Complete Abstract (PDF -
513KB)
IMPACTS OF GAS
DRILLING ON
HUMAN AND ANIMAL HEALTH
MICHELLE BAMBERGER
ROBERT E. OSWALD
ABSTRACT
INDEX
Environmental
concerns surrounding drilling for
gas are intense due to expansion of
shale gas drilling operations.
Controversy surrounding the impact
of drilling on air and water quality
has pitted industry and
leaseholders against individuals
and groups concerned with
environmental protection and public
health. Because animals often are
exposed continually to air, soil,
and groundwater and have more
frequent reproductive cycles,
animals can be used as sentinels to
monitor impacts to human health.
This study involved interviews with
animal owners who live near gas
drilling operations. The findings
illustrate which aspects of the
drilling process may lead to health
problems and suggest modifications
that would lessen but not eliminate
impacts. Complete evidence regarding
health impacts of gas drilling
cannot be obtained due to incomplete
testing and disclosure of chemicals,
and nondisclosure agreements.
Without rigorous scientific studies,
the gas drilling boom sweeping the
world will remain an uncontrolled
health experiment on an enormous
scale.
At what point does
preliminary evidence of harm become definitive evidence of harm? When someone
says, “We were not aware of the
dangers of these chemicals back then,”
whom do they mean by we?
Sandra Steingraber,
Living Downstream
(Da Capo Press,
2010)
Communities living
near hydrocarbon gas drilling
operations have become de facto laboratories
for the study of environmental
toxicology. The close proximity of these
operations to small communities has
created a variety of potential hazards to
humans, companion animals, livestock
and wildlife. These hazards have become
amplified over the last 20 years,
due in part to the large-scale
development of shale gas drilling
(horizontal drilling with high-volume hydraulic
fracturing), encouraged by the
support of increased drilling and
exploration by U.S. government
agencies [1]. Yet this large-scale industrialization of
populated areas is moving forward
without benefit of carefully controlled
studies of its impact on public
health. As part of an effort to obtain public health
data, we believe that particular
attention must be paid to companion animals,
livestock, and wildlife, as they may
serve as sentinels for human exposures, with
shorter lifetimes and more
opportunity for data collection from necropsies.
All phases of
hydrocarbon gas production involve
complex mixtures of chemical substances.
For example, in hydraulic fracturing
fluids, chemical substances other than
water make up approximately 0.5 to 1
percent of the total volume; however, the
very large volumes used require
correspondingly large volumes of a variety
of compounds. These substances range
from the relatively benign to the highly
toxic. Some of these are reported to
the public and others are not, but the
quantities and proportions used are
largely considered trade secrets. In addition
to these added chemicals, naturally
occurring toxicants such as heavy metals,
volatile organics, and radioactive
compounds are mobilized during gas extraction
and return to the surface with the
gas/chemical mix (wastewater); of the 5.5
million gallons of water, on
average, used to hydraulically fracture a shale gas
well one time [2], less than 30
percent to more than 70 percent may remain
underground [3]. Hydraulic
fracturing takes place over 2 to 5
days and may be repeated
multiple times on the same well over
the course of the potential 25- to
40-year lifetime of a well [4]. Many
of these chemicals are toxic and have known
adverse health effects, which may be
apparent only in the long term. A discussion of
these compounds and their health
effects is beyond the scope of this
article; however, Colborn et al. [5]
have analyzed this topic in depth.
The large-scale use
of chemicals with significant
toxicity has given rise to a great deal of
public concern, and an important
aspect of the debate concerns the level of proof
required to associate an
environmental change with activities associated with gas
drilling. Environmental groups
typically invoke the precautionary principle
[6]. That is, if an action is
suspected of causing harm to the environment, then in
the absence of a scientific
consensus, the burden of proof falls on the
individual or organization taking
the action. The oil and gas industry has typically
rejected this analysis and has
approached the issue in a manner similar to the
tobacco industry that for many years
rejected the link between smoking and cancer.
That is, if one cannot prove beyond
a shadow of doubt that an environmental
impact is due to drilling, then a
link is rejected. This approach by the tobacco
companies had a devastating and
long-lasting effect on public health from which we
have still not recovered [7], and we
believe that a similar approach to the
impacts of gas drilling may have
equally negative consequences.
Although reports of
petroleum hydrocarbon exposure in
humans [8-14], primates [15], and
several other species, including
ruminants [16-26], horses [27], wildlife [28],
and a dog [29], have been cited in
the literature, there are few reports on
exposure of animals to gas
operations, and to our knowledge, no case reports on
exposure of humans to hydrocarbon
gas operations [30]. Adler et al. [31] observed
aspiration pneumonia in sheep
following exposure to gas condensate. In
another study, Waldner et al. [32]
found no association between the productivity of
cattle and exposure to a sour gas
pipeline leak; while in a longer-term study
[33] in cattle, the same group
reported associations between sour-gas flaring and
increased risk of stillbirth across
three of the four years studied, as well as
increased risk of calf mortality in
one of the years studied.
In a study of habitat
selection, Sawyer et al. [34] found
that mule deer tended to move away from
areas of gas development, and in a
recent report [35] from the same author, the
deer population dropped by 45
percent in one year, and the survival rate
decreased.
Just as epidemiologic
studies linked smoking to human
health impacts, such studies could be used
to assess the health impacts of gas
drilling operations on human beings. Studies
in laboratory animals have also been
a powerful tool for linking
components of tobacco smoke to
cancer, not only because controlled studies can be done
but also because breeding cycles are
short and the age at which cancer develops
is within a range accessible to
laboratory studies. Though such controlled
animal studies of the effects of gas
drilling are not feasible, animals can
nevertheless serve as sentinels for
human health impacts. Animals,
particularly livestock, remain in a
confined area and, in some cases,
are continually exposed to
an environmental threat. Further,
effects on reproduction can be more readily
assessed in a herd of cattle than in
a human population, simply due to the
higher rates of reproduction.
