Each week we will feature a new landslide in Washington State. Washington State is covered with dynamic and sometimes quirky landslides.

Racehorse Creek Landslide, Whatcom County

On January 6th, 2009, a Pineapple Express (actually to have formed off of Hawai’i) flowed into Washington State, hitting the northern counties first and moving into the southern counties. Whatcom and Skagit Counties were first to report landslides late on January 6th as warm rains melted away snow and thawed the ground. As the rain continued, a small rain on snow type event occurred, spawning over 1,500 landslides. Debris flows and debris avalanches were the most common landslide to have formed from the storm event and the majority of the landslides occurred on the flanks of the Cascade Mountain Range.

January 2009 storm map with an incomplete landslide inventory

January 2009 storm map with an incomplete landslide inventory

This map shows the storm intensity overlain with landslide initiation points, primarily from DNR/DGER aerial surveys after the storm and reported landslides from public and private folks.

One of the largest landslides during the January 7-9th storm occurred along Racehorse Creek in Whatcom County.

Racehorse Creek Landslide; DNR/DGER Photo

Racehorse Creek Landslide; DNR/DGER Photo

The landslide occurred in two major components, the main debris avalanche and near lateral debris flows. The main debris avalanche is over 160,000 square yards and moved a significant amount of trees into Racehorse Creek. The debris flows scoured into the ground, removing timber in its way, also reaching Racehorse Creek. Once in the swollen waters of Racehorse Creek, the moved debris moved downstream, forming a massive logjam.

Logjam formed by the Racehorse Landslide, looking towards Kendall Creek; DNR/DGER Photo

Logjam formed by the Racehorse Landslide, looking towards Kendall Creek; DNR/DGER Photo

Logjam formed by the Racehorse Landslide, looking towards Racehorse Creek; DNR/DGER Photo

Logjam formed by the Racehorse Landslide, looking towards Racehorse Creek; DNR/DGER Photo

The size of the landslide has caused many to scratch their heads as to what possibly might have triggered this landslide. Some point to an earthquake as a possible trigger (one did occur on January 6th, 2009), others, natural factors of erosion and saturation. Or, as is common, a combination of saturated ground, erosion of the toe and a bit of shaking from an earthquake.

Whatcom County Earthquake Map; DNR/DGER

Whatcom County Earthquake Map; DNR/DGER

Thunderstorms over Chelan County have caused a small landslide that destroyed a driveway on the Stemilt Creek Road. According to the Wenatchee World:

” “I wouldn’t say our driveway is damaged. I’d say its nonexistent,” Tammy Liebert said of the 10-foot by 30-foot mud slide that swept across the gravel road to her home at 3550 Stemilt Creek Road. “My husband’s out there with a backhoe now.”

Liebert, her husband Terry Liebert and their 17-year-old daughter were home during the slide, which happened around 8:30 p.m.

“It rained huge drops about the size of chicken eggs, then it hailed about the size of quarters and then there was nothing, and then a huge downpour,” Liebert said. “We heard a noise like a semi skidding down pavement. It was the rumbling of rocks rolling together down the gully.”

Besides garbage can-sized boulders, tree trunks and mud, the Lieberts’ road is also covered in tires.
Liebert said the slide took with it part of a hill along Wenatchee Heights that is a popular teenage drinking spot.

The teens take tires up there and roll them over a guardrail to see how far they go, Liebert said.

“The kids strictly do it for entertainment,” she said. “Once we saw one bounce about 10 feet in the air.”
She said neither the home nor their vehicles were damaged in the slide, but neighbors also received mud damage to their driveways and an orchard.

The storm itself doesn’t appear to be too devastating with rain, so this probably barely reached a threshold limit on an area that is already very unstable. ”

Precipitation Map of Eastern Washington

Precipitation Map of Eastern Washington

The rainfall was probably under .5 inches of rain in the area. However, the area that the landslide initiated out of came is the lateral scarp of the Stemilt Landslide and is a fairly unstable area.

