USGS Study of Marcellus Shale Wastewater Radioactivity Levels

Radium Content of Oil- and Gas-Field Produced Waters
in the Northern Appalachian Basin (USA):
Summary and Discussion of Data

Scientific Investigations Report 2011–5135

U.S. Department of the Interior
U.S. Geological Survey


By E.L. Rowan, M.A. Engle, C.S. Kirby, and T.F. Kraemer

Scientific Investigations Report 2011–5135


KEN SALAZAR, Secretary

Marcia K. McNutt, Director

U.S. Geological Survey, Reston, Virginia: 2011

For more information on the USGS—the Federal source for science about the Earth, its natural and living
resources, natural hazards, and the environment, visit http://www.usgs.gov or call 1–888–ASK–USGS.
For an overview of USGS information products, including maps, imagery, and publications,
visit http://www.usgs.gov/pubprod
To order this and other USGS information products, visit http://store.usgs.gov

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the
U.S. Government.

Although this report is in the public domain, permission must be secured from the individual copyright owners to
reproduce any copyrighted materials contained within this report.

Suggested citation:
Rowan, E.L., Engle, M.A., Kirby, C.S., and Kraemer, T.F., 2011, Radium content of oil- and gas-field produced waters
in the northern Appalachian Basin (USA)—Summary and discussion of data: U.S. Geological Survey Scientific
Investigations Report 2011–5135, 31 p.
(Available online at http://pubs.usgs.gov/sir/2011/5135/)

iii

Contents
Abstract............................................................................................................................................................1
Introduction.....................................................................................................................................................1
Background.....................................................................................................................................................2
Data Sources and Analytical Methods.......................................................................................................5
New York State Department of Environmental Conservation Report (1999)................................5
New York State Department of Environmental Conservation, Draft Supplemental
Generic Environmental Impact Statement (2009)...............................................................6
Pennsylvania Department of Environmental Protection Report (1992).........................................6
Pennsylvania Department of Environmental Protection Reports
(Unpublished Data, 2009–2010)..............................................................................................7
Dresel and Rose (2010).........................................................................................................................7
This Study................................................................................................................................................7
Results ..............................................................................................................................................................8
Salinity and Radium...............................................................................................................................8
Gross Alpha and Beta Particle Emissions.........................................................................................9
Discussion........................................................................................................................................................9
Salinity and Dilution...............................................................................................................................9
Radium Activities in Context..............................................................................................................12
Summary........................................................................................................................................................15
Acknowledgments........................................................................................................................................15
References Cited..........................................................................................................................................15

Figures
1. Radioactive decay chains for U-238 and Th-232......................................................................3
2. Map showing locations of wells with data compiled for this study......................................4
3. Differences between measurements of duplicate and replicate analyses of Ra-226
and Ra-228 in produced water samples in relation to the mean activity of the sample
for data from Gilday and others (1999).......................................................................................6
4. Measured activities for total radium (Ra-226 + Ra-228) and Ra-226 for each
of the data sources used in the study........................................................................................8
5. Gross alpha and beta particle activities in relation to the activities of Ra-226
and Ra-228, respectively............................................................................................................10
6. Total radium activity, Ra-228/Ra-226, and total dissolved solids (TDS) as a function
of time since initiation of flowback...........................................................................................11
7. Activities of Ra-226 and total radium (Ra-226+Ra-228) in relation to total dissolved
solids (TDS)...................................................................................................................................13
8. Total radium and Ra-228/Ra-226 plotted against the age of the producing formation... 14

iv

Tables
1. Well locations and related information compiled for samples used in this study............19
2. Ra-226, Ra-228, gross alpha, and gross beta activities measured in samples
of produced water for wells listed in table 1...........................................................................26

Units and Conversions
pCi/L – picocuries per liter
dpm – disintegrations per minute
Bq – becquerels

1 pCi = 0.037 Bq; 1 Bq = 27.03 pCi
1 pCi = 2.22 dpm; 1 dpm = 0.4505 pCi


By E.L. Rowan,1 M.A. Engle,1 C.S. Kirby,2 and T.F. Kraemer1

Abstract Introduction
Radium activity data for waters co-produced with oil and Radium forms naturally from the decay of uranium and
gas in New York and Pennsylvania have been compiled from thorium, elements that commonly occur in sandstones and
publicly available sources and are presented together with new shales in sedimentary environments. Radium has been docu-
data for six wells, including one time series. When available, mented in the formation waters in many sedimentary basins
total dissolved solids (TDS), and gross alpha and gross beta (for example, Fisher, 1998). In the northern Appalachian
particle activities also were compiled. Basin, radium has been measured in the water co-produced
Data from the 1990s and earlier are from sandstone and with gas and oil (that is, produced water3) from reservoirs
limestone oil/gas reservoirs of Cambrian-Mississippian age; of Cambrian-Mississippian age. Radioactive isotopes are
however, the recent data are almost exclusively from the commonly quantified in terms of “activity concentration” or
Middle Devonian Marcellus Shale. The Marcellus Shale simply “activity,” which in this context refers to a number
represents a vast resource of natural gas the size and of disintegrations per unit time. For consistency with the
significance of which have only recently been recognized. studies cited, activity units of picocuries per liter (pCi/L) are
Exploitation of the Marcellus involves hydraulic fracturing used here to define the activity of radium in produced water
of the shale to release tightly held gas. Analyses of the water samples.
produced with the gas commonly show elevated levels of In surface and shallow subsurface environments, radium
salinity and radium. can be relatively soluble and, therefore, mobile in groundwater
Similarities and differences in radium data from reser- over a range of pH and Eh (redox) conditions (Langmuir and
voirs of different ages and lithologies are discussed. The range Riese, 1985; Sturchio and others, 2001). Radium also may
of radium activities for samples from the Marcellus Shale be adsorbed onto clay particles or onto oxide grain coatings
(less than detection to 18,000 picocuries per liter (pCi/L)) (Krishnaswami and others, 1982; Ames and others, 1983;
overlaps the range for non-Marcellus reservoirs (less Sturchio and others, 2001). As a radioactive element, radium
than detection to 6,700 pCi/L), and the median values are may represent a potential health hazard if released into the
2,460 pCi/L and 734 pCi/L, respectively. A positive correla- environment. The half-lives of the two principal isotopes of
tion between the logs of TDS and radium activity can be radium, Ra-226 and Ra-228, are 1,600 and 5.75 years, respec-
demonstrated for the entire dataset, and controlling for this tively (Akovali, 1996; Artna-Cohen, 1997), and approximately
TDS dependence, Marcellus shale produced water samples 10 half-lives are required for a radioactive element to decay to
contain statistically more radium than non-Marcellus samples. negligible quantities. Chemically, radium behaves in a manner
The radium isotopic ratio, Ra-228/Ra-226, in samples from similar to calcium and is capable of bioaccumulation in plants
the Marcellus Shale is generally less than 0.3, distinctly lower and animals. There is a significant body of research aimed at
than the median values from other reservoirs. This ratio may quantification of radium uptake in crops and livestock that
serve as an indicator of the provenance or reservoir source of make up the human food chain (for example, Tracy and others;
radium in samples of uncertain origin. 1983; Bettencourt and others, 1988; Linsalata and others,

1
U.S. Geological Survey, Reston, Virginia.
2
Bucknell University, Lewisburg, Pennsylvania.
3
The term “produced water” in this report represents water produced from an oil or gas well at any point during its life cycle. The term, therefore, includes
waters produced immediately after hydraulic fracturing, with compositions close to those of the injected fluid, as well as waters produced after months or years
of production, whose compositions resemble formation water.

2 Radium Content of Oil- and Gas-Field Produced Waters in the Northern Appalachian Basin: Summary and Discussion

1989). Most of these studies were conducted in areas where the Marcellus Shale, from less than 1,500 mg/L to greater than
uranium mining had previously taken place; however, it is not 300,000 mg/L. The lower salinities may be attributed in part to
known whether similar investigations have been conducted dilution with less saline fluid injected during hydraulic fractur-
in regions where oil- and gas-field produced waters are the ing, but the upper end of the salinity range is comparable to
source of radium. The purpose of this report is to compile the waters produced from the underlying Lower Devonian and
and present data from multiple sources to facilitate ongoing older reservoirs as well as some of the overlying Devonian
research. reservoirs (Rowan and others, 2010).
Activity data for radium-226 (Ra-226) and radium-228 The Marcellus Shale is an organic-rich shale that is both
(Ra-228) in oil- and gas-field produced waters from New the source rock and the reservoir for an extensive natural gas
York and Pennsylvania have been compiled from publicly resource (Harper, 2008). Shale-gas accumulations, such as
available sources and combined with new data for six wells the Marcellus, are termed “unconventional” or “continuous”
(tables 1 and 2, p. 19–31). Measurements of total dissolved because the gas is dispersed within a stratigraphic interval
solids (TDS) and of gross alpha and beta activities were also rather than confined by a conventional structural or strati-
tabulated when available. Unstable (radioactive) isotopes graphic trap. The process of “hydraulic fracturing” commonly
decay by emitting alpha and beta particles; therefore, alpha is used to access the gas in a continuous reservoir. In this
and beta activities can serve as rough indicators of the process, water is pumped into a well at pressures high enough
presence of radioactive elements. to fracture the rock, and the newly created fracture network
The publicly available radium data were obtained allows gas that is tightly held in micropores or adsorbed
from the New York State Department of Environmental onto clay particles to be released. The injected fluid may be
Conservation (NYSDEC), the Pennsylvania Department of freshwater or relatively dilute, or alternatively, it may have
Environmental Protection (PA DEP), and the Pennsylvania been recycled, that is, produced from one well and then used
Geological Survey. Most of these data are available online, to hydraulically fracture a new well. The water flowing from
although the most recent Marcellus Shale produced water hydraulically fractured wells initially reflects the composition
data were available only from the regional PA DEP offices. of the injected fluid, but with time shifts toward salinities
Three of the studies, Gilday and others (1999), Pennsylvania and inorganic chemical compositions similar to the fluids in
Department of Environmental Protection (1992), and Dresel adjacent formations (for example, Rowan and others, 2010).
and Rose (2010), provide data from wells producing from Hayes (2009), for example, examined the chemistry of
reservoirs of Cambrian-Devonian age. In contrast, the analyses produced water samples collected from 12 Marcellus Shale
reported by the New York State Department of Environmental wells at 1-, 5-, 14-, and 90-day intervals following hydraulic
Conservation (2009) and by the Pennsylvania Department of fracturing. The water injected into these wells was essentially
Environmental Protection (unpub. data, 2009–2010) are for fresh, with a median TDS of less than 1,000 mg/L, but within
produced waters predominantly from the Devonian Marcellus 90 days, the salinities had increased to a median value exceed-
Shale. ing 200,000 mg/L TDS.
Ra-226 and Ra-228 are the decay products of U-238 and
Th-232, respectively (fig. 1; Ivanovich, 1992). Once formed,
radium may remain within the original host mineral or other
Background solid phase, or may be released into the adjacent pore water.
Lithologies that contain substantial amounts of uranium and
The Appalachian Basin comprises a vast accumulation (or) thorium can, therefore, have measurable amounts of
of sedimentary rock west of the Appalachian Mountains, radium dissolved in their pore waters. The data compiled in
extending from Quebec and Ontario south through New York, this report span most of the oil- and gas-producing regions of
Pennsylvania, Ohio, West Virginia, to Alabama. Hydrocarbons the Appalachian Basin in Pennsylvania and New York (fig. 2),
are produced throughout the basin from reservoirs of Cam- and show significant levels of radium in produced water
brian-Pennsylvanian age (Legall and others, 1981; Milici and samples from Cambrian-Mississippian reservoirs.
others, 2003). In recent years, however, the Middle Devonian Dissolved radium occurs predominantly as the Ra+2
Marcellus Shale has become the focus of gas exploration and ion, but also forms complexes with chloride, sulfate, and
production, particularly in Pennsylvania, New York, and West carbonate ions (Rose and Korner, 1979; Kraemer and Reid,
Virginia. 1984; Langmuir and Riese, 1985; Sturchio and others, 2001).
A regional comparison of produced water salinities Aqueous radium may remain in solution, be adsorbed from
indicates that Appalachian Basin salinities are high relative pore water onto oxide grain coatings or clay particles by ion
to other oil- and gas-producing basins in the United States exchange, or may substitute for cations, such as Ba+2, Ca+2,
(Breit, 2002). The compilation yielded a median TDS of about and Sr+2, during precipitation of mineral phases, such as barite,
250,000 milligrams per liter (mg/L) for the Appalachian Basin anhydrite, and calcite. Radium sulfate (RaSO4) is much less
(USA), which was exceeded only by the median salinity for soluble than barite, anhydrite, and other sulfate minerals,
the Michigan Basin (about 300,000 mg/L). The data presented but rarely occurs as a separate mineral phase. When alkali
here indicate a wide salinity range for water produced from earth sulfates precipitate, however, radium present in solution

