SBAS aps about gmtsar

[Xiaohua-Eric-Xu]

Hi Simon,
The SBAS program has a built-in symmetry threshold (if I remember right) for atmospheric phase estimation. This was designed for satellites with more regular acquisitions like Sentinel where you could maintain seasonal deformation even after taking out turbulent atmospheric component. This is probably the reason why all your aps grids are zeros. You can probably change this by modifying sbas_utils.c at line 712&723. For now, it's acceptable if the two scenes are 5 days off.
712 if (hit[n * S + j] == 1 && fabs((time[n] - time[i]) - (time[j] - time[n])) < 5)
change 5 to a larger number if you really want this though I would not recommend it.

Xiaohua(Eric)

[simonst7]

Thank Eric, i supposed that problem, is there any way to calculate atmospherics delays between scenes with large temporal acquisition? Because you not recommend the change sbas_utils.c, My problem is the little data ALOS to the study area, i have only small perpendicular baseline between scenes in 1 yr or more.
Simon

[Xiaohua-Eric-Xu]

Hi Simon,
If you are only interested in the deformation velocity, stacking might be a better option.
We do have an ionosphere correction module, but it requires some tuning for the dataset in order to use it.
Another alternative is to look for atmospheric models.
Xiaohua(Eric)

[simonst7]

I understsnd, thank Eric for you support. I Will try that solution