In September of 2010 there was a blowout during early stage hydraulic fracturing of an oil and gas well in Dunn County, North Dakota just a couple of miles from the town of Killdeer. You’ve seen the result of a blowout. Think Gulf of Mexico, British Petroleum, Deep Water Horizon. Strangely the Deep Water Horizon blowout also occurred in 2010. You probably saw those pictures on television of oil flowing from the borehole at the bottom of the gulf. It really made a mess. The reason it made such a mess was the spill was difficult to contain because of the depth of water. When a blowout occurs on land, it is easier to contain but it makes a mess as well. All sorts of hydraulic fluid, oil and formation water come barreling up the borehole and spill all over the ground. If the casing in the borehole is ruptured as it was in this blowout, the fluids are also ejected into other formations, including possibly groundwater aquifers. In Dunn County the blowout spewed up some 90,000 to 100,000 gallons of fluid depending on the press you read. There are no reports of how much might have been lost in the subsurface. The blowout was contained and the North Dakota environmental regulators quickly had the situation at the surface cleaned up and monitoring wells installed in the groundwater aquifer.

 

The Environmental Protection Agency (EPA) chose this blowout site near Killdeer, North Dakota to conduct a “retrospective” study on whether there were impacts to the groundwater from hydraulic fracturing. If you are a regular reader of this blog, you know I am reviewing EPA “retrospective” studies. This is the fourth report I have reviewed and it is titled: “Retrospective Case Study in Killdeer, North Dakota: Study of the Potential Impacts of Hydraulic Fracturing on Drinking Water Resources.” EPA calls these retrospective studies because they are being done after the contamination has occurred without the type of previous groundwater quality characterization and hydrologic study that a scientist would like to have in order to make a solid determination of how groundwater contamination occurred. However these types of studies are done all the time at the EPA when they seek to determine contamination from industrial sites, so they do have a lot of experience and some good tools to use in making their determinations. But with a limited data set, scientists can only go so far in determining how the contamination got to a well.

 

A lot of people probably don’t think about North Dakota much. When they do think of it, they think it is cold and snowy and very flat. They do not load up the motor home and head to North Dakota on summer vacation. But the area around Killdeer is really pretty. It has your typical grasslands of course, but it has badlands as well. There are a number of mesas in the area which they call mountains and just west of the town of Killdeer are the Little Missouri National Grasslands and the Teddy Roosevelt National Park. Set among this landscape is one of the largest oil and gas booms going on in the country.

 

I have actually been to Killdeer, North Dakota. Not many people have. It is a very small town in a lightly populated county. I think EPA’s report said there were about 4000 people in the whole county. I worked at one time down the road in Dickinson, North Dakota. There are a number of towns like Killdeer in the area – small, rural and isolated. Oil and gas production in the area has increased so greatly now that it is the chief industry in the area. I visited long before the ramp up in the recent oil and gas boom. At the time there were not a lot of strangers who visited Killdeer. I stopped in one day at the grocery store to pick up some snacks and a drink after a long day in the field. Everybody in the grocery store turned around to look at me when I walked into the store. Obviously this was a town where everybody knew everybody else. When I got to the checkout counter, the clerk asked me even before she rang me up: “Who are you?” I had to spend ten minutes explaining myself and my presence in town. I can guarantee you it is not like that now. With the exploitation of the Bakken Shale in North Dakota, this previously isolated rural area has changed forever. You read about the expansion of oil and gas in North Dakota in the paper: not enough workers, man-camps because there’s no housing, fast food workers getting paid New York City wages, crime because high-paying jobs attract all kinds of people including criminals. North Dakota is the United State’s second largest producer of oil and gas. Crazy huh?