Background 3

A. Uranium-238

Uranium U-238 U-234 EXPLANATION
4.5 x 109 y 2.5 x 105 y
Alpha decay
Beta decay
Protactinium Pa-234
6.7 h

Thorium Th-234 Th-230
24.1 d 7.5 x104 y

Radium Ra-226
1600 y

Radon Rn-222
3.8 d

Polonium Po-218 Po-214 Po-210
3.1 m 1.6x10-4s 138.4 d

Bi-214 Bi-210
Bismuth 5.0 d
19.9 m

Pb-214 Pb-210 Pb-206
Lead 26.8 m 22.2 y (stable)

B. Thorium-232

Th-232 Th-228 EXPLANATION
Thorium
1.4 x 1010 y 1.9 y
Alpha decay
Beta decay
Ac-228
Actinium
6.15 h

Radium Ra-228 Ra-224
5.75 y 3.6 d

Radon Rn-220
55.6 s

Po-216 Po-212
Polonium
0.145 s 3.0 x10-7 s

Bi-212
Bismuth 60.6 m

Pb-212 Pb-208
Lead 10.6 h (stable)

TI-208
Thallium 3.05 m

Figure 1. Radioactive decay chains for (A) U-238 and (B) Th-232. Times shown are half-lives: y, years; d, days;
h, hours; m, minutes; s, seconds. Ra-226 and Ra-228 (shaded) are the primary isotopes of interest in this study.
Half-lives were obtained from the National Nuclear Data Center (http://www.nndc.bnl.gov/chart/ ).
Figure 1

80° 78° 76°

Jefferson
44° New
York
Lewis

Hamilton
New Jersey LAKE ONTARIO
Pennsylvania Oswego
West Orleans Oneida
Niagara Herkimer
Virginia Monroe Wayne
Genesee
Onondaga
Madison
Ontario
Erie Cayuga
Wyoming Livingston Yates Otsego
Seneca Cortland Chenango
Map Area
qua Tompkins
au Schuyler
aut
Ch Allegany Steuben
LAKE ERIE Cattaraugus
Tioga Delaware
Chemung Broome
NEW YORK
42° Erie
Warren PENNSYLVANIA
Bradford
Susquehanna
McKean Potter
Sullivan
Crawford Tioga Wayne
Forest Cameron Wyoming
Elk Sullivan Lackawanna
Venango
Mercer Lycoming Pike

OHIO
Clarion Luzerne
Clinton
M

Jefferson

PENNSYLVANIA
Columbia Monroe
on

Lawrence Clearfield Union
tou
r

Butler Armstrong Carbon n
Centre d pto
Snyder erlan am
mb rth
Beaver Mifflin orthu Schuylkill No
N
N

Indiana Juniata Lehigh P
Dauphin
EW

EN
Allegheny Cambria Blair
JE
R

Westmoreland Perry Berks Bucks NSY
SE

Lebanon Mo LV
Y

Huntingdon ntg AN
om IA
Washington Cumberland ery
Lancaster Philadelphia
40° Bedford Chester
Fayette Fulton Franklin York
Greene Adams Delaware
PENNSYLVANIA
Somerset PENNSYLVANIA
WEST VIRGINIA MARYLAND Cecil Gloucester
0 Harford25 50 75
Salem 100 MILES
New Castle
Harford
0 40 80 120 KILOMETERS

Base from U.S. Geological Survey digital data
EXPLANATION

PA DEP (unpub. data, 2009–2010) NYSDEC (2009) Dresel and Rose (2010)
NYSDEC (Gilday and others, 1999) This study PA DEP (1992)

Figure 2. Locations of wells with data compiled for this study.

Data Sources and Analytical Methods 5

coprecipitates as a solid-solution, preferentially enriching the reliable indicator of Ra-228 activity, and Pb-212, which occurs
solid phase and depleting the solution of radium (Langmuir lower on the decay chain (fig. 1), was seldom used.
and Riese, 1985). The values listed in table 2 are consistent with the
approach of Gilday and others (1999), but several instances
differ from the values highlighted in their report as representa-
tive of a given sample. At one well (no. 76), the Pb-212
Data Sources and Analytical Methods activity was anomalously high, 23,900 pCi/L, relative to a
corresponding Ac-228 activity of 1,500 pCi/L. Gilday and
The sources of data in this report (tables 1 and 2) are others (1999) concluded that the Pb-212 value was erroneous,
discussed below together with the available information although this was the value they highlighted as representative
on quality assurance/quality control (QA/QC), analytical of the sample. At a second well (no. 82), a Pb-212 activity of
methods, and uncertainty. The U.S. Environmental Protection 7,650 pCi/L also appeared to be anomalously high relative
Agency (USEPA) method codes refer to standard analytical to the Ac-228 activity of 1,110 pCi/L. In both instances, the
procedures defined by the USEPA (Krieger and Whittaker, Ac-228 rather than the Pb-212 activities are used to represent
1980; Eaton and others, 2005). Ra-228 in table 2. Pb-212 activities were used in only five
instances where Ac-228 was not reported. In wells where
New York State Department of Environmental duplicate analyses were available, (nos. 38, 56, 79, and 80),
Conservation Report (Gilday and others, 1999) the averages are given in table 2.
The New York State Department of Environmental All of the samples collected by Gilday and others (1999)
Conservation (NYSDEC) conducted a study titled “An were analyzed by an outside contract laboratory, and a subset
Investigation of Naturally Occurring Radioactive Materials of nine samples was also analyzed by the NYSDEC Bureau
(NORM) in Oil and Gas Wells in New York State,” in which of Pesticide and Radiation laboratory. Some interlaboratory
produced water, oil, sludge, and other waste materials were comparison and QA/QC information was provided in that
sampled from oil and gas wells in New York State (Gilday report and is discussed below. Ideally, metrics of both analyti-
and others, 1999). Analyses were reported for a total of cal accuracy (proximity of measured value to the “true” value)
57 brine samples collected from 48 oil or gas well sites, with and precision (measurement reproducibility) are presented.
9 duplicate or replicate samples (table 1). The NYSDEC report Because no analyses of reference materials or other standards
indicates that the brines were sampled from storage tanks, but were reported, the analytical accuracy for the included data
the length of time between production and sample collection is is unknown. Sample precision was examined by comparing
unknown. The wells in this study produced hydrocarbons and data for analyses of duplicate4 and replicate5 samples (fig. 3).
water from formations of Cambrian through Lower Devonian Despite the reported “internally consistent results” from
age, with one sample of possible mixed Lower Silurian and each laboratory, the measurement uncertainty ranges did not
Upper Devonian reservoir origin (table 1). Several of the wells overlap in five out of nine brine samples analyzed by both
produced from the Lower Devonian Oriskany Sandstone and laboratories. A single outlier exhibited an exceptionally high
Helderberg Limestone. Silurian reservoirs provided samples difference of 143 percent between replicate analyses for
from the Akron Sandstone, Bass Islands Dolomite, Medina Ra-226.
Sandstone, and Rochester Shale. Ordovician reservoirs These findings indicate that sample precision is generally
included sandstones within the Queenston Shale. better (less than 20 percent discrepancy between duplicate
Analyses of radium activity in the NYSDEC report or replicate samples) for samples that contained greater than
were determined using gamma-spectrometry as well as 500 pCi/L, but poor agreement in interlaboratory comparisons
alpha-spectrometry in some cases. Gamma-spectrometry indicates there may be bias between data sources. The
compares the gamma-ray wavelengths emitted by radioactive magnitude of the biases, however, appears to be in the tens of
material with the emission spectra of known radioactive percents while radium activities in brine samples range over
elements. In some instances, the signal emitted by a daughter more than four orders of magnitude. This comparison suggests
product can be more accurately identified and quantified than that even the higher end of analytical imprecision observed in
that of its parent isotope. Laboratories may therefore elect the data does not significantly affect the magnitude of radium
to report a daughter product activity as representative of its activities reported.
radium parent’s activity in an appropriately prepared sample.
Gilday and others (1999) considered that the Ra-226 daughter
products Pb-214 and Bi-214 were the most reliable indicators 4
Duplicate refers to individual samples from a single source collected at the
of Ra-226 activity, and they selected the larger of the Pb-214 same place and time.
and Bi-214 values to represent the Ra-226 activity. Gilday 5
Replicate refers to a repeat analysis made on the same sample or aliquots of
and others (1999) considered Ac-228 activity to be the most the same sample.


EXPLANATION
600 Ra-226

Absolute measurement difference,
Ra-228

in picocuries per liter 400 >20%

10–20%
200

<10%
0
0 500 1,000 1,500 2,000
Mean activity, in picocuries per liter

Figure 3. Differences between measurements of duplicate and replicate analyses
of Ra-226 and Ra-228 in produced water samples in relation to the mean activity of
the sample for data from Gilday and others (1999). The solid lines represent 10 percent
and 20 percent relative difference between duplicates/replicates using the method of
Thompson and Howarth (1978). Samples with higher radium activities generally have
better measurement precision, that is, lower percentage differences.

