Local and visiting researchers, along with city and state agency representatives, discussed landslide hazard mitigation and demonstrated a community sharing of knowledge during a forum held last week at Kachemak Bay Campus. The conversation also highlighted important roles that community members play in monitoring landslide hazards.
The workshop was organized by the City of Homer, the Homer Soil and Water Conservation District, the Kachemak Bay National Estuarine Research Reserve, and in-state experts. Presenters and panelists included City of Homer Community Development Director Julie Engebretsen; Noah Finnegan, a geologist with UC Santa Cruz; Roy Kaimo, a researcher and engineer associated with the University of the Philippines Diliman; local geologist Bretwood “Hig” Higman with Ground Truth Trekking; permafrost hydrologist Anna Liljedahl with the Woodwell Climate Research Center; Rick Dembroski, State of Alaska Hazard Mitigation Section Emergency Management Specialist; and Jillian Nicolazzo, a geologist with the State of Alaska Landslide Hazards Program.
Liljedahl started off the meeting by thanking those in attendance.
“It’s wonderful to see you all here tonight, because it means you’re curious,” she said. “You recognize that we have something here in our community that brings us together, that we have the opportunity to address together, talk about and figure out how we want to go about it.”
Higman, serving as forum facilitator, said that one of the “really striking” things that he’s seen come out of a number of recent discussions and presentations on landslide hazards in Homer is “how much cultural knowledge there is of this subject.”
“Plenty of people (are) asking questions where I haven’t thought about that, or making observations that show how I kind of oversimplified things, so that cultural knowledge is very significant,” he said. “One of the things I want to convey is that … (our) expertise is not particularly superior. It just provides some other little pieces of the story, and my hope is in this kind of conversation we can start seeing those different expertises come together.
Higman introduced Kaimo as a leader in the development of the Dynaslope Project, a research program implemented by the Philippine Institute of Volcanology and Seismology to develop an early warning system, using sensor technology and community participation, for deep-seated and catastrophic landslides. The project currently has 53 monitoring sites in the Philippines.
Kaimo highlighted the involvement and participation of local community members in the project, calling it a “people-centered early warning system.”
“Our project relies on landslide sensor technology and the network that comes with it. But (our Secretary of the Department of Science and Technology) said that technology alone is not enough,” Kaimo said. “We have to empower the local minds, and we have to recognize that our people are not just beneficiaries, but they are also powerful agents of innovation.”
The Dynaslope project develops, implements and innovates an end-to-end early landslide warning system using a multi-disciplinary and people-centered approach, he said, involving geology, engineering, social science and community development. Project members place instruments at “disturbed” sites so they can forecast when the slope may fail and a landslide may occur, and involve the participation of a landslide early warning committee, at-risk communities and local government units.
“The landslide early warning committee … are our partners when it comes to implementing the early warning system, and they have their roles and responsibilities,” Kaimo said. “When we send out warning information, they’re the first to receive that information, and then they relay the information to the at-risk communities.”
He said that the project began involving local communities in landslide risk assessment in order to equip residents with a better understanding of the kinds of risk present in their area.
“You lower the risk when the people are capacitated,” he said. “This public agenda increases your capacity because it gives you knowledge on what kind of hazards you’re dealing with.”
Kaimo further explained that community members also help provide monitoring data for the project.
“The community actually has a role in the early warning system, because half of our data comes from there, and the other half comes from the sensors that we develop and install in the ground,” he said.
For example, project members installed steel bars in a crack in the ground at one of their monitoring sites, and community members measure the bars regularly as a “very simple, traditional way of monitoring.”
“We also teach them how to monitor manifestations of ground movement or landslide features or precursors,” he said. “Basically, we want them to understand that landslides are complex processes. It’s important that they observe, and know what to observe, especially in events like high-intensity typhoons and prolonged rainfall.”
Data gathered from the sensors and communities are stored, analyzed and validated before an alert is generated at one of four established levels in the form of a bulletin and SMS text and then relayed to stakeholders. Kaimo said that project members also conduct response capability assessment and public awareness and education.
“If we want to develop a similar early warning system here in Homer, we need to focus not just on one aspect, but on all of these,” he said.
Nicolazzo said that Alaska does have a landslide reporter app that came out in February, which allows residents to act as citizen scientists and provide critical information to geologists about changing landscapes and landslide hazards. Users can also upload photos of landslide hazards using their computer, smartphone or tablet.
Kaimo said that the Dynaslope Project’s implementation has resulted in about 35 evacuations. Project leaders generally recommend that local government units and communities evacuate on alert level three, but he said that they also emphasize that local government units and communities can evacuate based on their own assessment.
“Our alert three is based on a measured threshold … but if the local government unit is confident that it’s not safe for you to be here, they can implement the evacuation. It’s ultimately their power and within their responsibility,” he said.
He noted that there have only been about 15 “alert three” events since 2013 or 2014.
As a related example, Higman described a massive landslide that occurred earlier this year in Switzerland that “destroyed a village,” but caused minimal casualties due to a “successful” evacuation prior to the landslide event.
“That case actually parallels what we’re talking about here, where it was a community observer who noted initial changes in the mountain,” he said. “There was no active monitoring prior to that — there was a little bit of incidental monitoring with a time-lapse camera. That initial community observer then led to more action, plus the slope got a lot more active, and a few days later they decided to issue an evacuation.
“It was a really, really severe event, and the success there was, in part, dependent on that local observer aspect.”
In his presentation, Finnegan reviewed the classification of landslides and landslide behavior, what the connection is between landslides and weather, and how landslide hazards are evaluated based on measurements.
“That’s a lot of what I do, is go in the field and install instruments on active landslides and make detailed measurements with an eye towards both better hazard mitigation and better understanding of the physics that actually govern the failure of these landslides, which is something we know we have some big holes in our knowledge about,” he said.
Regarding hazard mitigation, Finnegan referred to a landslide early warning system model in Sitka which was developed in response to a recent spate of debris flow events, or what he described as long, run-out landslides. Debris flow events are often referred to as mudslides, but Finnegan said that term wasn’t necessarily accurate because of the movement and makeup of the event — debris flows flow rather than slide, and are generally a sediment and water mixture, up to 80% sediment by volume, that becomes a slurry comparable to wet concrete and can achieve very high velocities and carry with it large pieces of debris, even boulders.
In Sitka, Finnegan said, the people who created the model looked at periods where debris flows occurred and plotted the rainfall that occurred within a three-hour period.
“If you look at the days when those debris flows happened, versus all the other days in the record where there were no landslides or debris flows … there’s a pretty good separation of the data,” he said.
Based on those observations, Sitka developed a landslide dashboard which utilizes National Weather Service forecast data for rainfall intensities over three-hour time windows, looks at the past rainfall intensities that have produced landslides, and then produces a forecast based on that time scale.
“While this is a unique place because of the unique geology here, the processes that are happening here are not unique,” Finnegan said. “We understand a lot about them, and you can look to other places to help with developing strategies here. Sitka would be the obvious place to do that, from the perspective of thinking about an early warning system for shallow landslides.”
Higman acknowledged that Homer has two distinct hazards — one illustrated at Bluff Point, which he called a “very deep landslide,” and shallow debris flows occurring in canyons across the area.
“They both have their importance,” he said.
Learn more about landslide hazard mapping and other ongoing efforts in Homer at www.cityofhomer-ak.gov/planning/join-us-presentation-landslide-hazard-mitigation.
