Using spatial data to model biodiversity
summary
Spatial species distribution models combine occurrence records and environmental layers to make actionable maps. Examples: using R and QGIS with OpenStreetMap and Google Maps traffic data produces collision-risk estimates for 600,000 road segments (≈150,000 km); a planning workflow using predictive species maps plus a tool called donation reduces projected remnant vegetation loss from 20% to 12% in Perth Peel; global and national data sources feed models (79 million records in the Atlas of Living Australia, ~1 billion in GBIF).
AI-written summary · about
chapters (11)
- 0:00Introduction and thanks
- 1:03What are species distribution models
- 5:40Challenges in species data collection
- 8:53Predictor variables for modeling
- 11:21Conservation planning in Perth
- 15:44Wildlife-vehicle collision risk
- 19:20Disease mapping with spatial models
- 22:15Importance of data quality
- 24:04eBird as a data model
- 27:38Handling uncertainty in models
- 30:48Reliable predictions and future work
related talks
30mUnlocking Geospatial Data Science: A Strategic Overview
25mOpen Geospatial and Big Data for Airspace Risk Analysis
25mEverything but maps with spatial tools
50mGeospatial Analysis
30mLocation Intelligence and Spatial Data Science: from geographic coordinates to business insights
33mBuilding Data Science Tools for Sustainable Transformation
22mR & Space : GI Science with R
01h 36mFrom Maps to Models: An End-to-End Journey in Geospatial Machine Learning with Python