New York State Department of Environmental Pennsylvania Department of Environmental Protection
Conservation, Draft Supplemental Generic Report (1992)
Environmental Impact Statement (2009) In 1991, the PA DEP conducted field work for a study of
In 2009, the NYSDEC released a study titled “Draft salinity and radium activities in produced waters, sludge, and
Supplemental Generic Environmental Impact Statement other related waste from oil and gas wells in Pennsylvania.
related to Marcellus Shale Gas Development” (New York The results were compiled in a report titled “NORM Survey
State Department of Environmental Conservation, 2009). Summary” and released the following year (Pennsylvania
Appendix 13 of the document, “NYS Marcellus Radiological Department of Environmental Protection, 1992). The wells
Data from Production Brine,” lists gross alpha, gross beta, and sampled for the study produced hydrocarbons and water
activities of Ra-226 and Ra-228 for water samples collected from Lower Silurian–Upper Devonian Formations, with one
from 12 gas-producing Marcellus Shale wells in New York sample thought to be from an Ordovician reservoir. Although
State. Appendix data were presented in table form without the Marcellus Shale falls within this stratigraphic interval, the
accompanying text, information relating to QA/QC, or analyti- study long pre-dated the recent (2005–present) focus on the
cal methods. However, well lease names and API numbers, Marcellus Shale as an unconventional gas resource. Among
towns, and counties were provided, allowing well locations the most commonly sampled reservoirs were sandstone in the
and related information to be obtained from the State database Silurian Medina Group, the Lower Devonian Oriskany Sand-
(http://www.dec.ny.gov/; fig. 2; tables 1 and 2). Activities stone, Huntersville Chert, and Onondaga Limestone, as well as
of uranium, thorium, and the anthropogenic isotopes, Upper Devonian sandstones (table 1). About three-fourths of
cesium-137, cobalt-60, ruthenium-106, and zirconium-95, the samples were taken from storage tanks, or separator tanks,
were listed in the appendix, but are not compiled in this report. and the remaining samples were collected from surface pits or
diked areas (table 1). The length of time between hydrocarbon
production and sample collection is unknown, and therefore,
Ra-228 activity may have been markedly reduced by natural
decay. Brines that accumulated in open pits presumably would
have been subject to evaporation and (or) dilution by rain.

Data Sources and Analytical Methods 7

In addition to brine samples, samples of sludge, drill (Alpha-Emitting Radium Isotopes in Drinking Water), and
cuttings, and pipe scale from brine treatment facilities, pipe EPA Method 903.1 (Radium-226 in Drinking Water Radon
yards, disposal wells, and other facilities were analyzed, but Emanation Technique). Radium-228 was analyzed using
these results were not compiled in this report. No information similar methods: EPA Method 901.1 (Gamma Emitting
on the laboratory, analytical methods, uncertainties, or QA/QC Radionuclides in Drinking Water) and EPA Method 904.0
was included with the PA DEP (1992) report. (Radium-228 in Drinking Water). For the four sets of duplicate
Ra-226 and Ra-228 analyses, the discrepancies were less than
Pennsylvania Department of Environmental Protection 7 percent, with one exception: Ra-226 analyses in duplicate
Reports (Unpublished Data, 2009–2010) samples from well no. 1 differed by 72 percent.

A number of the annually filed “Form 26R” (Chemical
Dresel and Rose (2010)
Analysis of Residual Waste, Annual Report by Generator)
waste reports related to shale gas production were obtained A recent publication by Dresel and Rose (2010) reports
from the PA DEP. The forms and accompanying chemical the produced water analyses originally conducted as part of
analyses are filed annually with the PA DEP by generators a Master’s thesis at Pennsylvania State University (Dresel,
of liquid or solid waste, including oil and gas well operators. 1985). Of the 40 samples collected, Ra-226 analyses are
The 26R forms can be viewed at the DEP regional offices reported for six wells producing hydrocarbons and water from
by appointment, or photocopies can be requested from the Lower Silurian–Upper Devonian sandstone reservoirs; Ra-228
DEP. The DEP offices in Williamsport and Pittsburgh were values are not reported. Most of the samples in this study were
visited during the spring and summer of 2010, and the collected from the wellhead rather than secondary storage
available 26R forms pertaining to liquid waste generated at units (table 1). The Ra-226 activities reported were determined
gas well sites were electronically scanned. Additional data by measurement of radon-222 activity at secular equilibrium
were obtained by correspondence with the Meadville, Pa., (Rose and Korner, 1979), using a method equivalent to USEPA
office. Radium activities from the 26R forms were included in Method 903.1 (Krieger and Whittaker, 1980). Detailed QA/QC
this report only when the well name and related information information was not available.
could be obtained for a given sample. Information obtained
from 26R forms filed with the PA DEP during 2009–2010 This Study
for a total of 23 wells was compiled and included in tables 1
Radium activities have been determined at the U.S. Geo-
and 2. In most instances, the TDS values of the samples were
logical Survey (USGS) for samples from six additional
also available. Time series data were available for four wells
Marcellus Shale gas wells in Pennsylvania. Samples were
(table 2). When duplicate analyses were provided, the average
collected from five of the wells (nos. 127–131, tables 1
value is shown in table 2.
and 2) as part of a study by Pritz (2010). The precise localities
Laboratory notes accompanying 26R forms reported to
of these wells in Bradford County are confidential, and they
the PA DEP varied substantially between individual wells,
are represented in figure 2 by a single point. Well no. 132 was
but all included the laboratory name and, in some cases,
sampled jointly by the USGS, the Department of Energy, and
the analytical method and QA/QC information. Despite the
industry collaborators on successive dates, thus providing time
numerous different reporting entities, the radiochemical data
series information. Analyses of the samples were conducted
reported in the 26R forms were obtained from only four
at the USGS radiochemistry laboratory in Reston, Virginia.
different laboratories, and all are accredited in accordance with
Two to four duplicates of each sample from well no. 132 were
the National Environmental Laboratory Accreditation Program
prepared and analyzed, and the average values are reported in
(NELAP).
table 2.
Gross alpha and beta emission measurements included
In the samples from well no. 132, radium was chemically
in the PA DEP 26R forms were determined by methods that
separated from the water by coprecipitating it with barium
include standard and modified versions of EPA Method 900.0
sulfate. The precipitate was then placed in the well of a high
(Gross Alpha and Gross Beta Radioactivity in Drinking
purity germanium detector, and quantitative analysis of the
Water) and Standard Method 7110C (Eaton and others, 2005).
Ra-226 and Ra-228 content of the precipitate was performed
No duplicate samples, replicate analyses, or other QA/QC
by gamma-spectrometry using a technique adapted from
information were available for either the gross alpha or beta
Moore (1984). As discussed above for the New York State
results.
data of Gilday and others (1999), Ra-228 was quantified by
When methods for radium analysis were reported, Ra-226
measuring the intensity of gamma rays emitted by Ac-228,
activity typically was measured using gamma-spectrometry,
and Ra-226 was quantified by measuring the intensity of the
and in some cases by alpha-spectrometry, using standard
gamma rays emitted by Pb-214 and Bi-214. As described in
USEPA methods: EPA Method 901.1 (Gamma Emitting
Kraemer (2005), the gamma-ray spectrometry systems were
Radionuclides in Drinking Water), EPA Method 903.0
calibrated using standardized radium isotopic solutions.


Uncertainties for these analyses are listed in table 2 as the Marcellus produced water data for New York State (fig. 4;
+/– one standard deviation from the mean peak intensity and New York State Department of Environmental Conservation,
represent the “counting error” for a specific analysis. When 2009; Pennsylvania Department of Environmental Protection,
duplicate samples were prepared, that is, reprecipitated, and unpub. data, 2009–2010; this study). For comparison, the
analyzed, the range of the discrepancies matched closely with total radium limit for industrial effluent is 60 pCi/L, and the
the range for the counting error: 0.2–8.5 percent. However, drinking water limit is 5 pCi/L (U.S. Environmental Protection
the discrepancies between analyses of duplicate samples were Agency, 1976; Hess and others, 1985; U.S. Nuclear Regula-
most commonly 2–4 percent higher than the counting error. In tory Commission, 2011).
all cases, the maximum error did not exceed +/– 8.5 percent. In the NYSDEC (2009) study, salinities were not
reported; however, two wells, no. 28 and no. 33, were
resampled and analyzed by Osborn and McIntosh (2010),
yielding respective salinities of 206,446 and 205,102 mg/L
Results TDS. Samples at two additional wells, no. 24 and no. 25,
both from depths of approximately 2,600 feet (ft), exhibited
Salinity and Radium very low total radium activities (less than 1 pCi/L), although
the activities of the remaining sites exceeded 1,900 pCi/L
Salinities, reported as TDS, were available for approxi- (fig. 4; table 2). The reason for the low radium content of
mately one-half of the produced water samples and ranged these samples is unknown, but they may have been composed
from 1,470 to 402,000 mg/L with a median of 157,000 mg/L largely of water injected for hydraulic fracturing, which often
TDS (table 2). The median total radium (defined here as is of lower salinity and radium content than the formation
Ra-226 + Ra-228) activity for the non-Marcellus Shale pro- water.
duced water samples is 1,011 pCi/L compared with 2,460 for In Pennsylvania, the range of total radium activities
Marcellus Shale produced water samples and 5,490 pCi/L for for the Marcellus Shale samples (Pennsylvania Department

100,000
)

l)
al

ta
t
26

26
(to

(to
-2

-2
Ra
Ra

Ra
Ra

l)
l)

ta
ta

26
(to
l)
(to
26

ta

-2
26
10,000
-2

Ra
(to
Ra

Ra
26

-2
Ra

Ra
-2

Ra
Ra (total), Ra-226, in picocuries per liter

Ra

1,000

100

10
Marcellus Shale data
PA DEP (2009–2010, NYSDEC (2009) this study NYSDEC PA DEP (1992) Dresel and Rose
unpub. data) (13) (14) (Gilday (37) (2010)
(25) and others, 1999) (6)
1 (48)

0

Figure 4. Measured activities for total radium (Ra-226 + Ra-228) and Ra-226 for each of the data sources used in
the study. The three datasets for produced water from Marcellus Shale wells are shown on the left; the remaining
three datasets are for non-Marcellus Shale wells. The number of points in each dataset is shown in parentheses,
and the median values are plotted as heavy black lines. For reference, the dashed line shows the industrial effluent
discharge limit (60 pCi/L) for Ra-226 (U.S. Nuclear Regulatory Commission, http://www.nrc.gov/reading-rm/
doc-collections/cfr/part020/appb/Radium-226.html).

