Bayesian Change-Point Analysis of Tropical Cyclone Activity: The Central North Pacific Case*

• DOI: 10.1175/jcli-3248.1
• Publication Date: 2007-12-28T19:54:37Z
• Authors: Pao-Shin Chu; Xin Zhao
• Keywords: 13. Climate action; 16. Peace & justice; 01 natural sciences; 0105 earth and related environmental sciences

Abstract Bayesian analysis is applied to detect change points in the time series of annual tropical cyclone counts over central North Pacific. Specifically, a hierarchical approach involving three layers—data, parameter, and hypothesis—is formulated demonstrate posterior probability shifts throughout from 1966 2002. For data layer, Poisson process with gamma distributed intensity presumed. hypothesis “no intensity” “single are considered. Results indicate that there great likelihood point on rates around 1982, which consistent earlier work based simple log-linear regression model. A also provides means for predicting decadal variations. higher number cyclones predicted next decade when possibility early 1980s taken into account.