xcube-smos

User-defined datacubes from SMOS Level-2 data

xcube-smos is a Python package and xcube plugin that adds a data store named smos to xcube. The data store is used to access ESA SMOS Level-2 data in form of analysis-ready geospatial datacubes with the dimensions time, lat, and lon. The datacubes are computed on-the-fly from the SMOS data archive s3://EODATA/SMOS hosted on CREODIAS.

Usage

After installation, data access is as easy as follows:

from xcube.core.store import new_data_store

store = new_data_store("smos", **credentials)
datacube = store.open_data(
    "SMOS-L2C-SM", 
    time_range=("2022-01-01", "2022-01-06"),
    bbox=(0, 40, 20, 60)
)

Above, a datacube of type xarray.Dataset for SMOS Soil Moisture has been obtained.
To access SMOS Ocean Salinity data use the identifier "SMOS-L2C-OS".

Features

The SMOS data is provided using a geographic projection. Users can choose from five spatial resolutions by specifying a resolution level ranging from zero to four. Zero refers to a resolution of 360/8192 ~ 0.044 degrees ~ 4.88 km. Higher levels subsequently increase the resolution by a factor of two.

Note

The native spatial resolution of SMOS data is roughly 25 km. xcube-smos performs an oversampling of the data to ensure no information is lost during spatial projection. It therefore uses a nearest-neighbor resampling at higher resolution involving pixel duplication.

xcube-smos does not perform any aggregation in the time dimension. Data is provided as-is, that is, up to 29 SMOS Level-2 data products are included per day.