Numerical experiments relating to the “Great Salinity Anomaly” of the seventies in the Greenland and Norwegian seas

Abstract

Results are presented from numerical experiments with a primitive-equation ocean general circulation model. The experiments were carried out in order to investigate some aspects of the “Great Salinity Anomaly” of the seventies which was observed in the northern North Atlantic from 1968 to 1982 and in the Greenland and Norweigan seas from 1975 to 1982. The model used here extends from the Fram Strait to the Greenland-Scotland Ridge system. Thus, only the Greenland and Norwegian seas' part of the problem is considered. Two experiments were run in which the model was forced with actual 6-hourly wind stress data. The first one was intended to simulate the spread of the anomaly in the model domain. The simulated patterns of its distribution and arrival times compare reasonably well with observations. However, the amplitudes are too small by at least a factor of 2. The second experiment was run to investigate the potential influence of wind forcing on the salinity distribution. The wind stress fields of 1979, the year the observed anomaly died away, were considerably stronger than those of 1976. This latter year is in the middle of a 3-year period with anomalously low salinity values measured in the North Atlantic Water entering the Greenland and Norwegian seas through the Faeroe-Shetland Channel. One to two years later, anomalously low salinities were measured further downstream. Forcing the model with the wind stress data of 1976 and 1979, respectively, and starting each run with the same initial data, a salinity distribution is obtained which is lower at the end of 1976 than at the end of 1979 by about 0.06 below approximately 70m on a section running from Bear Island to the Fugløy Bank on the Barents Shelf. This is about half of the observed amplitude of the anomaly. Thus, differences in the wind forcing between individual years may be one of the “processes which are both possible and necessary to account for observed persistence” (DICKSON, MEINCKE, MALMBERG and LEE, 1988) of the anomaly in the Greenland and Norwegian seas.