EDA's Arsenic Removal Technology by Dr. Samaresh Mohanta
Abstract EDA's Arsenic removal technology by Dr. Samaresh Mohanta, Berkeley, California
Abstracrt
March 2000
EDA's
Arsenic Removal Technology by Dr. Samaresh Mohanta
Abstract
EDA's Arsenic removal technology by Dr. Samaresh Mohanta, Berkeley, California
This
study showed that under the laboratory conditions the EDA’s proprietory
ceramic media absorbs Arsenic extremely well and preferentially in the
presence of various tramp ions as would be present in Bengal Basin ground
water. There is no foreseen scientific and/or manufacturing hurdle for
this technology. A study was initiated to determine the functional applicability
of a Ceramic Media to remove Arsenic from Bengal Basin Groundwater. The
primary objectives of this study were: Determine the effect of background
ions present in the water, namely iron, phosphate and sulphate, on the
Arsenic removal capacity of the media Determine the loading capacity of
the media in packed bed configuration Build and test a prototype for tubewell
application for Bengal Basin Compare EDA media with activated alumina in
this application A typical tubewell in Bangladesh Background shows a typical
village pond and a residence This report describes the laboratory work
carried out and the results obtained thereby. The principal conclusions
of this work are: The interference of background ions: The interference’s
of the following background ions were tested: iron, bicarbonate, hydronium,
phosphate and sulfate. The concentration levels chosen were twice that
of the maximum present in 13 of the Bangladesh water samples. None of these
ions showed any influence on the effectiveness of removal of Arsenic by
EDA-media. The loading capacity (defined by wt. % of Arsenic that can be
absorbed) of the media in the three surrogate water was found to be 1.3%-1.5%
in a column test. Activated Alumina was not at all effective in removing
Arsenic (III) or it needed longer contact time and bed depth in the packed
bed column test. A unit showing the main working principles of a filter
has been produced. This is delivering about 8 lpm at a head of 13.5 in.
water. The removal of Arsenic is also demonstrated in this unit. The output
of the device was analyzed in an outside lab and was found to contain non-detect
level of Arsenic. Detection limit was 5 ppb. Unit size is 2ft. high and
1 ft-square. The test on this unit is continuing. A paper study on manufacturing
cost showed that the commerciallization of the device is very feasible.
A patent has been filed. Based on these conclusions, there is no foreseeable
technical or engineering impediment to the use of this process for arsenic
removal applications in the Bengal Basin. Moreover, given that the paper
study confirmed our working assumptions on manufactured cost, commercialization
of this technology is strongly recommended.
Abstract:
March 2000
Chronic
and acute arsenic poisoning through drinking water
Dr.
Ratna Chatterjee, UCL London
Chronic
and acute arsenic poisoning through drinking water: impact on male and
female reproductive health and future children.
Abstract:
March 2000
Chemotherapy
as a model to understand arsenic toxicity in human reproductive
Dr.
Ratna Chatterjee
Chemotherapy as a model to understand arsenic toxicity in human reproductive and feto-maternal health
Abstract/Paper
is under preparation
Abstract:
March 2000
The
Problem of the Century
Dr
Ratna Chatterjee
Department
Of Reproductive Medicine
UCL,
London
Arsenic
and reproductive and sexual health issues: what do we know about it? The
Problem for the century
Abstract:
March 2000
River
Water Quality: The Source of Arsenic Free Drinking Water in Bangladesh
Islam,
Md. Riajul.1, Islam, Muhammad, Anwarul.2 Rojstaczer, Stuart*1 Islam, Md.
Riajul
1.
Islam, Muhammad Anwarul
2.
Rojstaczer, Stuart
1.
Division of Earth and Ocean Sciences
Duke
University
Box
90227
Durham,
NC 27708-0230, U.S.A.
2.
Department of Geology and Mining
University
of Rajshahi
Rajshahi
6205, Bangladesh
Recently, attention is focused on extensive arsenic contamination of groundwater in Bangladesh but surface water bodies throughout Bangladesh are subject to potential water quality hazards associated with metals due to both intense chemical weathering and anthropogenic activities. The issue of surface water quality is very crucial since the Government has suggested that inhabitants use surface water in areas where groundwater has been severely affected by arsenic. The main objectives of this study are to evaluate dissolved inorganic materials including As in major rivers and to identify the environmental factors that control their hydrogeochemistry. The geography of Bangladesh is dominated by the three great rivers Ganges (Padma), Brahmaputra-Jamuna and Meghna and a dense network of branches, distributaries and connecting channels which is estimated to be about 24,000 km (Rashid, 1977). The stratigraphic succession of Bangladesh is composed of Tertiary sediments, occasionally covered by Quaternary overburden. Sandstone, siltstone, shale and claystones are the main rock types existing all over the country. In the present study, 16 water samples were collected from the three major rivers - Padma (Ganges), Brahmaputra-Jamuna and Meghna and from other two branches - Mahananda and Buriganga during the flood season and analysed by IC and ICP-MS.
