satellite_imagery.md•2.05 kB
### Satellite Imagery (Microsoft Planetary Computer via `pystac-client`)
The **Satellite Imagery** tool in GIS-MCP enables downloading analysis-ready satellite scenes (e.g., Sentinel-2, Landsat) directly from the [Microsoft Planetary Computer](https://planetarycomputer.microsoft.com/).
It automatically selects the least-cloudy image matching your search criteria and prepares a multi-band GeoTIFF.
---
#### Installation
To enable satellite imagery downloads, install GIS-MCP with the **satellite-imagery** extra:
```bash
pip install gis-mcp[satellite-imagery]
```
#### Parameters
- **collection** – STAC collection ID (e.g., `"sentinel-2-l2a"`, `"landsat-8-c2-l2"`)
*(default: `"sentinel-2-l2a"`)*
- **assets** – Bands or asset keys to download (list or comma-separated string).
Common Sentinel-2 keys:
- `B04` → Red
- `B03` → Green
- `B02` → Blue
- `B08` → Near Infrared (NIR)
*(default: `["B04","B03","B02"]`)*
- **datetime** – Date or date range in ISO 8601.
Examples:
- `"2025-08-05"` → single day
- `"2025-08-01/2025-08-31"` → range
*(default: `"2024-01-01/2024-12-31"`)
- **cloud_cover_lt** – Maximum cloud cover percentage (integer).
Use `None` to disable filtering.
*(default: `20`)*
- **bbox** *(optional)* – Bounding box string `"minx,miny,maxx,maxy"` in WGS84 coordinates.
- **geometry_geojson** *(optional)* – A GeoJSON geometry (string).
If provided, the image is clipped precisely to this geometry.
- **out_crs** *(optional)* – Target CRS for output (e.g., `"EPSG:4326"`).
If omitted, the asset’s native CRS is preserved.
- **filename** *(optional)* – Custom filename for the output GeoTIFF.
*(default: auto-generated from collection, item ID, and asset keys)*
- **path** *(optional)* – Output directory.
*(default: `./data/satellite_imagery` inside the package)*
#### Example Usage
```bash
Using gis-mcp download Sentinel-2 RGB imagery for Iran during August 2025 with less than 15% cloud cover.
```