As you read the article, answer the following questions.

Description

As you read the article, answer the following questions. You could find both the article and the questions in the attachments.

2 attachmentsSlide 1 of 2attachment_1attachment_1attachment_2attachment_2

Unformatted Attachment Preview

Unconventional natural gas
resources in Pennsylvania:
The backstory of the modern
Marcellus Shale play
AUTHORS
Kristin M. Carter Pennsylvania Geological
Survey, 400 Waterfront Drive, Pittsburgh, Pennsylvania 15222; krcarter@pa.gov
Kristin M. Carter, John A. Harper, Katherine W. Schmid,
and Jaime Kostelnik
ABSTRACT
Pennsylvania is not only the birthplace of the modern petroleum industry but also the focus of the modern Marcellus Shale gas play. For
more than 150 yr, Pennsylvania has experienced a rich history of oil
and gas exploration and production, witnessed the advent of modern
petroleum regulations, and now sits deep in the heart of the largest
domestic shale gas play the United States has ever seen. Although a
known source rock for decades, the Marcellus Shale was not considered a viable gas reservoir until Range Resources Corporation
(Range) discovered the play with its completion of the Renz No. 1
well in Washington County in October 2004. Using horizontal drilling and hydraulic fracturing techniques used by operators working
the Barnett Shale gas play, Range has gone on to complete hundreds
of horizontal shale gas wells in Washington County alone. Other operators have followed suit in counties from one corner of the state to
the other, and as of June 2011, the Commonwealth has issued nearly
6500 Marcellus Shale gas well permits. Based on publicly reported
well completion and production data, an average Marcellus Shale
gas well requires 2.9 million gal of water during the hydraulic fracturing process and produces 1.3 mmcf gas/day. Furthermore, the U.S.
Energy Information Administration has estimated that as of mid2011, daily Marcellus Shale gas production in Pennsylvania exceeds
2.8 bcf. Because of the level of drilling activity and production associated with the Marcellus play, Pennsylvania has become the nexus
of shale gas production and water management issues.
INTRODUCTION AND PURPOSE
Like any notable subject, Pennsylvania’s modern Marcellus Shale
gas play has an interesting and complex backstory. The birthplace
of the modern petroleum industry, Pennsylvania has experienced
Copyright ©2011. The American Association of Petroleum Geologists/Division of Environmental
Geosciences. All rights reserved.
DOI:10.1306/eg.09281111008
Environmental Geosciences, v. 18, no. 4 (December 2011), pp. 217–257
217
Kristin Carter has been a geologist with the
Pennsylvania Geological Survey since 2001 and
currently serves as chief of the Petroleum and
Subsurface Geology Section in Pittsburgh, Pennsylvania. Kristin researches oil, gas, and subsurface
geology for the Commonwealth and also enjoys
petroleum history. Before her employment with
the Survey, Kristin worked for nearly a decade
in the environmental consulting field, where she
investigated matters of groundwater flow and
quality, contaminant fate and transport, and mine
reclamation. Kristin received her M.S. degree in
geological sciences from Lehigh University in 1993
and her B.S. degree in geology and environmental science (double major) from Allegheny
College in 1991.
John A. Harper Pennsylvania Geological
Survey, 400 Waterfront Drive, Pittsburgh, Pennsylvania 15222; jharper@pa.gov
John Harper has been a geologist with the Pennsylvania Geological Survey for more than 34 yr
and currently serves as chief of the Geologic Resources Division. His duties include overseeing
the work of the Survey’s Mineral Resource Analysis and the Petroleum and Subsurface Geology
sections, which have been studying various geologic resources in Pennsylvania, including petroleum geology and engineering, coal resources,
industrial minerals, and both organic and inorganic geochemistry. During his tenure with the
Survey, John has worked on numerous projects
dealing with the Commonwealth’s oil, gas, and
subsurface geology, including Department of
Energy–sponsored studies such as the Eastern
Gas Shales Project, the atlas of major Appalachian
gas plays, A Geologic Playbook for Trenton–
Black River Basin Exploration, and the Midwest
Regional Carbon Sequestration Partnership. He
holds a 1968 B.A. degree in geography and earth
science from Indiana University of Pennsylvania,
a 1972 M.S. degree in geology from the University
of Florida, and a 1977 Ph.D. in paleontology
and paleoecology from the University of Pittsburgh.
Katherine W. Schmid Pennsylvania Geological Survey, 400 Waterfront Drive, Pittsburgh,
Pennsylvania 15222; kschmid@pa.gov
Katherine Schmid joined the Pennsylvania
Geological Survey’s Petroleum and Subsurface
Geology Section in 2010 as a geologic scientist
and conducts oil, gas, and subsurface geology
research for the Commonwealth. Before coming
to the Survey, Katherine was employed in the
oil and gas industry for 5 yr, where her work focused on both organic-rich shale and siliciclastic
reservoirs across the state. Katherine received her
M.S. degree in geological sciences from the University of Pittsburgh in 2005 and her B.S. degree in
geology from the Ohio State University in 1999.
Jaime Kostelnik Weatherford Laboratories,
16161 Table Mountain Parkway, Golden, Colorado
80403; jaime.kostelnik@weatherfordlabs.com
Jaime Kostelnik joined the Pennsylvania Geological Survey in 2002 and, until July 2011, served
as a senior geologist for the Survey’s Petroleum
and Subsurface Geology Section. During her tenure with the Commonwealth, her research activities included Appalachian Basin petrology and
petrography, sedimentary geology, and organic
geochemistry. Jaime recently accepted a position
with Weatherford Laboratories in Golden, Colorado, as a senior geologist. She holds an M.S.
degree in geology from Wright State University
(2001) and a B.S. degree in geology from Juniata
College (1999).
ACKNOWLEDGEMENTS
We thank each of our technical reviewers, whose
experience and insightful reviews greatly improved the content of this article. In addition, we
extend our gratitude to John E. Egers, Jr., who
provided critical input to the legal issues discussed
herein; to Stuart O. Reese, who contributed valuable information, statistics, and maps regarding
private water supplies in Pennsylvania; and to
Austin Mitchell, Ph.D. candidate at Carnegie Mellon
University’s Department of Engineering & Public
Policy, who assisted greatly with our analysis of
reported hydraulic fracturing water usage for
completed Marcellus wells. Last but not the least,
we thank the Pennsylvania Department of Conservation and Natural Resources executive staff,
including our state geologist, George E. W. Love,
for supporting our research endeavors in the field
of petroleum geology.
EDITOR’S NOTE
This manuscript was reviewed by six reviewers
and accepted for publication by James A.
Drahovzal.
218
a long history of oil and gas exploration and production, much of
which predates modern regulation. To be sure, Pennsylvania has
been the place of many firsts and important industry developments
with respect to both oil and gas. For the purpose of this article, however, we have chosen to focus on discoveries and advancements because they mainly relate to gas drilling and primary production, with
the intent of demystifying the Marcellus Shale gas play.
The backstory of the modern Marcellus Shale gas play intertwines science and perception, law and regulation, and begs to be
told. Accordingly, this article highlights critical issues regarding
the history, geology, water issues, and regulation of Pennsylvania’s
oil and gas industry, in general, and the Marcellus Shale play, in particular. The intent of this approach is to give readers a better appreciation for improvements in technology, evolution of geologic
understanding, and changes in regulatory oversight—all with a view
toward how these are being applied today to shale gas resources in
the Commonwealth.
HISTORY OF PETROLEUM DRILLING IN PENNSYLVANIA
Pennsylvania is no stranger to oil and gas. Centuries ago, Seneca Indians gathered oil emanating from seeps along Oil Creek in Venango
County by installing crude dams along the stream bank using blankets, bark, or cloths to skim oil from the top of the water column
(Carll, 1887; Giddens, 1947; Carter and Flaherty, 2011). The early
settlers of northwestern Pennsylvania saw how the Seneca made
good use of the oil—as a medicine, an illuminant, and more—and
found value in this resource as well. It was not until 1768, however,
that the first written account of petroleum in Pennsylvania was prepared by David Zeisberger, a Moravian missionary, who observed oil
at the confluence of Tionesta Creek and the Allegheny River in Forest County (Giddens, 1947; Carter and Sager, 2010). In fact, it was
the oil seeps found throughout the Oil Creek Valley that attracted
many investors to investigate this potential resource.
Early Oil
Perhaps the most noteworthy investment in the Oil Creek Valley
during the 1850s was made by the Seneca Rock Oil Company, a
group of Connecticut-based investors that included Dr. Francis H.
Brewer, George H. Bissell, Benjamin Silliman, Jr., and James M.
Townsend. The company leased land called the Hibbard farm in Titusville and subsequently retained “Colonel” Edwin Laurentine Drake
to visit the site in April 1858 (Carter, 2009; Harper, 2009). No doubt
Drake observed the natural oil seeps along Oil Creek and saw the
shallow pits dug by Indians to gather the product, but his particular
charge, as conceived by Bissell, was to drill a well to extract oil from
the ground (Carll, 1887; Giddens, 1947; Brice, 2009). In subsequent
months, Drake investigated the drilling techniques of salt-well drillers
The Backstory of the Modern Marcellus Shale Play
Figure 1. Map of Pennsylvania illustrating the location of notable wells and other relevant features discussed in this article. EGSP =
Eastern Gas Shales Project.
operating in the Tarentum area north of Pittsburgh; procured a steam engine and other necessary equipment;
and after many efforts to identify a competent, dependable driller, eventually retained William A. “Uncle Billy”
Smith. Drilling commenced in June 1859 but encountered several obstacles along the way, such as slow drilling
rates, shallow groundwater, and poor borehole integrity. Eventually, Drake struck oil in Titusville (Figure 1)
on August 27, 1859, at a depth of 69.5 ft (21.2 m)
(Giddens, 1947). Drake’s well produced about 40 bbl
oil/day initially, but by October 1865, the well was producing only 5 bbl oil/day (Boyle, 1898).
And so with some ingenuity, hard work, and luck,
Drake inaugurated the modern petroleum industry by
successfully demonstrating the application of salt-well
drilling techniques to the extraction of oil from subsur-
face reservoirs. Soon after the completion of Drake’s
well, an oil boom ensued throughout northwestern
Pennsylvania. Several farms were lease for petroleum
drilling and many wells, such as the Crossley, McClintock No. 1, Fountain, Woodford, Sherman, and Noble,
became famous as they spewed black gold, producing
volumes of tens to thousands of barrels of oil per day
(Carter, 2009).
Early Gas
Even before Drake’s efforts in Titusville, the other
form of petroleum, natural gas, was being explored in
northeastern United States. Known as the first commercial natural gas well in the world, the well hand
dug by William Hart in Fredonia, New York, in 1821
Carter et al.
219
was completed to a depth of 27 ft (8 m) using a pick and a
shovel. By 1825, it supplied several users in the area with
enough gas for lighting purposes. In 1850, Hart’s well
was deepened to 50 ft (15 m), at which time it was piped
to Fredonia’s Main Street to light not only the street
lights but also the stores and local hotel (Henry, 1873;
Brice, 2009; Harper and Kostelnik, 2010). Of course,
after the Drake well, both oil and gas wells were drilled.
The earliest shale gas wells of record in Pennsylvania were drilled in 1860 in Erie County (Figure 1)
(Ashley and Robinson, 1922). These wells, drilled near
the Lake Erie shoreline, produced only minor amounts
of gas from the Upper Devonian Ohio Shale, yet were
enough to spawn further drilling. Between 1860 and
1878, three gas pools were identified: Erie, Fairview,
and Northeast (Ashley and Robinson, 1922). Although
no specific well records could be located for these early
wells, White (1881, p. 120) provided some level of
detail:
“The gas and oil wells at Erie vary in depth from
450′ to 1200′. Deming’s planning mill well got gas at
453′ (i.e., in Portage layers 650± from the top down).
The average depth of gas flow is about 600′…”
By all accounts, the first commercial gas well
drilled in Pennsylvania was the Newton Well (Figure 1)
in Oil Creek Township, Crawford County. This well was
completed to a depth of 786 ft (240 m) on May 11,
1872, and produced from the Upper Devonian Venango
Group (Third Sand of drillers) (Henry, 1873). From the
Pennsylvania Geological Survey (the Survey)’s historical well card file:
“The flow of gas from this well when first struck
has been estimated at 5,000,000 ft3/day. In 1877, an
attempt to make an accurate measurement of it was
made by means of a geometer prepared for the purpose, but the volume of gas was so great that the effort
failed. Shortly after the well was struck, pipes were laid
to Titusville, and the gas was introduced into many
dwellings and used by refiners and others for heating
purposes. It is still used in this manner as far as the well
is able to supply them. The flow has gradually decreased from the start and is now (March 1877) comparatively small.”
Indeed, the Newton Well made headlines for both
its open flow and the fact that it necessitated a 5.5-mi
(8.9-km) pipeline to Titusville. Noted by John Carll
(1887, p. 600) as “the first natural gas plant of the kind
in the country” in his annual report on the oil and gas
region of Pennsylvania, the Newton Well and pipeline
furnished heat and light to approximately 250 residents
220
The Backstory of the Modern Marcellus Shale Play
and local industry. Subsequent Survey reports also referred to the Newton Well’s commercial-scale production, but because no detailed records were kept on
Pennsylvania oil and gas production until 1882, the actual productivity of the Newton Well cannot be quantified (Sisler et al., 1933).
The famous Haymaker No. 1 well (Figure 1) was
completed on November 3, 1878, on the Remaley
farm in Murrysville, Westmoreland County (Johnson,
1925), and was reported by Carll (1887, p. 673) as a
flowing gas well that was producing “with unabated
volume.” The well produced gas at a depth of 1320 ft
(402 m) from the Upper Devonian Murrysville Sandstone, discovering the gas field of the same name (Carll,
1887). Based on the success of this well, several others
were drilled in the same area, along the Murrysville anticline, and recorded initial open flows of several million
cubic feet of gas (Johnson, 1925). The Haymaker well
has been credited with the success of electric light utilities (Westinghouse and others) and many glassmaking
factories throughout southwestern Pennsylvania.
Another gas producer of comparable notoriety was
the Speechley Well (Figure 1) in Pinegrove Township,
Venango County. This well was drilled on the Samuel
Speechley farm at Coal Hill in 1885 and is identified on
historical maps as United Natural Gas Company Well
No. 157 (Dickey et al., 1943; Dickey and Matteson,
1945) and in Pennsylvania’s current-day Pennsylvania
Internet Record Imaging System/Wells Information
System (PA*IRIS/WIS) as Permit No. 121-01747. No
production records for the Speechley Well are available
because the well was never gauged, but Dickey et al.
(1943) reported that the well’s open flow was very
large and that it supplied enough gas for all of Oil City.
The notable wells discussed above are only a few of
the many thousands of gas wells that have produced
from the Upper Devonian during the past 100 yr.
Pennsylvania has produced more than 12 tcf of gas during the last century, most of which (∼10 tcf) was produced from these shallow reservoirs (Pennsylvania
Internet Record Imaging System/Wells Information
System, 2011).
Twentieth Century and Beyond
The extent and scope of Pennsylvania’s petroleum exploration continued to grow in the 20th century. At the
height of the Great Depression in the 1930s, operators
were drilling deep wells in north-central Pennsylvania
to extract gas (likely sourced by the Marcellus Shale)
from the Lower Devonian Oriskany Sandstone. The
first Oriskany gas well, the North Penn Gas Company
No. 1143 (Figure 1), was completed on September 11,
1930, discovering the Tioga gas field in Tioga County
(Kostelnik and Carter, 2009). This activity continued
and expanded throughout western and north-central
Pennsylvania at large for several decades. By 1980, the
Oriskany reservoir was found to produce in different
physiographic settings and under multiple trapping
mechanisms and also served as a very effective gas storage reservoir (Pennsylvania Department of Conservation and Natural Resources, 2009). Of the approximately 1700 Oriskany gas wells drilled in Pennsylvania
to date, more than 1000 are still active as gas producers,
gas storage wells, or observation wells (Kostelnik and
Carter, 2009).
Another frequently drilled siliciclastic gas-producing
reservoir is the Lower Silurian Medina Group of northwestern Pennsylvania and its equivalent Tuscarora Sandstone in central Pennsylvania. Although surrounding
states and Canada discovered and were producing this
particular reservoir since the late 1800s (McCormac
et al., 1996) and Pennsylvania operators first discovered
gas in the Medina in 1947, operators did not produce
the Medina extensively until the mid-1970s (Laughrey,
1984). Many of the Medina wells completed in the late
1970s and early 1980s in Pennsylvania and surrounding
states were related to the Federal Energy Regulatory
Commission’s designation of the Medina Group play
as a “tight” sand in these areas. This enabled producers
to receive tax credits for wells producing from Medina
reservoirs (McCormac et al., 1996). The reservoir characteristics of two particular Medina-producing fields in
Pennsylvania, the Athens and Geneva fields of Crawford
County, were evaluated in detail by Laughrey (1984).
The discovery well for the Athens field, the Alden and
Roberta Post No. 1 (Figure 1), was completed by Columbia Gas Transmission Company on September 20,
1974. The Greenwood pool of Geneva field was discovered by the Kebert Developers No. 1 (Figure 1) and
completed by N-Ren Corporation on November 7,
1975 (Laughrey, 1984). Although not as prolific as it
once was, the Medina Group is still drilled for gas in
Pennsylvania today.
Fast forward to the 21st century, Pennsylvania
found itself in the middle of another prospective deep
gas play. Although recognized as a viable gas play along
the perimeter of the Appalachian Basin since the mid1880s (Wickstrom, 1996), Ordovician carbonate rocks
of the Trenton and Black River formations were not ex-
plored seriously in Pennsylvania until the early 2000s.
Historical Trenton-Black River production occurred
along the Lima-Indiana trend of Ohio and Indiana (westward of the basin) between mid-1885 and the 1930s
( Wickstrom, 1996). The more recent Trenton-Black
River play resulted in the organization of a consortium
of universities, operators, producing companies, and
geologic surveys that prepared a comprehensive basinwide geologic play book for the fractured and sometimes hydrothermally altered reservoir. In Pennsylvania,
Trenton-Black River rocks have been structurally and
hydrothermally altered to varying degrees, depending
on location (Patchen et al., 2006). To date, less than a
dozen gas wells have reported production from the
Trenton-Black River reservoir in Pennsylvania.
The modern Marcellus Shale gas play was established with the completion of the vertically drilled Renz
No. 1 (Figure 2) in Mount Pleasant Township, Washington County, Pennsylvania, by Range Resources Corporation (Range) in 2004 (Pennsylvania Internet Record
Imaging System/Wells Information System, 2011). Originally drilled in 2003 with deeper Devonian and Silurian targets in mind, Range revisited this well in October
2004 to hydraulically fracture (or “frac”) the organicrich parts of the Hamilton Group, which reported very
large gas shows during drilling operations the year before. Range began producing shale gas at the Renz No. 1
in 2005 and subsequently used drilling and hydraulic
fracturing techniques used by Texas operators in the
Barnett Shale gas play to drill hundreds of horizontal
shale gas wells in Washington County alone (Harper
and Kostelnik, 2010; Pennsylvania Internet Record Imaging System/Wells Information System, 2011). The
Marcellus play truly represents an unconventional gas
play in that (1) the organic materials that generated
the natural gas have remained in the shale, (2) the reservoir and seal characteristics are markedly different from
those in a conventional petroleum system, and (3) reservoir porosity and permeability are very low. The terms
“Marcellus Shale” and “Marcellus play,” as referred to in
this article, represent all organic-rich shales in the Middle Devonian Hamilton Group and not strictly to the
Marcellus Formation alone.
This section has provided only a brief overview of
the many oil- and gas-producing reservoirs within Pennsylvania’s borders. Note that although the research
community, industry, and media may periodically focus
on one play or another, the industry continues to develop
these many resources throughout Pennsylvania to the
current day.
Carter et al.
221
Figure 2. Location of the Renz No. 1 well (Mount Pleasant Township, Washington County), the discovery well for the modern Marcellus
Shale gas play.
EVOLUTION OF WELL
STIMULATION TECHNOLOGIES
Before the wide-scale deployment of hydraulic fracturing, Pennsylvania’s shallow oil and gas wells were typically stimulated or “torpedoed” by shooting with nitroglycerin or black powder to break up rock exposed in
the open part of the borehole, thereby creating fractures that enhanced the flow of hydrocarbons (Fettke,
1951, 1952a, 1953, 1954; Lytle, 1965). Because shooting was dangerous and commonly unpredictable, the introduction of hydraulic fracturing was a welcome change,
and the practice of shooting dropped off precipitously
by the late 1950s except in certain areas of eastern Kentucky and southern West Virginia (Moore, 1959).
The practice of hydraulic fracturing to enhance oil
and gas well production dates back to 1949, when Stanolind Oil and Halliburton conducted commercial222
The Backstory of the Modern Marcellus Shale Play
scale hydraulic fracturing at wells in Oklahoma and
Texas (Range Resources LLC, 2010; Petroleum Transfer
Technology Council, 2011). In fact, Halliburton was the
first to patent the so-called “hydrafrac” process, which
gave exclusive rights at the time to pump relatively large
volumes of fluid and sand (proppant) under high pressure into oil and gas wells to stimulate production.
Pumping rates during these hydraulic fracturing jobs increased from a couple barrels of fluid per minute in the
1950s to 20 bbl/min by the 1960s (Petroleum Transfer
Technology Council, 2011). The most commonly used
frac fluids between 1949 and 1954 were crude oil, kerosene, and acid either in their natural state or as a gel, but
by the late 1950s, more than 20% of the hydraulic fracturing jobs in the Appalachian Basin were using water as
the frac fluid (Moore, 1959). Changes in the size of frac
jobs occurred as well, as the petroleum industry observed
how the use of larger volumes injected at higher pressures
Figure 3. A tale of two wells in PunxsutawneyDriftwood field, Clearfield County (see inset
map). Cumulative production as measured in
1963 was as follows: the hydraulically fractured well produced 625 mmcf (6-yr period),
whereas the shot well produced 242 mmcf
(9-yr period). (Modified from Lytle, 1965, and
reprinted with permission of the Interstate
Oil and Gas Compact Commission).
resulted in better oil and/or gas production from the
reservoirs. As a comparison, data gathered at the end
of 1954 showed that the average volume of fluid used
during the hydraulic fracturing process was 2000 to
4500 gal (7571–17,034 L). In just 5 yr, these volumes
had increased an order of magnitude from 12,000 up to
20,000 gal (45,425–75,708 L) in the deep gas-producing
areas of Pennsylvania (Moore, 1959).
