the role of natural gas in a low-carbon energy economy (3 briefing
TRANSCRIPT
The Role of Natural Gas in a Low-Carbon Energy Economy
Natural Gas and Sustainable Energy Initiative
Christopher Flavin
Saya Kitasei
April 2010
Briefing Paper
1
The Role of Natural Gas in a
Low-Carbon Energy Economy*
Christopher Flavin and Saya Kitasei
Executive Summary
Growing estimates of natural gas resources, including a new category of ―unconventional‖ gas,
suggest that accessible supplies of this least carbon-intensive of the fossil fuels may be far more
abundant than previously assumed. This unexpected development creates opportunities for
deploying natural gas in a variety of sectors—including power generation, industry, and
transportation—to help displace oil and coal, thereby reducing greenhouse gas emissions and
improving air quality.
Beyond providing a cleaner, market-ready alternative to oil and coal, natural gas can facilitate
the systemic changes that will underpin the development of a more energy-efficient and
renewable energy-based economy. For example, smaller, distributed generators, many producing
usable heat as well as electricity, could generate economical, low-emission replacements for a
large fraction of currently operating conventional power plants, providing flexible back-up to the
variable output of the solar and wind generators that will comprise a growing share of the electric
power system.
All of these gains are contingent on the development of sound public policy to incentivize and
guide the transition. Critical policy decisions that are now pending include: electric power
regulation at the local, state, and federal levels; effective federal and state oversight of the natural
gas exploration and extraction process; future Environmental Protection Agency (EPA)
regulatory decisions under the U.S. Clean Air Act; and putting a price on greenhouse gas
emissions.
* This is the first in a series of briefing papers to be issued by the Worldwatch Institute’s Natural Gas and
Sustainable Energy Initiative examining the complementary roles of natural gas, renewables, and efficiency. This
first paper provides an overview of the role that natural gas currently plays in the energy system and a roadmap for
the role that gas could play in spurring the transition to a low-carbon economy in the decades ahead. Future papers
will focus on a range of specific issues, from the local environmental problems caused by shale gas development to
options for integrating natural gas generation with large wind farms.
2
I. The Renaissance of Gas
Natural gas was first developed as a modern fuel, together with oil, in the late 19th century. Most
of the early gas resources were co-located with oil, and this associated gas was extracted almost
as an afterthought as the oil industry took off in the early 20th century. Like oil, natural gas
began to be used to a limited extent in the industrial, residential, and commercial sectors as a
feedstock and to heat buildings prior to World War II.
Following the war, the United States and a few other countries began to build the extensive and
expensive pipelines needed to make gas a mainstay of the U.S. energy economy, and the first
generation of gas-fired power plants was built. As a byproduct of oil production, natural gas was
cheap, and by the early 1970s, provided 30 percent of the U.S. energy supply, most of it in
industry and buildings.1 (See Figures 1 and 2.) But that was the peak. As U.S. oil supplies
dwindled, so did gas, hampered by government price controls that discouraged exploration.
By the late 1970s, most experts believed
that natural gas had entered a period of
inevitable decline. Policymakers were so
worried that, for a time, Congress made it
illegal to build gas power plants in the
United States.2 While gas maintained its
dominant position as an industrial fuel and
the most economical means of heating
homes, by the 1990s, it had fallen to less
than 24 percent of the U.S. energy supply
and stayed close to that level for the next
decade and a half.3 Modest demand
growth in the 1990s and early 2000s was
met by Canadian imports.4
The 1990s were marked by relatively low and stable gas prices as U.S. and Canadian suppliers
easily kept up with demand growth. But
soaring oil prices, together with falling
reserves of conventional natural gas,
drove gas prices from just over $2 per
million BTU in 2002 to as high as $13 per
million BTU in 2008, making many
potential users reluctant to invest in the
fuel.5 Since then, gas prices have
moderated somewhat—ranging between
$2.50 and $6 per million BTU in 2009
and 2010.6 Still, price volatility remains
the Achilles’ heel of natural gas,
particularly when compared with coal.
0
20
40
60
80
100
120
19
49
19
54
19
59
19
64
19
69
19
74
19
79
19
84
19
89
19
94
19
99
20
04
20
09
Qu
ad
rill
ion
Btu
Figure 1. Trends in U.S. Primary
Energy Consumption, 1949-2008
Renewables
Nuclear
Natural Gas
Petroleum
Coal
Source: EIA
0
2
4
6
8
10
19
49
19
54
19
59
19
64
19
69
19
74
19
79
19
84
19
89
19
94
19
99
20
04
20
09
Tri
llio
n C
ub
ic F
eet
Figure 2. U.S. Natural Gas
Consumption by End Use, 1949-2009
ElectricityIndustrialResidentialCommercialVehicle Fuel
Source: EIA
3
Tempering coal’s price advantage are the substantial environmental advantages of natural gas,
which have gained economic significance as clean air standards have become progressively
tighter in recent decades. Burning natural gas produces virtually none of the sulfur, mercury, or
particulates that are among the most health-threatening of pollutants that result from coal
combustion.7 A National Research Council study published in 2009 estimated that the
environmental damages associated with electricity from natural gas are 95 percent lower than
from coal.8 Although natural gas does produce nitrogen oxides and carbon monoxide and is an
important contributor to ozone pollution in some areas of the United States, these can be reduced
substantially with widely available emissions controls.
Growing concern about climate change in recent years has also worked in favor of natural gas.
Gas contains 25 percent less carbon than oil and half as much carbon as coal.9 Planned and
proposed federal and state actions to curb greenhouse gas emissions—from stricter requirements
for emissions control technology to renewable or clean energy portfolio standards to a cap on
carbon—all expose oil and coal investments to much higher risk than natural gas.
Environmental considerations have helped revive interest in natural gas as a source of electricity
in recent years. Since the 1990s, 65 percent of the new capacity added to the U.S. power grid has
consisted of a new generation of efficient gas-fired power plants, compared with 2 percent for
coal.10
(See Figure 3.) While much of this capacity remains underutilized due in part to relatively
high gas prices, the decline in prices in 2009 boosted natural gas to 23 percent of U.S. power
generation, up from 20 percent in 2007 and just 12 percent as recently as 1990. During the same
period, coal declined from 52 percent of U.S. electricity to 45 percent.11
(See Figure 4.)
Outside of the power sector, other applications of natural gas have begun attracting interest as
well—particularly in the face of dramatically higher oil prices. From 1995 to 2005, oil cost an
0
10
20
30
40
50
60
70
80
Gig
aw
att
s
Figure 3. Existing U.S. Generating Capacity by Fuel Type and Initial
Year of Operation
Wind, Solar, Geothermal and BiomassHydroelectric ConventionalNuclearNatural Gas (Peaking Plants)Natural Gas (Steam Turbines and Combined Cycle)PetroleumCoal
Source: EIA
4
$0
$5
$10
$15
$20
$25
$0
$20
$40
$60
$80
$100
$120
$140
1994
1995
1996
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
doll
ars
per
MM
Btu
doll
ars
per
bar
rel of
oil
equiv
alen
t
Year
Figure 5. U.S. Oil and Natural Gas
Prices, 1994-2009
Henry Hub Natural Gas Futures Price
Crude Spot Price (Cushing)
average of 34 percent more than natural
gas.12
(See Figure 5.) And in the past few
years, as world oil prices have
skyrocketed, North American gas prices
have not risen as rapidly. In 2008 and
2009, the average price of oil was more
than double the price of gas, and by
March 2010, oil was nearly three times as
expensive.13
Transportation, the sector where oil is
dominant, will likely be affected most by
the widening price gap between oil and
gas. Boosted by a new generation of
compressed-gas fuel tanks, natural gas
vehicles have already become popular in countries such as Italy and Pakistan, where they are
seen as an economical way to reduce dependence on oil.14
In the United States, where gasoline
prices have been relatively low by
international standards, natural gas
vehicles have never been as popular,
but many local governments have
turned to gas-powered buses to
reduce fuel costs and the local air
pollution from diesel buses.15
Texas
businessman T. Boone Pickens has
proposed a nationwide effort to
convert heavy-duty trucks to run on
natural gas, in part to minimize U.S.
dependence on foreign oil.16
Recent studies conclude that, beyond
their ability to reduce local air
pollution, natural gas vehicles also
lower greenhouse gas emissions by roughly 25 percent compared with oil, far less than the
reductions possible in power generation but significant nonetheless.17
The big question now
facing energy planners is whether sufficient natural gas will be available at a competitive price to
allow for significant displacement of oil in transportation and coal in power generation. The
answer to that question will likely be determined in large measure by efforts to develop new
sources of natural gas, which has already had profound effects on the U.S. energy industry in
recent years.
0%
10%
20%
30%
40%
50%
60%
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Figure 4. Share of U.S. Electric
Generation from Coal and Gas
1990-2009
Coal
Natural Gas
Source: EIA
Source: EIA
5
II. The Unconventional Gas Revolution
A newfound abundance of natural gas promises to tip the fossil fuel balance further in its favor.
Gas production in the United States peaked in the early 1970s, along with oil, but in recent years
technology advances have dramatically reversed the decline.18
Advances in horizontal drilling
and hydraulic fracturing have unlocked gas resources in ―unconventional‖ reservoirs, such as
tight sands, coal bed methane, and shale rock rich in organic materials. As a result, resource
estimates have increased sharply, and as lessons learned in the United States are applied to
exploration and production of unconventional resources internationally, natural gas has the
potential to shed the supply, price volatility, and energy security concerns that have surrounded it
during the last few decades.
Unconventional gas is found in low-porosity sedimentary rock formations that act as both
sources and reservoirs for hydrocarbon deposits. Because of their low porosity, gas is more
difficult to extract from unconventional formations than from conventional gas reservoirs, which
generally contain stores of hydrocarbons that originated in other formations. But as conventional
resources have been exhausted, the industry has turned its attention to new sources of gas that
were previously dismissed as too difficult and expensive to extract. In the 1970s, gas producers
began to develop tight sands that they had discovered in the course of exploration for
conventional gas. Using hydraulic fracturing and horizontal drilling, they were able to recover
gas from these resources economically, largely in the Rocky Mountain states. Since then, tight
sands have grown to account for more than 30 percent of all gas production in the United
States.19
Natural gas is also found in coal seams, where it can pose serious health and safety risks to coal
miners and can, if leaked to the atmosphere, contribute to climate change. Methane is adsorbed
onto the pores of the coal, which has very low porosity. This methane, which would otherwise
leak into the atmosphere over time, can be extracted economically by drilling into the coal seam.
Coal bed methane development, most of it at relatively shallow depths, has been expanding since
1989, starting in Alabama, New Mexico, and Colorado, and later in Utah, Virginia, and
Wyoming. Total U.S. production of coal bed methane reached almost 2 trillion cubic feet in
2008—10 percent of total U.S. gas production.20
In the past few years, the focus of the gas industry has turned to a third unconventional source:
deep shale formations, or non-porous sedimentary rock that mostly lies thousands of feet
underground. Starting in the 1990s, independent gas producers began to develop a technique,
known as hydraulic fracturing, for injecting high-pressure water into these deep formations,
allowing the gas to be released and brought to the surface. First deployed on a large scale in
Texas’s Barnett Shale, the technique has subsequently been adapted to a range of shales in other
parts of the country, each of which has its own geological distinctions. When natural gas prices
shot upward after 2005, the shale ―gold rush‖ was on. The largest gas-bearing shale formation,
the Marcellus Shale, extends across five states from West Virginia to New York, and is attracting
great attention in the northeastern region where energy prices are high and most gas is imported
from over 1,000 miles away.21
(See Map 1.)
6
While shale rock does not give up its
methane easily, this is more than balanced
by its abundance. The Potential Gas
Committee, an independent authority on
gas supplies based at the Colorado School
of Mines, estimated potential U.S. natural
gas resources in 2008 to be 1,836 tcf, up
39 percent from 2006—with the difference
due mainly to a steep increase in estimates
of recoverable shale gas.22
Proven reserves
have increased 13 percent to 238 tcf,
bringing total gas resources to 2,074 tcf.
Assessments by ICF International, the U.S.
Energy Information Administration (EIA),
and Navigant Consulting Inc. confirm the magnitude of this resource.23
These figures suggest
that U.S. supplies of natural gas could last for 90 years at current rates of consumption.24
And
some experts expect the resource estimates to continue rising as exploration proceeds and as
extraction techniques are further developed.25
Since 1990, unconventional gas
production has already increased fourfold,
with an even steeper rise in the past few
years contributing to a sharp decline in
gas prices and a collapse in the North
American market for imported liquefied
natural gas.26
(See Figure 6.) Surprisingly,
the boom has only slowed marginally in
the face of a steep recession and a sharp
decline in the price of natural gas since
2008, suggesting that unconventional gas
may be cheaper to produce than
conventional gas.27
The breakeven price
for shale gas in various U.S. basins is
reported to range from just under $3 to
$4.50 per million BTU.28
Notably, some
of the most recent basins to be tapped, including the Marcellus, are among the least expensive.
Moreover, unconventional gas costs are likely to continue to decline since the technology is still
relatively immature and is continuing to advance. If these new gas supplies are sufficiently
abundant and economical to end the boom-and-bust cycle that’s marked the industry for decades,
gas could make major inroads in energy markets in the years ahead.
Source: Navigant Consulting
Map 1. Major U.S. Shale Gas Resources and
Existing Pipeline Infrastructure
0
2
4
6
8
10
12
tril
lion
cu
bic
feet
Figure 6. Unconventional Gas
Production in U.S. Lower 48 States,
1990–2009
Shale
Coalbed
Tight Gas
Source: EIA
7
The rise of gas stands in sharp contrast to the three-decade decline in U.S. oil production. Since
1990, total U.S. gas production
has increased 20 percent while
oil production fell 33 percent.29
(See Figure 7.) Today, the
United States produces more
than twice as much gas as it
does oil, and that gap will
almost certainly widen in the
coming years.30
After decades
of selling their domestic fields
to independent producers,
major oil companies such as
ExxonMobil and BP have
signaled a significant shift in
their thinking about the future
evolution of energy markets by purchasing tens of billions of dollars of gas reserves from those
same independents in the past few years.
III. Generating Low-Carbon Electricity
The prospect of more abundant and economical gas supplies, together with the increasing
urgency of the climate problem, is drawing increased attention to the role that natural gas might
play in the transition to a low-carbon power sector. In addition to the emissions reductions it
offers over coal, natural gas is a more flexible fuel, with the ability to provide backup power on a
range of scales to an electricity system that will include a rising share of variable wind and solar
energy, combined heat-and-power, and distributed generation.
Generating electricity from natural gas rather than coal yields dramatic reductions in carbon
dioxide emissions. Natural gas contains only half as much carbon per unit of energy as coal does,
and gas also lends itself to a more efficient form of power generation known as combined-cycle
technology. This consists of one or more combustion turbines (similar to jet engines) that are
equipped with heat-recovery steam generators to capture heat from the combustion turbine
exhaust. The heat-recovery steam generator powers a steam-turbine generator to generate
additional electric power. Use of the otherwise wasted heat in the turbine exhaust gas results in
high thermal efficiency compared to other combustion technologies, yielding efficiencies above
45 percent (compared with 30–35 percent for most coal plants). New combined-cycle gas plants
produce 55 percent less carbon dioxide than new coal plants do and 62 percent less than the
average U.S. coal plant.31
(See Figure 8.)
Although coal is the leading source of electricity in the United States, most of the new power
plants added to the U.S. electricity grid since 1990 are powered by natural gas.32
This includes
201 gigawatts (GW) of highly efficient combined-cycle power plants and 107 GW of relatively
inefficient gas-turbine peaking plants that are typically turned on only when needed during peak
demand periods.33
Altogether, natural gas power plants now represent 31 percent of U.S.
generating capacity (excluding gas-fired peaking plants, which contribute another 13 percent),
0
2
4
6
8
10
12
0
10
20
30
40
50
60
70
1950 1960 1970 1980 1990 2000 2010
Cru
de
Oil
(m
illi
on b
arre
l o
f o
il
equiv
alen
ts p
er d
ay)
Nat
ura
l G
as (
bil
lio
n c
ub
ic f
eet
per
day
)
Figure 7. Annual Oil and Gas Production in
the U.S. Lower 48 States,
1950–2009
Lower 48 Crude Oil Production
Lower 48 Natural Gas Production
Source: EIA
8
compared with 33 percent for coal.34
Even with the peaking plants excluded, gas-fired power
generators are under-utilized, operating at an average of only 42 percent of their capacity.35
The carbon emissions of the U.S. power sector could be decreased significantly simply by
running some of the existing plants more frequently and operating coal plants less, which would
have a significant impact on carbon emissions. In a 2010 study, the Congressional Research
Service found that if existing combined-cycle plants could be operated at 85 percent of their
capacity, gas could replace nearly one-third of coal generation and reduce power sector carbon
dioxide emissions by 19
percent. Taking into
account transmission and
siting constraints, however,
the author estimated that
the amount of current coal
generation that could be
displaced by natural gas
might be closer to 9
percent.36
As the disparity
in these numbers illustrates,
reforming power
generation will necessitate
a systemic approach to the
entire power sector.
Besides being more efficient and cleaner than their coal counterparts, combined-cycle power
plants are also cheaper and quicker to build. A survey of actual fossil, nuclear, and renewable
power projects in 2008 determined that natural gas combined-cycle plants had the lowest
construction costs of any available generating technology, under half the cost of a new
pulverized coal plant and just one-fifth the estimated cost of a new nuclear plant.37
Under most assumptions for construction costs, government incentives, and carbon controls,
combined-cycle plants are an extremely competitive source of electricity.38
However, the cost of
gas-fired power is extremely sensitive to the price of gas, which has been highly volatile in
recent years.39
(See Figure 9.) At the average price of gas in 2009, the levelized cost was 5.5
cents per kilowatt-hour, compared with 8.6 cents per kilowatt-hour based on average 2008 gas
prices.40
The recent decline in gas prices has already led utilities to increase the utilization of their gas
plants, raising the gas share of generation in 2009 to 23 percent, higher than at any time in the
past three decades.41
The resulting drop in coal-fired power generation was responsible for
almost half of the nearly 10-percent decline in U.S. carbon dioxide emissions from energy
consumption between 2007 and 2009.42
And some utilities are deciding to make these changes
permanent. Faced with the steep cost of installing pollution controls on its coal plants, North
Carolina-based Progress Energy announced plans to permanently close 11 of its dirtiest coal
plants over the next eight years, a total of almost 1,500 megawatts (MW), replacing them
primarily with natural gas plants.43
982
855
804
778
420
362
0 500 1000 1500
Average U.S. Coal Plant
New PC Boiler (Subcritical)
New PC Boiler (Supercritical)
New IGCC Turbine
Average U.S. Gas Plant
New NGCC Turbine
CO2 Emissions based on net output (kg/Mwh)
Figure 8. Carbon Dioxide Emissions from Gas
and Coal Plants, 2007
PC = Pulverized coal; IGCC = Integrated Gas Combined Cycle;
NGCC = Natural Gas Combined Cycle. Source: DOE, EIA
9
In addition to the emissions savings they represent over coal plants, natural gas generators are
better suited to play a complementary role in a generation mix that includes a growing amount of
wind and solar power. Unlike coal plants, gas plants can be more easily turned on and off,
enabling utilities to use them to balance variable generation from renewable energy sources. For
this reason, nine solar thermal power plants built in California during the 1980s and early 1990s
were designed as gas-solar hybrids, with auxiliary natural gas boilers or heat transfer fluid
heaters to provide backup generation.44
Existing combined-cycle and peaking plants already
provide de facto backup electricity for wind power in some parts of the country.
