Grid systems in spherical coodinates are proposed and a few tests of them are carried out, using a barotropic fluid with a free surface on the finite difference schemes presserving total energy, total absolute angular momentum and total mass. Although grid points of spherical grid system A tested in this paper are skipped near the pole, it is computationally stable enough to integrate for a long run, and the computational noises are not so serious. Also the grid is not so inferior to the Kurihara-type grid of very high-resolution, so it is useful for numerical experiments of large-scale motions of the atmosphere.
From the samples used for the statistical survey on the formation of the intermediate scale disturbance appearing near Japan and the vicinity during five years (Nitta and Yamamoto, 1973), seven examples are taken up for the case study about the synoptic structure of the disturbance. By means of a simple method, the deviations of the geopotential and temperature fields are analyzed and investigated. The longitude-height and the latitude-height cross sections of the deviations are made across the cyclone center. According to these cross sections, we may conventionally classify the intermediate scale disturbance observed near Japan and the vicinity into four types as follows; Type I: The warm air is in the east side of the cyclone and the cold air in the west side. The trough axis shows westward tilting with altitude in the lower troposphere. The concentrative warm core is observed in the mid-troposphere above the cyclone center. Type II: The synoptic structure in the lower troposphere looks like that of type I, but the warm air in the mid-troposphere is broad and vague. Type III: Similar to types I and II about the vertical structure in the lower troposphere, but the cold air occupies the mid-troposphere above the cyclone center. Type IV: Contrary to types I, II, and III, the cold air exists in the east side of the cyclone and the warm air in the west side. The trough axis shows trend to tilt eastward with altitude. Cold air is situated in the mid-troposphere above the cyclone center. Especially, the difference between type IV and other three types is remarkable. Formation and evolution of the intermediate scale disturbance are also illustrated in pressure pattern.
An atmospheric turbulence in stable stratification was observed at Syowa Station in Antarctica during the period from February to December in 1970. Its characteristics are analyzed in relation to the stability (Richardson number, Ri) in the surface air layer. The vertical mean wind speed at the height of 20m is observed to be about one-third of the longitudinal mean wind speed at the same height above the slope with an inclination of about 10 degrees. The standard deviations of the longitudinal, lateral and vertical wind speed fluctuations decrease exponentially with the increase of Ri. The ratios of the standard deviations of the longitudinal, lateral and vertical wind speed fluctuations to the friction velocity are independent of Ri. The energy dissipation rate increases roughly in proportion to the 0.5 power of Ri. The scale of the atmospheric turbulence seems to be a solid body of a long and narrow cubic vortex, i.e., length:breadth:height=6.3:1.9:1. The characteristic scale of the atmospheric turbulence which is the largest turbulon is independent of the stability of Ri. Furthermore the scale of the smallest turbulon increases exponentially with the value of Ri. The nondimensional frequency at the maximum spectral density of the longitudinal wind speed increases with the value of Ri. On the other hand, however, the peak frequency of the lateral wind speed is independent of Ri, and that of the vertical wind speed decreases with the increase of Ri.
The flat size distributions of riandrops were observed in heavy rainfalls during the last stages of Baiu at Hitoyoshi in southern Kyushu. Most flat distributions are fairly frequently seen especially in the beginning stage of shower rain, but such flat shape remained as to appear in the medium size range of the average distribution of each individual convective rainfall. Generation processes of those characteristic spectra were investigated by numerical experiments. Cloud models were considered with some assumptions from the results of the radar observations which were simultaneously made in the raindrop sampling. Intermittent supply of liquid water with a certain drop size spectra, which was assumed to be corresponding to the observation fact of combination of cell echoes by radar, was introduced in the well mixed cloud model by the process of drop disintegration and coalescence. The results represent well the average size distribution as observed in each rainfall at Hitoyoshi, which is characterized by the numerous space density of drops smaller than 1mm diameter, the high population of large drops, and the flat shape of the distribution curve in medium size range. Furthermore, the very flat distribution was obtained by the one dimensional sedimentation model in which the rain with the abovedescribed average distribution was released at the upper level. The Z-R relation derived from the computational results may be well represented by Z=360R1.6 which was observed at Hitoyoshi.
