Statistical characteristics of non-volcanic tremor distributions along the Mexican Subduction Zone

Abstract. We analyze statistical characteristics of non-volcanic tremor sequences determined in the Mexican subduction zone. To achieve this objective, we used different techniques such as the Gutenberg-Richter relationship, non-extensive statistics, and the multifractal detrended moving average analysis to extract information on the magnitude and interevent‐time distributions. The b-value results reveal that b fluctuates from 1.25 to 2.42, with the highest values corresponding to the plate interface down-dip regions. On the other hand, the q-value shows an inverse behavior, having the highest values in the inter-plate coupling region. Similar to tectonic earthquakes, the non-volcanic tremor sequences show a multifractal structure in magnitude and interevent data. The multifractality analysis suggests that multifractality may be associated with long-term correlations, the probability distribution of the data, and the presence of nonlinearities. Regarding the existence of apparent and intrinsic multifractality, our results indicate that both sources are present in the sequences, with the former being the most common. Our estimates of the Hurst exponent are in the interval of 0.65 to 1.06, with the majority indicating an exceptionally high persistent memory (H > 0.95). With respect to the estimation of the distribution that better describes the interevent sequences, we find that most sequences can be described by a Lognormal distribution and, to a lesser extent, by a Gamma distribution. Our investigation also showed that observations of the duration of tectonic tremors present a large scatter, resulting in scaling relationships with low values of the determination coefficient. The source of this variability may be related to the generation mechanism of NVT or to the process of detection and description of the signals.