Concentrations
of major nutrients, NO3-N, PO4-P and heavy metals are much lower than the
WHO-(1996) guidelines. Arsenic concentration ranges in between 4.6 - 4.9
ppb in water from all the major rivers in Bangladesh. The more or less
constant concentration is somewhat surprising, as flow volumes differ dramatically
and dilution of dissolved constituents should increase downstream. The
lack of dilution suggests that concentration of As is at saturation for
the geochemical conditions present in the river water. If this is so, concentration
during the dry season would likely be similar to that observed in this
samples and river water, aside from pathogens, should be safe for drinking
year round. We will be testing river water during the dry season in the
spring of 2000. The non-filtered and acidified water samples show higher
concentrations of metals which indicate that the metals released due to
intense chemical weathering have co-precipitated and/or adsorbed on to
the colloidal clay fractions and organometalic complexes which were filtered
out by 0.45 micrometer net. This difference also suggests that dissolved
concentration of metals are at saturation.
Poster:
March 2000
Mahfuzar
Rahman, MD, Ph.DA: Arsenic in drinking water and skin lesions
POSTER
Arsenic in drinking water and skin lesions
Mahfuzar
Rahman, MD, Ph.D
Division
of Occupational and Environmental Medicine
Faculty
of Health Sciences
Linköping
University
S-581
85 Linköping, Sweden
Chronic
arsenic intoxication from drinking water as contaminated from geological
sources has caused a devastating health crisis, in Bangladesh. A similar
situation can be observed not only in Bangladesh, but also in some other
parts of the world. Skin lesions are the hallmark of high exposure to arsenic
and pose a public health problem in Bangladesh. A cross-sectional study
was performed by door to door visits, interviewing families with arsenic
exposure and skin lesions. We interviewed 1146 individuals irrespective
of age and sex who had at least one sign of arsenical skin lesions, i.e.,
keratosis, leocomelanosis or melanosis. The mean arsenic level in their
drinking water was 144.4 mg/L (non-detectable limit to 4727 mg/L) and the
sex ratio was 1.5: 1 (men and women). Clinical examinations of the 1146
subjects who all had some typical skin lesion, in term of melanosis (99/100)
and keratosis (66.8/100). This study shows a higher percentage of arsenic
skin lesions in men than women. These skin lesions are an alarming sign
of high arsenic exposure. There is an urgent need for a technical solution
to provide good quality drinking water and requests for urgent remedies,
especially regarding future skin cancer.
Abstract:
March 2000
Mahfuzar
Rahman, MD, Ph.D Environmental exposure to arsenic and nonmalignant health
effects.
ORAL
Environmental
exposure to arsenic and nonmalignant health effects.
DivAbstract:
March 2000
W.G.Burgess1
and K.M.U.Ahmed:The variability of arsenic in groundwater of southern Bangladesh
- keys and constraints for sustainable developmentW.G.Burgess1 and K.M.U.Ahmed2
The
variability of arsenic in groundwater of southern Bangladesh - keys and
constraints for sustainable development
W.G.Burgess1
and K.M.U.Ahmed2
London-Dhaka
Arsenic in Groundwater Project
1
Dept. Geological Sciences,
University
College London,
Gower
St.,
London
WC1E 6BT, UK william.burgess@ucl.ac.uk
2
Dept. Geology,
University
of Dhaka,
Dhaka,
Bangladesh
kmahmed@du.bangla.net
Arsenic
is a widespread pollutant of groundwater in southern Bangladesh. Response
to the arsenic threat must take account of the nature of its occurrence,
and the scale and patterns of variability that are evident spatially, with
depth and in time. Hydrochemical relationships have demonstrated that chemically
reducing conditions favour the release of arsenic from sedimentary iron
oxyhydroxides which are the source of the pollution. How is the source
distributed within the alluvial sediments? What is the manner of its movement
to tube-wells? Detailed observations of the spatial variability of arsenic
in pumped groundwater, and depth profiles of arsenic in groundwater, pore-water
and aquifer sediments, have been used to develop a conceptual model of
arsenic movement to tube-wells. Preliminary numerical models reproduce
the variability observed spatially and with depth. The results constrain
answers to the questions on how successfully tubewell placement, design
and pumping regime can be managed to minimise the arsenic concentrations
in pumped groundwater, how arsenic concentrations may change with time,
and the effectiveness of monitoring. These are the key issues for sustainable
development of groundwater in southern Bangladeshon of Occupational and
Environmental Medicine,
Faculty
of Health Sciences,
Linköping
University,
S-581
85 Linköping, Sweden
Abstract
A
series of studies concerning environmental exposure to arsenic and some
novel chronic health effects of this element, namely diabetes mellitus,
glucosuria and hypertension. Substantial prevalence of the well-known skin
manifestations of arsenic ingestion was also found to occur as a result
of environmental exposure through drinking water. A cross-sectional study
was carried out in Bangladesh, where a fairly large part of the population
is exposed to inorganic arsenic in drinking water. The prevalence of diabetes
mellitus among subjects with keratosis (n = 163) was compared with unexposed
subjects (n = 854); keratosis was considered to be a definite sign of exposure.
