Spatial distribution of spectral SST oscillations over the equatorial pacific in the period 1888–2014

Abstract

AbstractThis study identified and analysed the spatial and temporal patterns of spectral oscillation ranges of the monthly sea surface temperature, SST, over the tropical Pacific region, from 1888 to 2014, through wavelet analysis. The strategic steps considered the variance analysis, obtained from the wavelet computation, for each single grid point instead of the analysis of a spatial mean variance, as is commonly done, allowing us to visualize the spatial distribution of SST variance during different time periods. The mean results indicate that lower‐frequency oscillations (2–4 and 4–8 years) over the considered area are becoming more intense while higher‐frequency oscillations (1–12 months) are becoming less intense through time, which can lead to more persistent climates around the globe. While the 1–2 year oscillations feature the highest positive trend along the equatorial belt over central and eastern locations, the 4–8 year oscillations feature the strongest signal closer to the western South American coast, and the 2–4 year oscillations present positive signals over areas closer to the continental coast and over more distant western locations intercalated by negative signals. Throughout time, the position of the highest EN signals occurs either over central or eastern areas of the equatorial Pacific. The SST variance in the Niño areas mainly oscillates within the 2–4 and 4–8 year spectral ranges. Niño1 + 2 and Niño3 regions, in opposition to Niño3 and Niño3 + 4 regions, present the highest SST variance values for the four considered spectral ranges. The strongest El Niño events present their main oscillations in the 2–4 and 4–8 year spectral ranges, especially in the Niño1 + 2 and Niño3 regions, while weaker El Niño events show more association with the 1–2 year spectral range. Although there are some patterns identified for distinct El Niño/Southern Oscillation, ENSO, intensities and the associated frequency oscillations, the distinct spatial spectral energy distribution over the equatorial areas suggests the importance of the spatial distribution of the spectral energy through time in order to better characterize the ENSO events.