EXPERIMENTS AIMING AT MONTHLY AND SEASONAL NUMERICAL WEATHER PREDICTION¹

Abstract

The author's thermal model for monthly and seasonal numerical prediction of temperatures is generalized, so that besides radiation other forms of heating (or the anomalies of 'heating) are generated within the model. This is done by expressing such heating a s a linear function of variables predicted in the model. The anomalies directly incorporated are those in the storages of thermal energy which are introduced by prescribing in the previous interval the temperature of the surface water in the oceans and the temperature of the midtroposphere, a s well as the anomalies in the shortwave radiation absorbed by the surface; which in turn are introduced by prescribing the albedo (snow cover) a t the end of the previous interval. The numerical experiments show that important anomalies of the evaporation a t the surface, of the vertical turbulent transport of sensible heat from the surface, of the condensation of water vapor in the clouds, and of the cloudiness are introduced by the anomalies of the computed temperature fields. Furthermore, these induced anomalies of the heating functions and of the cloudiness in turn introduce changes in the anomalies of the temperature fields. 1 This work was carried out ill. the Extended Forecast Division, National Meteorological Center, U.S. Weather Bureau, where the author is a consultant.

Simulation of the surface temperature anomalies in the Northern Hemisphere during the last 300 years of the Little Ice Age using a thermodynamic model

“…In the experiments reported here, we use the thermodynamic climate model (TCM), which is described in detail in previous works (Adem 1964, 1965, 1970, 1979, 1991, Adem et al 2000. The TCM consists of an atmospheric layer of about 10 km thickness, which includes a single variable horizontal cloud layer, an oceanic mixed layer of 60 m in depth and a continental layer of negligible depth.…”

“…The surface temperature perturbation (T ′ s ) on the oceans and continents can be expressed as a function of the mid-atmospheric temperature perturbation (T ′ m ), the cloud cover (ε) and the surface albedo (α) through the following linear algebraic equation: (2) where T ′ sp is the surface temperature perturbation in the previous month and F ′ 8 , F 8 , F 9 , F 10 and F 11 are known functions of the map coordinates, which are given by Adem (1965). F ′ 8 is zero in the continents and different from zero in the oceans, so that T ′ s depends on T ′ sp in oceanic regions.…”

“…Moreover, we suppose that the clouds absorb and emit longwave radiation as a black body, as is true for the Earth's surface. The changes in cloud cover, internally computed, are a linear function of the changes of the heat released by condensation of water vapour, which is a linear function of the computed changes in midtropospheric temperature (Adem 1965). An initial increase of temperature produces a decrement in the cloud cover, which favours an additional heating of the troposphere and the surface; the opposite occurs with an initial decrement of temperature (i.e.…”

“…The factor 0.693 in Eq. (4) is the result of the superposition of the normal cover of high clouds (13.5%) and middle clouds (19.9%) The cloud cover changes, ε DN , are calculated as determined by Clapp et al (1965) and incorporated to the TCM (Adem 1965). These changes are given by the following multiple regression equation: (5) where d 2 is a constant, T mN is the normal value of T m , and b′, d ′′ and c ′′ are the correlation coefficients, which depend on the space coordinates x and y and the season.…”

“…where D 2 is an empirical constant and G 5DN is the change or anomaly of the latent heat released by condensation of water vapour computed by the model (Clapp et al 1965, Adem 1965.…”