Creating the Datasets Metadata

The records of datasets in the ODC-Core catalog are made in the same way as the products: given a definition in YAML file, it is possible to make the addition. In this case, stac2odc offers a utility to map BDC-STAC items in datasets.

Note

Before executing the utility, it is necessary to understand that this process requires the BDC data repository; this is because the path to the data files is required during the datasets’ description. With this in mind, stac2odc offers the option --download, which allows you to download the data and then make its description. For this tutorial, this facility will be used.

To extract the metadata from the datasets, it is necessary to define the collection, the amount of data considered, and the options of unit and instrument presented previously:

sudo stac2odc item2dataset -c CB4_64_16D_STK-1 \
                           -o ~/products/CB4_64_16D_STK_1/datasets/ \
                           --units m \
                           -p CBERS4 \
                           --instrument WFI \
                           -m 3 \
                           --download \
                           --download-out /data \
                           --access-token BDC_ACCESS_TOKEN

The above command will download 03 items from the data cube CB4_64_16D_STK-1, saved in the /data directory. The descriptions of these items will be placed in the directory ~/products/CB4_64_16D_STK_1/datasets/. The directory tree created in the /data folder follows the same organization used in the BDC.

Note

Access to the data, done by the above command, depends on authentication to the Brazil Data Cube project services. Replace the value BDC_ACCESS_TOKEN with your access key. If you do not have this key, it can be generated through the BDC Portal .