Rancher Stacey Mills realized something was wrong when his cattle started falling sick and dying. Half of his 70,000 acres in the southeastern corner of New Mexico drew from an aquifer that had the cleanest, sweetest water. But tests in 2006 revealed that his water well was contaminated with brine.
A month earlier, EOG Resources Inc. had begun disposing of brine, a waste product of drilling oil and gas wells, in an abandoned oil well nearby. It was never established that the EOG brine had polluted the aquifer; the leak was sealed before officials tested for links between the well and the aquifer, Mills said. An EOG spokeswoman stressed that tests did not show communication between the brine well and aquifer. But in an effort to be a good neighbor, the company built Mills a new water well on his property and the infrastructure to go along with it.
The takeaway from the incident for James Goodbar, who leads the Bureau of Land Management's caves and karst resources program, was that oil wells do fail after a few decades.
A rash of new applications to drill had been cropping up since 2004 in southeastern New Mexico, in the Permian Basin oil and gas deposits. The region contains karst, a type of geology composed of rocks that dissolve in mildly acidic water over time, creating grottos that've made New Mexico a destination for cavers. Water flows quickly through these underground drainage systems, and a contaminant spilled in one place can end up in water wells miles away. Spills and leaks, if they happen, are harder to contain and clean up in karst areas than elsewhere.
Goodbar wanted to make sure that oil and gas wells do not leak, a century later, into the aquifers that supply Carlsbad, N.M., with its water supply. Other industries have contaminated springs and wells in karst terrain before. Now, with the new industrialization of America's oil and gas, scientists worry about similar contamination as more applications to drill on karst come in.
"We are looking in terms of contaminations that are 100, 200 years from now," Goodbar said. "Probably, I think, we will still be drinking water a couple of hundred years from now."
When drilling a well, companies use steel pipes -- "strings of casing" in industry jargon -- to line the borehole and prevent the rock sides from caving in. The operator drills the bore to a certain depth, then inserts the casing and attaches it to the surrounding rock with cement. Once the cement sets, the well is drilled further, and a second string with a smaller diameter is inserted into the bore. This string is cemented, or "tied back," to the first steel pipe. The process continues until the source rocks are reached.
If properly done, a few casing strings and cementing tie-backs are sufficient in most geologic structures. For example, out of the 33,000 wells drilled in Ohio between 1983 and 2007, there were only 185 reported cases of groundwater contamination, according to a report from the Ground Water Protection Council. Of these, only 41 were due to old wells leaking.
But this construction can pose a greater danger where the well bore intersects caves. Exposed to the elements because there is no surrounding rock, the cement and metal can corrode and leak over time. Threats to the springs and the wildlife that depend on them may be significant.
So, the BLM updated its requirements for karst drilling in 2006, requiring at least three layers of high-grade steel and cementing all the way to the surface. The more layers isolating the well from the caves, the better. When abandoning an old well, the operator has to plug the well with cement from the lowest karst zone, 2,500 feet deep, all the way to the surface.
But the standards are not applicable on nonfederal karst lands in places like West Virginia and New York's Schoharie County. And concerns remain about the thousands of older wells around Carlsbad.
The new requirements "would still not take care of the old wells," said Wesley Ingram, supervisory petroleum engineer at BLM's Carlsbad field office. "I mean, with the old wells, the only way you are gonna eliminate any potential difficulties from them is to actually plug the wells."
John Krohn, a spokesman for the industry group Energy in Depth, said drillers in the Marcellus Shale implement at least three strings of casing with many installing additional layers of protection beyond this minimum requirement. Included in these layers is the outmost layer of the well, called the "conductor casing" -- cemented to the surface. Casing failure is, in any case, a rare event, he said.
Goodbar pointed out that BLM does not consider the conductor casing a separate string. Under the BLM's definitions, most operators have only two casing strings cemented all the way to the surface, he said.
Linking oil wells to water wells
BLM embarked on a series of experiments in New Mexico to justify its special casing rules and prove there are links between oil wells and aquifers. Goodbar and his colleagues asked oil companies to include certain nontoxic dyes in their drilling fluid. The participating companies drilled 23 wells in 2007.