In some cases, utilities may also be able to retrofit existing conventional power plants with
renewable generators to reduce fossil fuel consumption and greenhouse gas emissions. The
Florida Power and Light (FPL) Company is adding a 75-megawatt solar thermal field to a much
larger natural gas plant in Indiantown, Florida.45
However, the potential of such large-scale
retrofits will be limited by the land and resource requirements of renewable generating
technologies. In the future, a new generation of gas-fired generators—from gas turbines to fuel
cells—can be deployed as a complement to wind and solar power, both in dedicated gas-
renewable hybrid systems and as independent components of a renewable-rich energy portfolio.
Natural gas also lends itself to applications that increase overall energy efficiency. Like all
thermal power plants, natural gas plants create heat as a byproduct while generating electricity.
While most plants discard this heat as waste, they can instead be designed to capture it for use in
space heating, a process called cogeneration or combined heat-and-power (CHP). Because it can
be scaled more easily than coal, natural gas is the most common fuel used in combined heat-and-
power applications, which are typically industrial scale or smaller. Cogeneration has enjoyed
little policy support in the United States, and as a result it provides just 8 percent of the country’s
electricity.46
However, the figure is much higher in other countries where cogeneration has
received more support: 39 percent in Finland and 52 percent in Denmark.47
A new generation of smaller, ―distributed‖ gas-fired generators that harness waste heat for
heating and cooling can provide better environmental performance than even the most efficient
central-station plants while adding an economical and flexible element to the power grid.
5.1 to 8.9
5.7
5.8
6.5 to 9.6
7.3 to 7.7
10.5 to 10.8
10.8
11.0 to 15.0
13.1
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0
Gas (CC)
Wind
Geothermal
Biomass
Coal (Supercritical)
Coal (IGCC)
Nuclear
Gas (Fuel Cell)
Solar PV
Cents per kWh
Figure 9. Levelized Cost of Energy From New Power Plants, 2008
Capital
O&M
Fuel - Low
Fuel - High
Source: Based on Lazard
10
Technologies ranging from reciprocating engines (similar to those used in automobiles) to gas
turbines and fuel cells can be added to factories, commercial buildings, and even family homes.
Located within the local power distribution system, micro-power plants avoid the need to add
expensive and hard-to-site transmission lines and—unlike ―baseload‖ coal and nuclear plants—
can easily be turned on and off as needed to meet fluctuating power demand. Compared with
electricity from a conventional power plant and heat from a separate gas-fired furnace, a
cogeneration system typically has an efficiency of between 65 and 80 percent and would allow
even greater emissions reductions than combined-cycle plants.48
Volkswagen announced plans in 2009 to produce 10,000 miniature 20-kilowatt gas-fired power
plants per year, based on the internal combustion engines it uses in its Golf automobiles. These
units are designed for use in individual residences and will operate at up to 94-percent efficiency,
providing heat, hot water, and electricity.49
Dubbed ―schwarmpower,‖ this network of tiny power
plants could, within a decade, provide 2,000 MW of capacity (equivalent to two nuclear plants)
that will be digitally controlled and used to back up the variable wind power that already
provides some 40 percent of the electricity in three German states.50
The prospect of widespread deployment of small-scale solar power plants in and near the world’s
cities in the years ahead will likely spur growing interest in micro-power plants using natural gas.
As small-scale solar and gas generators are integrated into local, low-voltage power systems,
both will require new laws that allow small businesses and consumers to access the local grid at
a competitive price.
IV. Overcoming Environmental Challenges
While natural gas has many environmental advantages over the other fossil fuels, it is not
without problems of its own. The rapid development of unconventional gas in recent years has
raised a host of environmental and health concerns, generating extensive controversy at the local,
state, and national levels. Gas development has been particularly controversial in the
northeastern states of Pennsylvania and New York, where state regulators and citizens who had
no experience with oil and gas development were ill-prepared for the unconventional gas boom.
Communities that welcome the jobs and income that are flowing from the new industry are also
struggling with the disruptions and environmental problems that often accompany expanded gas
development.
To extract natural gas from tight sand, shale, and some coal bed formations, engineers utilize two
key technologies: horizontal drilling and hydraulic fracturing. Shale gas extraction is begun by
drilling a vertical well to the depth of the reservoir, then gradually turning the drill bit 90 degrees
until it is oriented parallel to the productive layer. Horizontal wells offer greater contact area
with the reservoir than vertical wells, providing an important boost to production in strata that
have low permeability. In addition, they greatly reduce the surface impact of drilling operations
because engineers can drill multiple wells from a single well pad and extend those wells laterally
for thousands of feet. According to estimates from the U.S. Departments of Interior and Energy,
in the Fayetteville Shale a four-well horizontal drilling pad with roads and corridors would
disturb about 7.4 acres on the surface, whereas the 16 vertical drilling pads that would be
11
necessary to produce the same square mile of the formation, together with roads and utility
corridors, would disturb some 77 acres.51
To free up the gas that is tightly bound in the impermeable rock, developers typically inject wells
with millions of gallons of water mixed with chemical additives and sand under high pressure.
The fracturing or ―fracking‖ fluid widens and props open tiny fractures in the shale, increasing
the reservoir’s permeability and allowing gas to escape more freely. Fracking fluids can contain
small concentrations of toxic chemicals that improve the effectiveness of the procedure,
including biocides, corrosion inhibitors, and thickening agents.52
No federal law currently
requires companies to disclose the chemicals used in fracturing fluids—a condition that
companies argue is necessary to protect their trade secrets.
Once fracking fluid has come into contact with the rock formations through which the wellbore
travels, it can mix with methane, highly concentrated salts, and naturally occurring radioactive
materials (NORM).53
While some portion of injected fracking fluid remains underground,
produced water brought up from the target formation must be disposed of safely. Depending on
the state regulations in place, companies may be required to re-inject produced water into
disposal wells or to send it to wastewater treatment facilities, where it must generally be
transported in tankers. If a leak or spill occurs at any point during the production, transportation,
or disposal processes, produced water can pollute groundwater and surface waters. Similarly, the
volumes of produced water generated by increased levels of gas drilling can overwhelm
wastewater treatment facilities, as occurred in Pennsylvania’s Monongahela River during the fall
of 2008.54
As gas production expands to new regions, improving wastewater disposal and
treatment practices and capacity will be critical.
Produced water or natural gas can contaminate underground aquifers if improperly lined and
cased wellbores to leak under the pressure of hydraulic fracturing. In 2007 in Bainbridge, Ohio, a
well that had been drilled almost 4,000 feet into a tight sand formation through a layer of gas-
bearing shale was not properly sealed with cement, allowing gas from the shale layer to leak into
an underground source of drinking water. The methane eventually built up until an explosion in a
resident’s basement alerted state officials to the problem.55
The sheer volume of water consumed during hydraulic fracturing could make unconventional gas
production costly and unsustainable in many areas of the world that are water-constrained. Each
well requires an average of 2–4 million gallons of water to fracture, depending on the
characteristics of the shale formation. Although these volumes are significant, the Department of
Energy estimates that they will represent less than 1 percent of all water usage in each
basin.56
Nevertheless, producers must work with regulators to ensure that shale gas production
does not encroach on other regional demands for water. Gas companies have begun
experimenting with reusing produced water in subsequent fracturing jobs, a practice that could
greatly reduce water consumption, transportation costs and emissions, and contamination risks.
The extraction and transport of natural gas also generates local air pollution and greenhouse
gases. Natural gas itself is made up mostly of methane, a greenhouse gas 23 times more potent
than carbon dioxide. During the production process, natural gas may be intentionally vented or
unintentionally leaked. According to the EPA, natural gas systems were responsible for 178.9
12
million metric tons of CO2-equivalent of methane in 2008—61 percent of the energy sector’s
methane emissions and 24 percent of total U.S. methane emissions.57
Efforts are underway to
capture this methane, including the EPA’s Natural Gas STAR program, which has worked with
industry to reduce methane emissions from the U.S. gas industry by 822 billion cubic feet, or 334
million metric tons of CO2-equivalent, since 1993.58
Aside from methane, the natural gas production process also emits carbon dioxide and other air
pollutants. Diesel-powered compressors, which enable gas to be transported via pipeline, emit
significant amounts of CO2 and smog-forming pollutants if they are not equipped with control
technologies. Diesel fuel, drilling equipment, and water for hydraulic fracturing all must be
trucked to drilling sites, adding additional emissions from vehicle exhaust. An Environmental
Defense Fund study found that oil and gas production in the Barnett shale basin generates more
smog-forming compounds than motor vehicles in the five counties it occupies, as well as high
levels of air toxics, and greenhouse gases equivalent to the expected impact from two 750 MW
coal-fired power plants.59
These could likely be reduced significantly if pollution controls were
required. And if widely dispersed unconventional resources allow gas to be produced closer to
the point of use, the emissions associated with transporting it could be significantly reduced.
Increased seismic activity from hydraulic fracturing is another concern. Thus far, seismic activity
from fracturing is well below the level that would be noticeable to humans and can be detected
only by very sensitive instruments. Data from these instruments can be used to predict whether
there is a risk of a larger earthquake being triggered by hydraulic fracturing. This, too, is an area
that requires responsible oversight from industry and regulators.60
Important legal and regulatory issues surrounding the production of shale gas remain unresolved.
For example, although the underground injection of fluids produced during fracturing activities
are regulated by the EPA under the Safe Drinking Water Act, the hydraulic fracturing procedure
itself is exempt, and as a result is only regulated at the state level.61
State-level regulation
currently varies widely, and the sharing and emulation of best practices among states—
particularly those in which these resources are first developed—are essential. Additional research
and industry transparency are needed to improve understanding of and decision-making about
hydraulic fracturing at the local, state, and federal levels. Unless trust can be established between
local stakeholders and gas producers, natural gas’s ability to fulfill its potential contribution to a
low-carbon energy system will be weakened.
V. Beyond North America
Although the natural gas industry has its deepest historical roots in North America, it has become
an important global fuel in recent decades. In 2008, global production of natural gas totaled 137
trillion cubic feet, with the United States and Russia responsible for 19 and 18 percent of global
production, respectively. Other leading producers include Canada, Iran, and Norway.62
(See
Figure 10.) In most countries, however, natural gas plays a much smaller role than it does in
North America, which often means higher levels of dependence on oil and coal and consequently
higher emissions of greenhouse gases.
13
25.8
24.4
7.4
7.1
6.4
5.0
3.2
3.1
3.0
2.9
0 5 10 15 20 25 30
United States
Russia
Canada
Algeria
Iran
Norway
Qatar
Saudi Arabia
Netherlands
Indonesia
trillion cubic feet
Figure 10. Top Ten Global Producers of
Natural Gas (2008)
Global proven reserves of natural gas—defined as gas that geological and engineering analyses
indicate are recoverable from known reservoirs under existing economic and operating
conditions—have risen significantly in the last three decades. According to the International
Energy Agency (IEA), only 14 percent of the world’s ultimately recoverable conventional
resources have been extracted. At current global rates of production, remaining conventional gas
resources alone could last up to 130 years.63
Many countries with extensive gas resources have
hardly begun to exploit them, and in
some cases burn them off as an
unwanted byproduct of oil
production. This is because using
gas requires extensive investment in
pipelines for distribution while
exporting it means building
complex and expensive facilities for
super-cooling and liquefying it.
The recent growth of
unconventional gas resources in the
United States has already had a
significant impact on global
markets. Surging gas supplies
contributed to a 30-percent reduction in net imports of gas to the United States between 2007 and
2009, putting downward pressure on the price of internationally traded gas in Europe and Asia.64
At the same time, efforts to identify and acquire unconventional gas resources have risen in
many countries where domestic gas has until now played little role in meeting energy needs.
Much of this effort is being led by major oil companies such as BP, ExxonMobil, Statoil, and
Total, all of which have recently gained technical expertise via acquisitions and partnerships in
North America. And much of the technology for unconventional gas production is held by
international oil and gas service companies such as Halliburton and Schlumberger, which are
actively deploying it for client companies around the globe.
Much of the exploration activity outside of North America has occurred in Europe. Most
European countries have legal systems that allow and encourage private development of gas
resources. In addition, concerns about over-dependence on Russian gas, highlighted by recent
supply disruptions caused by pricing disputes, have encouraged European governments to seek
new sources of gas. Early assessments suggest that unconventional gas resources are significant
in Europe, though not likely as abundant as in North America. Among the countries where
exploration efforts have shown the most promise are Austria, Germany, Hungary, Poland, and
Sweden.65
The IEA’s 2009 World Energy Outlook estimates that global coal bed and shale gas production,
which at 13 tcf in 2007 contributed 12 percent of worldwide natural gas supplies, will rise to 22
tcf in 2030, or 15 percent of global supplies, with most of the predicted growth coming from
North America.66
These estimates now appear conservative. Knowledge of international
Source: EIA
14
unconventional resources is extremely limited in most countries, and more research will be
required to quantify the location and volume of available supplies.
In many developing countries, even modest new supplies of natural gas could significantly
reduce dependence on imported oil and gas, providing more energy security and improving the
balance of trade. In addition, early evidence suggests that some of the world’s most coal-
dependent countries, including China, India, and South Africa, may have extensive natural gas
resources that could contribute to reduced greenhouse gas emissions. As recently as 2007,
natural gas provided only 8 percent of power generation in India and 1 percent in China,
suggesting a large potential for expansion.67
In New Delhi, the municipal government has
required that most small two-wheelers be fueled with natural gas in order to reduce air pollution,
and the city recently announced plans to replace its coal-fired power plants with natural gas.68
The United States and China have already begun to collaborate on unconventional gas resources.
On November 17, 2009, Presidents Barack Obama and Hu Jintao announced the launch of the
U.S.-China Shale Gas Resource Initiative.69
Through this program, Chinese experts will be able
to benefit from U.S. expertise in shale gas science and technology to assess and develop Chinese
shale gas resources. Similar collaborative initiatives could speed the development of
unconventional gas in other parts of the world.
VI. Unlocking the Potential
For decades, natural gas has been a neglected element of the U.S. energy portfolio. In policy
deliberations, natural gas has been linked closely to oil and sometimes to coal. The distinction
between gas and the other fossil fuels is often blurred, as is its potential to accelerate the
transition to low-carbon energy.
Together with renewable energy and energy efficiency, natural gas could transform the energy
economy over the next few decades, drastically reducing climate pollution and lowering
dependence on imported oil. Natural gas lends itself to a range of high-efficiency applications,
and it can provide the flexible backup power that will allow high levels of reliance on wind and
solar power even before economical storage technologies are developed. Moreover, in the future,
fossil natural gas could be supplemented by renewable methane gas that is extracted from
landfills, feedlots, and other biological sources.
Unfortunately, the logical alliance between natural gas and the clean energy community has been
strained in recent years as both grew rapidly at a time when electricity demand was falling. The
growth of wind power in Texas, for example, has led to charges by natural gas generators that
wind farms are getting preferential treatment from state regulators. At the same time, the
environmental controversies being stirred up by the questionable practices of some shale gas
developers have exacerbated mistrust in local communities and led some environmental groups
to oppose additional gas development. Unless these tensions can be resolved and effective clean
energy alliances are created, the potential for natural gas to contribute to a low-carbon economy
will never be realized.
15
To reach that potential, building new policy frameworks will be essential. And for that to occur,
an innovative and strategic partnership between the gas, renewables, and efficiency industries—
and the environmental community—is needed. For environmentalists, gas can broaden the range
of tools available to reduce carbon emissions and bring a strong industry to the alliance that
supports climate legislation. And for the gas industry, allying itself with those who are working
to build a low-carbon economy will facilitate a policy environment in which gas plays a growing
role even as the United States gradually reduces its dependence on oil and coal.
Important policy changes will be needed to achieve these goals:
1. Putting an Effective Price on Carbon
By attaching a cost to carbon dioxide emissions, a cap-and-trade system or a carbon tax will tend
to favor natural gas at the expense of coal and oil. According to some analysts, fuel switching in
the power sector could contribute significantly to the 17-percent emissions reductions called for
by climate legislation passed by the U.S. House of Representatives in June 2009.70
However,
other studies have found that because that bill includes a range of allowance giveaways to coal-
burning power companies based on their historic emissions, it would actually discourage fuel
switching, forcing utilities to turn to more expensive alternatives such as carbon capture and
storage (CCS). Moreover, by protecting the market for coal, such legislation could leave gas,
efficiency, and renewables fighting among themselves for limited market share, rather than
working together to build a better energy system.71
As the U.S. Senate considers a range of approaches to assembling climate and energy legislation
that can garner 60 votes, it should seek to avoid repeating the House bill’s mistakes. One
alternative is the CLEAR Act introduced by Senators Cantwell (D-WA) and Collins (R-ME) in
December 2009, which auctions all emission allowances rather than giving any for free to
traditional coal plants, as the House bill would. Senators Kerry (D-MA), Graham (R-SC), and
Lieberman (I-CT) are meanwhile negotiating an alternative bill that is designed to attract
bipartisan support. Their opportunity is to create a carbon market with a level playing field,
reducing the cost to consumers and spurring rapid emissions reductions. Such a policy would
allow the replacement of the oldest, least-efficient coal-fired power plants with a robust
combination of gas and renewable generators.
2. Advancing Clean Air Standards
Installing pollution controls can significantly increase the construction and operating costs of
coal-fired power plants.72
Both the EPA and Congress are moving forward on measures to
mandate large reductions in electric utilities’ emissions of sulfur dioxide (SO2), nitrogen-dioxide
(NOx), particulate matter, and mercury—all air pollutants associated with coal. In addition, a
multi-pollutant power plant bill introduced in the Senate in February 2010 would create the
Clean Air Interstate Rule (CAIR), a cap-and-trade program for SO2 and NOx.73
The need to purchase allowances for these pollutants would make coal plants even more
expensive. More recently, a Supreme Court decision has extended the EPA’s jurisdiction under
the Clean Air Act to cover carbon dioxide. As the EPA and Congress consider stricter
16
regulations on SO2, NO2, mercury, and CO2, coal plants will become increasingly riskier
investments for utilities and rate-payers. In general, fuel-neutral standards that do not allow
indefinite grandfathering of older plants are likely to have the biggest impact on emission
trends—in part because they will motivate fuel switching from coal to gas.
3. Reforming Electric Utility Dispatch Rules
Utility regulation at the state level and to a lesser extent at the federal level has a major impact
on utility decisions regarding which plants they build and dispatch, and consequently on their
emissions. In most states, electric utilities are strictly required to ―dispatch‖ their power plants
based on the cost of generation, which means that a gas-fired plants will be idled if it is even
slightly more expensive to operate than a coal plant. Because natural gas plants are generally
more efficient, an analysis by the Energy Information Administration concluded that in some
areas of the United States, even a slight convergence in coal and gas prices would move many
gas plants up in the dispatch order.74
Shifting dispatch requirements so that environmental
performance is a consideration in these decisions—with resulting costs passed through to
consumers—could have a substantial environmental benefit even beyond the impact of putting a
price on carbon.
4. Strengthening Environmental Controls and Transparency in the Gas Industry
Environmental problems caused by the natural gas extraction process are damaging the gas
industry’s reputation in many communities and must be addressed promptly. During the past
decade, the industry successfully obtained key exemptions for hydraulic fracturing, including
under the Safe Drinking Water Act. Although several other parts of the shale development
process are federally regulated under the Safe Drinking Water Act, Clean Water Act and Clean
Air Act, hydraulic fracturing is left to the states, not all of whose environmental agencies are
adequately equipped to deal with the range and scale of environmental issues posed by the rapid
development of unconventional gas.
A bill introduced in the U.S. House and Senate last year, known as the FRAC Act, would require
producers to publically disclose a list of all chemical constituents, though not proprietary
formulas, in their fracking fluids. It also demands that companies disclose the details of their
proprietary formulas to treating physicians in the case of medical emergencies. In the meantime,
the EPA has embarked on a new study of the potential environmental and health impacts of
hydraulic fracturing.75
Members of Congress have also requested eight service companies to
provide information about the chemicals they use in fracturing fluids.76
The industry has so far resisted efforts to regulate hydraulic fracturing at the federal level,
creating concern among local stakeholders and environmental groups about the process’s lack of
transparency. Gas companies would be well advised to take a more cooperative approach to
these issues, both at the state and federal levels. More transparency and tighter regulations are
needed if unconventional natural gas is to play a constructive, sustainable role in a low-carbon
energy future.