For the past year, we
have been documenting cases of
animal and owner health problems with
potential links to gas drilling.
Many cases are currently in litigation. To
protect individuals' privacy and due
to ongoing legal action, the discussion will not
include personal identifying
information. We summarize the results of our
investigation, provide several case
studies, and conclude with recommendations for
minimizing or preventing similar
problems in the future.
This study is not an
epidemiologic analysis of the health
effects of gas drilling, which could proceed
to some extent without knowledge of
the details of the complex mixtures of
toxicants involved. It is also not a
study of the health impacts of specific
chemical exposures related to gas
drilling, since the necessary information cannot be
obtained due to the lack of testing,
lack of full disclosure of the International
Union of Pure and Applied Chemistry
(IUPAC) names and Chemical
Abstracts Service (CAS) numbers of
the chemicals used, and the industry's use of
nondisclosure agreements.
Nevertheless, the value of this
study is twofold. First,
clear health risks are present in
gas drilling operations. These cannot be eliminated
but can be decreased by commonsense
reforms. Second, our study illustrates not
only several possible links between
gas drilling and negative health effects, but
also the difficulties associated
with conducting careful studies of such a link.
Again, simple commonsense policy
reforms could facilitate the collection of data
that would lead to a careful
assessment of the health consequences of gas
drilling on both humans and animals.
SUMMARY OF THE
EFFECTS OF GAS DRILLING ON PRODUCTION AND
COMPANION ANIMALS AND ANIMAL
OWNERS
To describe how
exposures may occur, and to report
health effects, we conducted interviews
with animal owners in six states
(Colorado, Louisiana, New York, Ohio,
Pennsylvania, Texas) affected by gas
drilling. In all but one case, we spoke directly
with animal owners. The exception
was a case that had previously been
documented by the state
environmental regulatory agency
[36].
When possible, we
interviewed the owners'
veterinarians. Where available, we have obtained the
results of water, soil, and air
testing as well as the results of laboratory tests
on affected animals and their
owners. Documentation was obtained from the
animal owners, the veterinarians
(with permission of the owners), drilling
company representatives, state
regulatory agencies, and a Freedom of
Information Act (FOIA) request from
the Pennsylvania Department of Agriculture. Cases
were identified by requesting
referrals from environmental groups and
individuals actively involved in
influencing shale gas policy and studying its effects.
For each case, a standard series of
questions was asked, including the exact
location of each owner's property;
details on wells in the area (subsequently
verified by crosschecking with state
records and, using software developed for this
project, mapping the wells relative
to the owner's property); details of seismic
testing and well flaring; location
of wastewater impoundments; results of water,
soil, and air testing; details of
animal husbandry and medical records preceding,
during and following drilling,
depending upon the individual case; a list of
animals (species, breed, age, sex,
use (e.g., livestock)), sorted into those healthy and
those unhealthy; health history for
all animals; observations of wildlife in the
area; and health histories of the
humans living in the household.
As each case is
different, the standard form was
used as a starting point, with additional
information invariably supplied by
individuals being interviewed.
More than one-third
of the cases involved conventional
wells (shallow or deep vertical wells), with
the remainder comprising horizontal
wells subjected to high-volume hydraulic
fracturing. Because of the scale of
the horizontal well drilling operations,
such wells were more commonly
associated with animal health problems.
However, conventional wells have
also had problems associated with
faulty well casings and failure of
blowout preventers; in our study, wastewater
dumping and leakage, failure of a
blowout preventer, and affected well water
involving conventional gas wells
were associated with both animal and human
health problems.
By the standards of a
controlled experiment, this is an
imperfect study, as one variable could
not be changed while holding all
others constant. It also is not a systematic
study that will provide the
percentage of farms with problems associated with gas
drilling, but the design is such
that the study can illustrate what can happen in
areas experiencing extensive gas
drilling. It is also possible to observe temporal
correlations between events such as
well flaring and air quality, or hydraulic
fracturing and water quality leading
to toxicity. In two cases, spatial differences
(cows in a single herd, with some
allowed access to a creek or pond and others
not allowed access) could be used to
compare outcomes.
Table 1 summarizes
the types of wells involved and the
sources of exposure, and Table 2 describes
the details of each individual case.
In some cases, exposure was due to accidents
or negligence, but at other times,
it was a consequence of normal operations.
Direct exposure to hydraulic
fracturing fluid occurred in two cases: in one, a
worker shut down a chemical blender
during the fracturing process, allowing the
release of fracturing fluids into an
adjacent cow pasture, killing 17 cows in
one hour; the other was a result of
a defective valve on a fracturing fluid
tank, which caused hundreds of
barrels of hydraulic fracturing fluid to leak into a
pasture where goats were exposed and
suffered from reproductive problems
over the following two years.
Exposure to drilling chemicals occurred
during a blowout when liquids ran
into a pasture and pond where bred cows were
grazing; most of the cows later
produced stillborn calves with congenital
defects. Exposure to wastewater
occurred through leakage or improper fencing of
impoundments, alleged compromise of
a liner in an impoundment to drain
fluid, direct application of the
wastewater to roads, and dumping of the
wastewater on creeks and land. The
most common exposure by far was to affected
water wells and/or springs; the next
most common exposure was to affected ponds
or creeks. Finally, exposures also
were associated with compressor station
malfunction, pipeline leaks, and
well flaring. In addition to humans, the animals
affected were: cows, horses, goats,
llamas, chickens, dogs, cats, and koi. Other
than photographing and recording the
presence of dead and dying wildlife (deer,
songbirds, fish, sala manders, and
frogs) in the vicinity of affected pastures,
creeks and ponds, the effect on
wildlife has not been well documented.
Because production
animals were exposed to the
environment for longer periods and in
greater numbers than companion
animals, and because most of the farms we documented
raised beef cattle, cows were
represented to a greater extent than other animals.
Exposures through well water, ponds,
springs, dumping of wastewater into
creeks, and spills or leakage of
wastewater from impoundments were believed by
farmers to result in deaths over
time periods typically ranging from one to three
days, with cows going down and
unable to rise despite symptomatic
treatment. The most commonly
reported symptoms were associated with reproduction.
Cattle that have been exposed to
wastewater (flowback and/or produced water) or
affected well or pond water may have
trouble breeding. When bred cows were
likewise exposed, farmers reported
an increased incidence of stillborn calves with
and without congenital abnormalities
(cleft palate, white and blue eyes). In each
case, farmers reported that in
previous years stillborn calves were rare (fewer than
one per year). In most cases where
diagnostics were pursued, no final
diagnosis was made; in other cases,
acute liver or kidney failure was most commonly
found. Of the seven cattle farms
studied in the most detail, 50 percent of the
herd, on average, was affected by
death and failure of survivors to breed. In one
case, exposure to drilling
wastewater led to a quarantine of
beef cattle and
significant uncompensated economic
loss to the farmers.
The most dramatic
case was the death of 17 cows within
one hour from direct exposure to hydraulic
fracturing fluid. The final necropsy
report listed the most likely cause of
death as respiratory failure with
circulatory collapse. The hydraulic fracturing
fluid contained, among other
toxicants, petroleum hydrocarbons and
quaternary ammonium compounds
(tetramethylammonium and
hexamethylenetetramine). Although
petroleum hydrocarbons were reported
to be found in the small
intestine, lesions in the lung,
trachea, liver and kidneys suggested exposure to
other toxicants as well, and
quaternary ammonium compounds have been
described as producing similar
lesions [37].
Two cases involving
beef cattle farms inadvertently
provided control and experimental groups.
In one case, a creek into which
wastewater was allegedly dumped was the source
of water for 60 head, with the
remaining 36 head in the herd kept in other
pastures without access to the
creek. Of the 60 head that were exposed to the
creek water, 21 died and 16 failed
to produce calves the following spring. Of
the 36 that were not exposed, no
health problems were observed, and only
one cow failed to breed. At another
farm, 140 head were exposed when the
liner of a wastewater impoundment
was allegedly slit, as reported by the
farmer, and the fluid drained into
the pasture and the pond used as a source of water
for the cows. Of those 140 head
exposed to the wastewater, approximately 70 died
and there was a high incidence of
stillborn and stunted calves. The remainder
of the herd (60 head) was held in
another pasture and did not have access
to the wastewater; they showed no
health or growth problems. These cases
approach the design of a controlled experiment, and strongly implicate wastewater
exposure in the death, failure to
breed, and reduced growth rate of cattle.
Companion animals
were defined as those animals that
were kept as pets, and included horses,
dogs, cats, llamas, goats, and koi.
Companion animal exposures typically
occurred when animals ingested
affected water from a well, spring, creek or
pond. Reproductive problems
(irregular cycles, failure to breed, abortions, and
stillbirths) and neurological
problems (seizures, incoordination, ataxia) were the most
commonly reported. Other commonly
reported symptoms included those of
gastrointestinal (vomiting,
diarrhea) and dermatological (hair and feather
loss, rashes) origin.
In the majority of
cases, owners of animals were
exposed upon using their well or spring water
for drinking, cooking, showering and
bathing. Upper respiratory symptoms
(including burning of the nose and
throat) and burning of the eyes were the
most commonly reported. Headaches
and symptoms associated with the
gastrointestinal (vomiting,
diarrhea), dermatological (rashes), and vascular
(nosebleeds) systems were commonly
reported.
CASES ILLUSTRATING
THE EFFECTS OF GAS DRILLING ON
PRODUCTION AND COMPANION ANIMALS AND THEIR OWNERS
Case 1
Two homes (A and B)
are located within two miles of
approximately 25 shale gas wells. The
closest pad, drilling muds pit, and
wastewater impoundment are within one mile of
both homes; the impoundment is
approximately 4.5 acres in area and is at a
higher elevation than either home.
Two compressor stations are located within one
mile of both homes. The owners have
a variety of companion and farm animals, and
reported no unusual pet morbidity or
mortality preceding drilling operations.
Predrilling tests on water sources
were not done for either home. Soon after
drilling began, the owner of Home B
noted that her well water had an odor and black
sediment, and the owners of Home A
observed a decreased quantity of their
water sources (a well and a spring).
Once the wastewater impoundment was
constructed, the owners of Home A
noted a dramatic decrease in quantity, as well
as poor quality, of both the well
and spring water. The spring served as the sole
source of water for the owners' farm
animals. Approximately nine months after
drilling began, the owners of Home A
began hauling water from a nearby creek,
to supplement the spring water.
Since drilling
operations began, both owners have
observed wastewater being spread on the roads
during all weather conditions, and
noted that cats and dogs in their neighborhood
licked their paws after walking on
the road, and also drank from wastewater
puddles; some of these animals
became severely ill and died over a
period of one to three days
following these exposures. According to the owner of Home
B, the wastewater impoundment was
not initially fenced and animals had
direct access to the wastewater. An
accident involving the wastewater
impoundment was noted by both
owners; after filling, a truck carrying wastewater
drove away from the impoundment site
with an open valve, releasing
approximately 20 gallons of
wastewater onto the impoundment
access road and onto the
road near the property of Home A.