Discussion 9

of Environmental Protection, unpub. data, 2009–2010) is in bromide and can be distinguished from brines formed by
similar to the Marcellus data from New York but is more dissolution of evaporites on the basis of relations among
evenly distributed (less clustered) over the range. Dilution Na, Cl, and Br (Walter and others, 1990). Brines produced
of formation water with the relatively freshwater from the with gas from Marcellus Shale wells after salinities have
hydraulic fracturing process may have been an important reached a plateau share similar major ion chemistries with
factor influencing the distribution of both salinity and radium formation waters from the overlying and underlying Devonian
content. The time interval between hydraulic fracturing and formations and show similar Na-Cl-Br relations (Osborn and
sample collection is known in only a few cases. McIntosh, 2010; Rowan and others, 2010). On the basis of
these chemical similarities, a similar origin for the salinity of
waters produced from the Marcellus Shale and from adjacent
Gross Alpha and Beta Particle Emissions overlying and underlying formations can be hypothesized.
Blauch and others (2009), however, reported small lenses
Emission of alpha and beta particles accompanies the
of halite and other salts in core from the Marcellus Shale and
decay of Ra-226 and Ra-228, respectively (fig. 1), and the
suggested that dissolution of these minerals contributed to the
USEPA has established the measurement of gross alpha
salinity of the produced waters. They also described minor
and beta as a method of screening samples for the presence
volumes of salts, but noted that similar occurrences have not
of radium (Hess and others, 1985; Buckwalter and Moore,
previously been reported in the literature on the Marcellus.
2007, p. 48). Gross alpha and beta data were available for
Where present, salt lenses would contribute to total salinity,
two datasets (New York State Department of Environmental
but it is difficult to assess their distribution or quantify their
Conservation, 2009; Pennsylvania Department of Environ-
contribution to total fluid salinity. The elevated bromide
mental Protection, unpub. data, 2009–2010) and are plotted
concentrations and Na-Cl-Br relations suggest that the
with Ra-226 and Ra-228, respectively (figs. 5A–B). On log-log
dominant source of salinity for Marcellus Shale waters, and
scales, gross alpha and gross beta activities are linearly
for other formations in the stratigraphic section, originated as
correlated with Ra-226 and Ra-228, confirming their value
evaporatively concentrated seawater.
as indicators of radium activity. Although these isotopes are
Dilution of formation water with relatively freshwater
unlikely to be the only sources of alpha and beta particles, the
injected during the hydraulic fracturing may account for some
correlations shown in figures 5A–B suggest that they are likely
of the lower salinity values. For example, in well no. 11 salini-
to be the dominant sources for these samples.
ties were measured 14 and 90 days after hydraulic fracturing
and showed an increase with time (fig. 6A; table 2). In well
no. 5, successive salinity measurements made 17 days apart
Discussion also showed increased salinity with time (table 2). In a more
detailed study by Hayes (2009), repeated measurements of
Salinity and Dilution produced water salinity up to 90 days after hydraulic fractur-
ing showed increases in salinity with time from less than 1,000
Several studies of Appalachian Basin formation water mg/L to greater than 100,000 mg/L TDS. The marked increase
chemistry have shown general trends of increasing salinity in salinity with time is interpreted to represent a decreasing
with depth and age of the reservoir (for example, Stout and proportion of the lower salinity injected fluid and an increas-
others, 1932; Poth, 1962; Breen and others, 1985); however, ing proportion of the saline formation water returning to the
high salinities can occur even at relatively shallow depths. A surface. As mentioned previously, dissolution of mineral
salinity-depth curve for Mississippian-Devonian formation phases such as halite, if present, could also contribute salinity.
waters in eastern Ohio showed greater than 100,000 mg/L For data compiled from the PA DEP 26R forms, when the
TDS at 1,000 ft (Stout and others, 1932, p. 18). Poth (1962, sample collection date occurred less than 90 days from the
p. 37–38, table 6) noted that on the basis of a limited set of initiation date of drilling, it seems plausible that salinities less
samples, an equilibrium salinity had apparently been reached than 100,000 ppm TDS may have been affected by dilution of
in Middle Devonian and older reservoirs, and water produced the formation water with the water injected during hydraulic
from these units have a dissolved solids content of about fracturing.
300,000 mg/L. In the dataset compiled here, produced water Like salinity, radium in the produced waters increases
salinities from the Middle Devonian Marcellus Shale com- with time following hydraulic fracturing. A well producing
monly range from 100,000 to 200,000 mg/L TDS (table 2). gas from the Marcellus Shale in Greene County, Pa. (no. 132),
Despite the presence of evaporites in the Silurian Salina was sampled daily for the first 5 days, then on days 7, 15,
Group in parts of New York and Pennsylvania, many authors and 20 following hydraulic fracturing. The radium data for
consider that the salinity of the produced waters in much of these samples are discussed below, but the salinity data are
the Appalachian Basin originated from the evaporative con- not yet available. The hydraulic-fracturing supply water was
centration of seawater (for example, Stout and others, 1932; a mix of water recycled from similar gas wells nearby and
Sanders, 1991; Dresel and Rose, 2010; Osborn and McIntosh, more dilute surface water. Following hydraulic fracturing, the
2010). Brines derived from seawater evaporation are enriched total radium activity in the produced water increased sharply


A Gross alpha, in picocuries per liter
1 10 100 1,000 10,000 100,000
5.0 100,000

4.0 10,000
Log Ra-226, in picocuries per liter

Ra-226, in picocuries per liter
3.0 1,000

2.0 100

1.0 10
y = 1.35x – 2.05
R2 = 0.78
0.0 1

–1.0 0.1
0.0 1.0 2.0 3.0 4.0 5.0
Log gross alpha, in picocuries per liter

B Gross beta, in picocuries per liter
1 10 100 1,000 10,000 100,000
4.0 10,000

3.0 1,000
Log Ra-228, picocuries per liter

Ra-228, in picocuries per liter
2.0 100

1.0 10

y = 1.34x – 2.14
0.0 R2 = 0.54 1

–1.0 0.1

–2.0 0.01
0.0 1.0 2.0 3.0 4.0 5.0
Log gross beta, in picocuries per liter

Figure 5. (A) Log of gross alpha particle activity in relation to the log of Ra-226 activity and (B) log of gross beta
Figure 5.
particle activity in relation to the log of Ra-228 activity. Gross beta activities below the reported detection limit in
well nos. 17 and 33 are not plotted or used in the best fit line.

Discussion 11

A
4,000 200,000

3,500

Total dissolved solids, in milligrams per liter
Total radium, in picocuries per liter

3,000 150,000

2,500

2,000 100,000

1,500

EXPLANATION
1,000 50,000
Total radium
500 Total dissolved solids

0 0
0 10 20 30 40 50 60 70 80 90
Days since initiation of flowback

B
7,000 0.40

6,000
0.35

Ra-228/Ra-226, in picocuries per liter

5,000
0.30
EXPLANATION
4,000
Total radium
0.25
Ra-228/Ra-226
3,000

0.20
2,000

0.15
1,000

0 0.10
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Days since initiation of flowback

Figure 6. (A) Total radium activity and total dissolved solids related to time since initiation of flowback for
well no. 11, Washington County, Pa. (B) Total radium activity (left axis, squares) and Ra-228/Ra-226 (right axis,
diamonds) related to time since initiation of flowback for well no. 132, Greene County, Pa.


during the first week from the activity of the supply water brine production at the wellhead. A year after production at
(about 1,600 pCi/L) to a plateau at about 6,100 pCi/L (fig. 6B). the wellhead, for example, the activities of the shorter-lived
The increase in radium activity is interpreted as the result Ra-228 isotope would be reduced by approximately 11 percent
of equilibration between the injected water, whose radium because of natural decay.
activity is relatively low, and the radium that is present in the Analysis of covariance (ANCOVA6) was used to statisti-
reservoir, either adsorbed onto mineral surfaces or dissolved cally examine the effect of Marcellus versus non-Marcellus
in pore water. An anomalously low value on day 7 remains sample origin on the linear relationship between TDS and
as yet unexplained; close agreement between multiple repeat radium activity. The resulting linear regression models yield
analyses of the original sample conducted on different dates nearly identical slopes in the trends of log total radium in
rules out an instrumental or analytical error as an explanation. relation to log TDS and log Ra-226 in relation to log TDS.
The Ra-228/Ra-226 ratio for this fluid decreased from the However, the regression intercepts for the Marcellus Shale
initial value of 0.23 in the injected water to 0.12. Following data are 0.4 and 0.55 log units (2.5 and 3.5 times, respectively)
an unexplained increase on day 7, the ratio returned to about higher for total radium and Ra-226, respectively, than for
0.16. Low isotopic ratios reflect the low Th/U ratio that the non-Marcellus samples. Produced water samples from
generally characterizes the Marcellus Shale. The evolution of the Marcellus Shale are, therefore, enriched in radium to a
total radium and Ra-228/Ra-226 with time displays a fairly statistically significant degree (p<0.05) relative to samples
consistent pattern, with the exception of day 7. The area from other formations in the basin. This relative enrichment is
being drained by the well on this day may have intersected also illustrated in figure 8A.
a “pocket” of chemically distinct water, possibly a sandy An important mechanism by which salinity controls
horizon in the shale or a fracture intersecting a distinct facies. radium activity involves competition between Ra+2 and
other multivalent ions for adsorption sites primarily on clay
minerals (Kraemer and Reid, 1984). In low salinity fluids,
Radium Activities in Context radium tends to be adsorbed onto mineral surfaces and with
increasing salinity radium is progressively desorbed and
In a study of NORM (naturally occurring radioactive
released into solution. Sturchio and others (2001) presented
material) in oil- and gas-producing regions, Fisher (1998)
a salinity-dependent distribution coefficient (K) between
compiled radium activity data for nine sedimentary basins
dissolved radium and radium adsorbed onto clay particles and
in the United States and Europe. In separate studies, Ra-226
oxide grain coatings. The logs of K and TDS show a linear
activities were reported for formation water samples from
relationship with negative slope, indicating less adsorption
clastic aquifers in the U.S. Gulf Coast (Kraemer and Reid,
of radium at higher salinities and, therefore, more radium in
1984) and from carbonate aquifers in the U.S. midcontinent
solution (Sturchio and others, 2001, fig. 7).
(Sturchio and others, 2001). Radium activity ranges for these
In a study of saline groundwater systems in the mid-
regions generally are comparable to those compiled here for
continent with TDS concentrations reaching 250,000 mg/L,
the Appalachian Basin. The highest reported values found
Sturchio and others (2001, fig. 5) used equilibrium speciation
in literature are from the Donieper-Donets Basin, Ukraine
calculations to demonstrate that Ra+2 was the predominant
(Gutsalo, 1964, cited in Kraemer and Reid, 1984) and from
dissolved radium species, independent of salinity. In their
the Texas Panhandle (see references cited in Fisher, 1998); the
analysis, Ra+2 never accounted for less than 77 percent of the
high end of the Ra-226 activities ranges exceeded 4,500 pCi/L
total dissolved radium over a range of fluid chemistries. The
in the Ukraine and 5,000 pCi/L in the Texas Panhandle. In this
next most abundant species, RaCl+, gained significance with
report, several Ra-226 activities of approximately 4,000 pCi/L –
increasing salinity and Cl /SO4–2 ratios. Additionally, Sturchio
have been compiled for samples from non-Marcellus reser-
and others (2001) reported that radium forms strong organic
voirs, but the Marcellus Shale data range higher, with several
complexes at elevated salinities, which may be significant
activities exceeding 10,000 pCi/L (table 2; fig. 7).
because limited, unpublished data for organic compounds in
Relationships between salinity and radium activity
Appalachian Basin produced waters have shown significant
have been documented in a number of studies (for example,
concentrations of acetate and other anions of carboxylic acids.
Kraemer and Reid, 1984; Fisher, 1998; Sturchio and others,
Dissolved radium measured in produced water samples
2001). Fisher (1998), however, pointed out that chloride
originates from the decay of the parent isotopes, U-238
or TDS concentrations “best predict radium activity in
and Th-232, in uranium- and thorium-bearing minerals or
waters from reservoirs that are lithologically relatively
organic material contained in the host formation. The physical
homogeneous.” Despite their origin in reservoirs of varying
transfer of radium from the solid to the aqueous phase is
ages and lithologies, the non-Marcellus Shale data indicate a
discussed and illustrated in Fisher (1998, fig. 2) and Sturchio
positive correlation between salinity and both total radium and
and others (2001). Uranium and thorium, unlike radium, are
Ra-226 activities. Salinity (TDS) is plotted with total radium
poorly soluble in the oxygen-poor, reducing conditions that
and Ra-226 in figures 7A–B. Use of the longest-lived isotope,
Ra-226, may reduce some of the scatter induced in the data by 6
See Crawley (2007), for example, for additional discussion of the
sample collection at varying, or unknown, time intervals since ANCOVA and related statistical methods.

