Using R, Python, and Cloud Infrastructure to Battle Aquatic Invasive Species
summary
Minnesota has ~18–24k registered water bodies, about 9,000 with inspection data and roughly 10% infested; the project stitches four models — a Python risk model (predicting ~8‑year invasion probability), an intervention simulator, an inspection optimizer, and a collaboration scenario model — into a Shiny AIS Explorer. Hosting uses EC2 with AMIs, autoscaling + load balancer, precomputed outputs and a processing instance that can run 36–48 hours. Change detection runs via Lambda + EventBridge + S3, and heavy jobs are packaged for OCPU and returned by email.
AI-written summary · about
chapters (11)
- 0:00Intro
- 2:32Background
- 5:48Accessibility
- 8:25Keeping it Relevant
- 12:37Chrome Job
- 12:48Performance
- 13:42Load Balancing
- 15:37Performance tuning
- 16:08Running expensive models
- 17:02Conclusion
- 17:56Questions
related talks
21mAquawatch Data Systems a cloud solution for water quality monitoring and forecasting
48mAccelerate science, research, and innovation with AWS
20mHow the cloud changed the way I do analytics and research
18mData science and machine learning in wastewater intelligence
17mIt's All in the Cloud
33mBuilding Data Science Tools for Sustainable Transformation
19mHow to win an AI Hackathon, without using AI
28mTurn Data Into Insights