Human health and environmental considerations

This is an excerpt from a study by the Tyndall Centre for Climate Change Research, Shale gas: a provisional assessment of climate change and environmental impacts.
A report by researchers at The Tyndall Centre University of Manchester.
www.tyndall.ac.uk

Human health and environmental considerations


Importance of cumulative impacts


Perhaps unsurprisingly, the processes and operations involved in the extraction of
shale gas from wells are not without their human health and environmental
implications. For example, as is discussed in more detail later, the human health
and environmental risks associated with hydraulic fracturing in particular have risen
in prominence in the US. Here there have been a number of incidents and reports of
contamination from shale gas developments and, on 3 March 2010, the US EPA
announced that it will conduct a comprehensive research study to investigate the
potential adverse impact that hydraulic fracturing may have on water quality and
public health.


However, whilst the new risks associated with hydraulic fracturing of wells may be
the subject of debate, such risks and impacts are not the only potential drawback of
shale exploration, particularly when considering relatively highly populated countries
such as the UK.


Here, whilst there is the temptation to focus on the risks associated with individual
processes involved in shale gas production and reported incidents, it is also
important to consider the impact of shale gas as a whole.


More ‘run of the mill’ impacts including vehicle movements, landscape, noise or
water consumption, may be of significant concern, particularly in more populated
countries where there is greater competition for resources, such as the UK.


Cumulative impacts may be a particular issue too, when one considers the
development of shale gas at a scale sufficient to deliver gas at meaningful volumes.




Key risks and impacts


The key risks and impacts of shale gas and shale gas processes and development
can be divided as follows:


contamination of groundwater by fracturing fluids/mobilised contaminants arising
from:
o wellbore/casing failure; and/or
o subsurface migration;


pollution of land and surface water (and potentially groundwater via surface
route) arising from:
o spillage of fracturing additives; and
o spillage/tank rupture/storm water overflow from liquid waste storage,
lagoons/pits containing cuttings/drilling mud or flowback water;


water consumption/abstraction;


waste water treatment;
land and landscape impacts;


impacts arising during construction:
o noise/light pollution during well drilling/completion;
o flaring/venting; and
o local traffic impacts.


Pollution impacts


Introduction


Pollution impacts from shale gas development are closely connected with the
hydraulic fracturing process, the fracturing fluid chemicals used, transformation
products and subsurface contaminants that are mobilised during the process.


At present, there is little information available on fracturing additives and risks
associated with hydraulic fracturing. US Federal law currently exempts the
underground injection of fluids for hydraulic fracturing purposes from regulation
(Congressional Research Service, 2009) and a significant number of formulations
have been justified as trade secrets as defined and provided by Public Officers Law
(New York State, 2009).


Owing to recent expansion of the shale gas industry and increasing concerns raised
by the US public, media and Congress, the US EPA announced in March 2010 that it
will conduct a comprehensive research study to investigate the potential adverse
impact that hydraulic fracturing may have on water quality and public health. US
EPA notes that “there are concerns that hydraulic fracturing may impact ground
water and surface water quality in ways that threaten human health and the
environment” and is re-allocating $1.9 million for the study in the financial year 2010
and requesting funding for 2011 in the president’s budget proposal.


US EPA is still in the early stages of the hydraulic fracturing research program and
initial results will only be available towards the end of 2012. Whilst it, and other
assessments, are being completed some regulators are moving towards moratoria
on hydraulic fracturing. In New York State, for example, on 3 August 2010 the State
Senate passed a Bill to suspend hydraulic fracturing for the extraction of natural gas
or oil until 15 May 2011 (and to suspend the issuance of new permits for such
drilling). On 11 December 2010, the New York State Governor vetoed the Bill and
issued an Executive Order directing the Department of Environmental Conservation
(DEC) to “conduct further comprehensive review and analysis of high-volume
hydraulic fracturing in the Marcellus Shale”. The Executive Order requires that highvolume, horizontal hydraulic fracturing would not be permitted until 1 July 2011 at the earliest..


The issue of hydraulic fracturing and environmental and human health risks is, then,
under the spotlight in the US. In the meantime, however, there is a paucity of
information and data on which to base a quantified assessment of environmental and
human health risk.


That said, this short study seeks to draw together what information is available and
provide an overview of key issues, concerns and challenges from a UK perspective,
in particular.


Fracturing fluids and flowback water


Multi-stage fracturing operation involves injecting fracturing fluids at very high pressure into the wellbore to generate fractures in the target rock formation.


Fracturing of a single well requires a considerable volume of water and, with chemical additives of up to 2% by volume, around 180-580 m3 of chemical additives (or 180-580 tonnes based on relative density of one). After fracturing, a proportion of the fluid returns as flowback water.


