Toni Del Hoyo wondered about Sentinel data publication latency, that is how long it takes for Sentinel data to become publicly available after acquisition:
The idea for a Sentinel Latency project had been sitting in my backlog for months. I kept wondering: how fast and predictable is Sentinel data publication in practice? Does latency vary by geography? Is Sentinel-1 different from Sentinel-2? Do areas closer to polar ground stations get published faster? Are there patterns that suggest technical constraints, operational policies, or geopolitical factors?
So Toni built a web app that tracks publication latency for Sentinel-1 and Sentinel-2 data and visualises the results on a map for either a specific day or a date range1. Watch out: Visualising long date ranges can be slow, as the app queries many STAC2 pages and writes many H33 records.
You can find considerably more context in Toni’s Medium article about the project, including a discussion of the results.
The source code behind the map is available on GitHub. The app has obvious use cases in the remote sensing industry. But I think it could also be generalised to other frequently updated STAC-indexed geodata assets where time is of importance. Copying and modifying the app’s source code is dependent on written permission from Toni.
Footnotes
Date ranges are entered in the
dd/mm/yyyyformat.↩︎STAC stands for “SpatioTemporal Asset Catalog”, a modern, open specification for cataloging spatio-temporal data.↩︎
H3 is a hexagonal (mostly, except for 12 pentagons – think: football) discrete global grid system (DGGS) (Apache 2.0-licensed) created by Uber.↩︎