Low-Frequency Variability of the Equatorial Pacific Ocean Using a New Pseudostress Dataset: 1930–1989

Abstract
Interannual and interdecadal variability of the equatorial Pacific are examined using a new pseudostress dataset. The monthly mean pseudostress fields (1930–89)are derived from Comprehensive Ocean-Atmosphere Data Set (COADS) pseudostresses using climate basis functions obtained from the Florida State University pseudostress product (1966–90). To validate the new wind fields, a two-tier validation scheme was used. The new wind fields were first examined to see if they exhibited characteristics that have been shown to be important in terms of exciting El Niño events. Next, the new wind fields are used to force an ocean model, thereby obtaining model estimates of tropical Pacific currents and model upper-layer thickness (ULT). Observed sea level and spatially averaged SST anomalies are used to validate the hindcasts. The new wind fields were found to have a significant El Niño mode (accounting for 41% of the variance), which possessed features consistent with those that theory and numerical simulations dictate. Interannual fluctuations in modeled and observed sea level fluctuations are compared at Galapagos and Truk, yielding correlation r values of 0.73 and 0.71, respectively. The comparison of the interannual fluctuations in modeled ULT and observed SST anomalies, which are both spatially averaged over a subdomain in the eastern Pacific basin, results in an r value of 0.64. Interdecadal fluctuations in eastern Pacific model ULT are found to be qualitatively consistent with those in the spatially averaged observed SST anomalies. Stronger El Niño events are observed to occur during periods of higher Pacific ULT (for decadal and longer timescales). Comparison of interdecadal fluctuations in global mean land air temperature and eastern Pacific ULT suggests a connection between eastern Pacific SST and tropical mean land air-temperature warming for interdecadal timescales.

This publication has 0 references indexed in Scilit: