A landslide warning system can’t predict the time of an event — only the conditions that may cause one to occur. Making residents aware of a hazard requires the right kind of information and messaging. “We’re here working on a social network approach to make sure that the communication network is resilient, has redundant pathways, and reaches as many people as possible,” said social scientist Ryan Brown. (NWS photo/Joel Curtis)

A three-year project to develop a landslide warning system in Sitka is entering its second year — and moving from a hardware phase to something a bit softer: Social networking. This week a group of researchers is in the community to learn more about how Sitkans inform each other during emergencies.

Note: The researchers with the landslide warning project will be at the Backdoor Cafe to meet with residents beginning at 7:30 a.m.

When I hear the term “Social Network,” I always think about the bad break up scene in the movie of the same name.

Actor Jesse Eisenberg portrayed Facebook founder Mark Zuckerberg in the movie “The Social Network.” Although his love life suffered for it, Zuckerberg’s obsession with building Facebook resulted in a tool that researchers at RAND believe can save lives. (Merrick Morton/Columbia Pictures)

Clip – Dating you is like dating a  StairMaster...

Ouch. I don’t know if Facebook founder Mark Zuckerberg counts as a scientist, but people with his single-minded focus are blindsided all the time. A team is in Sitka right now to make sure that landslides don’t blindside us.

“We’re here working on a social network approach to make sure that the communication network is resilient, has redundant pathways, and reaches as many people as possible,” said Ryan Brown, one of two social scientists visiting Sitka hoping to understand how we collectively respond to warnings. He doesn’t think that there’s a universal approach.

“Even if a message or warning is sent to everybody, it often takes multiple warnings through multiple channels and sources,” he said. “So hearing it on the radio, and maybe through a text message, for people to get jogged out of their sense of normalcy and to take preparatory action: evacuate, for example.”

The work in Sitka is funded by a $2.1 million National Science Foundation “Safe and Connected Communities” grant. Brown is involved in the “connected” part. Annette Patton, a post-doc in Geoscience at the University of Oregon is the “safe” part. Last summer she led a team that installed three innovative landslide sensors on the slopes above Sitka. The sensors have been transmitting rainfall and soil moisture data to the Sitka Ranger District and to the firehall. Integrating that data into a warning system is the next step.

“One of the things that we know is that as we build this data set, near the beginning we’re going to have a pretty sparse data set,” Patton said. “And there’s going to be a lot of uncertainty. And so we want to weigh the questions: Is it better to have failed alarms or false alarms? What balance works well for this community?”

As a rule, Sitkans are responsive to warnings. When a magnitude 8 earthquake rocked the seafloor of the Gulf of Alaska in January 2018, Sitkans living near the beach quickly relocated into emergency shelters set up in local schools. This is something old-timers have been doing every few years since the Great Quake of 1964. Landslides, though, are a different animal: Usually a culminating event, following days of heavy rains and wind — landslides are very localized. Ryan Brown says that Sitka’s social network can utilize outside tools to enhance warning capabilities — like a weather app developed by NASA called Global Observer.

“Anyone with a cell phone can make observations of either landslide-relevant conditions, or evidence of a new slide,” said Brown. “Yeah, we know there are aviators, the fishing community, the outdoors community — and they see slides, or hazardous weather conditions before anyone else would, especially because the conditions are ultra-local.”

All of this is the brainchild of Rob Lempert, a community risk analyst with the RAND Corporation in Santa Monica, and the principal investigator under the National Science Foundation grant. Introducing a social networking component into a landslide warning system seems like it may stir some information chaos into the sauce. But Lempert believes the results will ultimately be more effective.

“A good warning system has risk knowledge, has monitoring, has communications, and has response,” Lempert said. “And so our project is trying to enhance all those different pieces and integrate them together — both within the city, and with state and federal agencies.”

Lempert says the social networking component of the system won’t be perfected in this trip — or maybe even in the next. He says this research is iterative: His team will ask questions, take feedback, refine the model, and then return to ask more questions. One member of the team, doctoral fellow Max Izenberg, will remain in Sitka for the entire month of April working on shortening the learning curve, and closing “structural gaps” in the social networking system to make sure that everyone is covered in the event of a warning.

Sitka lost three community members in August, 2015, when a landslide tore through a new neighborhood development. While far from the worst of landslides, the experienced galvanized Sitkans’ thinking about where and how we choose to live on this remote coastline. It’s a near-certainty that residents will embrace whatever social networking strategy emerges from this research.

And then, RAND hopes to export the model to other landslide prone areas across the country. Geoscientist Patton says Sitka is an ideal environment for this human-powered, data-driven warning system.

“If it doesn’t work in Sitka, it’s probably not going to work anywhere,” said Patton. “So if we can dial it in here — this system — it probably has the best chances here of anywhere in the country.”

And a chance of success is better than no chance, as our unlucky-in-love hero discovers in The Social Network.

Clip – I was just being polite. I have no intention of being friends with you.