How could this happen? Was everything done to rescue people? Did Snohomish County officials or residents know that such a threat existed? Will all the victims be recovered?
The slide was followed soon by theories and speculation, often without context, even as searchers still combed the debris field for bodies.
Here is what we know and what we don't:
What caused the slide?
We don't know exactly what triggered this slide. However, we have a good idea of the characteristics of the Hazel Slide, the name of the hill that collapsed, based on several studies conducted since the 1980s. None of those studies did the exhaustive analysis required to precisely know how the site functioned.
In the most basic terms, gravity pulled down a steeply sloped hillside of clay and waterlogged, loosely bonded soil.
We can't say exactly how much any one factor led to the slide. Some — such as soil composition and rainfall — definitely played a role. It isn't so clear for other factors, such as river erosion and logging.
The U.S. Geological Survey hopes to get some clearer answers. A USGS team is monitoring the site's stability during recovery operations and has an agreement with the county and state to analyze the slide after that mission ends, said Jonathan Godt, a geologist with the team.
March was one of the wettest on record, with Oso getting about 14.86 inches of rain by early afternoon March 22, most of which fell that day, according to the National Weather Service. The heavy rain actually prompted the National Weather Service to issue a landslide risk warning for the west side of the state just days before the slide.
River erosion is another factor for scientists to consider. For thousands of years, the restless North Fork Stillaguamish River had been eating into the foot of the Hazel Slide, weakening its stability. Authorities have tried at least twice in the past 50 years to stop the process. After a 2006 slide, crews built up the base of Hazel. Their effort appeared to be successful based on aerial photos of the site, which don't show the river carving away Hazel's foot, which ended at the river's north bank.
Scientists must also consider logging. The area to the north and west of the slide was heavily logged in the late 1980s, and a 7-acre slice above the slide was clear-cut in 2005. One of the geologists most familiar with the area, Dan Miller, said logging can't be ruled out as a factor, but at most it would be a small one. The clear-cutting in the 1980s would probably not affect the landslide area, based on analysis he and a colleague did in 1997.
However, a USGS expert told The Wall Street Journal last week that the slide was too deep to have been caused by logging.
Seismic activity earlier in the month has been ruled out by geologists as a contributing factor.
What happened March 22?
The day was notable for several hours of intense rain, which came on top of an already wet March. The waterlogged soil was too heavy for the hill to support, and at 10:37 a.m., the base gave way.
About 10 million cubic yards of material — soil, sand, trees, rocks, clay — raced across the valley floor.
"It looked like the trees were just sinking into the ground — that's how fast they were falling," said Amy Miles, a Darrington High School student who watched the slide race toward her before stopping about 20 feet away.
Uneven topography contributed to the varying depths in the debris field, which is as much as 75 feet deep in places.
Seismic data indicate two slides occurred, one at 10:37 a.m. and another at 10:41 a.m., according to the USGS.
There is mixed information about the exact size of the debris field, depending on how it is measured. In most instances, it is described as a square mile. This much is clear: It is huge. It engulfed 30 homes. Roughly a mile of Highway 530 was buried.
How many people died?
That is still unclear. As of Saturday, 30 people are confirmed as having been killed. The names of 28 victims have been released. Another 13 officially are missing. Those numbers are expected to change as search crews find more human remains and investigators develop more information about who was in Oso the morning the hillside fell.
Were there warning signs?
It doesn't appear there were warning signs readily apparent in the days leading up to the slide. People knew that the hill had slid several times in recent decades, including big slides in 1967 and 2006.
While experts had raised concerns in studies about future slides, they didn't imagine an event as big as the one March 22.
Their concerns came up in various studies commissioned by state, federal and tribal agencies. It is not yet clear how well those concerns were communicated. Over the years, flooding was a greater worry for the county and the people living there, records show.
Didn't planning codes keep people from living there?
None of the homes hit by the slide was in a designated landslide area. To trigger a special review under county building regulations, inspectors must identify a landslide area within 200 feet of a building site.
The January 2006 slide did not reach any homes — the closest was about 400 feet away. The seven homes built immediately below the slide area in the years since all were 500 to 2,300 feet from the toe of the slope.
Most of the homes destroyed March 22 were built before 1996, when Snohomish County first adopted comprehensive building regulations under the Growth Management Act. Some of the homes date from the 1930s and 1940s, according to county records.
Even if the county had wanted to clear the area for fear of landslides, officials couldn't have forced people to abandon their homes. At best, they could have attempted a voluntary buyout, something the county has done before when homes are threatened by flood waters. A geologist suggested a buyout in 2000, and the county considered the possibility in 2004. Again, the emphasis was on flood control, and the county wound up obtaining $1.9 million in federal money to buy out another neighborhood upstream that was threatened by the wandering river, county records show.
