National Geographic News
A photo of crews cleaning up the landslide in Oso, Washington.

The site of the March 22 Oso mudslide shows the extent of the destruction.

Photograph by Ellen M. Banner, The Seattle Times via AP


NG Staff

Warren Cornwall

for National Geographic

Published July 22, 2014

A deadly landslide that killed 44 and obliterated a riverside neighborhood in Washington state last March was fueled by rain soaking the site of an eight-year-old landslide, while logging in the area may have also played a role, according to a scientific report released Tuesday.

The most detailed published scientific account of the mudslide, which was the deadliest in U.S. history, suggests that the disaster was years in the making in a valley with a history of huge landslides dating back thousands of years.

"That landslide mass prior to March 22 was really in some sense poised to fail. It was really unstable," says Joseph Wartman, a University of Washington engineer and a lead author of the report.

The report doesn't offer a definitive explanation for why the mountainside collapsed on that day. But it describes a devastating chain reaction sparked by rain and groundwater on a hillside left unsettled by years of smaller slides.

The report comes from a team of university and private-sector researchers who are part of the Geotechnical Extreme Events Reconnaissance (GEER), a National Science Foundation-funded initiative to quickly dispatch scientists to evaluate natural disasters.

The slide plowed through part of the town of Oso, in the Cascade Mountain foothills northwest of Seattle, shortly after 10:30 a.m. on Saturday, March 22. It killed 43 people and demolished sections of State Highway 530.

The new report raises questions about how local governments take into account risks of this sort, and how the newest, best science finds its way to the local level. The hillside that collapsed had been the scene of multiple smaller slides since the first housing development was built there in the 1960s. Government agencies repeatedly attempted to buttress the hillside, though engineers warned it might do little good.

At the time of the slide, scientists were shocked at its speed and the distance it traveled—more than half a mile (over a kilometer). Yet recent images of the valley taken using LIDAR, a laser-based technology that can "see" through vegetation to the rock below, revealed multiple ancient landslides, including one more massive than the Oso slide. The GEER team also found the landslide's magnitude wasn't unusual for slides of this type.

"I don't think it was fully comprehended how far out that landslide could have run. So I don't think that even local officials understood the gravity of that," Wartman says.

The incident also underscores the importance of focusing on the potential damage of a natural disaster when government officials make decisions, the scientists say. Historically, governments have tried to assess such hazards but haven't always focused on places where the consequences would be most dire.

"The very first thing is to communicate very clearly what the risks are," says Robert Gilbert, a civil engineer at the University of Texas in Austin. "Not saying at one meeting every decade that 'Hey, there's a risk of landslides.' But constantly and persistently communicating what the risks are so that people are making decisions in an informed way."

The study could become fodder for legal claims already filed by survivors and families of some of the victims against Snohomish County and the state of Washington. Those focus on the state's role in regulating logging near the slide area, and the county's part in permitting homes to be built there. The county recently imposed a six-month ban on construction in the area.

The new report found that a 2006 landslide likely set the stage for the much bigger slide, creating a loose bed of rocks and soil that soaked up water more readily and was more prone to failing. Then heavy rains in the weeks leading up to the disaster drenched the region.

The slide was really two interconnected events, the report says. First, the 2006 landslide debris liquefied, a phenomenon in which solid earth becomes a liquid as the water pressure in the soil increases, pushing dirt particles apart. That then sent a mass of mud and debris shooting across the valley, removing support for a large chunk of the mountain, which fell away and added to the slide.

In the tragedy's aftermath, suspicion fell on recent logging on the mountain just above the slide and erosion along the Stillaguamish River, which flows along the base of the slide.

The new report suggests there was no single cause. "Our work did not conclude that only one path could have led to the Stage 1 failure," it says, referring to the first stage of the slide.

While the report doesn't definitively point a finger at the timber industry, it suggests that logging above the slide area might have changed the way rain soaked into the hillside, adding more water to the unstable slope.

"It is possible that in 2014, the location, size and maturity of growth in these [logging] tracts was such that groundwater discharge to the slope was greater in 2014," the report states.

Getting a clearer picture of logging's role would take more intensive analysis and modeling of the groundwater flow and forestry practices there, said David Montgomery, a University of Washington geomorphologist and member of the GEER team.

The GEER scientists largely discounted river erosion as a major factor in the current slide. Someone had inspected the area just days before the slide and reported seeing little evidence of erosion along a log wall built to shield the hillside from the river, Wartman says.

