.png)
Scientists search for ‘recipe’ driving landslides near Klukwan
By Lizzy Hahn
Chilkat Valley News
We’ve heard of seismometers identifying earthquakes, but what about them being used to identify and track landslides? This is something that students with the University of Oregon are working with, in conjunction with a slew of other devices to better understand the landslides in the Chilkat Valley.
In mid-August, University of Oregon graduate student Maia Seeley and newly graduated Ph.D. Maryn Sanders fiddled with a listening device on a landslide fan above the Haines Highway, manually swapping out its SD card.
As they worked, they joked that if a debris flow, containing tire-sized rocks, ever hit one of the monitors, they’d need a metal detector to find it.
The students are in a field team with University of Oregon professor Josh Roering. The group traveled to Klukwan to set up sensors as part of two ongoing landslide observation programs. The sites they’re studying, at 23-Mile and 19-Mile, are along the Haines Highway, part of a notorious avalanche and landslide corridor in the Chilkat Valley.
The 23-Mile fan sits directly above Klukwan; the Haines Highway runs between the two. The potential impact of that slide, and others nearby, is part of what’s driving the research team’s work to help Klukwan accurately assess and respond to the hazards in the future. Part of that emergency management is understanding what events lead to large landslides.
So the tribal government began working with regional organizations and international researchers to understand more about the local conditions. Now the village is taking part in two ongoing landslide observation programs: the Kutí program and the Center for Land Surface Hazards, or CLaSH, program. Through the Kutí program, Roering and other scientists partnered with communities and observed landslide areas in Skagway, Prince of Wales, Hoonah, Yakutat and Klukwan to monitor landslide areas. The CLaSH program is much broader; in addition to Klukwan it also monitors areas in Puerto Rico, California, Kentucky and Appalachia.
A long history with slides
One of Shawna Hotch’s first memories as a young child in the early 2000s was playing in front of her grandmother’s house in Klukwan. That summer, she remembers watching her parents work on the house. As she grew older, she realized her parents had been digging it out after a landslide.
“You could find my grandma’s house with like a bunch of rocks. Under it was silty and muddy; they had to raise her house up,” said Hotch, who lives in Klukwan with her husband and children. She drives between the village and Haines, through that unstable corridor, sometimes three times a day.
Elders who talked to Roering and his students described flows in the 1980s and ’90s that would come over the highway, through the village and then out to the river. These landslides didn’t have big boulders in them, making them less destructive. However, families like Hotch’s had to dig out their houses, some of which had mud on one side, causing them to tilt.
“It was super traumatic, really impacted the community and created a lot of concern,” Roering said.
In the late ’90s, a berm was built to keep any debris flows in the main channel; the smaller channels led to more debris ending up in the village while the main channel bypasses over the highway, north of Klukwan. Roering said this channel has worked very well, but flows have started to bounce off the other side of the mountain.
Like a ping-pong ball, the barrier guides the channel away from the village, but now it is being pushed back toward the village. In 2022, a debris flow from the fan above Klukwan ended on the highway. Roering said that if this had more volume, this flow would have gone down to the school and other structures in town.
Marvin Willard Jr. has been recording changes on the landscape for four years, taking pictures of how the river changes in front of his house. “In the springtime, I’ll go walk the beach and take pictures because I want to see what happens, because that’s where I fish,” he said.
When the sediment last came to his smokehouse, Willard had to shovel almost a foot of it just to open the door. It’s a smokehouse that he inherited from his father and Willard said he hasn’t built barriers to prevent sediment from getting into it because “we don’t mess with her [Mother Nature].”
But, using what he has learned, Willard is planning a new design for his future smokehouse which he said will have “two or three layers of concrete, with a little point on it, so it [sediment] can divert right around.”
That planning for the future, particularly one shaped around living with an unstable slide nearby, is happening all over Klukwan. Hotch, who used to lead Klukwan’s emergency operations, said village leadership started focusing on emergency management after a series of landslides in December of 2020. That year in Haines, a landslide killed two people when it slid down over Beach Road. A third died during the emergency response trying to find survivors.
Often when landslides or avalanches come down along the highway, they cross it and people get stuck on either side, sometimes overnight or longer depending on how long it takes the Alaska Department of Transportation to clear the debris. This happened to Hotch in 2024, when a 20-foot avalanche came down over the highway, stranding her in Haines.
Because of this, Chilkat Indian Village consultant Jess Kayser Forster said “we need to start thinking about this as a whole valley.”
The science
Roering and the teams working with him are focusing specifically on the 23-Mile and 19-Mile slides, and their fans, or accumulation of debris that occurs from repeated slides in an area. These two particular slide fans have been growing for more than 10,000 years and are some of the largest in the area, with small fans lining the Haines Highway. The Department of Transportation is already monitoring a growing fan that could cause future problems at 19.5 Mile.
While relatively close to each other, Roering describes dramatically different geology between them. The debris flow from the 19-Mile fan is made up of granite boulders, about the size of car tires. One team is studying why the boulders aren’t larger in this flow. Roering’s theory is that they break up as they fall down the mountain. The 19-Mile fan is steep, rocky and constantly eroding.
“These areas are just peeling away faster than any life can try to take hold,” Roering said as he stood in the fan in mid-August. “So the result of all that erosion, of rock falls and so forth, is the collection of all that loose material in these valleys.”
That loose material is one of the ingredients needed for a debris flow. The other is water which can come in intense floods with atmospheric rivers that sit over the Chilkat Valley.
“It flows 20 miles an hour, so faster than you can run,” Roering said.
The 23-Mile fan that looms over Klukwan has both large rocks and sand. Roering said its “super variable” rock sizes is part of what makes this fan interesting. Bigger rocks collect on the upper parts and most of the volume that comes down to the lower part is sand.
