Trouble Down Below

As if Florida and other coastal states didn’t have enough problems, what with frequent hurricanes, tropical storms, coastal erosion, and flooding, some of them are also being undermined from below.  Earlier this year, a sinkhole caused the evacuation of a Florida neighborhood, a result of the collapse of the underlying limestone. Any place with karst topography, in fact, can experience sinkholes. Missouri, about as far inland as you can get, has hundreds of areas of collapse, the largest affecting about 700 acres. And even though people living in these areas might be aware of the risks, it has been hard to predict exactly where or when one will occur. The phenomenon has caused least one death—in Tampa a few years ago, when a house was suddenly swallowed up—and people are sometimes forced to flee with little warning.

That might be changing, though. On the other side of the world, along the coast of the Dead Sea, researchers have devised a way to predict sinkholes years before they occur. The causes of the 6,000 or so sinkholes that have occurred there are a bit different from most of those in the US; as tributaries of the Dead Sea are diverted for human consumption, the sea level is dropping, and the coastal lands—porous like karst, and with many rock-salt formations—begin to draw in freshwater from surrounding sources. The freshwater dissolves the underground formations and leads to their collapse, along with whatever structures are on the surface above them.

Until fairly recently, the best prediction of where a sinkhole might occur came from interferometric synthetic-aperture radar, which, according to this article, can give a few weeks’ notice of the changes that signify sinkhole formation. In 2012, though, researchers began drilling holes, anywhere from 11 to 25 meters deep, in an area on a beach and installing sensors called geophones, which detect and transmit tiny tremors in the earth. The lead researcher, Meir Abelson of the Geological Survey of Israel, says he can determine which tremors are from the surface layers versus those that come from the deeper tectonic plates; most of what he’s detected come from the surface layers. With enough strategically placed sensors, his team has been able to figure out the “epicenter” of the tremors and predict where the earth is beginning to collapse, even though the collapses they’ve predicted didn’t occur until 2–4 years later.

Dr. Abelson believes that a permanent array of sensors installed in the region would mean the difference between hasty evacuation once a collapse is imminent or underway and long-term planning to relocate homes, businesses, and other infrastructure in the danger zone. “Not all of the tremors they recorded have yet been followed by the formation of sinkholes,” according to the article. “But, crucially, all the sinkholes that have formed so far have been at the epicenters of tremors.” A similar network of geophones should work in any susceptible area, such as Florida or Missouri, no matter the cause of the potential collapse.