The first application of hydraulic fracturing to
deep gas wells in Pennsylvania was performed in December 1953 at the State of Pennsylvania Tract 28
No. 3 well, situated in Punxsutawney-Driftwood field,
Elk County (Figure 1). The well produced 1.1 mmcf
after fracing (Fettke, 1954). By the middle of 1954, hydraulic fracturing was becoming more popular and successful in Pennsylvania (Moore, 1959; Lytle, 1965). Hydraulic fracturing of deep gas reservoirs focused mainly
on the Oriskany Sandstone and Middle Devonian Huntersville Chert and used large volumes of water, up to
20,000 gal (75,708 L) of water containing as much as
20,000 lb (9072 kg) of variously sized sand grains (Moore,
1959). An example of improvements in production
created by hydraulic fracturing is shown in Figure 3
using data from two Oriskany wells in PunxsutawneyDriftwood field, Clearfield County. Even several years
after treatment, the fraced well’s monthly production
was approximately 11 times greater than the shot well.
By 1963, more than 70% of the deep gas wells drilled in
Pennsylvania were stimulated using hydraulic fracturing
techniques (Lytle, 1964).
Since the mid-1950s, hydraulic fracturing became
more frequently applied not just to newly drilled deep
gas wells, but also to new shallow oil and gas wells, producing and service wells associated with secondary oil
recovery, and even gas storage wells (Moore, 1959; Lytle,
1965). Shallow oil and gas wells were typically fraced
with crude oil or kerosene gel because crude was readily
available from other locally producing wells (Moore,
1959). In 1955, fracing was applied to 40% of the shallow gas wells completed in Pennsylvania and was responsible for a 15-fold increase in average open flow (Fettke
and Lytle, 1956). In 1956, 62% of Pennsylvania’s shallow
gas wells were hydraulically fractured, which increased
their average open flow by 23 times (Lytle, 1957). By
1963, 85% of newly drilled shallow gas wells were fraced,
and their corresponding open flows increased as much
as 27 times (Lytle, 1964, 1965) (Figure 4).
The hydraulic fracturing of shallow oil sands, in contrast, was not used until 1963. The Cemetery Lot Well
No. 6, completed in Pleasant Township, Warren County,
produced oil from the Upper Devonian Bradford Group
(Glade sand of drillers) and was the first oil well in primary production to be successfully stimulated via hydraulic fracturing in Pennsylvania (Rough and Eckard,
Carter et al.
223
Figure 4. Summary of hydraulic fracturing
activity (bar graph) and corresponding production volumes (linear plot) in Pennsylvania’s
shallow gas wells during the period of 1955
to 1963 (modified from Lytle, 1965, and
reprinted with permission of the Interstate
Oil and Gas Compact Commission).
1963; Lytle, 1964). The well produced 50 bbl oil/day
after fracing and, after several months, had produced
more than 1600 bbl, giving Warren County a great
boost in oil production in 1963 (Lytle, 1965). Data presented by Rough and Eckard (1963) for these Glade
sand-producing wells indicated that newly completed
wells had higher production rates postfracing than older
wells stimulated using the same technique and that newly
drilled oil wells in this area of Warren County produced
12 times more than those completed using more traditional shooting techniques.
ADVENT OF MODERN OIL AND GAS
REGULATIONS
Introduction
The statutory and regulatory history of the oil and gas
industry in Pennsylvania is not as extensive as its legacy
of drilling and production. The state’s regulatory agency,
the Pennsylvania Department of Environmental Protection (DEP), has estimated that since the completion of
the Drake well in 1859, more than 350,000 oil and gas
wells have been drilled in the Commonwealth (Depart224
The Backstory of the Modern Marcellus Shale Play
ment of Environmental Protection, 2009a), of which
the state has some form of record for approximately
180,000 (Pennsylvania Internet Record Imaging System/Wells Information System, 2011). Countless statutes and regulations have been enacted during this time
frame, but only a small number pertain to the exploration
and development of petroleum resources. This section
offers a historical synopsis of oil and gas laws and regulations from the late 1800s through the current day, noting
those that have particular relevance to the development
of Marcellus Shale gas resources. It is by no means intended to provide a comprehensive discussion of the legal
or regulatory issues associated with the exploration and
development of oil and gas resources in Pennsylvania.
As a preface to this discussion, the following definition of terms is warranted. The body of documentation
used to regulate the oil and gas industry in Pennsylvania is
composed of both statutes and regulations. “Statutes” are
laws enacted by the Pennsylvania legislature. In contrast,
“regulations” are certain rules written by those working
with the state executive branch, sometimes in conjunction
and collaboration with industry and the public, and are
adopted into the Pennsylvania Code to enforce statutes
in more detail. Ultimately, all statutes and regulations
are consistent with the Pennsylvania State Constitution.
The Industry’s First Century in Pennsylvania
The earliest oversight of Pennsylvania’s petroleum industry came in May 1891, at which time the state legislature passed two statutes; one relating to the plugging
of abandoned wells and the other granting right of action to those who provided labor and/or materials for
oil and gas well installation or production (Burcat,
2008; U.S. Department of Energy, 2009). These statutes were followed by four more in 1921, which related
to well owner rights, well-plugging methods, and penalties for mismanagement of oil and gas wells (Burcat,
2008). Although these particular laws are roughly a century old, they are still in effect in Pennsylvania.
Originally enacted in June 1937, Pennsylvania Act
394, the Clean Streams Law, is another important statute that relates to petroleum development. The Clean
Streams Law has been amended several times, as recently as October 2006, and has as its goal not only to
protect but also enhance the quantity and quality of
water available to Pennsylvanians (Department of Environmental Protection, 2011d).
The next interjection of Commonwealth statutes
occurred during the mid-20th century. On November
30, 1955, the Pennsylvania legislature passed Act 225,
the Gas Operations, Well-Drilling, Petroleum and Coal
Mining Act. The specific purpose of this statute was to
address standards for drilling oil and gas wells in areas
where coal mining was occurring (Department of Environmental Protection, 2001). Nearly 6 yr later, in July
1961, Pennsylvania Act 359, known as the Oil and Gas
Conservation Law, was passed. The intent of this statute was to encourage efficient means of petroleum exploration and development while protecting the rights
of both producers and subsurface rights owners (Burcat,
2008; Department of Environmental Protection, 2011c).
In particular, Act 359 applies only to those oil or gas
wells that have penetrated to the top of the Middle Devonian Onondaga Limestone (or its stratigraphic equivalent) and have been drilled to depths in excess of 3800 ft
(1158 m). The act established permitting, well spacing,
and unitization requirements for such wells (Department of Environmental Protection, 2011c).
Modern and Current Regulation
In December 1984, two additional statutes were passed
by the Pennsylvania legislature, the Coal and Gas Resource Coordination Act and the Oil and Gas Act.
The first of these, the Coal and Gas Resource Coordina-
tion Act (Pennsylvania Act 214), is very specific in
scope; it applies only to those instances where gas wells
may be drilled in areas where workable coal is present
and intends to minimize legal disagreements between
the subsurface rights owners where both resources will
be extracted (Burcat, 2008; Department of Environmental Protection, 2011a). This act was amended in
May 2011 to address shale gas drilling techniques that
have been increasingly used in Pennsylvania in recent
years and is now referred to as Act 2 of 2011. As an example, the current act addresses the shale gas industry’s
use of centrally located well pads to complete multiple
horizontal wells, otherwise defined as “well clusters”
(E. Draper, 2011, personal communication).
The second statute enacted in 1984 was the Oil and
Gas Act (Pennsylvania Act 223). It repealed Act 225 of
1955 altogether and, for nearly 30 yr, has served as the
Commonwealth’s primary legislation for regulating the
petroleum industry and protecting the environment.
The Oil and Gas Act not only provided the state with
a more comprehensive approach to regulating the industry but also required operators to report drilling
and production details, necessitated bonding requirements, and provided safeguards for water and wetlands
(Burcat, 2008; Department of Environmental Protection, 2011b).
Based on the authority of these and other Pennsylvania statutes, the executive branch adopted further
provisions and rules related to oil and gas exploration
and development. Specifically, chapter 78 of the Pennsylvania Code was adopted in August 1987 and is based
on the authority provided, in part or in whole, by the
Oil and Gas Act of 1984 and the Coal and Gas Resource
Coordination Act of 1984. Chapter 79 of the Pennsylvania Code addresses the issue of oil and gas conservation, under the authority of the Oil and Gas Conservation Law of 1961, and was adopted in August 1971.
Federal Oversight
By the late 1980s, work was being done at the federal
level to evaluate whether waste derived from the oil
and gas industry should be exempted under the 1976
Resource Conservation and Recovery Act (RCRA)
Subtitle C or its 1980 amendments (U.S. Department
of Energy, 2009). In July 1988, the U.S. Environmental
Protection Agency (EPA) issued a determination that
such wastes should not be regulated under RCRA Subtitle C and that the EPA would support state efforts to
enhance their respective regulatory programs (U.S.
Carter et al.
225
Department of Energy, 2009). In 1989, the Interstate
Oil Compact Commission (predecessor to today’s Interstate Oil and Gas Compact Commission) formed
the Council on Regulatory Needs and worked with the
EPA, the states, industry, and various stakeholders to
evaluate state oil and gas regulatory programs across
the country. This study, which addressed state regulation of petroleum exploration and production wastes,
was published in 1990 and is more commonly referred
to as the “1990 Guidelines” (U.S. Department of Energy, 2009; State Review of Oil and Natural Gas Environmental Regulations, 2010). As part of their continuing work, the Council revised and updated these
guidelines in 1994. This multidisciplinary committee
of regulators, industry, and stakeholders eventually became a nonprofit oversight group in 1999, now named
State Review of Oil and Natural Gas Environmental
Regulations, Inc. (STRONGER) (U.S. Department of
Energy, 2009; State Review of Oil and Natural Gas Environmental Regulations, 2010). Since then, STRONGER
has revised the guidelines for review of state oil and gas
regulatory programs twice and, in 2009, formed a hydraulic fracturing workgroup to evaluate fracing issues
and put forth guidelines to address them (State Review
of Oil and Natural Gas Environmental Regulations,
2010). To this end, the DEP worked with STRONGER
in 2010 to formally evaluate Commonwealth’s hydraulic fracturing program (State Review of Oil and Natural
Gas Environmental Regulations, 2010).
Recent Regulatory Changes in Pennsylvania
For the last couple of years, the DEP has been working
on changes to chapter 78 of the Pennsylvania Code to
address certain issues related to Marcellus Shale exploration and development. Issues of particular concern
include incidents of well control and stray gas migration, new ways of drilling and completing unconventional gas wells (specifically, horizontal drilling and
massive frac jobs), disclosure requirements for additives
used in frac fluids, and the production reporting mandated by Pennsylvania Act 15 of 2010 (English, 2011).
On February 5, 2011, the new rulemaking went into
effect and is reflected in significant revisions to subchapter D of chapter 78, as well as certain changes to
subchapters A, C, and E. The most noteworthy changes
to subchapter D involve well control, well construction
standards (in particular, casing and cementing work),
existing well integrity, response to stray gas issues, and
matters of well reporting (English, 2011).
226
The Backstory of the Modern Marcellus Shale Play
As of June 2011, the DEP continues to evaluate
further revisions to chapter 78 of the Pennsylvania Code,
intending to revise regulations involving well plugging,
environmental protection performance standards, and
other necessary improvements, as needed (English,
2011).
