Detecting Slow-Slip Events in GNSS Position Time Series Using Relative Strength Index

Abstract

Slow slip (slow earthquake, silent earthquake) is an important way to release the pressure caused by the movement of the tectonic plate, and the other way is the earthquake in the conventional sense. Compared with conventional earthquakes, slow slip occurs deeper, lasts longer, and recurs periodically, but the stress intensity released is much weaker, and even does not generate seismic waves which is difficult to be detected by seismographs. The subduction zone where slow slip happens may breed super large earthquakes of magnitude 8 or higher. Studying slow slip can not only deepen the understanding of the movement of plate boundaries, but also increase the ability to predict earthquakes. GNSS technology has been widely used in monitoring plate motion because its high precision and high time resolution, we can detect slow-slip events from GNSS position time series. In this paper, the single-site transient signal detection method based on relative strength index (RSI) proposed by Brendan is used to detect the slow-slip events of the GNSS sites on the HIKURANGI subduction zone in New Zealand. The results show that the method can detect most slow-slip events, but exist missed and wrong detections within acceptable limits, and some GNSS sites have detection anomalies. This paper explored the reasons why the method detected anomalies at some sites, and got some beneficial conclusions, and gave some corresponding strategies which can strengthen the applicability and capability of the method.