Chemical composition of fracturing fluids


Additive Purpose


Proppant “Props” open fractures and allows gas / fluids to flow more freely to the well
bore.


Acid Cleans up perforation intervals of cement and drilling mud prior to fracturing
fluid injection, and provides accessible path to formation.


Breaker Reduces the viscosity of the fluid in order to release proppant into fractures
and enhance the recovery of the fracturing fluid.


Bactericide / Biocide
Inhibits growth of organisms that could produce gases (particularly hydrogen
sulfide) that could contaminate methane gas. Also prevents the growth of
bacteria which can reduce the ability of the fluid to carry proppant into the
fractures.


Clay Stabilizer / Control
Prevents swelling and migration of formation clays which could block pore
spaces thereby reducing permeability.


Corrosion Inhibitor
Reduces rust formation on steel tubing, well casings, tools, and tanks (used
only in fracturing fluids that contain acid)


Crosslinker
The fluid viscosity is increased using phosphate esters combined with metals.
The metals are referred to as crosslinking agents. The increased fracturing
fluid viscosity allows the fluid to carry more proppant into the fractures.


Friction Reducer
Allows fracture fluids to be injected at optimum rates and pressures by
minimising friction.


Gelling Agent
Increases fracturing fluid viscosity, allowing the fluid to carry more proppant
into the fractures.


Iron Control
Prevents the precipitation of metal oxides which could plug off the formation.


Scale Inhibitor
Prevents the precipitation of carbonates and sulfates (calcium carbonate,
calcium sulfate, barium sulfate) which could plug off the formation.


Surfactant
Reduces fracturing fluid surface tension thereby aiding fluid recovery.




The composition of the fracturing fluid varies from one product to another and the
design of the fluid varies depending on the characteristics of the target formation and
operational objectives. Fracturing fluid used in modern slickwater fracturing is
typically comprised of around 98% water and sand (as a proppant) with chemical
additives comprising 2%.


Owing to the fact that US Federal law currently exempts the underground injection of
fluids for hydraulic fracturing purposes from regulation, there is no information on the
identity and concentration of substances in hydraulic fracturing formulations.
Disclosure of the identity of chemicals used in hydraulic fracturing may be required
on a case by case basis and, in New York State, for example, the Department of
Environmental Conservation requires operators to disclose chemicals as part of the
permitting procedure. However, the New York State (2009) also notes that full
disclosure of chemicals and composition of formulations is not possible owing to
trade secrets exemptions from public disclosure. In this way, and as is identified in
comments on New York State (2009) by New York City, “involved stakeholders such
as City and local health departments do not have any knowledge of the chemicals
that are released into the environment near water supplies”.


In terms of disclosure to the wider public, operators are required to produce Material
Safety Data Sheets (MSDSs) of chemicals stored in quantities of >10,000pounds
(4.5t) under the US Emergency Planning and Community Right to Know Act of 1986
(EPCRA). However, this is unlikely to provide full coverage of chemical composition
nor does it provide data on concentration of substances.


Owing to the lack of detailed information on chemical composition, this assessment
must rely on information extracted from the MSDSs submitted by operators to
regulators. Here New York State (2009) provides a list of 260 chemical constituents
and their CAS numbers that have been extracted from chemical compositional
information for 197 products as well as Material Safety Data Sheets submitted to the
NYSDEC.


A review of this list has been undertaken by cross checking CAS numbers in the
NYS list with the following lists on the European chemical Substances Information
System (ESIS):


toxicity classification: for the purposes of classification and labelling (according
to Annex VI of Regulation (EC) No 1272/2008 and the Globally Harmonised
System);


presence on List 1-4 of priority substances: since 1994, the European
Commission has published four lists of substances requiring immediate attention
because of their potential effects to man or the environment. There are 141
substances on the lists;
presence on the first list of 33 priority substances: established under Annex
X of the Water Framework Directive (WFD) 2000/60/EC - now Annex II to the
Directive on Priority Substances (Directive 2008/105/EC). Member States must
progressively reduce pollution from priority substances; and


presence on the PBT list: substances which have been subject to evaluation of
their PBT properties under the Interim Strategy for REACH and the ESR
program. For substances which are persistent, bioaccumulative and toxic (PBT)
or very persistent and very bioaccumulative (vPvB) a "safe" concentration in the
environment cannot be established with sufficient reliability.
This analysis suggests that 58 of the 260 substances have one or more properties
that may give rise to concern and:


15 substances are listed in one of the four priority lists;


6 are present in list 1 (Acrylamide, Benzene, Ethyl Benzene, Isopropylbenzene
(cumene), Naphthalene, Tetrasodium Ethylenediaminetetraacetate);


one is currently under investigation as a PBT (Naphthalene bis (1-methylethyl));


2 are present on the first list of 33 priority substances (Naphthalene and
Benzene);


17 are classified as being toxic to aquatic organisms (acute and/or chronic);


38 are classified as being acute toxins (human health);


8 are classified as known carcinogens (Carc. 1A=1, Carc. 1B = 7);


6 are classified as suspected carcinogens(Carc. 2 = 6);


7 are classified as mutagenic (Muta. 1B); and


5 are classified as having reproductive effects (Repr. 1B=2, Repr. 2=3).


