Geospatial data analysis by its very nature lends itself to solving Big Problems: drought mitigation aimed at preventing mass starvation, national defense surveillance to prevent invasion. Urban planning to make cities more livable, and sustainable. High on the list of consequential challenges is environmental protection, use of data analysis to prevent water and air pollution – particularly at the mining and oil extraction sites that have been responsible for some of the world’s worst environmental disasters.
Sometimes, though, geospatial data analysis shifts to event postmortem, rather than preventive planning, to provide insight into the missing steps (and missteps) that better planning could resolve the next time.
One of the greatest environmental threats from mining and other resource extraction ventures are the hundreds – perhaps thousands – of so-called “tailings ponds” adjacent to minerals and metals mines and oil sands facilities around the world. These facilities are artificial ponds (hence the term) built to contain toxic waste combined with water, from mining and oil sands extraction. From the sky, most appear benign, and untrained eyes can easily mistake one for a natural body of water. Until something goes wrong.
On the morning of August 4, 2014, approximately 60 billion gallons of toxic waste from an open pit gold and copper mine in British Columbia, Canada began pouring into the nearby Polly Lake, Hazeltine Creek, and Quesnel Lake, home to local trout and salmon, when the dam holding it in place failed. The wastewater consisted of 326 tons of nickel, more than 400 tons of arsenic, 177 tons of lead, and other toxic material, all of which threatened the drinking water for a nearby community and local wildlife.
The investigation of the disaster later determined that mining engineers had failed to take note of the sloped glacial lake, partly consisting of silt, they chose as the location for the four square kilometer tailings pond, and that both the shape and composition of the site led to structural deficiencies and the ultimate failure of the containment dam. A news report of the following investigation noted, “One independent geotechnical engineer described the location and design of the tailings pond as loading a gun and pulling the trigger.”
A more thorough review of available GIS data – detailed annotation and analysis – might have alerted planners to the inherent dangers, though there has never been a full accounting of the events leading up to the disaster. Even more troubling are indications that the disaster was not caused by isolated and rare conditions.
A 2019 investigation by Reuters found that more than a third of the world’s 1,700 known tailings dams are at high risk of causing catastrophic damage to nearby communities if they crumble. A tailings pond dam collapse on Brumadinho, Minas Gerais, Brazil in 2019 resulted in 259 deaths – and is the human disaster that mining industry executives cite most often when they discuss both public and government regulatory imperatives for more careful planning.
Monitoring all the known sites, examining the geospatial data that might reveal higher risk facilities – even identifying and cataloging the unknown facilities – remains a daunting task. One Canadian researcher using GIS data and statistical analysis to catalog Canadian tailings ponds across the country, points out the degree of difficulty: “We’re getting some data from mining companies, (but for the rest) we just don’t know what’s out there. This is all about risk assessment, and right now the risks are high.”