17
This is the first in a series of briefing papers that the Worldwatch Institute Natural Gas and
Sustainable Energy Initiative will produce on critical environmental and policy issues
surrounding natural gas. For more information on this paper and the Worldwatch initiative,
please contact:
Christopher Flavin, President, at [email protected]
Heidi VanGenderen, Senior Energy Advisor, at [email protected]
Saya Kitasei, MAP Sustainable Energy Research Fellow, at [email protected]
For further information about the Worldwatch Institute and its energy programs, please contact
Communications Manager Julia Tier at [email protected] or (+1) 202452.1999 x594.
About the Authors
Christopher Flavin is president of the Worldwatch Institute. Chris is co-author of three
books on energy, including Power Surge: Guide to the Coming Energy Revolution, which
anticipated many of the changes now under way in world energy markets. He is also a
founding member of the Board of Directors of the Business Council for Sustainable Energy
and serves as a board member of the Climate Institute. He is on the Advisory Boards of the
American Council on Renewable Energy and the Environmental and Energy Study Institute.
Saya Kitasei is a MAP Sustainable Energy Research Fellow at the Worldwatch Institute. She
has previously worked on climate and energy at the Center for American Progress, the
Climate Institute, and INFORM. She holds an M.A. in Russian, East European and Eurasian
Studies and a B.S. in Earth Systems from Stanford University.
About the Worldwatch Institute:
Worldwatch Institute, a global think tank based in Washington D.C., delivers the insights and
ideas that empower decision makers to create an environmentally sustainable society that meets
human needs. Worldwatch focuses on the 21st-century challenges of climate change, resource
degradation, population growth, and human nutrition by developing and disseminating solid data
and innovative strategies for achieving a sustainable society. For more information,
visitwww.worldwatch.org.
18
Endnotes 1 Figure 1 from ―Table 1.3: Primary Energy Consumption by Source, 1949-2008,‖ in U.S. Department of Energy
(DOE), Energy Information Administration (EIA), Annual Energy Review 2008 (Washington, DC: 26 June
2009), and from EIA, March 2010 Monthly Energy Review (Washington, DC: 31 March 2010). Figure 2 from
―Table 6.5, Natural Gas Consumption by Sector, 1949-2008,‖ in idem, and from EIA, Natural Gas Navigator,
―Natural Gas Consumption by End Use,‖ at tonto.eia.doe.gov/dnav/ng/ng_cons_sum_dcu_nus_a.htm, viewed
19 March 2010.
2 See, for example, ―Senate Approves Ban on Plants Using Oil, Gas,‖ Chicago Tribune, 9 September 1977.
3 EIA, ―Table 1.3: Primary Energy Consumption by Source, 1949-2008,‖ op. cit. note 1.
4 ―Table 6.3: Natural Gas Imports, Exports, and Net Imports, 1949-2008,‖ in ibid.
5 NYMEX historical natural gas prices from EIA, ―Natural Gas Futures Prices (NYMEX),‖ available at
http://tonto.eia.doe.gov/dnav/ng/ng_pri_fut_s1_d.htm, viewed 16 April 2010.
6 Ibid.
7 U.S. National Research Council (NRC), Hidden Costs of Energy: Unpriced Consequences of Energy
Production and Use (Washington, DC: National Academies Press, October 2009).
8 Ibid. The NRC calculated the average non-climate costs of electricity generation from coal and natural gas in
2005 to be 3.2 and 0.16 cents/kilowatt-hour, respectively. These figures primarily reflect health damages from
air pollutants like sulfur dioxide, nitrogen oxides, and particulate matter, but do not reflect damages from
mercury, a pollutant emitted during the combustion of coal.
9 According to emissions factors used by the DOE, the combustion of pipeline natural gas emits 117.08 pounds of
CO2 per million Btu, whereas coal emits between 205.3 and 227.4 pounds of CO2 per MMBtu, per EIA,
Voluntary Reporting of Greenhouse Gases Program, ―Fuel and Energy Source Codes and Emission
Coefficients,‖ www.eia.doe.gov/oiaf/1605/coefficients.html, viewed 23 February 2010.
10 Capacity data and Figure 3 from EIA, ―Form EIA-860 Database Annual Electric Generator Report‖
(Washington, DC: March 2010), at www.eia.doe.gov/cneaf/electricity/page/eia860.html.
11 Figure 4 from ―Table 8.2a: Electricity Net Generation: Total (All Sectors), 1949-2008,‖ in EIA, Annual Energy
Review 2008, op. cit. note 1. Electricity generation by energy source from ―Table 1.1. Net Generation by
Energy Source: Total (All Sectors), 1995 through December 2009,‖ in EIA, Electric Power Monthly with data
for December 2009 (Washington, DC: 15 March 2010).
12 Figure 5 based on the following sources: WTI-Cushing, Oklahoma spot prices from EIA, Petroleum Navigator,
available at http://tonto.eia.doe.gov/dnav/pet/pet_pri_spt_s1_d.htm , viewed 16 April 2010, and Henry Hub
Contract 1 futures prices from Natural Gas Navigator, available at
http://tonto.eia.doe.gov/dnav/ng/ng_pri_fut_s1_d.htm, viewed 16 April 2010 (same as note 5).
13 Ibid.
14 Country-specific statistics for number of natural gas vehicles are available from International Association of
Natural Gas Vehicles, ―Natural Gas Vehicle Statistics,‖ www.iangv.org/tools-resources/statistics.html, viewed
16 April 2010, updated April 2010.
15 For example, the Washington, D.C. Metropolitan Transit Authority (WMATA) runs over one-third of its bus
fleet on compressed natural gas, per ―WMATA’s CNG Fleet Expands with 22 60-Footers,‖ NGV Global, 30
October 2008.
16 See Pickens Plan, available at www.pickensplan.com.
17 M. Melendez et al., Emission Testing of Washington Metropolitan Area Transit Authority (WMATA) Natural
Gas and Diesel Transit Buses (Golden, CO: National Renewable Energy Laboratory, December 2005).
18 EIA, Natural Gas Navigator, ―U.S. Natural Gas Gross Withdrawals and Production,‖ available at
http://tonto.eia.doe.gov/dnav/ng/ng_prod_sum_dcu_NUS_m.htm , viewed 16 April 2010.
19
19
EIA, Annual Energy Outlook 2009 (March 2009), available at www.eia.doe.gov/oiaf/archive/aeo09/index.html,
viewed 24 February 2010; EIA, Annual Energy Outlook 2010 Early Release Overview (Washington, DC: 14
December 2009), available at www.eia.doe.gov/oiaf/aeo/index.html, viewed 24 February 2010; and from EIA,
Natural Gas Navigator, ―Coalbed Methane Production,‖ available at
http://tonto.eia.doe.gov/dnav/ng/ng_prod_coalbed_s1_a.htm, and ―Shale Gas Production,‖ available at
http://tonto.eia.doe.gov/dnav/ng/ng_prod_shalegas_s1_a.htm, both viewed 24 February 2010.
20 EIA, Natural Gas Navigator, ―Coalbed Methane Production,‖
http://tonto.eia.doe.gov/dnav/ng/ng_prod_coalbed_s1_a.htm, updated 29 October 2009.
21 Map 1 from Navigant Consulting, ―North American Natural Gas Supply Assessment: Executive Summary and
Update,‖ presentation prepared for American Clean Skies Foundation, July 2008, available at
www.afdc.energy.gov/afdc/pdfs/ng_supply_assessment_2.pdf.
22 Table 1 from Colorado School of Mines, ―Potential Gas Committee Reports Unprecedented Increase in
Magnitude of U.S. Natural Gas Resource Base,‖ press release (Golden, CO: 18 June 2009).
23 U.S. Department of Energy, Energy Information Administration (EIA), ―Shale Gas Proved Reserves,‖ available
at http://tonto.eia.doe.gov/dnav/ng/ng_enr_shalegas_s1_a.htm, viewed 27 April 2010; Navigant Consulting,
North American Natural Gas Supply Assessment, available at www.cleanskies.org/pdf/navigant-natural-gas-
supply-0708.pdf, viewed 27 April 2010; Kevin Petak and Bruce Henning, ―Economy, Shales Key in Price
Outlook,‖ The American Oil & Gas Reporter (December 2009), available at www.icfi.com/docs/economy-
shales-price-outlook.pdf, viewed 27 April 2010.
24 Worldwatch calculation based on estimates of total remaining natural gas resources of 2074 trillion cubic feet
(tcf) from Colorado School of Mines, op. cit. note 22, and on 2008 U.S natural gas consumption of 23.2 tcf, per
EIA, Natural Gas Navigator, ―Natural Gas Consumption by End Use,‖
http://tonto.eia.doe.gov/dnav/ng/ng_cons_sum_dcu_nus_a.htm, updated 29 March 2010.
25 Vello Kuuskra and Scott Stevens, ―Worldwide Gas Shales and Unconventional Gas: A Status Report‖
(Arlington, VA: Advanced Resources International, 7 December 2009).
26 Figure 6 from EIA, Annual Energy Outlook 2009 (March 2009), available at
http://www.eia.doe.gov/oiaf/archive/aeo09/index.html, viewed 24 February 2010; EIA, Annual Energy Outlook
2010 Early Release Overview (Washington, DC: 14 December 2009), available at
http://www.eia.doe.gov/oiaf/aeo/index.html, viewed 24 February 2010; and from EIA, Natural Gas Navigator,
―Coalbed Methane Production,‖ available at http://tonto.eia.doe.gov/dnav/ng/ng_prod_coalbed_s1_a.htm, and
―Shale Gas Production,‖ available at http://tonto.eia.doe.gov/dnav/ng/ng_prod_shalegas_s1_a.htm, both viewed
24 February 2010.
27 Ibid.
28 ICE, BENTEK Energy, ―Lowest Breakeven Prices Associated with the Most Productive Basins,‖ cited in
Gregory Staple and Joel Swerdlow, ―A New Energy Option: North America’s New Natural Gas Resources and
their Potential Impact on Energy and Climate Security‖ (Washington, DC: American Clean Skies Foundation,
12 December 2009), p. 12.
29 Reserves and production data and Figure 7 based on EIA, Petroleum Navigator, ―Crude Oil Production,‖
available at http://tonto.eia.doe.gov/dnav/pet/pet_crd_crpdn_adc_mbbl_m.htm, viewed 24 February 2010, and
on EIA, Natural Gas Navigator, ―U.S. Gross Withdrawals of Natural Gas,‖ available at
http://tonto.eia.doe.gov/dnav/ng/ng_prod_sum_dcu_NUS_m.htm, viewed 16 April 2010, and ―U.S. Dry Natural
Gas Reserves,‖ available at http://tonto.eia.doe.gov/dnav/ng/ng_enr_dry_dcu_NUS_a.htm, viewed 24 February
2010.
30 Ibid.
31 Figure 8 based on the following sources: average carbon emissions from coal and gas plants based on total
carbon dioxide emissions from natural gas and coal in the power sector in 2007, from EIA, ―Energy-Related
Carbon Dioxide Emissions by End-Use Sector, and the Electric Power Sector, by Fuel Type, 1949-2007,‖
www.eia.doe.gov/environment.html, viewed 25 February 2010; electricity generation data from natural gas and
20
coal in 2007 from ―Table 8.2a: Electricity Net Generation: Total (All Sectors), 1949-2008,‖ in EIA, Annual
Energy Review 2008, op. cit. note 1; emissions factors for modern pulverized coal, IGCC, and NGCC plants
from DOE, Cost and Performance Baseline for Fossil Energy Plants (Washington, DC: 2007).
32 EIA, op. cit. note 10.
33 Ibid.
34 Ibid.
35 Stan Mark Kaplan, ―Displacing Coal with Generation from Existing Natural Gas-Fired Power Plants‖
(Washington, DC: Congressional Research Service, 19 January 2010).
36 Ibid.
37 Stan Mark Kaplan, ―Power Plants: Characteristics and Costs‖ (Washington, DC: Congressional Research
Service, 13 November 2008).
38 Ibid.
39 Figure 9 adapted from Lazard, ―Levelized Cost of Energy Analysis – Version 3.0‖ (New York: June 2009).
Except where noted, Lazard’s low-end assumptions for capital and O&M costs are used. For gas fuel-cell
figures, Lazard’s high-end assumptions are used, as low-end represents combined heat-and-power applications.
Natural gas and coal fuel cost ranges calculated using plant heat rate assumptions in Lazard and the minimum
and maximum delivered prices during 2007–35 from ―Table 13. Natural Gas Supply, Disposition, and Prices,‖
and ―Table 15. Coal Supply, Disposition, and Prices,‖ in EIA, Annual Energy Outlook 2010 Early Release, op.
cit. note 19. Natural gas low- and high-end fuel costs were $4.25 and $9.34 per million Btu, respectively. Coal
low- and high-end fuel costs were $1.80 and $2.15, respectively. Biomass fuel cost range calculated using
Lazard’s low- and high-end fuel cost assumptions ($0 and $2 per million Btu, respectively). Lazard capital cost
figures reflect investment tax credit (fuel cell and solar) and production tax credit (wind, biomass, geothermal).
They assume 2008 dollars, 20-year economic life, 40 percent tax rate, and 5–20 year tax life; 30 percent debt at
8.0 percent interest rate and 40 percent equity at 8.5 percent cost and 30 percent common equity at 12 percent
cost for alternative energy generation technologies. They assume 60 percent debt at 8.0 percent interest rate and
40 percent equity at 12 percent cost for conventional generation technologies. Solar PV reflects single-axis
tracking crystalline. Solar thermal represents solar tower. Pulverized coal represents advanced supercritical
pulverized coal with no carbon capture and compression.
40 Worldwatch calculation based on average delivered prices of natural gas in 2008 and 2009 of $9.02 and $4.77
per million Btu, respectively, per EIA, Electric Power Monthly…, op. cit. note 11, and on Lazard levelized cost
estimates, per Lazard, op. cit. note 39.
41 ―Table 1.1: Net Generation by Energy Source: Total (All Sectors),‖ in EIA Electric Power Monthly…, op. cit.
note 11.
42 Emissions data from EIA, March 2010 Monthly Energy Review, op. cit. note 1.
43 Progress Energy, ―Progress Energy Carolinas Plans to Retire Remaining Unscrubbed Coal Plants in N.C.,‖
press release (Raleigh-Durham, NC: 1 December 2009).
44 U.S. Department of Energy, National Renewable Energy Laboratory (NREL), ―U. S. Parabolic Trough Power
Plant Data,‖ available at www.nrel.gov/csp/troughnet/power_plant_data.html#segs_ix, viewed 27 April 2010.
45 FPL, ―Martin Next Generation Solar Energy Center,‖ available at www.fpl.com/environment/solar/martin.shtml,
viewed 27 April 2010; and Jad Mouawad, ―The Newest Hybrid Model,‖ The New York Times (4 March 2010),
available at www.nytimes.com/2010/03/05/business/05solar.html?pagewanted=1, viewed 27 April 2010.
46 Worldwatch calculation based on EIA, Electric Power Annual 2008, ―Table 2.1. Net Generation by Energy
Source by Type of Producer, 1997 through 2008,‖ released 21 January 2010, available
http://www.eia.doe.gov/cneaf/electricity/epa/epa_sum.html, viewed 16 April 2010.
47 Amanda Chiu, ―One Twelfth of Global Electricity Comes from Combined Heat and Power Systems,‖ Vital
Signs 2009 (Washington, DC: Worldwatch Institute, 2009).
21
48
Efficiency from U.S. Department of Interior, Environmental Protection Agency (EPA), ―Combined Heat and
Power Partnership,‖ http://www.epa.gov/chp/basic/efficiency.html, viewed 16 April 2010. Greenhouse gas
reductions based on comparison with conventional power plant with a 31-percent efficiency running on the U.S.
fossil mix, from EPA, ―Conventional Generation vs. CHP: CO2 Emissions,‖
http://www.epa.gov/chp/basic/environmental.html , viewed 16 April 2010.
49 Frank Dohmen, ―A Power Station in Your Basement,‖ BusinessWeek, 9 September 2009.
50 Germany Wind Energy Institute (DEWI), ―Auswirkungen der Finanzkrise moderat – Weltmarkt Wächst weiter,‖
press release (Wilhelmshaven: 31 December 2008).
51 DOE and ALL Consulting, Modern Shale Gas Development in the United States: A Primer (Washington, DC:
April 2009).
52 DOE and ALL Consulting, op. cit., note 51.
53 Abrahm Lustgarten, ―Is New York’s Marcellus Shale Too Hot to Handle?‖ ProPublica, 9 November 2009.
54 Joaquin Sapien, ―With Natural Gas Drilling Boom, Pennsylvania Faces an Onslaught of Wastewater,‖
ProPublica, 3 October 2010.
55 Ohio Department of Natural Resources, Division of Mineral Resources Management, ―Report on the
Investigation of the Natural Gas Invasion of Aquifers in Bainbridge Township of Geauga County, Ohio‖
(Columbus, OH: 1 September 2008), available http://www.dnr.state.oh.us/Portals/11/bainbridge/report.pdf,
viewed 16 April 2010.
56 DOE and ALL Consulting, op. cit., note 51.
57 EIA, ―Emissions of Greenhouse Gases in the United States 2008,‖ www.eia.doe.gov/oiaf/1605/ggrpt/index.html,
revised 8 December 2009.
58 EPA, ―Natural Gas STAR Program,‖ www.epa.gov/gasstar/accomplishments/index.html, viewed 2 April 2010.
59 Al Armendariz, ―Emissions from Natural Gas Production in the Barnett Shale Area and Opportunities for Cost-
Effective Improvements‖ (Austin, TX: Environmental Defense Fund, 26 January 2009).
60 Mark Zoback and Brad Copithorne, Stanford University, personal communication with Saya Kitasei,
Worldwatch Institute, 12 March 2010. For a suspected fracturing-induced seismic event, see Ben Casselman,
―Temblors Rattle Texas Town,‖ Wall Street Journal, 12 June 2009
61 See, for example, Amy Mall, ―Drilling Down: Protecting Western Communities from the Health and
Environmental Effects of Oil and Gas Production‖ (New York: Natural Resources Defense Council, October
2007), p. 17; DOE and ALL Consulting, op. cit., note 51, p. 25.
62 EIA, ―World Dry Natural Gas Production,‖ www.eia.doe.gov/emeu/international/gasproduction.html, viewed
26 February 2010.
63 International Energy Agency, ―Ultimately recoverable conventional natural gas resources by region, end-2008,‖
in World Energy Outlook 2009, available at www.iea.org/country/graphs/weo_2009/fig11-4.jpg, viewed 26
February 2010.
64 Import and export data from EIA, Natural Gas Navigator, ―U.S. Natural Gas Imports by Country,‖ available
http://tonto.eia.doe.gov/dnav/ng/ng_move_impc_s1_m.htm, viewed 2 April 2010. For the effect of North
American unconventional gas supplies on global gas prices, see ―Natural Gas: An Unconventional Glut,‖ The
Economist, 11 March 2010.
65 Vello Kuuskra and Scott Stevens, op. cit. note 25.
66 ―IEA Expects Gas Glut as Unconventional Output Rises,‖ International Energy, 11 November 2009.
67 International Energy Agency (IEA), World Energy Outlook 2009 (Paris: 2009).
68 ―Delhi to Convert Coal Plants to Natural Gas,‖ ClimateWire, 22 January 2010.
22
69
The White House, ―Fact Sheet: U.S.-China Shale Gas Resource Initiative,‖ press release (Washington, DC: 17
November 2009).
70 Joseph Romm, ―Game Changer, Part 2: Why Unconventional Natural Gas Makes the 2020 Waxman-Markey
Target So Damn Easy and Cheap to Meet,‖ ClimateProgress.org, 10 June 2009.
