The effect of distribution of stations upon location error: Statistical tests based on the double-difference earthquake location algorithm and the bootstrap method

In this letter, we investigate the effect of station distribution (including the number and azimuthal gap of stations) upon location error based on the field data observed at Northern California Seismic Network (NCSN) using a double-difference earthquake location algorithm and a bootstrap method. The earthquakes relocated by all 117 stations are set as reference and the error of location is defined as the RMS of the difference to the reference. We find that the location error has a nonlinear relationship with the distribution of stations. The results may be used as guidelines for building seismic network.