Quality control measures and analysis of EarthScope's USArray Transportable Array in the conterminous United States
Abstract
Between 2004-2015, the Incorporated Research Institutions for Seismology (IRIS) built and operated the National Science Foundation sponsored EarthScope Transportable Array (TA), a network of 400 seismographs that moved west to east across the conterminous ("Lower 48" or L48) United States to occupy 1679 sites with an average deployment span of 22 months at each location. The goal was to gather robust and uniformly spaced measurements of the seismic wavefield at a range of scales and bandwidths for use in the investigation of the geologic structure and geodynamics of the North American continent. To accomplish this, IRIS designed the TA to meet a data performance metric of >85% uptime, minimize gaps in the recorded data, and produce seismic records with low noise to maximize both earthquake detection and seismic imaging techniques.
Here we show the quality control measures applied throughout the TA data handling process, which included data accuracy, data integrity, and signal quality. The station calibration and orientation information, or data accuracy, is typically within the nominal instrument response and known within 0.2° with Octans gyroscopes [e.g., Ekström and Busby, 2008], respectively. Data integrity, or time series completeness, includes data availability, number of gaps, and dead channel information, with metrics obtained through the IRIS MUSTANG (Modular Utility for STAtistical kNowledge Gathering) web service. We find that on average, stations produced a usable time series 99.7% of the time, and >120 stations operated without any gaps throughout their deployment. Lastly, we assess signal quality of the time series with the composite probability density function of hourly-calculated power spectral density measurements. We observe that the average noise level is well below the high noise model of Peterson [1993] for all three components. The L48 TA outperforms other temporary and permanent deployments in terms of noise due to aspects of its siting, design, and emplacement. We find that its performance, when considered as individual stations or as an entire network, either met or exceeded expectations related to predefined metrics and overall data quality. We also highlight other aspects of the L48 TA which are fully documented in a summary report, available here: <a href="https://bit.ly/2LIg0Yj>"Lower-48 TA Report"</a>.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFM.S43E0674S
- Keywords:
-
- 7294 Seismic instruments and networks;
- SEISMOLOGY