Trend patterns in global sea surface temperature

S.M. Barbosa, Ole Baltazar Andersen

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Isolating long-term trend in sea surface temperature (SST) from El Nino southern oscillation (ENSO) variability is fundamental for climate studies. In the present study, trend-empirical orthogonal function (EOF) analysis, a robust space-time method for extracting trend patterns, is applied to isolate low-frequency variability from time series of SST anomalies for the 1982-2006 period. The first derived trend pattern reflects a systematic decrease in SST during the 25-year period in the equatorial Pacific and an increase in most of the global ocean. The second trend pattern reflects mainly ENSO variability in the Pacific Ocean. The examination of the contribution of these low-frequency modes to the globally averaged SST fluctuations indicates that they are able to account for most (>90%) of the variability observed in global mean SST. Trend-EOFs perform better than conventional EOFs when the interest is on low-frequency rather than on maximum variance patterns, particularly for short time series such as the ones resulting from satellite retrievals.
Original languageEnglish
JournalInternational Journal of Climatology
Volume29
Issue number14
Pages (from-to)2049-2055
ISSN0899-8418
DOIs
Publication statusPublished - 2009

Keywords

  • trend empirical orthogonal functions
  • SST
  • ENSO

Fingerprint Dive into the research topics of 'Trend patterns in global sea surface temperature'. Together they form a unique fingerprint.

Cite this