Earthquakes Precursors and Earthquake Prediction: Recent Advances

Seismic hazard assessment and medium/long-term earthquakes forecasting have long been key research topics for the national governments interested in giving scientific bases to their building codes in earthquakes prone areas. Methods like the PSHA (Probabilistic Seismic Hazard Assessment, e.g. Cornell, 1968) were used to extrapolate from the (often very small) frequencies of past earthquakes the annual probability of occurrence of (even very large) earthquakes in the future. Such methods were strongly criticized for “… the significant consequences of their failures in terms of human and economic losses…” ( Wyss et al., 2012). In Japan “…since 1979, earthquakes that caused 10 or more fatalities … actually occurred in places assigned a relatively low probability” ( Geller, 2011). In the same country the underestimation of the Tohoku earthquake (March 11th 2011, MW = 9.0) intensity by the Japanese National Seismic Hazard Maps and the intrinsic limits of PSHA were at last officially recognized (e.g., Kagan and Jackson, 2013 and Fujiwara et al., 2013). Together with the 2011 Tohoku event, Stein et al. (2012) report other highly destructive earthquakes (e.g. 2008 Wenchuan, 2010 Haiti, etc.) occurred in areas predicted by to be relatively safe. The ground accelerations of most strong earthquakes occurred in 2000–2010 were significantly underestimated (Kossobokov and Nekrasova, 2012) by the GSHAP (Global Seismic Hazard Project) maps (based on PSHA). Not rarely, and often after a major earthquake, the probabilistic maps of seismic hazard require a revision with a general increase of the seismic hazard in the affected area (e.g. Bommer and Abrahamson, 2006). Such hazard underestimations (dramatically recognized only after disastrous earthquakes) are particularly frequent in areas where neither historical or instrumental events are reported in seismic catalogues. Their possible consequences were properly described in Peresan et al., 2013 and Artioli et al., 2013. In this context the interest in alternative observational techniques and appropriate data analysis methods to improve the seismic hazard assessment in the short-medium term increases every day. Various independent observations have been reported for many decades to support models describing the earthquake generation process as a non-random phenomenon that involves wide spatial and temporal scales often (even if not always) culminating in a large event in correspondence to some critical point. Anomalous changes in the physical/chemical state of the Earth lithosphere, atmosphere and ionosphere, which could be timely identified provided that adequate measurement systems are deployed and appropriate data analysis methods are used, can be associated to those processes. Several geophysical parameters (see for instance Tronin, 2006 and Cicerone et al., 2009, and reference herein) have been proposed as possible earthquake precursors for decades. Among them ground deformations (uplift and tilt), abrupt changes in gas emission rates, underground water level, temperature and chemical composition, electrical properties of rocks, atmosphere and ionosphere, near surface air temperature and relative humidity, and the Earth’s thermal emission were observed. A large number of theoretical models and laboratory experiments, which could explain the occurrence of the observed anomalies in relation with the preparatory phase of an earthquake (e.g. Scholz et al., 1973, Tronin, 1996, Pulinets and Boyarchuk, 2004, Freund et al., 2006, Pulinets and Ouzounov, 2011 and Tramutoli et al., 2013, etc.), were proposed at the same time. This special issue of the “Physics and Chemistry of the Earth” journal contains twenty-four articles, most of them presented at 2014 EGU Assembly, offering an updated review of the recent advances in the study of earthquake precursors and related physical phenomena.