A dose-response relationship was found between categories of time-weighted
arsenic exposure (mg/L in drinking water) and the prevalence of diabetes
mellitus (p < 0.001), and the crude overall prevalence ratio amounted
to 4.4. Despite the lack of detailed individual exposure data and information
on potential confounders other than age, sex, and body mass index (BMI),
the association seems strong enough to support a causal relationship, because
the adjusted overall prevalence ratio was 5.9 (95% confidence interval
2.9-11.6). One study from Bangladesh indicated a significantly increased
risk of hypertension in connection with exposure to inorganic arsenic in
drinking water (1481 exposed and 114 unexposed subjects). The overall crude
prevalence ratio of hypertension amounted to 1.7, and the adjusted (for
age, sex, and BMI) ratio was 1.9 (95% confidence interval 1.0-3.6). A significant
trend in risk (p << 0.001) was observed between an approximate time-weighted
mean exposure to arsenic, considered in milligrams per liter or milligram-years
per liter, which strengthens the possibility of a causal association. One
of the other studies included 1481 exposed individuals, 430 exhibiting
keratosis showed a somewhat higher prevalence rate of skin lesions in males
(31%) than females (26%) due to chronic arsenic toxicity. The crude overall
prevalence was 29% in the studied villages, and there was a distinct dose-response
relationship between arsenic concentrations in drinking water and skin
lesions (p < 0.01). A clear dose-response relationship was also observed
between arsenic exposure and glucosuria for subjects both with and without
skin lesions (p < 0.01). The possibility of using the skin lesions for
initial screening for glucosuria was considered. However, the appearance
of dermatological signs of chronic arsenic toxicity proved to be a poor
marker in this respect, because glucosuria also occurred in the absence
of skin lesions. Overall, the results of these studies provide evidence
that arsenic exposure may play a role in the development of diabetes mellitus,
however, the mechanism underlying the ability of inorganic arsenic to induce
these disorders is still unclear. Various sources of exposure should be
taken into consideration in further confute or refute the indicated effects
of arsenic. POSTER Arsenic in drinking water and skin lesions Mahfuzar
Rahman, MD, Ph.D Division of Occupational and Environmental Medicine, Faculty
of Health Sciences, Linköping University, S-581 85 Linköping,
Sweden Abstract Chronic arsenic intoxication from drinking water as contaminated
from geological sources has caused a devastating health crisis, in Bangladesh.
A similar situation can be observed not only in Bangladesh, but also in
some other parts of the world. Skin lesions are the hallmark of high exposure
to arsenic and pose a public health problem in Bangladesh. A cross-sectional
study was performed by door to door visits, interviewing families with
arsenic exposure and skin lesions. We interviewed 1146 individuals irrespective
of age and sex who had at least one sign of arsenical skin lesions, i.e.,
keratosis, leocomelanosis or melanosis. The mean arsenic level in their
drinking water was 144.4 mg/L (non-detectable limit to 4727 mg/L) and the
sex ratio was 1.5: 1 (men and women). Clinical examinations of the 1146
subjects who all had some typical skin lesion, in term of melanosis (99/100)
and keratosis (66.8/100). This study shows a higher percentage of arsenic
skin lesions in men than women. These skin lesions are an alarming sign
of high arsenic exposure. There is an urgent need for a technical solution
to provide good quality drinking water and requests for urgent remedies,
especially regarding future skin cancer.
Abstract:
March 2000
Dr.
Barin Chatterjee: A GEOTECHNICAL APPROACH TO THE PROBLEM IN BENGAL BASIN
A
GEOTECHNICAL APPROACH TO THE TOXICITY PROBLEM IN THE BENGAL BASIN
By Barin Chatterjee
Before
arriving to any possible solution to the problem, it should be known why
this contamination problem is concentrated in the Bengal basin, what is
the source and does it occur in any correlatable way ? The present endeavor
is to throw some light on a possible correlation from the geotechnical
observation. In course of working in the lower Damodar basin for its endemic
flood problem, it could be observed that there are at least two mappable
arcuate micro-relief bands, almost swerving like bird’s track. These bands
could be picked up towards the upper part of the delta by means of exaggerating
the vertical scale with respect to the horizontal scale while contouring
at 1 m interval on a 1:250,000 scale. By subsequent geotechnical analysis
with undisturbed soil samples collected in a grid pattern, it could be
established that the micro-relief zones represent higher values of pre-consolidation
pressure to the tune of 3 kg/sq. cm in comparison with the background value
of 0.9 kg/sq. cm. The topography being almost flat, this led to infer zones
of initial depth of burial followed by subsequent squeezing up of sediments
( Chatterjee,B : Geotechnical aspect of endemic flood problem in the lower
Damodar basin, West Bengal, India ; VI International Congress of the Association
of Engineering Geology, IAEG,1990 published by Balkamara, Rotterdam, ISBN
90 61911303, pp. 2785-2790). Such deposition and uprising, basically a
long drawn and contemporaneous process, are manifested as drape over reefs
along the palaeo-shore line ( Morgan, J.P, 1968 ; Mud lump : Diapiric structures
in Mississipi delta sediments; Mem.8, Am.Assocn.Pet.Geologists, Tulsa,
Oklahama, pp. 145-161 ). Due to rapid deposition of denser mouth bar sand
over the less denser marine clay in the palaeo-shore line areas, the differentially
weighted clay subsequently gets squeezed up in the form of Diapiric type
of intrusion. In the upper part of the delta, it takes the shape of micro-relief
on the surface, while down the slope, in the subsurface, it causes barriers
to the path of subsurface infiltration, resulting in localized pockets
of toxic ground water aquifers. Due to non-existence of any surface expression,
it is very difficult to identify such zones in the down-delta region. As
ground water is now being extensively used for drinking and irrigation
purposes, it calls for identification of such zones by a systematic extensive
geotechnical investigation , keeping in view the possible palaeo-shore
lines.
