Evaluate tropospheric corrections

[Federico Carballo]

Hi all,

I want to know if mintpy generates any product or has any tool to evaluate the performance of tropospheric corrections applied to the time series. I understand that the residual RMS is applied beyond the correction and is not used for this purpose. Which in principle I assumed it was intended for this very purpose.
I am trying to analyze which is the best tropospheric correction in my area of interest or to have some support for the choice of it.

Thanks in advance.

[Zhang Yunjun]

Hi pulpero,

The residual RMS can be used to evaluate the performance of the tropospheric correction, as you said, but only to some extent because the RMS value also depends on how well the temporal deformation model you assigned during the `dem_error` step, which generates the residual phase time-series.

Tropospheric correction is still a very active area of research, which means it's not very straightforward or objective from computer's eyes to tell which method or not yet. I usually check the performance of different tropospheric corrections, by plotting the time-series before and after corrections (using view.py --wrap --wrap-range -5 5 to ensure they are in the exact same color range to help the comparison) and manually inspecting the results based on the human knowledge of the study area. 

I hope this helps,

Yunjun

[Federico Carballo]

Hi,  Yunjun

I'll try that then.

Thanks for your reply.

Fede