71 Gregory Staple and Joel Swerdlow, ―A New Energy Option: North America’s New Natural Gas Resources and
their Potential Impact on Energy and Climate Security‖ (Washington, DC: American Clean Skies Foundation,
12 December 2009), p. 12; Steven Levine, Frank Graves, and Metin Celebi, ―Prospects for Natural Gas Under
Climate Policy Legislation,‖ Discussion Paper, The Brattle Group, March 2010.
72 Stan Mark Kaplan, op. cit., note 37.
73 Robin Bravender, ―Carper, Alexander Unveil Multipollutant Power Plant Bill,‖ E&E News, 4 February 2010.
74 EIA, ―The Implications of Lower Natural Gas Prices for the Electric Generation Mix in the Southeast,‖ Short-
Term Energy Outlook Supplement (Washington, DC: May 2009).
75 EPA, ―EPA Initiates Hydraulic Fracturing Study: Agency Seeks Input from Science Advisory Board,‖ press
release (Washington, DC: 18 March 2010).
76 Henry Waxman and Edward Markey, ―Examining the Potential Impact of Hydraulic Fracturing,‖ memorandum
to the House Subcommittee on Energy and Environment, 18 February 2010), available at
http://energycommerce.house.gov/Press_111/20100218/hydraulic_fracturing_memo.pdf.
Addressing the Environmental Risks from Shale Gas Development
Briefing Paper 1
Mark Zoback
Saya Kitasei Brad Copithorne
July 2010
Natural Gas and Sustainable Energy Initiative
1
Addressing the Environmental Risks from
Shale Gas Development
Mark Zobacka, Saya Kitasei
b, Bradford Copithorne
c
I. Executive Summary
The rapid development of shale gas resources in the past few years has already dramatically
affected U.S. energy markets—lowering energy prices and carbon dioxide emissions—and could
offer an affordable source of low-carbon energy to reduce dependence on coal and oil.1 However,
the development of shale gas has been linked to a range of local environmental problems,
generating a public backlash that threatens to bring production to a halt in some regions. While
hydraulic fracturing in particular has been the focus of much controversy, our analysis indicates
that the most significant environmental risks associated with the development of shale gas are
similar to those associated with conventional onshore gas, including gas migration and
groundwater contamination due to faulty well construction, blowouts, and above-ground leaks
and spill of waste water and chemicals used during drilling and hydraulic fracturing.
Many technologies and best practices that can minimize the risks associated with shale gas
development are already being used by some companies, and more are being developed. The
natural gas industry should work with government agencies, environmental organizations, and
local communities to develop innovative technologies and practices that can reduce the
environmental risks and impacts associated with shale gas development.
Stronger, fully-enforced government regulations are needed in many states to provide sufficient
protection to the environment as shale gas development increases. In addition, continued study
and improved communication of the environmental risks associated with both individual wells
and large scale shale gas development are essential for society to make well-informed decisions
about its energy future.
This briefing paper, part of an on-going series on the role of natural gas in the future energy
economy, provides an overview of how horizontal drilling and hydraulic fracturing are used to
extract shale gas, examines the environmental risks, associated with shale gas development, and
provides an overview of the industry best practices and government regulations that are needed if
shale gas is to contribute its full potential to help build a low-carbon economy in the years ahead.
Cover photo: A drilling rig near Shreveport, Louisiana, by danielfoster437. a Benjamin M. Page Professor of Geophysics, Stanford University
b MAP Sustainable Energy Fellow, Worldwatch Institute
c Finance Specialist, Environmental Defense Fund
2
II. Extracting Natural Gas from Shale Geologists have long been aware that large amounts of natural gas lie trapped in some
formations of shale, a sedimentary rock formed from deposits of mud, silt, clay, and organic
matter. Over time, that organic matter breaks down, creating molecules of methane, also known
as natural gas. While some of this natural gas migrates into other formations over millions of
years, much of it remains trapped in its shale source rock.
Although the first producing U.S. natural gas well was drilled into a shale formation in New
York (in 1821), most commercial drilling during the 19th and 20th centuries targeted gas that has
migrated out of its source rock and accumulated in permeable reservoirs such as sandstone
formations.2 Unlike these ―conventional‖ reservoirs, whose relatively high permeability enables
producers to extract gas using vertical wells, shale is a much ―tighter,‖ less permeable rock. As a
result, methane molecules cannot flow easily through shale and a vertical well is only able to
drain gas only from a very small volume of the rock surrounding it, which generally prevents
vertical wells from producing sufficient gas to be economical.
Over the past decade, however, the application of two techniques, horizontal drilling and
hydraulic fracturing, has enabled operators to extract gas economically from shale formations
thousands of feet deep. Although both technologies originally were developed to increase
production from conventional wells, their use in the Barnett Shale, near Fort Worth, Texas,
revealed that they could be the key to unlocking the trillions of cubic feet of natural gas
estimated to exist in shale gas plays throughout the United States.3 (See Figure 1.) At year-end
2009, the five most productive U.S. shale gas fields – the Barnett, Haynesville, Fayetteville,
Woodford, and Marcellus shales – were producing some 8.3 billion cubic feet a day, the
equivalent of nearly 1.6 million barrels of oil a day, or 30 percent of total U.S. crude oil
production during 2009.4
Figure 1. Map of Shale Gas Plays, Lower 48 States
Source: EIA
3
Oil and gas drilling generally begins in the same way in both vertical and horizontal wells.
Operators insert an initial length of steel pipe, called ―conductor casing,‖ into a vertical wellbore
soon after drilling begins in order to stabilize the well as it passes through the shallow, often
unconsolidated sediments and soils near the Earth‘s surface.5 (See Figure 2.) Then, operators
continue drilling vertically and insert surface casing, which most states require to extend from
the ground‘s surface past the depth of all underground sources of drinking water (USDW‘s).6
Operators then pump cement into the casing, followed by water, to push the cement out through
the bottom of the casing and back up into the space between the surface casing and the wellbore
(called the ―annulus‖) until it is entirely filled. Almost all states require the surface casing to be
fully-cemented before drilling is allowed to continue.7 After the surface casing has been
cemented into place, regulators may require operators to install blowout prevention equipment
(BOPE) at the surface to prevent any pressurized fluids encountered during drilling from moving
up the well through the space between the drill pipe and the surface casing.8
After allowing the cement behind
the casing to set, operators continue
drilling for a short distance,
typically 10 to 20 feet, and test the
integrity of the cement by
pressurizing the well. They then
continue drilling vertically until
state regulations may require the
insertion of intermediate casing,
which can be used to help stabilize
deep wells. In addition, between the
base of the surface casing and the
target gas-bearing shale formations,
wellbores pass through thousands of
feet of rock formations. These
formations may contain
hydrocarbons, including natural
gas, or briny water containing
highly concentrated salts and other
contaminants. Intermediate casing
is designed to isolate such
formations from each other and the
wellbore, preventing contamination
of the gas that will be produced and of freshwater aquifers near the Earth‘s surface.
When drilling a horizontal well, operators begin turning or ―kicking off‖ the drill when they near
the top of the target formation or ―production zone,‖ until the wellbore runs through the
formation horizontally. Horizontal drilling, which can extend up to 10,000 feet, vastly increases
the wellbore‘s contact with the gas-bearing formation relative to vertical drilling, which would
be limited to the thickness of the formation—less than 300 feet in most major U.S. shale plays.9
Source: GWPC. Not to scale.
Figure 2. Casing and Cementing of a Horizontal Well
4
After drilling the horizontal section of the well, operators run a string of ―production casing‖ into
the well and cement it in place. They then ―perforate‖ the production casing using small
explosive charges at intervals along the horizontal wellbore where they intend to hydraulically
fracture the shale.
Hydraulic fracturing was first used in the late 1940s, and has since become a common technique
to enhance the production of low permeability formations, especially unconventional reservoirs
such as tight sands, coal beds, and deep shales.10 Hydraulic fracturing is a technically complex
process. Because most horizontal wells are quite long, operators conduct fracturing in stages,
starting at the tip or ―toe‖ and proceeding toward the end closest to the vertical portion or ―heel‖
of the foot-shaped wellbore. A wellbore that extends 5,000 feet horizontally within a shale layer,
for example, might be hydraulically fractured 10 to 15 times at intervals several hundred feet
apart. Each perforation interval is isolated in sequence so that only a single section of the well is
hydraulically fractured at a given time.
During a hydraulic fracturing operation, operators pump fracturing fluid at high pressure through
the perforations in a section of the casing. The chemical composition of the fracturing fluid, as
well as the rate and pressure at which it is pumped into the shale, are tailored to the specific
properties of each shale formation and, to some extent, each well. When the pressure increases to
a sufficient level, it causes a hydraulic fracture or ―hydrofracture‖ to open in the rock,
propagating along a plane more or less perpendicular to the path of the wellbore.11 (See Figure
3.) A typical hydrofracture is
designed to propagate
horizontally about 500 to 800
feet away from the well in each
direction and vertically for the
thickness of the shale.
Operators monitor and control
the fracture pressure to prevent
vertical propagation beyond the
thickness of the gas-producing
shale layer.12
One of the most novel
discoveries in the Barnett Shale
was the possibility of using
―slickwater‖ as a fracturing
fluid in deep shale formations.
Unlike the highly viscous gels
used previously to fracture
conventional formations,
slickwater is a more dilute,
low-viscosity water-based fluid
designed to carry a small
amount of sand into fractures to
prop them open after the pumping stops, allowing gas to escape. Chemical additives are designed
When multi-stage hydraulic fracturing is performed, the induced
microearthquakes generated during each stage are so small they can be
detected only using highly sensitive seismometers placed in nearby
monitoring wells. The microseismic events occur in the rock distributed
around each of the hydraulic fracture planes. The hydraulic fracturing is
done sequentially in 10 to 20 stages. Only four stages are shown here
for simplicity. The figure is not to scale.
Figure 3. Schematic of Multi-stage Hydraulic Fracturing
5
to inhibit scale and bacterial growth in the wellbore, reduce friction, and generally improve the
effectiveness of the fracture job. Slickwater works well in shale gas reservoirs because its low
viscosity allows the fracturing fluid to leak out of hydraulic fractures into many small, naturally
occurring fractures in the shale.
Slickwater increases water pressure in these microfractures, inducing shear-slip, or micro-
seismic events that generally have magnitudes of less than -1.5 on the Richter scale―about as
much energy as is released by a gallon of milk dropped from chest height to the floor. Because of
the small magnitudes of these events, which represent micro-earthquakes about one-millionth the
size of tremors that might be detected by inhabitants of a populated area, operators must deploy
ultrasensitive seismometers in nearby monitoring wells in order to detect them.13 (See Figure 4.)
Figure 4 shows microseismic data from a well drilled in the Barnett Shale and hydraulically
fractured with slickwater in 11 stages. The locations of the microseismic events generated during
slickwater hydraulic fracturing provides a picture of where the hydrofractures propagated. This
information is important to operators because the microseismic events define the portion of the
reservoir stimulated during hydraulic fracturing, increasing the shale‘s permeability and allowing
gas molecules to flow more easily into the production casing.
The above-mentioned well targeted a portion of the Barnett Shale about 330 feet thick and at
depths between about 5,600 and 5,930 feet below the surface. The horizontal wellbore is roughly
3,800 feet long. Monitoring detected microseismic activity over the entire thickness of the shale,
about 150 feet above and 200 feet below the wellbore (Figure 4A), and about 500 to 700 feet to
its sides (Figure 4B). Monitoring did not detect microseismic activity any significant distances
above or below the shale formation, suggesting that the design of this fracture job successfully
confined stimulation to the target formation. In this case, the propagation of fractures into the
underlying Ellenberger Limestone, which contains highly saline brine, would have allowed brine
to contaminate the gas in the Barnett Shale, decreasing the efficiency and increasing the cost of
its extraction. No microseismic events with magnitudes greater than -1.6 were detected.
Drilling and fracturing a typical horizontal well in the Marcellus shale takes about three weeks to
complete and costs about $3.5 to $4.5 million.14 After hydraulic fracturing is complete, gas
begins to flow out of the well to the surface, where it is processed, compressed, and transported
to markets through pipelines. During this period, maintenance may be performed on the well, but
much of the equipment used for drilling and fracturing the well is used to drill another horizontal
well from the same well pad and wellbore or removed for use at other sites. Each unconventional
well‘s production rate declines rapidly after the first few months of production. While the great
majority of gas is produced during the first few years of production, a well could continue to
produce for five to ten years before becoming uneconomical.15 In some cases, a well may be
fractured again to restimulate production, but while research is underway to improve the
performance of refracturing, it is not currently used in most shale gas wells.16
When a well becomes uneconomical, state regulations require operators to permanently plug it
with cement or another material. The majority of gas-producing states require plugs to be placed
through producing zones and from the surface to the base of ground water. Plugs are intended to
prevent fluid, which might include hydrocarbons, formation water, and fracturing fluid absorbed
6
by the target formation, from migrating along the wellbore to other layers of rock and potentially
contaminating ground water after the well has been abandoned.17
Figure 4. Microseismic Diagrams of Typical Hydraulic Fracturing Job in the Barnett
Shale
Each dot in Figure 4A and B represents a microseismic event induced during hydraulic fracturing of an actual
well in the Barnett Shale, with each color representing a distinct fracturing stage. Figure 4C displays the
distribution of these microseismic events by magnitude. Figures are not to scale.
Source:Data courtesy of the Stanford Department of Geophysics
5120
5320
5520
5720
5920
6120
6320
0 1000 2000 3000 4000 5000
Dep
th B
elo
w S
urf
ace
(fe
et)
Horizontal Distance (feet)
A. Side View of Wellbore
WellboreMarble Falls Limestone
Duffer Shale
Barnett
Shale
Ellenberger Limestone
5120
5320
5520
5720
5920
6120
6320
-1000 0 1000
Dep
th B
elo
w S
urf
ace
(fe
et)
Horizontal Distance (feet)
B. View Along
Wellbore Axis
Marble Falls
Limestone
Duffer Shale
Barnett
Shale
Ellenberger
Limestone
1
10
100
1000
10000
-3.6 -3.4 -3.2 -3 -2.8 -2.6 -2.4 -2.2 -2 -1.8 -1.6 -1.4
Cu
mu
lati
ve
Nu
mb
er o
f
Mic
rose
ism
ic E
ven
ts
Magnitude (Richter Scale)
C. Distribution of Magnitudes of Microseismic Events
7
III. Environmental Risks and Best Practices
Shale gas has received a good deal of attention recently for the potential negative impacts that its
development may have on the environments and communities in which it occurs. Instances of
water contamination, air pollution, and earthquakes have been blamed on gas extraction
activities. A thorough understanding of the techniques used to extract gas from shale formations
and the safeguards that exist to prevent environmental damage is critical to assessing the sources
and magnitudes of risk involved in shale gas development.
Subsurface Contamination of Ground Water
A frequently expressed concern about shale gas development is that subsurface hydraulic
fracturing operations in deep shale formations might create fractures that extend well beyond the
target formation to water aquifers, allowing methane, contaminants naturally occurring in
formation water, and fracturing fluids to migrate from the target formation into drinking water
supplies. With the notable exceptions of the shallow Antrim and New Albany Shales, many
thousands of feet of rock separate most major gas-bearing shale formations in the United States
from the base of aquifers that contain drinkable water.18 (See Figure 5.)
Because the direct contamination of underground sources of drinking water from fractures
created by hydraulic fracturing would require hydrofractures to propagate several thousand feet
Figure 5. Target Shale Depth and Base of Treatable Groundwater in Select Shale Plays
Source: GWPC
8
beyond the upward boundary of the target formation through many layers of rock, such
contamination is highly unlikely to occur in deep shale formations during well-designed fracture
jobs. For example, the top of the Marcellus Shale, which runs from upstate New York through
Pennsylvania, West Virginia, and parts of Ohio, lies from 4,000 to 8,500 feet below the surface.19
The deepest underground sources of drinking water in this region lie about 850 feet below the
surface.20 Geologists estimate that there is at least a half mile of rock between the natural gas
deposits and the groundwater, including nine layers of impermeable shale, each of which acts as
a barrier to vertical propagation of both natural and artificial fractures.21
As mentioned earlier, seismic monitoring is an essential tool for assuring that hydraulic
fracturing is inducing microseismic activity only within the shale gas reservoir. Yet only about
three percent of the ~75,000 hydraulic fracturing stages conducted in the United States in 2009
were seismically monitored.22 Public confidence in the safety of hydraulic fracturing would be
greatly improved by more frequent microseismic monitoring and public dissemination of the
results.
Failure of the cement or casing surrounding the wellbore poses a far greater risk to water
supplies. If the annulus is improperly sealed, natural gas, fracturing fluids, and formation water
containing high concentrations of dissolved solids may be communicated directly along the
outside of the wellbore among the target formation, drinking water aquifers, and layers of rock in
between. For example, in 2007, a well that had been drilled almost 4,000 feet into a tight sand
formation in Bainbridge, Ohio was not properly sealed with cement, allowing gas from a shale
layer above the target tight sand formation to travel through the annulus into an underground
source of drinking water. The methane eventually built up until an explosion in a resident‘s
basement alerted state officials to the problem.23
A variety of tools exist to help producers and regulators minimize the risk of cement and casing
failures. The American Petroleum Institute (API) develops and updates standards and
―recommended practices‖ for oil and gas exploration and production activities.24 Many state
regulations require steel casing and cement used in oil and gas well construction to meet
standards set by API or other organizations.25 Frequent monitoring and testing also allow
producers and regulators to check the integrity of casing and cement jobs. Many states require
operators to perform a test such as a cement bond log, which measures the quality of the cement-
casing and cement-formation bonds.26 Ensuring that these tests are conducted and heeded in
accordance with regulations, and requiring them in states where they are currently voluntary, are
essential to preventing accidents such as occurred in Bainbridge.
Blowouts
Recent gas well blowouts in Pennsylvania and West Virginia during drilling operations in the
Marcellus Shale, set against the backdrop of the recent offshore blowout and oil spill in the Gulf
of Mexico, underscore the environmental and public risks associated with drilling into highly
pressurized zones of hydrocarbons and introducing pressurized fluids during hydraulic
fracturing.27 At the time of writing this article, the causes of all three blowouts were still under
investigation. Operators in Pennsylvania reported that that blowout occurred because the blowout
preventer proved inadequate to deal with higher-than-anticipated pressures.28 In West Virginia,
9
drillers reportedly encountered an unexpected pocket of methane in an abandoned coal mine only
about 1,000 feet below the surface, and a blowout preventer had not yet been installed.29
Such disasters stress the need for gathering accurate information about the subsurface and
ensuring that personnel on drill sites are trained to deal with unusual and unexpected situations,
including blowouts. Even if drilling and well construction are carried out in full compliance with
local, state, and federal regulations, and industry best practices are followed, many decisions
during drilling and fracturing operations must be made by individuals, and training and
experience, together with full enforcement of strong regulations and adoption of industry best
practices, are critical to the protection of the public and the environment.
Seismic Risks
Another subsurface risk that has received attention recently is the possibility that drilling and
hydraulically fracturing shale gas wells might cause low-magnitude earthquakes. In 2008 and
2009, the town of Cleburne, Texas, experienced several clusters of weak earthquakes all
registering 3.3 or less on the Richter scale.30 Since the town had never registered an earthquake in
its 142-year history, some residents wondered if the recent increase in local drilling activity
associated with the Barnett Shale might be responsible. A study by seismologists with the
University of Texas and Southern Methodist University found no conclusive link between
hydraulic fracturing and these earthquakes but indicated that the injection of waste water from
gas operations into numerous saltwater disposal wells that were being operated in the vicinity
could have caused the seismic activity.31 Over 200 such wells exist in the Barnett Shale, and are
the preferred means of waste water disposal for operators in the area.32
While the hydraulic fracturing process does create a large number of microseismic events, or
micro-earthquakes, the magnitudes of these are generally too small to be detected at the surface.