Discussion 13

A Total dissolved solids (TDS), in milligrams per liter
1,000 10,000 100,000 1,000,000
5.0 100,000
EXPLANATION
Marcellus Shale Data
Log total radium, in picocuries per liter

4.0 Non-Marcellus Shale Data 10,000

Marcellus ND
Non-Marcellus ND
3.0 1,000

2.0 Log total Ra = 1.55 x Log TDS – 4.86 100

1.0 10
Log total Ra = 1.55 x TDS – 5.26

0.0 1.0

–1.0 0.1
3.0 4.0 5.0 6.0
Log total dissolved solids (TDS), in milligrams per liter

B Total dissolved solids (TDS), in milligrams per liter
1,000 10,000 100,000 1,000,000
5.0 100,000
EXPLANATION
Marcellus Shale Data
4.0 Non-Marcellus Shale Data 10,000
Log Ra-226, in picocuries per liter

Marcellus ND

Ra-226, in picocuries per lite
Non-Marcellus ND
3.0 1,000

2.0 100
Log Ra-226 = 1.56 x Log TDS – 5.05

1.0 10

Log Ra-226 = 1.56 x Log TDS – 5.60
0.0 1.0

–1.0 0.1
3.0 4.0 5.0 6.0
Log total dissolved solids (TDS), in milligrams per liter

Figure 7. (A) Log activity of total radium (Ra-226 + Ra-228) in relation to log total dissolved solids and (B) log activity
of Ra-226 in relation to log total dissolved solids. Linear regression lines are shown for data from the Marcellus
Shale (red), and for non-Marcellus Shale (blue) data; the lines are solid over the range of the data and dashed where
extrapolated. Radium analyses listed as “ND” or not detected (well nos. 2 and 14) indicate values below the reported
Figure 7.
level of quantification (LOQ), 1 pCi/L. These points were replaced by one-half of the LOQ, or 0.5 pCi/L, and are plotted
as open squares, but are not included in the regression. A reported Ra-226 activity of 0 (well no. 123) was replaced
by 0.5 pCi, and is plotted as an open circle, but was not included in the regression.


A
Devonian Upper

Middle

Lower

Upper
Silurian

Middle
Formation Age

Lower

Upper EXPLANATION
Ordovician

Upper Devonian sandstones
Middle Middle Devonian, Marcellus Shale
Middle Devonian, Huntersville Chert
Middle Devonian, Onondaga Limestone
Lower Lower Devonian, Oriskany/Ridgely Sandstone
Lower Devonian, Helderberg Limestone
Cambrian

Upper Upper Silurian, Bass Islands/Akron Dolomite
Lower Silurian, Medina/Tuscarora Sandstone
Upper Orodvician, Queenston Shale
Middle Ordovician, undifferentiated
Upper Cambrian, Theresa Sandstone
Unknown Unknown age Median

0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000

B
Upper
Devonian

Middle

Lower

Upper
Silurian

Middle
Formation Age

Lower

Upper EXPLANATION
Ordovician

Upper Devonian sandstones
Middle Middle Devonian, Marcellus Shale
Middle Devonian, Huntersville Chert
Middle Devonian, Onondaga Limestone
Lower
Lower Devonian, Oriskany/Ridgely Sandstone
Lower Devonian, Helderberg Limestone
Cambrian

Upper Upper Silurian, Bass Islands/Akron Dolomite
Lower Silurian, Medina/Tuscarora Sandstone
Upper Orodvician, Queenston Shale
Middle Ordovician, undifferentiated
Upper Cambrian, Theresa Sandstone
Unknown Unknown age Median

0.0 1.0 2.0 3.0 4.0
Ra-228/Ra-226

Figure 8. (A) Total radium and (B) Ra-228/Ra-226 plotted against the age of the producing formation. For formations with
large numbers of analyses, the median is shown as a solid black dot.

Figure 8.

References Cited 15

are typical of oil- and gas-producing horizons and are likely The radium activities in non-Marcellus produced waters
to be more concentrated in mineral phases or organic matter in this report are broadly comparable to those reported in
than in solution (Langmuir and Herman, 1980; Kraemer and other studies of deep sedimentary basins with highly saline
Reed, 1984; Fisher, 1998; Sturchio and others, 2001). Thus, formation water. In the produced water dataset for the
dissolved radium that is in secular equilibrium with its parent Marcellus Shale in New York, total radium activities have a
isotopes of uranium and thorium at depth in a reservoir may be distinctly higher median (5,490 pCi/L) than reported for other
isolated from its parents when pumped to the surface. formations in the Appalachian Basin, and range to higher
In addition to displaying higher radium activities for a values than reported in other basins. Produced waters from
given salinity, produced water samples from the Marcellus the Marcellus in Pennsylvania have similar ranges to the New
Shale have distinctly lower Ra-228/Ra-226 ratios (median of York data, but a lower median value (1,727 pCi/L), interpreted
0.16) than those of non-Marcellus samples (median of 1.1; as being due, at least in part, to dilution of formation water by
figs. 8A–B), reflecting the Th/U ratio of the reservoir litholo- formation water injected for hydraulic fracturing.
gies. Organic carbon has long been known to play a role in In the data compiled here, Ra-228/Ra-226 ratios in
concentrating uranium (Swanson, 1960, 1961; Szalay, 1964), produced water from the Marcellus Shale are most commonly
and recent work by Bank and others (2010) has documented less than 0.3, and samples from non-Marcellus reservoirs
a close spatial association between the organic matter and generally have Ra-228/Ra-226 ratios greater than 1. Elevated
uranium in the Marcellus Shale. As an organic-rich black total radium activities combined with low Ra-228/Ra-226
shale, the Marcellus is readily identified on geophysical logs ratios characterize produced waters from the Marcellus Shale,
by its high gamma-ray signal (Schmoker, 1981; Harper, 2008). and these characteristics might be used to constrain the origin
In eastern Pennsylvania, numerous minor occurrences of samples of unknown provenance.
of uranium have been reported in upper Paleozoic sandstones
surveyed at roadside outcrops (Klemic, 1962). If the uranium
enrichment is assumed to extend into the subsurface, it
provides a potential source for the radium reported in non- Acknowledgments
Marcellus Shale produced waters, and as discussed above,
Funding for this project was provided by the USGS
high formation water salinity can account for elevated radium
Energy Resources Program and the USGS Toxic Substances
activities in solution.
Hydrology Program. Reviews by Zoltan Szabo and James
The Ra-228/Ra-226 ratios span a wide range among the
Otton and comments provided by Robert Zielinski are
non-Marcellus produced waters, and the highest values occur
gratefully acknowledged. Insightful comment and discussion
in samples from the Lower Silurian Medina Group/Tuscarora
was provided by Richard Hammack and Daniel Soeder
Sandstone. The median ratio for Medina/Tuscarora produced
(U.S. Department of Energy National Energy Technology
water samples is 1.61 (fig. 8B). Interestingly, this ratio is
Laboratory).
consistent with the data compiled by Vengosh and others
(2009, fig. 4), which suggest an average isotopic ratio of
approximately 1.6 for sandstones worldwide. The sandstones
of the Upper Devonian Bradford Group have lower isotopic References Cited
ratios (median, 0.78), possibly because of the interfingering of
sandstone with siltstone and shale beds in this interval.
Akovali, Y.A., 1996, Nuclear data sheets for A = 226: Nuclear
Data Sheets, v. 77, p. 433–470, accessed July 13, 2011, at
http://www.nndc.bnl.gov.
Summary Ames, L.L., McGarrah, J.E., and Walker, B.A., 1983, Sorption
of trace constituents from aqueous solutions onto second-
Produced water salinities from reservoirs in rocks of
ary minerals II. Radium: Clays and Clay Minerals, v. 31,
Cambrian-Devonion age in the Appalachian Basin commonly
p. 335–342.
exceed 100,000 mg/L, and far exceed the salinities of many
other oil- and gas-producing regions in the United States, Artna-Cohen, Agda, 1997, Nuclear data sheets for A = 228:
including basins in California, the Great Plains, and Colorado Nuclear Data Sheets, v. 80, p. 723–786, accessed July 13,
Plateau. In many basins, radium activity is correlated with 2011, at http://www.nndc.bnl.gov.
salinity, and particularly among samples from lithologically
homogeneous reservoirs, salinity may be used as an indicator Bank, Tracy, Malizia, Thomas, and Andresky, Lisa, 2010,
of radium activity. The data compiled for Pennsylvania Uranium geochemistry in the Marcellus Shale—Effects
indicate a relationship similar to that described in other basins; on metal mobilization: Geological Society of America
total radium and Ra-226 activities are linearly correlated with Abstracts with Programs, v. 42, no. 5, p. 502, accessed
TDS. Salinity was not reported in the datasets for New York. July 13, 2011, at http://gsa.confex.com/gsa/2010AM/
finalprogram/abstract_181465.htm.


Bettencourt, A.O., Teixeira, M.M.G.R., Elias, M.D.T., and Gilday, W.M., Edick, R.G., Rommel, R.E., Tetley, W.C.,
Faisca, M.C., 1988, Soil to plant transfer of radium-226: Kadlecek, J.A., Zeh, J.B., and Youngberg, B.A., 1999, An
Journal of Environmental Radioactivity, v. 6, p. 49–60. investigation of naturally occurring radioactive materials
(NORM) in oil and gas wells in New York State: New York
Blauch, M.E., Myers, R.R., Moore, T.R., Lipinski, B.A., and State Department of Environmental Conservation, 35 p.
Houston, N.A., 2009, Marcellus Shale Post-Frac Flowback + appendices., accessed July 13, 2011, at http://www.dec.
Waters—Where is all the salt coming from and what are the ny.gov/docs/materials_minerals_pdf/normrpt.pdf; Executive
implications?: Society of Petroleum Engineers, SPE Eastern summary only: http://www.dec.ny.gov/chemical/23473.html.
Regional Meeting, September 23–25, 2009, Charleston,
W. Va., 20 p. Harper, J.A., 2008, The Marcellus Shale—An old “new”
gas reservoir in Pennsylvania: Pennsylvania Geological
Breen, K.J., Angelo, C.G., Masters, R.W., and Sedam, A.C., Magazine, v. 38, no. 1, p. 2–13, accessed July 13, 2011,
1985, Chemical and isotopic characteristics of brines from at http://www.dcnr.state.pa.us/topogeo/pub/pageolmag/
three oil- and gas-producing sandstones in eastern Ohio, pageolonline.aspx.
with applications to the geochemical tracing of brine
sources: U.S. Geological Survey Water-Resources Investi- Hayes, T., 2009, Sampling and analysis of water streams asso-
gations Report 84–4314, 58 p., also available at http://pubs. ciated with the development of Marcellus Shale gas, Final
er.usgs.gov/djvu/WRI/wrir_84_4314.djvu. Report, prepared for Marcellus Shale Coalition (formerly
the Marcellus Shale Committee): Gas Technology Institute,
Breit, G.N., 2002, Produced waters database: U.S. Geological 44 p. + appendices. (Report is available from the Pennsylva-
Survey, accessed July 13, 2011, at http://energy.cr.usgs.gov/ nia Department of Environmental Protection.)
prov/prodwat/index.htm.
Hess, C.T., Michel, J., Horton, T.R., Prichard, H.M., and Coni-
Buckwalter, T.F., and Moore, M.E., 2007, Ground-water glio, W.A., 1985, The occurrence of radioactivity in public
resources and the hydrologic effects of petroleum occur- water supplies in the United States: Health Physics, v. 48,
rence and development, Warren County, northwestern p. 553–586.
Pennsylvania: U.S. Geological Survey Scientific Investiga-
tions Report 2006–5263, 86 p., available at http://pubs.usgs. Ivanovich, M., 1992, The phenomenon of radioactivity, in
gov/sir/2006/5263/. Ivanovich, M., and Harmon, R.S., eds., Uranium Series
disequilibrium—Applications to Environmental Problems
Crawley, M.J., 2007, The R book: Hoboken, N.J., John Wiley in Earth Sciences (2d ed.): Oxford, Clarendon Press,
& Sons, Inc., 492 p. chap. 1, p. 1–33.
Dresel, P.E., 1985, Geochemistry of oilfield brines from Klemic, Harry, 1962, Uranium occurrences in sedimentary
western Pennsylvania: University Park, Pennsylvania State rocks of Pennsylvania: Geological Survey Bulletin 1107–D,
University, M.S. thesis, 237 p. p. 243–288.
Dresel, P.E., and Rose, A.W., 2010, Chemistry and origin of Kraemer, T.F., 2005, Radium isotopes in Cayuga Lake, New
oil and gas well brines in western Pennsylvania: Pennsylva- York—Indicators of inflow and mixing processes: Limnol-
nia Geological Survey, Open-File Report OFOG 10–01.0, ogy and Oceanography, v. 50, p. 158–168.
48 p., accessed July 13, 2011, at http://www.dcnr.state.
pa.us/topogeo/pub/openfile/ofog10_01.aspx. Kraemer, T.F., and Reid, D.F., 1984, The occurrence and
behavior of radium in saline formation water of the
Eaton, A.D., Clesceri, L.S., Rice, E.W., Greenberg, A.E., and U.S. Gulf Coast region: Isotope Geoscience, v. 2,
Franson, M.A.H., eds., 2005, Standard methods for the p. 153–174.
examination of water & wastewater (21st ed.): American
Public Health Association, 1368 p. Krieger, H.L., and Whittaker, E.L., 1980, Prescribed proce-
dures for measurement of radioactivity in drinking
Fisher, R.S., 1998, Geologic and geochemical controls on water: U.S. Environmental Protection Agency,
naturally occurring radioactive materials (NORM) in EPA–600/4–80–032, 111 p.
produced water from oil, gas, and geothermal operations:
Environmental Geosciences, v. 5, p. 139–150. Krishnaswami, S., Graustein, W.C., Turekian, K.K., and
Dowd, J.F., 1982, Radium, thorium, and radioactive
isotopes in ground waters—Application to the in situ
determination of adsorption-desorption rate constants
and retardation factors: Water Resources Research, v. 18,
p. 1633–1675.