The vertical distribution of the heating rate due to absorption of the visible solar radiation by aerosols is studied by solving the equation of radiative transfer in turbid atmospheres. Realistic model atmospheres with respect to size and vertical distributions of aerosols as well as surface reflectivity are assumed, and latitude-altitude variation of the heating rate is calculated as a function of atmospheric turbidity and complex refractive index of aerosols. It is shown that the heating rate increases in the lower atmosphere of the low latitude region, and becomes comparable to the heating rate due to absorption of near infrared solar radiation by gaseous constituents there. It is shown, however, that the heating rate is greatly affected by the imaginary part of the refractive index of aerosols. A possivility of determing the imaginary part from field observations is suggested.
Comparison is made between hydrostatically calculated and directly measured heights which were obtained by echo-sonde observations There exists a large systematic height difference which has the magnitude of order of 200m in the troposphere. The fact that such difference differs station to station seems to suggest that directly measured heights contain systematic error, as pointed out by Nyui and Matsuhashi (1967). Next, characteristic features of meteorological conditions are examined in the cases where anomalous height difference was observed. It is shown that it occurs in the dry region of low Richardson number. This indicates that anomalous height difference may be associated with a wave-like disturbance in the clear atmosphere revealed by recent radar observations.
Recently Maki (1974) reported the observations of the atmospheric turbulence above the sloping surface at Syowa Station in Antarctica in 1970 by means of an ultra sonic anemometer which was mounted at the 20m level of a meteorological tower (21m height). He shows that the turbulent intensities (σu/u*, σν/u* and σω/u*) are almost independent of Ri in stable conditions, and their averaged values are 1.80, 1.55 and 3.00, respectively, where σu, σν and σω are the standard deviations of the wind fluctuation in longitudinal, lateral and vertical directions, respectively, and u* is the friction velocity. And he also shows that the mean wind vector has an inclination of 18°16' (tan-1 0.330) from the horizontal plane, (the broken line of Fig. 2 in this paper.) The present author has made the observations of wind and temperature fluctuations at the same tower of the same place in 1971. The results of the observation do not always coincide with Maki's results. The following are the summary of the present results (Adachi, 1973).
The forcing of the second-order zonally-averaged wind and temperature fields due to the quasiresonant growth of planetary waves in a two-layer model is considered. Dissipation in the form of linear drag and Newtonian cooling is allowed for. Calculations are made in a mid-latitude β-plane channel bounded by two vertical walls. The linear or first order solution is constrained so as to not transport momentum meridionally. Three neutral modes in the adiabatic two layer model are brought to resonance separately by adjusting the basic zonal flow configuration. For a dissipation time scale of 2 days, the quasi-resonant enhancement of the response is negligible. A time scale of 4 days is used. For resonance involving the two layer equivalent of the Rossby wave, the strong westerly zonal flow configuration causes most of the second order effects to be confined to the lowest few scale heights, where a slow easterly acceleration of the zonal flow occurs. A second resonance associated with weak westerlies surmounted by moderate westerlies leads to a very rapid high level easterly acceleration lasting about two weeks. A final resonant wind configuration with easterlies above westerlies is associated with very rapid temperature rises near the pole. Calculations suggest that the quasi-resonant enhancement of the planetary wave-zone flow interaction resulting from certain basic flow configurations might be the explanation of the stratospheric warming. It is hypothesized that the timing of warmings is determined by how well tuned the zonal flow structure is in the sense that appreciable secondorder effects will result in response to small influxes of planetary-wave energy from below. Furthermore, it is suggested that the subsequent evolution of a warming into a full scale warming depends on the ability of the zonal flow configuration to evolve through a series of tuned or resonant states.