Using Python to assess landslide risk: A matter of life and death
summary
Python analysis of a local USGS river gage shows a simple, testable claim: when the river >=2.5 ft rise with a >=6 in/hr sustained rate, conditions often match historical landslides. Historical review finds about a dozen critical events with at least five associated slides, notification lead times of ~30–60 minutes, and several false positives over five years. A public real-time Django tool posted in 2021 visualizes those thresholds and runs on a $5 server.
AI-written summary · about
chapters (20)
- 0:00Landslide risk assessment intro
- 1:09Speaker background Alaska geology
- 3:39Recent landslides community impact
- 5:34River gauge as landslide proxy
- 7:19Watershed monitoring approach
- 9:30Simplified analysis method
- 13:28Real-time monitoring tool
- 18:21Django public-facing tool
- 21:43Strengths and limitations
- 24:09Future improvements
- 25:36Q&A scalability variables
- 27:00Ethics public release
- 28:04False positives vs negatives
- 29:53Optimizing slope thresholds
- 30:37Community actions critical events
- 32:23Landslide data challenges
- 34:52Sensor integration Python
- 36:15Project transition collaboration
- 37:21Student experiences classroom
- 39:09Closing remarks Q&A
related talks
31m"How we used Python to try and save lives"
36mGeoMonitoring with Python: Innovative Solutions for Latin America’s Challenges
31mA proof of concept Python toolkit for effective landscape fuel hazard management
26mFrom Data to Decision: Monitoring and Improving Water Networks using AI/ML
22mPython Applications in Infrastructure Planning and Civil Engineering
43mPython at City Scale
33mPython for Water Forecasting Services
37mSpatial Data Analysis With Python