Most recently, both the drilling company and
the state environmental regulatory
agency were notified of a spill from the
wastewater impoundment that flowed
past temporary barriers and into a creek; based
on soil erosion patterns, the owners
of Homes A and B reported that this
spill had been ongoing for months.
Soon after this accident, a malfunction occurred
in the wastewater impoundment
aeration system, producing a raw
sewage smell that persisted in the
air around Homes A and B for days and sickened the
families in both homes. When the
owner of Home A complained, the
drilling company offered to pay
motel expenses for her and her family; this offer
was declined because the owner
refused to leave her animals.
Approximately a year
after drilling began, an 18-year-old
intact female American Quarter
Horse in Home A had an acute onset
of anorexia, malaise, rapid weight loss,
and mild incoordination after
testing normal on a physical examination a few
weeks earlier. The horse was treated
symptomatically with an antibiotic,
steroid, and antihistamine. A few
days later, the horse had become ataxic, and was
treated for equine protozoal
myeloencephalitis, although no diagnosis was made.
The horse did not improve after
three to four days and was treated again. Within
a few days, the horse's neurological
symptoms had progressed such that the
horse was unable to rise. Blood and
clinical chemistry parameters indicated
acute liver failure due to toxicity.
The veterinarian suspected heavy metal
poisoning as a cause of the horse’s
sudden illness; this was not confirmed, as
toxicology tests were not done. The
horse was euthanized two weeks after onset due
to poor prognosis and failure to
respond. Similar neurologic signs were reported
in another case in this study that
involved two horses living adjacent to a deep,
vertical gas well operation.
In addition, both
homeowners were caring for animals
that were bred at this time: the owner of
Home B had a three-year-old intact
female Boer goat that aborted two kids in
the second trimester, and the owners
of Home A had a five-year-old intact
female Boxer that experienced
dystocia with a fourth litter (after previously
whelping three normal litters),
producing one stillborn pup and one pup with cleft
palate that died soon after birth.
This same dog subsequently whelped a fifth
litter of 15 pups in which seven
pups were stillborn and eight pups died within 24 hours.
All the pups were afflicted with
congenital hypotrichosis; that is, they were
born with the complete or partial
absence of normal hair.
Soon after drilling
and hydraulic fracturing began for
the first well, a child living in Home B
began showing signs of fatigue,
severe abdominal pain, sore throat, and backache.
Six months later, the child was
hospitalized with confusion and delirium and was
given morphine for abdominal pain.
After the deaths of several animals as
cited above, the child's physician
suspected that the child’s symptoms were
of toxicological origin. A
toxicology test revealed arsenic poisoning as the
cause of the child’s sickness. The
family stopped using their well water despite
test results indicating that the
water was safe to drink, and the child gradually
recovered after losing one year of
school.
During high-volume
hydraulic fracturing, substances
that occur naturally in the shale, including
arsenic, come to the surface in
wastewater. In this case, the wastewater was stored
in the impoundment, where aerators
misted the chemicals into the air,
increasing the chances of inhalation
by animals and people; also, surface spillage of
wastewater, as noted above, could
have contaminated the ground water. Tests
on well water from both Homes A and
B, and the spring from Home A, did not show
elevated levels of arsenic; however,
it is possible that, given fluctuations in
the water table and water quality,
high levels of arsenic may have initiated
symptoms in the child in Home B and
then dropped to low levels before water
testing was done more than one year
later. Also, reported arsenic levels may be
deceptively low because arsenic can
be converted to arsine-a toxic gas
that dissipates rapidly [38]. In
people, both acute and chronic oral exposure to
inorganic arsenic causes
gastrointestinal effects as well as effects on the
nervous system: short-term effects
include headaches, weakness, and delirium, while
long-term effects include peripheral
neuropathy [39].
Acute exposure of
people to arsine can produce many
effects including abdominal pain and
headaches [39]. Animals exposed
acutely to inorganic arsenic may show many
symptoms including staggering gait,
extreme lethargy, and intense abdominal
pain, while animals exposed over a
longer period of time may manifest
signs including anorexia,
depression, and partial paralysis of the rear limbs
[40]. Animal studies show that
arsenic can also cause fetal malformations and
fetal death [41].
As the family in Home
B continued to be screened for
toxicants, random urine tests on all family
members were positive for phenol, a
metabolite of benzene, with dramatic
increases over a period of a few
months. Based on occupational health
studies [e.g., 42], the testing
laboratory judged these results to
be consistent with chronic
exposure to 0.5 to 4.0 ppm benzene
in the air. The most recent symptoms
observed by families in both homes
include extreme fatigue, headaches,
nosebleeds, rashes, and sensory
deficits (smell and hearing). The child in Home B also
had difficulty breathing, and again
had to be taken out of school. Doctors of
the families in both homes warned
them to leave their homes for at least 30 days
or suffer more severe health
consequences. The owner of Home B followed her
doctor's advice, and moved her
children out of her home, returning each day to
care for her animals; the owners of
Home A elected to remain at their home
to care for their animals. After one
month of being away, the phenol levels as
well as the symptoms of the children
in Home B decreased, while the owner of
Home B, who returns to the home for
a few hours each day, has increased phenol
levels and worsening of symptoms.
One of the owners in Home A, who works at
home, has experienced worsening of
symptoms.
This case illustrates
the importance of considering both
animal and human health. Animals live
among us and are exposed to the same
environmental influences; however,
they tend to suffer more direct
exposure and have shorter life and reproductive
cycles. If it were not for the
numerous deaths of animals soon after shale gas
operations began in this
neighborhood, the child’s doctor might not have
ordered toxicology tests, as arsenic
poisoning is not a common diagnosis.
