0

BACKGROUND:

I am downloading Sentinel-2 imagery using the Semi-Automatic Classification Plugin, QGIS version 3.22.12.

Some of the granules contain "gap" areas where all of the pixel values = 0, as shown in the black area in the screenshot below. Unfortunately, there are no adjacent granules with correct data that overlap this gap area.

The gap areas occur at both Levels 1C and 2A, and for bands 2, 3, 4, 8, and 12 (I have not downloaded any other bands).

These gaps extend into adjacent granules. enter image description here

However, there are some dates that do not have the gap. Below is the same granule, three days earlier: enter image description here

I've reviewed the same granule up to a year earlier, and the same pattern exists: some - but not all - dates have gaps.

The Sentinel User Guide regarding coverage makes no mention of these gaps.

QUESTION:

Why are these gap areas appearing, and how can I acquire correct pixel data for those areas?

2
  • these are the boundaries of the satellites acquisition orbit. An orbit is split into multiple tiles, and the same tile can be either covered fully or partial. For these missing areas, you can’t get valid pixels for the specific point in time.
    – Val
    Nov 7, 2022 at 19:10
  • Your comment seems legit, so feel free to convert it to an answer and I will be happy to give you credit.
    – Stu Smith
    Nov 9, 2022 at 1:01

1 Answer 1

3

This is due to orbit geometry and the way orbits are split into tiles. For any orbit, there will be tiles that are cut from the orbit's border and these tiles will have these nodata portions. For these areas, there's simply no data and therefore none can be "recovered". To get data for this area, you need to select a successive overlapping orbit.

Here's a nice plot showing an orbit (black) split into tiles (green and red) which showcases the tiles on the orbit's borders having only partial data coverage:

https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.sciencedirect.com%2Fscience%2Farticle%2Fpii%2FS0034425718301792&psig=AOvVaw1quB1PybZgatgxFfTv60dk&ust=1668069223622000&source=images&cd=vfe&ved=0CA0QjRxqFwoTCPj0_Z_YoPsCFQAAAAAdAAAAABAP

Image is from L. Yan, D.P. Roy, Z. Li, H.K. Zhang, H. Huang, Sentinel-2A multi-temporal misregistration characterization and an orbit-based sub-pixel registration methodology

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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