xarray loader

Opens any file format supported by xarray.open_dataset: NetCDF, Zarr, GRIB, HDF5, and more. Accepts local paths, gs://, s3://, https://, or any fsspec-compatible URI.

Fields

Field	Description
urlpath	Path or URL to the dataset. Required.
engine	xarray engine to use: zarr (default), netcdf4, scipy, cfgrib, …
open_kwargs	Extra keyword arguments forwarded verbatim to xarray.open_dataset.
chunks	Dask chunk sizes for lazy loading.
mapping	Dict of {source_name: target_name} variable renames applied after opening.
sel	Label-based selection. Values can be scalars, lists, or {start, stop} dicts (converted to slice).
isel	Integer-index selection. Same value types as sel.

Example

source:
  type: xarray
  urlpath: /data/wave_hindcast.zarr
  engine: zarr
  chunks:
    time: 100
    latitude: 50
    longitude: 50
  mapping:
    Hs: hs
    Tp: tp
  sel:
    latitude: {start: -50, stop: -10}
    longitude: {start: 110, stop: 180}

API reference

gridstats.loaders.xarray.XarrayLoader

Load datasets using xarray.open_dataset / open_zarr.

load(config: XarraySourceConfig) -> xr.Dataset

Open, rename, and slice a dataset from a file or URL.

Parameters:

Name	Type	Description	Default
config	XarraySourceConfig	xarray source configuration.	required

Returns:

Type	Description
Dataset	Lazily loaded, preprocessed xarray Dataset.