References Cited 17

Langmuir, Donald, and Herman, J.S., 1980, The mobility of Poth, C.W., 1962, The occurrence of brine in western Penn-
thorium in natural waters at low temperatures: Geochimica sylvania: Pennsylvania Geological Survey, Fourth Series,
et Cosmochimica Acta, v. 44, p. 1753–1766. Bulletin M–47, 53 p.

Langmuir, Donald, and Riese, A.C., 1985, The thermodynamic Pritz, M.E., 2010, Geochemical modeling and analysis of the
properties of radium: Geochimica et Cosmochimica Acta, frac water used in the hydraulic fracturing of the Marcellus
v. 49, p. 1593–1601. Formation, Pennsylvania: Lewisburg, Bucknell University,
B.S. Honors Thesis, 228 p.
Legall, F.D., Barnes, C.R., and MacQueen, R.W., 1981, Ther-
mal maturation, burial history and hotspot development, Rose, A.W., and Korner, L.A., 1979, Radon in natural waters
Paleozoic strata of southern Ontario-Quebec, from conodont as a guide to uranium deposits in Pennsylvania, in Watter-
and acritarch colour alteration studies: Bulletin of Canadian son, J.R., and Theobald, P.K., eds. Proceedings of the Sev-
Petroleum Geology, v. 29, p. 492–539. enth International Geochemical Exploration Symposium:
Golden, Colo., p. 65–75.
Linsalata, P., Morse, R.S., Ford, H., Eisenbud, M., Franca,
E.P., deCastro, M.B., Lobao, N., Sachett, I., and Carlos, M., Rowan, E.L., Engle, M.A., and Kirby, C.S., 2010, Inorganic
1989, An assessment of soil-to-plant concentration ratios for geochemistry of formation waters from Devonian Strata
some natural analogues of the transuranic elements: Health in the Appalachian Basin—Preliminary observations
Physics, v. 56, p. 33–46. from Pennsylvania, New York, and West Virginia [abs.]:
Geological Society of America, Annual meeting,
Milici, R.C., Ryder, R.T., Swezey, C.S., Charpentier, R.R., October 31–November 3, 2010, Paper No. 204-8, Abstracts
Cook, T.A., Crovelli, R.A., Klett, T.R., Pollastro, R.M., and with Programs, v. 42, no. 5, p. 487, accessed July 14,
Schenk, C.J., 2003, Assessment of undiscovered oil and 2011, at http://gsa.confex.com/gsa/2010AM/finalprogram/
gas resources of the Appalachian Basin Province, 2002: abstract_174638.htm.
U.S. Geological Survey Fact Sheet FS 009–03, 2 p., avail-
able at http://pubs.usgs.gov/fs/fs-009-03/FS-009-03-508.pdf. Sanders, L.L., 1991, Geochemistry of formation waters from
the Lower Silurian Clinton Formation (Albion Sandstone),
Moore, W.S., 1984, Radium isotopic measurements using eastern Ohio: AAPG Bulletin, v. 75, p. 1593–1608.
germanium detectors: Nuclear Instruments and Methods in
Physics Research, v. 223, p. 407–411. Schmoker, J.W., 1981, Determination of organic-matter con-
tent of Appalachian Devonian shales from gamma-ray logs:
National Nuclear Data Center, [n.d.], Chart of nuclides data- AAPG Bulletin, v. 65, p. 1285–1298.
base, accessed July 14, 2011, at http://www.nndc.bnl.gov/
chart/. Stout, W.E., Lamborn, R.E., and Schaaf, D., 1932, Brines of
Ohio (Preliminary Report): Geological Survey of Ohio Bul-
New York State Department of Environmental Conservation letin 37, 123 p.
(NYSDEC), 2009, Draft Supplemental Generic Environ-
mental Impact Statement (SGEIS) on the oil, gas, and Sturchio, N.C., Banner, J.L., Binz, C.M., Heraty, L.B., and
solution mining regulatory program (September 2009), Well Musgrove, M., 2001, Radium geochemistry of ground
permit issuance for horizontal drilling and high-volume waters in Paleozoic carbonate aquifers, midcontinent, USA:
hydraulic fracturing to develop the Marcellus Shale and Applied Geochemistry, v. 16, p. 109–122.
other low-permeability gas reservoirs: New York State
Department of Environmental Conservation, Division of Swanson, V.E., 1960, Oil yield and uranium content of black
Mineral Resources, Bureau of Oil and Gas Regulation, shales: U.S. Geological Survey Professional Paper 356–A,
Appendix 13, NYS Marcellus radiological data from 49 p.
production brine, accessed July 14, 2011, Full document: Swanson, V.E., 1961, Geology and geochemistry of uranium
http://www.dec.ny.gov/energy/58440.html. in marine black shales, A review: U.S. Geological Survey
Osborn, S.G., and McIntosh, J.C., 2010, Chemical and isoto- Professional Paper 356–C, 112 p.
pic tracers of the contribution of microbial gas in Devonian Szalay, A., 1964, Cation exchange properties of humic acids
organic-rich shales and reservoir sandstones, northern Appa- and their importance in the geochemical enrichment of
lachian Basin: Applied Geochemistry, v. 25, p. 456–471. UO2++ and other cations: Geochimica et Cosmochimica
Pennsylvania Department of Environmental Protection Acta, v. 28, p. 1605–1614.
(PA DEP), 1992, NORM survey summary, September 1, Thompson, Michael, and Howarth, R.J., 1978, A new
1992; reproduced in IOGA NEWS (Independent Oil and approach to the estimation of analytical precision: Journal
Gas Association of Pennsylvania), April 1995, available at of Geochemical Exploration, v. 9, p. 23–30.
http://www.dep.state.pa.us/dep/deputate/minres/OILGAS/
NORM.pdf.


Tracy, B.L., Prantl, F.A., and Quinn, J.M., 1983, Transfer of Vengosh, A., Hirschfeld, D., Vinson, D., Dwyer, G., Raanan,
226
Ra, 210Pb, and uranium from soil to garden produce— H., Rimawi, O., Al-Zoubi, A., Akkawi, E., Marie, A.,
Assessment of risk: Health Physics, v. 44, p. 469–477. Haquin, G., Zaarur, S., and Ganor, J., 2009, High naturally
occurring radioactivity in fossil groundwater from the
U.S. Environmental Protection Agency, 1976, National Middle East: Environmental Science and Technology, v. 43,
Interim Primary Drinking Water Regulations, U.S. Envi- p. 1769–1775.
ronmental Protection Agency, Office of Water Supply,
EPA/570/9–76/093. Walter, L.M., Stueber, A.M., and Huston, T.J., 1990, Br-Cl-Na
systematics in Illinois basin fluids—Constraints on fluid
U.S. Nuclear Regulatory Commission, [n.d.], Radium-226: origin and evolution: Geology, v. 18, p. 315–318.
U.S. Nuclear Regulatory Commission, accessed July 14,
2011, at http://www.nrc.gov/reading-rm/doc-collections/cfr/
part020/appb/Radium-226.html.

Table 1. Well locations and related information compiled for samples used in this study. The Well/Sample ID column assigns a unique number to each sample;
digits to the right of the decimal (for example, “5.1,” “5.2”) indicate a time series or multiple samples taken from a well on different dates to characterize changes
over time.—Continued
[Sh., shale; Ss., sandstone; Dolo., dolomite; Fm., formation; Gp., Group; L., lower; M., middle; U., upper; undiv., undivided]

Well / Sample Description of Producing Producing Well
State County Township Longitude Latitude
Sample ID collection date sample site formation formation age type
Source: PA DEP (2009–2010)
1 11/18/2009 PA Clinton Chapman –77.56 41.37 Storage tank Marcellus Sh. Devonian, M. Gas
2 11/20/2009 PA Clinton Beech Creek –77.68 41.20 Storage tank Marcellus Sh. Devonian, M. Gas
3 6/1/2009 PA Bradford Burlington –76.60 41.74 Marcellus Sh. Devonian, M. Gas
4 8/24/2009 PA Lycoming Penn –76.63 41.28 Marcellus Sh. Devonian, M. Gas
5.1 3/18/2009 PA Lycoming Penn –76.66 41.27 Marcellus Sh. Devonian, M. Gas
5.2 3/30/2009 PA Lycoming Penn –76.66 41.27 Marcellus Sh. Devonian, M. Gas
6 12/21/2009 PA Tioga Charleston –77.21 41.79 Marcellus Sh. Devonian, M. Gas
7 12/21/2009 PA Tioga Richmond –77.13 41.78 Marcellus Sh. Devonian, M. Gas
8 9/8/2009 PA Centre Burnside –78.05 41.13 Impoundment Marcellus Sh. Devonian, M. Gas
9 1/8/2010 PA Forest Jenks –79.16 41.55 Tank or lined pit Marcellus Sh. Devonian, M. Oil
10 12/30/2009 PA Potter East Fork –77.88 41.61 Marcellus Sh. Devonian, M. Gas
11.1 4/9/2009 PA Washington Cross Creek –80.39 40.26 Marcellus Sh. Devonian, M. Gas
11.2 6/29/2009 PA Washington Cross Creek –80.39 40.26 Marcellus Sh. Devonian, M. Gas
12 12/30/2009 PA Tioga Gainesville –77.56 41.69 Marcellus Sh. Devonian, M. Gas
13 12/30/2009 PA Tioga Gainesville –77.58 41.68 Tuscarora Fm. Silurian, L. Gas
14 1/7/2010 PA Potter West Branch –77.62 41.67 Marcellus Sh. Devonian, M. Gas
15 12/16/2009 PA Clearfield Lawrence –78.45 41.17 Marcellus Sh. Devonian, M. Gas
16 12/22/2009 PA Westmoreland Washington –79.57 40.49 Marcellus Sh. Devonian, M. Gas
18 11/13/2009 PA Westmoreland Bell –79.55 40.51 Marcellus Sh. Devonian, M. Gas
19 9/18/2009 PA Westmoreland Hempfield –79.65 40.28 Marcellus Sh. Devonian, M. Gas
20 7/16/2009 PA Westmoreland Hempfield –79.57 40.50 Marcellus Sh. Devonian, M. Gas
21 7/23/2009 PA Indiana Rayne –79.04 40.75 Marcellus Sh. Devonian, M. Gas
23 8/13/2009 PA Westmoreland Bell –79.54 40.50 Marcellus Sh. Devonian, M. Gas
Table 1 19