County Council members say what happened in Oso will lead to a full review of policies for building in landslide areas. They also say it's inappropriate to begin that debate while emergency workers still are looking in the debris field for victims. When that discussion begins, they'll have to wrestle with how to address risks that might not materialize for hundreds or even thousands of years.
Are other areas in danger?
Landslides are common in Western Washington, where weakly bonded soils, steep slopes and heavy rains often come together.
In Snohomish County, about 30,000 people live in areas designated as landslide hazard zones, according to a 2010 study by the county.
These slide-prone areas are mostly along the Puget Sound shoreline and steep-sided river valleys.
These hazard zones have been mapped using basic data. A more accurate view of how likely a particular slope is to slide requires much more detailed and costly analysis.
Was enough done to save lives after the slide?
Snohomish County sheriff's helicopter rescue crews were on scene within minutes. Right away, they could tell survivors likely only would be found on the south edge of the slide's path. They pulled eight to safety. A Navy helicopter rescued six more.
Much of the area was unsafe for rescuers to cover on foot.
While a statewide wildfire-type mobilization was being requested and denied, a regional incident-management team got into place by nightfall. That team supported local leaders already working the disaster, including crews from the Darrington, Oso and Arlington Rural fire districts. Firefighters and police responded from throughout Western Washington, followed by state and national resources.
Separate teams also were sent to Darrington in the days that followed, and they deployed local volunteers in the debris field doing digging and working heavy equipment such as dump trucks, chain saws and road-graders.
"I think we did everything we could have possibly done in those first couple of days," Oso Fire Chief Willy Harper said Friday.
Why wasn't the slide attacked the same way firefighters respond to wildfire?
Under state law, officials have to deny any requests for fire-asset mobilizations that don't involve smoke and flames. Some are seeking to change that law in light of what happened in Oso. County officials got more help through a separate process, getting the slide declared a disaster.
Could more people have been saved if search efforts were better organized?
People immediately on the scene reported hearing screaming and believed it came from trapped victims. Recordings of radio chatter that first day document confusion about whether noises were coming from animals or people. Officials last week began to gently raise doubts that people might have survived by being trapped in air pockets. Autopsies of all the dead found so far showed they were killed by blunt-force trauma.
Questions have been raised about the level of resources that reached Darrington and why volunteers initially were turned away. Locals found ways around roadblocks to search for loved ones. Emergency officials insist the lives of would-be rescuers likely were saved by limiting well-intentioned missions into terrain that has challenged seasoned rescuers.
Why has the number of missing people fluctuated so much?
Fear and confusion, particularly in the early hours, played big roles. Phone service to Darrington largely was knocked out. People unable to reach loved ones called to report them missing. Others contacted a special phone line to pass along tips about people they thought lived in the slide area. As the hours became days, the list kept growing. Even after eliminating duplicate reports, detectives from the sheriff's office had to sift through a list of more than 500 names. They began the task three days after the slide. They used databases, worked the phones and knocked on doors to cut the missing list to less than 20 names by late last week.
The work was complicated. Every house in Oso held a different story the morning the slide hit. Some people were home. Some had visitors. Some houses sat empty. Meanwhile, people were driving both directions along Highway 530.
Will all the victims be recovered?
Although the search continues, officials say it seems unlikely all victims will be found. There's a lot of debris, more than can be excavated.
When will life get back to normal?
The state is helping to figure out housing assistance for survivors. State officials have said Highway 530 will be rebuilt, but it is too early to know when or exactly where.
Survivors with mortgages on homes lost in the landslide face difficult financial futures, as most homeowner insurance policies don't cover landslides.
Recovering from a traumatic and tragic event will be different for everyone.
For victims' families and many people living in the affected area, there is a new normal.
"It won't go back to the same way it was, but it will go back to a situation where they're able to manage daily life in a way that feels comfortable," said Martha Read, a licensed counselor who's leading the mental-health response for the American Red Cross in the wake of the landslide.
Read and other counselors are on the ground helping people carry on after the disaster and, hopefully, preventing long-term psychological harm, such as post-traumatic stress disorder.
More Local News Headlines
Three weeks out, only four 911 texts have been valid County deputy director named president of state engineers association Accident victim has endured, thanks to his family, police officer Crews try to save man pulled from Stillaguamish River State’s oldest full-time worker on the job at EdCC Lord Hill levee repairs to start July 29 Less than a week before election, only 9.1 percent of ballots returned Red Cross on the lookout for heroes
Our new comment system is not supported in IE 7. Please upgrade your browser here.