User Guide
This user guide is currently being written.
Obtaining the Data Store
After installation and once you have your
credentials at hand, you can use the SMOS data store
using its class exported by the xcube_smos.store module:
from xcube_smos.store import SmosDataStore
store = SmosDataStore(key="your access key", secret="your secret")
However, the preferred way to obtain the store object is by its name "smos"
and using the xcube new_data_store() function, because many other xcube
data stores can be used in the same way:
from xcube.core.store import new_data_store
store = new_data_store("smos", key="your access key", secret="your secret")
Note
You can avoid passing key and secret arguments if you set the environment
variables CREODIAS_S3_KEY and CREODIAS_S3_SECRET accordingly.
The new_data_store() function in its general form can take arbitrary keyword
arguments. The store identifier, here "smos" determines the allowed keywords.
You can inspect the allowed data store keywords by using the xcube function
get_data_store_params_schema(), which outputs the allowed parameters as a
JSON Schema object:
from xcube.core.store import get_data_store_params_schema
get_data_store_params_schema("smos")
Data Store Parameters
Using the data store's get_open_data_params_schema() method you can
inspect the allowed parameters for the store.open_data() method, which is
used to access the SMOS data in various forms.
open_schema = store.get_open_data_params_schema()
print(open_schema)
Data Access
All xcube data stores provide a open_data() method to access the data.
It has one required positional argument data_id which identifies the
data(set) to be opened. The SMOS store provides two datasets, they are
"SMOS-L2C-SM"- SMOS Level-2C Soil Moisture"SMOS-L2C-OS"- SMOS Level-2C Ocean Salinity
In the xcube data store framework, the different data representations
are provided by dedicated data openers. Hence, a common and optional
argument is opener_id, which is used to control how the data is
represented. The SMOS data store can currently provide three
different representations of SMOS data addressing different use cases:
"dataset:zarr:smos"(the default) - represent data as a datacube including all observations in the given time range at a fixed spatial resolution;"mldataset:zarr:smos"- represent data as a multi-resolution datacube including all observations in the given time range including 5 spatial resolution levels;"smosdsiter:zarr:smos"- represent data as an iterator providing datasets for all the observations in the given time range at a fixed spatial resolution.