It is clear that the presence of a number of the substances in fracturing fluids may
present cause for concern, particularly given the intended use and the volumes
being used. The level of risk associated with the use of these substances will be
related to the quantity and concentration of substances, their fate, and routes of
exposure of people and the environment, the latter of which is considered in
subsequent sections.


All first fracturing operations (i.e. without re-fracturing) on a single six well pad
require a total of around 1,000-3,500m3 of chemicals. Based on 1.25-3.5pads/km2,
3,780-12,180m3 (or 3,780-12,180tonnes based on relative density of one) of
fracturing chemicals would be required per km2 of shale development.


Based on the data, around 140-400km2 of shale development
comprising 2,500-3,000 horizontal wells would be required to deliver 9bcm/year
(10% of UK gas consumption in 2008). This, in turn, represents high pressure
injection of around 0.5-2.2million m3 (or tonnes based on relative density of one) of
fracturing chemicals.


Flowback water


Some 15-80% of injected fluid returns to the surface as flowback (and, by
implication, 20-85% remains underground). Whilst flowback fluids include the
fracturing fluids pumped into the well, it also contains:


chemical transformation products that may have formed due to reactions
between fracturing additives;


substances mobilised from within the shale formation during the fracturing
operation; and


naturally occurring radioactive materials (NORMs).


The nature and concentrations of different substances will clearly vary from one
shale formation to another and, for the UK, it is difficult to predict what the
composition of flowback fluid is likely to be. In terms of example compositions, New
York State (2009) provides limited sample data on composition of flowback fluids This analysis was based on limited data from Pennsylvania and West Virginia. The analytical methods and detection levels used were not uniform across all parameters and it is noted that the composition of flowback from a single well can also change within a few days of the well being fractured.


When visually compared with substances in fracturing fluids the data on flowback
water would tend to suggest mobilisation and presence of elevated concentrations
of:


heavy metals (of varying types);


radioactivity and NORMs;


total dissolved solids; and


perhaps, hydrocarbons including benzenes (unclear whether this represents
mobilised hydrocarbons or fracturing additives).


Altogether, the toxicity profile of the flowback fluid is likely to be of greater concern
than that of the fracturing fluid itself, and is likely to be considered as hazardous
waste in the UK. Volumes of waste generated and associated requirements for
storage and industrial waste water treatment are also large. Table 4.1 provides
ranges based on recovery of 15-80% of fracturing fluid as flowback (accounting also
for the range in values of volumes of fracturing fluid used. This suggests that, for
shale development delivering 9bcm/year, 5-89million m3 of hazardous waste water
would be recovered and would require treatment or storage. Importantly, the same
water use and percentage recovery ranges would also imply that, if 15-80% of fluid is
recovered, then between 20-85% of fluid is not recovered and, therefore, remains underground.




Groundwater contamination


Significance of groundwater pollution


Groundwater is water that collects in rock formations known as aquifers. Water
naturally fills the aquifer from the bottom upwards, occupying rock spaces with water
and creating what is known as the saturated zone of the aquifer, towards the bottom,
and in the upper sections (where rock spaces contain air and water) an unsaturated
zone.


The boundary between saturated and unsaturated zones is the 'water table'.
Groundwater is not stationary but flows through and along rock crevices from the
area where water enters the aquifer (recharge zone) to an area where water leaves
the aquifer (discharge zone). Where this is near the surface, springs occur and
support the flow of rivers and grounded wetlands such as fens and marshlands.


Groundwater quality is generally high and requires little or no treatment before use
as drinking water. In England and Wales groundwater provides a third of drinking
water on average and also maintains the flow of many rivers. In parts Southern
England, groundwater supplies up to 80% of needs (Environment Agency, 2010).
Owing to its importance as both a source of drinking water and as source for rivers
and wetlands, preventing its pollution is vital. If it becomes contaminated and
pollution runs deep it can lead to long-term deterioration.


The fracturing and ‘flowback’ fluids (including transformation products and mobilised
subsurface contaminants) contain a number of hazardous substances that, should
they contaminate groundwater, are likely to result in potentially severe impacts on
drinking water quality and/or surface waters/wetland habitats. The severity will
depend on, for example, the significance of the aquifer for abstraction; the extent and
nature of contamination; the concentration of hazardous substances; and connection
between ground and surface waters.