But it's possible that even a little erosion could provide a final nudge on a hill already poised to collapse, Montgomery says.

Follow Warren Cornwall on Twitter.

Updated 7/21/14 at 6:45 p.m.

yen tran
yen tran

the landslides are really dangerous, annually there are hundreds of people in the world die due to landslides everywhere Baixar facebook movel

Carter Fox Jr.
Carter Fox Jr.

Summation of a mudslide: Wet mud on side of a mountain! Warming temparatures, moisture in soil, humidity and gravity = mudslide


Steve Pipkin
Steve Pipkin

I am not a soil scientist. I merely worked in a technical position at a Geotechnical Engineering firm for three years.

Here's the skinny as I understand it: The geology of the mountain regions of western Washington is largely volcanic bedrock often overlain with LOTS of glacial material. As ice dams melted at the end of ice ages this glacial material got carried downstream by huge floods and was deposited all over the place.

This glacial material is soil, sand, gravel and rock. It is not solid. Technically it is a colloid composed of disparately sized particles. If it gets wet enough it acts like a liquid instead of a solid. It flows toward the path of least resistance - downhill. It's called a slope failure. I point them out to my wife when we're driving through road cuts. She rolls her eyes a lot.

Engineers perform slope stability analyses to ensure that engineered slopes do not fail. Mother nature does not. So beware if you build in areas where there are steep alluvial slopes.

As to the potential of logging practices influencing the liquification of the soil, I can only speculate. As a botanist (I have a BS degree) I believe that the maintenance of a sound forest ecosystem helps lower discharge of excess water. If the forest above the failed area was compromised by logging then the amount of water coming into the unstable slope area could be greater. That said, I have seen fully treed slopes with slope failure. From what I've witnessed, a slope of alluvium (and colluvium) that is too steep will fail if it gets too wet regardless of the condition of the fauna. The question is, would it have gotten wet enough without the addition of extra water from logged areas above the failure?

Brent Foust
Brent Foust

Here we go. I read the report and mostly speculation.  Lets blame logging!! I wish that National Geographic wasn't so politically motivated when publishing these reports.  It could have been caused by a lot of variables, but lets headline with LOGGING CAUSED THE SILDE THAT KILLED EVERYONE!!!.  Shame on you Cornhole!

The Commenter
The Commenter

@Chris Crawford: Hmm, if there was no intent to call attention to logging then there was no reason to subhead with that at all when there were other contributing factors. "The lady doth protest too much, methinks." Perhaps this is a bit above third grade level, though, and participation in a high school level class covering critical thinking skills would be more appropriate for you.

Chris Crawford
Chris Crawford

@Brent Foust Mr.Foust, you claim that the headline says “LOGGING CAUSED THE SILDE THAT KILLED EVERYONE!!!”. That is incorrect. It says “Causes of Deadly Washington Mudslide Revealed in Scientific Report” with the subheading “The report says that logging in the area may have played a role.”.

Now let me explain what that means. First, let's consider the phrase “may have" 

"May have" in this context means 'might have' or 'might possibly have’. It indicates uncertainty or possibility. It rules out certainty.

Next comes the phrase ‘played a role’. That indicates that the logging was one of several factors.

Thus, the headline and its subheading state only that there is a possibility that logging might have been one causal factor among several.

Perhaps a consultation with your third grade English teacher would be in order.

Swiftright Right
Swiftright Right

I don't understand why loggers don't leave like 5% of the trees standing instead of clear cutting everything (including saplings). Leaving some trees would allow the forest to grow back quicker, would help prevent and would help prevent mudslides and help prevent fires by preventing brash accumulation. You just need a few trees to cut down on the amount of think underbrush that fires tear through.

Doug Hornby
Doug Hornby

The headline reads "Causes of deadly Washington mudslide revealed.

Where are these definitive causations? I read speculations, nothing definite.

Another reason not to trust the internal headline writers. Sheesh.

Gwendolyn Mugliston
Gwendolyn Mugliston

I am very glad scientists are looking at the causes of this slide.  There are many, many places in WA and OR and northern CA where such scrutiny should be done.  I dread driving some of the roads in spring after big rains. 


Popular Stories

The Future of Food

  • Why Food Matters

    Why Food Matters

    How do we feed nine billion people by 2050, and how do we do so sustainably?

  • Download: Free iPad App

    Download: Free iPad App

    We've made our magazine's best stories about the future of food available in a free iPad app.

See more food news, photos, and videos »