Zonally symmetric global general circulation models with and without the hydrologic cycle

Abstract

A zonally symmetric general circulation model of the Earth's atmosphere has been developed based on the “box” finite difference formulation of Kurihara & Holloway (1967). The finite difference grid consists of 40 meridional points with 18 vertical levels extending up to the lower stratosphere; the model is bounded by “walls” at each pole. Realistic radiation and diffusion processes are included, but no topography. Two versions of the model were explored, one dry and one containing the hydrologic cycle, each being integrated for several hundred days for annual mean conditions. The calculated temperature and zonal wind distributions were very similar for both models. The temperatures were reasonably similar to observation, but the high latitude troposphere was about 30°K too cold, while the tropical tropopause was 20°K too warm. The zonal wind distribution, although consistent with the temperature distribution, was rather surprising. It consisted of an equatorial jet with a maximum west wind of over 110 m/sec. in the stratosphere, with surface easterlies at high latitudes and westerlies at low latitudes. To a first approximation this peculiar zonal wind distribution can be attributed to the omission of the large scale eddy flux of heat from the model, as the resulting temperature distribution constrains the possible zonal wind distribution. A detailed discussion is given concerning how the model maintains this wind distribution. The cellular structures of the dry and moist models were strikingly different. The dry model had an equator to pole Hadley cell in the troposphere, but in the region of the tropical tropopause a reverse cell existed. The moist model unexpectedly had Hadley, Ferrel and polar cells, and, in addition, a fairly intense indirect cell in the tropical troposphere. By comparing the results of the zonally symmetric model with other three-dimensional calculations a more precise evaluation is possible of the role of large scale eddies in the maintenance of the actual general circulation. In addition, since the results obtained in this study were noticeably different to those generally hypothesized in the literature for this type of model, it would appear to be of some value to document these results in order to rectify this situation. DOI: 10.1111/j.2153-3490.1973.tb00618.x