THE EASTERN GAS SHALES PROJECT
In the 1950s, the United States had a seemingly unlimited supply of oil and natural gas. By 1968, however,
that supply began to diminish as demand exceeded
the reserves being added by exploratory drilling (Roen,
1993). In October 1973, members of the Organization
of Arab Petroleum Exporting Countries began an oil
embargo in response to the U.S. decision to resupply
the Israeli military during the Yom Kippur war earlier
that month. This was the beginning of an “energy crisis”
that lasted throughout the 1970s. As a result, the U.S.
began to recognize the necessity of expanding research
and development into alternative forms of energy. To
facilitate this, the Energy Research and Development
Administration (ERDA) was created in 1975 to combine the federal government’s activities into a single
agency. The following year, ERDA initiated the Enhanced Gas Recovery (EGR) Program to focus research
efforts on unconventional natural gas resources such as
“tight” gas sands, ultradeep reservoirs, and organic-rich
shales. As the energy crisis extended into the late 1970s,
all of the federal energy agencies were reorganized into
the U.S. Department of Energy (DOE). The DOE’s
primary goal was to provide incentive to overcome
the inertia associated with commercial development
of the nation’s energy resources, including unconventional resources (Smith, 1977).
One of the EGR programs within DOE’s purview,
the Eastern Gas Shales Project (EGSP), was initiated to
evaluate the gas potential of, and to enhance gas production from, the extensive Devonian and Mississippian organic-rich black shales within the Appalachian,
Illinois, and Michigan basins in the eastern United
States. The EGSP program had two purposes: (1) determine the extent, thickness, structural complexity, and
stratigraphic equivalence of all Devonian organic-rich
shales throughout the three basins and (2) develop
and implement new drilling, stimulation, and recovery
technologies to increase production potential (Harper,
2008). Teams of geologists from the U.S. Geological
Survey, various state geological surveys, and universities
Figure 5. Correlation diagram
of Middle and Upper Devonian
formations in western and northcentral Pennsylvania (modified
from Carter, 2007) FM. = Formation; SH. = Shale; GR. = Group;
LS. = Limestone; CHT. = Chert.
correlated and mapped the stratigraphy and structural
geology. For example, the Survey team spent several
years correlating wireline logs and performing basic
mapping of the Upper Devonian Ohio, Java, West Falls,
Sonyea, and Genesee formations and of the Middle Devonian Hamilton Group, which includes the Marcellus
Shale at its base (Figure 5). Geophysicists developed
new technologies to locate fracture systems in the shales
that could be potential reservoirs. Teams of geochemists investigated ways to modify the shale matrix to increase gas flow. Engineers derived and tested both
mathematical and practical models of various fracturing techniques and directional drilling procedures.
And while all of these were occurring, industry drilled
and cored numerous test wells in each of the states involved in EGSP. Five wells were drilled in Pennsylvania
(Figure 1; Table 1), and cores of the Devonian shales
collected from each provided a wealth of data regarding
bedding, mineralogy, fracture systems, and so forth.
Gamma-Ray Log Interpretation
Eastern Gas Shales Project researchers used gamma-ray
logs as the primary source of data for correlating strata
and determining net thickness, as outlined by Piotrowski
and Harper (1979) and Schmoker (1981). Almost all
of the natural gamma radiation emitted from rocks is
caused by the radioactive potassium isotope (K-40) found
in feldspars, micas, and other common silicate minerals.
Shale, composed primarily of clay minerals, has a significantly higher K-40 content than sandstones or carbonates, so that it is relatively easy to distinguish most of
these rock types on wireline logs. In addition, elements
of the uranium and thorium series can be found in minor
amounts in many sedimentary rocks. Marine organicrich black shales, in particular, have higher than normal
radioactivity responses. Although thorium is insoluble
in water and remains stable during diagenesis, uranium
forms highly soluble uranyl compounds in oxidizing
Carter et al.
227
Table 1. List of EGSP Wells in Pennsylvania*
EGSP** Well No.
Permit No.
County
Well Name
Date Logged
PA1
PA2
PA3
PA4
PA5
083-37291
003-20980
049-20846
063-25073
073-20022
McKean
Allegheny
Erie
Indiana
Lawrence
Minard Run Exploration No.1
Combustion Engineering No. 1
Presque Isle State Park No. 1
Glen McCall No. 5
Sokevitz No. 1
February 1979
March 1979
October 1979
November 1979
December 1979
*See Figure 1 for locations.
**EGSP = Eastern Gas Shales Project.
environments that reduce to tetravalent uranium ions
by organic matter in reducing environments such as
organic-rich muds (Adams and Weaver, 1958; Leckie
et al., 1990). This uranium is then concentrated in the
sediments by being fixed in organic complexes, adsorbed
on organic material, or adsorbed on clay minerals. In addition, Veeh (1967; also Veeh et al., 1974) found that
some sedimentary rocks absorb uranium from seawater.
Marine black shales typically have uranium concentrations between 4 and 25 ppm and may approach
100 ppm, although values between 25 and 55 ppm are
considered anomalous and values greater than 55 ppm
are considered to be ore concentrations (Kochenov and
Baturin, 2002). Thus, relatively high gamma-ray log responses typically define organic-rich shales (Figure 6).
In those areas where shale-gas production had been
Figure 6. Diagram of parts of a typical wireline log showing the standard stratigraphic nomenclature of the Marcellus Formation in the
subsurface of Pennsylvania. Indicated log signatures include CAL = caliper; GR = gamma ray; Nϕ = neutron porosity; Dϕ = density
porosity; DEN = bulk density.
228
The Backstory of the Modern Marcellus Shale Play
developed and studied historically (e.g., the Big Sandy
field in eastern Kentucky), researchers found that an
empirical relationship exists between high gamma-ray
response and gas production from organic-rich shales
(Bagnall and Ryan, 1976; Martin and Nuckols, 1976;
Majchszak, 1978; Smith, 1978). In addition, Schmoker
(1978, 1979, 1980; also Meyer and Nederlof, 1984)
has shown that gamma-ray, formation density, and
other wireline log signatures can be used to calculate
the organic richness or total organic carbon (TOC) of
a source rock.
Findings and Deliverables
The EGSP products included numerous cross sections,
maps, and technical reports (e.g., Piotrowski and Harper,
1979) showing structure, total formation thicknesses,
net feet of black shale, and net feet of sandstone throughout the entire Middle and Upper Devonian sequence in
western and north-central Pennsylvania. At the time, it
was determined that the Devonian organic-rich shales
could be important gas reservoirs, at least in northwestern Pennsylvania where they were both thick (especially the Huron Member of the Ohio Shale and the
Rhinestreet Member of the West Falls Formation; see
Figure 5) and close to the surface. Homeowners, small
businesses, and even a few large manufacturing plants
willing to drill a few hundred to 1000 ft (305 m) within
a few miles of the shore of Lake Erie found a cheap domestic source of natural gas. This had been occurring
since the 1860s (White, 1881). By 1890, more than
200 wells were known to have been drilled in the vicinity of Erie, Pennsylvania (Harper, 1980), and interest
spurred by EGSP resulted in numerous shallow wells
being drilled in Erie County during the 1980s and
1990s. The wells did not provide a lot of volume—flows
generally were, at best, only a few tens of cubic feet per
day and at a few ounces to a few pounds of pressure—
but, aside from the coldest months of the year, they
typically provided the owner with a steady and adequate supply of natural gas for cooking and heating.
The EGSP produced much optimism about Devonian shale. All of the shale formations were thought to
have excellent potential, especially with the expected
development of better technology for inducing and enhancing fracture systems. This was particularly true of
the shallower formations such as the thick Huron and
Rhinestreet shales. The deeper shales, especially the
Marcellus Formation, were considered to be much less
attractive targets. Development of natural gas from
these rocks would essentially be in limbo until gas prices
increased and technology advanced enough to make
drilling and completion competitive with more conventional targets.
The Eastern Gas Shales Project Legacy
Although the DOE continued its research and testing of
Devonian shales through 1992, the furor experienced
during the heyday of EGSP faded during the early
1980s because of low gas prices and lack of sufficiently
useful technologies for extracting the gas. The numerous reports that resulted from EGSP, as well as other
independently sponsored studies, added a wealth of
new information to an already overwhelming volume
of literature on Devonian and Mississippian organicrich shales. In addition, it significantly enhanced our understanding of these complex rocks and their potential
for gas production. The complete EGSP library has remained relatively obscure until recent years because of
a general lack of interest. It is quite extensive and includes a variety of physical, chemical, geologic, and engineering information. Published in 2007, the National
Energy Technology Laboratory’s compendium of natural gas archives (National Energy Technology Laboratory, 2007) contains a significant part of this library. In
addition, some excellent summaries, such as those by
Roen and Kepferle (1993), have been published that
provide very useful information on shales, in general,
and EGSP, in particular.
GEOLOGIC INTERPRETATIONS THROUGH TIME
Since the late 1980s, our understanding of the geologic
history, depositional environment, and lithology of Devonian organic-rich shales has continued to evolve.
This section provides the Survey’s current evaluation
of Marcellus lithostratigraphy followed by our assessment of sequence-stratigraphic concepts as applied to
Devonian rocks including the Marcellus Formation.
Introduction: The Hamilton Group
The Marcellus Formation is the lowest stratigraphic
unit within the Middle Devonian Hamilton Group in
the central Appalachian Basin (Figure 7). Dennison
and Hasson (1976) described the Hamilton Group as
a classic example of intricate facies changes that have
resulted in a complexity of stratigraphic nomenclature
Carter et al.
229
Figure 7. Standard terminology of Marcellus strata in western and central New York. Mbr. = Member.
over its extent. In New York, the standard section includes the Marcellus (currently considered a subgroup),
Skaneateles, Ludlowville, and Moscow formations. In
Pennsylvania, in those areas where the Skaneateles,
Ludlowville, and Moscow cannot be separated or contain coarse clastics, correlative strata above the Marcellus
are called Mahantango Formation (Willard, 1935, 1939;
Faill et al., 1978).
In much of New York and Pennsylvania, the Hamilton Group, consisting primarily of terrigenous clastic
rocks with some limestones and calcareous mudstones,
extends from the top of the Onondaga Limestone to the
base of the Tully Limestone (Figure 8). In most of the
Appalachian Plateau and Valley and Ridge province of
Pennsylvania, where the Tully is present and mappable
(Heckel, 1969), recognition of the upper Hamilton
230
The Backstory of the Modern Marcellus Shale Play
boundary presents no problems. The Tully is absent
in northwestern Pennsylvania and western New York,
however, and darker colored Upper Devonian mudrocks and carbonates disconformably overlie the Hamilton Group. In parts of south-central Pennsylvania,
where the Tully is represented, if at all, by a concretion
zone, the upper boundary of the Hamilton Group is
placed at the first appearance of black mudrocks of the
basal Upper Devonian (Burket Member of the Harrell
Formation). In fact, Faill et al. (1978) considered the
Tully to be a member of the Mahantango Formation in
the Valley and Ridge province of central Pennsylvania.
In Virginia and adjacent parts of West Virginia, the entire Hamilton Group and the overlying Burket shales are
considered parts of the Millboro Shale (Butts, 1940;
Dennison and Hasson, 1976; De Witt and Roen, 1985;
Figure 8. Correlation of Middle Devonian rocks in Pennsylvania, with comparison with the New York standard. Nomenclature for most
of Pennsylvania is based on wireline log correlations. Nomenclature for northeastern, south-central, and east-central Pennsylvania, respectively, is based on Sevon et al. (1989), Cate (1963), and Faill et al. (1978). FM. = Formation; LS. = Limestone; GR. = Group; SUBGR. =
Subgroup; MBR = Member.
De Witt et al., 1993). Dennison and Hasson (1976; also
Hasson and Dennison, 1979) even extended the name
Millboro into southwestern Pennsylvania.
