search

tobiscode / disstans

Package repository for the Decomposition and Inference of Sources through Spatiotemporal Analysis of Network Signals (DISSTANS) toolbox.
https://tobiscode.github.io/disstans/
GNU General Public License v3.0
22 stars 7 forks source link

CME plot shows abnomal large value with graphical_cme() in example_1_longvalleycaldera.py ? #7

Closed GEOSCIENCELXG closed 3 days ago

GEOSCIENCELXG commented 1 week ago

Hi, Tobis

I try net.graphical_cme(ts_in="raw_filt_res", ts_out=None, annotate_stations=True, save=True, save_kw_args={'format': 'png'}, gui_kw_args={}, method="ica",rng=np.random.default_rng(0))

to get the CME timeseries, but there is abnormally large value, do you have any idea about this?

cme_spatial cme_temporal

tobiscode commented 3 days ago

Hi, looking at the timeseries for the station, this is purely a data issue. CME requires that the input doesn't have large steps anymore, and that all major outliers are removed. P727 does at least still have the latter: http://geodesy.unr.edu/NGLStationPages/stations/P727.sta, compare the raw versus cleaned plots. This is why in Example 1, there's a cleaning step beforehand: https://tobiscode.github.io/disstans/examples/example_1.html#cleaning-the-timeseries - if this step still leaves such large outliers, you would need to adjust your settings. Cheers