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

SUN-Ravenclaw · January 16, 2020, 2:48pm

Hi Abraun, I need to put all the exported dem, diff0, geo, rslc files in the INSAR_20190208 folder that has been created. Then I can execute mt_prep_snap. Is that right? Because there is always an error that the file or directory cannot be found.How do I execute this command?
Note:My directory is listed below:
INSAR_20190208
├── S1A_IW_SLC__1SDV_M20190208_S20190103_StampsExport
│ ├── dem
│ ├── diff0
│ ├── geo
│ └── rslc
├── S1A_IW_SLC__1SDV_M20190208_S20190115_StampsExport
│ ├── dem
│ ├── diff0
│ ├── geo
│ └── rslc
├── S1A_IW_SLC__1SDV_M20190208_S20190127_StampsExport
│ ├── dem
│ ├── diff0
│ ├── geo
│ └── rslc
└── S1A_IW_SLC__1SDV_M20190208_S20190220_StampsExport
├── dem
├── diff0
├── geo
└── rslc

ABraun · January 16, 2020, 5:43pm

actually it should only be

INSAR_20190208
├── dem
├── diff0
├── geo
└── rsl

SUN-Ravenclaw · January 17, 2020, 9:54am

Sir, do you mean to process each exported data separately? Or merge all the exported data in the dem, geo, diff0, rslc folders, and then process them together?

ABraun · January 17, 2020, 9:56am

the merging is actually done during the export.

In the StaMPS export you enter

stack of all backscatter images
stack of all interferograms (incl. DEM and lat/lon)

this produces one output folder INSAR_ with the four sub-directories.

SUN-Ravenclaw · January 17, 2020, 1:05pm

Yes, sir, you are right. But you misunderstood my question here, let me describe it to you in detail:
First I have successfully exported the data, as shown in the following structure directory (a total of 5 periods of image data, and then select 1 of them as the master image, and the rest as slave images).
First-level directory names are defined.
├── S1A_IW_SLC__1SDV_M20190208_S20190103_StampsExport
│ ├── dem
│ ├── diff0
│ ├── geo
│ └── rslc
├── S1A_IW_SLC__1SDV_M20190208_S20190115_StampsExport
│ ├── dem
│ ├── diff0
│ ├── geo
│ └── rslc
├── S1A_IW_SLC__1SDV_M20190208_S20190127_StampsExport
│ ├── dem
│ ├── diff0
│ ├── geo
│ └── rslc
└── S1A_IW_SLC__1SDV_M20190208_S20190220_StampsExport
├── dem
├── diff0
├── geo
└── rslc
Then your answer above suggested that I change the first-level directory name (S1A_IW_SLC__1SDV_M20190208_S _StampsExport) to INSAR_20190208.（as shown below）
My problem is that I should change the first-level directory names of the four folders to INSAR_20190208, and then separately run the mt_prep_snap command on each modified folder. Is this correct?
INSAR_20190208
├── dem
├── diff0
├── geo
└── rsl
INSAR_20190208
├── dem
├── diff0
├── geo
└── rsl
INSAR_20190208
├── dem
├── diff0
├── geo
└── rsl
INSAR_20190208
├── dem
├── diff0
├── geo
└── rsl

ABraun · January 17, 2020, 1:14pm

sorry, I don’t get your point. The master image is already defined during the coregistration. You should only have one first level directory (INSAR_20190208) after the StaMPS export and it contains files of all slaves.

Abdel · January 17, 2020, 1:34pm

Hello,
Let me step in this discussion.

@SUN-Ravenclaw If i well understand you have separated stacks of images paires and interfergrams.
You should export all your images+interferograms to the same directory : INSAR_20190208 that you create manually

On “Stamps Export” enter : Your_Working_Directory/INSAR_20190208
in the field “Target Folder” as shown bellow :

Use the same directory each time you will export a image+interferogram pair.
It would warn you that the directory exists already and it will be overwritten. Just ignore that and make YES. It will not overwrite the files but increment theme in each export

ABraun · January 17, 2020, 1:46pm

thank you for commenting.
Another solution is to create a stack of all image pairs (AB, AC, AD), both for the backscatter and the interferogram, and use these two stacks in the StaMPS export.

