Values of b and p: their Variations and Relation to Physical Processes for Earthquakes in Japan

Abstract

最近の日本における3つの地震群に対して解析したb値およびp値についてのレビューを行い，さらに，新しい解析結果を付け加えて，それらのパラメ―タと地下構造および震源過程との関連について調べた。1995年兵庫県南部地震については，約2～3年前からb値が変化したが，この変化は他の観測地の変化と対応している。また，この前兆的なb値の変化は震源域だけでなく，広い範囲で観測された。2000年鳥取県西部地震については，余震のb値とp値は本震付近で大きく，余震域の北部で小さい。この変化は地域的な構造の相違か応力変化のいずれかだと思われる。また，1989，1990，1997年に起きた前駆的地震活動ではb値は0.6程度で小さかった。これらのb値はp値とともに時間とともに増加した。1998年飛騨山地の群発地震については，相対的震源決定を行い，深さを含めて地域的なb値の変化を調べた。さらに，その結果，b値は山脈の北部で小さく南部で大きい。また，18個の大きな地震についてクーロン応力を計算したが，地震の80%は応力が増加した地域で発生している。B値の変化は応力変化と対応するように思われる。

This work reviews some results obtained already for the variations of the seismicity parametersb and p in different seismogenic and tectonic regions in Japan. We bring as well newevidence that the time and space changes in seismicity parameters are correlating well with thecrustal structure and/or some parameters of the earthquake process. In the first part of the paperwe show that several seismicity precursors (clear b-value changes, quiescence and clustering)occurred about two years before the 1995 Kobe earthquake and they correlate well with othergeophysical premonitory phenomena of the major event. The precursory phenomena occurredin a relatively large area, which corresponds probably with the preparation zone of the futureevent. In the second part, we analyze the b and p value spatial and temporal distribution for theaftershocks of the 2000 Tottori earthquake. The results indicate significant correlations betweenthe spatio-temporal pattern of b and p and the stress distribution after the main shock, aswell as the crustal structure. The swarm-like seismic sequences occurred in 1989, 1990 and1997 showed significant precursory b and p values. In the third part of the paper we analyzethe seismicity during the 1998 Hida Mountain earthquake swarm. The double-difference-relocatedevents are analyzed for their frequency-magnitude distribution and stress changes. While again the b-value is significantly different in south comparing with the north part of the epicentral area, the physical interpretation is difficult and complex. The changes in the Coulomb failure stress (?CFF) can explain the b-value distribution features, but the crustal structure may be also important. The seismicity distribution and migration, in relation with ?CFF is also discussed. We refer as well to other world-wide studies.