Case 2
In this case, a beef
cattle farmer had a herd of 96
cattle (Angus Limousine cross) that was
divided among three pastures. The
farm is located in an area of intensive gas
drilling, with two active shallow
vertical gas wells on the farmer’s property and
approximately 190 active gas wells
within five miles of the property; of
these, approximately 11 are shale
gas wells and approximately 26 are deep vertical
gas wells. In one pasture, 60 cows
(a mixed herd, mostly 5- to 10-year-old bred
cows) had access to a creek as a
source of water. In a second pasture, 20 cows
(bred yearlings) obtained water from
hillside runoff, and in a third pasture, 14
feeder calves (8 to 14 months old)
and two bulls had access to a pond. Over a
three-month period, 21 head from the
creek-side pasture died (17 adult bred cows and 4
calves). All the cattle were healthy
before this episode.
Despite symptomatic
treatment, deaths occurred 1 to 3
days after the cows went down and were unable
to rise. Basic diagnostics were
done, but no cause of death was determined.
On rendering, 16 of the 17 adults
were found to have dead fetuses, nearly
doubling this farmer’s losses. Of
the 39 cows on the creek-side pasture that
survived, 16 failed to breed and
several cows produced stillborn calves with white and
blue eyes. The health of the cattle
on the other two pastures was unaffected; on
the second pasture, only one cow
failed to breed. Historically, the health of the
herd was good, the farmer reporting
average losses of 1-2 cows a year in his herd of
nearly 100 cattle.
This is an
interesting case because it has a
natural control group. That is, the cattle that were kept
along the creek suffered severe
problems while the cattle in pastures at a higher
elevation and away from the creek
experienced no morbidity or mortality. As
discussed below, the contamination
of the creek may have been caused by illegal
dumping of wastewater. Fortunately,
these cows were not taken to slaughter, as they
died on the farm. However, they
still may have entered our food chain as well as
that of our pets: rendering plants
produce feed for many non-ruminants
including chickens, pigs, cats, dogs
and horses, so it is possible that chickens, raised
for egg production or meat, and pigs
were fed the flesh from these cattle.
Case 3
This case concerns
farmers that have raised beef cattle
(Herford Simmental cross) for the past
21 years. Before drilling operations
began the farmers lost one or two animals out of
a closed herd of 33 (yearlings,
heifers, mature cows, two bulls) every few
years to illness or accident. There
is one active shale gas well on the farmers’ 530-acre
property, and approximately six
active shale gas wells within two miles of
their property. A private well
provides water for the family’s use; the water for
the herd comes from a creek that
originates from springs above and below the well
pad, and spillover from a pond below
the well pad. The gas wellhead is 300 feet
from the farmers' house and 250 feet
from their water well.
The well pad is 75
feet from their barn at higher
elevation, and slopes directly down to the door. A
one-acre impoundment, used to
collect wastewater from the high-volume hydraulic
fracturing operations, and a
1/3-acre drilling muds pit, used to collect the
chemicals and fluids brought to the
surface during drilling operations, were both
within 350 feet of the farmers'
water well, and within 200 feet of the creek
and the pond where the cattle drink.
Soon after hydraulic
fracturing operations concluded, the
farmers noticed that on the far bank of
the wastewater impoundment, two dark
spots could be seen adjacent to a 20-acre
cow pasture. According to the
farmers, these two spots were a concern as they
grew in size from day to day;
approximately one month after first observing these
spots, the farmers found ankle-deep
water in one-third of an acre of the
pasture with the wet area extending
another one-quarter of an acre into the
pasture; the pasture grass in these
areas appeared to be burned.
Fearing their herd
drank the wastewater, they
voluntarily quarantined their farm and notified the
state environmental regulatory
agency.
According to the
farmers, drilling company workers
informed them that the liners of both the
wastewater impoundment and the
drilling muds pit had two-foot tears, and that the
tear in the liner of the wastewater
impoundment had caused the leak into the cow
pasture. Except for the two bulls,
the entire herd was exposed to the
wastewater leakage.
Four notices of
violations were issued to the
drilling company by the state environmental
regulatory agency: failure to notify
the agency, improperly lined impoundment (pressure
testing of liner revealed a failed
patch), pollution of a spring and farm pond
due to leakage of the impoundment,
and mismanagement of residual waste
(wastewater leaked from the
impoundment onto the ground and surfaced in an
adjacent pasture).
Testing of the
wastewater in the impoundment
indicated the presence of calcium, iron,
magnesium, manganese, potassium,
sodium, strontium, fluoride, chloride, sulfate,
and bromide; there was no reported
testing for any organic compounds. Strontium
was of most concern: it can be toxic
to both animals and people because it
replaces calcium in bone, especially
in the young, and because it may take years to
be eliminated from the body [43].
The state environmental regulatory
agency placed a quarantine on the
herd such that mature cows would be held from
slaughter for six months, yearlings
would be held for nine months, calves
exposed in utero would be held for
eight months, and growing calves would be held
for two years. Six of the exposed
cows eventually went on to slaughter,
and, according to the farmers, there
was no testing before or after slaughter.
Pre-drilling tests
were not done on any of the cattle’s
sources of water; post-drilling tests
were done and revealed no
significant findings. Soil tests
done on the cow pasture
contaminated by the leaked
wastewater revealed high levels of chloride, sulfate,
sodium, and strontium when compared
to background samples. The liners
from both the wastewater impoundment
and drilling-muds pit were removed, the
affected soil removed, and areas
remediated; sulfate concentrations
remained at high levels in the cow
pasture despite remediation.
During the spring of
the first calving season following
the leakage of wastewater into their
cow pasture, the farmers lost two
calves: one calf was aborted late-term,
and the other calf lived for
approximately seven days before dying [44]; both
calves were exposed in utero to the
wastewater. In the second calving season
post-drilling, the farmers lost 11
out of 17 calves: seven were stillborn, three died
a few months after birth and one was
born alive but severely ill; the dams of all
the calves had previously been
exposed to the wastewater.