Stemilt Road Landslide

Stemilt Road Landslide

This photo shows a fairly steep scarp with older landslides and scree along the cliff side. A failure in one of these gullies or an area that ponded water could easily turn into a debris flow or debris avalanche and be able to do some damage downhill.

Each week we will feature a new landslide in Washington State. Washington State is covered with dynamic and sometimes quirky landslides.

Hazel Landslide, Snohomish County

The Hazel Landslide is a persistent deep-seated landslide that is probably driven by groundwater and erosion by the North Fork of the Stillaguamish River.

Hazel Landslide Location Map

Hazel Landslide Location Map

The landslide has caused some headaches for DNR, as a catastrophic failure and partial blockage of the Stillaguamish River around 1988 spurred many to consider logging as the culprit.

Hazel Landslide

Hazel Landslide

Logging in the northwest area above the landslide came into question as a area that was a groundwater recharge area. The addition of water from the removal of trees was considered to be at least a partial catalyst for the failure of the Hazel Landslide and it is seen as a poster child of what groundwater recharge in sensitive areas can do.

Since 1988 the forest land has recovered and groundwater recharge should have been diminished, but in January of 2006, following a period of prolonged precipitation, the Hazel Landslide once again moved, diverting the river into a small community of houses. This landslide was another large landslide and has been called the Steelhead Haven Landslide. During that month, about 8 inches of rain fell, well above the average 4.5 inches that typically falls during the month of January (in 93 months of record). This would overwhelm any recovery that may have occurred with the maturing forest.

Hazel Landslide 2009 - DNR/DGER Photo

Hazel Landslide 2009 - DNR/DGER Photo

Hazel Landslide 2009 - DNR/DGER Photo

Hazel Landslide 2009 - DNR/DGER Photo

The landslide produced a lot of sediment, which I hear from some of my sport fisherman up there, caused quite a poor year of fishing. The long term effects might not have been devastating, but with a weakening population of fish, it certainly hasn’t helped any.

In the landslide world, the Hazel Landslide is certainly one of the more well known landslides. The unstable nature of the glacial lakebed lithology it sits in has caused countless landslides throughout Western Washington and is a legacy of our glacial history. However, it has taught us some valuable lessons in how groundwater affects lacustrine beds and its potential sensitivity to water.

Each week we will feature a new landslide in Washington State. Washington State is covered with dynamic and sometimes quirky landslides.

Sultan River Debris Avalanche, Snohomish County

When I first started at DNR, I lived in Seattle (commuting to Olympia), but most of my field work was in or near the Sultan Basin. The Sultan River watershed was the first watershed I worked on for the LHZ Project. My first year working was almost comical, every day I was in the field, it rained. In the summer, we would have long stretches of dry weather and one day of rain every couple of week and those were the days I ended up in the field.

On December 10th, Pat Pringle and I were in the field investigating the Sultan River basin. We didn’t hike or really get out of the car much due to a torrent of rain coming down. We never quite got over to the area of the Sultan River debris avalanche, but we were just about across the way from it. I did end up heading out there after it failed.

Sultan River Hydrology

Sultan River Hydrology

On December 11th, a group of Kayakers decided to ride the Sultan, mostly because of the elevated water levels from the storm. During their ride down, they ran into more than they bargained for.

Sultan River Debris Avalanche Location Map

Sultan River Debris Avalanche Location Map

Sultan River Debris Avalanche

Sultan River Debris Avalanche

A large debris avalanche/fall came down right after the kayakers passed by. As you can see in the video, the landslide dammed the Sultan River for a short time, but eventually over topped/breached the dam. The kayakers, a bit shaken up by the landslide decided to leave the river and hike out and head back to civilization. Unfortunately, they didn’t know the area and headed east of the the Sultan Basin, eventually reaching a small nudist community. They all make it out safely.

To make matters worst, the City of Sultan received a report of a large landslide damming the Sultan River. Their main concern was a dam burst flood that would damage the city. The Snohomish County sheriffs office had a helicopter in the air trying to find the landslide and see how bad the blockage was. This was a real thing to worry about. The Sultan River is entrenched and a dam could potentially block a large amount of water. Strangely enough, many of the rapids (that the kayakers seem to enjoy so much) came from old landslides that probably partially or fully dammed the Sultan River and slowly eroded out, just like this landslide.

