A
quiet, rural region of southern Colorado was shaken up in recent
weeks when a series of small earthquakes caused minor damage --
and raised some serious questions about possible causes, including
industry activity. Some are concerned that natural gas production
is causing the tremors.
The little town of Segundo, Colo., sustained minor
damage to buildings during one of the earthquakes, and area residents
can be excused for wondering just why their quiet little corner
of the world suddenly seems like it's sitting atop the San Andreas
fault. Colorado has only one major fault, so there should be little
danger of major earthquakes. The Sange de Christo fault, not far
from Segundo, is a major fault. A maximum credible earthquake of
7.5 Mw has been assigned to it.
The state has had, however, a history of small tremors,
and the U.S. Geological Survey's National Earthquake Information
Center in Golden, Colo., along with the Colorado Geological Survey
is currently working to gather and analyze additional information
about these recent events.
Clearly, this recent cluster of seismic events has
gotten the USGS' attention. The National Earthquake Information
Center typically records an average of eight to 10 quakes a year
in all of Colorado, so these recent tremors are significant.
The USGS has installed 11 portable seismographs to
gather additional information in the three- to five -- mile wide
earthquake area. From Aug. 28-Sept. 21 there were 12 earthquakes
reported, ranging in magnitude from 2.8 to 4.6 on the Richter scale.
The epicenters of the earthquakes are calculated to lie about 10
miles west/southwest of Trinidad, and data from the seismographs
indicate that the earthquakes occurred along a previously unknown
fault.
Prior to the installation of the portable seismographs,
the USGS was unable to precisely determine the earthquakes' epicenters
because the only data available was from seismographs located over
150 miles from Trinidad in Albuquerque, Idaho Springs and Golden.
Following analysis of the data from the portable
seismographs and seismic data, and geologic information from Evergreen
Resources, an oil and gas operator active in the area, Vince Matthews,
said that there is a northeast-to-southwest-trending high running
through this area of Colorado.
The earthquakes occurred on the southeast side of
this high, according to Matthews, who is a senior science advisor
with the Colorado Geological Survey and an AAPG member -- right
where the earthquakes would project to the surface.
"It appears there is a structure running through
this region that was not previously recognized with surface mapping,"
Matthews said. "So there is a geologic explanation for the seismic
activity."
It's a Mystery
What is still not known -- and may never be completely
understood -- is what set off the earthquakes.
The USGS is still working on velocity models to accurately
determine the depth of the earthquakes. The character of the earthquake
seismograms, however, implies that they occurred in the uppermost
10 miles of the earth's crust.
"Many people in the area are questioning whether
natural gas activity could be causing the earthquakes, but until
we pinpoint the depth of the events and gather additional data we're
not able to say one way or the other what the possible causes might
be," said Waverly Person, chief of the USGS National Earthquake
Information Service.
"Typically if the earthquakes are centered deeper
than the wells in the area, it can be determined that the activity
was not a cause. That's why it's important to determine the depth
of these earthquakes."
Person did say Colorado has experienced earthquake
clusters in the past. In 1984 a series of tremors was felt in the
Carbondale area, and two years later the area around Crested Butte
recorded a number of earthquakes.
"There's a lot of unanswered questions at this time,"
he added, "and we are working closely with the Colorado Geological
Survey to find the causes for these earthquakes. We hope to work
with the companies active in the area to further our efforts. "It's
also important to note that the last recorded earthquake was on
September 21 (then 10 days previous), so the events seem to have
quieted down."
Whole Lotta' Shakin'
There have been earthquakes in this region previously.
Tremors similar in size to the recent earthquakes
have occurred in southern Colorado and northern New Mexico since
seismographs have monitored the region. For example, a magnitude
4.6 shock in October 1966 produced minor damage in Trinidad, and
in the broader region there have been several dozen shocks of magnitudes
greater than 3.0 since 1961.
"It's also important to note that in recent years
the population of this region of Colorado has increased significantly,
which increases public awareness of seismic events," Person said.
The first known reference to a Colorado earthquake
was in December 1870 near Pueblo -- but the first seismographs were
not installed in the state until 1909, and sufficient quality and
quantities of seismographs were not installed until 1962.
The largest earthquake in Colorado was recorded in
1882 -- at the peak of Colorado's mining era -- and was likely centered
in the Front Range near Rocky Mountain National Park. The tremor
is estimated to have been about 6.6 Mw, and was the first to ever
cause damage in Denver.
It was felt as far away as Saline, Kan., and Salt
Lake City.
The recent series of earthquakes are in the Raton
Basin, and coalbed methane operations are the biggest concern to
area residents. The Vermejo coals -- flat lying units -- are located
at the surface in the area of the quakes. There has been coal mined
from the area, and in recent years an extremely successful coalbed
methane play has developed in the basin.
Matthews said a water injection well is located in
line with the earthquake activity and the previously unknown structure.
The well was drilled to about 4,100 feet in April 2000, and over
the last year 2.5 million barrels of water have been injected.
Evergreen Resources, the most active coalbed methane
operator in the Raton Basin, drilled the well to re-inject fluids
from its de-watering programs on the coalbed methane wells.
"The state geological survey is charged with protecting
the welfare of the people of the state," Matthews said, "so it's
important that we coordinate all the available information to try
and find out the cause of this seismic activity -- and whether there
is a problem that should be rectified."
Evergreen is certainly aware of the concern that
natural gas operations in the Raton Basin could be contributing
to the earthquakes and is cooperating with the USGS and the Colorado
Geological Survey. Matthews said he has been working with Evergreen
scientists to get a better picture of the geology in the region
and the firm has supplied detailed information as well as seismic
data.