Marcellus Lithology and Lithostratigraphy
Hall (1839) first used the name Marcellus for black and
gray, thinly laminated shales exposed near the village
of Marcellus in Onondaga County, New York, where
it rests disconformably on the Seneca Member of the
Onondaga Limestone. He described a variable formation, with a lower mass of black “slate” containing limestone nodules or concretions and an upper dark-colored,
slaty, fossiliferous shale grading upward into compact
blue shale and fissile olive shale. Hall marked the top
of the Marcellus where a “true Hamilton fauna” first appeared (Cooper, 1930). Later, Hall (1843) further divided the Marcellus into two significantly different
divisions. The lower division was very black and bituminous. Vanuxem (1840) described it as having the appearance of coal; Lesley (1892, p. 1196) stated that it
was “so bituminous in places that it flames when thrown
upon a fire of hot coals.” These shales contained an
abundance of pyrite and several beds of concretions
and terminated upward in a thin limestone. The upper
division started as a black, more fissile shale above the
limestone and graded upward to somewhat lighter
(olive or slate) color. Hall (1843) noted that the abundant fossils seen in the underlying Onondaga Limestone
either ceased entirely or were succeeded by others of a
totally different character.
In Pennsylvania, the First Geological Survey (1836–
1858) referred to Marcellus-equivalent strata as the
“Cadent Lower Black Slate” and described them as
“a black and highly bituminous slate, graduating upward into a dark-blue argillaceous shale” (Rogers, 1858,
p. 107–108). It took until the Second Geological Survey
(1874–1889) before the name “Marcellus” came into
general use in the state. Because of its relatively persistent lithology, the Marcellus is one of the easiest units
recognizable in Pennsylvania outcrops. Willard (1939,
p. 168) noted that “… its obviously distinctive features
could hardly escape even the least observant of (Second
Carter et al.
231
Survey State Geologist J. P.) Lesley’s band” in the late
1800s.
Marcellus lithostratigraphy has undergone numerous changes during the past 170+ yr. Hall (1839) originally considered the Marcellus to include all of the rocks
between the Seneca Limestone (uppermost Onondaga)
and “Ludlowville shales.” Vanuxem (1840) subsequently removed the upper rocks from the Marcellus
and renamed them “Skaneateles shales.” Cooper (1930)
subdivided the Marcellus Formation in ascending order
into the Union Springs, Cherry Valley, and Oatka Creek
members in the western part of New York (additional
members occur as eastern facies equivalents). This tripartite division of the Marcellus is generally accepted,
although the stratigraphic ranks of the formation and
its members have recently changed in New York as a result of highly detailed outcrop work (Ver Straeten et al.,
1994; Ver Straeten and Brett, 2006; Brett et al., 2011)
(see Figure 7 for the current standard nomenclature
in western New York). These rank changes generally
have not been accepted in the subsurface of New York,
Pennsylvania, and Ohio (Lash and Engelder, 2011),
however, because of the difficulty in recognizing and
separating some of the very thin members and beds in
drill cuttings and on wireline logs. In fact, Piotrowski
and Harper (1979) made no stratigraphic distinction
among any of the members of the Marcellus Formation
and even included organic-rich black shales from overlying Hamilton Group formations in their “Marcellus
facies.” When Range completed the Renz No. 1 gas
well in Washington County in 2004, they hydraulically
fractured the black shales of both the Marcellus Formation and the overlying Skaneateles Formation (the Levanna Member of New York). Thus, the “Marcellus
play” actually incorporated the “Marcellus facies” from
the beginning. This section of the article, however, focuses only on true Marcellus Formation strata.
In this article, we use a modified version of the
western New York terminology of Cooper (1930)—
Marcellus Formation, Union Springs Member, Cherry
Valley Limestone, and Oatka Creek Member. In addition to Cooper’s (1930) members, Ver Straeten et al.
(1994) recognized a new member, the Hurley, which
they placed at the top of the Union Springs Formation.
Brett et al. (2011), however, considered the Hurley to
lie at the base of the Oatka Creek Formation (Figure 7).
The Hurley, if it exists at all in Pennsylvania, appears
to be inseparable from the Cherry Valley Limestone
on wireline logs; therefore, it is not considered further
herein.
232
The Backstory of the Modern Marcellus Shale Play
Union Springs Member
Where it occurs in the subsurface of Pennsylvania, the
Union Springs Member consists of all of the black organicrich shale, with thin interbedded limestones and altered
ash fall beds, occurring between the Onondaga Limestone and the Cherry Valley Limestone or its equivalent. It is, therefore, essentially identical with the
Bakoven Member of the Union Springs Formation of
New York (Figure 7). The Union Springs Member appears on wireline gamma-ray and density logs as much
more radioactive and less dense than surrounding strata
(Figure 9), making it very easy to recognize and correlate. The lower part is particularly radioactive as a result
of both high TOC content and the occurrence of one or
more K-bentonite beds that are part of the Tioga ash
zone (Fettke, 1952b; Way et al., 1986).
Lash and Engelder (2011) mapped the extent of the
Union Springs in the subsurface of the central Appalachians and showed that it thickens significantly in northeastern Pennsylvania to more than 160 ft (>49 m). They
also suggested that the Union Springs is locally absent
along a northeast-southwest–trending axis in western
New York into northwestern Pennsylvania. We are preparing a series of preliminary cross sections across Pennsylvania that indicate that the Union Springs is essentially absent throughout much of western New York
and westernmost Pennsylvania north of the BeaverWashington County boundary. It typically overlies a
“corrosional discontinuity” (Baird et al., 2000) that separates it from the Onondaga Limestone (Figures 7, 8).
The discontinuity, juxtaposing the top of the Onondaga
with increasingly younger strata from east to west, represents the beginning of a major basin deepening event,
Ettensohn’s (1985) “second tectophase.” This marked
the end of Onondaga carbonate deposition and the beginning of organic-rich shale deposition as a result of
transgression and an associated sediment starvation
(Ver Straeten et al., 1994) related to tectonically induced subsidence (Ettensohn, 1998; Brett et al., 2011).
Cherry Valley Limestone and Purcell Member
The Cherry Valley Limestone and, at least partially,
equivalent Purcell Member of the Marcellus Formation
are very good marker horizons where they occur. The
Cherry Valley is a widespread relatively thin carbonate
unit, typically dark colored, fossiliferous, and petroliferous (Cooper, 1930; Baird et al., 2000). The limestone
truncates the upper Union Springs Member across western New York (Baird et al., 2000; Brett et al., 2011).
Our preliminary cross sections indicate that the Cherry
Figure 9. Generalized cross section of Marcellus Formation (FM.) and adjacent strata from Washington County to Bradford County,
Pennsylvania, showing the general west-to-east thickening trend. Gamma-ray log signature is shown in green, with the most organic-rich
radioactive sections colored in green and red. Bulk density signature is shown in blue. Heavy red lines indicate disconformities. GR. =
Group; Mbr. = Member; Ls. = Limestone; Sh. = Shale; CHT. = Chert.
Valley log signature can be traced in the subsurface across
most of Pennsylvania, making it an excellent marker
horizon.
Relatively thin limestones or calcareous siltstone
and shale packages as much as 45 ft (14 m) thick occur
in the lower half of the Marcellus in south-central Pennsylvania. Gas operators found one of these to be a useful
subsurface marker in wells drilled in Bedford County in
the late 1950s and named it the “Purcell Limestone”
(Cate, 1963). Dennison and Hasson (1976) changed
the name to Purcell Member because of its varied lithology; this is the terminology we use here. Drill cuttings
from the type well in Bedford County indicate that the
interval consists of dark-colored, silty, calcareous shale
essentially indistinct from much of the rest of the Mar-
cellus, but drill cuttings are not especially valuable for
detailed lithologic description and correlation because
each sample typically represents a vertical distance of
10 ft (3 m) or more. De Witt et al. (1993) described
the Purcell in outcrop as gray silty shale and mudrock,
with some siltstone, an abundance of limestone nodules,
and some barite nodules about 1 to 2 in. (2.5–5 cm) in
diameter. Dennison (1963) traced a persistent zone of
interbedded limestone and shale occurring between
40 and 100 ft (12–30 m) above the base of the Marcellus
through outcrops in eastern West Virginia. He apparently was unaware that the zone had been named Purcell because that name does not appear anywhere in his
report. Ver Straeten (1996) found the Purcell to be a
distinctive interval visible in almost all mid-Marcellus
Carter et al.
233
outcrops from central Pennsylvania to Virginia, with the
lithofacies in at least one locality being very similar to
that of the Cherry Valley Limestone in New York. In
fact, the Purcell typically is considered to be stratigraphically equivalent to the Cherry Valley (Figure 8).
Cate (1963) traced the Purcell Member into western Pennsylvania at least as far west as Laurel Hill anticline (coincident with the boundary between Somerset
and Westmoreland counties). We have been able to
trace the wireline log signature of the Cherry Valley–
Purcell interval as far west as western Greene County,
where it disappears as the result of the corrosional discontinuity that similarly affects the Union Springs Member. Based on the above research, the Purcell and Cherry
Valley are very distinct lithologic units with only a small
number of similarities. The larger issue is not whether
they are equivalent but, instead, where does the Cherry
Valley end and the Purcell begin? Drill cuttings are not
particularly useful for answering this question, and the
types of modern wireline logs that could be useful (e.g.,
photoelectric log) are generally not publicly available.
As this issue is beyond the scope of this article, it will
not be addressed here further.
Lash and Engelder (2011) mapped the Cherry
Valley Limestone from less than 10 ft (43 m) thick in
northeastern Pennsylvania and southeastern New
York. Dennison (1963) found the Purcell Member to
range from 6 to 29 ft (2–9 m) thick in his West Virginia
outcrops. Based on our preliminary cross sections, the
Cherry Valley–Purcell interval, where it occurs, ranges
from 3 ft or less (≤1 m) near the zone of the corrosional
discontinuity to 70 ft or more (≥21 m) in south-central
Pennsylvania.
Oatka Creek Member
The Oatka Creek Member, as defined here, begins with
a sharp, corroded, and mineralized boundary on top of
the Cherry Valley Limestone that Brett et al. (2011) interpreted as a drowning unconformity. The Oatka
Creek in the subsurface of most of the Appalachian Plateau of Pennsylvania is dominated by organic-rich shale
(the Chittenango Member of the Oatka Creek Formation in New York; see Figure 7). Like the Union Springs
Member, this shale has a high TOC value (up to 18%)
(Baird et al., 1999) and displays high radioactivity and
low density signatures, respectively, on gamma-ray and
bulk density wireline logs (Figure 9) in most areas where
it is developed.
234
The Backstory of the Modern Marcellus Shale Play
In westernmost Pennsylvania, especially in those
areas where the Union Springs and Cherry Valley limestones are missing, the Oatka Creek is most radioactive
in the lowest 10 to 15 ft (3–4.5 m). Lash and Engelder
(2011) found the Oatka Creek to be only about 30 ft
(∼9 m) thick in western New York and northwestern
Pennsylvania but thickened greatly to the east, exceeding 550 ft (168 m) in northeastern Pennsylvania. Our
own investigations indicate that this member is closer
to 90 ft (27 m) thick in westernmost New York and
westernmost Pennsylvania north of the Ohio River.
The top of the Oatka Creek Member (and of the
Marcellus Formation) is marked by the base of the Stafford Member of the Skaneateles Formation, a fossiliferous carbonate unit that occurs throughout much of New
York and Pennsylvania. We have traced the Stafford in
the subsurface as far east as the Allegheny Front in central Pennsylvania and Susquehanna County in northeastern Pennsylvania. In south-central Pennsylvania,
the Stafford appears to grade laterally, or be replaced entirely, by a zone of siltstone and shale equivalent to the
Turkey Ridge Member of the Mahantango Formation.
