The large, deadly landslide that struck near the town of Oso, Washington on March 22 was a slow-motion tragedy. The first few days were the worst part of itemergency responders were working without rest to find survivors, yet much of the ground was too dangerous to step on. The anxiety and frustration must have been terrible. So I grant slack to the weary manager of the county's emergency management department, who told the press, "The area was mitigated very heavily. It was considered very safe. This was a completely unforeseen slide. This came out of nowhere." Many people who don't know geology think that way. To geologists, what he said was mostly incorrect.
This is an area of large, well-mapped landslides. Washington geologist Dan McShane writes a blog, Reading the Washington Landscape, where he's been pulling up background information on the slide and its neighborhood. (I've had it in my Washington Geology resource list for some time.) The landscape has landslides written all over it. As McShane put it on the day of the slide, "Landslide wonks knew exactly where this slide was as soon as it made the news." Later he posted lidar maps of the area, showing the landscape with all trees and buildings removed. Take a look.
Outlines in red, made by McShane from county data, show some of the landslide areas. The area of last weekend's slide is the one on the right. That is to say, the latest slide was in the scar of an older one. In fact, the lower part of that older slide moved again in 2006, pushing the river southward. The area of the 2006 slide is the part that was "mitigated very heavily," meaning that the ground was drained and wreckage cleared up. Such steps can stabilize small landslide deposits, but they can't stop unstable mountainsides from collapsing over and over.
I have no doubt that the geotechnical engineers and geologists who dealt with that 2006 slide knew the overall situation perfectly well. But scientists, for better or worse, aren't in charge of things, and people determined to do what they want can't easily be made to understand the risks they're taking. Until the ideal world arrives, people need to learn on their own.
The technology that could help is easily envisioned. Sprinkle a bunch of RFID chips (or even just distinctive reflectors) on hillsides like this and map them regularly with GPS and lidar. The premonitory signs of landslides like this can be detected and warnings can be issued with good data behind them. The largest non-volcanic landslide in North American history happened just last year, inside the massive open pit of a copper mine in Utah. But because the pit was closely monitored, the workers cleared out 7 hours beforehand and not a soul was injured. The technical details were published January in the open-access journal GSA Today.
On the other hand, it would be cheaper just to keep housing away from danger zones.