Using the Noise Correlation Function to Determine Relative Timing of Analog Seismograms

Session: Back to the Future: Innovative New Research with Legacy Seismic Data [Poster]
Type: Poster
Date: 4/28/2020
Time: 08:00 AM
Room: Ballroom
Description: 

Much attention needs to be given to the wealth of legacy seismic data available from around the world. These data have great value as they span a significant portion of the recording history of seismology and provide a wealth of information. However, in comparison to the GPS-timed seismic instruments of today, individual clocks or local timing systems were used to provide time information of analog recordings. Due to this uncertain synchronization of data across stations, comparison of relative timing information between analog era stations is difficult.

While several programs currently allow for extraction of digital time series from analog seismograms, achieving accurate relative timing is still critical for the application of modern analysis methods such as ambient noise cross correlations or high-precision local earthquake relocations using analog data. We demonstrate how relative time corrections between stations can be determined by leveraging the symmetry present in ambient noise cross-correlation functions. These procedures are applied to paper seismograms from the Hawaiian Volcano Observatory network during 1988. There are challenges associated with analog data, such as removal of the time mark signals and record-edge issues. These problems require new methodologies that are specifically designed to process digitized analog records. After data preprocessing, the noise correlation functions provide relative time corrections between two stations on the order of seconds for a period spanning several days. Such time corrections will help in the utilization of legacy data to extend contemporary analyses over previously inaccessible timescales.

Presenting Author: Thomas A. Lee

Authors