Eliassen and Palm (1961) has pointed out that two dimensional steady internal gravity waves do not accelerate the zonal mean flow if there is no external forces or dissipation, and Andrews and McIntyre (1978 a, b) have generalized this theorem. It is pointed out that the so-called “Eliassen-Palm relation” is the direct result from the conservations of energy and momentum of material layers by using the fact that energy and momentum fluxes through a material surface are connected by the phase velocity. From Lagrangian view point, it is shown that the difference between the Reynolds stress and the radiation stress results from the Lagrangian transversal flow due to the waves, and the“Eliassen-Palm relation” means in general that steady waves do not accelerate the mean flow in Lagrangian meaning. In cases of three-dimensional rotational flow and quasi-geostrophic flow, this fact explains the meaning of Uryu's results (1973) and Charney and Drazin's results (1961), respectively. We can also obtain the mean flow acceleration when there are external forces and dissipation from the Lagrangian view point. The theorem proved by Eliassen and Palm (1960) and generalized by Andrews and McIntyre (1976a) and also by Boyd (1976), is discussed from a somewhat different viewpoint from that so far.
The broad-scale fluctuations of cloudiness over the Eastern Hemisphere during the northern summer monsoon were investigated by using daily satellite mosaic pictures taken from June 1 to September 30, 1973. Spectral analysis revealed two dominant periodicities, of around 40 days and around 15 days. Cross-spectral, time-sectional, time-lag correlation and phase-lag vector analysis were applied to reveal the characteristics of these two modes in the time-space field. The fluctuation of 40-day period shows marked northward movement of cloudiness from the equatorial zone to the mid-latitudes (around 30°E) over the whole Asian monsoon area, and southward movement over Africa and the central Pacific. The northward movement is most apparent over the India-Indian Ocean sector. The fluctuation of this mode is associated with the major “active”-“break” cycle of the monsoon over the whole Asian monsoon area. The fluctuation of 15-day period shows similar features to that of 40-day period, but includes two clockwise rotations, one over India and Southeast Asia and the other over the western Pacific. A southward movement from the equatorial zone to the Southern Hemisphere middle latitudes is also prominent to the east and west of Australia. The fluctuation of this mode seems to correspond with the movements of equatorial, monsoon (or tropical), and westerly disturbances. It is also suggested that the fluctuation of 40-day period may be closely connected with the global-scale zonal oscillation in the equatorial zone and that of 15-day period may exist as a result of meridional wave interactions.
The O3 density and temperature in the Earth's paleoatmosphere are investigated using a numerical model for the photochemical-radiative equilibrium. The model atmospheres resemble the present atmosphere in most respects but contain different amounts of O2, ranging from 10-6 to 10 times the present atmospheric level (PAL). It is assumed that the photochemical processes take place in a pure oxygen atmosphere, and that there is no effect of transport either of composition or of heat by atmospheric motions. The results show that the O3 column density reaches a maximum for an O2 level of about 10-1 PAL, and that the effect of temperature feedback on the O3 column density is relatively small. It is also found that for O2 levels less than 10-3 PAL, the temperature decreases almost monotonically with altitude, i.e., neither the tropopause* nor stratopause* is found. As the O2 level increases, the temperature above 30km increases notably and the tropopause and stratopause appear clearly. The temperature change near the surface with changing O2 level is relatively small compared with that in the upper region. The contribution to the heat budget above 20km by the solar heating and terrestrial cooling due to O3 is found to increase remarkably with increasing O2 level, while below 15km, the contribution by H2O always predominates as in the present atmosphere.
An occurrence of a gravity wind was found on a snow patch of smaller than 1.2km in length, on a warm summer day. The fundamental properties of this wind obtained from the observation are as follows. The maximum wind speed (Umax) was 1.5 to 4.0m/sec., and the height of the maximum wind speed (Zmax) was between 0.7 to 1.2m above snow surface. The Zmax and Umax had a linear relation. The wind speed of the gravity wind at a certain height, which can be considered to be proportionate to Umax was stronger when the difference of the air temperature between the free air and the cooled air layer above the snow surface was larger. The wind speed of the gravity wind was stronger at lower parts of the snow patch, having a relation of U∞√l where U is the wind speed and l is the distance from the upper end of the snow patch. The occurrence of the gravity wind was controlled by the strength of the general wind in the valley. It occurred only when the general wind was weak. The vertical profile of the observed wind speed can not be explained by the classical theoretical profile obtained by Prandtl (1952), but can be explained by the empirical profile obtained by Martin (1975).
In order to derive simple relations expressing the vertical profiles of turbulence quantities such as the rms values of vertical speed δw and temperature δT, the vertical diffusion coefficient for momentum KM, and the dissipation rate of kinetic energy ε in the planetary boundary layer (PBL), the similarity hypothesis, which was attributed to Monin and Obukhov (1954) and verified in the constant flux layer (CFL), is extended. Unlike the CFL, the horizontally homogeneous PBL is, in general, non-stationary. However, here, approximately stationary conditions are dealt with by taking sampling times for the turbulence quantities which are relatively short when compared with the time change of the conditions. As in the case of the CFL, the fundamental quantities which determine the structure of the PBL are the vertical flux of sensible heat q, and momentum u*, and scale of turbulence l. However, in the PBL, these differ from their CFL behavior q and u* vary with height z, and l is not linearly proportional to z. In the present PBL model, the height variations of q, u*, and l are expressed by suitable empirical relations and the layer thickness is defined by the height at which the extrapolated values of q and u* tend to zero. The vertical profiles of the turbulence quantities in the PBL are derived by replacing q0, u*0, z (suffix o denotes surface value) in the relations for the CFL by q, u* and l. The derived relations are compared with observation and the characteristics of the scale l are considered.
During a period of observation of snow crystals at Inuvik (68°22'N, 133°42'W), N.W.T., Canada, rimed snow crystals, snow crystals with small frozen rain drops and ice pellets were observed for a period of several days. It was found that even when the diameters of frozen cloud droplets were larger than 40μ, sometimes a considerable number of them were single crystals. The average size of polycrystalline frozen cloud droplets was larger by 20 to 30% than that of the single crystal frozen cloud droplets. Whenever the polycrystalline frozen cloud droplets prevailed there were water saturation layers aloft with air temperature below -15°C. Although the shape of snow crystal was spatial dendritic type, 80 to 90% of frozen cloud droplets were single crystal with the same axis to substrate snow crystal. The spatial branches were found to grow from the part where the number of polycrystalline and single crystal droplets with different axis to substrate snow crystals were large. Small frozen rain drops and snow crystals with small frozen rain drops of 200μm to 300μm in diameter were observed. Although the frozen cloud droplets on the surface of snow crystals were single crystal with 80% or more with the same axis to substrate snow crystal, 60% or more of coexisting frozen rain drops on the surface of the snow crystals and free falling frozen rain drops were polycrystalline. Mean diameter of the polycrystalline frozen rain drops was larger by 20% to 30% than that of single crystal frozen rain drops. This tendency was similar to that of frozen cloud droplets.