Landslides are rarely caught on video and when caught, they are a valuable source of scientific information. In this video, you can see how the landslide overcomes what almost looks like a small rock that is holding back the mass of material break apart shortly before movement. You can almost calculate the acceleration as it heads towards the river and what happens as a mass of material impacts into the water, the size of the waves and so forth. We can also see how the river responds to the debris dam and how it returns to normal flow without breaching the dam.

This video also caught the attention of others and it was part of a Discovery Channel series called Raging Nature. I was interviewed by the show, but I didn’t appear on camera for the show (probably the background river sound, which was really loud). The narrator during the show did say much of the exact words that I said (which was kinda creepy for me).


Geology

Bedrock Units

Regional bedrock that includes the Sultan River watershed belongs to the Western Mélange Belt, part of the Western and Eastern Mélange Belts (WEMB) terrain. The WEMB includes Mesozoic (late Jurassic to early Cretaceous) marine sedimentary rocks, along with lenses of Paleozoic limestones, Mesozoic intrusives, and other rich types in fault-bounded bodies that were tectonically juxtaposed (Tabor et al, 1993). The WEMB rocks underwent high pressure, low temperature metamorphism in the late Cretaceous orogeny at about the time they were juxtaposed against the Northwest Cascade System terrain to the North.

Bedrock in the Sultan River watershed is mainly composed of the Western Mélange Belt (Phipps et al., 2003; Dragovich et al., 2002). These rocks were deposited during the late Jurassic to early Cretaceous (170 to 100 million years ago) periods (Carithers and Guard, 1945). Sediment was thickly deposited in a marine setting, comprising mostly of silt and mud. Hydrothermal systems and submarine eruptions (similar to black smokers) formed from intruding magma, creating large pyritic deposits (such as the Lockwood Pyrite deposit) and overlaid the marine sediment with volcaniclastic and mafic flows (for example, basalt) material (Olson, 1995; Snohomish County, 1979). This magma chamber underwent differentiation, where the heavier mafic material (rich in iron and other metallic minerals) filtered to the bottom of the chamber and lighter felsic material (rich in silica, such as quartz and feldspar) rose to the top (Stewart, 2005). These rocks were then metamorphosed (exposed to heat and pressure), folded, uplifted and eroded. The metamorphism changed the marine sedimentary and volcaniclastic rocks into argillite (metamorphosed siltstone) and phyllite (metamorphosed mudstone). The granitic magma chamber also experienced metamorphism, altering the granitic rocks into meta-tonalites (light colored granitic rock), meta-gabbros and meta-peridotites (dark colored granitic rock).

As the rocks experienced pressure from the west (most likely from the oceanic plate colliding with the North American continental plate), they tilted the stratigraphic section to the northeast. This tilting, along with erosion of the overlying rock, exposed the relict magma chamber (gabbro and peridotite in the west, grading east to tonalite) in the western part of the Sultan River watershed. The metamorphic marine rock, which overlies the relict magma chamber, can be found primarily in the southern and eastern parts of the watershed. The metamorphic volcaniclastic rock, which overlies the marine rock, is located primarily on Blue Mountain, in the northeast part of the watershed.

The meta-tonalite rocks, where not overlain by glacial drift, is very stable, even with slopes steeper than 60% (A prime example of this is the large hill, located in T. 28N R. 8E, section 2 and 11). The meta-marine rocks can be unstable, especially when the beds are tilted to near vertical. The north flank of Blue Mountain is an excellent example, where the meta-sedimentary rocks are tilted to near vertical and failures are frequent within the section. The meta-volcanic rocks can be very unstable and appear to be very susceptible to slope failures when the rock is exposed to water. A prime example of this is the water run-off from the radio tower located at the highest peak on Blue Mountain; many debris flows initiated from this deposit, independent from harvest or road construction.

