Melbourne: land change, 2015–2025
Two Sentinel-2 satellite images of a 20 km square around Melbourne, from 2015-11-18 and 2025-01-21, were classified by a ResNet18 model trained on EuroSAT. Every 640 m cell is put in one of four land-cover groups, and the two dates are compared cell by cell.
- Before
- 2015-11-18
- After
- 2025-01-21
- Area analysed
- 393 km²
Change
+6.5 km²net change in built-up area
In cells the model is sure of on both dates (confidence ≥ 0.8): 0 km² became built-up, 1.2 km² changed in total.
The model is undecided on this scene; treat the results with caution.
The two images are from different seasons. Seasonal differences can look like change; pick the same month for a more reliable result.
Land cover on both dates
| 2015 | 2025 | ± | |
|---|---|---|---|
| Agriculture no notable change | 0.1% | 0.2% | +0.1 |
| Natural vegetation decreased | 7.6% | 5.3% | -2.3 |
| Built-up increased | 76% | 77.7% | +1.7 |
| Water no notable change | 16.2% | 16.8% | +0.6 |
Largest transitions (confident cells)
- Built-up → Water0.8 km²
- Built-up → Agriculture0.4 km²
Limits
- Every value shown is a model estimate, not a measurement.
- Areas are not measurements but approximations counted from cells. The figures are always computed on 640 m cells; the scale only changes how the map is drawn.
- The model was trained on European images only; results for places tagged "experimental" are less reliable.
- Clouds, shadows and seasonal differences can show change that is not there.