In this study, global sea surface temperature anomalies within ±30° latitudes of the equator were correlated with the time series of the major rotated principal component analysis (RPCA) modes of the seasonal rainfall over East Africa (Kenya, Uganda and Tanzania) for the period 1950-79. Regionally averaged rainfall anomalies were also correlated with the SST anomalies. The physical reality and climatological stability of the computed correlations were investigated using 6° by 6° gridmesh SST records instead of the original 2° by 2° values. The stability of the patterns were further tested by random removal of a maximum of upto five pairs of the SST and rainfall records from the original data sets. The results from the study indicate significant instantaneous (zero lag) and time lagged correlations between SST anomalies over portions of the global oceans and some of the principal seasonal rainfall modes in East Africa. The maximum instantaneous correlations occur in the boreal autumn between SST anomalies in the Pacific Ocean and the autumn rainfall RPCA mode, which is dominant over the coastal regions. The spatial patterns of the significant correlations indicate a 'see-saw' pattern between the eastern Pacific Ocean and the Indonesia region which coincides with positive rainfall anomalies over the coastal regions of East Africa, and indicates a relationship between rainfall variability in this region and the El Nino/Southern Oscillation (ENSO) phenomena. Lower spatial and temporal persistence is observed between SST anomalies and the rainfall RPCA modes that dominate inland. The maximum variance of the seasonal rainfall that could be accounted for by the SST anomalies was about 40%.
The fine structure of super clusters was analyzed using the 3-hourly Geostationary Meteorological Satellite (GMS) Infrared (IR) data. A super cluster is defined as an ensemble of cloud clusters having different life stages with an observed horizontal scale of several thousand kilometers found near the equator during the active phase of intraseasonal variations (ISVs). Super clusters move eastward with a phase speed of about 10-15ms-1. GMS IR data reveal that the life time of each cloud cluster within a super cluster is about 1-2 days and its phase propagation is westward along the equator. Although each cloud cluster moves westward, a super cluster moves eastward due to the successive formation of a new cloud cluster east of the mature-stage cloud cluster. The above results suggest the existence of a hierarchy of convective activity in the tropical atmosphere. During the onset phase of the 1986/87 ENSO event in November 1986, an eastward moving super cluster was observed over the western Pacific to the dateline.
Snow crystal growth in free fall by vapor diffusion under water saturation and riming was studied up to 30 min in a supercooled cloud tunnel. The results showed that: 1) the growth rates were pronouncedly highest along a-axis at-15°C and along c-axis at-5.5°C. Each peak coincided with the growth of dendrites and needles, respectively; 2) polycrystalline snows, consisting of spatial dendrites and spatial plates, were observed very frequently below-15°C; 3) the variation of fall velocity with growth time could be expressed by two straight lines, except for the temperature zone of dendritic crystal growth, where a curve and straight lines were fitted; 4) the empirical relations between crystal fall velocity and growth time under constant temperatures were divided into two groups. In the higher fall velocity group, isometric crystals grew and were rimed; in the lower fall velocity group, shape enhancement occurred during continuing vapor diffusional growth. The crystals in the former group dropped from 700 to 800m, and those in the latter only 300 to 400m in 30 min; 5) isometric crystals grown at around-10°C changed into graupel. The growth mode was represented by three stages: vapor diffusional, transitional and graupel growths. The crystal mass was roughly proportional to t1/2, t2 and t6 in each corresponding stage. In the vapor diffusional and graupel growth stages, the particles followed Stokes' and Newton's laws of resistance, respectively.
Characteristic features of medium-scale waves (zonal wavenumbers 4-7) are investigated for the Southern and Northern Hemispheres by the use of two global data sets from the troposphere to stratosphere for the period July 1982 through April 1986. In the Southern Hemisphere, the medium-scale waves with a regularly spaced distribution about latitude circles are frequently observed throughout the year with relatively small zonal wavenumbers in the winter season. From various aspects obtained in this study such as the vertical wave structure and eastward migration, these are considered to be baroclinically unstable waves. On the other hand, in the Northern Hemisphere, the medium-scale waves with the zonally homogeneous property are observed only in the summer season. They are stationary and considered to be topographically unstable waves. In other seasons, medium-scale disturbances are longitudinally localized and do not have the zonally homogeneous nature.
