Workflow between SNAP and StaMPS

Good afternoon everyone, I want to ask you a question regarding this post (of muhannad), and it can be very basic but I am an archaeologist and the ASCII format escapes my knowledge. That e+02, would correspond to the number of decimals that moves to the right? that is, e+01 in an ASCII case -1.123 would be -11.23? and if it is e+02 would it be -112.3?

Thanks in advance,

Best regards

It’s the scientific notation : e+02 means 10^2 which is 10x10=100.
You multiply this number by -1.123 : 100x(-1.123)=-112.3. And so on.

-1.123e+01 is -1.123x10^1=-1.123x10=-11.23
-1.123e+02 is -1.123x10^2=-1.123x10x10=-1.123x100=-112.3
-1.123e+03 is -1.123x10^3=-1.123x10x10x10=-1.123x1000=-1123

If you want it’s a shift of the decimal (.) to the right.

2 Likes

It is the same in all the steps untill the StaMPS_Export step.

I have never paid attention to the internal values of the heading. This is a good question for the developers.

However, once the StaMPS processing is completed and you plot the results on a map, points match features, so I am not sure what happens on the StaMPS export operator, but the final results looks fine to me, as each point has individual lat/lon coordinates, but maybe on your case is different.

Feel free to ask your questions to the developers, indeed, and keep us updated. Now probably we are all curious

Hi Jennifer,
Can you please share solution for Step 1,
I am facing here with similar error,

Index in position 2 exceeds array bounds.

Error in ps_load_initial_gamma (line 108)
*rg=rgn+ij(:,3)rps;

Error in stamps (line 263)
*** ps_load_initial_gamma;***

or you maybe have some another advice for solution,
Thanks,

  • Zoran

Zoran, in case you used Sentinel-1 data, have you used the mt_prep_snap script for your data?

Hi ABraun,
Yes, I am using s1 data, I run mt_prep_snap,


and I have got these six patch folders, with file structure below
,
And after that when I run stamps(1,1) from matlab I’ve got an error mentioned above,
any suggestion would be helpful,
Thank you for your fast replay,

  • Zoran

I don’t see a problem at first sight, sorry.
But maybe it is not a good idea to use StaMPS\bin folder as the working directory.

You can execute mt_prep_snap in a new folder to keep your files clean and separate.

Hi ABraun,
Can you please take a look at my geo folder,
I have got it from SNAP/…/StaMPS export, after making interferograms,
Should I replace lat and lot files…, inside geo folder?
I have not used PARTA from the picture bellow, linked in the brackets, is it necessary in SNAP7.0 (https://www.researchgate.net/publication/326655316_ESA_SNAP_-_StaMPS_Integrated_Processing_for_Sentinel-1_Persistent_Scatterer_Interferometry)?

Thanks in advance,

  • Zoran
1 Like

Sorry ABraun,
I meant on PARTB TOPSAR-Deburst(2), instead of PARTA,
Cheers,

  • Zoran

PART B will not work unless you update the xml for the ESD operator, as SNAP 7.0 changed it

I believe you have done that already. But this was just a note.

besides mdelgado’s point it looks fine to me.

Thank you both,
Yes, everything is up to date,
I will have to do deeper look on matlab,
I use matlab R2018a, it shouldnt be a problem I think, witch version you guys use?
Thank you very much once more, it is good to talk with someone about this,

  • Zoran

the matlab version shluldn’t cause any problems - I use many different and haven’t eperienced differences in the StaMPS processing.

Are there enough pixels with non-zero amplitude? (you get messages per interferogram when running mt_prep_snap)


Hi ABraun,
As you can see only my masterdate has zero amplitude,
,
I add couple screenshots, maybe you can see something what I cannot,
Cheers,

  • Zoran

I am not sure if this causes the error, but the master date should not be included in the slaves, because otherwise it is coregistered with itself and produces a null inteferogram (we discussed this here).

But I have seen workflows where this is the case and it didn’t cause an error at least, simply a single color interferogram at this point.

Excellent Mr ABraun, seems you are right,


I simply delete interferogram 20190531-20190624 from diff0 and rslc folder witch has zero amplitude, by your suggestion.
Is that allowed, or I have to go back to SNAP and redo again all stack without image 20190624?
Thank you very very much,
Cheers,

  • Zoran

very good. If all interferograms were loaded correctly I don’t think deleting the master one will cause an error at a later step within StaMPS. Go on, with each step separately [stamps(2,2), stamps(3,3),…) and see if you encounter any strange messages. If no error occures, you are good to go!

1 Like

3 has anyone had done this graph?can tell me how to create it?thank you!

first you create a file called small_baselines.list using sb_find
then you can modify the list and plot the graph with plot_sb_baselines

This is the theory, I have never used it so far, because SNAP currently does not allow to create multi-master products (only single master stacks).


souce: M.Motagh 2016