source
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
-3.44
-3.42
-3.43
-2.64
Ba
-3.04
-3.42
df
3
3
1
3
21
32
63
SS
67.0096
7.4962
2.2720
6.0764
4.2794
0.1179
87.2515
MS
22.3365
2.4987
2.2720
0.0255
0.2038
0.0037
F P
12.3 .0001
11.1 .0032
9.9 .0003
#4 Ni 3 Weils, 4 Devices, 2 Filtration Methods
Table of Means (log transformed)
Deyice
1
2
3
4
Filter
1
2
Analysis of Variance Table
Spurce
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
Ni
-3.93
-3,98
-3,52
-3.51
Ni
-3.39
-4.07
df
2
3
1
3
14
24
47
SS
12.5101
2.3317
5.5574
1.8784
11.2685
0.7777
34.3238
MS
6.2550
0.7772
5.5574
0.6261
0.8049
0.0324
F P
0.97 .43
6.90 .0199
0.77 .53
70
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#6 Ba 4 Wells, 3 Devices, 2 Filtration Methods
Table of Means (log transformed)
Device Ba
1 -3.4363
2 -3.4200
3 -3.4284
Filter Ba
1 -3.4177
2 -3.4388
Analysis of Variance Table
Source
Wells
Devices
Filter •
DxF
Exp. Error
Samp. Error
Total
df
3
2
1
2
14
24
47
SS
50.3986
0.0021
0.0053
0.0021
0.0231
0.0055
50.4367
MS
16.7995
0.0011
0.0053
0.0010
0.0016
0.0002
F
0.65
3.24
0.63
P
.54
.093
.55
#6 Ni 3 Wells, 3 Devices, 2 Filtration Methods
Table of Means (log transformed)
Device Ni
1
2
3
Filter
1
2
Analysis of Variance Table
Source
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
-3.9316
-3.9752
-3.5188
Ni
-3.5663
-4.0507
df
2
2
1
2
10
18
35
SS
13.06333
1 .52237
2.11185
0.49371
4.12528
0.77766
25.41991
MS F P
6.53167
0.76118 . 1.8 .21
2.11185 5.1 .048
0.24686 0.6 .94
0.41253
0.04320
#8 Ba 4 Wells, 4 Devices, Unfiltered Samples Only
Table of Means (log transformed)
Device
Ba
1
2
3
4
Analysis of Variance Table
Source
-3.4221
-3.4187
-3.4123
-1.9159
_S£
Wells
Devices
Exp. Error
Samp. Error
Total
3
3
9
16
31
34.476204
13.531530
3.567998
0.110944
51.686676
11.492067
4.510510
0.396444
0.006934
11.38 0.002
71
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#8 Ni 3 Wells, 4 Devices, Unfiltered Samples Only
Table of Means (log transformed)
Device
1 -3.7397
2 -3.8444
3 -3.1143
4 -2.8742
Analysis of Variance Table
Source df
Wells
Devices
Exp. Error
Samp. Error
Total
2
3
6
12
23
SS
8.27328
4.02345
5.45940
0.25569
18.01181
MS F P
4.13664
1.34115 1.47 0.313
0.90990
0.02131
#8 Fe 7 Wells, 4 Devices, Unfiltered Samples Only
Table of Means (log transformed)
Well Fe.
1 -1.6597
3 0.4144
4 -2.0435
5 -1.8125
6 2.6092
7 1.0037
8 2.9653
Pevjc? Fe
1 -0.5061
2 -0.9954
3 -0.2283
4 2.5737
Analysis of Variance Table
Source df
Wells 6
Devices 3
Exp. Error 18
Samp. Error 28
Total 55
SS MS F P
213.470 35.578
108.433 36.144 10.41 0.000
62.502 3.472
0.852 0.030
385.257
#8 Mn 7 Wells, 4 Devices, Unfiltered Samples Only
Table of Means (log transformed)
Well Mn
1 -3.9310
3 -2.1839
4 -0.2916
5 0.1403
6 0.8783
7 -3.0519
8 -1.0168
Device Mn
1 -1.6622
2 -1.7616
3 -1.4429
4 -0.3788
Analysis of Variance Table
$pgrce df
Wells 6
Devices 3
Exp. Error 18
Samp. Error 28
Total 55
SS MS F P
155,2640 25.8773
16.9783 5.6594 5.31 0.008
19.1922 1.0662
0.3333 0.0119
191,7678
72
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#8 As 7 Wells, 4 Devices, Unfiltered Samples Only
Table of Means (log transformed)
We!! As
1
3
4
5
6
7
8
Device
1
2
3
4
-6.2146
-5.8680
-1.9719
-2.2334
-4.4947
-6.0773
-5.0389
As
-4.6993
-4.7200
-4.5589
-4.2497
Analysis of Variance Table
Source
df
SS
MS
Wells
Devices
Exp. Error
Samp, error
Total
6
3
18
28
55
152.7663
1.9773
3.3941
0.0739
158.2116
25.4610
0.6591
0.1886
0.0026
3.50 0.037
#7 Fe 7 Wells, 4 Devices, 2 Filtration Methods, Omit 4,1
Table of Means (log transformed)
Device Filter Fe
1
1
2
2
3
3
4
1
2
1
2
1
2
2
-0.5061
-2.0565
-0.9954
-2.2037
-0.2283
-2.2166
-2.4637
Analysis of Variance Table
Source
df
SS
MS
Wells
Treatment
Exp. Error
Samp. Error
Total
6
6
36
49
97
470.107
71.433
149.663
2.196
693.398
78.351
11.905
4.157
0.045
2.86 0.022
#7 Mn 7 Wells, 4 Devices, 2 Filtration Methods, Omit 4,1
Table of Means (log transformed)
Device Filter Mn
1
1
2
2
3
3
4
1
2
1
2
1
2
2
-1 .6622
-1 .8990
-1.7616
-1 .8833
-1.4429
-1 .8807
-1.6811
Analysis of Variance Table
Source
df
SS
MS
Weils
Treatment
Exp. Error
Samp. Error
Total
6
6
36
49
97
351.7712
2.2922
13.2671
0.7200
368.0506
58.6285
0.3820
0.3685
0.0147
1.04 0.418
73
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#7 As 7 Wells, 4 Devices, 2 Filtration Methods, Omit 4,1
Table of Means (log transformed)
Device Filter As
1
1
2
2
3
3
4
1
2
1
2
1
2
2
-4.6993
-4.7086
-4.7200
-4.7573
-4.5589
-4.7492
-4.7267
Analysis of Variance Table
M§_
Wells
Treatment
Exp. Error
Samp. Error
Total
6
6
36
49
97
283.8582
0.3745
2.0123
0.0787
286.3238
47.3097
0.0624
0.0559
0.0016
1.12 0.372
#2 Fe 7 Wells, 4 Devices, 2 Filtration Methods
Table of Means (log transformed)
Device Fe
1
2
3
4
Filter
1
2
Device Filter
1 1
1 2
2 1
2 2
3 1
3 2
4 1
4 2
-1.2813
-1.5995
-1.2224
0.0550
Fe
0.2110
-2.2351
Fe
-0.5061.
-2.0565
-0.9954
-2.2037
-0.2283
-2.2166
2.5737
-2.4637
Analysis of Variance Table
Source
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
df
6
3
1
3
42
56
111
SS
436.533
44.817
167.531
64.813
205.953
2.248
921.895
MS F P
72.755
14.939 3.05 .039
167.531 34.16 < .001
21 .604 4.41 .0087
4.904
0.040
74
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#2 Mn 7 Wells, 4 Devices, 2 Filtration Methods
Table of Means (log transformed)
Device Mn
1
2
3
4
Filter
1
2
Device
1
1
2
2
3
3
4
4
Filter
1
2
1
2
1
2
1
2
-1 .7806
-1 .8224
-1.6618
-1.0300
Mn
-1.3114
-1.8360
Mn
-1.6622
-1.8990
-1.7616
-1.8833
-1 .4429
-1 .8807
-0.3788
-1.6811
Analysis of Variance Table
Source
df
SS
MS
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
6
3
1
3
42
56
111
351.272
11.427
7.707
6.002
30.122
0.773
407.303
58.545
3.809
7.707
2.001
0.717
0.014
5.31
10.75
2.79
0.0034
0.0021
0.052
#2 As 7 Wells, 4 Devices, 2 Filtration Methods
Table of Means (log transformed)
Device As
1
2
3
4
Filter
-4.7044
-4.7386
-4.6541
-4.4882
As
1 -4.5570
2 -4.7357
Device Filter As
1
1
2
2
3
3
4
4
1
2
1
2
1
2
1
2
-4.6993
-4.7096
-4.7200
-4.7573
-4.5589
-4.7492
-4.2497
-4.7267
Analysis of Variance Table
Source
df
SS
MS
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
6
3
1
3
42
56
111
314.711
1.035
0.894
0.962
5.448
0.126
323.176
52.452
0.345
0.894
0.321
0.130
0.002
2.66
6.89
2.47
0.060
0.012
0.075
75
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#5 Fe 7 Wells, Devices 1-3 Only, 2 Filtration Methods
Table of Means (log transformed)
Device Fe
1
2
3
Filter
1
2
-1.2813
-1.5995
-1.2224
Fe
-0.5766
-2.1589
Analysis of Variance Table
SS
MS
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
6
2
1
2
30
42
83
390.048
2.305
52.576
2.140