Biostratigraphy and Fossil Content
Few or no data exist on the fossil content of the Hamilton Group and Marcellus Formation in the subsurface
of Pennsylvania. The faunas are well known in outcrop
in New York (e.g., Linsley, 1994) and in central Pennsylvania (e.g., Willard, 1932, 1939), however, and have
been put to use in assessing the biota of the lower Middle Devonian in the Appalachian Basin.
Fossils are relatively sparse in the black-shale parts
of the Marcellus Formation, which have been shown to
have the highest geochemical signals for anoxia in the
Devonian of eastern North America (Baird et al., 1999).
As a result, most of the fossils found in these parts consist primarily of nektonic forms such as styliolinids,
cephalopods, and conodonts, as well as microscopic fossils of both phytoplankton and zooplankton. The occasional benthic fossil specimens include bryozoans, brachiopods, tentaculitids, crinoids, trilobites, bivalves, and
gastropods. The limestones and calcareous beds generally preserve a larger and more diverse fauna of benthic
animals.
The Devonian has been zoned on a global scale
using conodonts (e.g., Kaufmann, 2006). Conodont
zonation for Devonian strata in the eastern United
States, however, has been revised (Kirchgasser, 2000);
the currently accepted zonation for Middle Devonian
Figure 10. Sea level curves, conodont zonation, and sequence stratigraphy of Middle Devonian (Eifelian and part of the Givetian) rocks
in New York and Pennsylvania (based on Ver Straeten, 1996, modified by Bartholomew et al., 2006, and Brett et al., 2011). Euramerica
sea level curve (black) from Johnson et al. (1985). Revised sea level curve (red) interpreted from Brett et al. (2011): long bases = thirdorder sequence boundaries; short bases = fourth-order sequence boundaries. Eif = Eifelian; Giv = Givetian; EHST = early highstand
systems tract; ETST = early transgressive systems tract; FSST = falling stage systems tract; HST = highstand systems tract; LHST = late
highstand systems tract; TST = transgressive systems tract.
strata in the Appalachians is shown in Figure 10 (see
also Bartholomew et al., 2006; Brett et al., 2011). In addition, other faunal groups (e.g., goniatite cephalopods)
are also useful for correlation.
Sequence Stratigraphy
In a seminal article on Devonian eustatic cycles, Johnson
et al. (1985) recognized and described a series of
transgressive-regressive ( T-R) cycles in the Devonian
of the North American craton, combined into two larger
cycles labeled I and II that are separated by the Taghanic
unconformity (sensu Hamilton-Smith, 1993). A part of
this is illustrated in Figure 10 (left side). These T-R cycles represent the third-order cycles of Vail et al. (1977);
fourth-order cycles appear within those. Brett et al.
(2011, p. 22) viewed this work as “a useful synthesis,
correct in its main outline,” but that it was only intended
as a “work in progress.” Unfortunately, as they pointed
out, it has since become dogma despite some flaws. The
cycles defined by Johnson et al. (1985) were not based
on the modern concept of sequence stratigraphy. In an
attempt at defining Devonian sequences and their component systems tracts in a consistent manner, Brett and
Baird (1996; also Bartholomew et al., 2006, Brett et al.,
2011) have subdivided the Middle Devonian section
from the base of the Onondaga Limestone through
the Tully Limestone into sequences that can be traced
widely through the northeastern half of the United
States, including within the Michigan, Iowa, Illinois,
and Appalachian basins. A revised sea level curve, which
we have interpreted from their work, is shown in red in
Figure 10.
Each of the Middle Devonian third-order sequences
(Ic–IIb) in the Appalachian Basin starts at the erosive
base of a transgressive stratum indicating a sequence
boundary (transgressive systems tract [TST]). This stratum typically is a limestone, although some of the widely
Carter et al.
235
recognized Hamilton limestones (especially the Centerfield and Tichenor, which are standard formation boundary markers) cannot be recognized in many places in the
subsurface of Pennsylvania. In the Marcellus Formation
of Pennsylvania, two TSTs represented by the Cherry
Valley–Purcell interval and the beginning of deposition
in the Oatka Creek interval exist. A third TST, defined by
deposition of the Stafford Limestone of the Skaneateles
Formation, marks the top of the Marcellus Formation.
Each TST is followed by a highstand systems tract
(HST ) that encompasses the time between still-stand
and the beginnings of sea level fall. The HST strata typically record the deepest water in the cycle, although
what is meant by “deepest” is relative. Many authors consider the Marcellus sea to have been “deep” in the sense
of the Black Sea (that is, hundreds or thousands of meters of water column, e.g., Clarke, 1904; Rich, 1951;
Potter et al., 1981; Broadhead et al., 1982; and Kepferle,
1993). Harper (1999), however, suggested that the
Marcellus black shales were deposited in a variety of
shallow-water anoxic environments, possibly as shallow
as 150 ft (45 m). In such a case, the shallow-water column might have been stratified enough to keep warm
oxygenated surface water from mixing with hypoxic
(anoxic or euxinic) bottom water (Lash and Engelder,
2011). Modern hypoxic events have been documented
in water as shallow as 16 ft (5 m) over areas as large as
6000 mi2 (9500 km2) resulting from a combination of
increased nutrient input, decay of phytoplankton
blooms, and water column stratification induced by
warm water and a strong salinity gradient (Harper et al.,
1991; Rabalais et al., 1991). Within the Marcellus, the
most organic-rich black-shale facies of the Union Springs
and Oatka Creek members represent the HST. On wireline logs, they exhibit the highest radioactivity and lowest bulk density signatures within the formation.
Basal black-shale facies of the HST tend to grade
laterally and vertically into less organic-rich and coarser
grained shales, siltstones, and even sandstones as pulses
of continental sediment prograde seaward during stillstand. It should be noted that, although the flush of
coarser clastics into the depositional basin during early
progradation marks the beginning of a regressive phase
within the rock section, this might be caused more by a
tectonic than a purely eustatic event.
Falling stage systems tracts (FSST ), above and seaward of the HST, can be difficult to recognize because
of subsequent erosion. Bartholomew et al. (2006) found
that FSSTs in the Devonian of the Appalachian foreland
begin with a distinct shell bed that marks a shift to shal236
The Backstory of the Modern Marcellus Shale Play
lower water conditions. This condition is reflected twice
in the Marcellus, within the Union Springs and Oatka
Creek members, by a decrease in organic content and
an increase in silt upward and eastward in Pennsylvania.
In addition, an end-Marcellus FSST is recorded in outcrop in the Solsville and Pecksport members of the Oatka
Creek Formation in central New York (Brett et al.,
2011). In the subsurface of Pennsylvania, the FSSTs
within the Union Springs and Oatka Creek members
appear on wireline logs as decreasing radioactivity and
increasing bulk density signatures in the upper few meters or tens of meters.
Continuing regression/progradation ultimately results in a lowstand systems tract (LST ) that indicates
the shallowest water conditions. Typically, this is recorded as an unconformity at the base of the next
TST. Within the Marcellus Formation, the LSTs occur
at the interface between the Union Springs Member and
the Cherry Valley Limestone and between the Oatka
Creek Member and the Stafford Limestone of the overlying Skaneateles Formation.
Based on the above criteria, the Marcellus Formation
represents all or part of two third-order cycles, labeled
Eif-2 and Eif-Giv by Brett et al. (2011) (Figure 10) and
MSS1 and MSS2 by Lash and Engelder (2011). Sequence Eif-2 actually starts in the underlying Onondaga
Limestone but includes the entire Union Springs Member. Sequence Eif-Giv encompasses the Cherry Valley
Limestone and Oatka Creek Member. The exact durations of these sequences is unknown but probably is on
the order of 1 to 2 m.y. (Kaufmann, 2006; Lash and
Engelder, 2011). If the smaller scale fourth-order subsequences shown in Figure 10 spanned approximately
400 k.y. each (based on the long eccentricity Milankovitch cycle), sequence Eif-Giv (= MSS2) would have
been approximately 1.2 m.y. in duration. A similar assessment for sequence Eif-2 (= MSS1) is difficult because fourth-order subsequences have not yet been
recognized within it in the Appalachian Basin. But what
caused these T-R cycles? Although no overwhelming
evidence exists for large-scale glaciation during the Middle Devonian, Elrick et al. (2006, 2009) found that the
magnitude of changes in oxygen isotope values, determined from conodont apatite, matched the kinds of
global temperature shifts associated with Carboniferousand Pleistocene-type glacial cycles. The T-R cycles could
also have resulted from tectonics, the initial collisional
events in the early stages of the Acadian orogeny, but it
is difficult to understand how mountain-building episodes
could be regular enough to account for what appears to
be Milankovitch cyclicity. It is obvious that a great deal
more research is needed before this will be resolved.
THE MODERN MARCELLUS SHALE GAS PLAY
The Marcellus Shale covers approximately 95,000 mi2
(∼246,049 km2) in several states throughout the Appalachian Basin, but more than one-third of this area
(35,000 mi2 [90,650 km2]) is situated in Pennsylvania.
This organic-rich rock exists throughout the subsurface
of northern and western Pennsylvania, and it is approximately bounded to the east by the Topographic Front;
beyond the front, in the Ridge and Valley province, the
Marcellus crops out, exists in a structurally complex
state in the subsurface, or is absent altogether (Figure 1).
The Marcellus Shale is a known source rock for
most of the historical petroleum reservoirs in the Commonwealth. Furthermore, Pennsylvania operators have
known for decades that the unit contained significant
volumes of gas. As operators began exploring the Lower
Devonian Oriskany Sandstone in the 1930s, they commonly encountered large gas shows in the overlying
Marcellus. They sometimes attempted completions in
the shale but, more commonly, they shut down drilling
operations for a few hours or even days until the well
blew down enough to continue drilling safely (Harper
and Kostelnik, 2010). The difference between then
and now is that much has changed with respect to our
understanding of the Marcellus Shale’s source and reservoir characteristics, not to mention the business, technology, and engineering resources available to the petroleum industry today. This section chronicles the
modern Marcellus play, from its discovery in 2004
through mid-2011, and how geologic structure has
played an important function in the drilling of Marcellus
Shale gas wells permitted in different areas of the state.
The Discovery Well
In 2003, Range permitted and drilled the Renz No. 1
well (Figure 2). This vertical well was drilled to a depth
of 8470 ft (2582 m) into the middle Silurian Rochester
Shale. Large shows of gas were encountered when they
penetrated the Middle Devonian organic-rich shales,
and so Range returned to the well in October 2004, stimulating it in the organic-rich shales of the Marcellus and
Skaneateles formations from 6174 to 6284 ft (1882–
1915 m). The well was turned in line in 2005 and produced 5.5 mmcf in 31 days or an average of 300 mcf/day
(Pennsylvania Internet Record Imaging System/Wells
Information System, 2011).
Following the success of this discovery well, Range
initiated a pilot horizontal program, drilling five horizontal wells into the Marcellus that reported initial flows
ranging from 1.4 to 4.7 mmcf/day (Durham, 2008).
Based on these results, Range began using horizontal
drilling and slickwater hydraulic fracturing techniques
developed in the Barnett Shale of Texas to maximize
the economic viability of the Marcellus play in Pennsylvania. Continued refinement of drilling and stimulation
technologies has propeled the Marcellus Shale play into
its current position as a super giant gas field.