When debris flows down this fan, it tends to be a muddy slurry by the time it reaches the highway, and then Klukwan. He said this type of flow could be slightly less disruptive because the slurry can flow around buildings, versus big boulders which cannot be maneuvered easily.
In the flow
Scientists use maps to determine where active channels are. The map shows where the natural levees, or ridges along the channel, are made. Roering says that debris flows build up their own ridges by “plowing the material off to the sides,” which sets the course for the next ones to come.
An undergraduate student on the team is modeling what happens to a channel when the debris flow gets too big and overflows. Roering said this avulsion, or dipping into other channels, occurs when a big debris flow either gets clogged or so big that it flows over the levees.
“You can basically have years or decades of really well-behaved debris flows that all of a sudden, some change in the system causes it to change course, and occupy a whole different part of the fan,” Roering said.
That potential for a large channel of debris to shift is something Roering said should be a concern for everyone in the area, particularly the state’s department of transportation which is responsible for clearing the flows.
“It’s great if it says on course [the channel], but over geological time scales, we all know these things have been wanting to go all over,” he said.
One solution the department of transportation found was to raise the highway and install giant culverts to allow debris to go underneath the road, minimizing cleanup on the actual road itself. Roering said that the culverts are similar to drains, as they still require people to clear them to keep debris flowing through them.
Monitoring
Landslides happen on both fans every year, but not all of the debris falls make it down to the road. Part of this is thanks to the four large culverts under the bridge at the 19-Mile fan.
Each fan has 15 sensors: one seismometer to measure ground vibrations, four infrasound sensors to measure the sound, nine game cameras to visually document the debris flows, and one pressure transducer to measure the water level in the main channels.
During their recent trip, Roering and his students checked and installed new sensors to monitor slides, gathering the data from the previous year each summer.
Graduate student Seeley said they’ve been using listening devices sensitive enough to pick up rock that falls higher up in the fan. Those rocks and debris end up collecting in a catchment which, with heavy rainfall, can begin to move. So, similar to how earthquakes are mapped, the scientists will be able to track landslides as they move down the fan. “It will be able to hear the landslides,” Seeley said.
The students also put pressure transducers, or water-level measurement devices, into the creeks that flow in the main channel of each fan. The transducer records the river changes. Roering said that theoretically, if the transducer survives the debris flow going over it, it will be able to tell them the pressure associated with the flow.
A new sensor that the team introduced this year hopes to visualize the slides. Two students working with Roering, Charlie Holmes and Ries Plescher, are setting up game cameras, though it’s not yet clear how well they’ll work.
A few of these cameras are connected to cellular data with the idea being that Holmes and Plescher can watch a live stream of the fan from Oregon.
The duo said that they will share the livestream with the Chilkat Indian Village. These cameras cannot detect a slide in advance yet, Holmes said, however they are hoping to gather enough information to potentially get there with these cameras.
Aside from these cameras, most of the data cannot be seen by the scientists until the following year when they come back to check on the devices and swap their SD cards. So Roering said this process was all just delayed gratification.
How the science can help
These sensors cannot act as a warning system. Landslide warning systems are rare, only two or three exist in the world, in places with “massive infrastructure,” as Roering put it. One of these warning systems is in Kandersteg, Switzerland.
These instruments, like the seismometers Roering and his students have installed, need to be maintained to be a consistent warning system. He said that the warning system in Switzerland took “decades and millions and millions of dollars to build.” This kind of a warning system is difficult to do in Alaska.
The alternative to a warning system, he said, is figuring out what the recipe, or what weather patterns, create landslides that either end up on the highway or in the village. Roering said that his team has partnered with the Alaska Department of Transportation, which is sharing data on when they clean up the highway from a landslide. By getting the dates and locations of debris flows from DOT, Roering hopes to go back and look at the weather patterns that occurred, using information from those sensors.
“What is the recipe of the ingredients for these, [that] is what we want to find out,” Roering said.
Using historic landslide information and data gathered from these various sensors, Roering and his students were able to develop landslide-warning dashboard websites for Sitka and Prince of Wales. The dashboards rely on weather forecasts from the National Weather Service and the data Roering and his team are collecting. By combining these data sets, the dashboards can predict how much of a landslide risk there is in the next two to three days. The severity is ranked from low to medium to high.
“You need to take that science, you need to translate it on the ground and you need to apply it and better the lives of communities,” said Jess Kayser Forster, who has worked as a consultant with the Chilkat Indian Village in Klukwan since 2015. She has been a project manager for climate resilience since 2018. She is also working on the CLaSH project as a principal investigator, acting as the middleman between scientists and communities.
Over three years, CIV developed a climate resilience plan that focused on protecting its valuable infrastructure. Critical infrastructure like the village’s water treatment plants and clan houses are at stake with every landslide that comes down.
Using the information that Roering and his students are gathering, Kayser Forster said that Klukwan can create a long-term mitigation strategy. The Chilkat Indian Village is currently developing a small community emergency response plan using funding from FEMA.
“Landslides have come into the village. So the threat, the risk, is real to them and it’s going to get worse before it gets better,” Kayser Forster said.
One practical application from this project occurred when the village of Klukwan wanted to build new houses. Kayser Forster said that using the information Roering and his team had gathered, Klukwan built houses in an area more protected from landslides.
This has also brought up the question of how to protect existing infrastructure, including whether to build berms in front of some of it to deflect the debris flows.
“There’s a lot of communities around our country facing natural disasters and it’s only getting more intense,” said Kayser Forster. “Trying to find the funding to mitigate these types of things, it would benefit us all to do really great community planning moving forward.”
• This story originally appeared in the Chilkat Valley News.
_edited.jpg)
.png)
.png)
.jpg)