The 2015 Kramer Avenue Landslide in Sitka sliced through this neighborhood, destroying two houses, but narrowly sparing others. “Getting in on the front end of what triggers landslides… is really attractive to me,” says University of Oregon geomorphologist Josh Roering. He adds that there are many drainages on Sitka’s slopes with similar potential to slide. (SCS image capture/ Rob Dunbar)

A multimillion dollar effort to test the feasibility of a landslide warning system has hit the ground in Sitka.

A research team from the University of Oregon and the Rand Corporation was in Sitka recently (the week of April 29) to scout locations on the mountain slopes for sensors, and to build the social network that will rely on them.

Josh Roering is a geomorphologist — which is exactly what it sounds like. He studies changes in the surface of the Earth. Since 2011 Sitka has experienced a cluster of landslides — two with loss of life — which make it the right place to test a warning system. Roering says landslide prediction is a novel idea.

“For me, as someone who’s worked on landslides in a post-mortem capacity, in looking at their role in shaping landscapes and diagnosing the factors that caused them to occur,” said Roering. “Getting involved in the front end of trying to understand what triggers them, and whether we can come up with warning systems that are informed by the community, and have community involvement, is really attractive to me.”

Prof. Josh Roering (second from left) from the University of Oregon, uses high-resolution LIDAR maps to determine which drainages are most susceptible to slides. LIDAR “sees” terrain through the trees, with laser technology. (Sitka Sound Science Center photo)

Roering, a professor at the University of Oregon, joined the landslide warning team in 2018. The project is the brainchild of the Rand Corporation, and is funded with a $2.1 million grant from the National Science Foundation.

Rainfall is the root cause of Sitka’s landslide epidemic, and Roering says that the National Weather Service — another partner in the project — has become adept at forecasting the “rivers” of atmospheric moisture that can create massive rainfall events in a place that’s wet most of the time already.

Roering’s attention is not on rainfall per se, but on rainfall’s effect on land.

“We want to understand the evolution of the soil moisture through time,” said Roering. “When do soils wet up, how wet do they get, what does it take to push them over the edge to where they become so saturated that they fail. That they trigger a landslide that turns into a debris flow which can cause loss of life and damage to infrastructure.”

The monitoring equipment to do this work has been around a long time, but it hasn’t really been feasible to use on a widespread basis: expensive soil sensors, transmitters, receivers, heavy batteries that don’t last very long.

Innovation in our daily lives has changed all that.

“What’s new here is the fact that the sensors we’re putting in are low cost and they’re broadly available,” Roering explains. “These are the same sort of sensors and technologies that are being put into refrigerators and all kinds of appliances all over our lives. It’s generally referred to as ‘the internet of things,’ and it’s trying to leverage these new, low cost tools to do that.”

Landslide warning will harness all the latest technology and  tools, but is it a sure thing? Roering doesn’t think so.

But he’s confident that Sitka — and other places susceptible to landslides — will be better informed, and better off, when conditions add up to the potential for slides. And maybe we’ll get the word from a landslide app, or a notification on our phones.

“We wouldn’t be here if we didn’t think we could make a difference,” he said. “We now know from mapping that’s been done where debris flows are likely to go, and where they’re not likely to go. So we better understand the risk from a spatial pattern, and those maps are in the process of going through review and will be available to the community. Now temporally which one of those — for a given storm — has a greater likelihood of experiencing debris flow is something that’s a really big challenge, and I don’t think we’re going to get there. But we will be able to provide that realtime information to people — perhaps in these zones of potential debris flow runout — that conditions are approaching levels that might be associated with debris flow activity.”

Roering and his team will be back in Sitka this June to begin installation of the soil sensors on the slopes above the community.