Sinkholes appear seemingly out of nowhere, snarling traffic and causing costly repairs. But what if they could be predicted and, perhaps, mitigated?

Scott Acton, a professor of electrical and computer engineering in the University of Virginia's School of Engineering and Applied Science, is leading a federally funded $870,000 project to develop a suite of software tools that will enable transportation engineers to mitigate the road and bridge defects that lead to sinkholes.

U.Va. is partnering with the Virginia Center for Transportation Innovation and Research, which is headquartered  at U.Va., and Vancouver-based TRE Canada to develop the software that will soon be commercially available to state departments of transportation across the United States. TRE Canada specializes in using satellites to determine changes in elevation on the ground.

The project uses remote sensing technology that takes repeated satellite images of the earth's elevation. The sensors can detect changes in elevation down to a quarter of an inch. The technology combines synthetic aperture radar supplied by TRE Canada and novel image-analysis algorithms developed by Acton's lab.

Two graduate students, Andrea Vaccari and Mike Stuecheli, are developing the software that will highlight regions of potential risk. Their software will interface with popular display methods, such as typical geographical information systems and Google Maps.

The project initially will focus on a region of the Interstate 81 corridor plagued by sinkholes, while researchers develop the software for analyzing the images.

"This is the first time we've been able to predict sinkhole collapse," Acton said. "If we determine that we can use remotely sensed data to detect and monitor the subsidence that precedes sinkholes, bridge settlement and landslides, we can potentially save millions in highway repairs, reduce highway closures and enhance the safety of travelers."