Alos/Palsar L 1.1 Mosaic

I’m using two consecutive Alos/Palsar L1.1 dual-pol images but I have the following problem.
I apply for each one in SNAP: Calibrate -> Terrain Correction -> Subset , after I run SAR Mosaic.
Then I want to run Polarimetric Matrix generation but I get this:


Is it possible to do polarimetric processing with a SAR mosaic in SNAP or I must do it for each image separately???

Thank you.

  1. have you selected “complex output” during calibration? This is required for polarimetric analyses
  2. have you changed the data format during the processing steps? This causes loss of metadata.
  3. is it an option to process both image until the polarimetric matrix generation and create a subset in a final step?

Yes, during calibration I selected “Save a complex output” and did the same in Range Doppler Terrain correction selecting “Output Complex Data”.
I’ve saved the results of all processes as Beam-Dimap and tried with different workflows as you said in 3, but I always get the same problem when I want to run Polarimetric Matrix generation.

What if you apply matrix generation after calibration and terrain correct as a last step?

I am not sure but I have always thought that the polarimetric matrix should be generated from the SLC directly, without terrain correction or calibration or so on…

But maybe I am wrong. Can anybody tell ?

I always thought the same as you, @mdelgado.
In the PolSARpro course with Eric Pottier in January we also terrain corrected the complex data early for a better visual interpretation of the later polarimetric measures.
If he thinks that is okay, am willing to believe it :slight_smile:

But only for a better interpretation? or you continue with this data in the processing?

Yes, we derived all parameters after terrain correction.

1 Like

Good to know! Thanks for the hint!