Reference and mask

[ditafaith]

I applied step 0-4 to mask. Then, when I execute step 1-6, it mentioned no mask file .../results/mask which step I should back to generate mask file ?

[yumorishita]

.../results/mask is generated in step1-5. Step0-4 is irrelevant. I cannot answer the question about the multi-frame things because it is out of the scope of LiCSBAS.

[ditafaith]

It looks strange. I re-execute step 1-5, but still no generate results/mask LiCSBAS15_mask_ts.py -i 10 -l 3 -c 0.5 --noautoadjust -t TS_output

[yumorishita]

I recommend using batch_LiCSBAS.sh. Please show me the full log.

[ditafaith]

There is no log file, since I executed all steps one by one.

[yumorishita]

Then I cannot identify the cause of the error.

[ditafaith]

Then I cannot identify the cause of the error.

Sometime I just want to customize the parameter, then only want to start from step 0-5, since I already have the ML results. However, no mask file will make fail for steps 1-6, but it possible to perform by adding --nomask

[ditafaith]

Hi @yumorishita, What is residu value in /info/13rms.txt ? My understanding residu means obtained from substracted obs and model. z In this file, there is residu of each pairs..

[yumorishita]

Do you mean 13resid.txt? It is RMS of residual for each ifg. https://github.com/yumorishita/LiCSBAS2/blob/c9d2b0500d4e4c184164b324cf2016b11c3805c0/bin/LiCSBAS13_sb_inv.py#L48C6-L48C61

[ditafaith]

Do you mean 13resid.txt? It is RMS of residual for each ifg. https://github.com/yumorishita/LiCSBAS2/blob/c9d2b0500d4e4c184164b324cf2016b11c3805c0/bin/LiCSBAS13_sb_inv.py#L48C6-L48C61

I was assumed, the residu means an error during generating an interferogram. How this value come from ?

[yumorishita]

Obs - Calc, as you said above.

[ditafaith]

Obs - Calc, as you said above.

Hmm.. I am confusing now. How do we know, which obs data and which calc data ?

[yumorishita]

Obs is the input unwrapped interferogram value, Calc is calculated from the inverted solutions.

[ditafaith]

Obs is the input unwrapped interferogram value, Calc is calculated from the inverted solutions.

Thank you for your explanation and support.

[ditafaith]

Hi @yumorishita I would like to know if licsbas allows selecting interferograms with a maximum temporal baseline, for example 24 days (forward and backward ?

[yumorishita]

You can do it manually by removing interferograms with >24 days from GEOC* directory.

[ditafaith]

You can do it manually by removing interferograms with >24 days from GEOC* directory.

Thanks @yumorishita . What is the default in licsbas for temporal baseline ? Also, if I want to use plot_network, is it similar result with all figures in directories /network ?

[yumorishita]

There is no default in LiCSBAS because LiCSBAS just use the interferograms that users prepared. As for the LiCSAR products, the default is 3 or 4 connections as far as I remember.

Also, if I want to use plot_network, is it similar result with all figures in directories /network ?

I may not understand your question correctly, but you can create similar files to /network/*png using plot_network.

[ditafaith]

Not sure, why I can't produce a network figure even though all input files already correct. I'd like to hear your side @yumorishita, in this case mentioned ERROR: bperp for 20190702 not found! However, there is a clear value for 20190702 in both files. It sounds strange, if remove 20190702, then the figures can be obtained.

[yumorishita]

The first column of the bperp_list.txt must be the single prime date, not the multi prime date. https://github.com/yumorishita/LiCSBAS/blob/master/LiCSBAS_lib/LiCSBAS_io_lib.py#L127

[yumorishita]

I cannot guarantee the correct behavior if you don't follow the format.

[ditafaith]

I cannot guarantee the correct behavior if you don't follow the format.

Hi @yumorishita I followed your codes, and got this figures, is it correct that spatial baseline less than 1 meter ? The baseline file from LiCSAR portal not update and cover from my study period (2015). The file baseline in licsar portal from 2017 only. https://drive.google.com/file/d/1DLEdkms0htVOay7TEoqLC6Glesmri_8u/view?usp=sharing

[ditafaith]

I cannot guarantee the correct behavior if you don't follow the format.

Hi @yumorishita I followed your codes, and got this figures, is it correct that spatial baseline less than 1 meter ? The baseline file from LiCSAR portal not update and cover from my study period (2015). The file baseline in licsar portal from 2017 only. https://drive.google.com/file/d/1DLEdkms0htVOay7TEoqLC6Glesmri_8u/view?usp=sharing

Hi @yumorishita I'm still curious why the original figure on the y-axes mentions a dummy with a range of -1 ~ 1 meter? This happens in the descending case, while ascending works fine.

[yumorishita]

Because a dummy baseline file (-1 to 1) is created if the baseline file does not exist or is incorrect. https://github.com/yumorishita/LiCSBAS/blob/dfdb8fadf323c821f386ed60eb0e871288961a72/bin/LiCSBAS02_ml_prep.py#L376

[ditafaith]

Because a dummy baseline file (-1 to 1) is created if the baseline file does not exist or is incorrect. https://github.com/yumorishita/LiCSBAS/blob/dfdb8fadf323c821f386ed60eb0e871288961a72/bin/LiCSBAS02_ml_prep.py#L376

Your answer actually confirms the baseline that I got directly from licsar was incorrect. Oh.. is there any another way to get the correct one ?

[yumorishita]

Maybe you can request COMET to create the correct one.

[ditafaith]

Maybe you can request COMET to create the correct one.

Unfortunately, Milan said that the descending confirmed error, they missed to obtain this value. Then, it will not cover all periods.

[irmayolii]

Hi prof, I use LiCSBAS2 for InSAR timeseries, but when I try to extract by point, it's error

(licsbas2) irma@DESKTOP-N3C4NC0:~/LiCSBAS2/112A_09051_071616$ LiCSBAS_cum2tstxt.py -g 119.889/-0.820 -i TS_GEOCml1GACOSmaskclip/cum_filt.h5 --mask TS_GEOCml1GACOSmas kclip/results/mask /home/irma/miniconda3/envs/licsbas2/lib/python3.11/site-packages/numpy/core/getlimits.py:542: UserWarning: Signature b'\x00\xd0\xcc\xcc\xcc\xcc\xcc\xcc\xfb\xbf\x00\x00\x00\x00\x00\x00' for <class 'numpy.longdouble'> does not match any known type: falling back to type probe function. This warnings indicates broken support for the dtype! machar = _get_machar(dtype)

LiCSBAS_cum2tstxt.py ver1.2.2 20230302 Y. Morishita LiCSBAS_cum2tstxt.py -g 119.889/-0.820 -i TS_GEOCml1GACOSmaskclip/cum_filt.h5 --mask TS_GEOCml1GACOSmaskclip/results/mask

ERROR: option --mask not recognized

For help, use -h or --help.

[irmayolii]

It's already solved. I replace the LiCSBAS_cum2tstxt.py (LiCSBAS2) in bin folder with LiCSBAS_cum2tstxt.py (LiCSBAS)