Figure 4C shows the cumulative frequency distribution of microseismic events of different size
in a Barnett Shale well. Altogether, a downhole seismometer array deployed in a nearby well
detected about 1,000 micro-earthquakes. The biggest micro-earthquakes have a magnitude of
about -1.6. An earthquake of this size represents slip of less than a hundredth of an inch, about
the thickness of a human hair, on a pre-existing fault only a couple of feet across. The number of
extremely small earthquakes (less than a magnitude of about -2.8) tapers off because they are so
small that they cannot be detected.
Underground fluid injection is an integral part not only of hydraulic fracturing, but of waste
water disposal in injection wells, some geothermal energy projects, and carbon dioxide
sequestration. The seismic monitoring of hydraulic fracture jobs discussed earlier is critical to
improving understanding of how underground injection might spark unexpectedly high-
magnitude seismic activity.
Surface Water and Soil Contamination
Because of the quantities of chemicals that must be stored at drilling sites and the volumes of
liquid and solid waste that are produced, significant care must be taken that these materials do
not contaminate surface water and soil during their transport, storage, and disposal.
10
Fluids used for slickwater hydraulic fracturing are typically more than 98 percent fresh water and
sand by volume, with the remainder made up of chemicals that improve the treatment‘s
effectiveness, such as thickeners and friction reducers, and protect the production casing, such as
corrosion inhibitors and biocides.33 These fluids are designed by service companies that tailor
fracturing treatments to suit the needs of a particular job. In a 2009 survey of six service
companies and 12 chemical providers, the New York State Department of Environmental
Conservation received a list of some 200 chemical additives that companies might use in
fracturing fluids.34
Because the fluids in each fracturing treatment would contain a different subset of these
chemicals, and because these chemicals could be hazardous in sufficient concentrations, public
disclosure of the chemicals used in hydraulic fracturing on a site-by-site basis is necessary to
enable regulatory agencies, health professionals, and citizens to conduct baseline water testing
and respond appropriately should contamination or exposure occur. A number of companies are
investigating use of more environmentally benign fracturing fluids. 35 These would also help limit
the environmental and health risks posed by fracturing fluids in the case of contamination.
Chemicals to be used in fracturing fluids are generally stored at drilling sites in tanks before they
are mixed with water in preparation for a fracturing job. Under the Emergency Planning and
Community Right to Know Act of 1986 (EPCRA), companies must post Material Safety Data
Sheets (MSDSs) that list the properties and any health effects of chemicals stored in quantities of
more than 10,000 pounds.36 Disclosure of chemicals stored in smaller quantities is not currently
required by law, and access to MSDSs can often be limited. Several ongoing efforts would
require greater disclosure of fracturing fluids, including a provision in draft climate legislation
introduced by Senators John Kerry (D-MA) and Joe Lieberman (I-CT) in May 2010 that would
amend EPCRA to mandate the disclosure of all chemicals used on public websites.37
After each fracturing stage, the fracturing fluid, along with any water originally present in the
shale formation, is ―flowed back‖ through the wellbore to the surface. Flowback and water
produced during a well‘s lifetime can contain naturally occurring formation water that is millions
of years old and therefore can display high concentrations of salts, naturally occurring
radioactive material (NORM), and other contaminants including arsenic, benzene, and mercury.38
As a result, the water produced during hydraulic fracturing must be disposed of properly. The
―flowback‖ period typically lasts for periods of hours to weeks, although some injected water
can continue to be produced along with gas several months after production has started.39 In the
Marcellus Shale, approximately 25 percent of the water injected during hydraulic fracturing
operations may be produced during flowback.40
Flowback water is dealt with differently in different states. In the Barnett, Fayetteville,
Haynesville, Woodford, Antrim, and New Albany Shales, the primary disposal method has been
injection into underground saline aquifers, such as the Ellenberger Limestone that underlies the
Barnett formation.41 While injection is regulated at the federal level under the Safe Drinking
Water Act (SDWA), the availability of adequate disposal wells is a major issue that needs to be
addressed for shale gas development to take place. There are tens of thousands of licensed
11
injection wells in Texas, but because of political and geological constraints, many fewer exist in
the Marcellus Shale. The state of Pennsylvania currently only has about 10 Class II wells.42
As a result, one option for dealing with flowback water from wells in the Marcellus Shale is
disposal at municipal waste water treatment facilities, which generally discharge treated water
into surface water bodies such as rivers and streams.43 Current waste water treatment facilities in
the Marcellus are insufficient to handle the volumes of fluids that would be produced were shale
gas development to increase significantly. In addition, they may not be designed to handle the
highly saline water produced by gas drilling.
In late 2008 and 2009, there were significant spikes in the level of total dissolved solids (TDS) in
Pennsylvania‘s Monongahela River, which supplies drinking water to approximately 350,000
people. Since flowback contains large amounts of total dissolved solids (TDS), and drilling
fluids constituted up to 20 percent of the waste water being treated by some facilities, the
Pennsylvania Department of Environmental Protection (PADEP) ordered these facilities to
restrict their intake of drilling waste water.44 PADEP reported that TDS levels, which also can be
influenced by abandoned mine drainage, stormwater runoff, and discharges from industrial or
sewage treatment plants, exceeded standards at least twice more in 2009.45
Given the constraints on both underground injection and treatment and discharge in the
Marcellus Shale, serious investment will be needed in advancing treatment technologies that
enable companies to reuse fluids for subsequent fracturing jobs. As flowback comprises only 25
percent of the water injected into a given well in the Marcellus, treated flowback water could be
diluted with fresh water and re-injected. Recycling water minimizes both the overall amount of
water used for fracturing and the amount that must be disposed of. Many water treatment
processes are currently being investigated that could be potentially be used at large scale and
have a significant impact on this problem.46
Finally, one of the problematic aspects of handling flowback water is the temporary storage and
transport of such fluids prior to treatment or disposal. In many cases, fluids may be stored in
lined or even unlined open evaporation pits.47 Even if the produced water does not seep directly
into the soil, a heavy rain can cause a pit to overflow and create contaminated runoff.48 Storing
produced water in enclosed steel tanks, a practice already used in some wells, would reduce the
risk of contamination while improving water retention for subsequent reuse.49
In addition, equipment used to move fluids between storage tanks or pits and the wellhead must
be monitored and tested regularly to prevent spills, and precautions must be taken while
transporting produced water to injection or treatment sites, whether via pipeline or truck. In May
2009, PADEP discovered that two leaky joints in a pipeline carrying waste water from gas wells
to a disposal site had resulted in the release of about 4,200 gallons of waste water into Cross
Creek, causing the deaths of some fish and invertebrates.50 Range Resources, the owner of the
wells, was fined for this violation of Pennsylvania‘s environmental statutes, as well as for
another spill that occurred in October 2009.51
12
Other Surface Impacts
Drilling operations require significant above-ground development. In addition to the well pad
itself, roads may need to be built and gathering infrastructure installed to bring the natural gas
from the wellhead to a pipeline that, for a typical well in the Marcellus Shale, may require the
development of several acres of land. Total land use can be reduced by drilling multiple wells
from a single well pad, as is done in areas of steep topography or environmental sensitivity.
Nonetheless, because so many wells have to be drilled and appreciable infrastructure developed,
it is important to do as much as possible to minimize the overall impact on local communities.
Land use decisions affect a wide range of stakeholders including the landowners, neighbors and
surrounding communities. Permitting procedures will need to evaluate the needs of each of the
stakeholders and include clear and enforceable remediation strategies to ensure minimal impact
and maximum restoration of the land associated with natural gas production.
The trucks used to transport equipment, fracturing fluid ingredients, and water to the wellpad,
drilling rigs, compressors, and pumps all emit air pollutants, including carbon dioxide, nitrogen
and sulfur oxides (NOx and SOx), and particulate matter. Volatile organic compounds (VOCs)
and other pollutants associated with natural gas and fracturing fluids can enter the air from wells
and evaporation pits. In addition, natural gas, whose main component is methane, is itself a
greenhouse gas more potent than carbon dioxide and could represent a significant source of
emissions during the gas production process.52
Many technologies and practices to reduce venting and leakage during gas production and
transport have been compiled by the U.S. EPA‘s Natural Gas STAR program.53 Emissions of
gases that contribute to local air pollution, public health risks and climate change can be reduced
by available control technologies, improved monitoring, and more efficient production
operations. (The impacts of natural gas development with air quality will be the focus of a future
briefing paper by the Natural Gas and Sustainable Energy Initiative.)
Even compared with drilling, which might use up to a million gallons of water per well,
hydraulic fracturing is a water-intensive procedure, requiring between 2 and 8 million gallons per
well fractured.54 In the Barnett Shale, for example, an average of almost 3 million gallons of
water is used per well, the great majority of which is used for hydraulic fracturing.55 Since
development of this resource will require tens of thousands of shale gas wells to be drilled, the
required volumes of water are dramatic.
Any set of water use regulations must take into account local hydrology and competing uses for
the water in a given area. Operators and regulators must work together to explore opportunities
to reduce water use and increase recycling of produced water. Greater reuse of fracturing fluids
would reduce demands on community water supplies. Steps can also be taken to utilize excess
water during peak seasonal run-off and to try to use less water during slickwater fracturing
operations. (The water requirements for natural gas development will be the focus of a future
briefing paper by the Natural Gas and Sustainable Energy Initiative.)
While a well is being drilled and completed, operators are generally working around the clock
for several weeks. Drilling sites generate significant amounts of noise pollution, although noise
13
can be reduced through the construction of sound barriers.56 Gas development can also affect
communities in less tangible ways. While it may stimulate the local economy and provide jobs,
gas development may also lead to increased traffic and greater strains on public resources.
Operators must work with local stakeholders to minimize the impact of gas development
activities on a community‘s resources and quality of life.
BOX: Current Regulatory Framework Governing Shale Gas Development
Most regulation of oil and gas development is currently left to the states, where regulatory
bodies are in charge of enforcing state environmental laws as well as rules and regulations
specific to oil and gas production. Rules and regulations developed by state agencies such as
the Colorado Oil and Gas Conservation Commission, the Texas Railroad Commission, or the
Pennsylvania Department of Environmental Protection govern the specifics of gas production,
requiring producers to obtain permits before drilling, and requiring certain standards and
practices to be used during well construction, hydraulic fracturing, waste handling, and well
plugging. State regulations also deal with tanks and pits as well as any chemical or waste
water spills.
Currently, there is significant variation in the particulars of these rules and regulations from
state to state. For example, in a 2009 survey of the 27 largest gas-producing states, the
Ground Water Protection Council (GWPC) found that 25 states required surface casing to be
set below the deepest groundwater, 21 require a cement set-up period or test such as a cement
bond log, 10 require companies to list chemicals or pressures used during hydraulic
fracturing, and none requires companies to list an estimate of how much of this fracturing
fluid flows back to the surface after a well has been fractured. The non-profit STRONGER
(State Review of Oil and Natural Gas Environmental Regulations) has been updating
guidelines for reviews of state programs since 1999. As list of states that have completed
initial and follow-up reviews is available on STRONGER‘s website (www.strongerinc.org).
In addition to these state rules and regulations, some federal environmental regulations also
apply to shale gas development. For example, the Clean Water Act regulates contaminated
storm water runoff and surface discharges of water from drilling sites, and the 1986
Emergency Planning and Community Right-to-Know Act (EPCRA) requires companies to
post material safety data sheets describing the properties and health effects of any chemicals
stored in quantities that exceed 10,000 pounds. In some cases, states may obtain authority to
enforce a federal law. The Safe Drinking Water Act (SDWA), which regulates the
underground injection of waste water from gas wells, though not hydraulic fracturing, is one
example of a federal law which allows state regulatory agencies to obtain primacy over
enforcement if they demonstrate that they can do so to the minimum standards laid forth by
the Environmental Protection Agency.
Source: See Endnotes 2 and 5 for this section.
14
IV. Conclusion
New supplies of gas from shale could provide many U.S. states with an attractive, lower-carbon
transition fuel on the path to a fully renewable energy supply, while providing jobs and
generating appreciable revenue. However, these opportunities cannot be realized unless the
environmental risks posed by shale gas development are managed effectively. Our analysis
suggests that while shale gas development poses significant risks to the environment, including
faulty well construction, blowouts, and above-ground contamination due to leaks and spills of
fracturing fluids and waste water, technologies and best practices exist that can help manage
these risks.
Best practices are currently being applied by some producers in some locations, but not by all
producers in all locations. Enforcing strong regulations is necessary to ensure broader adoption
of these practices and to minimize risk to the environment. In addition, if increased shale gas
development is to be undertaken responsibly, the cumulative risks of developing thousands of
wells must be considered. Ongoing studies by the Environmental Protection Agency and others
examining the environmental impacts of hydraulic fracturing will arm state and federal decision
makers with critical information upon which to base future regulations.
By developing and adopting innovative best practices, industry can take a proactive role in
addressing the environmental risks associated with shale gas development. The Houston
Advanced Research Center and Texas A&M University are working with companies,
environmental organizations, universities, government laboratories, state and federal agencies,
and others to reduce the environmental impacts of drilling and production. The Environmentally
Friendly Drilling Systems Program focuses on solutions to reduce the footprint of drilling
activities, ensure the safe transport and disposal of drilling fluids and cuttings, lower air and
noise pollution, and minimize other risks to the environment.57
Robust regulatory oversight is an important ingredient to assure environmental and public
protection. Under current U.S. laws, some aspects of shale gas development are regulated by the
Clean Water Act, the Clean Air Act, and the Safe Drinking Water Act, but regulation of drilling
and hydraulic fracturing is left largely to the state level where regulatory capacity and
enforcement, as well as the regulations themselves, vary widely.
The state of Colorado recently revised its oil and gas rules to strengthen protections for the local
environment.58 The new rules, which went into effect on April 1, 2009, were devised after a
boom in gas production from coal bed methane and tight sands was linked to both environmental
and public health problems as well as permitting bottlenecks. Colorado Governor Bill Ritter has
argued that the public assurance that these rules created was as an important prerequisite for
adoption of Colorado‘s 2010 Clean Air-Clean Jobs Act. That Act requires Colorado‘s rate
regulated utilities to retire or re-power some 900 megawatts of coal-fired power plants,
displacing them primarily with natural gas.59 However, many independent producers feel that
they were excluded from what was touted as a multi-stakeholder process and argue that the
Colorado Oil and Gas Conservation Commission did not fully account for the increased costs the
new rules would impose, while some environmentalists feel that the revisions did not go far
enough.60
15
Colorado‘s example provides valuable lessons to other states pursuing their own reform of oil
and gas regulations. The Wyoming Oil and Gas Conservation Commission passed a package of
new oil and gas drilling rules on June 8, 2010. These rules make Wyoming the first state to
require operators to disclose the composition and concentration of chemicals used in hydraulic
fracturing.61 Other shale-producing states may soon follow suit.62
New York, a relative newcomer to the modern oil and gas industry, has been the site of a
contentious debate over future development of the state‘s gas resources in the Marcellus Shale.
The New York Department of Environmental Conservation (NYSDEC) has been charged with
updating rules regulating horizontal drilling and high-volume hydraulic fracturing and is
currently evaluating public comments on a draft Supplemental Generic Environmental Impact
Statement that it released in September 2009.63 In the meantime, 10 bills relating to shale gas
development, including one that would place a moratorium on drilling until 120 days after the
EPA‘s study of hydraulic fracturing is completed, are making their way through the state
legislature.64 In neighboring Pennsylvania, where over 564 wells were drilled in the Marcellus
Shale during the first half of 2010, Governor Ed Rendell has said that he would sign a bill calling
for a three-year moratorium on new leasing of state forest land for gas exploration while
potential environmental impacts are studied.65
The experiences of Colorado, Wyoming, Pennsylvania, and New York have demonstrated that
strong public pressure exists for stricter oversight of the oil and gas industry and that state
regulators can and will move forward in strengthening their own regulations. If they are
produced responsibly, shale gas resources in the United States could play a central role in
building a low-carbon energy economy. Greater outreach and public education about shale gas
development are clearly necessary to enable the many stakeholders engaged in shale gas
development to work together to find the most effective technological and regulatory solutions
for developing shale gas resources while protecting the environment and public interest.
16
Endnotes
1 Worldwatch Institute, The Role of Natural Gas in a Low-Carbon Energy Economy (Washington, DC:
April 2010). 2 Groundwater Protection Council (GWPC), Modern Shale Gas Development in the United States: A
Primer, prepared for the U.S. Department of Energy, National Energy Technology Laboratory (NETL)
with ALL Consulting (Oklahoma City, OK: April 2009), p. 13. 3 Figure 1 from U.S. Department of Energy, Energy Information Administration (EIA), ―Shale Gas
Plays, Lower 48 States,‖ available at www.eia.doe.gov/oil_gas/rpd/shale_gas.pdf, viewed 7 July 2010. 4 Calculated based on 2009 U.S. crude oil production of 5.31 million barrels per day, and conversion
factor of 1 billion cubic feet NG = 0.19 million barrels of oil equivalent. Shale gas production rates for
2009 from Vello Kuuskraa, Advanced Resources International, ―The ‗Paradigm Shift‘ in U.S. and
Worldwide Natural Gas Supplies,‖ presentation at Global Unconventional Gas Conference 2010,
Amsterdam, 17 June 2010. U.S. crude oil production from EIA, ―Crude Oil Production,‖ Petroleum
Navigator, available at www.eia.gov/dnav/pet/pet_crd_crpdn_adc_mbblpd_a.htm, viewed 7 July 2010.
Conversion factor from BP, Statistical Review of World Energy June 2010 (London: 2010). 5 Figure 2 from GWPC, State Oil and Natural Gas Regulations Designed to Protect Water Resources,
prepared for NETL (Oklahoma City, OK: May 2009), p. 20. 6 State regulations requiring surface casing to be set below the deepest ground water exist in 25 of the
27 oil and gas-producing states surveyed by the GWPC in 2008. Twenty-four states required surface
casing to be cemented along its entire length. GWPC and NETL, State Oil and Gas Regulations
Designed to Protect Water Resources (Oklahoma City: May 2009), p. 19. 7 Twenty-four of 25 states surveyed required surface casing to be cemented along its entire length, per
Ibid. 8 Not all states require blowout preventers for all wells. Colorado‘s new rules regulating oil and gas
drilling, which may be considered a best management practice, requires operators to: install blowout
preventer equipment (BOPE) on any well expected to flow, inspect it daily, ensure that it has a
sufficient rating to accommodate the maximum anticipated surface pressure, and ensure that rig
employees understand and can operate it. Operators must also pressure test the casing and BOPE after
each new string is added, and proceed with drilling only when this equipment has been tested and
found to be functional. Colorado Oil and Gas Conservation Commission (COGCC), ―Rules and
Regulations, 603 (i) (as of 1 April 2009),‖ available at http://cogcc.state.co.us/. 9 The Barnett Shale is a notable exception, with a thickness of 100 to 600 feet. GWPC, op. cit. note 2, p.
17. 10
GWPC, op. cit. note 5, p. 21. 11
Figure 3 created by Mark Zoback, Stanford University. 12
M.J. Economides and K.G. Nolte, eds., Reservoir Simulation (West Sussex, U.K.: John Wiley & Sons,
Ltd., 2000). 13
Figure 4 was created by Bradford Copithorne and Mark Zoback and is based on proprietary data made
available by a gas company currently operating in the Barnett Shale. 14
Thomas R. Driscoll, ―Marcellus Could Become Largest Gas Play,‖ Barclays Capital, 15 March 2010. 15
S.M. Currie, D. Ilk, and T.A. Blasingame, ―Continuous Estimation of Ultimate Recovery, SPE
132352,‖ paper presented at 2010 SPE Unconventional Gas Conference, Pittsburgh, PA, 23–25
February 2010. 16
E. Siebrits et al., ―Refracture Reorientation Enhances Gas Production in Barnett Shale Tight Gas
Wells,‖ presented at the 2000 Society of Petroleum Engineers Annual Technical Conference and
Exhibition, Dallas, TX 1–4 October 2000; Mukul Sharma, ―Improved Reservoir Access through
Refracture Treatments in Tight Gas Sands and Shales,‖ presentation (Golden, CO: 15 April 2009). 17
GWPC, op. cit., note 5, p. 27. 18
Figure 5 from GWPC, op. cit., note 2, p. 54.
