SWA Flood Mapping
summary
Rapid pipeline converts Sentinel-2 (and fallback Sentinel-1/SAR) imagery via Google Earth Engine into sedimentation-based flood masks, vectorizes them with GDAL/PostGIS and GeoPandas, intersects with Stats NZ polygons, and computes a weighted impact score per area that factors demographics and proximity. CI/CD via Docker + make auto-builds vector tiles with T-Rex, pushes to S3 and vector tiles on S3 with CDN restart so updates go live in ~20 minutes. Beta appears in 3 days, public live within one week; stack is repeatable and works.
AI-written summary · about
chapters (7)
- 0:00Introduction and project overview
- 2:28Project goals and requirements
- 7:03Team structure and workflow
- 8:27Technical stack and tools used
- 13:19Satellite imagery processing challenges
- 17:33Outcomes and beta delivery
- 20:48Live demo and Q&A
related talks
16mOpen Data to plan Flood management in Assam
45mFrom Radar Echoes To Real-Time Recognition: Deep Learning on Another Planet an Open Source Workflow
40mHow do you harness crowd sourcing for a community that also delivers for business?
17mFrom Swipes to Solutions: Exploring the Impact of MapSwipe on Humanitarian Response
23mFOSS4G Pasifika Roadmap: Then/Now/Beyond
28mDiscover UMap – A free QGIS Plugin for effortless Visualisation of bathymetric data
20mRecoding our workflows to FOSS4G
13mAdding Open Data to Rapid through Esri