A seamless and cloudless Sentinel-2 image of France in July 2018

=> UPDATE: resolution improved to 20m !  I had to tinker a couple of days with Theia’s Sentinel-2 monthly syntheses, in order to produce the mosaic below. For that, I used:

• monthly syntheses of Sentinel-2 L3A products delivered by Theia in July 2018; this products are made from Level2A products, corrected from atmospheric effects, and provided with a good cloud mask, thanks to MAJA processor. Such L3A products will be delivered every month.
• a script developed by Simon Gascoin
• good advice from Simon and Michel Le Page
• and Gdal, (Thanks Gdal !)

This mosaic best resolution is 20m. It is already requiring 8 GB. We could have provided it at 10m resolution, but it would have required 32 GB and several dozens of hours of computation. See it full screen

The drought which affected the North East of France is already visible, with a brownish colour. Our level 3 product tries to reduce the artifacts usually related to this kind of products, using only physics, without any cosmetic homogenisation of seamlines, because there are no seamlines.. Still you will find some artifacts

• some tile borders are visible, because we did not generate the July L3A products for the same date. The central dates go from the July 7 to 27. Starting from August, all the products will be generated for the 15th of each month
• a few mountains were cloudy all month, at least for Sentinel-2 overpasses. They are indicated as cloudy in the masks delivered with the product.
• a few orbit borders may appear, but I did not noticed much of them (there is a discrete one in Britanny)d
• some artefacts over water and snow, because their reflectances change with time
• a few small clouds forgotten by MAJA

If you notice artefacts outside of this list, please tell me ! Producing the mosaic was a little more complicated than for Belgium, because France, this great country ;), spans over 3 UTM zones (31, 31, 32). I needed to merge the data within the same projection. Anyway, it still only requires a few lines of bash code, as you can see on my github site.