Urban Expansion Monitor
Detect and quantify urban expansion from multi-temporal built-up rasters.
CLI Usage
python scripts/urban_expansion_monitor.py --before before.tif --after after.tif
python scripts/urban_expansion_monitor.py --before before.tif --after after.tif --threshold 0.6
python scripts/urban_expansion_monitor.py --before before.tif --after after.tif --output-dir my_output
Data Download
This skill can auto-fetch a pair of Landsat Collection 2 Level-2 scenes
from the Microsoft Planetary Computer when given a bounding box + date
range. The --date-range is split at the midpoint: the first half is
used for --before and the second half for --after.
python scripts/urban_expansion_monitor.py \
--bbox 116,39,117,40 \
--date-range 2020-01-01,2024-12-31 \
--output-dir ./expansion-output
The PYTHONPATH must include the parent of _geoskill_data_fetcher/
(the same directory the 50 skills live in). Set it once:
export PYTHONPATH="/path/to/行业Skill创意-20260727"
Parameters
| Argument | Required | Default | Description |
|---|---|---|---|
--before |
Yes | — | Path to "before" built-up raster (continuous built-up index or binary mask) |
--after |
Yes | — | Path to "after" built-up raster (same grid as --before) |
--threshold |
No | 0.5 |
Built-up classification threshold (values ≥ threshold are considered urban) |
--output-dir, -o |
No | expansion-output |
Directory to write outputs |
Output
| File | Description |
|---|---|
expansion-report.json |
Machine-readable expansion stats (areas, growth rate) |
report.html |
Human-readable HTML report |
output-manifest.json |
Run metadata + result summary |
Exit Codes
| Code | Meaning |
|---|---|
| 0 | Success |
| 2 | Argument error (e.g. missing input file) |
| 7 | Processing failure |