The European Space Agency

FORCE Time Series Analysis

Recap

Warning

You should use pystac_client based querying if your inputs encompass more than just a handful of items

L1C_COLLECTION_URL = "https://stac.dataspace.copernicus.eu/v1/collections/sentinel-2-l1c"
STAC_ROOT_URL = "https://stac.dataspace.copernicus.eu/v1"

w,s,e,n = 10.386, 44.437, 11.423, 44.973
spatial_extent = { "west": w, "south": s, "east": e, "north": n}
# ATTENTION: Inclusion of the second date in the search depends on the query method (see below).
temporal_extent=["2026-04-17", "2026-05-01"]
import openeo
connection = openeo.connect("openeo.dataspace.copernicus.eu").authenticate_oidc()

query_pg = openeo.processes.process(
    "query_stac",
    arguments={
        "url": L1C_COLLECTION_URL,
        "temporal_extent": temporal_extent,
        "spatial_extent": spatial_extent,
    }
)
Authenticated using refresh token.
import pystac_client

client = pystac_client.Client.open(STAC_ROOT_URL)
search = client.search(
    datetime=temporal_extent,
    collections=["sentinel-2-l1c"],
    bbox=[w, s, e, n]
)
item_collection = search.item_collection()

import getpass
from openeo.rest.stac_resource import StacResource
from openeo.internal.graph_building import PGNode

# Make sure to choose a unique name
processing_name = "Modena"

force_l2_stac_resource = StacResource(
    graph=PGNode(
        process_id="force_level2",
        arguments={
            "stac_document": query_pg,
            "name": processing_name,
            "do_brdf": True,
        }
    ),
    connection=connection,
)

merge = f"{getpass.getuser()}/experiments/{processing_name}"
force_l2_stac_resource.export_workspace(
    workspace="apex-force-results-workspace",
    merge=merge,
);
from openeo.internal.graph_building import PGNode
from openeo.rest.stac_resource import StacResource

query_catalog = transform_item_collection_to_catalog_with_links(item_collection)

processing_name = "Modena"

force_l2_stac_resource = StacResource(
    graph=PGNode(
        process_id="force_level2",
        arguments={
            "stac_document": query_catalog.to_dict(),
            "name": processing_name,
            "do_brdf": True,
            # level 2 parameters
        }
    ),
    connection=connection,
)

merge = f"{getpass.getuser()}/experiments/{processing_name}"
print(merge)
force_l2_stac_resource.export_workspace(
    workspace="apex-force-results-workspace",
    merge=merge,
)
hannes/experiments/Modena

l2_job = force_l2_stac_resource.create_job()
l2_job.start()

Time Series Analysis

The openEO process for FORCE Time Series Analysis (TSA) is called force_tsa. Like the level 2 process, it takes most of the parameters usually passed to the TSA module of FORCE using the parameter file. Parameters use snake_case format instead of SCREAMING_SNAKE_CASE (used in FORCE parameter files).

Details of parametrization are discussed in the dedicated article.

connection = openeo.connect("openeo.dataspace.copernicus.eu").authenticate_oidc()
Authenticated using refresh token.
connection.describe_process("force_tsa")
Authenticated using refresh token.

Using FORCE level 2 results as an input

The process graph for FORCE TSA built in the same way as the level 2 process graph is built. The main input for FORCE TSA is the STAC catalog (or item) from the level 2 processing.

We can access the STAC catalog produced by our level 2 job in the workspace, described in level 2 processing article.

