A Python tool to process MeteoSwiss radar products (ODIM HDF5 and Metranet, in ZIP archives) into a consolidated Zarr archive for efficient analysis with xarray.
- Reads daily radar products (ZIP/HDF5/Metranet).
- Converts to
xarray.Datasetwith proper coordinates and metadata. - Aligns products on a common grid and time reference.
- Adds missing variables as NaNs for consistency.
- Writes results into a chunked Zarr store (appendable by time).
- Parallel processing with configurable workers.
python radar2zarr.py \
--cfgpath config.yaml \
--starttime 20230101 \
--endtime 20230105
- populate_zarr_db.py: main script
- run.sbatch: example SLURM job to populate database
- config: examples of configuration files
- postproc: set of scripts to postprocess the database (concat zarrs, rechunking, etc)
- examples_extraction: set of scripts to perform operations on the zarr (watershed extraction for example)
- geodata: contains geojson files of watersheds