Causes of Atlantic Ocean Climate Variability between 1958 and 1998*

• DOI: 10.1175/1520-0442(2000)013<2845:coaocv>2.0.co;2
• Publication Date: 2002-07-27T12:28:32Z
• Authors: Richard Seager; Yochanan Kushnir; Martin Visbeck; Naomi Naik; Jennifer Miller; Gerd Krahmann; Heidi Cullen
• Keywords: Mixed layer; Forcing (mathematics)

Numerical experiments are performed to examine the causes of variability Atlantic Ocean SST during period covered by National Centers for Environmental Prediction–National Center Atmospheric Research (NCEP–NCAR) reanalysis (1958–98). Three ocean models used. Two mixed layer models: one with a 75-m-deep and other variable depth layer. For both heat transports assumed remain at their diagnosed climatological values. The third model is full dynamical general circulation (GCM). All coupled subcloud atmospheric (AML). AML computes air temperature humidity balancing surface fluxes, radiative cooling, entrainment cloud base, advection eddy heat, moisture transports. forced NCEP–NCAR monthly mean winds from 1958 1998. adequately reproduce dominant pattern climate in its spatial time dependence. This familiar tripole alternating zonal bands anomalies stretching between subpolar gyre subtropics. goes along wind that corresponds North Oscillation (NAO). Analysis results reveals changes speed create subtropical while higher latitudes fluxes important. An observational analysis boundary energy balance also performed. Anomalous very closely tied anomalies. horizontal damp anomalous vertical tend cool entire midlatitude NAO's high-index phase maximum cooling exactly where gradient strengthened most. SSTs simulated compared those dynamic GCM. In far equally important as generating they act constructively. associated Ekman drifts consequently changing fluxes. Elsewhere dominate over transport. These suggest almost all last four decades can be explained response caused circulation. Changes force dampen SST. Both interannual longer timescale this way. While authors were unable find evidence transport systematically leading or lagging development anomalies, leaves open problem explaining low-frequency variability. Possible discussed reference modeling results.