Drilling Activity: 2005–2011
We have tracked the drilling activity associated with the
Marcellus Shale play in terms of both permitted and
completed wells using the DEP Environment Facility
Application Compliance Tracking System (eFACTS)
and the Survey’s PA*IRIS/WIS database, respectively.
Although both systems track active oil and gas permits,
information contained in PA*IRIS/WIS incorporates
and archives critical paperwork associated with each
oil and gas permit as obtained from DEP, which enables
the Survey to determine when or whether a well has actually been completed by the operator. By the Survey’s
process, an oil or gas well is considered complete when
an operator has submitted a well completion report to
the Commonwealth that includes not only basic demographic and drilling data for the well, but also casing details, perforation and stimulation depths, water and
proppant volumes used in hydraulic fracturing, and initial open-flow data.
As of June 2011, DEP had issued 6488 permits to
drill Marcellus Shale gas wells (Department of Environmental Protection, 2011e); of these, 1098 have been
interpreted as known Marcellus completions by the Survey (Pennsylvania Internet Record Imaging System/
Wells Information System, 2011). The remainder of
this section uses this completed wells data set to analyze
the Marcellus Shale gas play with respect to geographic
location, operator involvement, and drilling type.
Geographic Distribution
The modern Marcellus play discovery well was drilled
in Washington County, Pennsylvania, and for the first
several years (2005–2008), more permits were issued
in Washington County than anywhere else in the state
( Table 2). Marcellus exploration was initiated in the
Carter et al.
237
Table 2. Summary and Distribution of Permits Issued and Wells Completed for Marcellus Shale Gas Activity in Pennsylvania, 2004 to
June 2011
Completion Event
H*
V*
2004
Total Permits
3
Total Completions
3
0
3
1
2
2005
Total Permits
13
Total Completions
2
0
13
1
1
2
39
2006
Total Permits
41
Total Completions
12
County
Westmoreland
1
Westmoreland
1
Elk
1
Elk
1
Somerset
1
Somerset
1
Washington
8
Washington
1
Armstrong
1
Butler
1
Washington
23
Butler
Potter
4
Mckean
Washington
0
12
2007
Total Permits
98
Total Completions
59
13
85
9
50
2008
Total Permits
411
Total Completions
167
136
275
64
103
2009
Total Permits
1702
Total Completions
411
1310
392
249
162
2010
Total Permits
2986
Total Completions
426
2602
384
340
86
as of June 2011
Total Permits
1229
Total Completions
0
1138
91
0
0
2
2
Warren
2
Lycoming/Butler
McKean
3
2
Armstrong, Brad, Elk Fayette,
Jefferson, Susquehanna
1
Washington
41
Greene
12
Butler
12
Washington
7
Lycoming
8
Butler
7
Potter/Susquehanna
6
Lycoming
6
Elk/Armstrong
5
Potter/Fayette
5
Washington
68
Washington
41
Susquehanna
56
Susquehanna
25
Greene
51
Greene
20
Bradford
46
Fayette
19
Fayette
36
Bradford
12
Bradford
422
Washington
63
Tioga
257
Bradford
62
Susquehanna
144
Susquehanna
58
Washington
161
Fayette
51
Bradford
791
Bradford
95
Tioga
538
Tioga
82
Susquehanna
222
Susquehanna
44
Lycoming
222
Lycoming
38
Washington
220
Greene
38
Washington
74
Washington
0
Susquehanna
98
Susquehanna
0
Tioga
130
Greene
0
Bradford
280
Fayette
0
Lycoming
150
Bradford
0
*Horizontal = horizontal; V = vertical.
238
2
Greene
1
Fayette
1
The Backstory of the Modern Marcellus Shale Play
Lycoming
86
Greene
42
Figure 11. Comparison of number of Marcellus Shale gas permits issued by the Pennsylvania (PA) Department of Environmental
Protection (DEP) and number of known
completions, as interpreted by the PA Geological Survey, for the most active counties in
the Commonwealth (Department of Environmental Protection, 2011e; Pennsylvania
Internet Record Imaging System/Wells Information System, 2011).
northeastern part of the state in 2006 when the first
permit was issued in Lycoming County, but it was not
until 2008 that drilling activity became more evenly
split between these two regions (i.e., southwestern and
northeastern Pennsylvania). Specifically, in 2008, 155
permits were issued in Washington, Greene, and Fayette
counties and 148 permits were issued in Tioga, Bradford,
Susquehanna, Wayne, and Lycoming counties. Since
2009, most of the permits have been issued in northeastern Pennsylvania, although several hundred continue to be issued in Greene, Washington, and Fayette
counties every year (Table 2). Of the 6488 Marcellus
permits issued by DEP as of June 2011, the most were
issued in Bradford County. Specifically, 1542 permits
have been issued in Bradford County, 937 in Tioga
County, 597 in Washington County, 527 in Susquehanna
County, and 503 in Lycoming County (Figure 11).
Activity by Operator
As of June 2011, 76 operators had applied for permits to
drill Marcellus Shale gas wells in Pennsylvania (Department of Environmental Protection, 2011e). The five most
active Marcellus Shale operators, as determined by the
number of permits issued by the DEP to date, include
Chesapeake Appalachia LLC (1268), Talisman Energy
USA, Inc. (620), Range (588), Shell Western Exploration
and Production Inc. LP (SWEPI, formerly East Resources
LLC, 525), and Chevron (formerly Atlas Resources LLC,
326) (Figure 12). Of these, Chesapeake, Talisman, and
SWEPI work in northeastern Pennsylvania, and Range
and Chevron have been focusing most of their efforts in
southwestern Pennsylvania (Pennsylvania Internet Record Imaging System/Wells Information System, 2011).
Drilling Type
The use of horizontal well drilling technologies in the
Marcellus Shale gas play has played a major part in its
widespread success. Compared with historical shale gas
well production (where shale gas was produced from
shallow low-pressure wells that lasted for many years
but returned less than notable production volumes),
the modern Marcellus play has exceeded all expectations
for shale gas production in the state. Horizontal well drilling technology clearly emerged as the most popular drilling approach in 2009 when 1310 horizontal permits
were issued by the DEP, exceeding the number of vertical well permits (392) by nearly 3.5 times (Figure 13). As
of June 2011, a total of 1287 Marcellus wells had been
permitted by DEP to be vertical completions, and 5201
had been permitted to be horizontal completions. Of the
1098 well completion reports received and processed by
the Survey as of this same date, 664 were horizontal
wells and the remaining 434 were vertical completions.
Effects of Geologic Structure on Marcellus Shale Gas
Drilling and Production
The structural grain of the rocks in western and northern Pennsylvania roughly parallels the structural grain
Carter et al.
239
Figure 12. Comparison of number of Marcellus Shale gas permits issued by the Pennsylvania (PA) Department of Environmental
Protection (DEP) and number of known
completions, as interpreted by the PA Geological Survey, for the most active operators
in the Commonwealth (Department of Environmental Protection, 2011e; Pennsylvania
Internet Record Imaging System/Wells Information System, 2011). SWEPI = Shell Western
Exploration and Production, Inc.
of the Ridge and Valley province of central Pennsylvania (Figure 14). Orientations of fold axes and faults do
not specifically relate to Marcellus Shale gas production,
however, because the primary pathway for fluid flow in
the shale is along a set of joints that are oriented essentially parallel with the contemporary tectonic stress field.
The Marcellus, like most shales, has very low permeability (100–500 nanodarcys). Success in producing shale
gas, therefore, depends mostly on the presence of both
naturally occurring joint sets and fractures induced by
the hydraulic fracturing process, instead of on folds,
faults, and/or tectonic fractures, as is the case for con-
Figure 13. Comparison of
vertical and horizontal Marcellus
Shale gas well permits issued
by the Pennsylvania Department
of Environmental Protection
over time (Department of Environmental Protection, 2011e).
240
The Backstory of the Modern Marcellus Shale Play
ventional reservoirs. In fact, the presence of intense fracturing caused by folding and faulting is considered undesirable for effective production from the shale, and the
industry tends to avoid them if at all possible. As such,
seismic surveying for the purpose of identifying tectonic
fractures to be avoided has become an integral part of
the shale-gas exploration process.
Engelder (2004; also see Engelder and Whitaker,
2006; Engelder, 2008; Lash and Engelder, 2009) recognized a pre- or early Alleghanian joint set (J1) that has the
same approximate east-northeast–west-northwest orientation as neotectonic joints (J3) and the contemporary
Figure 14. Map of Pennsylvania showing general orientations of structure axes, contemporary tectonic stress field, and J1, J2, and J3
joint sets of Engelder (2008).
tectonic stress field (Figure 14). The J1 joints were created by fluid pressure during hydrocarbon generation
within the Devonian organic-rich source rocks such as
the Marcellus (Lash and Engelder, 2009), as evidenced
by the occurrence of J1 joints folded along with bedding
in the Ridge and Valley province of the Appalachians
(Engelder, 2004). The J1 and J3 joints intersect a set
of Alleghanian cross-fold joints (J2) that are very prominent both at the surface (e.g., Nickelsen and Hough,
1967; Bench et al., 1977) and in Devonian shales in
the subsurface (Evans, 1980, 1994).
Successful drilling and production programs in the
Marcellus have occurred in vertical as well as horizontal holes. Vertical holes, however, are essentially hit or
miss in terms of the number and extent of J1 joints they
encounter. Assuming that a vertical hole intersects a set
of J1 joints, production will be limited to the area encompassing that set (Figure 15A). Similarly, a horizontal
well-oriented parallel with J1 will have limited produc-
tion (Figure 15B). The most successful wells have been
horizontal wells drilled normal to J1 (Figure 15C) because, in this manner, the lateral leg will intersect numerous joint sets and take full advantage of the interconnectedness that otherwise would not exist.
WATER MANAGEMENT ISSUES
IN PENNSYLVANIA
Typically, a petroleum geology developments article
might conclude here after having discussed the discovery, geology, extent, and drilling activity of the subject
reservoir. In this case, however, water management has
become a very important component of the Marcellus
Shale gas play and is integral to the discussion of how
the play is progressing in Pennsylvania. Specifically, water
management is an essential part of the play for the following reasons: scale, logistics, and contending uses.
Carter et al.
241
tal Marcellus wells completed between 2005 and 2010,
we have determined that an average Marcellus frac job
uses 2.9 millions of gallons (109,777 L) of water (Pennsylvania Internet Record Imaging System/Wells Information System, 2011).
Also very important is logistics. The extraction,
transport, handling, reuse, and ultimate disposal of
water and produced formation waters from the fracing
process have been argued to impact the state’s infrastructure, surface water and groundwater supplies, and
overall environmental health. For reasons such as these,
the state’s basin commissions have become very active
in the Marcellus Shale gas play.
Last but not the least is contending uses. Pennsylvania does not regulate private water supplies, so the use
of groundwater by operators for fracing in some areas
has raised concerns by citizens who are responsible for
their own drinking water supply’s quantity and quality.
In addition, some operators are extracting surface
waters in areas of the Commonwealth where Pennsylvanians fish, boat, and otherwise enjoy certain recreational uses of water. The use of water resources for
hydraulic fracturing has created a perceived competition for these resources, even when wise water management practices are being used or drought conditions do
not exist.
The remainder of this section focuses on key legal
constructs and private water resource issues specific to
Pennsylvania, which we would argue are mostly responsible for the existing water management “landscape”
that must be navigated by operators drilling shale gas
wells in the Commonwealth.
Legal Issues
Figure 15. Diagrams showing the results of vertical versus horizontal drilling in the Marcellus Shale: (A) drainage area for a vertical well, assuming it intersects a set of J1 joints; (B) drainage area
for a horizontal well drilled along a set of J1 joints; (C) drainage
area for a horizontal well drilled normal to the J1 joint orientation.
