Some studies—such as a 2011 paper by Stephen G. Osborn, Avner Vengosh, Nathaniel R. Warner, and Robert B. Jackson, who were then at Duke—linked fracking to groundwater contamination. There has been less evidence on surface-water contamination, other than due to isolated spills and leaks. The study by Bonetti, Leuz, and Michelon finds systematic evidence of increased salt concentrations in surface waters related to new hydraulic fracturing wells in the same watershed.
The researchers used a geocoded database that combined surface-water measurements with 46,479 hydraulic fracturing wells from 24 shales across 408 watersheds from 2006 to 2016. They specifically examined concentrations of bromide, chloride, barium, and strontium—ions usually found in high concentrations in flowback and produced water from wells. These ions do not biodegrade and have been found several years after spills. Using a statistical approach, Bonetti, Leuz, and Michelon identify anomalous changes in ion concentrations associated with new wells in the same watersheds.
The researchers find small but consistent increases in barium, chloride, and strontium concentrations, but not bromide, for many watersheds across the US.
The increases in salt levels were largest during the early phases of production, when wells generate large amounts of flowback and produced water. The salt concentrations were most pronounced for wells with larger amounts of produced water and for wells located in areas where the deep formations exhibited higher levels of salinity. This evidence as well as the time patterns tie the elevated ion concentrations more closely to the unconventional oil and gas development process.
“While the elevated levels we discovered were well below maximum EPA and health advisory levels, it is important to recognize that the water measurements were predominantly taken from rivers, and that not all wells are close to surface water. Moreover, not all monitors in a watershed are in locations where they could detect an effect,” says Leuz. He further notes that the study was hampered by the availability and measurement frequency of water-quality data. Hydraulic fracturing fluids contain chemical substances that are potentially more dangerous than salts, but Bonetti, Leuz, and Michelon weren’t able to analyze these chemicals, as they’re not widely covered by public databases.
Better and more frequent water measurement could improve understanding of the surface-water impact of unconventional oil and gas development, the researchers note. For instance, federal and state environmental agencies could consider placing monitoring stations in a more targeted way to improve tracking of changes in water quality.