 

This area of the country has been producing oil and gas since the 1920’s and development in Dunn County began sometime in the 1950’s. The whole area is part of the famed Williston Basin, which consist of thousands of feet of sedimentary rock. Groundwater is the principal source of drinking water in Dunn County and is found in the glacial tills and glacial outwash. This aquifer is called the Killdeer aquifer and can be some 200 feet thick in areas. So no matter how cold it is in North Dakota today, at one time it was even colder with thick continental glaciers the main feature on the landscape.  It is only about 20 to 30 feet to groundwater from the surface. The water is kind of hard and is generally a sodium bicarbonate or sodium sulfate type water. No matter how untasty the water is though, there’s a lot worse in the area from the underlying rock formations. The first time I went to a restaurant in one of the little towns west of Dickinson, North Dakota, the waitress brought a pitcher to the table filled with a brown liquid. I thought it was coffee, but my co-workers quickly told me it was water. A lot of the water at the time came from fractured lignite in the area. Lignite’s are kind of like underperforming coal. It’s one step up from peat. It burns but not well. Underlying the good glacial groundwater in Dunn County is just such a rock formation called the Sentinel Butte.

 

 

The nine monitoring wells that were installed after the blowout were mostly drilled into the Killdeer aquifer except for one, where EPA spying lignite in the drill log decided the well was drilled into the Sentinel Butte formation. EPA therefore only sampled and analyzed the 8 other monitoring wells. Two of the monitoring wells were upgradient of the oil and gas well that had the blowout. I need to explain what groundwater gradient is here. Scientists call it hydraulic gradient. Simply put it describes the direction of groundwater flow (yes groundwater flows). Mostly groundwater flows along the path of least resistance or pressure. The direction groundwater flows is called downgradient. Groundwater aquifers that are like the Killdeer are composed of sand, silt and gravel. The groundwater is actually in the pore spaces between the sand. The Killdeer aquifer is pretty horizontal lying with a slight southern tilt, which is the direction where groundwater is flowing. The Killdeer is what is called an unconfined aquifer, meaning it doesn’t have a layer of non-permeable rock above it creating a lot of pressure on it. The hydraulic gradient of the Killdeer is very low and the flow of water is very slow. EPA’s report cited references to movement of a foot a year.  Groundwater is moving away from the upgradient wells, which means samples from those wells will have uncontaminated groundwater. These samples can then be compared to samples from wells downgradient from the blowout, which is where the water is flowing to.

 

EPA conducted three rounds of sampling of the monitoring wells as well as a number of local domestic, municipal and supply wells in the area. The sampling occurred between July 2011 and October 2012. EPA recognized right away that two of the downgradient wells were producing anomalous results compared to the other wells they were sampling. They showed statistically significant water quality differences not only with the other wells in the area being sampled but with historical groundwater data as well. Chloride, calcium, magnesium, sodium and strontium were all much higher in the water from these two monitoring wells. EPA suspected from these analyses that brine had intruded into these two wells. EPA then applied standard analysis of the ratios of a number of these water quality parameters to determine from where the brine might have come. Interestingly enough, EPA found that the brine in the two monitoring wells matched the brine signature from the formation overlying the Bakken Shale. EPA’s curiosity on this point led them to look at the scientific literature on hydraulic fracturing of the Bakken Shale. EPA found out that hydraulic fracturing in the Bakken often produces what is called in the industry: “out of zone fracturing”. In other words, the fractures that are created in the Bakken Shale to get the oil and gas out of the rock actually extend into another formation; in this case the rock formation on top of the Bakken. So EPA seems to be implying that when the blowout occurred, brine from the rock overlying the Bakken Shale must have come up the borehole. I was a little confused about why brine from the Bakken didn’t also come up the borehole, but hey this is what EPA came up with, not me.

EPA also analyzed the monitoring wells for constituents found in hydrofracture fluids. For the first time in all of the EPA “retrospective” reports I have reviewed, they actually found a chemical they could link back to hydrofracture fluid. It is called TBA which stands for tert-butyl alcohol. TBA is not actually used in hydrofracturing fluid. It is what is called a degradation product. This means another more complicated organic chemical underwent a chemical reaction and became TBA. According to EPA, there are several chemicals used in hydrofracture that degrade into TBA. The one that was used in the well that had the blowout was tert-butyl hydroperoxide. The TBA was found (surprise, surprise) in the two monitoring wells where the brine intrusion was noted. The presence of TBA brought EPA to the conclusion that groundwater was contaminated in the vicinity of the well with the blowout; the very near vicinity. According to EPA, the monitoring wells were installed 20 feet from the well with the blowout. Thank goodness the groundwater moves so slowly in the area and no one’s drinking water wells were contaminated.