In Part II, the normal distributions of various quantities related to the radiation budgets of a tropospheric column, the earth's surface, and the troposphere-earth system are presented over the northern hemisphere for January and July. In general, the distribution of radiation quantities has mainly zonal characteristics and intense north-south gradients in January, but has mainly cellular characteristics in July closely corresponding to the distribution of sea-land or the cloud cover. The annual mean value of planetary albedo for the troposphere-earth system averaged over the northern hemisphere is 0.374, which is slightly larger than the 0.34 of Houghton but smaller than the 0.40 adopted by Budyko. Both spatial and seasonal variations of the radiation budget of the troposphere, which contribute directly to the diabatic heating field in the troposphere, are relatively small compared with those of the radiation budgets of the atmosphere-earth system and of the earth's surface.
In part III, zonal cross-sections of the heating field by radiative processes are presentedfor each 10 degree latitude in January and July. The role of radiational processes in the energy cycle of the atmospheric general circulation is also discussed. The zonal cross-section of the heating field by net radiative processes shows two significant maximum cooling layers : one is in the middle troposphere between 400 and 500mb and the other is in the lower troposphere. As for the middle tropospheric cooling layer, the height of this layer shows little change along the latitude circle, while it becomes lower from south to north and also from July to January. Its cooling rate is, in general, stronger over the ocean than over land. This maximum cooling layer is caused by the existence of middle clouds, nimbostratus or cumulonimbus. Another significant feature is that an isolated maximum cooling layer is found over high mountain regions. The radiation budget of the troposphere seems to act to cause the destruction of the zonal available potential energy the year round. However, the magnitude of the energy destruction is considerably smaller than the generation of that energy by the condensation heating. This result clearly suggests that the direct contribution by the radiative processes to the atmospheric general circulation is not so significant but that the indirect contribution from the earth's surface is of prime importance.
The inside air temperature variations of the glasshouse during the nighttime are discussed. The weighting functions of the glasshouse are derived from the responses to the unit step functions of outside air temperatures. Having linearized the effective radiation heat exchange, the response of the glasshouse to the outside air temperature and to the effective radiation are calculated. Then, the inside air temperature is obtained by making use of Duhamel's integral. Nighttime observations in the model experiment do not conform to the theoretical calculation. The departures are not surprising in view of the after-effect of the solar radiation which is not taken into account in the present study. The same method of analysis is applied to the examination of the cold protection of the glasshouses whose constructions are the same as that of the model in the present analysis. The following conclusions can be deduced from these results, that is, the phase lag of the inside air temperature behind the outside temperature increases with the increase of β (the ratio of floor area to total glass surface area) and the decrease of μ (the ratio of air volume of the glasshouse to the floor area), and the amplitude of the inside air temperature variation decreases with the increase of β and the decrease of μ. The effect of net radiation on the inside air temperature is shown not only through the parameter β but also through the view factor of the glass surface to the sky. It is also shown that in the present model, the air temperature observed in the glasshouse is lower than the out door temperature during the nighttime.
Basing on the aerological observations by smaller scale network which were set up in January of 1963, 1964 and 1965 in Hokuriku district, the Japan Sea coastal area of central Japan, the heat and moisture budgets were compared among these three years. The flux divergence of vapor assumes nearly the same values for three winters, whereas the amount of precipitation changes very much from year to year. Although the difference in the evaporation from the sea surface and that in the convective transfer are estimated to be of considerable amount, the precipitation is principally related to the net transport of condensed water either from or to the surrounding region. The vapor import in 6 hours is compared with 6 hourly precipitation within the region. Better relation is found in the precipitation onto the downstream side stations. It is shown that the sensible heat increment is nearly twice as much as the latent heat decrement if they are computed by mean flow flux divergence. This circumstance is observed well in the cloud layer regardless to the scale of network. The surplus of the heat energy must be transported by convective activity. It is suggested that, when heavier snowfall is observed, the convective activity is so predominant that the more heat energy than that supplied from the sea surface is transported into the cloud layer.
