How to prepare Sentinel-1 images stack for PSI/SBAS in SNAP

Hello @m477h145 and others.

In my study i also found 0 Ps pixels.

“PS_SELECT: 0 ps selected after re-estimation of coherence”

As a result of 0 PS pixel, steps not completed. Did you find any solution about this problem?

Thank you.

Hi @volkan ,

you might want to check if there are any empty files among your radar data and interferograms. This is not checked within StaMPS, so no warning is raised if one or more of your interferograms are somehow messed up . However, you should get a warning when calling mt_prep_snap (something related to related to having 0 mean amplitude, see posts #461ff above). Make sure to remove all files referring to the specified date (both in the rslc and diff0 directories).

Apart from that, there’s the obvious hint to slightly increase your amplitude dispersion index.

I looked the lines after mt_prep_gamma_snap command and there is no zero amplitude pixel. So, can we say the problem is not about 0 mean aplitude?

Amplitude Dispersion Threshold: 0.4
Processing 1 patch(es) in range and 1 in azimuth
 
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20141015.rslc...
Mean amplitude = 58.0443
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20141108.rslc...
Mean amplitude = 56.7398
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20141214.rslc...
Mean amplitude = 55.3928
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150119.rslc...
Mean amplitude = 50.2859
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150224.rslc...
Mean amplitude = 54.4745
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150308.rslc...
Mean amplitude = 52.1636
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150413.rslc...
Mean amplitude = 61.8206
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150507.rslc...
Mean amplitude = 60.4407
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150624.rslc...
Mean amplitude = 67.7829
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150706.rslc...
Mean amplitude = 64.6618
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150823.rslc...
Mean amplitude = 66.5205
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150916.rslc...
Mean amplitude = 62.4759
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20151010.rslc...
Mean amplitude = 69.5845
Number of pixels with zero amplitude = 4
Number of pixels with amplitude different than zero = 20331
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20151103.rslc...
Mean amplitude = 61.2597
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20151209.rslc...
Mean amplitude = 52.6889
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160219.rslc...
Mean amplitude = 55.5107
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160314.rslc...
Mean amplitude = 57.6025
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160501.rslc...
Mean amplitude = 56.7956
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160525.rslc...
Mean amplitude = 56.5111
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160630.rslc...
Mean amplitude = 58.5274
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160805.rslc...
Mean amplitude = 62.3036
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160829.rslc...
Mean amplitude = 57.3653
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160910.rslc...
Mean amplitude = 58.875
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20161028.rslc...
Mean amplitude = 55.6083
Number of pixels with zero amplitude = 0
Number of pixels with amplitude different than zero = 20335
245
83
mt_extract_cands Andy Hooper, Jan 2007
 
Patch: PATCH_1
file name for zero amplitude PS: pscands.1.ij0
dispersion threshold = 0.4
width = 245
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20141015.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20141108.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20141214.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150119.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150224.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150308.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150413.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150507.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150624.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150706.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150823.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20150916.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20151010.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20151103.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20151209.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160219.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160314.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160501.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160525.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160630.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160805.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160829.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20160910.rslc...
opening /home/volkan/Desktop/PSI/INSAR_20151010//rslc/20161028.rslc...
number of amplitude files = 24
number of lines per file = 83
patch lines = 83
patch width = 245
opening /home/volkan/Desktop/PSI/INSAR_20151010/...
opening pscands.1.ij...
width = 245
opening /home/volkan/Desktop/PSI/INSAR_20151010//geo/20151010.lon...
opening /home/volkan/Desktop/PSI/INSAR_20151010//geo/20151010.lat...
opening pscands.1.ij...
width = 245
opening /home/volkan/Desktop/PSI/INSAR_20151010//geo/elevation_dem.rdc...
opening pscands.1.ij...
width = 245
number of interferograms = 23
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20141015.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20141108.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20141214.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150119.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150224.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150308.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150413.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150507.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150624.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150706.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150823.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20150916.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20151103.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20151209.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160219.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160314.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160501.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160525.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160630.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160805.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160829.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20160910.diff...
opening /home/volkan/Desktop/PSI/INSAR_20151010//diff0/20151010_20161028.diff...
1 of 23 interferograms processed
2 of 23 interferograms processed
3 of 23 interferograms processed
4 of 23 interferograms processed
5 of 23 interferograms processed
6 of 23 interferograms processed
7 of 23 interferograms processed
8 of 23 interferograms processed
9 of 23 interferograms processed
10 of 23 interferograms processed
11 of 23 interferograms processed
12 of 23 interferograms processed
13 of 23 interferograms processed
14 of 23 interferograms processed
15 of 23 interferograms processed
16 of 23 interferograms processed
17 of 23 interferograms processed
18 of 23 interferograms processed
19 of 23 interferograms processed
20 of 23 interferograms processed
21 of 23 interferograms processed
22 of 23 interferograms processed
23 of 23 interferograms processed

What is left is to check the interferograms, unfortunately there you have 2 options:

  1. visual inspection opening them into SNAP
  2. create a script to do png of your interferograms

Let us know.

I used snap6.0 to prepare stamps.My chain:read,split,apply orbit,backgeocoding,deburst,and interforgram formation,stamps out.But i got an unexpected result after interforgram.the intensity image were almost black.i did’n know where is wrong .wish reply@FeiLiu@ABraun@ katherine

the dark intensity image is fine. I asked the same question here and mengdahl gave an answer: TopoPhaseRemoval and stack

thankyou for hlep me.but now i met another issue is that backgeocoding six splited-orbited dim file,why only the master one is ok,other slave i ,q,intensity image are all empty.wish help @ABraun@lveci. Much appreciate

1 Like

Yeah. This problem happening everywhere.

how did you solve it?can share it.much thaks.

I re-applied SNAP steps (split/orbit/back geocoding/deburst/subset/interferogram formation/export) with new images. And still i find 0 PS pixel (“PS_SELECT: 0 ps selected after re-estimation of coherence”) when i run stamps ( version 3.3b1 under Ubuntu)

I shared SNAP export screenshots and the original files with link below.

Stamps export files

Is there any problem until this stage? Thank you.

your files look extremely small (150 kb). How large is your study area?

Nearly 0.9x0.8 km. I also enlarged study area but nothing has changed.

Area.kml (1.4 KB)

your area (besides that it is very small) rarely has any suitable permanent targets, given that the original footprint of Sentinel-1 SLC products is 3x22 meters (range and azimuth). The longer your time-series, the smaller the chance to capture PS which are present throughout all images. You are observing almost 2 years.

To get more PS, you can:

  • increase the amplitude dispersion to 0.42 for mt_prep
  • have a larger area
  • reduce the time to one year for a start

Thank you for your suggestions Andreas. I will try as your said and write here the results. Best wishes.

Hi Andreas,
I reduced the time to one year, enlarged area and increased the amplitude dispersion to 0.42. It worked very well and I found 5-6 PS pixels in the study area. Thank you (:

In addition, to find more PS pixels in the study area, do you suggest any other specific hint?

how many images do you use within this year?

10 images between 2018.03.04-2018.12.05.

are these all available? Sometimes it is helpful to use denser image series to identify more PS.

I downloaded and used 10 of images. So, yes there are available. I will try the denser image series.

Hello, thank you very much for providing the step by step summary. however the link you provided seems broken. could you please provide the new link?