Abstract
Paper March 2000
NATURALLY
OCCURRING ARSENIC IN SANDSTONE AQUIFER WATER SUPPLY WELLS OF NORTHEASTERN
WISCONSIN Rebecca S. Burkel and Richard C. Stoll
NATURALLY
OCCURRING ARSENIC IN SANDSTONE AQUIFER WATER SUPPLY WELLS OF NORTHEASTERN
WISCONSIN
NATURALLY
OCCURRING ARSENIC IN SANDSTONE AQUIFER WATER SUPPLY WELLS OF NORTHEASTERN
WISCONSIN Rebecca S. Burkel and Richard C. Stoll Rebecca S. Burkel is the
District Environmental Coordinator for the Wisconsin Department of Transportation,
Division of Highways, 944 Vanderperren Way, Green Bay, WI 54304. Rebecca
is a graduate of the University of Wisconsin Green Bay with a BS (chemistry)
and a MS (environmental science) degree. Richard C. Stoll is the District
Hydrogeologist for the Wisconsin Department of Natural Resources, Lake
Michigan District office in Green Bay, Wisconsin. Rick is a certified Groundwater
Professional (CGWP #457) with the National Groundwater Association, a Certified
Professional Geologist (CPG #9157) with the American Institute of Professional
Geologists, and a State of Wisconsin Professional Geologist (#38). ABSTRACT
NATURALLY OCCURRING ARSENIC IN SANDSTONE AQUIFER WATER SUPPLY WELLS OF
NORTHEASTERN WISCONSIN
Rebecca S. Burkel and Richard C. Stoll
The
EPA maximum contaminant level (MCL) of 50 Fg/L for arsenic was exceeded
in 86 of 2125 water supply wells sampled over a broad geographic range
in parts of Brown, Outagamie and Winnebago Counties, Wisconsin. The hydrologic
and geochemical properties of the area were examined and the source of
arsenic determined to be natural. Groundwater collected from two geologic
formations, the St. Peter Sandstone and the overlying Platteville/Galena
Dolomite, were found to be the principal sources of the elevated arsenic
concentrations. These two formations supply a large portion of eastern
Wisconsin private wells their drinking water. Three wells were found within
Outagamie County to have an unusually low pH. Results suggest that the
cause of the low pH in these wells is of natural origin induced by the
oxidation of iron sulfide minerals. In this reaction iron sulfide minerals
are oxidized forming sulfuric acid causing a low pH and a high concentration
of various metals to leach from native rock formations into the water supply.
Based on the data gathered from this study an arsenic advisory area for
both Outagamie and Winnebago Counties was designated. Guidelines were developed
for well drillers and owners constructing new wells within the advisory
area to reduce the likelihood of arsenic presence in the water supply.
Fifteen wells containing arsenic exceeding the MCL were successfully reconstructed
or new wells were constructed based on the guidelines developed. These
constructions substantially reduced arsenic levels in the well water supplies.
INTRODUCTION
Arsenic (As) contamination in water supplies in Winnebago County, Wisconsin,
was first identified in two different locations in 1987. Following this
the Wisconsin Department of Natural Resources (WDNR) initiated a study
in 1991 to investigate the occurrence of As in private wells in Outagamie
and Winnebago Counties. The study was expanded to include parts of Brown,
Marinette, Oconto and Shawano Counties, Wisconsin, following the results
of investigation in Outagamie and Winnebago Counties. The objective of
the study was to determine the source and lateral and vertical distribution
of As occurring in groundwater and geologic formations. The study results
were used to develop special well casing and well construction criteria
for new wells in affected areas Arsenic enters the environment through
natural processes or via human activity (Eisler, 1988). Natural processes
that influence the presence of As are volcanic emissions and weathering
of arsenic-containing rocks with minerals such as arsenopyrite (FeAsS)
(Eisler, 1988). Currently, the drinking water standard for total As is
50 Fg/L, as established by the Safe Drinking Water Act (SDWA) in 1986.
Geology of Brown, Marinette, Oconto, Outagamie, Shawano and Winnebago Counties,
The Study Area There are five principal geologic units beneath the glacial
overburden in Brown, Marinette, Oconto, Outagamie, Shawano and Winnebago
Counties, Wisconsin. The Platteville/Galena Dolomites overlie the Saint
Peter Sandstone. Beneath the sandstone are the Prairie du Chien Dolomites
and the Cambrian Sandstones. The basement consists of crystalline Precambrian
rocks. The overlying Platteville/Galena Dolomites are composed of sandy-gray
to bluish-gray dolomite with fine to medium grained sandstone near the
base (Olcott, 1966). The formation generally yields little water to wells
(LeRoux, 1956). However, in certain locations sufficient quantities of
water exist in joints, bedding planes and fractures to furnish some private
wells. The St. Peter Sandstone is a productive water-yielding unit. It
consists of fine to coarse-grained dolomitic sandstone (LeRoux, 1957).
The St. Peter Sandstone rests on the Prairie du Chien Group filling in
low areas but it is absent on the Prairie du Chien highs (Olcott, 1966).
Water yields from the St. Peter Sandstone are limited by the presence of
shale and by the limited thickness of the formation. The Prairie du Chien
is a relatively unproductive water yielding unit. The Prairie du Chien
Group consists of dolomite with thin layers of white sandstone and green
shale (Olcott, 1966). The upper surface of the Prairie du Chien is highly
irregular (Olcott, 1966). A limited amount of water is found in fractures,
joints, and bedding planes (Olcott, 1966). The Cambrian sequence rests
on the irregular and highly eroded surface of the Precambrian rock (Olcott,
1966). The Cambrian System is made of fine to coarse grained sandstone.
