A few years back, while working on ecological research at Oklahoma State University, Brady Allred had what he calls a "crazy idea." What would it look like, he wondered, if you combined the physical footprint associated with oil and gas booms of the past decade across a broad geographic area? Not just the Bakken or the Permian Basin, but, say, the entire central swath of North America from the Gulf Coast in Texas to northern Alberta?
"I've done a lot of work before where we look at either something occurring at a very small scale—grazing or fire effects—up to a larger scale, a landscape or collection of landscapes," says Allred, now an assistant professor of rangeland ecology at the University of Montana. "But this was a conceptual or intellectual challenge, being able to think of these cumulative effects."
The answer surprised not only Allred but several fellow scientists from a number of universities. Last month, Allred co-authored a two-page study in Science magazine with Steven Running, Montana State University's Julia Haggerty and others examining the extent of ecosystem services lost to oil and gas development. Their research revealed, for starters, that new oil- and gas-related well pads, roads and storage facilities built between 2000 and 2012 occupied a land area equivalent to three Yellowstone National Parks.
The study, which relied heavily on high-resolution satellite data and detailed information on industry infrastructure, foregoes the more typical region-based analysis in favor of a much broader geographic investigation. While the findings indicate widespread ecological effects like severed migratory pathways, altered wildlife behavior and high water consumption by fracking in already water-stressed areas, they also reveal the resource boom across much of the U.S. and Canada has resulted in severe losses in vegetation. To put this in perspective, Allred says, researchers translated losses in net primary production (the amount of carbon accumulated in plants as biomass) into more relatable terms. In the case of rangeland, vegetation loss amounts were equivalent to roughly "five million animal unit months," the study reads; in croplands, the equivalent was upwards of "120.2 million bushels of wheat."
"We wanted to put it into numbers that people could understand and relate it to other land uses out there," Allred says.
The study makes a compelling case for policymakers and land-use planners to consider the broader cumulative impacts of permitted development. Lack of such insight in the 1930s, the study reads, contributed heavily to the Dust Bowl. For Allred, the technological advancements that made the study possible can foster increased understanding of how local decisions affect the big picture.
"We need to zoom out and look at it as a bigger picture so we can make these informed decisions," he says, "so we can decide what trade-offs we are comfortable with."