High-tech sensors find gaps and opportunities in polluted site cleanup

A study by the U.S. Geological Survey and the Environmental Protection Agency tells us two things: first, that both cleanup and monitoring of the country’s most polluted sites could be improved by state-of-the-art remote sensing techniques; and second, that sites we thought we had cleaned up often aren’t so clean after all.

Among the more than 1,300 sites in the EPA’s Superfund program, 375 have been “deleted,” meaning they were cleaned up to an appropriate extent following whatever extensive pollution led them to the list in the first place. The agency certainly conducts post-cleanup monitoring at various points and tries to keep an eye on things, but it looks like it might not always work as it should.

The new study, intended to test out new techniques involving aerial spectral imaging and other ideas, focused on 16 deleted sites in Pennsylvania. The goal was to determine how useful those techniques could be for monitoring of hazardous sites, but they also seem to have uncovered some problems in post-cleanup Superfund management.

The researchers flew over the sites and imaged them using a system called ARCHER: Airborne Real-Time Curing Hyperspectral Enhanced Reconnaissance. The hyperspectral system can record reflected and emitted energy in specific wavelengths, which are then interpreted to find chemical signatures of whatever is on the ground. Combine that with imaging of the vegetation on a site—which can grow in strange patterns in polluted areas—and one can paint a picture of pollution and contamination on the ground without ever having to visit. For this study, the investigators combined the aerial views with actual site visits to do soil testing.

They found that three of the 16 sites had some problems that one would hope not to see in a “deleted” Superfund site. The “hyperspectral anomaly detection algorithm” was able to see that Hranica Landfill and Taylor Borough Dump had large amounts of debris, metal equipment, and essentially trash just laying around, and Bruin Lagoon had “vegetation stress anomalies” detected by the aerial imaging.

On site visits, they confirmed some high concentrations of lead at one site, and plenty of theoretically off-limits areas without fencing or any security. This isn’t trivial: Taylor Borough Dump, for example, is near a residential area and a park. Check out the picture up top: does it look like Taylor should have been “deleted”? Interestingly, the vegetation anomalies seen on spectral imaging at Bruin Lagoon could not be found when the investigators actually walked around the site, suggesting that some of the potential problems at Superfund sites may need the top-down view.

The group has now tested the tech at a total of 21 deleted sites, and uncovered some potential issue at five of them. Of course, this is a small sample size relative to the long list of Superfund sites, but 24 percent isn’t exactly a strong record. But sending EPA workers out to every deleted, theoretically clean Superfund site all the time is a costly proposition. The dropping prices on those fancy aerial ideas start to sound better and better.

The study authors conclude that expanding this program to more sites and using a more complete spectral imaging technology could eventually yield a program that involves little or no actual field work. “This is the ultimate monitoring scenario that would be most helpful to the EPA,” they write. Newer sensors could even be deployed on drones, perhaps dropping costs further. The idea of Superfund, of course, is to yield sites that are no longer hazardous; monitoring them remotely could help us keep them that way. - Dave Levitan | April 29 2014

Source: Slonecker ET and Fisher GB (2014). An evaluation of remote sensing technologies for the detection of fugitive contamination at selected Superfund hazardous waste sites in Pennsylvania, U.S. Geological Survey Open-File Report 2014-1081. DOI: 10.3133/ofr20141081

Image: USGS

email-signup-header

Recommended

Leave a Comment

Like-what-you're-reading-Donate2