-
Notifications
You must be signed in to change notification settings - Fork 299
Description
I processed a stack of interferograms two times with data from Sentinel-1A.
The first stack included ~201 images, including one from 2015. The second stack has the same dataset but I did not used the image from 2015. I processed the time series using Mintpy and for the first case, it creates a decent time series, but for the second case, it creates an empty time series with only data for the reference point (all the displacement is 0).
I went over the processing, and I noticed that the merged/interferograms folder for each case has different contents. Meaning that in the first case, ISCE2 created all (most) the interferograms in each folder (e.g., filt_fine (.unw, .unw.xml, .unw.vrt, .unw.conncomp.xml, .unw.conncomp.vrt, .unw.conncomp)) when running the step run_11... But in the second case, it created the folders with just a few files.
I did not change anything in the configuration but the date of the reference date. In the first case, I used the date of the 2015 image, in the second case I used the date of the first image in 2016.
I thought that doing this change was going to improve the final results as the interferograms created with the image from 2015 were the only ones not conneted to the time series.
I am attaching a couple of screenshots showing the different files created for each case:
Example files in merge/interferograms/20160421-20160503 (in the stack with the 2015 image)

Example files in merge/interferograms/20160421-20160503 (in the stack without the 2015 image)

Any explination is appreciated!