The severely ill calf
and a stillborn calf were sent for
necropsy: the ill calf was diagnosed with E.
coli septicemia, and the stillborn
calf was diagnosed with goiter (diffuse
thyroid hyperplasia); both calves
were also diagnosed with low liver vitamin E and
selenium.
This case illustrates
several important points. First, the
testing was not complete. According
to the farmers, they were not
informed of the chemicals used during either
drilling or hydraulic fracturing
operations. Testing of the water
well and cattle’s sources
of water excluded organic compounds
except for a pasture spring; the
wastewater analysis also excluded
organic compounds. No toxicology tests were done on
live cattle, and the tests at
necropsy omitted volatile organic compounds, endocrine
disruptors, and many minerals
present in the wastewater.
The cattle’s sources
of water were tested only after the
farmers lost many calves.
Soil tests were not
done in the area affected by the
leakage of the drilling-muds pit. Second, the
cattle were exposed to sulfate in
the wastewater for at least one month and to
elevated sulfate in the grass and
soil [45, 46] for over a year.
Studies show that
increasing dietary sulfur decreases
the bioavailability of selenium [47-50], and
that Vitamin E and selenium
deficiency is associated with reproductive failure
in cattle [51, 52]. Third, the liner
tear and subsequent leakage of drilling fluids
onto the farmers' land were not
considered a potential problem and not officially
recorded as a violation by the state
environmental regulatory agency. Due to gas
drilling operations on their
property, the farmers now have 26 head of cattle
instead of 33, and have lost 40 to
50 acres of hayfields.
These farmers
received no compensation from the
drilling company for the loss of their animals,
damage to their land, or the
treatment of the animal health problems they have
encountered since gas drilling
began.
DISCUSSION
The most striking
finding of our investigations was
the difficulty in obtaining definitive
information on the link between
hydrocarbon gas drilling and health effects.
However, the results point to a
number of ways policies can be changed to
facilitate better data collection
and to avoid obvious risks to animal and human
health.
Practices for Providing Better Assessment of
Health Impacts
Nondisclosure Agreements
Nondisclosure
agreements between injured parties
and corporations make it difficult to document
incidents of contamination.
Compensation in the form of cash, payment for all
settlement expenses, an offer to buy
the property and/or payment for medical
expenses in exchange for a
nondisclosure agreement prevents information
on contamination episodes and health
effects from being documented and
analyzed. Nondisclosure agreements
are common in all areas of business and are
often essential to protect
intellectual property. However, when documentation of
health problems associated with gas
operations is shielded from public scrutiny
by a nondisclosure agreement, this
is clearly a misuse of this important
business tool and should be
prohibited. Likewise the lack of prior testing of air
and water, and of follow-up testing
during drilling and after incidents of
suspected contamination, impedes the
analysis of health impacts.
Even when testing is
done, the results are being withheld
from interested parties either by government
agencies (e.g., by incomplete
responses to FOIA requests) or by the industry.
If the industry, government
agencies, and the public truly want the facts, then
appropriate testing must be done,
and full disclosure of all data associated with
both baseline and incidents of
suspected contamination must be made. Without
full disclosure of all facts,
scientific studies cannot properly be done.
Science should drive decisions on
whether or not to use a practice such as
shale gas drilling, and until
scientific studies can proceed unimpeded, then an
accurate assessment cannot be made.
Food Safety
A major problem is
the lack of federal funding for food
safety research.
We documented cases
where food-producing animals exposed
to chemical contaminants have not
been tested before slaughter and
where farms in areas testing positive for
air and/or water contamination are
still producing dairy and meat products for
human consumption without testing of
the animals or the products. Some of
these chemicals could appear in milk
and meat products made from these animals.
In Case 3, a quarantine was
instituted after cattle were exposed to
wastewater. However, basic
knowledge, such as hold times for animals exposed to
chemical contaminants as a result of
gas operations, is lacking, and research
in this area is desperately needed
to maintain an adequate level of food safety
in our country [53]. Without this
information, contaminants in the water, soil
and air from gas drilling operations
could taint meat products made from these
animals, thus compromising the
safety of the food supply.
Routes of Exposure
The major route of
exposure in the cases documented
here is through water contamination. This
is perhaps the most obvious problem
(seen in all three case studies), but other
routes of exposure are of serious
concern. Soil contamination can be significant in
situations such as that described in
Case 3. Although the cases we have
documented thus far include only a
handful of exposures through affected air, the
actual incidence of health effects
may be underestimated due to a lack of air
sampling. In Case 1, toxicological
testing suggested high levels of ambient benzene
due to a nearby impoundment pond,
but air canister tests were not done at the
time. Neither drilling companies nor
state environmental regulatory agencies
routinely offer air canister tests
as a part of testing protocols, and due to the
expense, many property owners are
reluctant to pursue them on their own.
Nevertheless, the effects of air
pollution on cardiovascular and respiratory health
have been well documented [54], and
we believe that exposure to contaminated air
may contribute significantly to the
health problems of both people and animals
living near gas drilling operations.
In several cases where air
monitoring was done, the results
confirmed the presence of
carcinogens commonly known to
originate from gas industrial
processes such as exploration, drilling, flaring,
and compression. Thus, the
Environmental Protection Agency (EPA) must include a
study of air in its congressionally
mandated hydraulic fracturing study [55]
if it is to be complete.
