Finding cause of Oso mudslide will take time and science
Mark Mulligan / The Herald
Since the earliest days of the Oso mudslide, surveyors like Carl Hagaman have kept measurements of targets along the ridge line and on debris at the site.
Mark Mulligan / The Herald
Carl Hagaman locates a surveying target that is located high on the edge of the ridge line above the Oso mudslide area Thursday.
Mark Mulligan / The Herald
During a visit to the mudslide site Thursday, Dale Topham, supervisor of Snohomish County’s geotechnical group, picks apart a large lacustrine deposit, the material that formed the bed of an ancient lake that filled the Stillaguamish Valley during the last glacial period. Now chunks of the material are spread throughout the site of the mudslide.
Mark Mulligan / The Herald
Horizontal striations on the slide show various layers of clay along with glacial till.
Mark Mulligan / The Herald
Carl Hagaman peers at a survey target.
Birds tweeted in the trees, a breeze blew through the leaves and brush. A rooster crowed. The North Fork Stillaguamish River softly rushed through the valley.
The landscape, however, consisted of a square mile of upended terrain left by the March 22 mudslide that wiped out the Steelhead Haven community, killing 41 people and leaving two missing and presumed buried under tons of earth and debris.
On a small hill on the south edge of the slide area sat a survey station. Dale Topham, the supervisor of Snohomish County's geotechnical engineering group, and surveyor party chief Carl Hagaman last week were taking measurements of several “targets,” two-foot-square markers, each with a bright red “X,” that were placed on the edges of the new ridge line and down below on piles of debris.
“Looks like Seven is moving around a lot,” Hagaman said. “Every time I look at it there's a new branch in front of it.”
Target seven was on a tree about halfway up the northeastern edge of the slide. The handheld device that communicates with the survey gun mounted on a tripod showed that the target had moved 5.3 feet to the south and 3.1 feet to the west.
On the other hand, targets at the very top of the slope, three-quarters of a mile away and barely visible, had only moved a couple thousandths of a foot since the last measurements were taken a few days earlier.
Dark, vertical striations in the exposed slide showed where soil still sloughs down. Occasionally, Topham said, a larger chunk of earth calves off the scarp and tumbles down.
“You can hear it from here,” Hagaman said.
Work at the mudslide has entered a new phase. With the active search for victims finished, scientists now are working to find out what happened that Saturday morning when the mountain came down.
It will take time for researchers to fully understand what triggered the slide above Steelhead Haven, which sent a deadly torrent of slurry across the valley and up the slopes on the far side.
In coming months, teams of geologists, hydrologists, geomorphologists, engineers and other researchers will visit the site and observe, measure, record and analyze.
Then they'll come back and do it again and again.
The challenge then becomes making that abstract data accessible and understandable to the public, some of whom are wondering why they weren't warned, to government planners reconsidering land-use policies, and to politicians who are being asked to account for decisions made over decades.
Geologists have known for years that the hillside that came down March 22 was unstable. A 1999 draft report prepared for the Army Corps of Engineers points out that in the past century, there have been periods of alternating heavy slide activity and dormancy on the hillside.
Slides at Hazel partially blocked the North Fork Stillaguamish River in 1951, 1967, and 2006, and there have been several smaller events over the years.
“This kind of landslide is a tough one to categorize,” said Daniel Miller, the geologist who wrote the 1999 draft report.
“No one was able to anticipate the kind of runout that it would have,” Miller said.
His report suggested that the slope had the “potential for a large catastrophic failure,” which could involve runout of 880 feet, similar to what happened in the 1967 slide.
But that was based on assumptions such as an estimate that a slide would consist of 7,095 cubic yards of soil. The March 22 slide was estimated to have displaced 10 million cubic yards.
Miller's draft report made clear that more study was needed to make accurate predictions of the slide's behavior.
“I currently have no basis for estimating the probable rate or timing of future landslide activity,” Miller wrote.
The fact debris ran out a mile across the floor of the valley was a major surprise even to scientists who specialize in debris-flow analysis.
Richard Iverson, a senior research hydrologist at the U.S. Geological Survey's Cascades Volcano Observatory in Vancouver, Wash., has been building mathematical models of debris flows for 20 years.
The data that have gone into that modeling have come from other landslides around the state, especially those triggered by the 1980 eruption of Mount St. Helens, which Iverson said showed some similar behaviors to the Oso slide in that it ran fast and far.
“It's no surprise to any geologist that that slope failed. The surprise was what it did when it did fail,” he said.
The big question Iverson hopes to answer concerns the mobility of landslides, and he suspects the answer lies partly in understanding the role played by the geological process known as liquefaction, in which soil loses its solid consistency and behaves like a liquid.
Liquefaction is most commonly associated with earthquakes and is caused by seismic waves causing the soil to vibrate and collapse under its own weight.
In landslides, soil saturation during extreme weather plays a role in creating conditions ripe for a slide, but it's gravity that pulls down the hill and gives the slide its energy.
