著者

工藤 一嘉 坂上 実

出版者

東京大学地震研究所

雑誌

東京大学地震研究所彙報 (ISSN:00408972)

巻号頁・発行日

vol.59, no.2, pp.361-382, 1984-10-20

---

1983年日本海中部地震の発生に伴ない,震央から270kmも離れた新潟県下の大型石油タンクから石油が溢流した.地震により石油が溢流することは,石油コンビナートの耐震安全性を考える上で極めて重要な問題であり,盗流の原因となったスロッシングの発生状況,地震動の特徴について検討した.新潟における本震の地震観測記録類には,それぞれ欠陥があり,正確な地震動を把握できなかったが,以下のような特徴が見い出された.1.溢流の生じたタンクのスロッシング周期,8~11秒における速度応答は少なくとも100cm/sec以上で200cm/secは超えない.2.最大地動変位は10cmから20cmの間である.3.長周期で大振幅の地動変位は,表面波としての特徴を有し,新潟における厚い堆積層による増幅を受けた結果である.スロッシングに対する基本的理解は,新潟のデータを大勢としては説明するが,スロッシングが異常に大きい例もあり,基木的理解に検討を加える余地が残されている.溢流を発生させないために設定されている入力地震動レベル(速度応答=100cm/sec)は,全国平均で見れば相当安全側に考慮されていると言える.しかし新潟の事例から,地域性を含めたより詳細な規準値の設定が要請される.Oil sloshed from the huge oil tanks at Niigata due to the Nihonkai-Chubu earthquake of 1983, eventhough the epicenter was as far as 270km away. The oil sloshed out generally for periods of 8 to 11 seconds. The earthquake motions at Niigata were estimated by analyzing the displacement records and the accelerogram (SMAC) from the mainshock (M=7.7) and the accelerogram (digital recordings) from the June 21 aftershock (M=7.1). Because the horizontal components of displacement-meter were overloaded and the recorded amplitudes by the SMACB type accelerograph were too small to be analyzed exactly, the results obtained should have errors somewhat greater than we expected. However, it can be said with a certain degree of reliabilty that; 1) velocity responses of one mass system caused by the mainshock at Niigata in the periods from 8 to 11 sec. were over 100cm/sec but less than 200cm/sec, 2) the maximum ground displacement at Niigata was between 10cm and 20cm, 3) the long-period and large ground-displacements at Niigata were mostly composed of surface waves excited by the thick sedimentary layers at the site. The limit value prescribed by the regulation for the safety of oil sloshing from the tanks is defined as a velocity response of 100cm/sec. It can be said that the limit value of 100cm/sec is reasonable as an average in Japan, but it should be increased for some sites due to local seismic-wave amplifications. The data on sloshing obtained at Niigata roughly match the model for the sloshing of liquid in the cylndrical shell as it is generally understood, but the data that do not confirm to the model are included. So, clarification is needed on the scattering of the maximum sloshing wave heights as revealed in the data at Niigata.