Wj. Cai et al., RESPONSE OF A GLOBAL COUPLED OCEAN-ATMOSPHERE-SEA ICE CLIMATE MODEL TO AN IMPOSED NORTH-ATLANTIC HIGH-LATITUDE FRESHENING, Journal of climate, 10(5), 1997, pp. 929-948
The response of a coupled oceanic-atmospheric-sea ice climate model to
an imposed North Atlantic high-latitude freshening is examined. The i
mposed freshening lasts for 5 yr with a total salt deficit equivalent
to about eight times the observed Great Salinity Anomaly during the la
te 1960s and early 1970s. The thermohaline circulation associated with
North Atlantic Deep Water Formation (NADWF) initially weakens, but it
recovers within 20 yr of the imposed freshening being removed. The ef
fect of the weakened NADWF is gradually transmitted from high latitude
s to the entire Atlantic Ocean. The response at the equator lags that
at 62 degrees N by about 10 yr. In the midlatitude (from 30 degrees to
58 degrees N) region, the lag causes a warming during the initial wea
kening and a cooling during the recovery. Changes in the thermohaline
circulation significantly modify the large-scale North Atlantic circul
ation. In particular, the barotropic Gulf Stream weakens by about 18%.
An interesting feature is the dipole structure of the initial respons
e in sea surface temperature, with cooling in the sinking region and w
arming south of it. This dipole structure plays an important role for
the recovery of the NADWF once the imposed freshening is removed. It i
ncreases the surface density in the sinking region and increases the n
orth-south pressure gradient. Thus, the conditions set up during the i
nitial weakening contribute to the recovery process. Modifications of
the thermal structure of the ocean surface lead to changes in the atmo
spheric circulation, in particular, a weakening of the westerlies over
the midlatitude North Atlantic and a southward shift over Western Eur
ope. The North Atlantic oscillation (NAG) index under the imposed fres
hening is negative, consistent with findings from observational studie
s. The associated climate changes are similar to those observed with n
egative NAO values. Effects of various oceanic and atmospheric feedbac
ks are discussed. The results are also compared with those from ocean-
only models, where the atmosphere-ocean interactions and some of the o
ceanic feedbacks are excluded.