Routes of Exposure
The most obvious routes for exposure of groundwaters to contamination from shale
wells are:


catastrophic failure or full/partial loss of integrity of the wellbore (during
construction, hydraulic fracturing, production or after decommissioning); and


migration of contaminants from the target fracture formation through subsurface
pathways including:
o the outside of the wellbore itself;
o other wellbores (such as incomplete, poorly constructed, or
older/poorly plugged wellbores);
o fractures created during the hydraulic fracturing process; or
o natural cracks, fissures and interconnected pore spaces.




Wellbore failure/loss of integrity


Owing to the relatively significant depth of shale resources, wellbores are likely to be
drilled through several aquifers. At all stages in the lifetime of a well, the wellbore
therefore provides a continuous physical link between the target formation (where
high pressure hydraulic fracturing and subsequent extraction occurs), other rock
formations/saline aquifers, freshwater aquifers and the surface. Owing to this, the
wellbore itself probably provides the single most likely route of pollution of
groundwater.


To reduce the likelihood of contamination via the well itself, casings are installed to
isolate the well from the surrounding formations.


Notably, just as depth requirements vary from state to state, so do requirements for
cementing in of casings. As noted in Section 2.2, a method known as ‘circulation’
may be used to fill the entire space between the casing and the wellbore (the
annulus) from the bottom of the surface casing to the surface. However, according
to the GWPC:


circulation of cement on surface casing is not a universal requirement and in
some states cementing of the annular space is required across only the deepest
ground water zone but not all ground water zones;


although some states require complete circulation of cement from the bottom to
the top of the production casing, most states require only an amount of cement
calculated to raise the cement top behind the casing to a certain level above the
producing formation; and


in very deep wells (as is often the case for horizontally drilled shale wells), the
circulation of cement is more difficult to accomplish as cementing must be
handled in multiple stages which can result in a poor cement job or damage to
the casing if not done properly.


Clearly, once installed, wellbore casings provide the primary line of defence against
contamination of groundwater. As such, the loss or initial lack of integrity of the well
casing arrangement (at any point along the wellbore) has the potential to result in
contamination of rock formations including aquifers.


Anything from the catastrophic failure of a well casing (for example during high
pressure fracturing) through to partial loss of integrity of poor cement seals is likely to
result in a pollution event. The severity of such events will depend on the nature of
the loss of integrity, the contaminants and the receiving environment.
In terms of events linked to loss of casing integrity, contamination resulting from the
flowback of fracture fluids through the casing itself could occur but would require
physical failure of both steel casing and cement. More likely is upward flow via the
cemented annulus between the casing and the formation which, in GWPC’s view,
presents the greatest risk of groundwater contamination during hydraulic fracturing.






“It is the cementation of the casing that adds the most value to the process of ground water protection...consequently, the quality of the initial cement job is the most critical factor in the prevention of fluid movement from deeper zones into ground water resources”.


New York State (2009) ignores the role and significance of cementing (and,
particularly, the initial cementing work) when considering groundwater pollution. It
largely dismisses the issue by referring to a study it commissioned from ICF
International, which used an upper bound estimate of risk from a 1980s study by the
American Petroleum Institute (API). The API study analysed the risk of
contamination from properly constructed Class II injection wells to an Underground
Source of Drinking Water (USDW) due to corrosion of the casing and failure of the
casing cement seal. Using this, the ICF study (and New York State, 2009) identified
that the “probability of fracture fluids reaching a USDW due to failures in the casing
or casing cement is estimated at less than 2 x 10-8 (fewer than 1 in 50million wells)”.
On this basis the ICF study concludes that “hydraulic fracturing does not present a
reasonably foreseeable risk of significant adverse environmental impacts to potential
freshwater aquifers”.


Examination of this suggests that both the estimate and the conclusion may be
problematic on a number of counts. Most notable is that a thorough analysis of
process risk requires consideration of all (reasonably conceivable) circumstances,
events and failure nodes that could potentially result in adverse impacts. As such,
focussing only on an estimate of the risk of failure of properly constructed wells fails
to account for the risk of failure of improperly constructed wells. Whilst improper
construction of wells may be unintended, it does occur and has resulted in pollution
events (see later). As the study of risk requires the study of unintended
consequences, this is a serious omission particularly as poor construction is known
to represent the most significant risk to groundwater.


Another issue is the comparison between injection wells and hydraulically fractured
shale wells. Whilst the ICF study notes the difference between the two, it implies
that risk from shale wells is likely to be lower because injection wells work under
sustained pressure and hydraulically fractured shale wells are pressurised only
during hydraulic fracturing (after which pressure within the casing is less than the
surrounding formation). Whilst the operational differences are true, at 5,000-
10,000psi (345-690bar) the pressures applied in hydraulic fracturing are both higher
and are applied several times during fracturing of a well. This means that the well
and casings are put under repeated episodes of high pressure followed by total
pressure release, and negative pressure relative to surrounding rocks. Thus, it could
equally be argued that the stress on well casings and cement seals from repeated
‘inflation and deflation’ may be significantly higher, and damage and subsequent loss
of casing integrity is more likely for hydraulically fractured shale wells.