17
19
Daniel Arthur, Hydraulic Fracturing Considerations for Natural Gas Wells of the Marcellus Shale,
presented at the GWPC 2008 Annual Forum, Cincinnati, OH, 21–24 September 2008. 20
GWPC, op. cit., note 2, p. 17. 21
Arthur, op. cit note 19. 22
Kent Perry, Gas Technology Institute (GTI), personal communication with Mark Zoback, 9 June 2010. 23
Ohio Department of Natural Resources, Division of Mineral Resources Management, ―Report on the
Investigation of the Natural Gas Invasion of Aquifers in Bainbridge Township of Geauga County,
Ohio,” (Columbus, OH: 1 September 2008). 24
American Petroleum Institute, ―E & P Safety Standards,‖
www.api.org/standards/epstandards/index.cfm, viewed 12 July 2010. 25
―Addendum: State Oil and Gas Regulations Reference Document,‖ in GWPC op. cit notes 5. pp. 66-
264. 26
GWPC, op. cit., note 5, pp. 19-21. 27
Katie Howell, ―Marcellus Well in Pa. Blows Out,‖ E&E News PM,4 June 2010;Vicki Smith, ―7
Burned in W. Va. Gas Well Blast Likely to Survive,‖ Charleston Gazette, 7 June 2010. 28
Howell, op. cit., note 27. 29
Smith, op. cit., note 37. 30
Ben Casselman, ―Temblors Rattle Texas Town,‖ Wall Street Journal,12 June 2009. 31
Cliff Frohlich et al., ―Dallas-Fort Worth Earthquakes Coincident with Activity Associated with
Natural Gas Production,‖ The Leading Edge (Society of Exploration Geophysicists), March 2010, pp.
270-75. 32
Disposal well count from Ibid. 33
GWPC, op. cit., note 2, p. 63. 34
New York State Department of Environmental Conservation (NYSDEC), Draft Supplemental General
Environmental Impact Statement (Albany:30 September 2009), pp. 5-35 to 5-40. 35
Energy companies such as Halliburton and BJ Services are developing ―environmentally focused‖
fracturing fluids that do not contain BETX (benzene, ethyl benzene, toluene and xylene) components
and that meet the requirements of the Energy Act of 2005 and the Clean Water Act. Halliburton, ―Ultra
Clean Fracturing Fluid Technology‖(December 2007), available at
www.halliburton.com/ps/default.aspx?pageid=1844&navid=105&prodid=PRN::JSHSB7C4S; BJ
Services, ―Environmentally Friendly Fracturing Fluids‖ (2008) available at
www.bjservices.com/website/index.nsf/WebPages/Shale-Papers-Expandable-Section. 36
GWPC, op. cit., note 2, p. 41. 37
American Power Act, Sec. 4131. 38
Amy Mall, et al., Drilling Down: Protecting Western Communities from the Health and
Environmental Effects of Oil and Gas Production (New York: Natural Resources Defense Council:
September 2007), p. vi. 39
GWPC, op. cit., note 2, p. 66. 40
Thomas Hayes, GTI, personal communication with Mark Zoback, June 1, 2010. 41
GWPC, op. cit., note 2, p. 69. 42
Jeff Jollie, U.S. Environmental Protection Agency (EPA), Office of Ground Water and Drinking
Water, personal communication with Saya Kitasei. 21 April 2010. 43
GWPC, op. cit., note 2, p. 69. 44
Pennsylvania Department of Environmental Protection (PADEP), ―DEP Investigates Source of
Elevated Total Dissolved Solids in Monongahela River,‖ press release (Harrisburg, PA: 22 October
2008). 45
Ibid.; Riverkeeper, ―Impacts and Incidents Involving High-Volume Hydraulic Fracturing From Across
the Country,‖ Appendix 1 to comments on NYSDEC Draft Supplemental General Environmental
Impact Statement, available at www.riverkeeper.org/wp-content/uploads/2010/01/Riverkeeper-
DSGEIS-Comments-Appendix-1-Case-Studies.pdf.
18
46
A.W. Gaudlip, L.O. Paugh and T.D. Hayes, ―Marcellus Shale Water Management Challenges in
Pennsylvania, SPE -119898-PP,‖ paper presented at the 2008 Shale Gas Production Conference, Ft.
Worth, TX, 16-18 November, 2008; M.E. Blauch, ―Developing Effective and Environmentally
Suitable Fracturing Fluids using Hydraulic Fracturing Flowback Waters, SPE 131784,‖ paper
presented at 2010 SPE Unconventional Gas Conference, Pittsburgh, PA, 23-25 February 2010. 47
GWPC, op. cit., note 5, p. 29. 48
Delaware River Keeper, ―Natural Gas Well Drilling and Production: In the Upper Delaware River
Watershed,‖ www.delawareriverkeeper.org/resources/Factsheets/Drilling_and_Production.pdf 49
GWPC, op. cit., note 5, p. 28-9. 50
PADEP, Oil and Gas Management Program, Inspection Record #1802228 (Harrisburg, PA: 27 May
2009). 51
PADEP, ―DEP Penalizes Range Resources $141,175 for Spill in High Quality Waterway,‖ press
release (Harrisburg, PA: 14 May 2010). 52
Methane leakage could be responsible for 71 percent of total greenhouse gas emissions from the U.S.
natural gas industry if the IPCC‘s recent 20-year global warming potential is used. Jerome Blackman,
U.S. EPA, ―Methane Emissions Reductions: Barriers, Opportunities and Possibilities for Oil and
Natural Gas,‖ presentation at Natural Gas STAR Program 2009 Annual Implementation Workshop, 20
October 2009. 53
EPA, Natural Gas STAR Program, ―Recommended Technologies and Practices,‖
http://www.epa.gov/gasstar/, viewed 26 May 2010. 54
This range is given for a 4,000-foot lateral wellbore in the Marcellus, but concurs with estimates in
other plays. The amount of water required depends on both the length of the well and the properties of
the target formation. NYSDEC, op. cit., note 30, p. 5-72. 55
GWPC, op. cit., note 2, p. 64. 56
Ibid., p. 49. 57
Environmentally Friendly Drilling Systems Program Web Site,
www.efdsystems.org/Home/tabid/1253/Default.aspx, viewed 12 July 2010. 58
Colorado Oil and Gas Conservation Commission, ―COGCC Amended Rules,‖ available at
http://cogcc.state.co.us/, viewed 7 June 2010. 59
Office of Governor Bill Ritter, Jr., ―Gov. Ritter Signs Historic Clean Air-Clean Jobs Act,‖ press
release (Denver: 19 April 2010). 60
David Williams, ―Ritter‘s oil and gas rules one year later,‖ Colorado Independent (25 May 2010),
available at http://coloradoindependent.com/54104/ritters-oil-and-gas-rules-one-year-later, viewed 8
June 2010. 61
Phil Taylor, ―Wyoming set to Approve Fracking Disclosure Rules,‖ Land Letter, 3 June 2010 62
David Williams, ―Major Gas-producing States Debating Colorado-style Drilling Regulations,‖
Colorado Independent, 21 May 2010. 63
NYSDEC, Draft Supplemental Generic Environmental Impact Statement on the Oil, Gas and Solution
Mining Regulatory Program (updated 26 October 2009), www.dec.ny.gov/energy/58440.html,
updated 26 October 2009. 64
Jon Campbell, ―Bill to Halt Hydro-fracking Close to Assembly Vote,‖ PressConnects.com (4 June
2010). 65
PADEP, ―2010 Wells Drilled by Operator as of 06/03/2010,‖
www.dep.state.pa.us/dep/deputate/minres/oilgas/2010%20Wells%20Drilled%20by%20Operator.htm,
viewed 7 June 2010; Marc Levy and Joe Mandak, ―Rendell Backs Moratorium on Leasing Public
Land for Gas Drilling,‖ Ithaca Journal,11 May 2010.
How Energy Choices Affect Fresh Water Supplies: A Comparison of U.S. Coal and Natural Gas
Briefing Paper 2
Emily Grubert
Saya Kitasei
November 2010
Natural Gas and Sustainable Energy Initiative
1
How Energy Choices Affect Fresh Water Supplies:
A Comparison of U.S. Coal and Natural Gas
Emily Grubert and Saya Kitasei
I. Introduction
Water and energy are critical and interdependent resources. The production and use of energy
requires both the withdrawal and consumptiona of water and represents one of the largest
demands on fresh water in the United States. In 2005, U.S. power plant cooling systems
withdrew 143 billion gallons of fresh water per day, accounting for 41 percent of domestic fresh
water withdrawals. Mining and fuel extraction withdrew an additional 2 billion gallons per day. 1
Fresh water in turn requires energy to be pumped, treated, and transported before it can be used.
In a 2003 study by the Government Accountability Office, water managers in 36 of 47 surveyed
U.S. states predicted that their states or regions would face water shortages by 2013.2 The study
warned that the depletion of groundwater aquifers, the rising demand for fresh water, and the
potential impacts of climate change could all reduce water availability.
Declining water availability is already limiting energy choices. Over the past decade, concerns
about water availability have halted power plant construction or operation in the U.S. states of
Arizona, California, Colorado, Georgia, Massachusetts, Missouri, New Mexico, North Carolina,
Pennsylvania, Rhode Island, South Dakota, Tennessee, Texas, and Washington.3 As state and
local governments around the country plan their electricity generation mix for the coming years,
they will need to consider the water dimension of their decisions.
A shift from reliance on coal-fired steam-turbine generators (which provided about 44 percent of
U.S. electricity generation in 2009) to combined-cycle plants fueled by natural gas (about 19
percent of generation) could have a profound effect on the power sector’s water demands.4 The
relatively high efficiency of natural gas combined-cycle (NGCC) plants means that they generate
electricity using less fuel and creating less than half the greenhouse gas emissions that coal
plants do.5 Moreover, NGCC plants consume one-tenth to one-half as much fresh water as
conventional coal plants do to generate each unit of electricity—a critical advantage in regions
where water shortages present as urgent a concern as air pollution and climate change.6
A newfound abundance of economically viable natural gas from unconventional reservoirs,
combined with tightening air-quality standards and/or carbon constraints, could enable natural
gas to claim a large share of the U.S. power market from coal.7 This should reduce water demand
at the power plant. However, complete comparisons of coal- and natural gas-generated electricity
must account for water demands during the full fuel cycle. For example, extracting
unconventional natural gas, including shale gas, tight gas, and coalbed methane, often requires
a Withdrawal refers to the removal of water from a natural source, which may be either returned to the
source or consumed.
2
significantly more water than conventional natural gas extraction because of the use of hydraulic
fracturing, a water-intensive well-stimulation technique.8
This paper examines the impacts on U.S. fresh water resources of generating electricity from
coal and natural gas, from the point of fuel extraction through the fuel’s use at the power plant.
Although fuel extraction can require locally significant quantities of water, by far the largest
water consumer in the life cycle of electricity is power plants—which can be responsible for
more than 90 percent of the water consumed to produce a kilowatt-hour of electricity.9
NGCC power plants generally use less water per unit of electricity generated than coal power
plants due to higher efficiency and less need for emissions controls (which in many cases
represent an extra water use at coal plants). Thus, shifting generation from coal to natural gas
should reduce the electricity sector’s water needs whether unconventional or conventional
natural gas is used.
This paper also finds that coal extraction has higher potential for long-term degradation of water
resources than does natural gas extraction. However, the quantitative and qualitative water
impacts of fuel extraction vary by site and method. Using natural gas instead of coal will likely
involve less damage to the fresh water system, with localized exceptions.
Although this paper focuses on water, water is not the only resource affected by the extraction,
processing, transport, and use of coal and natural gas; air, land, and communities also face
impacts. All of these impacts must be considered in a holistic comparison of the effects of coal
and natural gas production and use. (See Table 1.)
Table 1. Lifecycle Environmental Impacts of Coal and Natural Gas Production and Use
Coal Natural Gas
Land use Land intensive
Reclamation can be difficult
Well pads are relatively small,
especially for horizontal wells
Solid waste Large volumes of processing waste
and combustion byproducts must be
disposed of
Limited volumes of drill cuttings
Air emissions CO2 emissions when combusted;
some methane leakage
Emissions of particulates, sulfur,
nitrogen, mercury, other metals at power plant
CO2 emissions (less than half those of
coal) when combusted; methane
leakage
Other emissions from drilling site,
pipelines, and service trucks
Water
pollution and disturbance
Chemical pollution from mining and
processing
Minewater discharge and
groundwater pumping at mines disturbs stream and groundwater
flows
Thermal pollution from power plants
Chemical pollution from accidental
spills or faulty well completions
Briny flowback and produced water
often has high solids content and sometimes has naturally occurring
radioactive material (NORM)
Thermal pollution from power plants
Water
consumption Limited water use during mining
Steam-turbine power plants require
large amounts of water for cooling
Water used for drilling and hydraulic
fracturing
Combined-cycle and gas-turbine power
plants use less water for cooling
3
II. Water Impacts from Fuel Extraction, Processing, and Transport
Evaluating the water impacts of displacing coal with natural gas in the power sector requires a
clear understanding of these fuels’ demands on water. Coal and natural gas must be extracted,
processed, and transported before they reach power plants. Each of these stages uses and affects
the supply and quality of water. (See Figure 1.) Unfortunately, quantitative data on how resource
extraction affects water quality and demand are scarce, as water impacts are site-specific
depending on how and where the extraction occurs. In particular, the quality of a coal or natural
gas resource and how close it is to water affects its need for water and its potential to pollute.
Coal
The two main methods for mining coal are surface mining and underground mining. For surface
mining, miners uncover coal by removing the rock at the surface, known as the overburden; for
underground mining, they dig beneath the overburden and work under a rock roof. In the United
States, surface mines account for 69 percent of total coal production, and underground mines for
31 percent.10
Both underground and surface mines are often situated at least partially below the water table, so
miners must pump out water from the working area during much of the mining process (from
pre-excavation until the mine is abandoned). This ―mine dewatering‖ includes removing water
from rain or snow in addition to water that is already in the coal formation. Dewatering
equipment discharges most of the water at the surface, although some operations capture or treat
this water for reuse for dust suppression and other needs.
Mine dewatering can lower water tables for decades, affecting groundwater levels and flow
patterns around the mine for miles. But it can also prevent the long-term exposure of water to
any contaminants in the coal. Coal is highly heterogeneous, with the full range of the world’s
coals containing 76 of the 92 naturally occurring elements.11 Both the combustion of coal and its
exposure to water can release contaminants to the environment, making remediation difficult.
Most U.S. coal mining takes place in two regions: the Appalachian Mountains in the mid-
Atlantic (33 percent) and the Powder River Basin in the Western states of Wyoming and
Montana (42 percent).12
Appalachian coals are characterized by high energy density and
relatively high sulfur content, whereas Western coals are typically lower energy density and low
in sulfur. The United States is expected to obtain a rising share of its coal from the West in the
future, in part because of the lower sulfur emissions released during combustion.
In addition to the coal that can be used for energy, mining involves removing large volumes of
waste rock from the ground. Removal of this rock can disrupt surface and groundwater flows,
and waste rock disposal can bury streams. Rock surfaces that are inert when surrounded by other
rock can oxidize and leach material into water when they are exposed to air and water.
4
5
Coals and waste rock in the eastern United States are particularly problematic, not just because
they release higher sulfur emissions during combustion, but because the sulfur compounds are
reactive when exposed to air and water. Waste rock generally contains more of these
contaminating compounds than coal, and Appalachia’s thin coal seams are often associated with
more waste rock than Western seams, which can be 10 times as thick. And because Appalachian
coal is located in wet, mountainous areas with many streams, coal-related contamination is more
likely to affect the water in this region than in the semi-arid West.
Environmental regulations require that most Eastern coals be processed before they are used in
power plants. Companies remove impurities by crushing the coal into smaller pieces in water,
which adds 1–2 gallons of water demand per million Btu of coal.13
Once used, this water is
discharged to holding ponds and often contains fine coal particles that are difficult to remove.14
Since coals in the West usually have fewer impurities as well as lower energy densities than
those in the East, such preparation before use is often not considered worthwhile for Western
coals.15
Most U.S. coal is transported by rail, barge, or truck, so water usage for transportation is low. At
mine sites, water consumption is usually limited to domestic services such as toilets and showers
for workers and dust suppression, which involves spraying water from the mine on coal piles and
roads to reduce airborne dust. Because contamination from this dust is typically transferred to the
water, the associated pollution is often moved but not eliminated, presenting a major challenge
for mitigation. Coal is usually stored in open piles, so precipitation can become contaminated
runoff as well.
One of coal mining’s greatest impacts on water quality comes from abandoned mines, as water
moves through pits or tunnels in rock surfaces that remain chemically active. Abandoned surface
mines often turn into lakes, whereas underground mines experience groundwater seepage.
Without control measures, new water continuously enters abandoned mines and piles of waste
rock, which means that a poorly remediated mine can contaminate water for decades or even
centuries. Such contamination is a bigger problem in the eastern United States than the West, so
the expected continued shift to Western coals could reduce the negative impacts on fresh water
quality.16 But because water associated with Western coals is often of high quality, containing
few contaminants, mining in the region can deplete already-scarce water resources that could be
used for other purposes.
After mining ceases, reclaimed mine sites continue to consume water to reestablish vegetation,
although the amount varies by climate. In the U.S. West, estimates of this usage range from
616,000 liters to 925,000 liters of high-quality water per acre of reclaimed land annually, over a
10-year period.17
In 2009, Central Appalachia produced some 196 million short tons of high-sulfur, bituminous
coal—18 percent of the U.S. coal supply—from 399 underground and 403 surface mines.18
Because coal seams in the region are thin—between 3 and 15 feet—mining companies may use
mountaintop removal to access seams that are less than 30 inches thick. This involves stripping
the tops of mountains and pushing waste rock into adjacent valleys. Mountaintop removal
generates large volumes of waste rock that may be stored in impoundment slurry dams or large
tailings piles, which can bury or eliminate streams.
6
Water flow following mountaintop removal can liberate contaminants such as pyrite and heavy
metals, creating acid mine drainage (AMD) and other contaminated runoff.19
AMD is associated
with certain sulfur compounds and is highly damaging to water quality, particularly since the
acid can dissolve other contaminants into water. AMD is also persistent: rocks containing sulfur
can produce AMD as long as air and clean water are in contact with an exposed coal seam.
Wyoming and Montana’s Powder River Basin is the largest U.S. source of coal, with only 17
mines accounting for 496 million short tons, or 42 percent, of domestic production.20
The basin
is expected to provide an even larger share in the future.21 Extremely thick seams (up to 150 feet)
of low-sulfur, sub-bituminous coal are exploited through open-pit surface mines.
Because coal-mining areas in the U.S. West are less mountainous than in the East, there are
fewer headwater streams that can be affected by mining. Even when Powder River Basin coal
does impinge on water resources, it does not pose as great a risk of contamination as
Appalachian coals because it contains lower levels of sulfur, heavy metals, and other
contaminants.22 In some areas, the coal even acts as a natural filter and holds potable water—
creating a tension between developing coal resources and preserving high-quality water supplies
for drinking, livestock, and irrigation.
Some of the water in Western mines has high levels of sodium, which can negatively affect soils
when the water is discharged.23 The potential negative impact on agriculture from pumping water
and sodium contamination is one reason that Montana has restricted mining in its portion of the
Powder River Basin.24
Natural Gas
The United States produced 21 trillion cubic feet (tcf) of natural gas in 2009.25
Up until 2008,
most of the nation’s natural gas was produced from ―conventional‖ reservoirs, which have
relatively high permeability, enabling natural gas to flow easily to drilled wells. More recently,
production in less-permeable ―unconventional‖ reservoirs, including tight sands, deep shales, and
natural gas-bearing coalbeds, has overtaken conventional production and is projected to grow
through 2030 at least.26
As with coal, the water needs and risks of natural gas extraction depend
on geology and geography. In general, natural gas production from unconventional reservoirs
requires more water than production from conventional reservoirs.
