Earthquake cycles and physical modeling of the process leading up to a large earthquake

Abstract

A thorough discussion is made on what the rational constitutive law for earthquake ruptures ought to be from the standpoint of the physics of rock friction and fracture on the basis of solid facts observed in the laboratory. From this standpoint, it is concluded that the constitutive law should be a slip-dependent law with parameters that may depend on slip rate or time. With the long-term goal of establishing a rational methodology of forecasting large earthquakes, the entire process of one cycle for a typical, large earthquake is modeled, and a comprehensive scenario that unifies individual models for intermediate-and short-term (immediate) forecasts is presented within the framework based on the slip-dependent constitutive law and the earthquake cycle model. The earthquake cycle includes the phase of accumulation of elastic strain energy with tectonic loading (phase II), and the phase of rupture nucleation at the critical stage where an adequate amount of the elastic strain energy has been stored (phase III). Phase II plays a critical role in physical modeling of intermediate-term forecasting, and phase III in physical modeling of short-term (immediate) forecasting. The seismogenic layer and individual faults therein are inhomogeneous, and some of the physical quantities inherent in earthquake ruptures exhibit scale-dependence. It is therefore critically important to incorporate the properties of inhomogeneity and physical scaling, in order to construct realistic, unified scenarios with predictive capability. The scenario presented may be significant and useful as a necessary first step for establishing the methodology for forecasting large earthquakes.

Spatio-temporal distribution of seismic moment release near the source area of the 2011 Tohoku-Oki earthquake

“…doi:10.5047/eps.2012.04.006 et al, 1993;Ohnaka, 2004;Oishi and Sato, 2007). However, some studies have obtained weak interplate coupling (20-40%) (Peterson and Seno, 1984;Pacheco et al, 1993;Matsumoto et al, 2008), whereas others have shown strong coupling along the Japan Trench (Nishimura et al, 2004;Suwa et al, 2006;Hashimoto et al, 2009).…”

Reply to “Comment on ‘Earthquake cycles and physical modeling of the process leading up to a large earthquake’”

“…The most recent comment by Bizzarri and Cocco (2006) represents their counterargument to this remark. The basis of their reasoning is that even if the +1 term was included in the argument of the logarithm in equations (1c) and (1d) described in the paper (Ohnaka, 2004), the traction dependence on slip rate would remain in the governing equation, because the state variable depends on the fault slip velocity (Bizzarri and Cocco, 2006). I feel the need to point out, however, that their counterargument still appears to be logically inconsistent.…”

“…In response to Cocco and Bizzarri's (2002) argument that "there is no need to assume that friction must become independent of slip rate at high speeds to resemble slip weakening", I remarked in my previous paper (Ohnaka, 2004) that "this argument seems logically inconsistent, because the effect of high velocity cutoff has been incorporated into the equation used in their simulation." The most recent comment by Bizzarri and Cocco (2006) represents their counterargument to this remark.…”

“…where 0 is state atḊ = 0; for the notation of the other parameters, refer to Ohnaka (2004). In their calcu-lation of dynamic rupture regime under the rate-and statedependent formulation, Cocco and Bizzarri (2002) The rate-and state-dependent constitutive law was derived by Dieterich and Ruina (for references, see Ohnaka, 2004) based on laboratory experiments on extremely low slip rates (of the order of 1 mm/year up to 1 mm/s; for the references, see Dieterich and Kilgore, 1996), from nearly a steady state value to another steady state value; as such, the law does not warrant its extension to high slip velocities of the order of 1 m/s or higher experienced during actual dynamic rupture processes.…”

“…In their calcu-lation of dynamic rupture regime under the rate-and statedependent formulation, Cocco and Bizzarri (2002) The rate-and state-dependent constitutive law was derived by Dieterich and Ruina (for references, see Ohnaka, 2004) based on laboratory experiments on extremely low slip rates (of the order of 1 mm/year up to 1 mm/s; for the references, see Dieterich and Kilgore, 1996), from nearly a steady state value to another steady state value; as such, the law does not warrant its extension to high slip velocities of the order of 1 m/s or higher experienced during actual dynamic rupture processes. However, Cocco and Bizzarri (2002) and Bizzarri et al (2001) have arbitrarily extended mathematically this law to include such high slip velocities in dynamic rupture regimes-without any consideration of the underlying physics.…”