topsApp (ISCE) to MintPy

[Kleanthis Karamvasis]

Hello MintPy community,

Recently, I processed ARIA products using Mintpy and I would like to do the same using topsApp (ISCE) generated interferograms. I have already computed ~2000 Sentinel-1 interferograms using topsApp (ISCE) of a given region and successfully extracted ground deformation using NSBAS from GiaNT. All interferograms have processed seperately using topsApp.py, are geocoded and have the same dimensions. I attached an interferogram (int_20160123_20160111.zip) with associated files.

I noticed that Mintpy already supports topsStack, stripmapStack and stripmapApp (from prep_isce.py) and I had a brief look at the functions. Right now i am looking at the FernandinaSenDT128 example and i am trying to replicate the required directories (e.g. using metadata from isce.log file) in order to use the same script (prep_isce.py).

I would like to ask some directions/suggestions on how to on how to prepare/import the topsApp interferograms to Mintpy.

Am I missing an already implemented solution?

is it a good practice to replicate the missing directories/files that Mintpy expects?

Is it better to write a new function in the prep_isce.py?

Thanks in advance,

Kleanthis

[Yunjun Zhang]

Hi Kleanthis,

There is no implemented solution for the stack of geocoded interferograms from topsApp. But it should be straightforward to implement it.

Regarding the directories, I would recommend the one from topsStack: https://mintpy.readthedocs.io/en/latest/dir_structure/#isce_topsstack. But this is not required by MintPy. You can specify the path pattern of each type of file in the template file, as long as the wildcard pattern could locate the unique file for each interferogram.

prep_isce.py generates, for every *.unw data file, a simple *.rsc file containing the required metadata for MintPy, which is described here: https://mintpy.readthedocs.io/en/latest/api/attributes/. The core functionality is to extract metadata from the IW*.xml file (for TOPS mode) using isce. It also extract some other metadata, i.e. perpendicular baseline, from the “baselines” directory; but if you could extract them from existing files, there is no need to re-create it.

I would recommend adding a new function, prepare_tops_ifgram() for example, inside prep_isce.py as you suggested, to prepare all required metadata and write them into an *.rsc file. 

Yunjun

--
You received this message because you are subscribed to the Google Groups "MintPy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mintpy+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mintpy/443b70d8-f32b-4e7f-90ad-b143888ea280%40googlegroups.com.
<int_20160123_20160111.zip>

[范成彦]

My solution:  Prepare ISCE topsApp interferograms for processing with MintPy_沉研的博客-CSDN博客

[Zhang Yunjun]

Thank you for sharing your solution! I took a quick look and it seems reasonable to me.

I would suggest turning your notes into a jupyter notebook, similar to what we have in the MintPy-tutorial repo (https://github.com/insarlab/MintPy-tutorial/blob/main/applications/water_mask.ipynb). Then it would be even more convenient for you and other users to check it out and try it themselves.

Cheers and best regards,

Yunjun

[范成彦]

Thank you for your advice. I put my solution on GitHub.
insar-process/2tops2mintpy.ipynb at master · Fanchengyan/insar-process (github.com)

[Zhang Yunjun]

This is great!

[Liuru Hu]

Hello MintPy community,

Recently, I processed Sentinel-1 using ISCE and Mintpy. I have finished the ISCE process, but I have a problem when the data from ISCE to Mintpy. I attached the problem.

Could you give me some solution to solve this problem? Looking forward to hearing from you!

Thanks in advance!

Best

Liuru

[Yuan-Kai Liu]

Do you have the .rsc files generated? It looks like the metadata is missing so while loading the datasets, MintPy cannot find the STARTING_RANGE and RANGE_PIXEL_SIZE.

Have you run pre_isce.py? That should generate the corresponding .rsc files with those metadata if I remember correctly. 

@fanch (https://groups.google.com/g/mintpy/c/WRxZ5aUuQFY/m/WenEL7EEAwAJ) has made a post of his solution in a jupyter notebook. Maybe check the prep_isce.py in that notebook would be helpful for you.

[Eric Fielding]

I had a problem like this recently, and I ran the `load_data.py` program separately to see more about the problem. I found that the program that converts the ISCE2 files to `.rsc` files was failing, most likely due to having a different version of ISCE2 in my `mintpy` environment than the version of ISCE2 that I used for the InSAR processing.