I’ve come across a weird issue and I must solve this problem. I wanted to collocate a Sentinel-1 Level-2 (OCN) image with SLSTR image from the same day, in order to compare SST and the owiWindSpeed data. However, not only is the original S1 data horizontally flipped (but it agrees with the lon band), but after collocation or reprojection to a Projected coordinate system the output is not reprojected properly. I compared the outputs with the TM_WORLD_BORDERS-0.3 shapefile that works great for me as a reference. in case of doubts like this time. I attach print screens below (top - S1, bottom - SLSTR).
How can I obtain properly collocated images? I would be grateful for an explanation, because I really need this data.
Regarding the display of S-1 image, it could be that your image was acquired with ascending pass and right antenna pointing (check the abstracted metadata of the product). In that case, the first line of the image corresponds to the bottom of the scene and the last line corresponds to the top of the scene. In SNAP, the image is always displayed with the first line on top and last line at the bottom. That’s why the image looks up-side-down
It might be the case. It was very misleading. I did not expect anything like that, especially in L2 products.
It’s me again. It looks like the upside-down position is not the only problem. After collocating images the OCN products are shifted. There are GCP points in the file and separate LAT, LON bands, but during reprojection or collocation SNAP seems not to use them. How should open the files / modify / process them when I want to collocate them with the e.g. SLSTR data? I know that I may visualize them using WorldWind Analysis (which apparently visualizes the iamges properly), but I really need to collocate OWI wind speed data, not only visualize it.
I want to add that it is the first time I have such problems and all the other Sentinel products collocate without problems, so it is probably not an issue of using the Collocate function in a wrong way.
I suggest you to have a look to this post: Reading and mapping OCN products - Windmap - #16 by carloc where the issue has been (partially…) solved.