GWR local R² — Earth 2026

Vulnerability & census

Where the three-predictor local regression explains heat risk index variation best. Use this to see who lives in hotter areas and where social sensitivity is highest.

Open this layer on the map All layers

Legend scale: 0.1 – 1 R²

Map preview of GWR local R² — Earth 2026 — Preview uses a continuous legend from 0.1 to 1 R²

What it shows

Local R² from local regression windows (0–1). Peripheral neighbourhoods often show higher fit than complex central zones.

Data sources

Same local regression specification as β_DSI layer.

How it was prepared

Map adaptation of the cited study — same neighbourhood unit and index structure, not the full 2001–2024 longitudinal analysis. Only Census 2021 is available in this project (no 2001 census neighbourhood tables). Paper: local R² for ΔHRPI (Model 1c). Here: R² for 2024 heat risk index levels — useful for spatial fit patterns, not directly comparable to published Fig. 5 values.

Citation

Bečić, D.; Gašparović, M. (2026). Disentangling Climate and Demographic Drivers of Urban Heat Risk: A Geographically Weighted Regression Analysis of Zagreb (2001–2024). Earth, 7, 72. DOI: 10.3390/earth7030072