GAMMA Questions

Hello,
To simulate phase, you should always use phase_sim_orb, not phase_sim!
phase_sim_orb uses an exact model for the baseline individually calculated for
each sample, where has phase_sim uses a simple polynomial baseline model, that
should only be used if there are no state vectors, or you are trying to model
residual baseline error due to inaccurate state vectors.

Best regards,
Charles

Hello Charles,

I also try to use "phase_sim_orb", however, this always product the unflattened interferogram when I need the flatten topography phase.
In addition,I found that the "phase_sim” is ok for TDX/TSX data, and I really want to understand the same baseline and DEM adjusted, but the results are different.
To make the phase (according to the base_perp and DEM, and using the linear baseline mode) consistent with the result (phase_sim), what changes do I need to make?

Kind regards.

Hello,
I do not know exactly what you are trying to do, but it if is trying to update a
DEM using corrections from the phase of a differential interferogram:  Take the
interferogram with the orbital and topographic phase removed (flatten is not a
good term for an interferogram, since we are using an ellipsoid, not the flat
earth!). and unwrap that. Then use the program dh_map_orb to get a height
correction.

If you want the phase due to topography alone, without the ellipsoidal phase:

Simulate the phase entering 0.00001 for the height as input (you can enter a
number for the height, rather than a file name).
and subtract that simulated phase from the output from phase_sim_orb including
the original DEM.  The program to subtract phase:
sub_phase. You can subtract unwrapped phase either from unwrapped phase or the
complex interferogram.
Best regards,
Charles

Hi Charles,

Thanks for your suggestion. I will give it to my study. However, I would like to ask one more question. Except for the ellipsoid effect, has the polynomial baseline model in the program of phase_sim considered the effect of the topography height on basedline?

Best regards.

Hello,

The baseline model is only defined relative to the track. The program
phase_sim_orb calculates the slant range from each satellite position to a point
on the ground and the interferometric phase is simply proportional to the
difference (-4pi (R1-R2)/wavelength). Hence there is no parametric baseline
model. The baseline vector is simply the vector difference of the slant range
vectors and changes from point to point in the scene.

The polynomial baseline model in phase_sim was developed when patches were small
and is limited to small scenes.

Best regards,
Charles

Hi Charles,

Recently, I am processing the TanDEM-X bistatic data with the routine InSAR workfolw (e.g., SLC_interp, SLC_intf, base_init, ... and so on.)

However, I found all the differential interforam always contains visible stripe information. Thus, I guess that the baseline estimation is inaccuracy.

(1) As the project (https://dep1doc.gfz-potsdam.de/boards/33/topics/230?r=235#message-235) says, I am not sure that is the "base_init" suitable for TanDEM-X data at present? If not, how to get the parameters more accurately, which will be important to calculate the vertical wavenumber (kz).

(2) For the routine InSAR processing, if I can obtain several ground control points (GCPs) that can be used to refine the inaccuracy baseline parameter, can the software be used to estracted the correspoding information in the position of the GCPs (as the extract_gcp)?

Best regards.