Abstract
By means of adequate parameterizations, the advection of heat by the mean wind and that by ocean currents are incorporated in a thermodynamic model for long-range weather prediction. Numerical experiments with the revised model for a single case (January 1968) show that nonnegligible anomalies of wind and ocean currents are generated. These introduce important changes in the predicted surface temperature and in the 700-mb temperature anomalies. Furthermore, the predicted monthly anomalies of temperature are in good agreement with the observations and are better than those obtained when advection by mean wind is neglected. An evaluation of the predictions by the model of the anomalies of the mean monthly surface air temperature for the whole year 1969 over the conterminous United States is presented; and it is shown that, for this period, the skill improves considerably when advection by the mean wind is included. Normal temperatures are computed using normal observed geostrophic mean winds in the midtroposphere and climatological seasonal values of ocean currents. For January, the effect of introducing advection by the prescribed observed normal geostrophic wind is to move toward the east the midcontinental troughs and midoceanic ridges that were obtained without advection. The resultant temperature distribution for the Northern Hemisphere at 700 mb is in remarkably good agreement with the observed values.