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

Dear @thho tremendous thanks to your efforts by providing me with the new cod, I succeeded to create CSV with a plenty of points, I don’t include the cod within this post, because I think you only have the privilege to post it, I know that it is possible to subset the area according to the R script, but still why still isn’t possible to visualize a huge number of sp?

Now, when I run the ui.R and then selecting my folder the machine goes so heavy and the R. is closed.

It is already included in the latest version on the manual page.

technically it is, but it depends on the hardware of your machine…rendering a huge amount of points may force your system to crash or could take long time

so, use the subset script, also provided in the manual page and do a spatial subset of the csv table outside of matlab in R. All steps are explained in the manual page…I thnik at the bottom, after the export description under troubleshooting or so

Cheers and I am looking forward to seeing some nice screenshots :wink:

2 Likes

Hi @thho

I’m now trying up to use R scripts to create small subset area, I did get this error, after applying the highlighted script

Any suggestions,

Your lon lat objects are empty, please go to the Environment (upper right) go to the pnts object and click on the table icon to the right (same row). Make a screen shot of the table (will be opened in the upper left window) and post it again.

The other errors in your console are due to copy pasting mistakes I guess, the last line however looks fine to me. I suggest open a new script (upper left corner green plus sign) copy paste the code from the StaMPS Visualizer Manual there and run it line by line using Ctrl+Enter for the line your courser is in. stop, when the first error occurs, this way it is easier to check your progress.

Please try to move your post to the stamps Visualizer thread, it is still part of this workflow and I think your error is related because of something you have done there…

Cheers

I dont find the file mt_prep_gamma_snap and ps_load_initial_gamma.m. in the bin folder of stamps after installation. Kindly help me out… I am deadly struck up at various processes…

Hey Jennifer,

you no longer need the mt_prep_gamma_snap script that has circulated around this forum.
if you clone the github for the stamps 4.1b release a new script called mt_prep_snap is in there.
run this script in your INSAR_masterdate folder with the following command:

mt_prep_snap MASTERDATE FULL_location_insar_masterdate_folder/ DA
DA is normaly between 0.4 and 0.42

check the StamPS manual and also this gitlab for extra help: https://gitlab.com/Rexthor/gis-blog/blob/master/Documentation/StaMPS-2_Workflow.md

Greetings,
Gijs

4 Likes

thank you so much for clarifying @Gijs

no problem @Jennifer !
the forum was helpful for me to get starting, glad i can now help someone else!

note that before you can run mt_prep_snap you have to add all nessecary paths (matlab, stamps, snaphu etc) to your PATH variable in linux, you can do this by editing the config.bash file to your paths and then run in a terminal:

source exact_location_of_config.bash

Template for this config.bash is in the gitlab i send you.

Another note: if processing reqs become to high you can devide your area in patches using the 4 numbers that come ofter the DA value. For example 2 3 50 200 where 2 and 3 make 6 patches and 50 and 200 are meters overlap to stitch the patches together in step 6.

Goodluck processing

3 Likes

As @Gijs mentioned the script will as below,

Read the manual of Stamps 4.1b to get the meaning of each parameter,

This is a linux script run outside matlab in the shell

1 Like

Thank you @Gijs

I dint start yet with the step by step processing yet as i was struck up in the installation and configuration itself… However your suggestions are much helpful which i am about to carry out in the near future. Moreover the area of my process is also a larger one… I think i would get to you sooner to know more about the patches…

Looking forward to carry out the PS processing successfully… And am sure it cant be done without the help of you wonderful people…

thank you @falahfakhri… Will take note of it…

@thho @mdelgado

Hi Thorsten and Jose, I found this Toolbox for Reducing Atmospheric InSAR Noise I think it is important to add up, but I didn’t quite understand from the manual how to add up to the other softwares in Linux, would you please to take a look, and in which installation step should be added up, aftermath install all for example the last one or it doesn’t matter?

Thanks a lot.

I found other easy way to view time series of any point, using Orange (could be used in case of data visualization out of the QGIS, or Arcgis) after exporting the *csv file from stamps. I shared a very simple example below,

TRAIN was made to work with StaMPS as an optional package see StaMPS manual 2.3.2 the configuration is explained in the TRAIN manual pages 6 pp. The package can be added to the StaMPS installation any time by running the APS_CONFIG.* file in your shell, as it is described in the TRAIN manual. Google’s MAINSAR group is the place where I found the most helpful information about the package handling, I would recommend to read some posts there for a start.

5 Likes

In the case where no reference point is provided, the manual indicates “If not set, the reference value is the mean value for the whole area.”
I would like to ask the “mean value of whole area” refers to each point to compare with its own mean value(PS point own mean value), or the total average of all PS points together(whole area to one value)?

Thanks for your help
JAMIE

@JAMIE if you refer to the displacement velocity, the reference mean value is calculated by averaging all values in your dataset. I think that is what you said with

Hello,
I have this problem,too.
What should I do?

could you refer to the problem you are asking about?

I want to detrmine horizental displacements in snap, but when I want to create stack, it told me " [Nodeld: 2-create stack] S1 TOPS SLC products should use TOPS Coregistration. "
What should I do?

I think SNAP doesn’t support E-W displacements detection, But let’s ask our colleague @ABraun is it possible to determine a time series of single interferograms (A+B, C+D, E+F,…) of an area of landslide for example and later on get the E-W displacement by apply something similar to offset tracking?

S-1-IW, TOPS cor-registration should be applied,

Would you please to clarify, what do you intend to do?

In this case we could help more, an important thing please try to use the search before posting your demand

1 Like