Testing
The most important
requirement for an assessment of the
impact of gas drilling on animal and human
health is complete testing of air
and water prior to drilling and at regular
intervals after drilling has
commenced. This includes chemicals used in the drilling
muds, fracturing fluid and
wastewater (the latter contains heavy metals and
radioactive compounds normally found
in a par ticular shale [56]). Currently, the
extent of testing (particularly for
organic compounds) is frequently inadequate
and limited by lack of information
on what substances were used during the
drilling process. In a number of the
cases that we have studied, drinking
water is clearly unsuitable for
human and animal consumption, based not only on the
smell and turbidity, but also on
pathological reactions to drinking the water.
Nevertheless, because of inadequate
testing, the water is deemed fit for
consumption and use, and neither
bottled water nor the large plastic containers known as
“water buffaloes” are typically
provided for the affected individuals-and even
less commonly for animals living on
those farms. In Case 1, water was reluctantly
provided for the humans (after
considerable effort) but not to the animals living
on the farm. Even when identified,
the health effects of chemicals associated
with the drilling process are
unknown in many cases. No Maximum Contaminant
Levels (MCLs) have been set by the
EPA for many of the compounds used, and
those that have been set are based
on older data that does not take into con
sideration effects at significantly
lower concentrations (e.g., endocrine dis ruption
[5]). Furthermore, the disclosure of
all chemicals involved in the drilling and
hydraulic fracturing processes is
not required if a component can be justified as a “trade
secret.” In order to be complete,
air, soil and all sources of potable water used
for humans and animals in the
vicinity of a well site (at least within 3,000 feet for
soil and water tests [57], and five
miles for air monitoring, based on dispersion
modeling of emissions from
compressor stations [58]) must be tested for all
components that are involved in
drilling and are likely to be found
in wastewater, before
any work on the site commences.
Sampling must then be repeated at intervals
following the commencement of
drilling as well as upon suspicion
of adverse effects. The
following practices must be part of
a testing protocol:
1. The sampling must
be done by a disinterested third
party with a clear chain of custody
between sampling and testing. A
certified independent laboratory must do
the testing, and the results must be
available to all interested parties.
2. All chemicals
(with IUPAC names and CAS numbers)
used in the hydraulic fracturing fluid at
any concentration for each well must
be disclosed to the property owners
within a five-mile radius, testing
laboratories, local governments, and
state agencies. Material Safety Data
Sheets (MSDSs) for each chemical and
chemical mixture must accompany this
disclosure.
Following this
procedure will allow prior testing
to be targeted to specific chemicals to be used
in the drilling process for a
specific well, as well as providing valuable
information to first responders and
hospital personnel in the case of an
accident.
3. Upon suspicion of
adverse health effects, testing must
include air, soil, wastewater, all
sources of drinking water, and
blood, urine and tissue samples from affected
animals and humans. If methane is
present in drinking water,
isotopic analysis to determine the
origin (thermogenic vs. biogenic) must be
done.
4. As illustrated by
several cases we documented, air
canister tests are essential. This must be
done as a baseline before drilling
begins and during and after well flaring.
It must also be done after a
wastewater impoundment and a compressor
station have been established.
5. Any fracturing
fluid chemicals and chemicals
released from the shale that are known or
possible human carcinogens, are
regulated under the Safe Drinking Water
Act, or are listed as hazardous air
pollutants under the Clean Air Act
must have MCLs, which are set by the
EPA. Many of the chemicals to which
both people and animals are exposed
as a result of high-volume hydraulic
fracturing are not listed as primary contaminants, and thus have no
enforceable MCL. More than half of
the chemicals listed as toxic chemicals in
a recently released U.S. House of
Representatives report [59] have no
MCL.
6. All testing
expenses must be a part of the cost
of doing business for gas drilling companies. Testing before and
during drilling operations is an
important part of documenting health
effects. If health effects are
related to a chemical pre-existing
in a pond or well, this
would prevent a false association
between drilling and water contamination.
Alternatively, if a change in
chemical composition is correlated to health changes,
then a strong justification for
compensation is provided. In numerous cases that
we documented, compensation was not
provided because adequate prior
testing had not been done. By doing
complete testing, at the proper
times, a clear scientific
justification can be made for
providing or denying compensation. Beyond
that, a better understanding of what
practices lead to water contamination can be
obtained. This will be a benefit to
people living in the midst of shale gas
drilling and will, in fact, benefit
the industry by providing consistent and useful
data to guide operations. The
current practice of undertesting and denying any link
between drilling and water, air, or
soil contamination is beneficial to
neither the public nor the industry.
Practices for Avoiding Animal and Human Exposure to
Environmental Toxicants
As shale gas drilling
expands across the northeastern
United States, exposure of animals and humans
to environmental toxicants can
result from negligence, illegal actions,
catastrophic accidents (at drilling
pads or compressor stations), or normal operations.
Negligence and illegal actions are
difficult to prevent and may have
contributed to the health problems
we documented. Suspected illegal dumping of
wastewater and the alleged
compromise of the liner of a wastewater
impoundment were most likely
responsible for cattle deaths in two instances that we
studied. Cases of alleged wrongdoing
[60] illustrate the vulnerability of
agricultural operations in the midst
of large volumes of toxic waste. Dumping and
other intentional violations are
difficult to prevent or regulate given the
large numbers of small companies
involved in servicing drilling operations
and the lack of willingness and
funding on the part of state environmental
regulatory agencies to investigate
and fine the gas industry. The prevalence of small
subcontractors increases the
possibility that best practices will not be followed
due to inadequate training and
supervision.
Although accidents
might be minimized with strict
safety standards and careful inspection,
regulatory agencies would require
sufficient staff to monitor operations. This is
obviously not the case in
Pennsylvania, where 666 environmental health and
safety violations have been reported
in 2011 as of June [61]. With a staff of
37 inspectors [62] and 64,939 active
wells (as of December, 2010), regulatory
oversight is essentially impossible.
