The European Space Agency

Selecting input data

When running FORCE level 2 locally, we are responsible to manually provide the necessary input data and reference their full path in the FILE_QUEUE parameter of the FORCE level 2 parameter file.

When running on CDSE, we do not have access to the excuting environment and its filesystem directly. Downloads, file management and creation of the file queue are managed by the integration. To specify which files should be processed, we provide a STAC document (Catalog, Collection, Item or ItemCollection) to the process.

Warning

Make sure that the STAC document provided to force_level2 only references the files that should actually be processed. The FORCE integration will attempt to download all products referenced in the document, so it is inadvisable to provide a reference to something like the entire Sentinel-2 L1C collection on CDSE.

STAC

The integrated FORCE and openEO more generally uses the SpatioTemporal Asset Catalog (STAC) specification to identify (groups of) EO products and annotate them with metadata.

If you are unfamiliar with STAC, we recommend that you familiarize yourself with the basic components of the specification (What are catalogs, items and feature collections?) on the official STAC specification website before proceeding.

The CDSE Sentinel-2 L1C STAC collection which we use as input for FORCE level 2 processing, is hosted at https://stac.dataspace.copernicus.eu/v1/collections/sentinel-2-l1c

In the following, we show how this collection may be queried to to obtain inputs for FORCE level 2 processing.

Querying STAC catalogs/collections

FORCE level 2 can work with any STAC URL or document referencing input products of the correct type (Sentinel-2 L1C). The following describes how to make a query to the CDSE STAC catalog. Be advised to query the CDSE STAC collection of L1C, as data access will be faster than external archives and the integration makes some assumptions about the metadata provided in the STAC items describing the products.

It is recommended to make the query using pystac_client for now, as using the query_stac process limits us to very small data sets See below for more details.

Area of interest

As an example, we selected an area in northern Italy, around the city of Modena.

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"]
Warning

When defining the temporal extent as shown above, be aware that different methods may interpret time ranges differently. OpenEO’s query_stac behaves just like load_collection: The second instant is excluded from the search.

Contrarily, pystac_client’s search method includes the second date when given as a list. Note that pystac_client’s search has many more options for specifying time ranges.

AOI plot 
import matplotlib.pyplot as plt
import contextily as ctx
import geopandas as gpd
from shapely import Polygon

def plot_area_of_interest(
    w, s, e, n, *,
    context=5,
    color="red",
    linewidth=2,
    title="Spatial Extent",
    figsize=(12, 8),
    show_tick_labels=True,
    ):
    polygon = Polygon([(w, s), (e, s), (e, n), (w, n)])
    gdf = gpd.GeoDataFrame(geometry=[polygon], crs="EPSG:4326")
    fig, ax = plt.subplots(1, 1, figsize=figsize)
    gdf.plot(edgecolor=color, facecolor="none", linewidth=linewidth, ax=ax)
    gdf.plot(edgecolor=color, facecolor="none", linewidth=linewidth, ax=ax)
    ax.set_xlim(w - context, e + context)
    ax.set_ylim(s - context, n + context)
    if not show_tick_labels:
        ax.set_xticks([])
        ax.set_yticks([])
    ctx.add_basemap(ax, source=ctx.providers.OpenTopoMap, crs=gdf.crs)
    fig.suptitle(title)
    fig.tight_layout()

plot_area_of_interest(
    w=w,
    s=s,
    e=e,
    n=n,
    context=3,
    figsize=(10, 6),
    title=f"Spatial Extent: Modena",
    show_tick_labels=False
)

Querying a catalog

import openeo

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

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.
Authenticated using refresh token.

We can directly use query_pg as a parameter for FORCE level 2, using openEO’s process chaining. To inspect the query results ahead of time, we may want to execute it ahead of time. If we do not do this, execution of the query will be delayed until we submit the entire process graph (including processing with FORCE).

