Different "geo_timeseries_ERA5_ram_demErr" and "geo_velocity" output

[Sven Borgstrom]

Hi Yunjun,

I just finished a MintPy processing and I produced the two products in the subject.

Geo_velocity is automatically produced but I get a new one inverting red with blu for a personal convention (blu=uplift).

I produced also the "geo_timeseries" product to check the deformation on single targets.

As you will see from the attached files, the two figures show different edges on the right, as the geo_velocity is cut off on that side.

This is the first time this happens, at least for me.

How can I fix this?

Cheers,

Sven.

[Eric Fielding]

Hi Sven,

I am not sure what steps you took to make the two figures. One thing that surprised me a few times is that the `view.py` program automatically applies a mask from the `maskTempCoh.h5` or the `geo_maskTempCoh.h5` file, so the plot output can be different from the original file `geo_velocity.h5`.

Cheers,

++Eric

[Sven Borgstrom]

Hi Eric,

I got "geo_velocity.kmz" with the command "save_kmz.py geo/geo_velocity.h5 -c jet_r", while "geo_timeseries_ERA5_ramp_demErr.kmz" with "save_kmz_timeseries.py geo/geo_timeseries_ERA5_ramp_demErr.h5 -c jet_r".

Also the original "geo_velocity.kmz" file automatically produced by processing in the PIC folder is cut off on the right.

No problem at all with "geo_timeseries... kmz file" (whole image).

I tried also with another processing (same sensor but on the desc track), and still the same problem for "geo_velocity.kmz".

Cheers,

Sven

[Zhang Yunjun]

Hi Sven,

Here are a few things to check to help the diagnose:

1. What's the InSAR processor/product that you are using before mintpy analysis? We know that save_kmz.py currently could not handle the UTM coordinates from ASF/HyP3 products, as you have seen it on github issues.

2. Does the automatically produced geo_velocity.kmz file have the issue?

3. is the issue there when plotting using view.py? If yes, then it should be that the geo_velocity.h5 is cropped and save_kmz.py output is correct. If no, could you share your geo_velocity.h5 file for me to re-produce the issue and test?

Best,

Yunjun

[Sven Borgstrom]

Hi Yunjun,

answers to your questions:

1) I got the stack of interferograms with ISCE before MintPy processing.

2) Also the automatically produced "geo_velocity.kmz" file has the same issue.

3) I didn't use "view.py" for plotting, just "save_kmz.py geo_velocity.h5" to get "geo_velocity.kmz" I opened in Google Earth (see the screenshot from my previous message): I'll check which is the output from "view.py" (not at the office now where I have the run, so I cannot send you "geo_velocity.h5").

I have to add something, though I don't know if this is relevant.

The processing comes from a "mixed" dataset, i.e. TSX/TDX data, jointly with only one PAZ image, so I excluded this image ("mintpy.network.excludeDate = PAZ date" in the template) to check if the issue comes from the only PAZ image but the issue still remains. Keep in mind the PAZ frame covers completely my AoI.

What is strange for me is the fact "geo_velocity.kmz" is incomplete whereas "geo_timeseries...kmz" gives the whole image: in case of cropping, I would expect an issue for both, not only one but of course I'm not a developer. 

I'll try "view.py geo_velocity.h5" and let you know.

Cheers,

Sven.

[Sven Borgstrom]

Hi Yunjun,

sorry for late reply.

In the attached file you'll find the output of "view.py" ("view.py geo_velocity.h5 -c jet_r")

As you can see, the result is the whole image, though I noticed a different reference point (?!?) compared to the "geo_velocity.png" figure from my prevoius message.

I've attached also the "geo_velocity.h5" file for your inspection.

Cheers,

Sven.

[Zhang Yunjun]

Hi Sven,

Thank you for sending the h5 file. I tried the save_kmz.py geo_velocity.h5 --zero-mask -c jet_r command, and it works fine for me, as shown in the attached screenshot.

Could you make sure that you are using the latest development version of mintpy, by running cd MintPy; git pull command?

Yunjun

[Sven Borgstrom]

Hi Yunjun,
at the end I fixed the issue.
Considering the "geo_velocity.h5" was not corrupted as I saw from your figure, the only source of error could be Google Earth:
GE on Linux (Ubuntu 18.04) still gave me the problem, whereas GE on Mac OS (Catalina) worked fine, quite strange.
About the MintPy developement version, I always pay attention to get the last one ;)
Cheers,
Sven.

[Zhang Yunjun]

Hi Sven,

I am glad to hear you found the cause of the issue. 

Cheers,
Yunjun

--
You received this message because you are subscribed to a topic in the Google Groups "MintPy" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mintpy/TEv5CGbaCXk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mintpy+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mintpy/b25f491b-39f6-4855-880b-68effc9bcb77n%40googlegroups.com.