PickAx: a simple seismic picker, when you just need to dig out a few arrivals.
pickax -h
Hi PickAx!
usage: pickax [-h] [-v] [-l LOADER] [-s SEIS]
Pickax, really simple seismic phase picker.
options:
-h, --help show this help message and exit
-v, --verbose increase output verbosity
-l LOADER, --loader LOADER
Initialization loader script, run at startup
-s SEIS, --seis SEIS Seismogram file, loaded at startup
For example, using simple.py to initialize pickax, load some data from files and open the picker window:
pickax -l simple.py
Or to view just a single miniseed file, ie quick look:
pickax -s JKYD.mseed
Picks are draggable with the mouse.
PickAx tries to be very simple and do just one thing, just like
a pickax is good for digging a hole in
red clay, but is just one of the tools in your toolshed. So don't forget
to use your shovels, screwdrivers and lawnmowers for the things they do better.
Basically all data management is handed off to the startup
script. By setting the finishFn, you can save picks however you want (or not
at all), and load in the next seismogram for display. Setting the filters
allows toggling between different ways of processing the seismogram.
The creation_info provides default creation info for the pick,
primarily for author or agency_id, defaulting to author=os.getlogin()
.
See simple.py for an example of setting up pickax and loading data.
The fdsnws_query.py is another example that uses FDSN web services to load data on the fly.
Filters are often a way of applying actual time series filters, but can
be anything that modifies the waveform.
They are each a simple dictionary with a name
and fn
.
The function will be called
with four arguments. First is a copy of the original stream, so modification in
place is safe. The second is the current filtered stream, so filters can build
on the previous if that is what you want. Third is the name for the current
filter function, and last is the filter index, just for reference.
If a stream is returned, that becomes
the current displayed stream, but if None is returned, then it assumes
the original was modified in place.
The finish function is called whenever the user quits, goes to next or previous,
ie q
, V
, v
, R
or r
. It is called with four arguments, first is the QuakeML
Event, which contains picks, including both new picks and any existing picks.
Second is the current stream, useful to get the channel. Third is the command,
one of "quit", "next", "next_quake", "prev" or "prev_quake" and the last is pickax itself.
For next and prev, you generally will call
pickax.update_data()
passing in a new stream and optionally
quake and inventory if they have changed.
The main window can fave an additional title string, generated by titleFn. The default is origin time, lat/lon depth and magnitude of the event. The titleFn takes 3 arguments, quake, inventory and stream, similar to the update_data function.
usgspicks: reload event from USGS by id to load picks as default query does not include picks. Likely only works if host is USGS server.
mergepicks: merge picks from one or more QuakeML files into another, optionally limiting to author
dumppicks: print picks from one or more QuakeML files
conda create -n pickax python=3.10
conda activate pickax
python3 -m pip install --upgrade build
/bin/rm -f dist/* && python3 -m build
pip3 install dist/seismic_pickax-*-py3-none-any.whl --force-reinstall
or if all deps are already installed, much faster:
pip3 install dist/seismic_pickax-*-py3-none-any.whl --force-reinstall --no-deps
to publish:
python3 -m twine upload dist/*
for testing, use code in current directory so updates on edit:
pip install -v -e .