Advertisement

Satellite maps reveal spread of mountaintop coal mining in Appalachia

The practice of mountaintop coal mining has been linked with a variety of environmental threats, including air and water pollution.

By Brooks Hays
Satellite maps reveal spread of mountaintop coal mining in Appalachia
Areas of yellow mark the tops of mountains that have been turned to bare earth and rubble by mountaintop coal mining. Photo by Christian Thomas/SkyTruth

July 26 (UPI) -- Coal mining is less common than it used to be, but mountaintop coal mining continues to transform Appalachia's topography.

To measure the physical scope of mountain coal mining in the region, researchers at Duke University designed a model to analyze satellite images of mountains in Kentucky, Tennessee, Virginia and West Virginia.

Advertisement

The new analysis -- detailed this week in the journal PLOS One -- suggests 1.5 million acres of mountain have been converted to bare earth and rubble by mining operations since the 1970s.

Scientists at Duke previously used satellite imagery to show mountaintop coal mining has made the region flatter, reducing the average slope of the land by 10 percent.

RELATED Land degradation threatens millions, according to first-ever land health report

The impacts of mountaintop coal mining are significant. Explosives and machinery are used to blast off the tops of mountains, expelling soil and bedrock to expose layers of coal underneath. The debris left behind by the destruction is pushed into adjacent valleys, and sometimes buried stream beds.

The practice has been linked with a variety of environmental threats, including air and water pollution.

Scientists say more accurate tracking of mountaintop coal mining activity will help monitor and mitigate the practice's environmental impacts.

Advertisement

To build their model, researchers used Google's Earth Engine cloud-computing platform to access 10,240 satellite images of the region take between 1985 and 2015. A computer algorithm scanned each image and measured for each pixel's level of greenness. Pixels that were insufficiently green and were not part of a city, road or body of water were identified as likely mining locations.

When researchers compared the latest predictions with on-the-ground data, they found the model achieved an average annual accuracy rate of 83 percent.

"It just took a matter of minutes to output the data set," Duke researcher Andrew Pericak said in a news release. "It's a huge time save."

As part of the study, researchers compared their data with coal production statistics. The analysis showed what many had predicted, mountaintop coal mining has become less efficient over time. Over time, more and more waste rock must be blasted out to access layers of coal.

"It takes more land to get the same amount of coal than it had in the past," Pericak said.

Pericak and his colleagues hope their work will prove useful to others who want to study the impacts of mountaintop coal mining. The mapping model can also be used by journalists and the public -- or anyone interested in ensuring ongoing mountain mining operations are adhering to the proper rules and regulations.

Advertisement

Latest Headlines

Advertisement
Advertisement

Follow Us

Advertisement