Pixel Value of Sentinel2 subset : which method

Hi everyone, this time I am submitting here not a technical problem, but a question about computation.
For a specific polygon over a geographic area, I obtain an intersection with two different tiles, so I obtain two different subsets for the same Time of Interest. This means that for every common point of the subsets I will have two different pixel values, whatever the band. Which is the best way to go on? how to choose which value to consider? the average of the two? or whatelse? maybe @ABraun could help me :wink:

Thanks to whoever will reply :wink:


The Mosaic tools in SNAP let you decide if you want to compute an average or take the values of a specific raster in the overlapping area.

Yes, but from a scientific point of view, besides the average which could be a reliable approach? :slight_smile: