A brief note

Coming from the gravitational-wave community, much of this code is written using code developed for the purpose of detecting and characterizing gravitational-waves. What that means is that, while many of the signal processing methods are similar, the packages used are non-standard in the seismic community. That means that, for now, if you'd like to use this code, please contact me (pmeyers279@gmail.com) and I will be happy to help.

I am hoping to include an interface that allows one to pull data using obspy and run this code. Presently, though, this code is written mainly to read "Gravitational-Wave Frames" (gwf), which is one data format in which we stored our data from the Homestake seismometer array.

A link to the related paper is here:

Also, I recognize that many of the names of classes, etc. are probably confusing. I'm working towards clearing up ambiguities.

Seismic radiometer code

I would recommend installing this in a conda environment.

This can be installed by:

1. clone repo
2. cd to top-level directory (i.e. with setup.py in it)
3. run pip install . (or pip install -e .)
4. run pip install -r requirements.txt

Hopefully then the notebooks in notebooks/ should work. There may still be an issue with certain matplotlib plotting libraries (one of which, if I recall, requires installing in a virtualenv or conda environment)

Creating a conda environment

Links

• Install conda: link
• Create environment: link

Example

• Once you've installed conda you should be able to run:

# create env
conda create --name myenv
# activate env
conda activate myenv
# Install some packages needed with conda
conda install basemap
# install pip because
# not everything is conda installable
conda install pip
# install requirements
pip install -r requirements.txt
# install seispy
pip install -e .

Running jupyter notebooks

There are quite a few example Jupyter notebooks. You can run these through your browser by running (from this directory):

conda activate myenv
cd notebooks/
jupyter notebook

It should then bring up a browser window with a file tree with all of the available notebooks.

General notes:

• Currently, we have included the locations for the homestake seismometer array in dict format in seispy/station/station.py. The specifications are units of meters, where the first value is EW, second value is NS, and third value is either depth (if coords=='depth', which is standard), or elevation (if coords=='utc'). Units are in meters. Accessing these are fairly straightforward, and examples are shown in the notebooks.

• Things like "fetch data" are currently only built for the Homestake array, and are written to fetch data from their location on a supercomputing cluster. I hope to update this in the near future to fetch stations from IRIS, for example.

• Plotting 3D maps requires matplotlib's basemap, which can sometimes be painful to install. I have also had issues pointing to it in my python installation (there may be residual evidence of this in the notebooks that you need to comment out). Please reach out if you have issues. It may be worthwhile to explore plotting with different projections that maybe don't require basemap.

FAQs: