A delayed action oscillator shared by biennial, interannual, and decadal signals in the Pacific Basin

Publication Status is "Submitted" Or "In Press: 
LDEO Publication: 
Publication Type: 
Year of Publication: 
Journal Title: 
Journal of Geophysical Research-Oceans
Journal Date: 
Mar 13
Place Published: 
Tertiary Title: 
Section / Start page: 
ISBN Number: 
ISSN Number: 
Short Title: 
Accession Number: 
LDEO Publication Number: 
Call Number: 

[1] Biennial, interannual, and decadal signals in the Pacific basin are observed to share patterns and evolution in covarying sea surface temperature (SST), 18degreesC isotherm depth (Z18), zonal surface wind (ZSW), and wind stress curl (WSC) anomalies from 1955 to 1999. Each signal has warm SST anomalies propagating slowly eastward along the equator, generating westerly ZSW anomalies in their wake. These westerly ZSW anomalies produce cyclonic WSC anomalies off the equator which pump baroclinic Rossby waves in the western/central tropical North Pacific Ocean. These Rossby waves propagate westward, taking similar to6, similar to12, and similar to36 months to reach the western boundary near similar to7degreesN, similar to12degreesN, and similar to18degreesN on biennial, interannual, and decadal period scales, respectively. There, they reflect as equatorial coupled waves, propagating slowly eastward in covarying SST, Z18, and ZSW anomalies, taking similar to6, similar to12, and similar to24 months to reach the central/eastern equatorial ocean. These equatorial coupled waves produce a delayed-negative feedback to the warm SST anomalies there. The decrease in Rossby wave phase speed with latitude, the increase in meridional scale of equatorial SST anomalies with period scale, and the associated increase in latitude of Rossby wave forcing are consistent with the delayed action oscillator (DAO) model used to explain El Nino. However, this is not true of the western-boundary reflection of Rossby waves into slow equatorial coupled waves. This requires modification of the extant DAO model. We construct a modified DAO model, demonstrating how the various mechanisms and the size and sources of their delays yield the resulting frequency of each signal.


666VVTimes Cited:24Cited References Count:58

Doi 10.1029/2002jc001490