Poorly-Consolidated Surficial Units

Surficial units in the study area consist of continental glacial drift. Other surficial deposits are composed of alpine glacial drift, colluvium, and alluvium. About 14,000 years ago, the Puget Lobe of the Cordilleran ice sheet, which represents the most recent advance of continental ice sheet, flowed into surrounding valleys. This advanced was named the ‘Vashon Glaciation’ locally. Tongues of the Vashon glacier dammed valleys that were tributaries to the Puget Lowlands, creating large ice dammed lakes. Glaciers advanced up the Pilchuck River system and the Sultan valley, covering the northwestern portion of the watershed (Tabor et al., 1993). This blocked the paleo-Pilchuck River, creating a large ice-dammed lake and depositing deltas and lake deposits on the north flanks of Blue Mountain to Bald Mountain. This rising lake eventually overflowed, washed over Olney Pass, and deposited fluvial outwash across the plains in the west and south parts of the Sultan River watershed.

Ice margins near Lake Chaplain and Echo Lake also produced significant outwash towards the town of Startup (Booth, 1990). As the glaciers retreated, the terminal moraine (called the Pilchuck plug) blocked the upper drainage of the Pilchuck River, creating the new Sultan River watershed (Coombs, 1969; Bliton, 1989). The Sultan River established a channel, rapidly incised into the glacial material, cut into the bedrock, and became entrenched. This incision is probably due to easily eroding glacial material and isostatic rebound of the bedrock in the area. Old meander bends and channels can be seen near the main channel of the present Sultan River.

Near the confluence of the Sultan and Skykomish River, glacial lakes formed by the advancement of the Cordilleran ice sheet, creating thick lake deposits in the southern extent of the watershed (Booth, 1990). These lake deposits formed low-permeability clay and silt layers that perch water and spawn large landslides during high precipitation. The silt and clay layers are commonly overlain by permeable glacial outwash from the paleo-Spada Lake and ice-margin flows. This combination of silt, clay and sand makes much of the hillsides in the southern part of the watershed susceptible to shallow and deep-seated landslides.

Reference

Booth, Derek B., 1990, Surficial geologic map of the Skykomish and Snoqualmie Rivers area, Snohomish and King Counties, Washington: U.S. Geological Survey Miscellaneous Investigations Series Map I-1745, 2 sheets, scale 1:50,000, with 22 p. text.

Bliton, William S., 1989, Sultan River project. IN Galster, R. W., chairman, Engineering geology in Washington: Washington Division of Geology and Earth Resources Bulletin 78, v. I, p. 209-216.

Carithers, Ward; Guard, A. K., 1945, Geology and ore deposits of the Sultan Basin, Snohomish County, Washington: Washington Division of Mines and Geology Bulletin 36, 90 p., 1 plate.

Coombs, H. A., 1969, Leakage through buried channels: Association of Engineering Geologists Bulletin, v. 6, no. 1, p. 45-52.

Olson, Duane F., 1995, Geology and Geochemistry of the Lockwood Volcanogenic Massive Sulfide Deposit, Snohomish County, Washington: Western Washington University Master of Science thesis, 118 p., 8 plates.

Phipps, Richard W.; McKay, Donald T., Jr.; Norman, David K.; Wolff, Fritz E., 2003, Inactive and abandoned mine lands–Spada Lake and Cecile Creek watershed analysis units, Snohomish and Okanogan Counties, Washington: Washington Division of Geology and Earth Resources Open File Report 2003-3, 36 p.

Snohomish County Public Utility District No. 1; Washington Department of Ecology, 1979, Sultan River project, Stage II; Application for amended license, FERC project no. 2157–State of Washington final SEPA EIS and FERC environmental report (exhibit W): Snohomish County Public Utility District No. 1, 2 v.

Stewart, Richard, May 27th, 2005, Personal Communication

Tabor, R. W.; Frizzell, V. A., Jr.; Booth, D. B.; Waitt, R. B.; Whetten, J. T.; Zartman, R. E., 1993, Geologic map of the Skykomish River 30- by 60-minute quadrangle, Washington: U.S. Geological Survey Miscellaneous Investigations Series Map I-1963, 1 sheet, scale 1:100,000, with 42 p. text.

