Quantifying Downstream Climate Impacts of Sea Surface Temperature Patterns in the Eastern Tropical Pacific Using Clustering

Abstract

El Niño–Southern Oscillation (ENSO) phases and flavors, as well as off-equatorial climate modes, strongly influence sea surface temperature (SST) patterns in the eastern tropical Pacific and downstream climate. Prior studies rely on EOFs (which characterize fractional SST variance) to diagnose climate-scale SST structures, limiting the ability to link individual ENSO flavors with downstream phenomena. Hierarchical and k-means clustering methods are used to construct Eastern Pacific patterns from the ERSST dataset spanning 1950 to 2021. Cluster analysis allows for the direct linkage of individual SST years/seasons to ENSO phase, providing insight into ENSO flavors and associated downstream impacts. In this study, four clusters are revealed, each depicting unique SST patterns influenced by ENSO and Pacific Meridional Mode (PMM) phases. A case study demonstrating the utility of the clusters was also carried out using accumulated cyclone energy (ACE) in the Atlantic and Eastern Pacific basins. Results showed that Eastern Pacific (EP) El Niño suppresses Atlantic tropical cyclone (TC) activity, while Central Pacific (CP) La Niña enhances it. Further, EP El Niño, coupled with positive PMM, amplifies ACE. Ultimately, the methods used herein offer a cleaner analysis tool for identifying dominant SSTA patterns and employing those patterns to diagnose downstream climatic effects.

Keywords

• clustering
• unsupervised machine learning
• ENSO
• PMM
• sea surface temperatures
• eastern tropical Pacific