Source: NYSDEC (2009)
24 4/1/2009 NY Steuben Avoca –77.41 42.40 Marcellus Sh. Devonian, M. Gas
25 4/1/2009 NY Steuben Avoca –77.42 42.41 Marcellus Sh. Devonian, M. Gas
26 4/2/2009 NY Chenango Oxford –75.61 42.45 Marcellus Sh. Devonian, M. Gas
27.1 10/7/2008 NY Steuben Caton –77.04 42.05 Marcellus Sh. Devonian, M. Gas
27.2 4/1/2009 NY Steuben Caton –77.04 42.05 Marcellus Sh. Devonian, M. Gas
28 10/8/2008 NY Schuyler Orange –77.08 42.28 Marcellus Sh. Devonian, M. Gas
29 4/1/2009 NY Steuben Woodhull –77.44 42.02 Marcellus Sh. Devonian, M. Gas
30 4/1/2009 NY Steuben Troupsburg –77.47 42.02 Marcellus Sh. Devonian, M. Gas
31 4/6/2009 NY Schuyler Dix –76.94 42.34 Marcellus Sh. Devonian, M. Gas
32 4/6/2009 NY Schuyler Dix –76.94 42.34 Marcellus Sh. Devonian, M. Gas
34 4/6/2009 NY Schuyler Reading –76.91 42.44 Marcellus Sh. Devonian, M. Gas
36 NY Cattaraugus –78.66 42.42 Brine tank Medina Gp. Silurian, L. Gas
(rusted)
42 NY Erie –78.47 42.93 Bottom of brine Medina Gp. Silurian, L. Gas
tank
43 NY Genesee –78.46 42.93 Bottom of brine Medina Gp. Silurian, L. Gas
tank
44 NY Genesee –78.46 43.03 Brine tank, Medina Gp. Silurian, L. Gas
subsurface

45 NY Genesee –78.45 43.02 Brine tank Medina Gp. Silurian, L. Gas
46 NY Genesee –78.25 43.05 Brine tank Medina Gp. Silurian, L. Gas


Source: NYSDEC (1999)—Continued
subsurface
subsurface
49 NY Genessee –78.30 42.88 Brine tank Medina Gp. Silurian, L. Gas
50 NY Wyoming –78.10 42.82 Brine tank Theresa Ss. Cambrian, U. Gas
(rusted)
52 NY Wyoming –78.12 42.82 Brine tank Medina Gp. Silurian, L. Gas
54 NY Wyoming –78.36 42.74 Brine tank Medina Gp. Silurian, L. Gas
55 NY Cayuga –76.64 42.91 Spigot, base of Queenston Sh. Ordovician, U. Gas
brine tank
56 NY Cayuga –76.65 42.91 Spigot, base of Queenston Sh. Ordovician, U. Gas
brine tank
57 NY Seneca –76.86 42.84 Spigot, base of Queenston Sh. Ordovician, U. Gas
brine tank
58 NY Seneca –76.85 42.84 Spigot, base of Queenston Sh. Ordovician, U. Gas
brine tank
59 NY Seneca –76.90 42.78 Spigot, base of Rochester Sh. Silurian, L. Gas
brine tank
60 NY Genesee –77.98 42.93 Brine tank Medina Gp. Silurian, U. Gas
61 NY Livingston –77.91 42.94 Brine tank Medina Gp. Silurian, L. Gas
62 NY Ontario –77.54 42.81 Brine tank Medina Gp. Silurian, L. Gas
63 NY Chautauqua –79.52 42.38 Bottom of brine Medina Gp. Silurian, L. Gas
tank
64 NY Chautauqua –79.72 42.27 Bottom of brine Medina Gp. Silurian, L. Gas
tank
65 NY Chautauqua –79.58 42.16 Brine tank Medina Gp. Silurian, L. Gas
Table 1 21


66 NY Chautauqua –79.54 42.08 Brine tank Bass Islands Silurian, U. Oil
Dolo.
68 NY Chautauqua –79.47 42.13 Brine tank Bass Islands Silurian, U. Oil
Dolo.
69 NY Chautauqua –79.42 42.16 Brine drain tank Bass Islands Silurian, U. Oil
Dolo.
70 NY Erie –78.89 42.54 Brine tank Onondaga Ls. Devonian, M. Gas
73 NY Chautauqua –79.19 42.31 Bottom of stock Devonian, U., Devonian, U. Gas–
tank undiv. Oil
74 NY Chautauqua –79.24 42.23 Bottom of stock Bass Islands Silurian, U. Oil
tank Dolo.
75 NY Chautauqua –79.36 42.17 Brine tank Bass Islands Silurian, U. Gas–
Dolo. Oil
79 NY Erie –78.79 42.56 Brine tank Medina Gp. Silurian, L. Gas
80 NY Allegany –77.92 42.01 Brine tank Oriskany Ss. Devonian, L. Gas
81 NY Allegany –77.95 42.01 Brine tank Oriskany Ss. Devonian, L. Gas
82 NY Tioga –76.26 42.03 Brine tank Helderberg Ls. Devonian, L. Gas
83 NY Tioga –76.31 42.00 Brine tank Helderberg Ls. Devonian, L. Gas


Source: PA DEP (1992)
84 PA Allegheny S. Fayette –80.21 40.36 Diked area Gas
85 PA Armstrong Cowanshannock –79.26 40.79 Tank Devonian, U. Gas
86 PA Armstrong Cowanshannock –79.25 40.78 Tank Catskill/Lock Devonian, U. Gas
Haven Fms.
87 PA Cambria Susquehanna –78.76 40.70 Tank Venango Gp. Devonian, U. Gas
88 PA Cambria Barr –78.83 40.66 Tank Lock Haven Devonian, U. Gas
Fm.
89 PA Centre Curtin –77.77 41.11 Tank Catskill/Lock Devonian, U. Gas
Haven Fms.
90 PA Centre Burnside –77.87 41.12 Tank Catskill/Lock Devonian, U. Gas
Haven Fms.
91 PA Clearfield Jordan –78.61 40.83 Tank Catskill/Lock Devonian, U. Gas
Haven Fms.
92 PA Clearfield Burnside –78.72 40.84 Tank Lock Haven Devonian, U. Gas
Fm.
93 PA Clinton Beech Creek –77.70 41.17 Tank Lock Haven Devonian, U. Gas
Fm.
94 PA Clinton Beech Creek –77.74 41.15 Tank Medina Gp. Silurian, L. Gas
95 PA Crawford Beaver –80.41 41.81 Tank Medina Gp. Silurian, L. Gas
96 PA Elk Highland –78.83 41.54 Separator tank Bradford Gp. Devonian, U. Oil
97 PA Elk Highland –78.93 41.51 Tank Devonian Oil
98 PA Erie Millcreek –80.17 42.08 Tank Medina Gp. Silurian, L. Gas
99 PA Erie Conneaut –80.44 41.89 Tank Huntersville Devonian, M. Gas
Chert
100 PA Fayette Springfield –79.36 39.97 Tank Oriskany Ss. Devonian, L. Gas
101 PA Fayette Springfield –79.40 39.96 Tank Huntersville Devonian, M. Gas
Chert
102 PA Forest Howe –79.14 41.56 Tank battery Warren sand Devonian, U. Oil
103 PA Forest Kingsley –79.29 41.58 Tank Devonian, U., Devonian, U. Oil
undiv.
Table 1 23

104 PA Indiana Cherryhill –78.82 40.74 Tank Kane sand Devonian, U. Gas


Source: PA DEP (1992)—Continued
105 PA Indiana Burrell –79.17 40.48 Tank Fifty Foot sand Devonian, U. Gas
106 PA Indiana White –79.19 40.65 Tank Gas
107 PA Jefferson Bell –78.90 40.97 Tank Devonian, U., Devonian, U. Gas
undiv.
108 PA McKean Wetmore –78.87 41.63 Separator pit Unknown Oil
109 PA McKean Lafayette –78.72 41.83 Pit Oriskany Ss. Devonian, L. Oil
110 PA Somerset Middlecreek –78.92 40.06 Tank Oriskany Ss. Devonian, L. Gas
111 PA Somerset Lincoln –79.07 40.08 Tank Huntersville Devonian, M. Gas
Chert
112 PA Tioga Union –76.96 41.57 Drill pit Ordovician Gas
113 PA Venango Cornplanter –79.59 41.48 Separator Red Valley sand Devonian, U. Oil
114 PA Venango Allegheny –79.55 41.57 Pit Venango Gp. Devonian, U. Oil
115 PA Warren Pleasant –79.19 41.81 Pit Medina Gp. Silurian, L. Oil
116 PA Warren Southwest –79.57 41.63 Tank Medina Gp. Silurian, L. Gas
117 PA Warren Watson –79.25 41.76 Pit Devonian, U., Devonian, U. Oil
undiv.
118 PA Washington Cecil –80.24 40.33 Separator Devonian, U. Oil
119 PA Westmoreland Washington –79.58 40.49 Tank Venango Gp. Devonian, U. Gas
120 PA Westmoreland Hempfield –79.53 40.26 Tank Venango Gp. Devonian, U. Gas
Source: Dresel and Rose (2010)
121 1982 PA Indiana Banks –78.85 40.87 Wellhead Devonian, U., Devonian, U. Gas
undiv.
122 1982 PA Indiana South Mahoning –79.14 40.78 Wellhead Devonian, U., Devonian, U. Gas
undiv.
123 1982 PA Warren Pleasant –79.21 41.82 Wellhead Glade sand Devonian, U. Oil
124 1982 PA Crawford Fairfield –80.14 41.49 Wellhead Medina Gp. Silurian, L. Gas
125 1982 PA Centre Boggs –77.84 41.00 Separator Tuscarora Fm. Silurian, L. Gas

126 1982 PA Somerset Black –79.11 39.93 Separator Ridgeley Ss. Devonian, L. Gas


Source: Pritz (2010), this study
127 04/09 PA Bradford Marcellus Sh. Devonian, M. Gas
Source: This study
132.1 12/8/2010 PA Greene –80.05 39.88 Separator Marcellus Sh. Devonian, M. Gas
Table 1 25

Table 2. Ra-226, Ra-228, gross alpha, and gross beta activities measured in samples of produced water for wells listed in table 1. Analytical uncertainties are included
when known.—Continued
[TDS, total dissolved solids; mg/L, milligram per liter; pCi/L, picocurie per liter; ND, not detected]