query_res = connection.execute(query_pg)

print(f"Found products ({len(query_res['features'])}):")
print("\n".join([f["id"] for f in query_res["features"]]), "\n")
Found products (18):
S2C_MSIL1C_20260430T102021_N0512_R065_T32TPQ_20260430T135954
S2C_MSIL1C_20260430T102021_N0512_R065_T32TNQ_20260430T135954
S2A_MSIL1C_20260429T101041_N0512_R022_T32TPQ_20260429T152547
S2A_MSIL1C_20260429T101041_N0512_R022_T32TNQ_20260429T152547
S2C_MSIL1C_20260427T101021_N0512_R022_T32TPQ_20260427T142022
S2C_MSIL1C_20260427T101021_N0512_R022_T32TNQ_20260427T142022
S2B_MSIL1C_20260425T101559_N0512_R065_T32TPQ_20260425T142712
S2B_MSIL1C_20260425T101559_N0512_R065_T32TNQ_20260425T142712
S2A_MSIL1C_20260422T102041_N0512_R065_T32TPQ_20260422T153540
S2A_MSIL1C_20260422T102041_N0512_R065_T32TNQ_20260422T153540
S2B_MSIL1C_20260422T101019_N0512_R022_T32TPQ_20260422T134442
S2B_MSIL1C_20260422T101019_N0512_R022_T32TNQ_20260422T134442
S2C_MSIL1C_20260420T102021_N0512_R065_T32TPQ_20260420T135843
S2C_MSIL1C_20260420T102021_N0512_R065_T32TNQ_20260420T135843
S2A_MSIL1C_20260419T100711_N0512_R022_T32TPQ_20260419T152521
S2A_MSIL1C_20260419T100711_N0512_R022_T32TNQ_20260419T152521
S2C_MSIL1C_20260417T101021_N0512_R022_T32TPQ_20260417T134842
S2C_MSIL1C_20260417T101021_N0512_R022_T32TNQ_20260417T134842 

import pystac_client
import time

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

client = pystac_client.Client.open("https://stac.dataspace.copernicus.eu/v1")
search = client.search(
    datetime=temporal_extent,
    collections=["sentinel-2-l1c"],
    bbox=[w, s, e, n]
)
item_collection = search.item_collection()

print("Found items:")
for item in search.items():
    time.sleep(1) # CDSE Stac API has a rate limit of 5 requests per second
    print(item.id)
Found items:
S2C_MSIL1C_20260430T102021_N0512_R065_T32TPQ_20260430T135954
S2C_MSIL1C_20260430T102021_N0512_R065_T32TNQ_20260430T135954
S2A_MSIL1C_20260429T101041_N0512_R022_T32TPQ_20260429T152547
S2A_MSIL1C_20260429T101041_N0512_R022_T32TNQ_20260429T152547
S2C_MSIL1C_20260427T101021_N0512_R022_T32TPQ_20260427T142022
S2C_MSIL1C_20260427T101021_N0512_R022_T32TNQ_20260427T142022
S2B_MSIL1C_20260425T101559_N0512_R065_T32TPQ_20260425T142712
S2B_MSIL1C_20260425T101559_N0512_R065_T32TNQ_20260425T142712
S2A_MSIL1C_20260422T102041_N0512_R065_T32TPQ_20260422T153540
S2A_MSIL1C_20260422T102041_N0512_R065_T32TNQ_20260422T153540
S2B_MSIL1C_20260422T101019_N0512_R022_T32TPQ_20260422T134442
S2B_MSIL1C_20260422T101019_N0512_R022_T32TNQ_20260422T134442
S2C_MSIL1C_20260420T102021_N0512_R065_T32TPQ_20260420T135843
S2C_MSIL1C_20260420T102021_N0512_R065_T32TNQ_20260420T135843
S2A_MSIL1C_20260419T100711_N0512_R022_T32TPQ_20260419T152521
S2A_MSIL1C_20260419T100711_N0512_R022_T32TNQ_20260419T152521
S2C_MSIL1C_20260417T101021_N0512_R022_T32TPQ_20260417T134842
S2C_MSIL1C_20260417T101021_N0512_R022_T32TNQ_20260417T134842

Inspecting query results

We may use pystac to inspect or process the feature collection further. In this example, we inspect the first of the returned items:

import pystac

item = pystac.Item.from_dict(query_res["features"][0])
item
<Item id=S2C_MSIL1C_20260430T102021_N0512_R065_T32TPQ_20260430T135954> 
next(iter(search.items()))
<Item id=S2C_MSIL1C_20260430T102021_N0512_R065_T32TPQ_20260430T135954>

Other methods

CDSE provides a graphical STAC browser that can be used to search for and select singular items.