Given these issues, it would appear problematic to conclude that there is no
reasonably foreseeable risk to potential freshwater aquifers, particularly since the
probability of contamination of aquifers given is the probability per well. As
thousands of shale wells in the US are drilled through aquifers the figure presented
as the probability of contamination of a USDW should have been presented as a
factor of thousands higher than the one provided.
Interestingly, New York State (2009) identifies that natural gas migration “is a more
reasonably anticipated concern with respect to potential significant adverse impacts”
owing to:


inadequate depth and integrity of surface casing to isolate potable fresh water
supplies from deeper gas-bearing formations;


inadequate cement in the annular space around the surface casing, which may
be caused by gas channelling or insufficient cement setting time; and


excessive pressure in the annulus between the surface casing and intermediate
or production casing.


Such pressure could break down the formation at the shoe of the surface casing and result in the potential creation of subsurface pathways outside the surface casing. Excessive pressure could occur if gas infiltrates the annulus because of insufficient production casing cement and the annulus is not vented in accordance with required casing and cementing practices.


Thus, on the one hand, the assessment of hydraulic fracturing in New York State
(2009) dismisses the possibility of contamination owing to poor construction but, on
the other, the possibility of the same poor construction is identified as “a more
reasonably anticipated concern”.


The omission is highlighted by the fact that there are a number of documented
examples of pollution events owing to poor construction and operator error. There
are reports of incidents involving contamination of ground and surface waters with
contaminants such as brine, unidentified chemicals, natural gas, sulphates, and
hydrocarbons such as benzene and toluene. In many cases the exact cause or
pathway of the contamination is yet to be identified owing to the difficulty in mapping
complex subsurface features (Hazen and Sawyer, 2009) but there are also several
where causes such as poor construction have been identified. These include the
following:


1) in 2004 in Garfield County Colorado natural gas was observed bubbling into a
stream bed. In addition to natural gas, groundwater samples revealed that
concentrations of benzene exceeded 200micograms/litre and surface water
concentrations exceeded 90micrograms/litre (also 90 times the state water
quality limit). The operator had ignored indications of potential problems while
drilling, failed to notify the regulators as required by the drilling permit, and failed
to adequately cement the well casing. This, in conjunction with the existence of
a network of faults and fractures led to significant quantities of formation fluids
migrating nearly 4,000ft (1,200m) and horizontally 2,000ft (600m), surfacing as a
seep. Although remedial casings installed in the well reportedly reduced
seepage, the resulting benzene plume has required remediation since 2004.
Subsequent hydrogeology studies found that ambient groundwater
concentrations of methane and other contaminants increased regionally as gas
drilling activity progressed, and attributed the increase to inadequate casing or
grouting in gas wells and naturally occurring fractures.


2) in 2007, a well that had been drilled almost 4,000ft 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;


3) groundwater contamination from drilling in the Marcellus shale formation was
reported in 2009 in Dimock, Pennsylvania, where methane migrated thousands of
feet from the production formation, contaminating the freshwater aquifer and
resulting in at least one explosion at the surface. Migrating methane has
reportedly affected over a dozen water supply wells within an area of 9miles2
(23km2). The explosion was due to methane collecting in a water well vault.
Pennsylvania Department of Environmental Protection (DEP) has since installed
gas detectors and taken water wells with high methane levels offline at impacted
homes to reduce explosion hazards. The root cause remains under investigation
and a definitive subsurface pathway is not known;


4) in July 2009 in McNett Township, the Pennsylvania DEP discovered a natural gas
leak involving a drilled well. Two water bodies were affected by the release of
methane gas which also impacted numerous private drinking water wells in the
area and one resident was forced to evacuate. A subsequent PA DEP report
identified that the “suspected cause of the leak is a casing failure of some sort.”
The investigation is ongoing (Riverkeeper);


5) in April 2009 in Foster Township, PA, drilling activities impacted at least seven
drinking water supplies. Stray gas became evident in numerous wells and
residents complained. Two of the affected water supplies contained methane and
five had iron and manganese above established drinking water standards. After
investigating, the PA DEP found that “the stray gas occurrence is a result of 26
recently drilled wells, four of which had excessive pressure at the surface casing
seat and others that had no cement returns” (Riverkeeper);