The situation is even worse in New York
State, where only 16 inspectors are
currently on the staff of the Department of
Environmental Conservation. Although
the number of staff positions required to
police this industry adequately
would necessarily be very large, hiring of new
inspectors is essential if
environmental and health damages are to be minimized.
New York, Pennsylvania, and Iowa are
the only active drilling states that
have no severance tax for drilling
operations. A severance tax could fund
additional inspectors and help
insure compliance with existing regulations, although
this will require the political will
to levy a tax sufficient to fund the required
number of inspectors. Given the high
probability that accidents will happen
[63], increasing setbacks between
homes, barns, schools, ponds, and streams
would provide some additional
security. The current regulation in
Pennsylvania is a setback of 200
feet from water supply springs and wells, 100 feet from
surface water bodies, and 200 feet
from wetlands. The revised draft
supplemental generic environmental
impact statement in New York indicates a 500-foot
setback from private water wells.
Increasing these setbacks 5- to 10-fold would
decrease but not eliminate the
impacts of accidents such as the April 20, 2011 spill
in Bradford County, PA [64].
Contamination of the air by compressor station
blowouts and contamination of
streams leave an imprint that cannot be easily
mitigated by even the most stringent
setbacks.
Normal practices can
be modified to reduce but not
eliminate exposure of humans and animals to
toxicants associated with gas
drilling. One of the important problems
associated with shale gas drilling
is the huge volume of wastewater generated.
This wastewater, which includes
flowback and produced water, contains at
different times in the process the
chemicals used in the hydraulic fracturing
fluid as well as compounds and
minerals extracted in the fluid flowing
back with hydrocarbon gas. The
materials extracted from underground can be
equally or more toxic than the
hydraulic fracturing fluid, and include radioactive
material (e.g., radium-226,
radon-222, and uranium-238), arsenic, lead,
strontium, barium, benzene, chromium
and 4-nitroquinoline-1-oxide [56]. However,
despite the actual toxicity of this
material, according to the EPA, “drilling
fluids, produced waters, and other
wastes associated with the exploration,
development, or production of . . .
natural gas” are considered “solid wastes which are not
hazardous wastes” [65]. This allows
the substances to be spread on roads as
deicing solutions and as solutions
to minimize dust and sets up a potentially
lethal threat, particularly to
companion animals, wildlife, and children. Typically
these solutions contain high salt
concentrations and attract dogs and cats, as was
illustrated in Case 1. This hazard
can be easily mitigated by not allowing
wastewater to be spread or sprayed
on roads.
Before wastewater is
removed from a drilling site, it is
often stored in large impoundments
(sometimes serving multiple well
pads) where the volume is decreased by
evaporation. This increases the
concentration of some toxic substances in the
impoundment (salts, heavy metals)
and also introduces other toxicants into
the atmosphere (e.g., volatile
organics such as benzene and toluene). In
addition, impoundments are
associated with a number of deaths of both cattle and
wildlife [66]. These effects raise
the question of whether wastewater should be
stored in open impoundments. Whereas
this may be economically
advantageous to the drilling
company, the environmental and agricultural
impacts are too great to allow this
practice to continue. In Pennsylvania, some
progress has been made in recycling
increasing fractions of the wastewater.
This decreases the total volume of
wastewater but increases its toxicity due to
the successive increase in the
concentrations of total dissolved solids. The
alternative is to store wastewater
in metal containers at the drilling site before
it is removed for disposal.
Finally, the disposal
of wastewater presents significant
environmental risks. Cases of alleged
dumping of untreated wastewater in
streams have been documented in the press
(e.g., [60]). In the southwestern
United States, wastewater is disposed of in
injection wells; however, the
prevalence of nonporous sandstones and shales in
Pennsylvania and New York State
largely precludes the use of disposal
wells. An earthquake of magnitude
3.2 was associated with injection into a
hydraulically fractured vertical
well on February 3, 2001 near Avoca, New York [67],
suggesting that seismic
considerations may further limit the development
of injection wells in New York
State. Similar seismic occurrences in other
parts of the country, most recently
in Ohio [68], may mean that New York and
Pennsylvania will have fewer options
for disposal of wastewater due to
shale gas drilling. In May 2011, a
voluntary moratorium was placed on the
acceptance of hydraulic fracturing
wastewater at sewage treatment plants in
Pennsylvania. These plants are not
equipped to handle either the radioactive and toxic
compounds or the high salt content
of this waste, and the increased use of
recycling has magnified the problem.
Discharge of water treatment plants into
the Monongahela River led to the
contamination of drinking water in Pittsburgh
in 2010 [63]. Sewage treatment
plants clearly are not a viable option for disposal
of wastewater, and despite the
industry's progress in recycling, suitable
injection wells are unlikely to be
located to support the scale of drilling planned in
Pennsylvania and possibly New York
State.
CONCLUSION
Animals, especially
livestock, are sensitive to the
contaminants released into the environment by
drilling and by its cumulative
impacts. Documentation of cases in six states
strongly implicates exposure to gas
drilling operations in serious health
effects on humans, companion
animals, livestock, horses, and wildlife. Although
the lack of complete testing of
water, air, soil and animal tissues hampers
thorough analysis of the connection
between gas drilling and health, policy
changes could assist in the
collection of more complete data
sets and also partially
mitigate the risk to humans and
animals. Without complete studies, given the
many apparent adverse impacts on
human and animal health, a ban on shale gas
drilling is essential for the pro
tection of public health. In states that
nevertheless allow this process, the
use of commonsense measures to reduce the impact
on human and animals must be
required in addition to full disclosure and
testing of air, water, soil,
animals, and humans.
Direct reprint
requests to:
Robert E. Oswald
Department of
Molecular Medicine
Cornell University
Ithaca, NY 14853
e-mail: reo1 [at] cornell.edu
Complete Abstract (PDF -
513KB)
