1

I'm new to GIS, so please be patient if my question is a bit naive. I would like to regularly monitor an area in the Czech Republic in multispectral image depth, with update at least once in a few days. I am willing to pay for it.

I checked the webpages intelligence-airbusds.com and sentinel-hub.com so far; both seem to be promising, but I was not able to get down to details and high-resolution samples.

Could you give me some hints where to start, how to see examples and evaluate the quality, before subscribing and paying?

More generally, which services and platforms would you recommend?

3 Answers 3

3

Welcome Petře.

Your best single source for what you describe is probably Sentinel 2 L2A images, which you can every 5 days and offer visible + one IR band at 10m resolution, and multiple other bands at 20m and 60m.

You can visualize data for free with signup at Copernicus. The data comes as large files which you can unpack or visualize directly with QGIS, etc., but which take a long time to download and will be far larger than your area of interest.

If you are willing to pay, one convenient service is indeed sentinel-hub. You can sample and see rendered on-screen for free here, and as a paid service you can download conveniently or import into QGIS and other places.

As an example, this link will lead you to choose what is available around the locality of Okrouhlice, chosen semi-at-random.

You can also look into other datasources there like Landsat and Modis, which each have their uses. I'm highlighting Sentinel-2 due to its frequency and high resolution on the 4 usually most crucial bands as a starting point.

Editing to add: Seems I'm behind the times. A (non-gis.stackexchange) colleague has told me that sub-metre imaging with reasonable pricing is available at LandInfo, possibly others. Uses sources such as Worldview, Ikonos, Pleiades satellites, for a minimum ROI purchase size of 25 sq km. Still won't be cheap, especially if you want frequent revisits/updates, but a lot better than when I looked and failed to find for my ROI a few years ago!

5
  • Thanks a lot! However, I somehow thought I could get 30-50 cm resolution; that is too far off to hope for? Apr 6, 2020 at 15:46
  • 1
    Petře, 10m is as good as it gets for regular satellite flyovers (at least what anyone shares). You can get sub-meter resolution for orthorectified airplane photography, but this is done at far wider time intervals. For instance, in Ontario, Canada where I am based, the government coordinates ~60cm imagery on a rotating cycle every 5 years.
    – Houska
    Apr 6, 2020 at 15:53
  • Thanks a lot! This was very helpful. Apr 6, 2020 at 15:54
  • 1
    If you are willing to pay it might be worth looking into Planet labs. As far I can recall they offer about 3 m RGB+NIR coverage for the entire earth daily.
    – JonasV
    Apr 7, 2020 at 8:42
  • 1
    @JonasV, great add. Can you make it a separate answer (maybe add a link, though it's easily Google-able).
    – Houska
    Apr 7, 2020 at 10:20
3

If you are willing to pay it might be worth looking into Planet labs. They offer daily global coverage with a resolution between 5, 3 and 0.75 m. However you will at most get RGB+NIR and sometimes Red Edge.

2

I will add another possibility how to easily visualize some data from different satellite missions. Firstly, it will be Google Earth Engine. It is quite easy if you know a little bit JavaScript. If you would like to download data for area of the Czech Republic, there is a great possibility to use Collaborative Ground Segment. Probably every member of ESA will have such a Segment which is not limited by a number of downloaded files. Finally, there is a website called DIAS (Data and Information Access Services), which provides five cloud-based platforms providing accessing data and probably some advanced analysis. However, I've never used it and it's quite a chaotic solution for me (one platform would be enough)

I hope that it will extend info from JonasV and Houska.

Not the answer you're looking for? Browse other questions tagged or ask your own question.