6) on December 12, 2006, PA DEP issued a cease and desist order to two
companies which had “continued and numerous violations” of Pennsylvania law
and had “shown a lack of ability or intention to comply with the provisions of the
commonwealth’s environmental laws.” Among the violations cited in the order
were “over-pressured wells that cause gas migration and contaminate
groundwater; failure to implement erosion and sedimentation controls at well sites
which has caused accelerated erosion; unpermitted discharges of brine onto the
ground; and encroachments into floodways and streams without permits”
(Riverkeeper);


7) in Fremont County, WY, in response to complaints of foul odours and taste in
residential wells, EPA Region eight funded an investigation into the source and
nature of the contamination. The report considered data collected from residential
and municipal wells in Pavillion, Wyoming in March and May 2009. The report
found heightened levels of hazardous contaminants in a number of drinking water
wells, including the same chemicals used in a nearby hydraulic fracturing
operation (Riverkeeper); and


8) on 3 June 2010 a gas well blowout in Clearfield County sprayed natural gas and
wastewater into the air for 16hours. The blowout reached as high as 75ft,
according to press accounts, before an emergency response team flown in from
Texas was able to cap the well. The blowout was blamed on untrained personnel
and improper control procedures, and the well operators were fined $400,000
and ordered to suspend all well operations in the state for 40days.


In addition to the evidence that contamination of groundwater via this route can (and
does) occur, the fact that voluntary action on the use of some toxic substances in
fracturing fluid has been taken on the basis of ‘unnecessary risks’ implies that there
is a risk of potential concern. Here GWPC report that diesel was cited as a
principal constituent of concern by the Oil and Gas Accountability Project (OGAP)
because of its relatively high benzene content. An agreement was reached to
discontinue its use as a fracture fluid media in coalbed methane (CBM) projects in
zones that qualify as USDWs. This action, then, also conflicts with the general
conclusion that “hydraulic fracturing does not present a reasonably foreseeable risk
of significant adverse environmental impacts to potential freshwater aquifers”.


Sub-surface migration of contaminants


The exposure routes outlined above may combine with other routes, for example, via
man-made or natural fractures, to produce contamination of ground or surface
waters.
The GWPC provide data on depths of formations and treatable water and identify that, outside New Albany and the Antrim, wells are expected to be drilled at depths greater than 3,000ft (900m) below the land surface.
On the basis of this some commentators seek to dismiss the potential for water contamination on the basis that target formations frequently lie at significant depths below aquifers and contaminants must migrate through the intervening rock.


Here, for example, reports such as New York State (2009) identify that the objective
of hydraulic fracturing is to limit fractures to the target formation as excessive vertical
fracturing is undesirable from a cost standpoint. The expense associated with
unnecessary use of time and materials is cited, as well as added costs of handling
produced water and/or loss of economic hydrocarbon (should adjacent rock
formations contain water that flows into the reservoir formation). Whilst this may be
true, it does not negate the possibility of fractures extending vertically beyond the
target formation and thereby creating or enhancing the pathways between previously
isolated formations. For example, New York State (2009) cites an ICF report that
identifies that, despite ongoing laboratory and field experimentation, the mechanisms
that limit vertical fracture growth are not completely understood.


Incidents such as those highlighted above serve to demonstrate that a combination
of exposure routes including the following can, and do, act together to result in
contamination of groundwaters via:


the outside of the wellbore itself;


other wellbores (such as incomplete, poorly constructed, or older/poorly plugged
wellbores);


fractures created during the hydraulic fracturing process; or


natural cracks, fissures and interconnected pore spaces.


Routes of exposure – surface water and land contamination


Routes of exposure of land and surface waters, and via both to groundwater, are
more straightforward.
The operations conducted at individual well pads requires the transport of materials
to the site; use of those substances; generation of wastes; storage of wastes; and
subsequent transport of wastes generated. For an individual well pad these can be
summarised as follows:


well cuttings/drilling mud: a single well drilled vertically to a depth of 2km and
laterally by 1.2km generates around 140m3 of cuttings. A six well pad will
generate around 830m3 of cuttings. These are typically stored in pits before
transport offsite;


transport and temporary storage hydraulic fracturing additives: based on 2%
content of fracturing fluid and water volumes provided previously, around 180-
580m3 of chemical additives (or 180-580tonnes based on relative density of one)
are required for each well. At the level of a well pad some 1,000-3,500m3 of
chemicals (or 1,000-3,500tonnes based on relative density of one).
The exact composition of such fracturing fluids is not disclosed but
analysis of chemical identities suggests a significant number of substances with
hazardous properties and priority substance status in the EU;


flowback fluid: each well on a multi-well pad will generate between 1,300–
23,000m3 of flowback waste fluid containing water, fracturing chemicals and
subsurface contaminants mobilised during the process (including toxic organic
compounds, heavy metals and naturally occurring radioactive materials or
NORMs). According to New York State (2009) approximately 60% of the total
flowback occurs in the first four days after fracturing and this may be collected
via:
(a) unchecked flow through a valve into a lined pit;
(b) flow through a choke into a lined pit; and/or
(c) flow to tanks.