Both conventional and unconventional natural gas drilling use water to lubricate and cool the
drill bit, consuming hundreds of thousands of gallons per well. The ―drilling mud‖ that results
can contain toxins, posing a disposal challenge: it may be injected underground; treated and then
reused or released; or dried and disposed of. Similarly, both conventional and unconventional
wells can produce naturally occurring water from the reservoir rock. This ―formation water‖ has
typically spent millions of years in contact with ancient rock formations and can therefore
contain high concentrations of salts, naturally occurring radioactive material (NORM), and other
contaminants including arsenic, benzene, and mercury.27 Produced water volumes vary by basin:
they are generally low for deep shale wells, due to the extreme temperature and pressure, and
high for coalbed methane wells, which must be dewatered.
All natural gas wells are subject to accidents such as blowouts, improper well construction and
abandonment, and contamination associated with the disposal of drilling mud and produced
7
water. Any structure that penetrates water aquifers, such as a well, has the potential to
contaminate these water sources.
After it is extracted, natural gas must be processed, transported, and stored for use. Natural gas
processing uses about two gallons of water per million Btu of natural gas, removing liquid
hydrocarbons, acid gases, carbon dioxide, and water vapor to produce a nearly pure methane
stream.28 Transportation in pipelines requires an additional one gallon of water per million Btu.29
Natural gas can be stored in oil or gas reservoirs, aquifers, or salt caverns. Salt cavern storage has
the highest water impact, as salt must be dissolved with a one-time use of 500–600 gallons of
water per million Btu of capacity, yielding a briny waste stream. This represents 4 percent (and
growing) of U.S. natural gas storage volume.30
Other storage options for natural gas use little
water.
For unconventional natural gas production, the major additional water need is associated with
―hydraulic fracturing,‖ a commonly used technique that enables drillers to extract natural gas
from rock with low permeability, or interconnected spaces. The goal is to give natural gas
molecules a pathway to the wellbore by stimulating and propping open fractures in the rock
formation containing the natural gas. The fracturing is accomplished by pumping 2–4 million
gallons of water mixed with sand and chemical additives into the gas-bearing layer of rock at
high pressures.31 The water used for fracturing is often transported in trucks and stored in tanks at
drilling sites. The substantial transportation needs associated with moving water can stress
nearby stream banks, contributing to erosion and adding sediment to surface water.32
Once in the ground, a large portion of the fracturing fluid may be trapped in the target
formation.33 The rest is pumped to the surface as ―flowback,‖ combined with any water
―produced‖ from the formation itself.34 Both flowback and produced water represent large waste
streams that must be disposed of in injection wells or evaporation pits, or municipal treatment
plants, or treated and reused to fracture future wells. Where injection or evaporation are not
locally tenable, waste water must be trucked to a treatment or disposal site (which increases truck
traffic) or recycled and reused in other fracture jobs. Numerous efforts to make produced water
reusable for fracturing are under way.35 If flowback and produced water are disposed of
improperly, or if the well is poorly constructed, waste water or natural gas (methane) can
contaminate surface water, threatening public and environmental health.36
With shale gas production, the two major pathways to water contamination are activities at the
surface and errors below ground. At the surface, water resources—particularly stream banks—
can be disturbed by truck traffic associated with wellpad construction, day-to-day industrial
activity, and in particular, trucking water to and from the site for fracturing and then disposal.
Other surface risks include chemical spills and leaching from produced water and flowback
stored above ground. Mitigation options do exist, however. Good road planning and reduced
truck traffic can protect stream banks. Using more benign chemicals or stricter handling
standards reduces the risk of harmful chemical spills. And better water-disposal practices,
including lining storage pits and treating water on site, can reduce contamination risk from
produced water.
Errors below ground can endanger water resources during shale gas production as well. Properly
casing wells mitigates substantially the risk of contamination when an aquifer is penetrated. One
element of this is identifying zones that need to be isolated in order to prevent potential shallow
8
pockets of natural gas in formations above the target layer from entering into ground water.37
Another way to mitigate the risk of contamination during aquifer penetration is by using
horizontal wells rather than vertical wells: horizontal wells allow drillers to produce natural gas
from a much larger region using fewer wells, thus penetrating aquifers less frequently. When
multiple horizontal wells are drilled from one well pad, the risk of contamination is reduced even
further since any problems will be localized to that area.38
Conventional natural gas has a low extraction-related water footprint, largely because
conventional wells do not require hydraulic fracturing. Shale gas, tight gas, and coalbed
methane, however, can use and affect much larger amounts of water during their extraction,
raising concerns that switching from coal to natural gas could be less benign for water supplies if
an increasing share of natural gas is produced from unconventional formations.
Unconventional natural gas has rapidly gained importance in the United States. Estimates of
potential shale gas resources have increased dramatically in the past two years and now stand at
616 tcf, or 33 percent of potential U.S. natural gas resources.39
Most natural gas-bearing shales in
the United States are located thousands of feet below the Earth’s surface, and all have very low
permeability, necessitating horizontal drilling and hydraulic fracturing. Approximately 163 tcf of
potential natural gas resources are thought to exist in coal seams as coalbed methane.40
Coalbed
methane basins are generally shallower than shales and can be located in drinking water aquifers,
meaning that wells must be dewatered rapidly and that hydraulic fracturing can pose a greater
risk of water contamination.
The Barnett Shale in Texas has served as a testing ground for hydraulic fracturing and
horizontal drilling techniques in shales. One concern in this region, which is highly urbanized, is
the high quality of water used for natural gas wells: potable water from both fire hydrants and the
Dallas-Fort Worth airport is used for drilling and fracturing. Barnett wells use an average of
250,000 gallons of drilling water per well and 3.8 million gallons of fracturing water each time a
well is hydraulically fractured.41
Water is disposed of through injection wells or recycled for
further fracturing jobs.42
Barnett wells produce very little formation water and an average of 2.7
billion cubic feet (bcf) of natural gas over their lifetimes.43
The Marcellus Shale underlying much of Appalachia presents challenging terrain in sensitive
watersheds. Marcellus wells are drilled with air mists and water- or oil-based muds, requiring
some 80,000 gallons of water for drilling and 3.8 million gallons for hydraulic fracturing per
well.44
Recovered fracture fluids are disposed of primarily through treatment and discharge or
recycling, although water treatment plants already are proving inadequate to deal with the
volumes and high salinity flowback.45
Safe fluid disposal is likely a greater challenge than water
availability, especially since few injection wells for water disposal exist in the Marcellus due to
challenging geology: water must be treated and recycled or discharged, which poses a
contamination risk for surface and ground water.46 However, water availability can be a barrier
when stream flows are low, as in the summer, and even relatively small withdrawals can affect
aquatic life. Marcellus wells produce about 3.7 bcf of natural gas over their lifetimes and almost
no formation water.47
The San Juan Basin is a mature coalbed methane field in the Four Corners area of the U.S.
Southwest. Unlike shale and tight gas reservoirs, coalbeds must be dewatered to reduce pressure
and maximize natural gas production.48 Water content and quality varies throughout the San Juan
9
Basin, with wells producing between zero and 10,000 gallons of water (average 1,000 gallons)
each day and produced water qualities ranging from potable to as saline as ocean water.49
Produced water is disposed of in injection wells and evaporation ponds.50
San Juan Basin
coalbed methane wells are historically vertical, long-lived, and almost all hydraulically fractured,
with individual well productivity ranging from as little as 1,000 to as much as 500,000 cubic feet
per day.51
Each fracturing job requires 55,000 to 300,000 gallons of water-based fluid, which can
be difficult to recover.52
Estimates for average lifetime recovery vary, with typical values
between 2 and 4 bcf per well.53
The Powder River Basin of Wyoming and Montana is a long-term target for coalbed methane
production. As with the region’s coal mines, water produced from some of the basin’s coalbed
methane wells could have been suitable for municipal consumption, so expansion of this
production in the Powder River Basin could contribute to rapid depletion of high-quality ground
water.54
Wells in the basin generally do not have to be hydraulically fractured and produce an
average of 17,000 gallons (but up to some 170,000 gallons) of high-quality water per day.55
This
water is usually discharged to surface waters, stock ponds, or reservoirs.56
Wells are small, with
average reserves of 0.4 bcf.57
III. Water Impacts at the Power Plant
Natural gas and coal are both used in thermoelectric power stations to generate electricity. Just as
the water implications of fuel extraction, processing, and transport are different for these two
fuels, so too are the water implications at the power plant.
Thermoelectric power represents a significant share of U.S. water usage. In 2005, it accounted
for 143 billion gallons (41 percent) of fresh water withdrawals per day and 58.1 billion gallons
(95 percent) of saline water withdrawals per day.58
Most of the water that is withdrawn is not
consumed: a power plant may take water from the ocean, add heat to it, and return it to the
ocean, withdrawing large amounts but consuming almost none. Even so, returning heated water
to its source can have negative environmental impacts.
In 2009, thermoelectric power plants were responsible for generating about three-quarters of
U.S. electricity.59
Steam-electric generation operates roughly the same way whether it is fueled
by coal, natural gas, biomass, nuclear, solar, or something else: the fuel is used to heat water to
steam using a circulatory system of tubes. This steam converts much of its heat energy to
mechanical energy by expanding through a turbine, which turns a generator that produces
electricity. The steam passes to a heat exchanger or ―condenser,‖ where it is cooled and
condensed to a liquid so it can be moved back easily to be reheated by fuel combustion in the
boiler.
Power plants generally use one of three main types of cooling systems to condense steam: open-
loop and closed-loop cooling (both of which use water), and dry cooling (which uses air). Some
plants use a hybrid dry-wet system to accommodate seasonal variability in water availability and
the plant’s cooling needs. The selection of a cooling technology has impacts on a plant’s water
withdrawal and consumption, construction costs, and efficiency. In general, cold-water cooling
systems allow for more efficient operation.60
10
11
Plant efficiency, or the amount of usable energy a plant creates from the chemical energy
contained in its fuel, also depends on the generating technology, fuel, elevation, age, ambient
temperature, and many other factors. No plant can convert 100 percent of its fuel’s energy into
electricity, and typical efficiencies are between 30 and 40 percent. The rest of the energy is lost
from the system as heat in flue gas or cooling water.61
Before 1970, most U.S. thermoelectric power plants used open-loop cooling, where water is
withdrawn from a lake, river, ocean, or other body of water, passed through the condenser, and
then discharged back to its source.62 (See Figure 2.) About 31 percent of current U.S. generating
capacity uses open-loop cooling, responsible for 92 percent of water withdrawals for
thermoelectric power.63
Although water withdrawals for open-loop cooling are high, the amount
of water consumed is generally minor.
However, open-loop cooling systems can damage aquatic ecosystems. Warm water is a form of
thermal pollution, as it reduces the amount of dissolved oxygen available to fish and other
species. Since the passage of the U.S. Clean Water Act in 1972, open-loop cooling systems have
become much less common in new power plants—only about 10 such systems have been built in
the United States since 1980.64
Open-loop cooling systems can also use seawater where it is
available.
Most new U.S. power plants today use closed-loop cooling, a system in which water is pumped
to a cooling tower or pond, where it is stored and cycled through the heat exchanger. Heat is
dissipated through evaporation from cooling towers, which replenish their water supply from a
nearby water source. Closed-loop cooling withdraws much less water than open-loop cooling,
but half or more of the water it uses is lost through evaporation.65
As a result, water consumption
is actually higher for closed-loop systems, although withdrawals and impacts on aquatic
ecosystems are lower.66
Using seawater in cooling towers reduces stress on fresh water
resources, but it also introduces maintenance challenges related to corrosion and mineral build-
up.67
Dry cooling systems use air instead of water to cool power plants. After the steam is collected in
a condenser, the condenser’s tubes are cooled using air that is typically blown across the
condenser with a fan. Dry cooling enables plants to operate in regions where water availability is
extremely limited for all or part of the year.68
However, air is less able to absorb heat than water,
so air cooling reduces overall plant efficiency. This means that more fuel must be consumed and
more emissions created for each unit of electricity.69
Different plant types in the existing U.S. fleet may be more likely to have one type of cooling
system or another. For example, coal-fired generation uses a greater share of open-loop or ―once-
through‖ cooling systems than does natural gas combined-cycle generation, in large part because
these facilities are more likely to predate the 1972 Clean Water Act.70
In addition to cooling systems, coal-fired power plants may use water in ―wet‖ or ―dry‖ flue gas
desulfurization (FGD) devices. These devices remove sulfur dioxide—an air pollutant that can
lead to smog and acid rain—from boiler exhaust. The use of an FGD device has been estimated
to add some 43 liters of water per megawatt-hour (MWh) for a dry system and 257 liters per
MWh for a wet system to a plant’s withdrawals, virtually all of which is consumed.71
The
incremental water consumption attributable to FGD might be equivalent to almost 50 percent of
12
all water consumption in a plant with a wet FGD process and an open-loop cooling system,
whereas it might be less than 10 percent of all water consumption in a plant with a dry FGD
process and a cooling tower.72
(FGD is not necessary in natural gas-fired power plants since
natural gas has its relatively low sulfur content stripped out at gas processing facilities.)
Combusting coal in a power plant produces solid wastes such as coal ash, the noncombustible
portion of coal. About 10 percent of the volume of coal burned becomes ash.73 The ash is
landfilled, recycled, or mixed with water and stored in impoundments, creating large reservoirs
of sometimes-toxic ash suspended in water.74 Ash must be isolated from aquifers and
precipitation to prevent leaching. Spills from impoundment dams can be damaging to surface
waters and the surrounding environment. In 2008, the breach of an impoundment dam at a
Tennessee coal plant released an estimated 5.4 million cubic yards of wet coal ash, destroying
three houses, flooding roads and rails with sludge, and contaminating drinking water with lead
and thallium.75 By contrast, natural gas combustion produces almost no ash.
The type of generating technology also affects a plant’s overall water requirements. U.S. coal-
fired power plants are almost all single-cycle steam-turbine plants that run most of the time.
Natural gas combined-cycle plants are much more efficient and may emit more than 60 percent
less CO2 per kilowatt-hour generated than the average U.S. coal plant.76 Although the United
States has installed some 142 GW of NGCC plants since 2000, these facilities have been
underutilized, due largely to the relatively high prices of natural gas and the persistent use of coal
plants for baseload generation. In 2008, NGCC plants ran at 41 percent of their capacity, while
coal steam-turbine plants ran at almost 73 percent, although this gap narrowed substantially
during 2009 and the first half of 2010.77
Coal-fired steam turbines are the most common power-plant generator technology, accounting
for some 44 percent of U.S. electricity generation in 2009.78
They dominate baseload electricity
generation in many parts of the country, and they historically have been expensive to build but
cheap to run, since coal prices recently have been lower and more predictable than natural gas
prices and the original loans on many older coal plants have now been paid in full. Pulverized
coal is combusted in a boiler, and the resulting heat is used to create steam, which powers a
steam turbine to generate electricity. The boiler operating temperature affects the plant’s
efficiency; supercritical boilers operate at higher temperatures and with consequently higher
plant efficiencies than subcritical boilers.
Some 81 percent of U.S. natural gas generation today takes place in combined-cycle plants,
which generate about 19 percent of the country’s electricity.79
In NGCC systems, a gas turbine is
used to generate electricity, and the waste heat is recovered and used to heat water in a heat
recovery steam generator. The steam is then used to power a steam turbine. Because a portion of
the gas turbine’s waste heat is captured and utilized, NGCC plants often have high thermal
efficiencies, approaching 50 percent. Water is required to condense steam from the steam
turbine, but because the plant also utilizes a gas turbine, which is air-cooled, the water used to
generate a kilowatt-hour of electricity is only about one-third of that required by a subcritical
pulverized coal plant.80
Figure 3 summarizes the estimated water needs for different generating technologies.81
13
IV. Analysis
Using basic assumptions about the heat rates of different power plant technologies makes it
possible to estimate the lifecycle water impacts associated with the electricity generated in coal
and natural gas power plants. This allows for a first-order comparison between the two plant
types from a water perspective.82 (See Figure 4 and Table 2.)
Figure 4 compares the water consumption from a range of coal and natural gas power plant
technologies equipped with wet cooling towers, the most common cooling systems for new
plants. With cooling towers, power plant cooling represents the largest point of water
consumption throughout the life cycle of a unit of electricity from coal or natural gas, regardless
of the type of plant or fuel used. On balance, the analysis suggests that, for electricity generation,
using natural gas consumes less water than using coal. In other words, the water savings from
NGCC plants relative to coal steam-turbine plants overwhelm differences in water consumption
from extraction, processing, and transportation.
The choice of cooling technology has a large impact on the overall water required to generate a
kilowatt-hour of electricity. NGCC plants have lower withdrawal and consumption rates than
their coal counterparts when cooling systems are held constant. However, many older U.S. coal
plants have once-through cooling systems. As Figure 4 illustrates, a unit of electricity generated
at one of these plants could require lower water consumption than a unit of electricity generated
0 0 490655
2,2001,970
2,900
5,300
4 15
1,900
3,1403,400
3,940 3,800
6,800
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
Wind Solar PV NGCC IGCC Nuclear Coal CSP Geothermal
Lit
ers
per M
egaw
att
-hou
rFigure 3. Water Consumed in Electricity Generation,
by Power Plant Type
Note: Figures assume that plants are equipped with wet cooling tower systems. PV = Photovoltaic; NGCC = natural gas combined-cycle; IGCC = integrated gasification combined-cycle; CSP = Concentrating Solar Power. Source: Fthenakis and Kim.
14
at an NGCC power plant. Thus, the type of cooling system employed plays an important role in
determining the overall water implications of choosing to generate electricity from coal or
natural gas.
Table 2. Estimated Water Consumption Throughout Fuel Cycle of Coal and Natural Gas
Plant/Cooling System/Fuel Estimated Water Consumption
(Liters per Megawatt-hour)
Extraction Processing Transport Generation Total
Coal steam turbine, cooling tower, PRB
11–53 0–109 Negligible 1,970–3,940 1,981–4,102
Coal steam turbine, once-
through, PRB 11–53 0–109 Negligible 450–1,210 461–1,372
Coal steam turbine, cooling tower, Appalachia
11–200 82–109 Negligible 1,970–3,940 2,063–4,249
NGCC, cooling tower,
conventional natural gas Negligible 57.5 28.8 490–1,900 576–1,986
NGCC, cooling tower, Marcellus Shale
29.4 57.5 28.8 490–1,900 606–2,016
NGCC, cooling tower, SJB
coalbed methane 0.8-2.1 57.5 28.8 490–1,900 577–1,988
Note: PRB = Powder River Basin Coal; NGCC = Natural Gas Combined Cycle; SJB = San Juan Basin.
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
Conventional
NG
Marcellus
Shale NG
San Juan Basin
CBM
PRB (Once
Through)
PRB (Cooling
Tower)
Appalachia
(Cooling
Tower)
Est
ima
ted
Ra
ng
e o
f W
ate
r C
on
sum
pti
on
(lit
ers
per m
egaw
att
-hou
r)
Figure 4. Estimated Total Lifecycle Water Consumption for Selected
Fuel–Plant Combinations
Note: Blue bars represent natural gas combined-cycle power plants; gray bars represent coal steam-turbine power plants. Assumed plant heat rates based on U.S. averages in EIA's Electric Power Annual 2008. All power plants assumed to be equipped with cooling towers unless noted otherwise. NG = Natural Gas; CBM = Coal Bed Methane; PRB = Powder River Basin. Source: See Endnote 85.