Green arrows indicate the direction of J1 joint orientation.
First and foremost, the magnitude of water usage by
the industry for Marcellus Shale gas development is
considerable. The volume of water needed to complete
the hydraulic fracturing of a single horizontal gas well is
measured in millions of gallons, instead of the thousands of gallons typical for vertical wells completed in
Pennsylvania. Using publicly available data for horizon-
242
The Backstory of the Modern Marcellus Shale Play
It is true that the Commonwealth of Pennsylvania is fortunate to have abundant natural resources, from wide
open spaces and state forests, to water and wildlife, to
those treasures beneath our feet—rich soils, mineral deposits, and fossil fuels. The Pennsylvania State Constitution (1968) addressed these natural resources and the
public estate in article 1, section 27:
“The people have a right to clean air, pure water,
and to the preservation of the natural, scenic, historic
and esthetic [sic] values of the environment. Pennsylvania’s public natural resources are the common property of all the people, including generations yet to
come. As trustee of these resources, the Commonwealth shall conserve and maintain them for the benefit of all people.”
Figure 16. Location and extent of the three major river basins in Pennsylvania: the Ohio River, Susquehanna, and Delaware.
Water, essential for life, and arguably the most important of these resources, has been subject to various
rules and management practices since the 1800s (Department of Environmental Protection, 2009b). The
progression of water management approaches to the
current day could be referred to as a “patchwork quilt”
of state and federal statutes stitched together by common law. An understanding of how these statutes and
legal concepts relate to issues of water management is
important for evaluating how productive the Marcellus
Shale gas play may ultimately be, and is, therefore, an
important chapter in the Marcellus backstory.
Common Law
Common law provides the basis for Pennsylvania’s earliest water management decisions and remains at the
core of its water management practices even today
( Weston, 2008). Common law, or case law, is built over
time through judicial rulemaking that generally addresses matters of nuisance, trespass, and negligence.
In other words, common law is not based on scientific
principles, always subject to change, and inherently
incomplete—that is, rulemaking voids remain because
the body of common law is aggregated from individual
legal decisions (Bishop, 2006; Weston, 2006).
Common law, as applied to water resource management, has defined four categories of water resources:
diffused surface waters, surface waters, groundwaters in
subterranean streams, and percolating groundwaters
(Weston, 2008). These admittedly awkward classifications result from the limited knowledge of hydrogeologic concepts that existed in the 1800s when this terminology was developed (Weston, 2008). In the simplest of
terms, however, water occurs as surface water and groundwater, and different rules apply to the allocation of each
under common law.
The “riparian rights” doctrine applies to the use of
surface water, that is, the use of water flowing in a stream
on or adjacent to riparian land. In Pennsylvania, the riparian rights doctrine also incorporates the “reasonable
Carter et al.
243
Table 3. Number of Private Wells and Population Served by Pennsylvania Counties*
County
Adams
Allegheny
Armstrong
Beaver
Bedford
Berks
Blair
Bradford
Bucks
Butler
Cambria
Cameron
Carbon
Centre
Chester
Clarion
Clearfield
Clinton
Columbia
Crawford
Cumberland
Dauphin
Delaware
Elk
Erie
Fayette
Forest
Franklin
Fulton
Greene
Huntingdon
Indiana
Jefferson
Juniata
Lackawanna
Lancaster
Lawrence
Lebanon
Lehigh
Luzerne
Lycoming
McKean
Mercer
Mifflin
Monroe
Montgomery
Montour
244
No. Wells
County
Population
HU**
Population
per HU**
No. People Served
by Private Wells†
People Served by
Private Wells† (%)
15,655
10,676
10,774
14,271
13,472
38,847
12,790
15,113
45,507
27,073
7149
1602
11,793
6986
49,316
6946
6842
3119
11,292
24,671
19,587
21,655
6757
3312
22,122
7523
4664
14,455
4444
3938
10,118
13,180
7548
5364
12,055
50,966
12,937
13,034
17,465
23,307
17,441
6321
17,996
6729
36,569
30,716
3315
78,274
1,336,449
73,478
186,093
47,919
336,523
130,542
60,967
541,174
152,013
163,029
5913
56,846
123,786
376,396
41,699
78,097
37,182
63,202
86,169
195,257
237,813
547,651
34,878
275,572
145,351
4802
121,082
13,837
39,550
44,164
89,994
46,083
20,625
219,039
422,822
96,246
113,744
291,130
328,149
118,710
47,131
121,003
46,197
95,709
678,111
17,735
30,141
580,738
31,757
76,336
21,738
134,482
54,349
27,058
199,934
59,061
67,374
4399
27,380
46,195
139,597
18,022
34,300
16,478
25,598
40,462
77,108
102,684
211,024
17,249
108,585
61,406
8445
48,629
6184
15,982
19,286
34,770
21,242
8505
91,707
156,462
38,844
44,634
118,335
138,724
49,580
21,454
48,689
19,641
54,823
265,856
6885
2.60
2.30
2.31
2.44
2.20
2.50
2.40
2.25
2.71
2.57
2.42
1.34
2.08
2.68
2.70
2.31
2.28
2.26
2.47
2.13
2.53
2.32
2.60
2.02
2.54
2.37
0.57
2.49
2.24
2.47
2.29
2.59
2.17
2.43
2.39
2.70
2.48
2.55
2.46
2.37
2.39
2.20
2.49
2.35
1.75
2.55
2.58
40,655
24,569
24,928
34,790
29,698
97,209
30,721
34,053
123,177
69,681
17,299
2153
24,484
18,720
132,971
16,072
15,578
7038
27,880
52,540
49,599
50,152
17,536
6697
56,142
17,807
2652
35,992
9944
9745
23,170
34,113
16,375
13,008
28,793
137,730
32,055
33,215
42,968
55,132
41,759
13,886
44,724
15,827
63,841
78,346
8539
51.9
1.8
33.9
18.7
62.0
28.9
23.5
55.9
22.8
45.8
10.6
36.4
43.1
15.1
35.3
38.5
19.9
18.9
44.1
61.0
25.4
21.1
3.2
19.2
20.4
12.3
55.2
29.7
71.9
24.6
52.5
37.9
35.5
63.1
13.1
32.6
33.3
29.2
14.8
16.8
35.2
29.5
37.0
34.3
66.7
11.6
48.1
The Backstory of the Modern Marcellus Shale Play
Table 3. Continued
County
Northampton
Northumberland
Perry
Philadelphia
Pike
Potter
Schuylkill
Snyder
Somerset
Sullivan
Susquehanna
Tioga
Union
Venango
Warren
Washington
Wayne
Westmoreland
Wyoming
York
No. Wells
County
Population
HU**
Population
per HU**
No. People Served
by Private Wells†
People Served by
Private Wells† (%)
17,456
9482
11,112
486
16,511
5273
14,685
6913
11,228
4254
12,682
9955
6178
9757
10,285
13,849
16,997
22,460
7804
43,441
247,105
96,771
41,172
1,585,577
27,966
16,717
152,585
36,680
78,218
6104
40,380
41,126
36,176
59,381
45,050
204,584
39,944
370,321
28,076
339,574
95,345
41,900
17,063
674,899
30,852
11,334
66,457
13,629
35,713
5458
20,308
18,202
12,886
26,961
22,236
84,113
28,480
153,554
11,857
134,761
2.59
2.31
2.41
2.35
0.91
1.47
2.30
2.69
2.19
1.12
1.99
2.26
2.81
2.20
2.03
2.43
1.40
2.41
2.37
2.52
45,241
21,899
26,813
1142
14,967
7777
33,717
18,605
24,591
4757
25,217
22,493
17,344
21,490
20,837
33,684
23,839
54,166
18,479
109,464
18.3
22.6
65.1
0.1
53.5
46.5
22.1
50.7
31.4
77.9
62.4
54.7
47.9
36.2
46.3
16.5
59.7
14.6
65.8
32.2
*Data based on 1990 U.S. Census report.
**HU = Housing units.
†
Estimated values.
use” doctrine. This means that within the Commonwealth’s borders, people have certain nonexclusive
rights of use, otherwise known as “usufructuary” rights,
with respect to water but they cannot claim absolute
ownership (Cunningham et al., 1993; Marquitz, 2003;
Bishop, 2006; Dismukes, 2006; Weston, 2008; Department of Environmental Protection, 2009b). Along these
same lines, riparian rights cannot be severed from the
land with which it is associated (Bishop, 2006).
The American doctrine of reasonable use (commonly referred to as the “American Rule” or “reasonable
user” doctrine) applies to the use of groundwater in
Pennsylvania. According to this doctrine, a landowner
can use as much groundwater beneath his property as
needed for “natural and ordinary use,” regardless of impact to neighboring groundwater supplies (Cunningham
et al., 1993; Marquitz, 2003; Bishop, 2006 Dismukes,
2006; Department of Environmental Protection, 2009b).
This usage would only be restricted if it were found by
the courts to be negligent or have caused malicious damage to another’s property (Marquitz, 2003; Bishop, 2006;
Department of Environmental Protection, 2009b).
State Legislation
Not one comprehensive piece of legislation regulates
water resources allocation in Pennsylvania. Over the
years, however, a short list of state laws has been interpreted to require permits, registration, and/or reporting
of withdrawals for certain uses and from specific water
sources, including the Water Rights Act of 1939, the
Pennsylvania Safe Drinking Water Act of 1971 (and
subsequent amendments), the Water Well Drillers License Act of 1956, and Water Resources Planning Act
of 2002 (Weston, 2008).
Perhaps the most relevant of these to the current
discussion are the Safe Drinking Water Act and the
Water Resources Planning Act—the former being directly applicable to water withdrawals that industry pursues with respect to the extraction of Marcellus Shale
gas and the latter likely becoming a major influence on
water management policy (and perhaps even natural gas
drilling and production practices) in the future. Based
on the 1996 court ruling in the case of Oley Township
et al. versus DEP and Wissahickon Spring Water, Inc.,
DEP is obligated under the Safe Drinking Water Act to
Carter et al.
245
Figure 17. Estimated percent of Pennsylvania population relying on private water wells (based on 1990 U.S. Census data; S. Reese,
Pennsylvania Geological Survey, compiler).
ensure that none of the water withdrawals it permits
will impact other water resources (Bishop, 2006), a very
specific directive. The Water Resources Planning Act
requires that the existing State Water Plan be periodically updated and expanded to address matters of water
budget, sustainable usage, and future water resource
planning. To this end, the DEP and various stakeholders
formed regional and statewide committees in 2003 and
began working to develop a new State Water Plan that
would replace the one created in 1983. The new document was signed into law in March 2009 (Department
of Environmental Protection, 2009b). One of the derivative products of the new State Water Plan is an online
basin-based water atlas for water resource planning. In
addition, the Plan has incorporated certain registration
and reporting requirements to track water usage; as an
example, all water withdrawals in excess of 10,000 gal/
day (37,854 L/day) must be reported to the Commonwealth (Carter, 2010). The reporting, management, and
246
The Backstory of the Modern Marcellus Shale Play
planning elements of the new State Water Plan could
very well be used as a model for oil and gas permitting,
unitization, and production in the future.
Federal Law: Implications for State Activities
In Pennsylvania, there exists an additional level of oversight when it comes to water resource allocation and
planning. Three different organizations associated with
the major river basins in the…

Purchase answer to see full
attachment

User generated content is uploaded by users for the purposes of learning and should be used following Studypool’s honor code & terms of service.