About this project

My name is Eric Matthes, and I've lived in Sitka since 2002. I have a background in physics, and I was a math and science teacher for many years. I'm the author of Python Crash Course, an introduction to the Python programming language. I've been a programmer all my life, with an interest in using data to make informed decisions. I've also been an active member of Sitka Mountain Rescue as long as I've lived here. When landslides began to affect the community, I started talking to many people about the issue - first responders, outdoor enthusiasts, scientists, people who live at the base of mountains, and many others. This project is a result of many of those conversations. For more about how this project has evolved, see the project timeline.

This is a pilot project to determine whether monitoring river levels in a small watershed can lead to a better understanding of when landslides are likely to occur, and just as importantly when they are unlikely to occur. This project is based on a historical analysis of data from the Ḵaasda Héen (Indian River) stream gauge operated by the USGS. This analysis is ongoing, and is being done in collaboration with the Sitka Sound Science Center Landslide Research project.

The purpose of this site is to help the working group determine how readings from the river gauge correspond to landslide risk in real time, and to determine the most appropriate messaging to accompany this visualization. The long term home for this project is likley to be the NWS, which is responsible for reporting on current weather events.

The source code for this site is in a GitHub repository, and you are welcome to report bugs that you find. The historical analysis that forms the basis of this project is in a separate repository. You can also download a report summarizing the original analysis of this approach to monitoring for slides. If you have questions about this project, you can also write to eric@khmpsitka.org.

There are a number of other pages that tell more about this project as well: