CLUSTER ANALYSIS OF EASTERN AND CENTRAL NORTH PACIFIC TROPICAL CYCLONES AND THE INFLUENCES OF ENSO AND MJO

Abstract

While the eastern and western North Pacific’s Tropical Cyclones (TCs) have been thoroughly studied, the central Pacific hurricanes are often overlooked, which poses a problem for those living in Hawaii. Using a mixture Gaussian model and an Expectation – Maximization (EM) algorithm, the Eastern and Central North Pacific TCs can be clustered into different track types. The best-track hurricane data from 1966 to 2019 has been sorted into four distinct tracks. Once separated, each track type is examined in terms of frequency, lifetime, accumulated cyclone energy, intensity, and maximum strength. Additionally, the relationships with El Niño Southern Oscillation (ENSO) and the Madden-Julian Oscillation (MJO), along with the environmental conditions, have been examined in order to gain a better understanding of regional TC activity over the Central North Pacific (CNP) and Eastern North Pacific (ENP). In particular, Central Pacific and Eastern Pacific El Niño events have been identified in an attempt to identify an evolving pattern of hurricanes in conjunction with changing sea surface temperature anomaly trends. The various phases of ENSO and MJO have been shown to influence which track type is more dominant, while the environmental conditions of a particular type can dictate the genesis location, maximum wind speed, intensity, and translation speed of TCs.