About time series analysis and mean LOS velocity

210 views
Skip to first unread message

Malvas

unread,
Nov 25, 2020, 12:56:48 PM11/25/20
to MintPy

Hello everyone,

I have always had a question about the time series analysis. For example, in order to estimate the parameters of a contracting volcanic source at depth, I would use the mean LOS velocity, right? But, could it not be possible that volcanic events are hidden after applying the time series analysis?

And regarding the mean LOS velocity, it could be that in a period of time two volcanic events occur. One of these ended in an uplift and another one in a subsidence. In this case the mean deformation pro period of time is zero. How to understand the time series analysis in this case?

I would appreciate an explanation about that!

Best regards,
Alejandra

Zhang Yunjun

unread,
Nov 25, 2020, 4:58:16 PM11/25/20
to MintPy
Hi Alejandra,

Good question. As a volcanologist myself, I do not use the average velocity of the whole time period, for the extract reason you said. I use the final displacement time-series file directly and grab the cumulative displacement for my time period of interest. save_gdal/roipac.py could do that. This is the strength of the mintpy time-series because it is non-parametric.

Cheers,

Yunjun

Ale

unread,
Nov 26, 2020, 6:52:19 PM11/26/20
to min...@googlegroups.com
Hi Yunjun,

thanks for your answer! Do you mean this command: save_gdal.py geo/geo_timeseries_ERA5_demErr.h5 -d 20190111_20190821 --of ENVI, where -d corresponds to the dates of my timeseries (first and last SAR-image)?

Best regards,
Alejandra

--
You received this message because you are subscribed to the Google Groups "MintPy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mintpy+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mintpy/c7a91a71-e745-403d-bbb6-da8e72e77d17n%40googlegroups.com.

Zhang Yunjun

unread,
Nov 27, 2020, 2:10:50 PM11/27/20
to MintPy
Yes.

Malvas

unread,
Dec 2, 2020, 3:21:09 PM12/2/20
to MintPy
Hello Yunjun,

I am trying to write a script to find the most likely parameters of a volcanic magma source using a basic Mogi source. At first I want to plot the displacement data from the time-series file, which I get with save_gdal.py, as you suggested me.  When masking with  os.system(f'mask.py {date12}.unw -m geo_maskTempCoh.h5 -o {date12}_msk.unw')  I get this:

  File "/scratch/uni/u253/u300850/MintPy/mintpy/utils/readfile.py", line 709, in read_attribute
    raise FileNotFoundError('No metadata file found for data file: {}'.format(fname))
FileNotFoundError: No metadata file found for data file: 20190111_20200821.unw


Could you please tell how could I fix this?

Thanks a lot!

Best regards,
Alejandra
Reply all
Reply to author
Forward
0 new messages