The dimensions and capacity of on-site pits and storage tanks are likely to vary
but, based on volumes calculated above, total capacity would have to be in
excess of the expected volumes of flowback water from a single well fracturing
operation, namely between 1,30–23,000m3.
New York State (2009) notes that one operator reports a typical pit volume of
750,000gallons (2,900m3). Based on a pit depth of 3m, the surface footprint of a
pit would be around 1000m2 (0.1ha). It also notes that, owing to the high rate
and potentially high volume of flowback water, additional temporary storage tanks
may need to be staged onsite even if an onsite lined pit is to be used. Based on
the typical pit capacity above, this implies up to around 20,000m3 of additional
storage capacity for flowback water from one fracturing operation on a single
well.


In terms of overall flowback water volume for a six well pad the data suggest a
total of 7,900-138,000m3 of flowback water per pad for a single fracturing
operation (with fracturing chemicals and subsurface contaminants making up to
2%, or 160-2,700m3).


The key operational hazards in these processes at an individual well pad site include
(but are not limited to) the following:


spillage, overflow, water ingress or leaching from cutting/mud pits owing:
o limited storage capacity;
o operator error;
o storm water or flood water ingress; or
o poor construction or failure of pit liner;


spillage of concentrated fracturing fluids during transfer and final mixing operation
(with water) that occurs onsite owing to:
o pipework failure;
o operator error;


spillage of flowback fluid during transfer to storage owing to:
o pipework or frac tree failure during the operation;
o insufficient storage capability and overflow;
o operator error;


loss of containment of stored flowback fluid owing to:
o tank rupture;
o overfilling of lagoons due to operator error or limited storage capacity;
o water ingress from storm water or floods;
o poor construction or failure of liner;


spillage of flowback fluid during transfer from storage to tankers for transport
owing to:
o pipework failure; or
o operator error.


In addition to the many onsite hazards listed above, the pooling and subsequent
treatment and discharge of hazardous waste water generated by well pads, and the
possible need for additional industrial wastewater treatment works, contributes to an
increase in the risk of contamination through this route. The likelihood of each of
these adverse events occurring varies from one hazard to another as do the
consequences. Given the toxic properties of fracturing/flowback fluids (or
constituents), however, any spillage onto land or surface water is likely to be of
concern.


Many of these hazards and routes of exposure are well known from other industrial
processes and action can be taken to reduce the likelihood of such events occurring.
Usually such risks persist in dedicated industrial facilities with significant investment
having been built into the design to reduce the impacts should incidents occur. In
contrast, the activities and hazards at well pads identified above are part of the
construction of the pad and, hence, occur over a short time relative to the lifetime of
the pad itself. Investment in permanent physical containment to the standard of
other hazardous installations is unlikely.


Given that the development of shale gas requires the construction of multiple
wells/well pads, the probability of an adverse event leading to contamination
increases accordingly. As such, the likelihood of pollution incidents associated with
wider development of shale increase from the ‘possible’ end of the spectrum at the
level of a well pad through to the ‘probable’ as the number of wells and pads
increases. As might be expected, there have been a number of incidents reported in
the US including (Riverkeeper, 2010):


in September 2009 in Dimock, PA. two liquid gel spills occurred at a natural gas
well pad polluting a wetland and causing a fish kill. Both involved a lubricant gel
used in the high-volume hydraulic fracturing process and totalled over
30,000litres. The releases were caused by failed pipe connections;


in Monongalia County, West Virginia in September 2009 a substantial fish kill
along the West Virginia-Pennsylvania border was reported to the West Virginia
Department of Environmental Protection. Over 30 stream miles were impacted
by a discharge, originating from West Virginia. The DEP had received numerous
complaints from residents who suspected that companies were illegally dumping
oil and gas drilling waste into the waterway;


in Dimock, PA, there have been two reports of diesel fuel leaking from tanks at
high-volume hydraulic fracturing drilling operations. The first leak was caused by
a loose fitting on a tank and resulted in approximately 3,000 litres of diesel
entering a wetland. The second leak resulted in approximately 400 litres of diesel
causing in soil contamination; and


on December 12, 2006, PA DEP issued a cease and desist order to two
companies owing to continued and numerous violations. Among the violations
cited in the order were unpermitted discharges of brine onto the ground.


A number of such incidents relate to failure to implement or conform to regulatory
controls and the provision of sufficient regulatory oversight to so many individual
sites and processes is both difficult and costly.
The lack of sufficient regulatory control has been an issue of concern in the US and
on 27 January 2010, the US EPA announced the opening of the ‘Eyes on Drilling’
Tipline for citizens to report non-emergency suspicious activity related to oil and
natural gas development.


