Automated Tsunami Source Modeling Using the Sweeping Window Positive Elastic Net

• 成果类型: Article
• 署名作者: Percival, Daniel M.; Percival, Donald B.; Denbo, Donald W.; Gica, Edison; Huang, Paul Y.; Mofjeld, Harold O.; Spillane, Michael C.
• 署名单位: Alphabet Inc.; Google Incorporated; University of Washington; University of Washington Seattle; University of Washington; University of Washington Seattle; University of Washington; University of Washington Seattle; National Oceanic Atmospheric Admin (NOAA) - USA; National Oceanic Atmospheric Admin (NOAA) - USA
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459
• DOI: 10.1080/01621459.2013.879062
• 发表日期: 2014
• 页码: 491-499
• 关键词: selection

摘要

In response to hazards posed by earthquake-induced tsunamis, the National Oceanographic and Atmospheric Administration developed a system for issuing timely warnings to coastal communities. This system, in part, involves matching data collected in real time from deep-ocean buoys to a database of precomputed geophysical models, each associated with a geographical location. Currently, trained operators must handpick models from the database using the epicenter of the earthquake as guidance, which can delay issuing of warnings. In this article, we introduce an automatic procedure to select models to improve the timing and accuracy of these warnings. This procedure uses an elastic-net-based penalized and constrained linear least-squares estimator in conjunction with a sweeping window. This window ensures that selected models are close spatially, which is desirable from geophysical considerations. We use the Akaike information criterion to settle on a particular window and to set the tuning parameters associated with the elastic net. Test data from the 2006 Kuril Islands and the devastating 2011 Japan tsunamis show that the automatic procedure yields model fits and verification equal to or better than those from a time-consuming hand-selected solution.