labs_title

Caldeira Lab Research:Ocean acidification and ocean carbon cycle

Sensitivity of simulated salinity in a three-dimensional ocean to upper-ocean transport of salt from sea-ice formation

P.B. Duffy & K. Caldeira

Inability to accurately simulate salinity is a repeated failing of several three-dimensional ocean models. In this paper, it is shown that when the sinking of salt rejected during sea-ice formation is explicitly represented in the model, its ability to predict salinity is greatly improved.


Duffy, P. B., and K. Caldeira, Sensitivity of simulated salinity in a three-dimensional ocean model to upper-ocean transport of salt from sea-ice formation, Geophysical Research Letters 24, 1323–1326, 1997.

Salinity of the Atlantic Ocean predicted by models and determined by observation: Salinity of the Atlantic ocean is diagrammed here against latitude and depth. The top panel shows salinity determined by observation and the middle panel shows a "control simulation," salinity simulated by a normal model. The bottom panel shows a "test simulation," a model where rejected salt is explicitly represented. The test simulation yielded results that were much closer to observed values.

Salinity graphed against depth: When salinity is graphed against depth, it is shown that the currently used model predicts salinity well in shallow water but creates a deep ocean that is much too fresh. In the test run, where rejected salt is evenly distributed, the salinity of the deep ocean comes much closer to matching observations.

Abstract

We show that explicit representation of sinking of salt rejected during sea-ice formation dramatically improves simulated salinity in an ocean general circulation model (OGCM). In our "control" simulation, rejected salt goes into the top model layer, and simulated salinities are typical of OGCMs: the deep ocean is too fresh, and the intermediate-depth salinity minimum associated with Antarctic Intermediate Water is absent. These problems are eliminated in our "test" simulation, in which we distribute rejected salt uniformly over the upper 160m. Also, the strength of the Antarctic Circumpolar Current is more realistic in this simulation. These results show the need for, but do not provide, a better representation of sinking of rejected salt. The sensitivity of our model to sinking of rejected salt suggests that a similar sensitivity may exist in the real ocean, and that loss of Antarctic sea ice might have major effects.