Doppler velocity data obtained from a single Doppler radar are used to examine fine structures of gust fronts associated with severe thunderstorms. Fields of inflection points of velocities (on surfaces scanned by a radar) are analyzed and are compared with the fields of spectrum width of Doppler velocity and horizontal shear. The analysis is performed on the cases of May 9, 1981 and April 13, 1981 in Oklahoma. In the first case, a gust front consisted of an evolving undulate bore with three separate waves or discontinuities. Two of these waves are detected as a 60km stretch of inflection points of Doppler velocity, whereas the third is marginally evident. For both cases, the length representing secondary features behind the main discontinuities are short (less than 20km) and transient. Careful examinations of the structures at low altitude show that large spectrum width and large shear areas exist slightly behind the main line of inflection points. However, there is a multitude of areas with large shear and spectrum width that have no corresponding inflection point lines. Thus, these two parameters, spectrum width and horizontal shear, alone are not suitable for detection of gust fronts. Besides showing the presence of gust fronts, inflection points can also be good indicators of atmospheric waves. We discuss a pattern of such waves that are embedded in the environmental flow ahead of a gust front.
Characteristics of the equatorial intraseasonal oscillation are studied with the use of a general circulation model which includes the Arakawa-Schubert (abbreviated as AS hereafter) model of penetrative cumulus convections. The AS model is modified by introducing the minimum value of cumulus entrainment rate of the environmental air, μmin, as μmin=α/D, where D is the depth of the planetary boundary layer (PBL) and a is a non-negative constant. The introduction of a positive α in the AS model suppresses the activity of deep penetrative cumulus in the area where D is not sufficiently thick, which allows, in turn, an accumulation of moist air in the large-scale low level convergence zone. This process is essential in maintaining the equatorial 30-60 day oscillations, and also in simulating the Pacific subtropical high during the northern summer. Experiments are performed by changing α from 0 (ie., the original AS model), to ∞ (i.e., no penetrative cumulus convection) under an aqua-planet condition. When α=0, the 30-60 day oscillation does not appear in the tropics. Instead, there exists a quasi -10 day eastward propagating oscillation with zonal wavenumber 1, which resembles a neutral Kelvin wave. Moist air is not accumulated in the 'low level east-west convergence longitude associated with the flow of zonal wavenumber 1 (LLCL)' due to the rapid response of the AS model to the evaporation and moisture flux convergence by small scale motions and also due to the resulting upward transport of water vapor by penetrative cumuli to the west of the LLCL before the moist air can be accumulated in the LLCL. When α=0.1, a quasi-30 day eastward propagating oscillation with zonal wavenumber 1 grows in the model, with the moist air and the major heating found around the LLCL. The change in the heating is mostly due to the increase of middle-level convection (i.e., moist convection between adjacent vertical layers within the free atmosphere) and to the decrease of deep penetrative cumuli to the west of the LLCL. Overall characteristics of the mode are close to the observed ones. When α=∞, a quasi-45 day eastward propagating oscillation grows in the model. The structure of the heating associated with the oscillation is similar to that of the quasi-30 day oscillation. Associated with the increase of α, the static stability decreases in the low latitudes. The maximum level of the zonally averaged heating lowers due to the suppression of deep penetrative cumulus and the increase in both the middle-level convection and large-scale condensation. The maximum amplitude level of the heating associated with the equatorial intraseasonal oscillation also lowers from 300mb for α=0 to 500mb for α=0.1-∞. These changes seem to provide favorable conditions for the occurrence of the equatorial intraseasonal oscillation, in agreement with linear stability studies of a CISK model.
An analytic expression is derived to estimate the zenith angle distribution of the downward longwave radiation from sky elements in terms of the surface air temperature, surface water vapor pressure and zenith angle, taking into account the anisotropy of the downward longwave radiation. In addition, a formula for estimating the sky view-factor is also derived for the special case in which the zenith angle of the skyline is constant. It is shown, on the basis of this formula, that the additional downward longwave radiation is a function of the surface air temperature, surface water vapor pressure, surface temperature of obstructions, and the zenith angle of the skyline. The values estimated by use of this function fit well with Kondo's (1982) screening factor F2, assuming that the surface air temperature is equal to the surface temperature of the obstructions. The relative deviation of the screening factor for isotropic radiation from that for anisotropic radiation is larger with respect to increasing surface water vapor pressure and decreasing skyline elevation. Applying this formula, a new method is proposed to calculate the view-factors of visible and obstructed skies from the fish-eye lens image using a digitizer in conjunction with a computer.
