Feedback
Science
Deadly Mudslide

Double-Barreled Blow: How the Oso Mudslide Went Down

Image: Seismic activity

Seismic readings from the Pacific Northwest Seismic Network show the two bursts of activity caused by the collapse of a hillside near Oso, Wash., on March 22. PNSN

Almost a week after the catastrophic landslide in Washington state, geologists are filling out their timeline for the double-barreled collapse β€” and mapping out how previous landslides extensively changed the surrounding landscape.

Mud and debris swept down a rain-saturated slope near the town of Oso, Wash., on March 22 β€” crossing the North Fork of the Stillaguamish River and burying homes and other buildings on the other side. Scores of people are dead or missing.

Readings from the Pacific Northwest Seismic Network reveal that the hillside collapsed in two stages: The lower part of the slope gave way at 10:37 a.m. PT, triggering seismic activity that lasted two and a half minutes. That material could have been disturbed and weakened by a landslide that occurred in 2006, Kate Allstadt writes in PNSN's Seismo Blog.

3:04

The first collapse let loose a surprisingly fast wave of debris that probably caused most of the damage and deaths, Durham University's Dave Petley writes in the American Geophysical Union's Landslide Blog. It also destabilized the upper part of the slope, which collapsed four minutes later and set off a smaller seismic shock. Smaller landslides continued for days afterward.

Was there a seismic trigger for the mudslide? The U.S. Geological Survey says there were no earthquakes in the area on March 22. A few days after the tragedy, a county emergency official suggested that a magnitude-1.1 earthquake on March 10 could have played a role, but seismologists say that wasn't strong enough or close enough to cause a collapse 12 days later.

However, the University of Washington's Bill Steele said the earlier quake could have been "indicative of deformation going on in the hillsides around that area." The area's recent rains played a role as well. The glacial moraine and sand in the soil "can form a slick surface when saturated," he noted.

"It's a nightmare for landslides in this area," he told NBC News.

That much is clear in a fresh series of maps issued by the USGS, based on elevation readings that were made using a laser-scanning technology known as lidar ("light" plus "radar").

Image: Slide zones
A USGS map based on interpretation of lidar data shows landslide zones in the area surrounding the March 22 slide, which is indicated by the red cross-hatched area. The zones are color-coded to indicate relative age. Red zones (also marked "A") are youngest, orange "B" zones are older, brown "C" zones are even older, and yellow "D" zones are the oldest. All of the indicated slide zones are younger than 14,000 years, when an ice sheet is thought to have retreated from the area. USGS

Computer-processed lidar observations can digitally strip away the vegetation from a landscape, revealing the terrain that lies beneath. The USGS lidar map, based on an interpretation of data collected during a 2013 aerial survey, shows a patchwork of previous collapses β€” including a landslide zone to the west that is substantially larger than the square-mile area that was devastated last week.

The USGS acknowledges that the lidar readings probably don't tell the whole story.

"The actual extent of landslide deposits is probably greater than shown," USGS research geologist Ralph Haugerud wrote.

Update for 10:15 p.m. ET March 28: I originally wrote that last week's mudslide was the only one that crossed the river, but reader Brian Maschhoff says that's not correct. Past landslides have shifted the course of the Stillaguamish, as can be seen in historical imagery. "This river has moved around a lot, and has seemingly moved mountains as well," Maschhoff writes.