If any drilling fluid escaped the well bore and migrated to a distant water well, the dye would theoretically be recovered there. The scientists placed dye-detection traps in three springs, two domestic water wells and the Capitan Reef aquifer that supplies Carlsbad's water.
While drilling, five operators lost their drilling fluid. The scientists later found traces of the injected dye in traps in some of the water wells.
The study "was to scientifically prove, without a shadow of a doubt, that there was a connection between oil and gas drilling and our water supplies," Goodbar said. "And once that was established and everybody was kind of like, sure, there's that connection all right, there was no need to continue that pilot project."
BLM updated its well casing requirements, keeping in mind the potential for contamination. In critical karst areas, the agency requires three casing strings -- four including the conductor pipe -- of a high grade of steel cemented externally all the way to the surface. Operators also are required to perform pressure tests yearly to ensure there are no casing failures or leaks, Ingram said. And under controversial rules proposed by the Interior Department for drilling on federal land, operators would have to get officials to sign off on cement bond logs, which contain records of the quality of the cement jobs. Requirements for abandoning older wells were also tightened.
And New Mexico does not allow pits and tanks holding wastewater on cave and karst areas in order to minimize the risk of surface spills.
New Mexico's Oil Conservation Division (OCD) has followed BLM's lead with stringent casing requirements on a case-by-case basis, according to officials. Now, it is up to the operators to ensure a proper casing and cementing job by conducting necessary pressure tests, especially because the agencies are significantly short-staffed. There are five inspectors overseeing 17,000 wells on BLM land, and the agency receives about 20 to 30 new permit applications every week, Ingram said. The OCD faces a similar staff crunch.
"Our inspectors are working incredibly hard, and they are able to make it to their well sites and continue their activity in this growing time," said Jami Bailey, director of the OCD.
Need for caution
Behind the need for stringent well casing in cave areas is the geology of the rock. Typical sandstone terrains hold water like a sponge, in microscopic spaces. The water moves slowly through these spaces, just an inch or so every year. Any pollutants stick to the sand, and the water gets filtered over time.
But karstic terrains are different because the water quickly flows through large conduits and channels, said Lewis Land, a cave and karst hydrologist at the National Cave and Karst Research Institute. If there are contaminants, they get transported to distant wells without any kind of natural filtering.
"So if you are drilling in a karstic area, there's always a greater potential for contamination; you have to proceed with much more caution when you are drilling," he said.
Well casing leaks related to oil and gas have so far been rare, but, when they have occurred, the consequences have been remarkable. In Kansas, corrosion of the casing of a brine disposal well allowed the brine to escape into the surrounding karst for many years. The brine eventually dissolved enough of the rocks to cause the Panning Sinkhole in 1959. Regulations for well construction have since been tightened in the state.
Worries about drilling surfaced in Monroe County, W.Va., when landmen began collecting mineral rights two years ago. The karst area there overlaps the edge of the Marcellus Shale, but activity has not yet ramped up due to the low price of natural gas today. Harold Parsons, a geologist formerly with the state's Department of Environmental Protection, carried out some dye tracer studies to explore the paths groundwater takes in his county.
Monroe County has entered into a memorandum of understanding with the oil company, asking it to triple case wells and use high-grade steel. The West Virginia Legislature also recently updated its drilling laws, suggesting additional well casings may be needed in karst areas.
Even so, Parsons is not certain the threat has been addressed. The area is too rural to get piped water, so groundwater contamination would be a problem, he said. His tracer studies revealed that some sinkholes that resemble innocuous, littered ditches link directly to water supplies.
"Even if you did triple casing, if those fail for some reason, if they weren't properly installed, for example, they could fail and you may have the same situation," Parsons said. "If a truck hauling frack fluid or any kind of chemical wrecks on the road and spills its contents into a sink that goes into the underground water system, then we've got a problem."
In New York state, where Gov. Andrew Cuomo (D) has suggested he will allow drilling on the Marcellus Shale from 2013, environmental scientists -- including one of the world's most respected speleologists, Arthur Palmer -- have asked the Department of Environmental Protection to ban drilling on karst terrain.
"Our aquifers didn't form overnight; they took at least a million years to form through geologic time and erosion," said Paul Rubin, hydrogeologist at the environmental consulting firm HydroQuest in New York. "And there's no reason to believe that they shouldn't operate, if we don't contaminate them, for the next million years."