Orbital ramps in Sentinel data

[Teije Horst, van der]

Hello there,

I have recently computed a stack of Sentinel-1 interferograms and the results seem quite satisfying. The only thing that seems really off are the orbital ramps which influence the results quite a bit. Does anyone have a suggestion what could be the reason behind this?

What is exactly estimated in the orbital ramps? Any linear slope of the phase in any direction?

Best regards, Teije

[Karsten Spaans]

Hi Teije,

Sentinel 1 orbits are very high quality, so there should be no orbital ramps. The orbit ramp estimation in StaMPS estimates a linear ramp in any direction. Any long wavelength (e.g. tropospheric delay) signal will therefore affect this estimation. I suggest you do not use the orbital ramp correction.

Cheers,

Karsten

--
You received this message because you are subscribed to the Google Groups "MAINSAR" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mainsar+u...@googlegroups.com.
To post to this group, send email to mai...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[n.abdel]

Try to do that : 

1-set parametre 'scla_deramp' to 'y'
2-repeat stept 6 and 7 several times

and try to plot u-dmo, that's the unwrapped phase with removed orbital rampe and other errors, and v-do for velocity.

I have a question, what is the density of your PS ? and witch parametres values you have used in Steps 2 and 3 ?

[Katleen Wils]

Hi Karsten,

Is this the same for Envisat data? I also have a long-wavelength trend in my data, but I'm afraid I would be deleting actual deformation signal by applying this orbit ramp estimation.

If I use the -o option, the entire signal is removed. Is there a way to be sure something is related to orbit ramp rather than deformation?

To check if it is related to atmosphere, I will try to use a weather model.

Thanks!

Katleen

Op maandag 17 oktober 2016 17:56:16 UTC+2 schreef Karsten:

[David Bekaert]

Envisat orbits are not as good as the Sentinel-1 orbits.

The orbit ramp correction is a de-ramping of the phase on the fly.

It would remove any long-wavelength signal present in the data as it cannot separate between the different signals.

e.g. troposphere, tectonic, orbit etc.

For envisat the oscillator drift (a ramp in range direction) is removed prior to the 'o' de-ramping. 

In terms of step 8, this is spatial temporal filtering.

One should be careful to select the input parameters such one does not remove deformation either.

Using auxiliary data to correct for the APS such as with TRAIN would be indeed one way to avoid deformation contamination in the correction.

Cheers,

D.

To unsubscribe from this group and stop receiving emails from it, send an email to mainsar+unsubscribe@googlegroups.com.