Scaling the Urban Forest: An Integrated Framework for Managing Cities by Fusing Raster(…)
summary
A reproducible pipeline converts LiDAR rasters and vector building/park data into building-level metrics across England, producing 190 million trees segmented with crown top, height and area, canopy cover per neighbourhood, and 3-30-300 rule quantified nationally. The stack uses R for LiDAR segmentation, Python libraries (GeoPandas, Sedona, xarray) for tiling and parallelism, plus graph-based Dijkstra on the road network to compute walking distance to park access points.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:07Why trees matter
- 2:113-30-300 urban forestry
- 3:13Mapping urban trees tools
- 4:41Tree segmentation process
- 5:26Canopy cover per neighborhood
- 5:47Distance to nearest park
- 6:31Satellite imagery integration
- 7:14Reproducible open-source tools
- 8:19Q&A: Distance to parks
- 9:01Q&A: Trees and deprivation
- 10:24Closing remarks
related talks
43mPython at City Scale
27mSmart Forestry: AI in the Service of Nature
22mUnderstanding our cities with FOSS4G
21mActual trees, not b-trees: or how I found PostgreSQL through PostGIS
31mUnpacking the geospatial engineering toolbox – an overview of data science techniques for spatial da
27mCartography on the Web
14mGeographies of Linear Infrastructure
26mWho Said Wrangling Geospatial Data at Scale was Easy?