12-3. Inspecting STAC data sources¶
In this exercise you will load a prepared configuration of STAC layers and inspect how each data source is defined.
Load the starting configuration¶
- Click Load in the Configuration Builder.
- Open the Examples tab.
- Under Test & development, select STAC datasets.
The configuration contains a set of layers whose data comes straight from STAC catalogues, alongside the usual base layers.
Recognise a STAC data source¶
- Open the layer card for STAC Tillage simple.
-
Look at its data source row. Two things mark it as STAC:
- the format badge shows stac, and
- a muted badge on the right names the asset and its real format —
here
data (flatgeobuf).
-
Open a few other layer cards — Austria LST STAC item, Africa January 2023, Sentinel 2 STAC Query - June 2026 — and note that each shows the same pattern: a STAC address plus one named asset.
Why the asset badge matters
The STAC address only says where to ask; the asset badge says what the Explorer will actually fetch and display. The same collection could offer a COG, a FlatGeoBuf and a thumbnail — the badge tells you which one the layer uses.
Inspect the metadata¶
- On the STAC Tillage simple data source row, click the (i) button to open STAC Metadata.
-
Work through the tabs:
- Collection — the title, description, licence and spatial and temporal extents of the whole collection.
- Item — the single item, or the first item returned by the layer's filters, with its date, footprint and properties.
- Asset — details of the selected asset, with a button that opens the COG or FlatGeoBuf file inspector you met in earlier tutorials.
- Queryables — the properties the collection allows you to filter on.
-
Close the dialog and open the metadata for Austria LST STAC item. Because this address points at one item, the Item tab describes exactly that scene.
Read-only
The metadata view never changes your configuration, and temporary or signed file links are never saved.
Compare the layer types¶
The configuration contains one of each common shape. Open each layer card and compare the data source addresses:
| Layer | Address shape | What it means |
|---|---|---|
| STAC Tillage simple | .../collections/tillage_type_detection |
A whole collection — every item is used. |
| Austria LST STAC item | .../collections/ECOSTRESS_LST/items/ECOSTRESS_MEAN_LST_2024 |
One specific item. |
| Africa January 2023 | .../items?datetime=2023-01-01.../2023-01-31...&limit=10000 |
An item list filtered by date. |
| Morocco, Time Series | .../items?limit=10000&bbox=... |
An item list filtered by area. |
| Sentinel 2 STAC Query - June 2026 | .../search?...&filter=cloud_cover < 10&filter-lang=cql2-text |
A search query with area, date and a property filter. |
Notice that every filter — area, dates, limit, cloud cover — is part of the address itself. Nothing about the query is hidden elsewhere in the configuration.
Look at the query editor¶
- On the Sentinel 2 STAC Query - June 2026 layer card, edit the data source to open the query editor.
- The filters from the address appear as badges: the area (AOI), the date
range (TOI), the limit and the
cloud_cover < 10property filter, plus a collection badge. - Find the Query method choice — Items endpoint or Search endpoint — and the small Test button beside each.
- Click Test for each method and compare the results. Some services only evaluate property filters on the search endpoint, so the two counts can differ; the test tells you whether the server honours your filter.
- Close the editor without saving.
Export your configuration
Save your progress before moving on to the next exercise.