GRD product from S1: wrong georeference

Hi all, I’m new to this community. I have a problem with the SAR scene from S1: it rotated 90 degrees, not the North oriented like the map

It should be like this

I do not know if this rotation will affect other processing step or not. Please help me !

Kind regards,
Alan

Hi Alan,

did you already perform a terrain correction on the image?
If not, please be aware of the specifics of SAR acquisition geometry; Sentinel-1 is a so-called right-looking satellite acquires images in ascending and descending orbit. Depending on the current orbit from the scene, the image appears “flipped”.

You can find more information in this related post:

For more information on the technical properties, please have a look at this detailed description of the technical properties of Sentinel-1:
https://directory.eoportal.org/web/eoportal/satellite-missions/c-missions/copernicus-sentinel-1
For more “starter information” you can also have a look here: https://saredu.dlr.de/
And here, an online course is starting next month on the topic of an “Introduction to Radar Remote Sensing”: https://eo-college.org/courses/echoes-in-space/
It may seem a little much when you’re just starting in the radar world, but just keep looking up things you don’t understand. We all started from scratch :wink:

Welcome to the community!

Regards,
Denise

1 Like

Please go through the tutorials.

2 Likes

Thanks Denise,

Before seeing your answer, I applied the “Geolocation Grid” correction and the img seemed to flip back to how it should be. However, after that operation, I could not apply Radiometric Flattening, error says that “Source product should not be map projected”.

I have not found anywhere to use the Flip. I think it is in NEST toolbox rather than SNAP/S1Tbx. I will try the Range Doppler Terrain Correction to see if the img be flipped back.

I have put my name in the EO-College course already.

What I intend to do with S1 SAR: The cloudy weather blocks me from having good Sen2 img. I used those images to visually detect the forest cutover on the ground. Now I hope with S1 SAR I can somehow see through the “cloud” to see those cutovers.

Our cutover site sometimes can be small 100m x 100m but I still can see it on Sentinel 2. I’m not sure if that dimension is too small for Sentinel 1 ?

Thank you very much for the support !

Regards,
Alan

Please go through the tutorials for S1TBX. Use Range-Doppler terrain correction to orthorectify the image into map coordinates.

Dear Alan,

For what concern your primary request on the fact the images are not properly oriented.
I have given an answer to that on that post:

Please have a look to it.

kind regards,
Nuno Miranda