GAMMA Questions

This is a multi-part message in MIME format.
--------------LNwjYBMwG5DHad5r4Ynizmrs
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Hello,

When reading S1 GRD images in the gamma software they are converted to MLI
images by the program par_S1_GRD.
Reading a stack of S1 GRD images can be performed using the script
S1_GRD_preproc. The set of MLI images must be coregistered first if they are
to use the same geocoding lookup table. Coregistering an MLI image pair can be
done using MLI_coreg.py, or an entire stack of MLI images using MLI_resamp_lt_all.

If you have a very large area then it should be possible to concatenate MLI
images produced by generating MLIs from sucessive GRD frames, though this must
be tested.?? The concatenated MLI images could then be concatenated.

Another alternative is to mosaic multiple geocoded frames using the mosaic command.

Best regards,
Charles

On 09.07.2025 17:30, Miguel Gonz??lez wrote:

GAMMA Questions - Sentinel-1: How can I identify unique GRD scenes to re-use
pivot table during RTC? <https://dep1doc.gfz.de/boards/32/topics/666>

/Miguel Gonz??lez /

Hi guys!

I'm struggling with terrain-correction of several GRD images of the same
area (obtained from same acquision geometry, ie. same relative orbit number
and footprint), or what is the same, a kind of time series images of same area.

Since the geometry is the same, I would like to re-use the look up table
(created with gcmap2) to save up time. For this purpose, I need to find
unique scenes then, what it's being quite hard since GRD images do not keep
bursts ID information. In contrast, only subswaths and footprints information
is available in manifest.safe file.

You cannot reuse the lookup table from different acqusitions. The nature of the
Sentinel-1 data is that different GRD images

My doubt then is:

_ How can I identify unique scenes ? May I do it just by relative orbit
number and footprint or shall I require other extra info?
_ Once identified this unique do I need to co-register all the images
belonging to the same unique scene ? If it is, is there an offset threshold
(in meters) beyond no co-registration is possible? I need to identify unique
scenes???I mean, scenes from the same relative orbit that cover the same
geographic area.

Thank you very much in advance!

---

You have received this notification because you have either subscribed to it,
or are involved in it.
To change your notification preferences, please click here:
http://dep1doc.gfz-potsdam.de <http://dep1doc.gfz-potsdam.de>

Hi Charles!

Thanks a lot for your quick response (and sorry for my late reply).

Ok, it's clear for me now that co-registration should be made between images of same area and different time. However, I do not understand why you say that is not possible to re-use the look-up table for images of same geometry acquisition but different times. As long as I now, this is a common practice, though--as you said--preliminary coregistration should be performed...

The issue for me is which parameters to take into account for identifying images with same acquisition geometry (since GRD images do not contain burst ID info) so that I can build that stack of images to co-register. For the moment I am considering relative orbit number, orbit direction, and approximated footprints. I am struggling with the latter because image footprints for different time steps do not usually fully match--there is always an offset--, and maybe is there an offset threshold for doing co-registration?

Best regards,
Miguel

This is a multi-part message in MIME format.
--------------h8TNR0D9VJoBtbHyxN9VSxVi
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Hello,

there are small shifts of a few lines from one acquisition to the other.

Best regards,
Charles

On 21.07.2025 09:30, Miguel Gonz??lez wrote:

GAMMA Questions - Sentinel-1: RE: How can I identify unique GRD scenes to
re-use pivot table during RTC?
<https://dep1doc.gfz.de/boards/32/topics/666?r=668#message-668>

/Miguel Gonz??lez /

Hi Charles!

Thanks a lot for your quick response (and sorry for my late reply).

Ok, it's clear for me now that co-registration should be made between images
of same area and different time. However, I do not understand why you say that
is not possible to re-use the look-up table for images of same geometry
acquisition but different times. As long as I now, this is a common practice,
though--as you said--preliminary coregistration should be performed...

The issue for me is which parameters to take into account for identifying
images with same acquisition geometry (since GRD images do not contain burst
ID info) so that I can build that stack of images to co-register. For the
moment I am considering relative orbit number, orbit direction, and
approximated footprints. I am struggling with the latter because image
footprints for different time steps do not usually fully match--there is
always an offset--, and maybe is there an offset threshold for doing
co-registration?

Best regards,
Miguel

---

You have received this notification because you have either subscribed to it,
or are involved in it.
To change your notification preferences, please click here:
http://dep1doc.gfz-potsdam.de <http://dep1doc.gfz-potsdam.de>