Understanding geospatial data with duckdb
summary
Shows that DuckDB can store and query vector geodata with the spatial extension and HTTPFS to load GeoJSON from URLs, run ST_Length/ST_Area/ST_Within and create 2.4 km square buffers in SQL, and join those to admin boundaries (Colombia admin0/1/2) and a 5,000-point RWI CSV. DuckDB has no native raster engine, so a Python UDF (rasterstats) computes zonal population sums over the buffered areas. Demo builds a Lambda pipeline that writes GeoJSON to S3 (Prism) and it works end to end.
AI-written summary · about
chapters (12)
- 0:00Introduction to GIS
- 8:03Vector vs raster data
- 12:19DuckDB geospatial setup
- 17:38Creating geometries in DuckDB
- 21:51Loading geospatial formats
- 23:58Visualizing geospatial data
- 25:24Spatial relations and joins
- 27:53Projection challenges
- 30:07Buffer creation and analysis
- 31:57Raster statistics with Python
- 35:10WFP Prism application demo
- 37:18Q&A session
related talks
25mA quack of all trades: DuckDB, a versatile analytical database to keep in your toolkit
28mAnalyze data easily with duckdb - and the implications on data architectures
40mDuckDB: Harnessing in-process analytics for data science and beyond
29m"Serpents and Ducks: wrangling data with Python and DuckDB"
34mData Wrangling with DuckDB
33mData preparation and analytics on your laptop with DuckDB
27mThe Duck and the DataFrame: A Data Engineer’s Journey with DuckDB
40mDuckDB beyond the notebook