These sandstones are a major source of groundwater especially for municipal
wells. The Precambrian crystalline rocks are composed primarily of granite
and except for fractures are generally impermeable. METHODOLOGY Well Water
Sample Collection and Analysis Initial water samples were collected by
the WDNR following the guidelines set in the WDNR Groundwater Sampling
Procedures Field Manual and the WDNR Groundwater Sampling Procedures Guidelines
both developed by Lindorf, Feld, and Connelly (1987). These were later
supplemented by a private-well-owner sampling program. All samples were
collected from a cold water tap situated prior to any in-line treatment
device (e.g. a water softener or filter). Samples were collected after
flushing the water supply tap for three to five minutes or a couple of
minutes after the pump started running. This procedure identified in WDNR
Groundwater Sampling Procedures Guidelines developed by Lindorf, Feld,
and Connelly (1987) was used to ensure the water sample represented in
situ groundwater and not water standing in the pipes. All samples were
collected, preserved with 2.5 ml/ 35% (8N) HNO3,to a pH of less than 2
and labeled in appropriate 250 mL sample containers supplied by the Wisconsin
State Laboratory of Hygiene (WSLH). Samples were analyzed for As by high
temperature graphite furnace atomic absorption method, 3113B in the Standard
Methods for the Examination of Water and Wastewater, 17th Edition, (1989).
The methods had either a 2 Fg/L or a 3 Fg/L lower detection limit for As.
All sampled well locations were plotted on 7.5 minute USGS topographic
maps or keyed with a global position system (GPS) for digitizing into a
geographic information system (GIS). Multiple information layers including
land net, railroads, trunk highway network, local roads, hydrology, bedrock
geology, glacial geology, potentiometric surface and all known point and
non-point potential groundwater contamination sources were created for
all of northeast Wisconsin. This information was analyzed with respect
to As contamination utilizing PC ARC/Info and ARC View GIS software. Geophysical
Well Investigation Methods - Inflatable Packer Tests Packer tests were
used to isolate and sample specific subsurface water-bearing zones within
wells and thus to identify potential subsurface horizons with poor water
quality. Two wells, 1 and 4, were chosen for packer testing, (Figure 1).
Both wells were selected based on their significant depths and the elevated
As concentration of their water. See Figures 3 and 4 for well 1 and 4,
respectively, lithology and construction. Furthermore, the family consuming
water from Well 1 was informed by their physician that they exhibited symptoms
associated with chronic As poisoning. Wells 1 and 4 are located about 24
km from each other. To minimize costs, packer test intervals of 9.1 and
4.6 m (Wells 1 and 4 respectively) were selected to best define vertical
variations in As concentration throughout the entire well column with the
fewest number of packer tests. Each Packer interval was pumped for approximately
10 minutes at 37.8 L/min so that a representative water sample could be
obtained from each interval. The integrity of the packer assemblies were
monitored using water level indicators throughout the packer tests to verify
that the water sampled during the test was actually collected from within
the designed packer interval. Samples were collected only from those packer
tests with confirmed packer seal integrity or those that showed little
fluctuation of water level measurements during pumping. However, due to
the type of packer assembly used at Well 4, the integrity of the bottom
packer seal could not be monitored. The bottom packer may not have sealed
when sampling took place nearest the bottom of the well column. The bottom
6.1m of the well was filled in with sediment. Sediment and colloidal material
clogged the pump and was incorporated into the sample that was analyzed.
Well 1, at packer interval 10, the sample was collected after the pressure
tank. The owner required temporary reservice of water. The pressure tank
had a torn bladder. This may allow for outside contamination. Also, the
water level indicator showed a leaking top packer at interval 10. This
may have allowed for contaminated water from the upper portion of the well
to drain into the area being packed off. All other samples from Well 1
were collected directly from the packer assembly, prior to the pressure
tank. The water samples from packer test wells were collected according
to WDNR guidelines. Packer test water samples were collected from a brass
tap located directly above the packer assembly prior to disposal to a holding
tank. Water samples from each packer were tested for field pH and field
temperature. The water sample was then filtered using a 0.45 micron filter
prior to being sent to the WSLH for analysis of As, cadmium (Cd), copper
(Cu), manganese (Mn), and zinc (Zn). The parameters As, Cd, Cu, Mn and
Zn were selected based on the elevated levels found in the original well
samples . All samples were collected, preserved with 2.5 ml/ 35% (8N) HNO3,
to a pH less than 2 and labeled in appropriate 250 mL sample containers
supplied by the WSLH. Samples were analyzed by high temperature graphite
furnace atomic absorption methods: and inductively coupled plasma, following
procedures 3113B and 3120, from the Standard Methods for the Examination
of Water and Wastewater, 17th Edition (1989).. RESULTS and DISCUSSION Private
Well Arsenic Levels The distribution of water well As concentrations in
Brown, Marinette, Oconto and Shawano Counties is markedly different from
Outagamie and Winnebago Counties. Arsenic exceedances do not occur in Marinette
and Oconto Counties while limited exceedances occur in part of Brown County.
