Sinkholes are significant, costly hazards to life and infrastructure, yet there is no national database of sinkhole occurrences or losses, said U.S. Geological Survey (USGS) research geographer and hazard expert Nathan Wood.
An oft-repeated figure is a 1997 estimate from the Federal Emergency Management Agency that put financial losses from sinkholes at $125 million per year, whereas another estimate, compiled by USGS, put them at $300 million per year. But after doing some digging, Wood and his fellow researchers discovered that Florida alone had $1.4 billion in sinkhole-related losses in only 5 years, from 2006 to 2010.
“There’s no national snapshot of what sinkholes are capturing or what sinkholes are costing the nation,” Wood said.
Wood and his colleagues recently mapped regions in the United States that are most vulnerable to sinkholes now and where sinkholes might occur more often in the future by examining areas where karst and pseudokarst predominate. They published their results in Frontiers in Earth Science.
Sinkholes are caused by many factors, explained coauthor Daniel Doctor, a research geologist and karst expert at USGS. They mostly happen because of voids in underground bedrock where water drains into but not out of closed depressions at the surface. When water flows through soil or sediments on top of these voids, the surface layer can collapse and create an open hole. Karst or pseudokarst geologies are more prone to sinkholes because water easily dissolves the carbonate minerals that make up the landforms, said Doctor.
“The surface can collapse either slowly over time or catastrophically,” he said.
There are two main types of sinkholes. The first type, called a suffosion or subsidence sinkhole, forms slowly and creates a bowl-shaped depression on the surface as water and sediments trickle downward. The second type, called a cover collapse sinkhole, forms quickly as the closed depression’s overlying soil or sediments stick together enough that, as the name indicates, collapse can happen very suddenly. “Those are the ones we’re most concerned about in terms of hazard,” said Doctor.
The new index is part of a project from the U.S. Department of the Interior that aims to build a database of hazard zones for everything from earthquakes to extreme weather.
Most Likely to Sink
The researchers found that leading sinkhole hot spots in the United States are counties in Kentucky, Tennessee, Missouri, and north central Florida, although every one of the coterminous 48 states plus Washington, D.C., have areas with a likelihood of sinkholes.
In addition to mapping sinkhole susceptibility in a karst landscape, the index also includes such data as rainfall, soil type, and intensity of development. “Even though we’re saying everything’s about karst, there really are different features, different occurrences, [even different] causes,” said Douglas Gouzie, a geologist at Missouri State University who was not involved with the study. “So you do need a national data set.”
To figure out where land is more susceptible to sinkholes, the researchers investigated their four major influences: underlying geology, soil features, amount of rainfall, and land use by humans. They used 10-meter-resolution 3D digital elevation models and other geospatial layers to map closed depressions in areas of karst and pseudokarst on a 6-kilometer grid. They mapped sinkhole susceptibility for current conditions and also created a map of potential sinkholes for the years 2070–2079.
“We wanted to show some of the major factors that come into [sinkhole formation], like the type of geology you’re on, the type of soils,” said Wood.
The researchers were limited by a lack of research on different geologies, which made it hard to predict what may happen in some areas. The Pacific Northwest, for example, has a lot of lava tubes, and how those features may react to events like extreme rainfall is unknown. “We don’t often get the opportunity to go out and investigate a sinkhole and what led up to its occurrence,” said Doctor. “There’s not a lot of information about the exact causes—it’s missing.”
Wood said the use of technologies like ground-penetrating radar to better understand geologic structure of vulnerable areas would be a valuable step forward. States could also use federal help in setting up their own sinkhole inventories, an important step in figuring out losses.
“Maybe this will help encourage that national conversation by identifying where we think there are higher-susceptibility zones, and maybe that’ll help states as they try to triage, because they don’t have unlimited resources either,” said Wood. “They might want to do more focused mapping and monitoring.”
—Danielle Beurteaux, Science Writer