Important

It is important to remember the merge parameter passed to export_workspace, this is the path under which we can find the catalog later.

processing_name = "Modena"
merge = f"{getpass.getuser()}/experiments/{processing_name}"
print(merge)
hannes/experiments/Modena

Remember, the dedicated workspace for FORCE has the URL

WORKSPACE_URL = "https://s3.waw4-1.cloudferro.com/apex-force-results-waw4-1-exotc5yuexi2c5tvwqhoivj62fz8v0uupy0me"

TSA Parameters

Some parameters can be set a-priori or from prior knowledge of our level 2 processing parameters

temporal_extent=["2026-04-17", "2026-04-27"]

For other parameters, we may want to inspect the metadata of the level 2 processing. For example, we can extract the tiles from the asset names of the STAC item describing the level 2 datacube.

merge = f"{getpass.getuser()}/experiments/{processing_name}"
l2_catalog_url = f"{WORKSPACE_URL}/{merge}/catalog.json"
print(l2_catalog_url)
l2_catalog = pystac.Catalog.from_file(l2_catalog_url)
l2_item = next(l2_catalog.get_items()) # there is just one item
l2_item
https://s3.waw4-1.cloudferro.com/apex-force-results-waw4-1-exotc5yuexi2c5tvwqhoivj62fz8v0uupy0me/hannes/experiments/Modena/catalog.json
<Item id=Modena-level2>

We can extract the tiles manually or programmatically

tiles = set([k.split(".")[0] for k in l2_item.assets.keys() if k.endswith("BOA")])
tile_x = set(int(tile[1:5]) for tile in tiles)
tile_y = set(int(tile[7:]) for tile in tiles)
tile_x, tile_y
({31, 32}, {29, 30})

Process Graph

The main process is force_tsa.

from openeo.rest.stac_resource import StacResource
from openeo.internal.graph_building import PGNode

tsa_processing_name = f"{processing_name}_TSA"

force_tsa_stac_resource = StacResource(
    graph=PGNode(
        process_id="force_tsa",
        arguments={
            "stac_url": l2_catalog_url,
            "name": tsa_processing_name,
            "date_range": temporal_extent,
            "x_tile_range": [min(tile_x), min(tile_x)],
            "y_tile_range": [min(tile_y), max(tile_y)],
            "stm": ["AVG"],
            "output_stm": True,
            # ... more TSA parameters (optional)
        }
    ),
    connection=connection,
)

We can then create and start the job as before:

tsa_job = force_tsa_stac_resource.create_job(title=tsa_processing_name)
tsa_job.start() # use start_and_wait to block until finished

Downloading assets works the same way as for level2:

tsa_results = tsa_job.get_results()
tsa_results
tsa_results.download_files("force-tsa-results")
[PosixPath('force-tsa-results/CITEME_0x31.txt'),
 PosixPath('force-tsa-results/X0031_Y0029/20260417-20260427_001-365_HL_TSA_SEN2L_EVI_STM.tif'),
 PosixPath('force-tsa-results/X0031_Y0029/20260417-20260427_001-365_HL_TSA_SEN2L_NBR_STM.tif'),
 PosixPath('force-tsa-results/X0031_Y0029/20260417-20260427_001-365_HL_TSA_SEN2L_NDVI_STM.tif'),
 PosixPath('force-tsa-results/X0031_Y0030/20260417-20260427_001-365_HL_TSA_SEN2L_EVI_STM.tif'),
 PosixPath('force-tsa-results/X0031_Y0030/20260417-20260427_001-365_HL_TSA_SEN2L_NBR_STM.tif'),
 PosixPath('force-tsa-results/X0031_Y0030/20260417-20260427_001-365_HL_TSA_SEN2L_NDVI_STM.tif'),
 PosixPath('force-tsa-results/datacube-definition.prj'),
 PosixPath('force-tsa-results/job-results.json')]

Further processing

Because level 2 results are provided by a STAC URL to the TSA process, it is possible run multiple TSA jobs with the same level 2 input, without expensively reprocessing the level 2 data. We just create a new TSA process graph and submit it, passing the same level 2 data cube URL.

Remember that the job assets will be stored for some time on the backend which allows you to do processing over a span of a few days, but will not be stored permanently. Please download and archive results you wish to keep on your own infrastructure.