【摘要】 侦测重大地震前相关的地震活动度异常,以此作为地震预测基础理论的研究,在近十年内常被提出讨论。其中有两个典型的算法,包括美国加州Rundle教授的"震模信息学"算法和北京尹祥础教授的"加卸载响应比值"算法。前者旨在揭示地震活动度的高异常变异性;后者则藉助于日、月引力的响应,来研究地壳块体的受损程度。文章从"临界转变理论"出发,认为不管是高异常地震活动度变异,或是象征地壳材料已严重受损的高加卸载响应比值,都是地壳即将出现临界转变(亦即重大地震)的早期征兆。也就是说,原由两个独立研究群体提出的地震活动度算法,实为临界转变的前兆信号的一体两面,可以在临界转变理论下统一起来。更多还原

【Abstract】 Seismicity-based earthquake forecasting/prediction techniques progressed rapidly in the last decade. Some were even claimed to be promising in predicting the occurrences of future large earthquakes. Yet, apparently ambiguous algorithms and lack of theory about seismicity precursor signals cast doubts on these techniques. Two fundamental issues stand out: is there any relationship between these apparently ambiguous algorithms, and is there any conceptual model that can unify these independently proposed algorithms? In this paper we suggest a unified framework based on the theory of critical transitions to link two seismicity-based predicting techniques, i.e., the loading-unloading response ratio (LURR) and pattern informatics (PI) algorithms. In terms of critical transitions, the loss of lithospheric resilience suggested by the LURR technique is theoretically associated with high seismicity variances portrayed in a PI hotspot map. This study thus implies the existence of generic seismicity precursors that could be also revealed by the recently proposed approaches of detecting early warning signals in the theory of critical transitions.更多还原