Designing or Upgrading a Seismic Network To Meet Specific Performance Criteria Using Array Modeling, a Case Study

More science, particularly related to hazard reduction and earthquake forecasting, is enabled via the availability of rich seismic datasets and event catalogs. Deployment of high performing monitoring networks, which produce high quality datasets, is an investment that enables ongoing and future science advancements.

One measure of network performance, magnitude of completeness (Mc), is determined by a number of factors including station density, network geometry, self-noise and passband of the system used, ambient noise environment and sensor installation method and depth.  Sensor installation techniques related to depth are of particular importance due to their impact on deployment cost and station performance. It is well established that deploying seismic sensors at greater depths reduces their exposure to cultural and environmental noise, improving seismic signal detection. When extended to overall network performance, this noise suppression results in improved (decreased) magnitude of completeness for the network. Using modeling tools, we assess the theoretical improvement in performance associated with an upgrade to borehole installations, increasing sensor depth, for a real world network in the Puget Sound area of Washington State.

The goals of monitoring networks and science are overlapping and dependent. Establishing measurable and achievable performance metrics for these supported networks helps the community understand the present distribution of performance and converge on recommendations for government agencies that will benefit both science and monitoring. For example, datasets from monitoring networks with reduced Mc are likely to inform and enable earthquake forecasting research with the potential to benefit hazard reduction for the general population.