Outagamie and Winnebago Counties exhibit wide spread As exceedances over
a 8.0 km area. Winnebago County had 23 of 827 wells sampled that exceeded
the SDWA MCL for As while Outagamie County exhibited 45 of 1116 wells over
the MCL (Table 1). Low pHs ranging from 2.5 to 3.8 were documented for
3 wells in Outagamie County. Sampling in Brown County was conducted where
wells intercept the upper St. Peter Sandstone similar to those in Outagamie
and Winnebago Counties. The As problem appears to be a localized problem
in parts of Brown County where 18 of 76 wells sampled exceeded the SDWA
MCL. Seventeen of these wells exceeding the MCL are located within the
same square 2.6 km or township, range and section. This enclave of impacted
wells is surrounded by an area of municipal water service such that additional
nearby water sample collection locations are limited. However, one well
with As exceeding the MCL is located about 3.2 km away from the 17 other
As-impacted wells in Brown County. Each of these wells draws water from
the upper St. Peter sandstone. One well with both a field pH as low as
3.1 and As over the MCL is documented in Brown County. The sample information
for Shawano, Oconto and Marinette Counties indicate this more northern
group of counties is much less likely to exhibit As-related drinking water
problems. While the sampling conducted in these counties was also along
the St. Peter Sandstone subcrop and conducted similarly, there were no
SDWA MCL exceedances or pH problems. Wells with elevated levels of As were
found principally in areas where St. Peter Sandstone is present, based
on existing bedrock geology maps (Figure 2). This is where the St. Peter
Sandstone is the predominant aquifer supplying private wells. There are,
however, some areas where wells with elevated As concentrations lie west
of the mapped St. Peter Sandstone trend. This is unexpected if the St.
Peter Sandstone is the primary source of naturally occurring As in the
groundwater. The following are potential explanations for this: (1) the
sandstone lenses of the Prairie du Chien or the older underlying sandstone
of the Cambrian formation may also contribute As to well water rather than
only the St. Peter Sandstone formation; (2) some of the St. Peter Sandstone
subcrop contacts shown on the bedrock map for this area are inferred. The
subcrop was inferred because there was not enough data to map it accurately.
These inferred areas correspond to the areas where wells with elevated
As concentrations exist to the west of the mapped subcrop; (3) localized
westward shallow groundwater flow, impacted by As, may move As down gradient
away from the regional divide to where it would not normally be suspected.
The regional groundwater divide parallels much of the St. Peter Sandstone
subcrop expression. Well 1 - Water Quality Well 1 was first sampled in
1992, due to the owners complaint of an iron (Fe) problem. The well was
sampled for both Fe and As because the well was located near the St. Peter
Sandstone trend. Initial test results revealed a high Fe concentration
(87 mg/L) and an As concentration of 1200 Fg/L, the highest As concentration
recorded at that time in a private water supply well in Outagamie County.
Two months later a follow-up water sample was collected and analyzed for
As, Cd, chloride (Cl), chromium (Cr), conductivity, pH, alkalinity, barium
(Ba), calcium (Ca), Cu, Fe, Mn, sodium (Na), Zn, hardness, lead (Pb), nitrate+nitrite-nitrogen,
selenium (Se), silver (Ag), sulfate, total solids, field pH, and field
temperature. Both the field pH and lab pH indicated normal ranges for groundwater
from 6 to 8 (standard units). Chloride, Cr, conductivity, alkalinity, Ba,
Ca concentrations were were 4.0 mg/L, 8.2 Fg/L, 743 umhos/cm, 60 mg/L,
<40 Fg/L, 87 mg/L, respectively. Copper, Na, hardness, Pb, nitrate+nitrite-nitrogen
and total solids concentrations were 390 Fg/L, 2.4 mg/L, 350 mg/L, 12 Fg/L,
<1.00 mg/L, 746 mg/L. Both Se and Ag were not detected. Arsenic and
Cd concentrations were 720 Fg/L and 53 Fg/L, respectively. Arsenic and
Cd exceeded the SDWA maximum contaminant levels (MCL). Iron, Mn, Zn, and
sulfate concentrations were 80 mg/L, 490 Fg/L, 20000 Fg/L, and 330 mg/L,
respectively). Iron, Mn, Zn, and sulfate concentrations also exceeded the
SDWA secondary standards set by the EPA. Furthermore, these results indicated
the need to analyze the packer test samples not only for As but for Cd,
Fe, Mn, Zn and sulfate to further delineate the zones of poor water quality.
Well 1 - Water Quality Packer Test The results of water samples collected
from the packer test intervals at Well 1 (Figure 3) show a general decline
in As concentrations with depth within the well column. The upper portion
of the St. Peter Sandstone (34.7 m to 43.9 m) exhibits higher As concentrations
than those found in the base of the St. Peter Sandstone (43.9 m to 62.2
m) and in the Cambrian sandstones (71.3 m to 91.4 m) found below it. While
As concentrations within the borehole water increase in the Cambrian sandstones,
they were below the SDWA and much below the concentration in the upper
St. Peter Sandstone. Iron concentrations follow the same trend as As with
higher concentrations in the upper St. Peter Sandstone. The Fe concentration
in the well water declines in the lower portion of the St. Peter Sandstone
and then rises somewhat within the Cambrian sandstones. The Fe concentration
in the samples collected from the packer test remains above the SDWA welfare
standard (0.3 mg/L) throughout the entire well column. However, the Fe
concentrations are significantly reduced from that of the original well
which had a Fe concentration of 80 mg/L. As a result of this packer test
the potable well was deepened to 91.4 m by casing through the upper St.
