A COUPLED ATMOSPHERE-OCEAN MODEL FOR TRANSIENT CLIMATE-CHANGE STUDIES

Citation
Gl. Russell et al., A COUPLED ATMOSPHERE-OCEAN MODEL FOR TRANSIENT CLIMATE-CHANGE STUDIES, Atmosphere-ocean, 33(4), 1995, pp. 683-730
Citations number
67
Categorie Soggetti
Oceanografhy,"Metereology & Atmospheric Sciences
Journal title
ISSN journal
07055900
Volume
33
Issue
4
Year of publication
1995
Pages
683 - 730
Database
ISI
SICI code
0705-5900(1995)33:4<683:ACAMFT>2.0.ZU;2-D
Abstract
A new coupled atmosphere-ocean model has been developed for climate pr edictions at decade to century scales. The atmospheric model is simila r to that of Hansen et al. (1983) except that the atmospheric dynamic equations for mass and momentum are solved using Arakawa and Lamb's (1 977) C grid scheme and the advection of potential enthalpy and water v apour uses the linear upstream scheme (Russell and Lerner, 1981). The new global ocean model conserves mass, allows for divergent flow, has a free surface and uses the linear upstream scheme for the advection o f potential enthalpy and salt Both models run at 4 degrees x 5 degrees resolution, with 9 vertical layers for the atmosphere and 13 layers f or the ocean. Twelve straits are included allowing for subgrid-scale w ater flow Runoff from land is routed into appropriate ocean basins. At mospheric and oceanic surface fluxes are of opposite sign and are appl ied synchronously. Flux adjustments are not used. Except for partial s trength alternating binomial filters (Shapiro, 1970), which are applie d to the momentum components in the atmosphere and oceans, there is no explicit horizontal diffusion. A 120-year simulation of the coupled m odel starting from the oceanic initial conditions of Levitus (1982) is discussed. The model dynamics stabilize after several decades. The ma ximum northward ocean heat flux is 1.4 x 10(15) W at 16 degrees N. The model appears to maintain the vertical gradients characterizing the s eparation between the upper and deep ocean spheres. Inadequacies in th e coupled model simulation lead to decreasing temperature and salinity in the high latitude North Atlantic and to a poor simulation of the n orthern North Atlantic thermohaline circulation. The mass transport of the Gulf Stream is about half of observed values, while the transport s of the Kuroshio and Antarctic Circumpolar Currents are similar to ob servations. Additional deficiencies include a climate drift in the sur face air temperature of 0.006 degrees C year(-1) due to a radiation im balance of 7.4 Wm(-2) at the top of the atmosphere and too warm temper atures in the eastern portions of tropical oceans. The coupled model s hould be useful for delineating modelling capabilities without the use of flux adjustments and should serve as a benchmark for future model improvements.