ABSTRACT Central Chile is exposed to tsunami hazard, and large earthquakes and tsunamis have occurred over the last 500 years. Tsunami hazard analysis in Chile has been traditionally implemented by means of a deterministic approach, which is based on historical events and uniform slip distribution. The objective of the present study is to improve tsunami hazard analysis in central Chile (30°S to 38°S). To encompass the purpose, stochastic earthquake scenarios of magnitude 8.8 to 9.2 were generated. Two different sets of stochastic tsunami scenarios were selected by means of the Stochastic Reduced Order Model (SROM), which were applied to Quintero bay to perform a Probabilistic Tsunami Hazard Analysis (PTHA). The results showed that PTHA of Quintero bay from stochastic tsunami scenarios agrees with paleotsunami records in the bay, while a deterministic tsunami scenario underestimated the hazard. Two sets (50 and 100 scenarios, respectively) give similar results when smaller return periods are analyzed. However, for larger return periods ( 2000 yr) the set of 100 scenarios show better results consistent with previous paleoseismological findings. The methodology implemented here can be replicated in other seismic regions in Chile as well as in other active subduction zones, thus, both near field and far field events can be analyzed.

中文翻译：

---

基于随机破裂情景的智利中部海啸灾害分析的改进

摘要 智利中部面临海啸灾害，在过去 500 年中发生过大地震和海啸。智利的海啸灾害分析传统上是通过确定性方法实施的，该方法基于历史事件和均匀的滑动分布。本研究的目的是改进智利中部（30°S 至 38°S）的海啸灾害分析。为了实现这一目的，生成了 8.8 至 9.2 级的随机地震情景。通过随机降阶模型 (SROM) 选择了两组不同的随机海啸情景，并将其应用于 Quintero 湾以执行概率海啸危害分析 (PTHA)。结果表明，随机海啸情景中昆特罗湾的 PTHA 与该湾的古海啸记录一致，而确定性的海啸情景低估了危害。当分析较小的回报期时，两组（分别为 50 和 100 个场景）给出了相似的结果。然而，对于更大的重现期（2000 年），100 种情景的集合显示出与先前古地震学发现一致的更好结果。这里实施的方法可以在智利的其他地震区以及其他活跃的俯冲带中复制，因此，可以分析近场和远场事件。