Savage Island, Franklin County

One of the interesting things about working here at the Washington Geological Survey is you never know what you are going to find. Last week as I was trying to find information on environmental impacts of landslides, I came across a large landslide along the Columbia River at Savage Island. The landslide did fail on the island itself, but the shoreline along the river did.

Savage Island Landslide Location Map

Savage Island Landslide Location Map

Savage Island is located in on a stretch of the Columbia River known as the Hanford Reach. Geologically, this area is know as the White Cliffs and across the way is the Hanford Dunes.

Savage Island Landslide Map

Savage Island Landslide Map

This area was originally inhabited by the Wanapum tribe prior to European settlers entering the area in the 1860’s. European settlements established farms, orchards, vineyards, similar to what we see today. However, in 1943, during the Manhattan Project, the Department of Energy confiscated the land and removed many of the established towns and forming the Hanford Site.
In modern times, agriculture still exists on the eastern side of the Columbia River (with the Hanford on the West). Thick loess deposits make up most of the area (hence the White Cliffs) and create the rich soil for agriculture. Loess can also be sensitive to large amounts of water, since Loess is a fine grained (silty) sediment. Water(from agriculture) + silt(loess) = landslides(usually earthflows).

Savage Island Landslide 1981 - Photo from USGS

Savage Island Landslide 1981 - Photo from USGS

So, it should be no surprise that in 1981 a section of the cliff east of Savage Island started to become active, catastrophically failing as a large earthflow in April of 1981. This landslide, besides clearing away farmland, also killed a large amount of swallows nesting in the cliff side (undoubtedly taking a large amount of rattlesnakes and rodents as well). Agriculture still plays a big role in instability along these cliffs and we continue to see many smaller landslides. Newer watering techniques and technology has helped reduce the amount of water needed (and hence, less water in the subsurface) and this has lead to a reduction in catastrophic landslides such as the Savage Island Landslide.

Hyak Landslide

June 3, 2009

One of the more talked about landslides from the January 7-8th storm was the landslide that occurred at Hyak. The landslide started at the Summit at Snoqualmie ski area and moved into the Hyak community.
This landslide got a lot of press and originally, it was thought that is might be an avalanche. I remember having a discussion about this at a NOAA/NWS video conference meeting. Although, looking at the photos, it seemed that instead of the snow scraping up the soil beneath it, the slope gave way and moved the snow along. Something like a debris/snow avalanche.

Hyak Landslide Location Map

Hyak Landslide Location Map

The landslide occurred at approximately 11:40 a.m. Wednesday, January 7, 2009. Heavy rains (probably at about its elevation limit before turning into snow) from the storm had reached the summit earlier, warming the hillside and inundating it with rain.

Here at DNR – Division of Geology and Earth Resources (Washington Geological Survey), we were in emergency mode. We mobilized all of our geologist and sent them into the field to document landslides, but more importantly, sent them to check on residences that were impacted from landslides and to make sure they were safe from future landslide movement. Unfortunately, I was in the office, directing geologists to specific areas and creating updated maps of where we had located landslides or had damaged houses or blocked roads. I sent one of our geologists on the east side to get to Hyak and investigate what had happened and determine if it was an landslide or a snow avalanche. Plus, it did damage a handful of houses and the hillside had the potential to continue moving.

Aerial photo of the Hyak Landslide - DGER Photo

Aerial photo fo the Hyak Landslide - DGER Photo

Oblique aerial view of the Hyak Landslide - DGER Photo

Oblique aerial view of the Hyak Landslide - DGER Photo

Our geologist arrived I think late on January 7th and found numerous other crews investigating the landslide. Talking to them and doing some investigation herself, it was determined that it was most certainly a landslide and further, the slope was completely saturated. The scarp and material had woody debris within it and within the scarp, casts of old logs could clearly be seen, most with water gushing out the casts. It turns out that the slope had been modified about 40 or so years earlier and it appears they incorporated woody debris into the material. Over 40 years, the wood deteriorated and probably allowed water to more easily infiltrate into the subsurface, probably to the contact between the fill and rock/soil.

