Investigated in this article is the dynamical response of a bounded barotropic ocean on the β-plane to zonally moving atmospheric disturbances whose characteristic time-scale is about one week and characteristic horizontal scale is several thousand kilometers. The solutions of the linearized vorticity equation are obtained for several cases with and without incorporation of physical processes such as the bottom friction, the horizontal diffusion, and the horizontal mass divergence. The result may be summarized as follows: (1) When the ratio of the atmospheric traveling velocity to the critical velocity exceeds unity, the mode which travels with the speed of atmospheric disturbances is dominant, as already indicated by Pedlosky (1965). The critical velocity is deter-mined by the latitudinal and longitudinal wave numbers of atmospheric disturbances. (2) When the ratio is unity, the flow patterns show the behaviour of standing wave. (3) When the ratio is smaller than unity, the reflection from the eastern and western boundaries is observed in the open sea. (4) When the ratio is negative, the atmospheric disturbances may excite the resonance on the oceanic motion. The effects of the bottom friction and the horizontal diffusion are shown to be minor over the range of time-scale considered here. The existence of the horizontal mass divergence dicreases the critical speed but hardly affects the amplitude of meridional velocity.
Upper-wind data over the Line Islands (mainly Christmas Island, N02°00', W157°23') during the two years 1957-58 were analyzed by spectral methods to study the behavior of disturbances in the troposphere and the lower stratosphere near the equator. Frequency-height diagrams of power spectral density were shown in time sequence. We found three kinds of disturbances: 1. Lower tropospheric disturbances, 2. Upper tropospheric disturbances, and 3. Lower stratospheric disturbances. The lower tropospheric disturbances and the lower stratospheric disturbances appear markedly in the meridional wind component with periods of about 4-5 days. The phase of disturbances slightly lags with height in the lower troposhpere and leads with height in the lower stratosphere. We suggest, from the phase difference between the zonal and meridional wind components, that these disturbances take the form of eddies centered on the equator. The upper tropospheric disturbances appear both in the zonal and meridional wind components with periods of about 6-12 days. The phase of disturbances is almost constant with height. The horizontal structure of these disturbances has not yet been clarified. We did not find any marked annual cycle in the behavior of disturbances.
A method for analysis of the ultra-long waves in the atmosphere, which are represented by Fourier harmonics of the geopotential field etc., is presented. Firstly, the ultra-long waves are separated into the quasi-steady and the fluctuating parts by applying several band-pass and one low-pass time filters to the time series of Fourier harmonics. Secondly, the fluctuating parts, which pass through the band-pass filters, are devided into the transient part and the quasi-stationary part with temporal change of the amplitude, with the procedure of shifting the cosine and sine-time series by a quarter of the period. This method is applied to the geopotential field, and some preliminary results are given as an example.
The author deals herewith, by the linear theory, the large-scale or even ultra large-scale steady topographical perturbation of a two-dimensional zonal current with friction in the rotating system for the rotating dishpan model experiment and the atmospheric prototype under the different conditions of the upper boundary, free or bounded, of the lateral boundary, periodic or infinite-distantly vanishing, and of the β-term, included or not included. Two types of analytical solution are obtained, periodic and exponentially damping, according to the periodic and infinite-distantly vanishing lateral boundary conditions. Using these two types of analytical solution, following three main results have been derived. Firstly, under the upper free boundary condition without theβ-term, the frcitional effect upon the large-scale steady topographical perturbation of a zonal current is not so much of amount as pointed out previously by several authors. This treatment is the most fitted for the topographical perturbation of this scale of the atmospheric prototype. Secondly, the result of customary use of constant height or upper-bounded treatment for this kind of perturbation is much deviated from that of the upper free one mentioned above and consequently seems to be unreasonable for the prototype. The frictional effect becomes large in this case. Thirdly, within the framework of the present study the two-dimensional topographical perturbation of the prototype is not appropriate to be treated on the β-plane since the behaviors of perturbation become quite fictitious in this case. The frcitional effect is also very large.
Detailed analyses of radar echoes, betanetwork data and upper-air soundings recorded on 3 April 1964 are made to investigate possible mechanisms of the movement of severe thunderstorms. One of the storms investigated moved significantly to the left of the mean wind while others moved in the direction of or moderately to the right of the mean wind. The left-moving storm was found to generally display a mirror image of the radar echo intensity structure, growth, and circulation character of right-moving storms. A similar relationship is found in the surface pressure and wind fields. Analyses of a diverging storm pair revealed that the left-moving storm echo indicated the presence of continuous echo growth and development along the left flank of a quasi-steady supercell. An area of low surface pressure, and converging surface wind occurred along the left-front quadrant beneath a low-level inflowing current. The right-moving storms were found to contain similar pressure and wind features along the right-rear flank. The analytical results give reasonable evidence that continuous propagation was occurring on opposing flanks of the storm pair. The continuous propagation mechanism, incorporated in a simple, qualitative model taking into account low-level momentum conservation, drag, and lift is demonstrated to be important in order to explain the observed movement of these storms.