Six case studies covering catchment hydrology and flood constraint, coastal erosion and defences, urban heat, and environmental change. These aren't just screenshots. For each one I've written up the question, the method, the limitations, and what I actually found.
The groundwork of flood-risk analysis: a full raster hydrology run on an SRTM DEM — fill, flow direction, accumulation, Strahler stream ordering, and watershed delineation across the Dovey catchment, validated against the real river.
Multi-temporal cliff-line mapping (2003–2025) with custom Python tools and DSAS. Mean retreat 4.72 m/yr across 18 transects (all R² ≥ 0.99). A 20-year projection of 80–113 m.
Hierarchical multi-criteria analysis over flood risk, terrain, geology, ecology and transport layers, screening out floodplain land before recommending Site B (Bow Street).
The same DSAS method on a defended, hard-rock coast. Mean LRR 0.20 m/yr at R² 0.34, which shows how sea defences scramble the erosion signal.
NASA ARSET training: ECOSTRESS land surface temperature in Google Earth Engine, Random-Forest sharpening to 10 m, and a nighttime urban-heat-island analysis in R.
Grinnell Glacier between 2009 and 2016. It lost 9.09% of its area, measured from free satellite imagery.
Analysis of shoreline change across the entire Holderness Coast. Details to be confirmed.