128.530
2.066
577.666
65.008
1.152
52.576
1.070
4.284
0.049
0.27
12.27
0.25
0.77
0.0015
0.78
#5 Mn 7 Wells, Devices 1-3 Only, 2 Filtration Methods
Table of Means (log transformed)
Device Mn
1 -1.7806
2 -1.8224
3 -1.6618
Filter Mn
1 -1.6222
2 -1.8877
Analysis of Variance Table
Source df
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
6
2
1
2
30
42
83
SS
309.0407
0.3890
1.4796
0.3581
10.6942
0.4378
322.3995
MS
51.5068
0.1945
1.4786
0.1791
0.3565
0.0104
F P
0.55 0.58
4.15 0.0505
0.50 0.61
#5 As 7 Wells, Devices 1-3 Only, 2 Filtration Methods
Table of Means (log transformed)
Device As
1
2
3
Filter
1
2
-4.7044
-4.4386
-4.6541
As
-4.6594
-4.7387
Analysis of Variance Table
SS
MS_
frf-X^L*^-
Wells
Devices
Filter
DxF
Exp. Error
Samp. Error
Total
6
2
1
2
30
42
83
240.3703
0.1013
0.1321
0.1320
0.7428
0.0729
242.5513
40.0617
0.0507
0.1321
0.0660
0.0581
0.0017
0.87
2.27
1.14
0.43
0.14
0.33
76
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#9 Major Cations
Table of Means
Well
1
3
4
5
6
7
8
Device N
1
2
3
4
Filter
1
2
Device Filter
1 1
1 2
2 1
2 2
3 1
3 2
4 1
4 2
7 Wells,
Na
78.6
136.7
8972.5
8471 .2
141.8
12.9
35.0
a
2535.1
2561.4
2565.8
2536.9
Na
2552.0
2547.7
Na
2538.1
2532.2
2565.2
2557.6
2570.1
2561 .5
2534.4
2539.4
4 Devices, 2 Filtration
K
1.65
6.17
125.57
112.93
8.68
1.52
32.28
K
40.449
41.660
41.915
41.003
K
41.461
41.053
K
40.469
40.429
41.754
41.566
42.171
41.659
41.449
40.557
Methods
Ca
16.67
164.85
897.50
861.50
110.74
15.28
99.35
Ca
295.75
305.03
307.39
329.48
Ca
318.84
299.99
Ca
296.32
295.17
305.17
304.89
311.63
303.16
362.24
296.73
Mg
8.05
285.00
435.75
399.63
97.37
11.59
58.12
Mg
182.12
184.88
182.84
190.46
Mg
187.62
182.53
Mg
181.85
182.39
185.20
184.56
184.80
180.87
198.63
182.30
Analysis of Variance for Na
Source
Well
Devices
Filter
DXF
Error
Total
Analysis of Variance for K
Source
Well
Devices
Filter
DXF
Error
Total
Analysis of Variance for Ca
Source
Well
Devices
Filter
DXF
Error
Total
Analysis of Variance for Mg
Source
Well
Devices
Filter
DXF
Error
Total
df
6
3
1
3
42
55
df
6
3
1
3
42
- 55
df
6
3
1
3
42
55
df
0.0026
3
1
3
42
55
SS
854425088
10803
259
418
40887
854477504
SS
142132
18
2
1
94
142248
SS
7418818
8581
4977
10303
40757
7483436
SS
1632763
600
363
627
1381
1635734
MS
142404192
3601
259
139
974
MS
23689
6
2
0
2
MS
1236470
2860
4977
3434
970
MS
272127
200
363
209
33
F
3.70
0.27
0.14
F
2.75
1.05
0.22
F
2.95
5.13
3.54
F
6.08
11.03
6.36
P
1.019
0.608
0.933
P
0.055
0.312
0.879
P
0.044
0.029
0.023
P
0.002
0.002
0.001
77
-------
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
#9 Major Anlons 7 Wells, 4 Devices,*2 Filtration Methods -.
Table of Means
Well • HC03 '
1 239.1 ' '
3 - 1831.9 ,.
4 830.9
5 795.1
6 1014.5
7 , ; 78.2
8 701.5 ;
Device » ' HC03
1 . " 767.81
2 761.63
3 764.44 "
4 843.97
Filter HC03
1 806.00*
2 762.92
Device Filter
1 1 ,
1 2
2 1 , .
2 2
3 1 ,
3 2
4 1
4 2
Analysis of Variance for HC03
Source df
Well 6
Devices 3
Filter 1
DXF 3
Error 42
Total 55
Analysis of Variance for C1
Source df
Well 6
Devices 3
Filter 1
DXF '3
Error 42
Total 55
Analysis of Variance for S04
Source df >'
Well 6
Devices 3 -
Filter ' 1'
DXF 3
Error 42
Total 55
Analysis of Variance for NO3
Source df
Well 6
Devices 3-
Filter 1
DXF 3
Error 42
Total 55
C1
8 • ' :
78 '
12555
11551 ; .