Well / Gross Gross Total
TDS Ra-226 Ra-228 Ra-228/
Sample alpha +/– beta +/– +/– +/– radium Method, method codes
(mg/L) (pCi/L) (pCi/L) Ra-226
ID (pCi/L) (pCi/L) (pCi/L)
Source: PA DEP (2009–2010)
1 54,000 436 32.2 121 8.2 556 0.28 SM2540C; EPA904.0, 903.0
2 16,200 14 2 1,322 86 ND 1.8 ND 0.3 SM2540C, 7110C; EPA 900.0, 903.0, 904.0
3 333,000 19,220 2,843 7,944 1,320 50 1.3 37 3.3 87 0.73 SM2540C; EPA 900.0 903.0, 904.0
4 61,800 6,159 743 1,325 190 430 11.0 51 8.9 482 0.12 SM2540C; EPA 900.0, 903.0, 904.0
5.1 38,200 454 126 149 78 66 4.0 2.2 0.9 68 0.03 SM2540C; EPA 900.0, 903.0, 904.0
5.2 82,600 1,644 371 745 242 239 9.7 38 6.3 277 0.16 SM2540C; EPA 900.0, 903.0, 904.0
6 40,880 7,512 750 732 16,920 3,283 1,125 227 18,045 0.07 EPA 903.1, 904.0
7 21,960 4,074 980 757 11,120 2,204 1,287 261 12,407 0.12 EPA 903.1, 904.0
8 124,000 1,525 110 657 76 2,182 0.43 SM18 2540C; EPA 901.1 Mod.
9 284,000 11,810 2,482 1,060 759 4,184 789 1,074 202 5,258 0.26 SM20 2540C; EPA 903.1, 904.0
10 157,000 7,330 460 1,180 180 8,510 0.16 SM18 2540C; EPA 901.1 Mod.
11.1 157,000 951 86 703 69 1,654 0.74 SM18 2540C; EPA 901.1 Mod.
11.2 200,000 1,280 130 1,110 120 2,390 0.87 SM18 2540C; EPA 901.1 Mod.
12 183,000 7,530 1,141 2,683 372 562 26 648 67 1,210 1.15 SM18 2540C; EPA 900.0, 903.0, 904.0
13 358,000 10,356 2,186 11,595 723 892 32 2,589 128 3,481 2.90 SM18 2540C; EPA 900.0, 903.0, 904.0
14 1,470 ND 3 78 4 ND 0.31 ND 0.39 1.00 SM2540C; EPA 900.0, 903.0, 904.0
15 288,900 19,240 7,049 1,268 106 1,374 0.08 SM2540C
16 24,700 318 453 340 590 103 24 168 32 271 1.63 SM2540C; EPA 900.0Mod., 903.1, 904.0
17 88,500 3,640 1,004 ND 631 1,042 197 298 59 1,340 0.29 SM2540C; EPA 900.0Mod., 903.1, 904.0
18 116,000 2,320 800 2,077 929 1,037 200 515 97 1,552 0.50 SM2540C; EPA 900.0Mod., 903.1, 904.0
19 32,500 733 175 81 61 554 104 5.5 1.9 559 0.01 SM2540C; EPA 900.0Mod., 903.1, 904.0
20 45,400 845 213 379 116 66 4.05 1.4 0.3 67 0.02 SM2540C; EPA 900.0, 903.0, 904.0
21 46,460 820 249 505 140 76 2.7 23 2.4 99 0.30 SM2540C; EPA 900.0, 903.0, 904.0
22 47,800 585 163 536 83 36 1.75 2.7 0.2 39 0.08 SM2540C; EPA 900.0, 903.0, 904.0
23 125,100 2,103 631 1,574 335 229 6.8 56 6.5 285 0.25 SM2540C; EPA 900.0, 903.0, 904.0


TDS Ra-226 Ra-228 Ra-228/
24 70 48 7 54 0.163 0.20 0.029 0.22 0.192 0.175
25 54.6 37 59 58 0.195 0.16 0.428 0.34 0.623 2.195
26 3,914 813 715 202 1,779 343 201 39 1,980 0.113
27.1 17,940 8,634 4,765 3,829 2,472 484 874 174 3,346 0.354
27.2 3,968 1,102 618 599 7,885 1,568 234 51 8,119 0.030
28 206,446 14,530 3,792 4,561 1,634 2,647 494 782 157 3,429 0.295
29 9,426 2,065 2,780 879 4,049 807 826 160 4,875 0.204
30 7,974 1,800 1,627 736 5,352 1,051 138 37 5,490 0.026
31 10,970 2,363 1,170 701 6,125 1,225 516 99 6,641 0.084
32 20,750 4,117 2,389 861 10,160 2,026 1,252 237 11,412 0.123
33 205,102 18,330 3,694 ND 654 13,510 2,655 929 179 14,439 0.069
34 16,550 3,355 1,323 711 15,140 2,989 957 181 16,097 0.063
35 123,000 23,480 12,000 2,903 16,030 2,995 912 177 16,942 0.057
36 669 88 1,100 250 1,769 1.644 γ-spectrometry
37 402 68 402 γ-spectrometry
38 1,164 93 429 27 1,593 0.369 γ-spectrometry
39 398 64 234 182 632 0.588 γ-spectrometry
42 413 61 856 222 1,269 2.073 γ-spectrometry
43 260 43 703 194 963 2.704 γ-spectrometry
45 169 86 565 254 734 3.343 γ-spectrometry
46 306 126 568 248 874 1.856 γ-spectrometry
47 175 100 255 179 430 1.457 γ-spectrometry
49 290 50 460 172 750 1.586 γ-spectrometry
Table 2 27


TDS Ra-226 Ra-228 Ra-228/
50 764 81 433 242 1,197 0.567 γ-spectrometry
51 450 66 326 319 776 0.724 γ-spectrometry
52 477 65 651 306 1,128 1.365 γ-spectrometry
53 708 71 350 297 1,058 0.494 γ-spectrometry
54 238 117 269 52 507 1.130 γ-spectrometry
55 1,240 100 1,290 130 2,530 1.040 γ-spectrometry
56 823 69 1,333 450 2,156 1.620 γ-spectrometry
57 557 67 933 230 1,490 1.675 γ-spectrometry
58 465 65 977 230 1,442 2.101 γ-spectrometry
59 ND γ-spectrometry
60 369 61 890 227 1,259 2.412 γ-spectrometry
61 538 72 625 207 1,163 1.162 γ-spectrometry
63 187 20 80 29 267 0.427 γ-spectrometry
64 324 36 503 30 827 1.552 γ-spectrometry
65 444 47 1,690 55 2,134 3.806 γ-spectrometry
66 1,550 110 319 60 1,869 0.206 γ-spectrometry
67 366 45 1,660 47 2,026 4.536 γ-spectrometry
68 654 36 1,440 50 2,094 2.202 γ-spectrometry
69 1,040 40 387 32 1,427 0.372 γ-spectrometry
70 64.2 27.0 83 12 147 1.294 γ-spectrometry
71 148 17 100 23 248 0.676 γ-spectrometry
72 160 35 574 31 734 3.588 γ-spectrometry
73 185 20 86 25 271 0.467 γ-spectrometry
74 953 64 444 45 1,397 0.466 γ-spectrometry
76 951 109 1,500 170 2,451 1.577 γ-spectrometry

77 484 32 1,420 50 1,904 2.934 γ-spectrometry
78 156 35 504 36 660 3.231 γ-spectrometry


TDS Ra-226 Ra-228 Ra-228/
79 111 35 60 23 170 0.538 γ-spectrometry
80 901 52 250 42 1,151 0.278 γ-spectrometry
81 691 26 154 30 845 0.223 γ-spectrometry
82 3,760 100 1,110 60 4,870 0.295 γ-spectrometry
83 1,620 110 1,790 60 3,410 1.105 γ-spectrometry
Source: PA DEP (1992)
84 143,432 512 202 714 0.395
85 315 165 480 0.524
86 20 13 33 0.650
87 175,296 1,408 904 2,312 0.642
88 195,404 1,154 1,083 2,237 0.938
89 222,672 163 126 289 0.773
90 252,980 1,489 636 2,125 0.427
91 185,146 2,015 1,749 3,764 0.868
92 230,924 107 77 184 0.720
93 194,902 731 491 1,222 0.672
94 153,096 811 530 1,342 0.653
95 396,012 599 1,683 2,282 2.810
96 31,502 15 18 33 1.150
97 25,159 23 26 49 1.143
98 390,928 628 1,478 2,106 2.355
99 378,148 588 1,483 2,072 2.521
100 341,918 4,685 2,038 6,723 0.435
101 354,034 566 2,110 2,676 3.728
102 130,588 42 42 84 1.000
103 86,988 39 56 95 1.427
104 186,736 2,019 2,196 4,215 1.088
105 198,668 2,575 1,866 4,441 0.725
106 121,928 450 313 763 0.696
Table 2 29


TDS Ra-226 Ra-228 Ra-228/
Source: PA DEP (1992)—Continued
107 1,280 848 2,128 0.663
108 36,470 8 12 20 1.446
109 185 184 370 0.995
110 176,676 203 1,543 1,746 7.601
111 182,274 1,988 499 2,487 0.251
112 1,137 1,457 2,594 1.281
113 114,208 12 30 42 2.500
114 36,282 54 77 131 1.426
115 59,554 34 27 61 0.794
116 395,440 795 968 1,763 1.219
117 125,264 275 187 462 0.680
118 134,164 255 456 711 1.788
119 221,134 170 46 216 0.271
120 402,148 857 71 928 0.083
Source: Dresel and Rose (2010)
121 253,000 1,900 1,900
122 244,000 200 200
123 91,000 0 0.0
124 257,000 500 500
125 259,000 5,300 5,300
126 302,000 5,000 5,000
Source: Pritz (2010), this study
127 122,527 2,653 11 318 22 2,971 0.120 γ-spectrometry
128 250,112 3,082 21 935 27 4,018 0.303 γ-spectrometry
129 134,880 1,958 26 572 37 2,530 0.292 γ-spectrometry
130 222,681 1,486 13 472 23 1,957 0.317 γ-spectrometry

131 117,259 1,756 6 377 10 2,133 0.215 γ-spectrometry


TDS Ra-226 Ra-228 Ra-228/
Source: This study
132.1 1,312 19 300 23 1,612 0.229 γ-spectrometry
132.2 3,363 21 542 19 3,905 0.161 γ-spectrometry
132.3 4,372 39 577 19 4,949 0.132 γ-spectrometry
132.4 4,892 38 599 29 5,491 0.122 γ-spectrometry
132.5 5,110 18 626 20 5,736 0.123 γ-spectrometry
132.6 5,210 31 614 32 5,824 0.118 γ-spectrometry
132.7 3,105 26 686 30 3,791 0.221 γ-spectrometry
132.8 5,446 28 820 17 6,266 0.151 γ-spectrometry
132.9 5,272 88 846 33 6,118 0.160 γ-spectrometry

Edit and layout by Kay P. Naugle
Raleigh Publishing Service Center

Graphics by Francisco A. Maldonado
Manuscript approved on August 4, 2011.

Prepared by the USGS Science Publishing Network,
Table 2 31

Rowan and others—Radium Content of Oil- and Gas-Field Produced Waters in the Northern Appalachian Basin—Scientific Investigations Report 2011–5135

USGS Study of Marcellus Shale Wastewater Radioactivity Levels

More Related Content

Viewers also liked (20)

Similar to USGS Study of Marcellus Shale Wastewater Radioactivity Levels (20)

More from Marcellus Drilling News (20)

Recently uploaded (20)

USGS Study of Marcellus Shale Wastewater Radioactivity Levels