Peter Sandstone to 45.7 m. The resulting As concentration after reconstruction
is 12 Fg/L, a substantial reduction from the original 1200 Fg/L As. Well
4 - Water Quality In addition to elevated As and Fe levels of 360 Fg/L
and 250 mg/L respectively, Well 4 also had an unusually low field pH of
3.8. The water sample was analyzed for Ca, Cl, conductivity, alkalinity,
hardness, Mg, Na, sulfate and total solids. Arsenic exceeded the SDWA maximum
contaminant levels (MCL). Iron and sulfate also exceeded the SDWA secondary
standards set by the EPA. A caliper log and gamma-ray log were run on Well
4 because the well had no subsurface information available below 36.6 m.
Well 4 - Water Quality Packer Test The results from the Well 4 packer test
indicated that the As concentrations throughout the entire well column
remained above the SDWA MCL of 50 Fg/L. The As concentration was highest
(>1,000 Fg/L) at the contact between the Prairie du Chien and the underlying
Cambrian sandstones (83.2 m to 87.8 m) (Figure 4). However, this concentration
likely resulted from the excessive colloidal sediment found in this zone.
The bottom 6.1 m of the well was filled in with sediment. Even though the
sample was filtered the WSLH noted abundant colloidal material which was
acidified and analyzed with the sample. Elsewhere throughout the well column
As concentrations were highest in the upper portions of the St. Peter Sandstone
(610 Fg/L). Arsenic concentrations declined to 51 Fg/L at the bottom of
the St. Peter Sandstone. The packer test results show the As concentration
increasing in the Prairie du Chien dolomite (55.8 m to 75.6 m). Iron concentrations
found in all packer test intervals exceeded the SDWA secondary standard
for Fe of 0.3 mg/L. Iron concentrations exceeding the standard typically
cause aesthetic nuisance problems such as odor and staining. In addition,
Cd levels exceeded the SDWA MCL of 10 Fg/L in all of the packer intervals
analyzed for Cd. Copper concentrations exceeded the SDWA MCL of 1300 Fg/L
in only the lowest interval, which contained large amounts of sediment
and colloidal material in the sample. Manganese concentrations remained
over the EPA SDWA secondary standard of 50 Fg/L in all of the packer intervals
analyzed. Zinc concentrations exceeded the SDWA secondary standard of 5000
Fg/L in all but one of the packer intervals tested. The possible presence
of drilling grease, thread goop and galvanized pipe are insufficient to
attribute these as the source for a problem of this magnitude and breadth.
An unusually low field pH of 3.8 was recorded for Well 4, which provides
an explanation as to why excessive amounts of metals were present in the
water supply. As pH decreases, the metal ion concentrations in the groundwater
tend to increase. Low pH values found in the well water may dissolve metallic
minerals in the aquifer, increasing the metal ion concentrations. A replacement
well has since been drilled on this property approximately 30.5 m away
from the original well and is cased into the Prairie du Chien dolomite.
The replacement well draws water from that dolomite and the underlying
Cambrian sandstone and has tested to be below 2.2 Fg/L As which is a substantial
reduction from the original 360 Fg/L As. The pH of this new well was not
determined. Two wells in Outagamie County, Wells 2 and 3, were found also
to have low pHs. Well 2, in Oneida Township, has a pH of 2.5 and well 3
in Greenville Township, had a pH of 3.0 . Similarly to Well 4, As, Fe,
Cd and Pb ( 5,900, 740,000, 210, 160 Fg/L, respectively) were excessively
elevated in well water obtained from Well 2. Also, through extensive investigation
using geophysics, boring, and monitoring well installation, the WDNR has
determined this pH is caused by pyrite oxidation (WDNR, 1994). A new well
was drilled to replace Well 2. This replacement well draws water from the
Platteville/Galena Dolomite, has an As value of 4 Fg/L compared to 5900
Fg/L and a pH of 8.22 compared to 2.5 . The original Well 3, which was
located within 1.6 km of Well 4 in Greenville Township, had a recorded
pH of 3.0 in 1967. Other nearby wells were sampled and did not exhibit
similar low pH values at that time. A new well was drilled to provide potable
water with a pH between 6.0 - 9.0. However a new well construction form
was not completed by the driller and further analytical information is
not available for the new well. The Well 4 packer test results showed that
the pH in the well varied throughout the well column. The pH of the water
from the upper portion of the St. Peter Sandstone was substantially lower
pH than that of lower portions of the well. It is likely that the acid
in the Well 4 water is derived from minerals within the St. Peter Sandstone
that have oxidized to form sulfuric acid. Lithological analyses have identified
arsenic-bearing pyrite and marcasite (FeS2) to occur within the Platteville/Galena
Dolomite and St. Peter Sandstone. Many wells in the study area with elevated
As concentrations may not exhibit low pH values due to the mixing and dilution
within the entire well column. Only a few natural reactions are known to
cause low pH in groundwater. Driscoll (1986) mentions that the acids from
mine waters are produced from the oxidation of iron pyrite or other metal
sulfide minerals to form sulfuric acid (H2SO4) . Oxidation of trace quantities
of pyrite and marcasite does contribute to the acidic nature of groundwater
(Bierens de Haan, 1991). Even the simple placement of a marble size field
sample of this sulfide material in 250 ml of distilled water reduced the
pH from 7.0 to 2.0 overnight. Oxidation of pyrite may also cause elevated
levels of As in groundwater because As tends to be associated with pyrite
and other sulfide minerals. The presence of marcasite and pyrite was documented
in Well 2, in abundant quantities from a rock core through the upper St.
