Sigma0 might be possible (Radiometric Calibration), but RVI and decompositions are technically impossible because they require the complex SLC information. Please ask your supervisor if there is a chance to get the Level 1.1 data of these products.
Also, RVI and decompositions based on dual-pol data are not comparable to the results gained from quad-pol because they are based on two polarizations only.
@ABraun
Ok, I will ask him because Im not sure if he still have quota to get level 1.1.
Please suggest or give me link on what can I do with this level 2.1 data. The main purpose are for studying the forest. or otherwise I could not make good use of this level 2.1 data. Thanks
This is a quite general description, can you please specify a bit more?
Your method should determine the selection of data, not the other way round.
@ABraun
How can I georeference the processing output of the quadpol. Is there any function to do it in Polsar pro or I just do it in envi software.
have you seen this tutorial? SAR Polarimetry and Analyses
When the quad-pol data is Level-1.1 you can do all processing in SNAP.
SNAP only has problems with Level-2.1 data which is already geocoded. There was a work-around reported in this topic which works on higher level products: ALOS-1 L1.5 Terrain Correction
@ABraun
Yes I did.
Ok thank you. I will read the topic. Basically here is how the outputs look like.
where have you generated these raster bands?
Generated from from PolSARpro v6.
I see. Many of these parameters could be derived directly in SNAP which would bring the advantage that subsequent terrain correction is no problem (as shown in the tutorial).
A tutorial on the exchange of data between PolSARpro and SNAP is planned, but not to be released within the next weeks.
Ok, im reading the forum posts about the “trick” still trying to understand what exactly it is meant. Also installed the ASF MapReady and trying to figure out how to use it to do the georeferencing. Hope you can share with me the tutorial when it is available soon. thanks
Level 1.5 or 2.1 can be geocoded in SNAP when you apply the mosaic operator on the image. But it does not include terrain correction.
The problem of gecoding quadpol data exported from PolSARpro is a different one, so let’s not mix this up.
Yes, understood this one is different case.
I tried the Radar --> Geometric --> SAR-Mosaic also tried the Update Geo Reference but error. Seems some required information (metadata) is missing.
Ya, its quad pol. Great!
Thank you @ABraun 
you saved my day. Regarding my TanDEM-X quiry the other day I will ask you about few details about the outputs.
There is slight difference with my SNAP Geocode parameter box. Addition to the selection is the “Save Auxiliary Files” Maybe I should try the DEM.
The geocoding was stuck and I closed PolSARpro. Then I wanted to repeat the process but I dont know how to set the data directory.
“Save Auxilliary Files” is not necessary here.
If you want to access the data again, you select the directory where your data was processed (not the sub-folders e.g. T3) under Environment > Single Dataset.
Hello, This is Enlan. I have the level 2.1 data like you and I want backscatter value from it. Could you please suggest me what to do as per your experience? I can’t do calibration. Thanks in advance.
Level-2.1 is an advanced product level where terrain correction has already been applied. Its calibration to Sigma0 is currently not supported in SNAP.
Have you tried to apply the calibration equation via the band maths?
10*log(10) * (DN^2) - 83
where DN is your raster image.source: https://www.eorc.jaxa.jp/ALOS-2/en/calval/calval_index.htm
The result should range between -40 and +10 dB