15
Many of the water impacts associated with fuel extraction, such as potential long-term
contamination or altered hydrology, do not represent water consumption and are difficult to
measure and compare. Coal mining likely has a greater potential for long-term water disturbance
than does natural gas extraction due to the large footprint of coal mines, acid mine drainage, and
other contamination from abandoned mines. However, fresh water impacts from extraction of
either fuel must be considered in a local context as well as a national context to obtain a true
sense of how energy choices affect fresh water supply and quantity. It is also important to note
that risks are not static: for natural gas in particular, near-term technological development may
mitigate many risks. Technologies for treating produced water, as well as less harmful additives
for hydraulic fracturing, are under active development. As drilling and fracturing techniques
improve, the water needs per well might decline as well.
Produced water represents the largest waste stream associated with natural gas extraction, as well
as one of the most significant potential sources of contamination. The volume and chemical
characteristics of produced water vary by geological formation. Coalbed methane wells in
particular produce a lot of water when they are first drilled but less and less over their lifetime.83
In some cases, produced water is of high-enough quality that it may be used for irrigation, such
as in the Powder River Basin. However, produced water from most conventional natural gas and
deep shale reservoirs is highly saline, can be toxic, and must be disposed of.
So far, most of the water produced by the U.S. oil and natural gas industry has been reinjected
into underground formations or evaporated. But these methods will not be viable on the scale
that shale gas drilling is anticipated to reach in the Marcellus Shale because of land constraints
for evaporation ponds and geology poorly suited to injection wells. The safe disposal or effective
treatment of produced water will be a significant challenge to natural gas development as it
moves into new regions.
Of course, the energy choices confronting the United States in the coming years are much
broader than whether the country will use coal or natural gas in steam turbines or combined-
cycle power plants. The United States is supporting extensive research and development into
―clean coal‖ technologies that will enable electricity to be generated from coal without the steep
CO2 emissions that conventional coal plants produce. One proposed alternative is to gasify coal
and use it in a combined-cycle plant. Because such Integrated Gasification Combined Cycle
(IGCC) plants are more efficient, they use about a third of the water that their steam turbine
counterparts do.84
Another potential solution to the large CO2 emissions associated with coal-fired electricity
generation is carbon capture and sequestration (CCS). However, CO2 capture technology
generally adds a significant parasitic load, reducing overall plant efficiency and indirectly
increasing the water intensity of generation through additional fuel needs. It can also directly
require additional water for cooling and other processes. A 2007 National Energy Technology
Laboratory study estimated that the use of CO2 capture technology increased water consumption
per kilowatt-hour by about 95 percent for pulverized coal plants and 37 percent for IGCC
plants.85 Although less frequently proposed, CCS can be used with NGCC plants as well, but this
could raise their water consumption by more than 80 percent.86 The significant water
requirements of carbon capture could make CCS-based ―clean coal‖ generation an unsustainable
option for water-constrained parts of the world.
16
Other power plant technologies have negligible water needs—including solar photovoltaic
panels, wind turbines, and gas turbines.87 If these technologies become more prevalent, they are
likely to further reduce the per-kilowatt-hour water needs of the U.S. power system.
V. Recommendations
Water and energy are valuable resources whose fates are closely linked. As the United States—
and the world—enter a 21st century marked by carbon, energy, and water constraints, managing
these two resources in isolation will become ever more challenging. Promoting technologies that
are less water intensive—and that have fewer negative impacts on water quality and quantity—
will become increasingly important as the demand for energy, clean air, and clean water grows.
Choosing natural gas over coal for electricity generation might be an option that simultaneously
reduces air emissions and water demand.
Energy production is only one of many competing consumers of limited fresh water supplies.
Fresh water is also needed to irrigate crops, supply households, and sustain aquatic ecosystems,
among other uses. Moreover, fresh water availability is not distributed evenly in time or space,
and even relatively small volumes of water may be locally significant. Decision makers should
consider both local impacts and their larger context. For example, using a more water-intensive
fuel might be a good choice if that fuel can be extracted in a water-rich region and reduce water
needs at a power plant in a water-scarce region.
A range of technologies can reduce water demand throughout the fuel cycle of electricity. At the
point of fuel extraction, recycling water simultaneously mitigates the need for fresh water
supplies and wastewater disposal. Both the coal and natural gas industries are exploring methods
for treating and reusing waste water. One Pennsylvania power plant is investigating the
possibility of using treated acid mine drainage water from local abandoned mines for cooling
water and boiler feedwater.88 And numerous natural gas companies are filtering their produced
water onsite for reuse in future fracturing jobs.89 Producers should work with communities and
local water authorities as well to shift their water usage to coincide with periods of relatively
high water availability.
Finally, improved efficiency can lower the cooling water requirements of any power plant
technology. One application that provides substantial efficiency gains is cogeneration, the
capture and utilization of the excess heat created during electricity generation. Often,
cogeneration supplies heat in the form of hot water or low-temperature steam, so the net water
impact depends on the application. Cogeneration may increase water use at the plant level, for
example, if waste heat in flue gas is captured in water for heat delivery, thereby displacing heat
from a natural gas furnace that was not using water. But cogeneration can save water if a hot
water discharge from a power plant is used to directly replace a separately fueled hot water
heating system. Cogeneration also mitigates the need for other energy sources to supply heat,
reducing water impacts associated with fuel extraction, processing, and transportation.
Decisions about energy resource use and policy should value both the quantity and quality of
water. Valuing water quality impacts is often subjective, however, and the short- and long-term
17
needs of communities should be considered. Water is inherently a fungible resource, and
considering its full context might improve overall resource utilization.
As energy extraction in the United States and elsewhere continues to affect water resources,
restoring these resources will require additional energy usage, which in turn will require greater
water demand for energy production. Utilities and policy makers considering natural gas as an
alternative to coal in the power sector should take into account the sizable difference in the fuels’
water footprints.
18
Endnotes 1 Joan Kenny et al., Estimated Use of Water in the United States in 2005, U.S. Geological Survey Circular 1344
(Reston, VA: October 2009). 2 U.S. Government Accountability Office, Freshwater Supply: States’ View of How Federal Agencies Could Help
Them Meet the Challenges of Expected Shortages (Washington, DC: 9 July 2003). 3 Benjamin Sovacool, ―Running on Empty: The Electricity-Water Nexus and the U.S. Electric Utility Sector,‖
Energy Law Journal, vol. 30, no.11 (2009), pp. 12–51. 4 Calculated based on U.S. Department of Energy, Energy Information Administration (EIA), ―Form 923 (2009)
– Preliminary,‖ (Washington, DC: 2010). 5 Sovacool, op. cit. note 3. 6 Worldwatch calculation based on estimates for water consumption by natural gas combined-cycle and coal
steam-turbine plants using cooling towers in V. Fthenakis and H. Kim, ―Life-cycle Uses of Water in U.S.
Electricity Generation,‖ Renewable and Sustainable Energy Reviews, vol. 14, no. 7 (2010), pp. 2039–48. 7 Christopher Flavin and Saya Kitasei, The Role of Natural Gas in a Low-Carbon Energy Economy (Washington,
DC: Worldwatch Institute, April 2010). 8 For other upstream environmental impacts associated with shale gas development, see Mark Zoback, Saya
Kitasei, and Brad Copithorne, Addressing the Environmental Risks of Shale Gas Development (Washington,
DC: Worldwatch Institute, July 2010). 9 Worldwatch calculation based on data in Table 2, see note 81. 10 EIA, ―Coal Production and Number of Mines by State, County, and Mine Type,‖
www.eia.doe.gov/cneaf/coal/page/acr/table2.html, viewed 21 July 2010. 11 Stanley Schweinfurth, ―Coal – A Complex Natural Resource,‖ United States Geological Survey (USGS),
Circular 1143 (Reston, VA: 2003). 12 EIA, ―Coal Production and Number of Mines by State and Mine Type, 2008–2007,‖
www.eia.doe.gov/cneaf/coal/page/acr/table1.html, viewed 21 July 2010. 13 DOE, Energy Demands on Water Resources: Report to Congress on the Interdependency of Energy and Water
(Washington, DC: 2006). 14 CQ Inc., Coal Cleaning Primer, at www.cq-inc.com/Coal_Primer.pdf, viewed 10 May 2010. 15 Ibid. 16 EIA, ―Coal Projections,‖ www.eia.doe.gov/oiaf/aeo/coal.html, viewed 9 November 2010. 17 I. White et al., Energy from the West: Energy Resource Development Systems Report, Volume II: Coal
(Washington, DC: U.S. Environmental Protection Agency (EPA), 2004 revision). 18 EIA, ―Table 1. Coal Production and Number of Mines by State and Mine Type, 2009, 2008,‖ and ―Table 6.
Coal Production and Number of Mines by State and Coal Rank, 2009,‖ in Annual Coal Report (2009), at
http://eia.gov/cneaf/coal/page/acr/table1.html and http://eia.gov/cneaf/coal/page/acr/table6.html. 19 A. Akcil and S. Koldas, ―Acid Mine Drainage (AMD): Causes, Treatment and Case Studies,‖ Journal of
Cleaner Production, vol. 14 (2006), pp. 1139–45. 20 EIA, op. cit. note 12. 21 EIA, op. cit. note 16. 22
Mikael Höök and Kjell Aleklett, ―Historical trends in American coal production and a possible future outlook,‖
International Journal of Coal Geology 78 (March 2009): 201-216. 23 Ibid. 24 Ibid. 25 Calculated based on 2009 dry natural gas production in EIA, ―June 2010 Natural Gas Monthly with Data for
April 2010,‖ www.eia.gov/oil_gas/natural_gas/data_publications/natural_gas_monthly/ngm.html. 26 In 2008, the United States produced about 6.6 tcf from onshore conventional reservoirs, 3.2 tcf from offshore
wells, 0.4 tcf from Alaska, 6.7 tcf from tight sands, 2.0 tcf from gas shales, and 1.8 tcf from coalbed methane.
EIA’s Annual Energy Outlook 2010 predicts that U.S. shale gas and coalbed methane production will grow to
7.4 tcf by 2030, or 32.8 percent of total natural gas production. Worldwatch estimates based on the following:
EIA, ―Coalbed Methane Production,‖ www.eia.gov/dnav/ng/ng_prod_coalbed_s1_a.htm; EIA, ―Shale Gas
Production,‖ www.eia.gov/dnav/ng/ng_prod_shalegas_s1_a.htm; EIA, Annual Energy Outlook 2009
(Washington, DC: March 2009); EIA, Annual Energy Outlook 2010 (Washington, DC: 11 May 2010). 27 Amy Mall et al., Drilling Down: Protecting Western Communities from the Health and Environmental Effects
of Oil and Gas Production (New York: Natural Resources Defense Council, September 2007), p. vi.
19
28 DOE, Energy Demands on Water Resources: Report to Congress on the Interdependency of Energy and Water
(Washington, DC: 2006). 29 Ibid. 30
U.S. Department of Energy, Energy Information Administration (EIA), ―Underground Natural Gas Storage,‖
available at http://www.eia.doe.gov/pub/oil_gas/natural_gas/analysis_publications/ngpipeline/
undrgrnd_storage.html, viewed 9 July 2010. 31 Groundwater Protection Council (GWPC) and ALL Consulting, Modern Shale Gas Development in the United
States: A Primer, prepared for NETL (Oklahoma City: April 2009). 32 E. Inlander, ―Innovative Construction and Maintenance Practices to Reduce Sedimentation from Unpaved
Roads during Gas Development for the Benefit of Yellowcheek Darter (Etheostoma moorei) and other Species
of Greatest Conservation Need in the Upper Little Red River Watershed, Arkansas,‖ 2009, available at
http://www.conservingarkansaswildlife.org/proposals/2010PreProposals/Sedimentation%2520and%2520the%2
520Yellowcheek%2520Darter%2520in%2520Fayetteville%2520Shale%2520development.pdf, viewed 15
November 2010. 33 Zoback, Kitasei, and Copithorne, op. cit. note 8. 34 GWPC and ALL Consulting, op. cit. note 31. 35 Pam Kasey, ―WVU Project Would Recycle Frack Water,‖ The State Journal, 9 September 2010; NETL, ―Pilot
Testing: Pretreatment Options to Allow Re-Use of Frac Flowback and Produced Brine for Gas Shale Resource
Development,‖ www.netl.doe.gov/technologies/oil-
gas/Petroleum/projects/Environmental/Produced_Water/00847_Pretreat.html, viewed 9 November 2010; NETL,
―An Integrated Water Treatment Technology Solution for Sustainable Water Resource Management in the
Marcellus Shale,‖ www.netl.doe.gov/technologies/oil-
gas/Petroleum/projects/Environmental/Produced_Water/00833_MarcellusWater.html, viewed 9 November 2010. 36 Zoback, Kitasei, and Copithorne, op. cit. note 8. 37 Duane Grubert, Susquehanna Financial Group, personal communication with Emily Grubert, 2010. 38 D. McMahon, ―The Real and Timely Issues Caused by the Marcellus Shale Play,‖ presentation at West Virginia
Water Conference, Morgantown, West Virginia, 6-7 October 2010. 39 Colorado School of Mines, ―Potential Gas Committee Reports Unprecedented Increase in Magnitude of U.S.
Natural Gas Resource Base,‖ press release (Golden, CO: 18 June 2009). 40 Ibid. 41 Chesapeake Energy, ―Water Use in Barnett Deep Shale Gas Exploration‖ (March 2010). 42 GWPC and ALL Consulting, op. cit. note 31. 43 J. Arthur et al., Evaluating the Environmental Implications of Hydraulic Fracturing in Shale Gas Reservoirs
(2008); Chesapeake Energy, op. cit. note 41. 44 GWPC and ALL Consulting, op. cit. note 31. 45 Ibid. 46 K.M. Currie and E.B. Stelle, ―Pennsylvania’s Natural Gas Boom,‖ Commonwealth Foundation Policy Brief,
vol. 22, no. 5 (2010). 47 Arthur et al., op. cit. note 43; Matthew Mantell, Deep Shale Natural Gas: Abundant, Affordable, and
Surprisingly Water Efficient, presentation at Water/Energy Sustainability Symposium, 2009 Groundwater
Protection Council Annual Forum, Salt Lake City, UT, 2009. 48 ALL Consulting and Montana Board of Oil and Gas Conservation, Coal Bed Methane Primer (February 2004). 49 Ibid; U.S. Department of Energy, Energy Information Administration (EIA), ―The San Juan Basin,‖ Evaluation
of Impacts to Underground Sources of Drinking Water by Hydraulic fracturing of Coalbed Methane Reservoirs
(2004), A1-1 to A1-33. 50 EIA, op. cit. note 49. 51 Ibid. 52 Ibid. 53 R. Williams, ―Hydrogen Production from Coal and Coal Bed Methane,‖ in Baldur Eliasson, Pierce Riemer, and
Alexander Wokaun, eds., Greenhouse Gas Control Technologies (Kidlington, United Kingdom: 1999); Vello
Kuuskraa, Advanced Resources International, personal communication with Saya Kitasei, Worldwatch Institute,
October 2010. 54 U.S. Department of Energy, Energy Information Administration (EIA), ―The Powder River Basin,‖ Evaluation
of Impacts to Underground Sources of Drinking Water by Hydraulic Fracturing of Coalbed Methane Reservoirs
(2004), A5-1 to A5-20.
20
55 Ibid. 56 Ibid. 57 ALL Consulting and Montana Board of Oil and Gas Conservation, op. cit. note 48. 58
Kenny et al., op. cit. note 2. 59 Worldwatch calculation based on Coal, Nuclear, Petroleum, and Natural Gas steam turbines and the steam
portion of combined cycle gas turbines, in EIA, ―Form-923 (2009) - Preliminary‖ op. cit. note 4. 60 NETL, Estimating Freshwater Needs to Meet Future Thermoelectric Generation Requirements: 2008 Update
(Washington, DC: September 2008), p. 11. 61 Ashlynn Stillwell et al., Energy-Water Nexus in Texas (Austin, TX: University of Texas-Austin and
Environmental Defense Fund, April 2009). 62 Figure 2 based on Stillwell et al., op. cit. note 61. 63 Kenny et al., op. cit. note 2. 64 DOE, Energy Demands on Water Resources: Report to Congress on the Interdependency of Energy and Water
(Washington, DC: 2006), p. 18. 65 NETL, op. cit. note 60, p. 14. 66 P. Torcellini, N. Long, and R. Judkoff, ―Consumptive Water Use for U.S. Power Production‖ (Golden, CO:
National Renewable Energy Laboratory, December 2003), p. 10. 67 John S. Maulbetsch and Michael N. DiFilippo, Performance, Cost, and Environmental Effects of Saltwater
Cooling Towers, prepared for California Energy Commission (Berkeley, CA: January 2010). 68 For example, NGCC plants using dry cooling systems might withdraw and consume 0.004 gallons/kWh,
compared to a consumption rate of 0.15 gallons/kWh and a withdrawal rate of 0.13 gallons/kWh for a NGCC
plant using a wet cooling tower closed-loop system. Thomas J. Feeley III et al., ―Water: A Critical Resource in
the Thermoelectric Power Industry,‖ Energy, January 2008, p. 4. 69 Stillwell et al., op. cit. note 61, p. 8. 70 For more about how the Clean Water Act impacts cooling systems, see NETL, Estimating Freshwater Needs to
Meet Future Thermoelectric Generation Requirements: 2010 Update (Washington, DC: September 2010). 71 Feeley et al., op. cit. note 68, p. 4. 72 Ibid. 73 Texas Comptroller of Public Accounts, ―Coal,‖ Window on State Government,
www.window.state.tx.us/specialrpt/energy/exec/coal.html, viewed 9 November 2010. 74 EPA, ―Fact Sheet: Coal Combustion Residues (CCR) – Surface Impoundments with High Hazard Potential
Ratings‖ (Washington, DC: June 2009). 75 Shaila Dewan, ―Tennessee Ash Flood Larger Than Initial Estimate,‖ New York Times, 26 December 2008. 76 NETL, Life Cycle Analysis: Power Studies Compilation Report (Washington, DC: 7 October 2010). 77 Worldwatch calculation based on EIA, ―Form 923 (2008) - Final‖ (Washington, DC: 2008) and ―Form 860
(2008) - Final‖ (Washington, DC: 2009). Preliminary generation data from 2009 and early 2010 had been
released at the time of publication, but final figures permitting a publishable estimate of capacity factors were not yet available.
78 Worldwatch calculation of generation based on EIA, ―Form 923 (2009) - Preliminary,‖ op. cit. note 4. 79 Worldwatch calculation based on EIA, ―Form 923 (2009) - Preliminary,‖ op. cit. note 4. 80 Feeley et al., op. cit. note 68, p. 4. 81 Figure 3 from Fthenakis and Kim, op. cit. note 6. 82 Figure 4 and Table 2 from the following sources: Feeley et al., op. cit. note 67; Fthenakis and Kim, op. cit. note
6; DOE, op. cit. note 64; GWPC and ALL Consulting, op. cit. note 31; Kuuskraa, op. cit. note 53; Mike
Hightower, Sandia National Laboratories, personal communication with Emily Grubert, September 2010; T.F.
Edgar, Coal Processing and Pollution Control, Gulf Publishing (Houston, TX:1983), Worldwatch and
University of Texas-Austin estimates. 83 John Veil et al., A White Paper Describing Produced Water from Production of Crude Oil, Natural Gas, and
Coal Bed Methane, prepared for NETL (Argonne National Laboratory: Argonne, IL, January 2004). 84 Worldwatch calculations based on Feeley et al., op. cit. note 68, pp. 1–11. 85 NETL, Cost and Performance Baseline for Fossil Energy Plants (Washington, DC: August 2007), pp. 6–7. 86 Ibid. 87 Fthenakis and Kim, op. cit. note 6. 88 Robert Zick, ―Mine Drainage: An Alternative Source of Water,‖ POWER Magazine, 1 September 2010. 89
GWPC and ALL Consulting, op. cit. note 31, p. 68.