79 '
8
16
C1
3476.3
3473.0 .. " ,
3483.0
3450.2
C1 ' i
3463.8
3477.4
HC03
770.27 -
765.34
762.89
760.37 "
772.41
756.47 '
918.43
769.51
SS
: 15642300
66372 '
25976
52635 '
645367
16432651
SS
1654049536
8523
2609
5246
62806
1654128768 • .
SS
300164384
595
12 "
284
13498 • '
300178752 ' - "
ss : '
246 - •- :
3 -
0
0 •
17 '
266 ; '
S04 '
1.8
285.4
5110.0 ' ;
5305.0
45.7
25.1
55.4
S04
1544.2
1541.3 '
1546.2
1550.2
S04 .
1545.0
1546.0
C1
• 3478.0 :
3474.6
3470.9
3475.1
3479.3
3486.6
3426.9
3473.4
MS
2607050
22124
25976
17545
15366
MS
275674912
2841
2609 ;
1749
1495
MS •
50027396
198
12
95 :
321 .:
MS
41
1
0
0
0
N03 •
0.0400
0.1255
0.0200
0.0200
0.0275
0.0688
6.0388
N03
1.2157
0.7529
1.0207
0.6321
N03
0.91679
0.89393
S04
1544.4
1544.0
1544.1
' 1538.4
1543.0
1549.5
1548.6
1551.9
F
1.44
1.69
1.14
F
1.90
1.74
1.17 .
F
0.62
0.04
0.29
F
2.42
0.02 '
0.06
N03
1.2514
, 1.1800
0.7686
0.7371
1.0614
0.9800
0.5857
0.6786
P
0.245
0.201
0.343
P
0.144
0.194
•'• 0.333
1
P
0.608
' 0.847
0.829
P
0.079
0.893
- 0.982
.=',, ^
78
-------
#9 Other Parameters
Table of Means
TABLE D-1. RESULTS OF STATISTICAL ANALYSIS CONT.
7 Wells, 4 Devices, 2 Filtration Methods
Well EC
1 419
3 2855
4 44000
5 41450
6 1704
7 200
8 1249
Device ED
1 13106
2 13143
3 13123
4 13130
Filter ED
1 13125
2 13126
Device Filter
1 1
1 2
2 1
2 2
3 1
3 2
4 1
4 2
Analysis of Variance for EC
Source df
Well 6
Devices 3
Filter 1
DXF 3
Error 42
Total 55
Analysis of Variance for pH
Source df
Well 6
Devices 3
Filter 1
DXF 3
Error 42
Total 55
Analysis of Variance for TDS
Source df
Well 6
Devices 3
Filter 1
DXF 3
Error 42
Total 55
Analysis of Variance for Si
Source df
Well 6
Devices 3
Filter 1
DXF 3
Error 42
Total 55
pH
17.284
15.391
14.585
14.711
16.265
13.310
13.711
pH
14.988
15.010
14.884
15.265
pH
14.997
15.076
HC03
13098
13113
13144
13141
13125
13121
13131
13128
SS
1.9688E+10
9922
23
904
183592
1.9688E+10
SS
94
1
0
0
4
99
SS
9159377920
23582
232
8871
197297
9159607296
SS
273105
13
13
3
69
273203
TDS
262
1716
29862
28229
938
186
754
TDS
8852.1
8843.4
8822.9
8880.0
TDS
8851.6
8847.5
C1
14.937
15.039
15:006
15.004
14.811
14.957
15.234
15.296
MS
3281361408
3307
23
301
4371
MS
16
0
0
0
0
MS
1526562944
7861
232
2957
4698
MS
45517
4
13
1
2
Si
23.44
53.95
203.00
195.50
52.38
34.64
82.21
Si
92.664
91.879
92.564
91.529
Si
91.671
92.646
S04
8864.1
8840.0
8834.4
8852.3
8811.0
8834.7
8896.9
8863.1
F
0.76
0.01
0.07
F
3.87
0.93
0.13
F
1.67
0.05
0.63
F
2.54
8.09
0.59
TOG
6.700
9.000
14.450
11.925
5.113
27.850
46.213
TOG
16.450
16.314
16.229
20.293
TOG
17.600
17.043
N03
15.557
17.343
16.071
16.557
16.071
16.386
22.700
17.886
P
0.525
0.943
0.976
P
0.016
0.340
0.944
P
0.187
0.825
0.600
P
0.069
0.007
0.627
*U.S. GOVERNMENT PRINTING OFFICE: 1994-550-001/00190 Region 1.
79
-------
------- |