Mark Sexton, president and chief executive officer
of Evergreen, is confident that his firm's operations are not impacting
the earthquake activity.
"Preliminary indications seem to suggest that the
seismic activity is caused by deep-seated basement fault movement,
and this is completely consistent with our geologic model of the
basement," Sexton said. "Our injection activity is only one to one
and a half kilometers in depth -- nowhere near the likely depth
of the geologic activity creating the newly mapped faults in the
area."
Also, the company injects into the Dakota and Purgatoire
formations, which are 100+-foot thick sandstone bodies that essentially
take the injected water on a vacuum.
Also, the water injection is not done with any surface
pressure, but is simply a gravity feed operation.
"We are not causing any fracture porosity," Sexton
said, "we are simply filling up void spaces in the sandstones."
More Studies Needed
Sexton said it is not surprising that structures
and faults in this part of Colorado have not been mapped.
"Prior to 1995, when a pipeline was installed that
opened the Raton Basin for oil and gas operations, there were less
than two dozen wells drilled throughout the basin," he said. "Historically
there was no incentive to intensely study the geology of the region."
Today the basin boasts more than 700 wells, but all
that activity is very shallow, and the vast majority of water from
the coalbed methane wells has been allowed to simply percolate back
into the earth, he said.
Evergreen also points to the fact that the earthquake
activity has died down since Sept. 21, although the firm has not
altered its water injection rates or drilling activity. This could
be an indication that other factors are at play in the region.
"This area has been the site of earthquake activity
in the past," Sexton said. "Until now there hasn't been a swarm
of earthquakes documented, but that could simply be a lack of data
and no knowledge of the periodicity of the fault movement.
"It may be that every 200 years it's common for this
area to have a swarm of small earthquakes," he added. "We just don't
have any way to know."
Matthews said the coincidence of a swarm of small
earthquakes along a fault that also happens to pass very close to
the injection well warrants additional study.
"In my view there may never be definitive proof one
way or the other on the cause of these earthquakes," Matthews said.
"But we have a responsibility to conduct further studies to get
a better geologic picture of the region."
Historical Precedence
There is definitely historical precedence for manmade
causes of earthquakes in Colorado.
"This state is the biggest natural laboratory in
the world for human-induced earthquakes," Matthews said. "There
have been three major experiments in the state concerning human-induced
events that prove human activities can indeed touch off earthquakes."
♦
The most famous episode of a human-induced earthquake began in 1961,
when a 12,000-foot disposal well was drilled in the U.S. Army's
Rocky Mountain Arsenal northeast of Denver. The well was used for
disposing of waste fluids from arsenal operations, and injection
commenced in March 1962.
Shortly thereafter an unusual series of earthquakes
erupted in the area, and by the end of December 1962 about 190 earthquakes
had occurred. None caused damage until December, when several structures
were damaged in Dupont and Irondale.
Over 1,300 earthquakes were recorded between January
1963 and August 1967. In April 1967 the largest earthquake since
the series began in 1962 occurred, and damage was recorded in the
arsenal, Derby and Boulder. This tremor measured 5.0 on the Richter
scale.
Even after the Rocky Mountain Arsenal waste dumping
practice stopped, earthquakes continued to be felt in the Denver
area, so in 1968 the Army began removing fluid from the arsenal
well very slowly in an effort to reduce the earthquake activity.
♦
The second episode was in the 1970s at the Rangley Field in northwest
Colorado, Matthews said. The USGS heard reports that earthquakes
in the area were touched off by water flooding in the field.
The USGS got permission from Chevron, the field's
operator, to conduct an experiment in part of the field on some
abandoned wells. The USGS stopped injection in the area, and earthquake
activity dropped from about 50 a day to one or two. Scientists then
began injection again to determine if the earthquakes would increase
again when the pressure built up. Sure enough, the episodes jumped
back up to about 50 a day.
The USGS shut down injection operations and the tremors
died down.
p The most recent experiment is in the Paradox Valley,
where the U.S. Bureau of Reclamation is attempting to prevent salt
from entering the Dolores River and then flowing into the Colorado
River, according to Matthews.
"The Paradox Valley is a salt anticline, and the
salt in the Delores River comes from the ground water," he said.
"The Bureau of Reclamation has drilled a series of wells all along
the river to intercept the ground water flowing into the river,
and that water is then injected into a 14,000-foot injection well.
"The bureau has a very talented seismo-tectonic group
that researches earthquakes related to dams and other projects,
so when this Paradox Valley injection program began they expected
to generate earthquakes," he continued. "They installed a network
of instruments to monitor any activity and, sure enough, tremors
did start when they began water injection."
The program has generated over 4,000 earthquakes,
but most are too small to be felt on the surface, he said. The earthquakes
built until June 2000 when there was a magnitude 4.5 event. That
earthquake was large enough that Bureau of Reclamation scientists
began looking for a remedy to the situation.
Today the USGS said the bureau is injecting water
every other month to minimize the earthquake activity.
"There's not another place in the world that's had
as many manmade earthquakes as Colorado," Matthews said. "For that
reason we have to look seriously at any series of tremors we have
and determine their cause."
There have been earthquakes in the Trinidad area
in the past, and no manmade cause for those earlier quakes has been
found. Also, there have been earthquake swarms in the state.
But this swarm is unique in that the events occurred
so close together. That's why residents are hopeful that data from
the USGS instruments will provide some answers -- fast.