Identifying those conditions and pinpointing the threshold at which liquefaction occurs will require studying the amount of water that was already in the soil, the depth at which the liquefaction occurred at Oso and the characteristics of the surrounding terrain and the river that influenced the debris flow once it started moving.
Some of this data can be obtained from surface measurements, but getting the full picture will require gathering as much information as possible, including digging in the dirt, what Iverson called “good old-fashioned field geology.”
The USGS's technical understanding of slide behavior has already had real-world effect in Oso.
The agency has spent years not just building complex mathematical models but testing how different material flows at a facility in central Oregon where it operates a 300-foot-long slide flume.
Repeated tests on different materials, and watching how they behave and mix during a slide, provided knowledge that ultimately led rescuers to search for victims of the slide near the edges of the zone and not where their homes used to be, Iverson said.
When Iverson visited the slide zone during the first week after the event, he saw affirmation that lab work and modeling could yield real-world results.
Remote-sensing imagery of the terrain before and after the slide enabled the USGS to add new data to the mathematical model.
“Within a couple days, the model results were also being used in the search efforts,” Iverson said.
Down on the floor of the valley, Dale Topham pointed out thick, gray blocks of claylike soil.
They are lacustrine deposits, he said, the material that formed the bed of an ancient lake that filled the Stillaguamish Valley during the last glacial period.
On the slope of the slide, horizontal striation shows where those clay layers were buried. Now they're tossed around with the tumbled hillocks of sandier glacial till in a seemingly random pattern on the valley floor, mixed in with broken trees and pieces of cars, houses and personal items.
Up the slopes and on top of the bluff, the USGS and the transportation department have installed remote monitoring equipment that can track minute movements in the earth.
Around the edges of the slide zone, standing trees show the level where liquid gray mud washed up before settling back to the floor.
“Where you see how high it's splashed in some areas, it's very impressive. It witnesses the fluidity of the flow,” Topham said.
Researchers will be looking at all of those things this summer. Some of it, such as the mud splashes on the tree trunks, will need to be documented before it washes away. Grass, and later trees, will start growing in the new terrain, obscuring other details of the slide.
One team of researchers from the Geotechnical Extreme Events Reconnaissance (GEER) Association, will soon start to document that evidence before it gets washed away or covered by vegetation.
GEER sends in a rapid-response team that doesn't design studies of its own. It gathers data to share with other scientists.
“The mission is to collect data and give it away,” said David Montgomery, a professor of earth and space sciences at the University of Washington who was asked to join the team investigating the Oso slide.
Another team of scientists from the University of Illinois at Urbana-Champaign's Department of Civil and Environmental Engineering is expected to arrive later this month. That group studies manmade and natural disasters with a goal of producing data that can be used in public policy.
One question that team would like to answer is whether the size of the Oso slide could have been predicted at all.
“If we could have predicted the size and extent, what would have been needed beforehand to predict the size and extent of the slide?” said Timothy Stark, the Illinois team's lead researcher.
An ultimate analysis
In the end, “we would like to have a comprehensive analysis of the slide,” said Steve Thomsen, public works director for Snohomish County.
Doing the field work to produce that report might cost $2 million to $3 million, he said, noting that a source of money hasn't been identified yet.
“Land-use decisions wouldn't be a part of the report, but it could inform policy decisions down the road,” Thomsen said.
At the same time, a team of researchers from the Federal Emergency Management Agency and the U.S. Army Corps of Engineers, and another from the state Department of Transportation, have been working to assess the danger of flooding to people still living and working in the area and to the highway.
Miller, the geologist who studied the slide area earlier, hopes research in Oso will help create a new understanding of landslide risks throughout Washington.
Thomsen pointed out that while hydrologists have developed models for 25-year and 100-year floods, there is no similar rating system for landslides.
Doing scientific work in the debris field will be challenging, not just because of the dynamic nature of the slide zone and the river but because of reminders — of lives lost, the persistent threat of flooding, even further disintegration of the slope.
“It's the old analogy of trying to repair the car while driving down the road,” Thomsen said.
It's complicated, there are many moving parts, but for scientists, the slide is also an opportunity to advance knowledge in their fields.
“The scientists are really hungry for data points,” he said.
Chris Winters: 425-374-4165; firstname.lastname@example.org.
- 43 trees mark lives lost along Highway 530 in Oso 9/28/14
- Oso shows its heart during ceremony honoring slide victims 9/22/14
- Oso survivor rebuilding his life six months after disaster hit 9/21/14
- Highway 530 will be back up to speed Monday, if not sooner 9/19/14
- 43 trees to honor those killed in Oso mudslide 9/17/14
- $75K grant will help rebuild trail wiped out by Oso slide 9/16/14
- County gets $1.5 million grant for mudslide buyouts 9/15/14
- Darrington schools get federal grant to pay for mudslide expenses 9/14/14
Our new comment system is not supported in IE 7. Please upgrade your browser here.