In 1995, Daniel J. Miller, a geologist based in Seattle, made his first trip an hour north to the rural town of Oso to study how clear-cutting the foothills of the Cascades might create conditions for landslides. There was concern that timber harvests were removing a critical buffer that absorbed heavy rainfall. If the terrain became overly saturated, it was feared, the loosened dirt would be more likely to slide. In 1999, he submitted a report to the Army Corps of Engineers about the threat. Miller warned that the effects of logging and river erosion had the potential to cause a "large, catastrophic failure" on a steep hill across the Stillaguamish River from Steelhead Haven, a close-knit community of outdoor enthusiasts, retirees, young families, and former city folk drawn to the area's mountainous splendor. "It's a nice place to live out there," says Miller. "The signs of danger would have only been apparent to a geologist."
This year, on March 22, a half-mile-wide portion of the hillside – 15 million cubic yards of earth, equal to 3 million dump-truck loads – suddenly gave way and engulfed the 49 homes in Steelhead Haven, killing more than 40 residents.
The scale of the devastation at Oso may have been unprecedented, but a number of factors, especially severe weather and demographic shifts, could make similar events more commonplace nationwide. "As people move into landslide-prone country and spend more time there, the potential increases for similar [incidents]," says Jonathan Godt, a geologist with the United States Geological Survey (USGS). In 2013, a massive mudslide and flash flood killed one man and closed a highway in Manitou Springs, Colorado. In the weeks after Oso, a "slow-motion" landslide in Jackson, Wyoming, falling a few inches every day, forced dozens of families to evacuate.
Scientists now have a number of new tools to forecast these types of disaster. Using data from plane-mounted lasers called "lidar" (which essentially take a giant X-ray of the landscape), and cutting-edge computer mapping tools that can account for variables like slope steepness, soil content, rainfall, erosion, and nearby development, they can pinpoint with uncanny accuracy where and when precarious geology is likely to shift.
Miller wants to put this kind of information into the hands of homeowners. His new company, TerrainWorks, is developing software that shows potential risks for specific locations – not just landslides but fire and flood hazards, as well. "We've got the technology now," says Miller. "We're trying to design something that would be more accessible to nonscientists and nonengineers."
In Oso, Miller had estimated, a huge amount of sand, silt, and clay above Steelhead Haven was liable to collapse, and a flood-management plan, commissioned by Snohomish County in 2004, recommended the use of eminent domain to get the community out of harm's way. Still, the county continued to issue home-building permits. Two years later, a massive slide dammed the river, though the run-out never reached Steelhead Haven. When Miller spoke at a town meeting to discourage further development in the area, at least one attendee accused him of "trying to take our land." Millions of dollars were spent on shoring up the hill, but officials failed to notify new residents of the threat. Seven of the Steelhead Haven houses destroyed on March 22 were built after the 2006 collapse. "We had no idea it could reach us," Debby Rule, who bought a home on Steelhead Drive in 2007, said to NBC News. "Why did they even authorize people to build homes there?"
Two days after the collapse, Snohomish County emergency management director John Pennington told reporters, "This was a completely unforeseen slide. This came out of nowhere." But Miller had foreseen it. Unfortunately, his report was filed away and forgotten until reporters from the Seattle Times dug it up after the slide. "We need a centralized way of making all this information easily available," says Miller. "Ideally, we would have a way of getting people to realize on their own that they shouldn't be living in these places."
Landslides kill an average of 25 to 50 people in the U.S. every year and cause nearly $5 billion in property damage. In the past three decades, 13 million housing units have been built in areas prone to landslides. Nationwide, about 84 million people – some 26 percent of the U.S. population – may live in hazardous areas. The only national map of landslide danger was drawn in 1982, and geologists now dismiss it as little more than a cartoon. In 2003, the USGS proposed a national plan to update it, which would have relied on the kind of technologies used by TerrainWorks, at an estimated cost of $25 million, but Congress only budgeted $3.5 million for the effort. As a result, our knowledge of the geological threats around us is woefully obsolete.
While all 50 states experience landslides, there's no unified approach for dealing with the issue. California has strict regulations on building codes and reporting risks. Oregon and Washington are both working on a landslide-alert system. Pennsylvania is the only state using lidar to fully catalog its geography. At the other extreme, North Carolina, with a deadly hazard in its western peaks, has no system to map where landslides are most likely to occur: Its program was scrapped when the Republican-led statehouse – backed by the real estate lobby – laid the geologists off. Other high-risk states that have scaled back mapping efforts include Nevada, Colorado, and Georgia.
"It's enormously political," says Lynn Highland, a geographer at the USGS Landslide Hazards Program. "Every state has an idea of how they want to track their landslides, and they're all different."
Learning where these events will probably occur requires extensive – and expensive – research, but it could also save lives. For instance, lidar images taken around the Stillaguamish River after the slide showed that something similar in magnitude occurred hundreds of years ago just downstream from Steelhead Haven – a hint that the tragedy had some precedent. "We didn't know that," says Miller. "Ideally we'd have a system that allowed this kind of information to get funneled right to the county."
Even with that information in hand, however, local officials and property owners need to be responsive to the dangers of building in certain locations. "People don't want to know about it, or they think it reduces property values if they're in a known hazardous area," says Highland. "So that's where you get into land use and private property rights. And that's a big issue."
For now, landslides smaller than Oso will continue to take lives and destroy property across the country, largely unnoticed, until the next big one. "People like to build their houses on these bluffs overlooking Puget Sound, for example, and all of those bluffs are eroding away," says Miller. "All of those houses are eventually going to fall, hundreds of feet sometimes, down to the shore." If the public were more aware of how freely decision-makers issue permits, Miller says, "maybe we would make more of an outcry about having developments in all of these places."