Question about using multi-frame HyP3 InSAR-GAMMA products for MintPy time-series

[m-AA1923]

Hi
I processed two adjacent HyP3 InSAR_GAMMA frames (Frame 460 and Frame 461). After subsetting both frames to my area of interest, I tried to run a MintPy SBAS time-series workflow. However, the time-series and velocity outputs only cover the reference frame, not the full area where the two frames overlap. As a result, parts of my study region remain outside the processed MintPy products.

My question is:

Is there an officially recommended workflow for combining multiple HyP3 InSAR_GAMMA frames so that MintPy can generate a complete time-series over the entire AOI?
(or alternatively)
Does MintPy support multi-frame HyP3 inputs directly, or must they be pre-processed in a specific way before running load_data?

Any advice, documentation, or recommended steps would be extremely helpful for my research.

Thank you very much for your time.

[Arctic-Ambrun]

Very good question. The reason for this is the absence of a certain period of time, generally due to sudden orbital changes, which causes mintpy to judge the mask file as missing. This is a problem with the input data. I suggest you check your interferogram stack and delete results like 93 and 94 (if this accounts for a small proportion of your interferogram processing data). I initially encountered this problem, but I didn't find a good solution because the fusion would cause obvious linear inconsistencies due to the reference point problem. Later, I directly used ISCE's stack to process the data locally (this is the best method I have confirmed). If you have GAMMA yourself, you can try processing it locally yourself.

[Arctic-Ambrun]

Another approach is to use ASF's burst products and then process the fusion yourself, which can effectively reduce the required space. It is best to use GAMMA+Mintpy for processing, because the ISCE stack encounters some kind of strange ESD issue...

[m-AA1923]

@Arctic-Ambrun Hello, thank you very much.
I was able to resolve this issue by using burst processing with GAMMA.

I would like to ask you another question, and I would really appreciate it if you could share your opinion.

Currently, I am interested in studying the overall deformation of urban areas. My aim is to analyze the general deformation pattern of an entire city, rather than focusing on individual buildings or very localized targets.

In your experience, would you recommend a specific processing workflow or methodology for this type of city-scale deformation analysis?

Any suggestions or guidance would be greatly appreciated. Thank you very much for your

[Arctic-Ambrun]

@maa192399-ops Hello, urban InSAR deformation monitoring is not my research interest, so I am sorry that I cannot give you any useful advice. I am not very familiar with the existing problems and methods of urban InSAR. I may prefer to use high-resolution X-band data for research.

[m-AA1923]

@Arctic-Ambrun Hello again,

Thank you for your previous reply. I would like to add that I am relatively new to using InSAR and I am currently trying to learn and work with ISCE2.

However, I am experiencing some difficulties in understanding the overall ISCE2 workflow, especially the processing steps and how they are connected (from SLCs to interferograms and time-series analysis).

I was wondering if you could recommend any tutorials, documentation, example workflows, or community resources that clearly explain the ISCE2 processing workflow for beginners.

Any suggestions would be greatly appreciated. Thank you very much for your time.

[Arctic-Ambrun]

@maa192399-ops Hello, there are many detailed tutorials and examples on GitHub; you can look for them yourself.

[ditafaith]

Hi @maa192399-ops there are many videos from Earthscope, you can explore them.