Natural Time Analysis: The Area under the Receiver Operating Characteristic Curve of the Order Parameter Fluctuations Minima Preceding Major Earthquakes

Abstract

It has been reported that major earthquakes are preceded by Seismic Electric Signals (SES). Observations show that in the natural time analysis of an earthquake (EQ) catalog, an SES activity starts when the fluctuations of the order parameter of seismicity exhibit a minimum. Fifteen distinct minima—observed simultaneously at two different natural time scales and deeper than a certain threshold—are found on analyzing the seismicity of Japan from 1 January 1984 to 11 March 2011 (the time of the M9 Tohoku EQ occurrence) 1 to 3 months before large EQs. Six (out of 15) of these minima preceded all shallow EQs of magnitude 7.6 or larger, while nine are followed by smaller EQs. The latter false positives can be excluded by a proper procedure (J. Geophys. Res. Space Physics 2014, 119, 9192–9206) that considers aspects of EQ networks based on similar activity patterns. These results are studied here by means of the receiver operating characteristics (ROC) technique by focusing on the area under the ROC curve (AUC). If this area, which is currently considered an effective way to summarize the overall diagnostic accuracy of a test, has the value 1, it corresponds to a perfectly accurate test. Here, we find that the AUC is around 0.95 which is evaluated as outstanding.

Assessing the Potential Earthquake Precursory Information in ULF Magnetic Data Recorded in Kanto, Japan during 2000–2010: Distance and Magnitude Dependences

“…The measurement of ultra-low-frequency (ULF: less than one Hertz) electromagnetic phenomena may be one of the most promising candidates, because of their deeper skin depths [ 1 , 2 , 8 , 17 ]. To date, a large number of ULF electromagnetic phenomena associated with earthquakes has been reported from different parts of the world, e.g., Greece [ 18 , 19 , 20 , 21 , 22 , 23 ], Italy [ 24 , 25 ], Russia [ 26 , 27 ]; America [ 28 , 29 , 30 ], Mexico [ 31 , 32 , 33 , 34 ], Taiwan [ 35 , 36 , 37 , 38 ], Indonesia [ 39 , 40 ], China [ 41 , 42 , 43 , 44 , 45 , 46 , 47 ] and Japan [ 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 ]. Furthermore, experiments and simulations have confirmed the existence of ULF seismo-electromagnetic phenomena [ 61 , 62 , 63 , 64 , ...…”

Natural Time Analysis of Seismicity within the Mexican Flat Slab before the M7.1 Earthquake on 19 September 2017

“…Note that for = 50 events, the global minimum appears during February 2016 with a value 0.089, but the minimum value attained during February 2017 is 0.096 which is the next deeper local minimum. Such minima in EQ catalogues have been shown to be statistical significant EQ precursors by various techniques like Monte-Carlo [ 91 ], random shuffling of EQ magnitudes [ 92 ], Receiver Operating Characteristics (ROC) [ 91 ], area under the ROC curve [ 93 ] and event coincidence analysis [ 38 ]. Thus, it appears that a minimum is observed several months before the strong M7.1 EQ in the Mexican flat slab.…”

“…The latter, as mentioned in Section 5.2 , is also unique during the whole period studied, see Figure 4 . On the more general question of the specificity of the variability minima as EQ precursors, one may consult the first paragraph and the ROC diagram in Figure 3 of Reference [ 93 ] that has led to an outstanding performance.…”

Detecting Earthquake-Related Anomalies of a Borehole Strain Network Based on Multi-Channel Singular Spectrum Analysis

“…Sarlis, et al investigated the behavior of seismicity after the observation of the Seismic Electric Signal activity using natural time analysis and revealed the fluctuations of the order parameter of seismicity exhibited a minimum a few days before a major earthquake [ 53 ]. In addition, this study recently has been verified to show statistical significance by means of the receiver operating characteristics (ROC) technique by focusing on the area under the ROC curve (AUC) [ 62 ] and event coincidence analysis [ 59 ]. In addition, entropy and negentropy, which serve as measures of the unknown external energy inflow and outflow the seismic system, also exhibited self-organization properties for catastrophic events [ 14 , 58 , 63 ].…”