I’m trying to create a catalogue of waveforms with associated picks information (mainly the type, uncertainty and pick time). The goal is to obtain something similar to the STEAD dataset for Machine Learning.
I was parsing quakeQML files available here Italian Seismic Bulletin » INGV Osservatorio Nazionale Terremoti containing events and picks, and then using the info in the files to query the fdns database for waveforms of stations that detected the event.
The process has been … long and not always satisfactory.
I was thinking of changing my approach and downloading events automatically with this sort of script
events = client.get_events(minlatitude=minlatitude, maxlatitude=maxlatitude,
minlongitude=minlongitude, maxlongitude=maxlongitude,
starttime=starttime,
endtime=endtime,
minmagnitude=minmagnitude, maxmagnitude=maxmagnitude)
but the issue is that I do not get picks information with this. If I specify includearrivals=True then I get an error that says that I need to specify an eventID. In fact if I download a list of events and then use one of the event IDs as an argument for the get_events function, then I’m able to get the picks for that event too.
Being kind of new to the fdns service, this kind of confused me. Can I do this automatically or I should fall back to my old method of downloading bulletins, parsing qml files, and then querying for waveforms?