Abdel · January 17, 2020, 2:09pm

It’s been a while i have not followed this topic.
Is one Stack for multiple slaves working fine now ?

ABraun · January 17, 2020, 2:20pm

now I’m not sure anymore either
I have only been using this python package for its automation: Snap2stamps package: a free tool to automate the SNAP-StaMPS Workflow
I’m not entirely sure how it handles the different pairs.

ABraun · January 17, 2020, 6:20pm

I checked the snap2stamps scripts again and this is how it is solved there, too. So you were right: The export has to be repeated for each pair, but the output folder is the same.

Abdel · January 17, 2020, 6:30pm

In a time i did it manually for a huge amount of images.
Thanks to snap2stamps authors this come to be more easier.
@SUN-Ravenclaw you can also use this tool. it can help you a lot Snap2stamps package: a free tool to automate the SNAP-StaMPS Workflow

SUN-Ravenclaw · January 20, 2020, 6:38am

Hi Abdel, thank you for the answer above, I have completed steps 1 to 8 of StaMPS in matlab. However, I encountered some warning messages in step 6 (shown below).
Warning: A’s rank is not sufficient to reach the computer precision range.

In lscov (line 200)
In uw_sb_unwrap_space_time (line 304)
In uw_3d (line 161)
In ps_unwrap (line 235)
In stamps (line 503)
Do you know why? Thanks again.

survivere13 · January 28, 2020, 3:36pm

Dear @ABraun ,
I tried what you recommended to do but I still cant get anu results! would you please take a look at these:

thank you

ABraun · January 28, 2020, 4:32pm

I don’t understand why you apply Create Stack when you have already applied Back Geocoding.

survivere13 · January 28, 2020, 6:16pm

I found this tutorial in forum, with these steps, I may be misunderstood:
" 1) For S1, apply TOPS split and then apply precise orbit
2) Add all products to the Backgeocoding
3) Create Stack of Backgeocoded images (master_Stack)
Create a stack of Interferograms (Follow the step one of Prepare for STACK AVERAGING) on the resulting stack. { STEP 1 : master (SAME MASTER)+ (Different Slaves)first slave -> TOPS Corregistration, if necessary Deburst (and possibly subset), Interferogram, topographic phase removal, } master_Stack_ifg_dinsar .
Deburst both the Backgeocoded stack (master_Stack_Deb), and the Interferogram stack (master_Stack_Deb_ifg¬_dinsar .).
5) Add an elevation band to the Interferogram stack (by right-click > add elevation raster.).
6) Add the Debursted stack and Debursted Interferogram stack to the stamps export "

ABraun · January 29, 2020, 6:46am

yes, sorry. You are right. It was a misunderstanding from my side yesterday.
Do you use the same master for all coregistered stacks?

By the way, to add elevation and lat/lon rasters can now be done in the interferogram generation step already. Unfortunately, there are many outdated instructions on StaMPS preparation in this forum. This one is the latest I know: Documentation on SNAP-StaMPS workflow

We collected them once in this topic: About the StaMPS category

Please make sure you are referring to the latest version.

survivere13 · January 29, 2020, 10:18am

Yes, I use a 201912 image as master for each of coregistered stacks, I still don’t know whats wrong
thank you so much I’m gonna use this tutorial and check if I did sth by mistake

gabrielaquintana77 · January 29, 2020, 5:28pm

Hi all, the following figure is my map in GMT with the results of my DInSAR PSI processing…
using the ps_mean_v file!!! @mdelgado

gabrielaquintana77 · January 29, 2020, 5:52pm

Hi again, I want to check with details the contents of folders PATCH 1… 2… of others people here! can somebody to list this please??? I am checking errors presented during the processing…