Water consumption


Each stage in a multi-stage hydraulic fracturing operation requires around 1,100-2,200m3 of water so that the entire multi-stage fracturing operation for a single well requires around 9,000-29,000m3 (9- 29megalitres). For all fracturing operations carried out on a six well pad, a total of between 54,000-174,000m3 (54-174megalitres) of water would be required for a first hydraulic fracturing procedure.


As such, large quantities of water must be brought to and stored on site. Local
conditions will dictate the source of water and operators may abstract water directly
from surface or ground water sources or it may be delivered by tanker truck or
pipeline. However, as has been noted elsewhere, well pads themselves are spaced
out in an array over the target formation, with around 3-4/square kilometre. As each
fracturing phase of the operation lasts around 2-5days/well, the provision of
dedicated pipelines to each well pad would appear unlikely in the UK situation and
transport via truck or abstraction is the most likely means of providing source water.


For provision of 9bcm/year shale gas for 20 years, it is estimated that total water
consumption is 27,000-113,000megalitres. Averaged over the 20 year period, this is
equivalent to an annual water demand of 1,300-5,600megalitres. Annual abstraction
by industry (excluding electricity generation) in England and Wales is some
905,000megalitres/year. As such, development of shale reserves at levels sufficient
to deliver gas at a level equivalent to 10% of UK gas consumption would increase
industrial water abstraction across England and Wales by up to 0.6%.


Clearly, this comparison relates to total abstraction across the whole of England and
Wales and shale development will be focussed in a much smaller area. Sourcing
such significant quantities of water sustainably from local sources will be difficult
owing to existing pressure on UK water resources. By way of example, the (as yet
exploratory) drilling being undertaken by Cuadrilla resources at Preese Hall in Fylde,
UK, is within the River Wyre catchment (and, incidentally, just on the boundary of the
flood zone). The catchment covers some 578km2 and the Environment Agency’s
Catchment Abstraction Management Strategy (CAMS) for the Wyre identifies that all
zones are classified as either ‘over licensed’, ‘over abstracted’ or ‘no water
available’.


Other impacts of and constraints on shale development


Overview


In addition to the very real issues surrounding shale gas development, chemical
pollution and abstraction, there are a number of other impacts that, from a UK
perspective, are likely to be significant. These impacts include:


noise pollution;


landscape impacts; and


traffic and road damage.


Of all of the impacts, these are likely to present the greatest constraint on
development of shale gas in the UK, whether at a local level or over a significant
area.


Noise and Visual/Aesthetic Impacts


In terms of noise impacts, Table 2.4 provides a summary of activities required at well
pads prior to production. On the basis of this, it is estimated that each well pad
requires a total of around 500-1,500days of noisy surface activity. Of all of these
activities, drilling of wells is likely to provide the greatest single continuous noise
(and, light) pollution as drilling is required 24 hours a day. Here, New York State
(2009) estimates that each horizontal well takes four to five weeks of 24hours/day
drilling to complete.


The UK operator Composite Energy estimates 60 days of 24 hour drilling. On the basis of this, each well pad will require 8-12 months of drilling day and night. This would be significant even if it were only a single pad that was being developed, but with 1.25-3.5 pads/km2 the noise impacts on a locality are likely to be considerable and prolonged.


Landscape Impacts


In terms of visual impacts, each well pad will be around 1.5-2ha in size and will be
equipped with access roads (New York State, 2009). During construction well pads
will comprise storage pits, tanks, drilling equipment, trucks, etc. making the
installations difficult to develop in a way that is sympathetic with surrounding
landscapes.


Given that 430-500 well pads would be required to deliver 9bcm/year of shale gas, it
is likely that in a UK context visual impacts will be contentious. As there is little that
can be done to alleviate the levels of visual intrusion (individually or collectively),
these impacts, along with noise and construction, may provide the greatest
constraints on development in the UK.


Traffic


In addition to impacts onsite, construction of well pads requires a significant volume
of truck traffic. Table 2.5 provides truck movements per well pad (based on a six
well pad) from New York State (2009). This suggests a total number of truck visits
4,300-6,600 for the construction of a single well pad. Local traffic impacts for 1.25-
3.5 pads/km2 are, clearly, likely to be significant, particularly in a densely populated
nation such as the UK.


In the US traffic damage to roads has been an issue. For example, it is reported that
West Virginia Department of Transportation has increased the bonds that industrial
gas drillers must pay from $6,000 to $100,000/mile. Pennsylvania is considering a
similar rule where the increased funds are needed to repair roads not designed for
the intense truck traffic associated with industrial gas drilling.