A family of remarkable mesoscale disturbances was observed on the western side of the cold vortex of smaller scale which had passed the central Japan on January 16, 1965. The cold air is bounded by a steep inversion layer and the mesoscale undulation is superposed on its western slope. By the special aerological observation network whose covering area is about one fifth of the routine network, a characteristic wind field is revealed. It is convergence and cyclonic vorticity in the lower part and divergence and anticyclonic vorticity in the upper part of the cold air and it implies a mesoscale convective system. Correspondingly lower level depression and upper level high are found in the pressure field. A series of disturbances of a 2-3 hour period which moved eastward with a velocity of 85 km/hr and was closely related to the precipitation was revealed in the surface pressure field. Detailed quantitative analyses were made on the pressure and wind anomaly field. The analyses were obtained by applying the 2.5hr running mean in order to pick up the disturbances mentioned above. The scale of disturbance was found to be of the order of 250×108m2. The convergence area followed the pressure depression with positive vorticity area to the south and negative vorticity field to the north. Thus the principal balance of the vorticity equation was established between the vorticity change and the twisting term. While the balance of the divergence equation was approximately obtained between the divergence change and the pressure field as was suggested in our previous paper. However a discrepancy was observed mainly in the convergent area. This fact is well explained by introducing the effect of cumulus convection activity in that area.
It is shown by the observational data and the numerical experiments, that the local strong wind appears under the lee of a mountain when air crosses over the mountain, being cooled from the earth's surface. It is also shown by the numerical experiment that strong wind appears under the lee when cold air passes over the mountain, but generally it is not so in the unstable layer, and a crowd of convection cells is produced in the unstable layer originated with heating from the ground surface in the daytime. The character of such crowd of convection cells as the gravity waves is discussed.
While Tsuchiya and Fujita (1966) were investigating mesoscale features of winter monsoon clouds over Japan by using satellite photographs and radiation data, a mesoscale upper-air disturbance was found to exist over the region of monsoon snowfall on the windward side of Hokkaido and northern Honshu as they block the strong northwesterly monsoon. A relatively dense upper-air network with an average spacing of 200km (120miles) permitted us to analyze mesosynoptically the vertical structure of the upper-air mesosystem. The system was characterized by a cold dome directly above which warm dry air descended from an eroded tropopause. Analyses of 1963 and 1964 data revealed that the upper-air mesosystem stays more or less over the regions of monsoon snowfall while high-level winds blow through it within a few hours. To explain the formation and development of this upper-air meso- system, a mechanism of mesoscale jet stream called “snowstorm mesojet ” has been proposed.
Introducing the characteristic scales of thermal convection, 1/k and 1/m, in the horizontal and vertical directions respectively, the time evolution of total kinetic energy of the system and the life time of thermal convection are qualitatively discussed in a dry air case. According to the linear theory of thermal convection, the disturbance in the unstable atmosphere develops exponentially with time. In this paper, it is qualitatively shown that the exponential amplification of amplitude of unstable disturbance with time is suppressed by the modification of mean static stability due to the upward eddy transport of entropy. In this process, the second-order effect on the mean static stability due to the presence of unstable disturbance is emphasized. The characteristic life time of thermal convection is roughly estimated and it is proportional to (1+m2/k2)1/2. In order to examine the qualitative discussion mentioned above, the numerical experiment of thermal convection in a dry air case is performed. The results show in good agreement with those obtained by the qualitative discussion.
In order to confirm the increase of natural ice nucleus count around the 28th day after a meteor shower, measurements of natural ice nuclei were made with an improved filter paper technique at sufficiently separated two or more sites. The measurements were made continuously for the two periods of more than ten days in May and November, 1962 and January-February, 1963. It was found that the concentration of ice nuclei increased at each site around the same day, that is, the 28th day after major meteor showers. Other increase by local source was seldom found during the periods. From 1960 to 1964, the time variation of all the concentration of measured ice nuclei during the several days around the 28th day after a meteor shower was examined by a mixing cloud chamber or filter paper technique. Almost always the ice nuclei showed unusual increase during the 27th to 29th day after the meteor shower. From the statistical examination, it was recognized that the increment of precipitation amount occurs for the period of the ice nucleus increase following the major meteor shower.