eqstats

A python package for performing statistical tests on earthquake catalogs

Overview

This package contains code useful for performing statistical tests on earthquake catalogs. This includes performing the statistical tests themselves, as well as tools for generating random catalogs in order to calibrate tests. The tests and methods are outlined in this (paper)[https://doi.org/10.1002/2014JB011777]. If you use this code in any further work, please cite this paper as well as this package.

Installation

To install the package, from the directory where you have downloaded the source code enter the following into the shell:

$ python setup.py install

This should copy the necessary files into your Python installation. The code requires numpy, scipy, and statsmodels, all of which can be installed using pip. I have mainly used the code with Python 3.

Usage

The package contains two submodules: catalogs and stattests.

• catalogs has functions for generating random event times, event magnitudes that follow a Gutenberg-Richter magnitude/frequency distribution, and event times that follow an Omori-like aftershock time decay. These can be combined to create synthetic catalogs that are random and/or contain aftershocks and can approximate real seismicity for calibrating statistical tests.
• stattests contains the various functions for computing test statistics on a given catalog. Each test requires inputs specific to that test, so it is up to the user to compute the correct inputs. Most tests require either recurrence or occurrence times of events, and some require hyperparameters such as number of bins. One test (bigtrig_test) requires event magnitudes as well as a magnitude threshold, as it makes the assumption that large events lead to a detectable increase in seismicity. This submodule also has functions for computing p-values and detection power.