Near-record rainfall gave Oso mudslide its destructive force, USGS finds

OSO — If the slope that collapsed during the deadly Oso mudslide had been slightly drier, it probably wouldn’t have caused anywhere near as much damage.

A new report published Jan. 8, written by U.S. Geological Survey scientists, says the amount of water in the soil is what gave the slide its power and violence.

The slide flowed down and across the North Fork Stillaguamish River valley. It wiped out a neighborhood and killed 43 people.

Liquefaction, a process in which compression of the saturated soil causes it to lose its coherence and flow like a liquid, gave the slide its unusual mobility.

Most significantly, mathematical modeling showed that had the soil been slightly less saturated, the base of the slide might not have liquefied at all, and the runout would not have inflicted anywhere near as much damage to property and lives.

“That is the big takeaway, that the mobility of the landslide was very large, not necessarily unprecedented, but very large,” said Richard Iverson, the lead author of the report and a senior research hydrologist at the USGS’s Cascades Volcano Observatory in Vancouver.

“The model results indicate that, for example, had the water content been about 5 percent smaller, that it might have been far less mobile,” Iverson said.

“That alternative case we simulated only travels about 100 meters. In that case the landslide would have crossed the river but wouldn’t have done much more than that,” he added.

That would have been similar to other slides at that location in modern times, including the most-recent in 2006.

The Oso slide occurred on what was a relatively dry day at the end of one of the wettest winters on record.

Weather readings taken at the Darrington Ranger Station for 180 days prior to March 22 showed that the period was wetter than 91 percent of similar periods in the past, going back 86 years, the USGS report said.

Iverson and his coauthors relied on a number of tools and methods. These included comparing Oso to other high-mobility landslides around the world, lidar and GIS mapping data of the area both before and after the slide and seismic measurements from 18 networked regional monitoring stations.

The team used mathematical modeling to analyze alternative scenarios, and supplemented their data with geological field work and eyewitness accounts.

The seismic measurements in particular support a different conclusion about the chronology of events during the slide compared with a scenario outlined in a report released in July 2014 by GEER. That team of geologists and engineers had a mission to collect raw data as quickly as possible after the slide for rapid distribution to the scientific community and the public.

The new interpretation of the seismic data shows an initial shock wave at 10:37 a.m. consistent with a large, slow-moving landslide, similar to the 2006 slide. *

The initial debris didn’t travel, but it undermined the slope above, which collapsed about 50 seconds later. That mass compressed the soils down below in the base of the slide, which then liquefied and rushed out across the North Fork Stillaguamish River and into the Steelhead Haven neighborhood.

A secondary shock wave, coming about five minutes later, didn’t have the wavelengths associated with the main slide, and is therefore thought to be just a further collapse of part of the headscarp, said Kate Allstadt, a seismologist working at the USGS on a National Science Foundation fellowship and one of 14 listed co-authors of the new report.

Previous reports postulated that the second shock wave was also part of the main slide activity.

“We think that’s the source of the deposits in the secondary debris fall,” Allstadt said, referring to pile of debris that came to rest high up the slope.

One unnamed eyewitness account in the new report corroborated the main debris flow roaring across the river less than a minute after the initial collapse.

“There were no eyewitness reports of anything big happening a few minutes later,” Allstadt said.

The new report’s focus on water was reinforced by comparisons to other landslides. The 2006 landslide on the same slope, as well as older slides, hadn’t exhibited unusual mobility.

Slides elsewhere that had large runouts typically originated on slopes of greater than 30 degrees, whereas the slope at Oso was less than 20 degrees when it collapsed.

That was a surprise to Iverson when he first made the trip to Oso and saw the relatively shallow, 600-foot-high Hazel hill that was the source of the slide.

“I was expecting this to have arisen from one of the big mountainsides up there,” Iverson said. “One of the surprises was it did come from a site that had so little potential energy.”

Another finding was that, of the approximately 10.8 million cubic yards of earth that moved during the slide, less than half that amount moved beyond the slide’s original footprint on the north shore of the river.

The initial wave of watery debris to cross the floodplain probably only amounted to 261,590 cubic yards, or less than 3 percent of the total volume of material in the slide.

The report suggests that the leading edge of debris might have gotten more energy from incorporating about 65,000 cubic yards of water from the river, which could have contributed to further scouring and liquefaction in the flood plain.

Comparisons with past landslides elsewhere reveal that the 2014 Oso slide, while having slightly less energy than other slides of its size, had a runout area about 50 percent larger than what would have been expected for a rock or debris avalanche of the same volume.

“That’s why the liquefaction is such a critical part of the story,” Iverson said.

But the extensive runout across the flat flood plain of the valley also makes the Oso slide an outlier in the historical record, posing a challenge to research, said Jonathan Godt, director for the USGS Landslide Hazards Program.

Being able to predict large slides requires a better understanding of how they’ve moved over thousands of years, and records over a few decades may not be particularly relevant to an event the size of Oso, he said.

“All the long-runout landslides we know of, there’s really no written down historical information. You’re really looking at the geologic record,” Godt said.

Assessing the risk a slide poses to people and property depends no underground soil conditions, the report says. Gathering that data is a significant challenge.

While identifying a hazard is comparatively easy, Iverson said, quantifying the risk the hazard poses requires information that is usually not available.

“If we had a three-dimensional picture of the distribution of pore space in the ground and also had a three-dimensional picture of evolving water pressures in the ground, we would have, I think, a significant portion of the information we need to make an accurate forecast,” he added.

“It sounds deceptively simple, but it’s very difficult information to come by,” Iverson said.

Chris Winters: 425-374-4165; cwinters@heraldnet.com. Twitter: @Chris_At_Herald.

* This article has been corrected since it was first posted to accurately state scientists’ description of the initial shock wave.

Oso slide report available online

The report, “Landslide mobility and hazards: implications of the 2014 Oso disaster,” was published online Jan. 9 by the peer-reviewed journal Earth and Planetary Science Letters and will be in the Feb. 15 print edition.

It can be read online at sciencedirect.com/science/article/pii/S0012821X1400781X.