Converts daily IceNet netCDF prediction outputs to COG files + STAC catalogs for icenet-dashboard
.
A command-line tool for generating Cloud Optimized GeoTIFFs (COGs) and STAC catalogs from IceNet forecast NetCDF files. It is outputs COGs in the necessary format to serve them using icenet-dashboard-preprocessor for use with the icenet-dashboard
.
- Converts IceNet NetCDF predictions into Cloud Optimized GeoTIFFs (COGs)
- Automatically reprojects to EPSG:3857 for Leaflet/web-map compatibility
- Builds STAC-compliant metadata catalogs for each forecast
- Supports compressed or uncompressed output
- Dynamically detects northern or southern hemisphere from input data
To use this tool, ensure the following dependencies are installed:
pip install xarray rioxarray pystac tqdm
icenet_dashboard_preprocess --input ./results/predict/*.nc
Flag | Description |
---|---|
--input , -i |
One or more paths to .nc files or directories (wildcards supported). |
--overwrite , -o |
Overwrite existing GeoTIFF files if they already exist. |
--no_compress , -c |
Disable compression in generated GeoTIFFs (default is compressed). |
icenet_geotiff_generator_gen_cloud_tiffs -i results/predict/fc.2024-11-11_north.nc -o
This will:
- Detect the hemisphere automatically
- Convert each leadtime slice to a COG
- Process the COG outputs into
data/cogs/{hemisphere}/{date}/
- Build a hierarchical STAC catalog in
data/stac/
icenet-dashboard-preprocessor
is licensed under the MIT license. See LICENSE for more information.