The turbulent exchange coefficient is assumed to be given as a linear combination of the exchange coefficient due to forced and that due to free convection. This assumption is in contrast with the assumption of the quadratic combination of the exchange coefficients which, according to Sellers, leads to the equation for the wind profile derived by Ellison, Yamamoto, and Panofsky et al. Then, a correction factor is introduced which makes the role of free convection in stable conditions less weighted than that in unstable conditions. Based on these assumptions, new formulae for wind and temperature profiles in diabatic conditions are derived. Under unstable conditions they are similar to those derived by above workers. Under stable conditions, by determining the values of parameters involved in the new formulae empirically, it is possible to make the calculated profiles agree with the observed ones.
The observation at the coast of and over the Japan Sea revealed as follows; (a) the amounts of sodium and chlorine in graupel pellets were several times as large as those in snow flakes, and the ratio of Na to Cl in the graupel was very close to that in the sea water, (b) a number of supercooled large cloud drops (several tens of microns in diameter) were present in clouds over the sea near the coast, where graupel showers occurred very often, (c) there was a positive correlation between the daily amount of snowfall and the concentration of giant chloride particles in the air, (d) the relation between NH4+ content and Na+ content of snow shower from convective cells formed over the sea was different from that of snow shower from orographic cells at an inland area. It may be concluded that these results provide the evidence for the importance of giant sea salt particles in the formation of snowfall from convective cells over the sea. The role of the giant salt particles was probably to enhance the growth of rimed snow crystals. Owing to the large relative fall velocity of the rimed snow crystals, they collected a number of ice crystals and formed snow flakes. Thus, a kind of colloidal stability of ice clouds where all ice crystals had nearly the same slow fall velocity could be destroyed and the substantial snowfall would be formed.
The present paper describes a part of the results of analysis of an intense upper cyclone, which developed over the North America. The thermal structure and the distribution of kinematical properties of the cyclone at its mature stage of development are discussed. The results are compared with analyses of upper cyclones by other investigators. Similar to many other cases, there was a funnel-like boundary separating upper stable and lower relatively unstable air. However, in this case, only a portion of this boundary may be regarded as tropopause. It is pointed out that the thermal structure, in the upper troposphere, of such a vortex is considerably influenced by its location with respect to the subtropical jetstream. We may consider that a complete funnel-like tropopause will be found in a W-E cross-section of an upper cyclone situated far north of the subtropical jetstream, whereas such a tropopause will not be found within an upper cyclone south of the subtropical jetstream. In the present case, the cold-dome boundary was not well defined near the vortex center and the north sector, over these regions there was no significant stable layer below tropopause. On the contrary, there were a number of inversions and stable layers to the south through the west of the center. Most of these are considered to be the subsidence inversions. Computations revealed a possibility of the reversal of vertical motion, due to very strong divergence in this region, within a thin layer near the funnel-like boundary. It is considered that this high-level strong divergence (or convergence) is related to formation (dissipation) and propagation of the tropopause. It is pointed out that the vertical motion field associated with the upper cyclone may not be described by simple models such as indirect circulation or direct circulation. Instead, actual vertical motion associated with a mature upper cyclone will be consistent with a fact that the system moves changing slightly its characteristic thermal structure, viz., a dome of cold air in the lower troposphere and a funnel-like warm air in the lower stratosphere. A simplified vertical circulation model for an eastward moving upper cyclone is given in Fig. 19. General appearance of vertical circulation in the W-E cross-section is similar to that obtained by Newton and Palmén (1963) for a very deep trough. Unlike to the earlier outoff cyclone models, a band of ascending motion was evident to the north of the center of the upper cyclone of the present case. It is considered that the difference occurs because the earlier models refer to the formation stage of a cutoff low, whereas the present paper deals with the upper cyclone nearly at the steady state.
Results of computations of vertical velocity and precipitation amounts for two winter time pre-frontal precipitation zones are presented. Since each extensive precipitation zone seemed to be consisted of a series of several meso-scale precipitation centers, computations are made to obtain averaged conditions for such meso-systems. It is found that there is a tendency that the level of maximum upward motion rises with time for such meso-scale precipitation systems of pre-frontal type. In one of the cases examined, the weakening of pre-frontal precipitation occurred as the system approached to the quasi-stationary anticyclone where large-scale low level subsidence was evident. In case of well developed precipitation systems, contribution of condensation above the 6 km (MSL), due to upward flux of water vapor through this level, may reach 1/4 of total precipitation amount. Some computations are made for a cold- and warm-frontal precipitation. The former showed similar characteristics with the pre-frontal type, but the result for the latter indicates the classical model of warm frontal rain is satisfactory for the interpretation of the computation.
To obtain a theoretical basis for the numerical simulation of tropical cyclones, the stability properties of symmetrical disturbances driven by the latent heat of condensation released in deep cumulus clouds under the existence of surface friction are discussed by solving a system of primitive thermohydrodynamic equations applied to a three-layer model including the friction layer. The processes of heat generation and transfer by cumulus clouds are incorporated by assuming that the total amount of latent heat released in any vertical column of air is proportional to the horizontal convergence in the friction layer. An analysis of linear perturbation equations shows that a typhoon-scale disturbancesimilar to those obtained by Ooyama (1963), Charney and Eliassen (1964) and Ogura (1964)using the balanced model is also found in the unbalanced (or primitive) model. Without surface friction the disturbance would be in geostrophic (or gradient wind) balance. Under some conditions unstable gravity waves as discussed by Haque (1952) and Syono (1953) still exist even in our model in spite of a different treatment of condensational heating. Moreover, under some conditions gravity oscillations are amplified with a large amplification rate and short period. These unstable gravity waves are found to be filtered out by use of the assumption that the pressure gradient force is balanced with the Coriolis force and the frictional force in the friction layer, as adopted by Ooyama (1963) and others. At the same time the use of this assumption modifies the growth rate of the typhoon-scale disturbance. Furthermore this assumption alters the condition under which the typhoon-scale disturbance exists. The analysis further shows that the vertical partition of the released latent heat as well as the amount of water vapor in the ascending air and the static stability of the atmosphere are important factors which determine properties of the disturbance.
Ice nucleus concentrations in Seoul were measured using millipore filters for the period of last four winters, 1962-1963, 1963-1964, 1964-1965 and 1965-1966. Concurrently, the microscopic dust count and the radioactivity of air borne dust caught on the filter surface were also examined. Taking into consideration the correlation among these factors, the sources of ice nuclei in Korea seem to be distant rather than local. The average value of ice nucleus concentrations in Seoul during the winter seasons was 0.12/l at -15°C, which is somewhat lower than the Australian average and also that of Japan.
The normal three-dimensional distributions of the heating fields by radiative processes in the troposphere over the northern hemisphere are presented for January and July. They are obtained from the data newly compiled for the present calculation by more reasonable procedures than those in the previous studies. The calculations of the long weve radiation are carried out numerically based on the procedure presented by Yamamoto. In the calculation of the solar radiation, absorption by water vapor, scattering by dry air and water vapor, depletion by atmospheric dust, absorption and reflection by clouds are taken into consideration. Despite many assumptions, the calculated total solar radiation at the earth's surface is in fairly good agreement with that estimated by Budyko mainly from its surface observations. 1. Introduction