According to Matt Cowan, Fire Chief of the Snoqualmie Pass Fire and Rescue, the landslide impacted eight houses, one which was pushed off its foundation, the other lightly damaged. Two people were injured.

(Photo from Hyak Flickr site)

May 27 photo of Hyak Landslide

May 27 photo of Hyak Landslide

The hillside might still pose a threat to future failures. If woody debris exists in the subsurface then continued weakness still exists. I am not sure if the ski resort is planning on regrading the hillside to make the slope usable to skiing, although I cannot imagine that they will abandon the ski area. Unfortunately, we have inherited a lot of legacy problems from early land-use modifications (from the early 1900’s to 1970’s) that still plague us today. They are rarely recognized as a hazard, but their results can be deadly.

Each week we will feature a new landslide in Washington State. Washington State is covered with dynamic and sometimes quirky landslides.

Piper Road Landslide

The Rock Creek Landslide (also known as the Skamania, Stevenson, or Piper Road Landslide) started in 2007 probably after heavy rains hit the area in November of 2006. Movement started to accelerate in February of 2007 and continues to today.

Piper Road Landslide Location Map

Piper Road Landslide Location Map

The Piper Road landslide is located north of the town of Stevenson, located in Skamania County. The landslide is situated on the eastern side of Rock Creek near a waterfalls.

Febuary 2007 Photo of Piper Road Landslide; Photo by DGER/DNR

Febuary 2007 Photo of Piper Road Landslide; Photo by DGER/DNR

This photo was taken shortly after the landslide started to become more active. The landslide itself has started to cut back into the bluff. DNR Division of Geology and Earth Resources Geologists investigated the landslide at this time to determine clues to the landslide and possibly mitigation to help reduce or stop the landslide movement.
By November of 2007, the landslide had progressed significantly upslope:

Piper Road Landslide November 2007; Photo by DGER/DNR

Piper Road Landslide November 2007; Photo by DGER/DNR

At this point, the landslide had destroyed one house (in July of 2007) and the town of Stevenson was desperate for aid and help to mitigate the landslide. The landslide was no longer threatening only houses, but now threatened to destroy a road (with utilities below) and inundate a sewage treatment plant. If left unchecked, sediment would continue to aggregate and threaten to take out the bridge of State Route 14, which is one of the few points that connect major utility lines between eastern and western Washington (such as a natural gas pipeline). Unfortunately, no aid was forthcoming and Stevenson was left to deal with the landslide on its own. I think the quote was (in a conference call regarding the landslide) something akin to “larger towns have been destroyed or abandoned by landslides”. It is, perhaps, the unfortunate result of building in unstable areas.
The long term problems might be bigger than losing a couple of bridges and a sewage plant. The Bonneville Dam is located just downstream of Stevenson and Rock Creek drains into Lake Bonneville. Sediment from the landslide is flowing into Lake Bonneville, which can impact fish populations and slowly inundate the lake and create a sediment headache down the road (of course, this could be dredged out). If the sewage plant is impacted, raw sewage could flow into Lake Bonneville (as far as I know, it isn’t used for drinking water), which could lead to things like an algae bloom or dead zones, anoxia, or just misery and destruction. Granted, probably not enough sewage to do that in such a big lake, but worst case scenario.

The landslide is located on the Bonneville/Cascade Landslide complex and is probably the result of continued movement and activity of the landslide.

Piper Road Landslide in the Bonneville Landslide Complex

Piper Road Landslide in the Bonneville Landslide Complex

As in the Greenleaf Basin Landslide, a combination of higher rainfall and geologic setting probably resulted in the landslide movement. Erosion of the toe of the landslide by Rock Creek, overtime, probably reduced the lateral strength of the landslide and eventually resulted in a breakdown in resistive forces.

The geology of the landslide is well covered in this report by Mark Yinger Associates.
The Rock Creek Landslide Website has been established for the Piper Road landslide (with webcams fixed on the landslide) and additional images.