ISCE output - MiaplPy input

[Basin-Ir]

I am a beginner in using MiaplPy, but I managed to run it successfully using the sample data. The sample file and example provided with the package are as follows:

wget https://zenodo.org/record/6539952/files/PichinchaSenDT142.zip
unzip PichinchaSenDT142.zip
cd PichinchaSenDT142
miaplpyApp.py PichinchaSenDT142.txt --dir ./miaplpy

Now, I have tried to create new input data for MiaplPy using ISCE2 based on my own Sentinel-1 images. My goal is to later use the MiaplPy output in the SARvey package. According to the documentation, the input for MiaplPy should have the following structure:

ISCE_processed_data
├─ reference
│ ├─ IW*.xml
│ └─ ...
├─ merged
│ ├─ SLC
│ │ ├─ YYYYMMDD
│ │ │ ├─ YYYYMMDD.slc.full
│ │ │ └─ ...
│ │ ├─ YYYYMMDD
│ │ ├─ YYYYMMDD
│ ├─ geom_reference
│ │ ├─ hgt.rdr.full
│ │ ├─ lat.rdr.full
│ │ ├─ lon.rdr.full
│ │ ├─ los.rdr.full
│ │ └─ ...
└─ baselines
└─ YYYYMMDD_YYYYMMDD
└─ YYYYMMDD_YYYYMMDD.txt

And this input should correspond to the output of ISCE2. After running ISCE2 for about five hours, a large number of massive files were generated on my system, and the processing completed successfully. However, I cannot find the files that MiaplPy requires as input among the ISCE2 outputs!

I used the following command and the output is very large. I expected it to only output for the defined BBox which is a small range!

python /home/behzadsarhadi/tools/isce/applications/topsApp.py --steps /mnt/c/WSLData/SARvey_Rey/topsApp.xml

[mirzaees]

Hi @Basin-Ir
topsApp is used to generate a single pair and what we require for timeseries analysis with miaplpy is a stack of coregistered SLCs. what we meant by isce2 outputs is the topsStack outputs. if you use topsStack, you will find the required inputs for miaplpy

[Basin-Ir]

Dear @mirzaees Yes, thank you very much for your excellent explanation. However, my final question is: what is the minimum number of images required? I don’t mean in terms of calculation accuracy, but rather to ensure that the code runs successfully to the final step without any errors. For example, would 12 images be enough? 30? Or how many? Because I encountered an error when using only 3 images — not during the use of MiaplPy, but later on when using SARvey.

[mirzaees]

@Basin-Ir, I am not aware of the limitation of sarvey but miaplpy should be able to process 3+ images.
To see a meaningful result, I recommend setting the ministack size to 15 and using at least 2 ministacks to process, meaning you need 30 images.