兵庫県南部地震による地震動と基盤構造 : 甲陽断層周辺について  [in Japanese] Ground motions during the 1995 Hyogo-ken Nanbu Earth quake and bedrock structure around Koyo fault  [in Japanese]

1995年兵庫県南部地震のいわゆる「震災の帯」の成因については, 地震工学・応用地震学の観点からは, 強い震動が生じた原因の究明が課題となる。本論では, これまで明らかにされてきた地震動の特徴とそれを生起した要因-震源破壊過程, 波動の伝播による減衰, 広い意味における地盤震動特性-を概観する。3次元シミュレーションによる強震動分布は, 神戸市域中部の被害分布などは概ねうまく説明するが, 東灘区以東の複雑な被害分布や甲陽断層の南西延長上付近の広義変位地形などは説明できない。甲陽断層周辺の基盤構造の詳細な研究が不可欠である。ブーゲー重力異常, 余震・脈動観測資料などからモデル化した基盤岩構造からは, 甲陽断層は2段の逆断層を形成し, 西の芦屋断層と共に走向を転じながら盆地境界を形成していることが示唆される。ただし, 甲陽断層が南西方向に「震災の帯」の下まで落差を減じながら伸びている可能性もあり, 今後の研究が待たれる。

The Hyogo-ken Nanbu (Kobe) Earthquake of January 17, 1995 (M=7.2) brought destructive damage to the Kobe and Hanshin area in the northern margin of the Osaka basin. The serious damage was distributed on E-W oriented belt-like zone 1-2 km far from the basin edge between Rokko mountains and Osaka Bay. From the point of view of engineering seismology, the question is how the strong motions were produced in the belt-like zone. We review several studies on the characteristics of ground motions and main causes producing the strong motions. After shock observations and 3-D simulations of wave field explain that strong motions were produced mainly by combined effects of propagating rupture process of seismogenic fault under the mountains and vibrational characteristics of thick soil-sediments lying on bedrock with step-like configuration along basin edge. The distribution of simulated strong motions corresponds to damage distribution in its central and western part. However, the simulation do not produce strong motions in the eastern part of the damaged zone, where the damage distribution was complex and systematic ground-failures were observed along SW extension of Koyo fault across the belt -like zone. From a seismological point of view, it is considered that Koyo fault did not moveas the seismogenic fault of the earthquake. To discuss detailed bedrock structure around Koyo fault, a dense gravity survey and observations of after shocks and microseisms (long-period microtremors) were carried out. Modeling the bedrock structure around Koyo fault with these geophysical data suggests that, bedrock subsides by 700-850 m across the Koyo fault, possibly making two reverse faults, and by 100-200 m across the Ashiya fault, and that these faults with NE-SW strike appear to change their strike to WSW and joint together to make basin edge in their western part. There is also a possibility that Koyo fault extends southwestward. Even so, subsidence of bedrock seems much smaller than that along the basin edge. In this case, soil deposits along SW extension of Koyo fault might be weak portions to ground vibrations, resulting in making the ground failures in the damaged belt-like zone.