Peter Sandstone (WDNR, 1994). Laboratory analysis of the sulfide rock material
obtained from these Well 2 location cores showed As was present at 150
mg/kg. It is plausible that pyrite and marcasite may also be found in larger
lenses or fractures of adjacent rock. Oxygen may be provided to the iron-rich
water flowing from the aquifer at the well/rock interface at the borehole
column, or by drilling and pumping (Smith and Tuovinen, 1985). Highly acidic
environments in aquifers are rare. However, wells with pH values as low
as 2.5 can occur as described in Wells 2, 3 and 4. New Well Construction
and Well Reconstruction Packer tests showed a marked increase in As concentration
in water from the upper portion of the St. Peter Sandstone. Water from
the lower St. Peter Sandstone and the Prairie du Chien units had much lower
As concentrations. To minimize As levels in groundwater supplies, the results
suggest that one should avoid extracting well water from the upper portion
of the St. Peter Sandstone aquifer (Figure 5) within the As advisory area
(shown in Figure 1). Arsenic concentrations in private well water have
been reduced to below the SDWA MCL by constructing new wells or reconstructing
existing wells. A total of 17 wells have been reconstructed or constructed
in a new location as part of this study. Two new wells were drilled shallower
to draw water only from the Platteville/Galena Dolomite, thus removing
the underlying St. Peter Sandstone as the source of water altogether. The
other wells were deepened and the casing was extended below the upper St.
Peter Sandstone preventing this zone from directly contributing water to
the well. Some wells utilized water exclusively from the Prairie du Chien
Dolomite. The other wells obtained water jointly from the Prairie du Chien
Dolomite and the underlying Cambrian Sandstone. Fifteen of 17 reconstructions
have successfully reduced As concentrations significantly below the SDWA
MCL. CONCLUSIONS AND RECOMMENDATIONS This study of As levels in private
wells in Brown, Marinette, Oconto, Outagamie, Shawano and Winnebago Counties
has shown that a significant number of wells are affected by high As levels.
The present SDWA MCL of 50 Fg/L As was exceeded on average of 4.0% of the
total wells tested in Brown, Marinette, Oconto, Outagamie, Shawano and
Winnebago Counties. This study has provided three lines of evidence that
the As found in the groundwater in Outagamie, Winnebago and Brown Counties
is of natural origin. First, the pattern of As contaminated wells is over
56.3 km long covering an area of approximately 951 km2. The aerial extent
of the contamination alone clearly suggests that the source of As is not
of an anthropogenic source. Also, historical storage of arsenic-based pesticides,
such as sodium arsenite used in Wisconsin in the 1930's and 1940's for
grasshopper control did not occur in these areas. Second, there are a number
of natural sources that can contribute to arsenic in groundwater. Lithological
analyses have identified the presence of pyrite and marcasite in the upper
St. Peter Sandstone in Well 2 and 4, previously mentioned. Pyrite as the
principal carrier of As in rocks, tends to be associated with mineral deposits
of sulfides and sulfo-salts (Eisler, 1988). Pyrite exists in layers within
the upper St. Peter Sandstone or lower Platteville/Galena Dolomite and
thus can contribute As to groundwater. Third, groundwater extracted during
packer tests show that the highest levels of As occur in the upper St.
Peter Sandstone. As previously mentioned, pyrite layers occur within the
upper St. Peters Sandstone and can contribute As to groundwater. Computer
mapping, utilizing a Geographic Information System (GIS) proved to be a
valuable tool to identify the approximate geographic regions in each county
where private wells are affected by naturally occurring As. Nearly all
sampled private wells which exceed the SDWA MCL for As are within a 8 km
zone east or west of the mapped St. Peter Sandstone subcrop extending SW
to NE through Outagamie and Winnebago Counties. Private wells exceeding
the SDWA MCL in Brown County are shown to be localized in a specific area.
No exceedances of the SDWA MCL for As was found in the wells sampled in
Oconto, Marinette and Shawano Counties. The aerial distribution of wells
within various As concentration ranges were used to develop an As advisory
area in Outagamie and Winnebago Counties. Since most private wells within
Outagamie and Winnebago Counties range between 30.5 m and 48.8 m deep,
wells within the advisory area commonly draw water from the St. Peter Sandstone.
Those private wells, located outside the 8.0 km boundary from the St. Peter
Sandstone, that are drilled deeper than normal may encounter the formation
and therefore may contain As, as has happened in Brown County 16 km downdip
eastward. Recommendations were identified through this project for private
well users to eliminate or greatly reduce their exposure to As in their
drinking water supply. Also, guidelines were developed for well drillers
in the area to eliminate and/or greatly reduce As exposure in water wells.
The guidelines are as follows: If As concentration exceeds SDWA MCL: (1)
purchase bottled water, (2) install a state approved water treatment device
such as a distillation or reverse osmosis unit, to remove As (3) reconstruct
the existing well, or drill a new well that avoids water from the upper
St. Peters Sandstone formation If drilling a new well in the advisory area:
(5) sample well water for As (6) construct the well to avoid water from
the upper St. Peter Sandstone formation To avoid high As, the St. Peter
Sandstone formation should not be penetrated, especially near the mapped
subcrop. If it is necessary to drill through the St. Peter Sandstone to
obtain a sufficient volume of water, t