This Icechunk repository provides virtual access to NASA PACE OCI Level-3 chlorophyll (short_name PACE_OCI_L3M_CHL) products for in-region AWS access. See README below for how to read in data. See the ipynb notebook for how this store was created.
This Icechunk repository provides virtual access to NASA PACE OCI Level-3 chlorophyll (short_name PACE_OCI_L3M_CHL) products for in-region AWS access. The data remain in the original NASA S3 bucket and are accessed through virtual references, avoiding duplication while providing a Zarr-compatible interface.
The repository is organized by temporal resolution and spatial resolution from the cooresponding files in the PACE_OCI_L3M_CHL collection.
Each temporal/spatial-resolution dataset is split into two groups based on the chunking of the original NASA NetCDF files:
chunks_512: Files with chlorophyll chunked as (time=1, lat=512, lon=1024).chunks_16: Files with chlorophyll chunked as (time=1, lat=16, lon=1024).The NASA archive transitioned from 512-row latitude chunks to 16-row latitude chunks during the record. Because Zarr currently requires arrays appended along a dimension to have identical chunk shapes, the two chunk layouts must be stored as separate datasets.
Each dataset includes a time coordinate derived from the start date in the filename, allowing it to be opened as a standard xarray dataset with xr.open_zarr().
See pace-icechunk-examples.ipynb for more examples and plot examples. See environment.yml for the basic environment needed.
NASA Earthdata requires authentication. Use earthaccess to get the s3 credentials.
We can concat the 2 groups to make on dataset with all times.