The impact of the sea surface temperature (SST) anomalies and cumulus parameterizations is investigated for the case of the 1984 northern summer with the use of the MRI general circulation model. Integrations are performed either with the observed SST for 1984 or with the chmatological SST, either with the original Arakawa-Schubert cumulus parameterization (the A-model) or with a modified scheme (the M-model) which imposes an additional constraint between the minimum entrainment rate and the depth of the predicted planetary boundary layer. Intraseasonal oscillations in the low latitudes are better simulated with the M than with the A. However, the forecast skill with the M depends sensitively on the initial conditions. In one case, the observed eastward moving wave in the velocity potential field at 200mb over the equator is forecasted well, both in its phase and amplitude up to 20 days. Using different initial conditions, the forecast skill for the transient fields was poor. The M-model also has many advantages over the A-model in its climate simulation. It has simulated the Pacific subtropical high in the proper position and succeeded in simulating a Baiu-like rain band. The M-model has a stronger monsoon activity and a stronger sensitivity to the boundary forcing (here the SST anomalies) than the A-model, due to more unstable stratification. The impact of the SST anomaly is better simulated with the M than with the A. Monthly mean forecast skill in the low latitudes varies from month to month, but overall useful skill is found in the three-month mean forecast with the M over the tropics except at the very beginning of the forecast. The SST anomaly impact is not clearly determined, mainly because of the limited forecast samples and partially because of the modest SST anomalies in 1984.
Growth of snow crystals, i. e. growth of ice crystals from vapor is considered to occur by the following three processes: (1) a surface kinetic process for incorporating water molecules into the crystal lattice, (2) a diffusion process of water molecules in air towards the crystal surface, (3) a heat conduction process of the latent heat of sublimation. The resistance of each process is not constant, but depends on growth conditions such as supersaturation, the diffusion constant of water molecules in air, and also on crystal size. The purpose of the present sudy is the quantitative investigation of the role of each process as a rate determining process, and a discussion of the dependence on crystal size of the growth rate in the case of two-dimensional nucleation growth and in the case of spiral growth under various growth conditions. The results are interpreted from the viewpoint of the change of resistances of individual processes with increasing crystal size.
This paper describes the internal structure of two convective snowbands, deduced by using mainly single-Doppler radar data, which developed over the Sea of Japan in winter, traveled nearly perpendicularly to their orientation, and showed common features. A typical snowband was formed in a convective mixed layer 4km deep over the relatively warm sea surface in the early stage of a cold-air outbreak. This snowband was a multicell system which contained two or three echo cells in a vertical plane normal to its orientation. Each echo cell developed aloft in the forward portion of the snowband and became a mature echo cell in the middle of the snowband. The upper portion of old cells remains in a small anvil with low reflectivity below the stable layer in the rear part of the snowband. The snowband had the main updraft in its forward portion. The updraft tilted upshear and penetrated the stable layer above the convective mixed layer. There was a descending current from rear to front relative to the snowband in the lower half of the anvil. Snow particles evaporated in the downdraft. The leading edge of the downdraft was observed on the ground as a gust front with a severe gust and a drop in temperature of -1°C. The circulation and maintenance mechanism of the convective snowbands was similar to those of tropical and midlatitude squall lines.
The interrelationships among the 700mb zonal wind in the tropics, the sea surface temperature (SST) in the tropical Indian and Pacific Oceans, and the sea water temperature (SWT) in the tropical western Pacific are discussed on a characteristic interannual time scale. The interannual variability of the main thermocline in low latitudes along 137°E longitude line is reflected primarily on the quasi-biennial oscillation (QBO) time scale. To investigate its physical process, each dominant QBO mode propagating eastward is deduced by applying the complex EOF analysis to monthly mean SST and 700mb tonal wind data over tropical Indian and Pacific Oceans. The QBO mode of the 700mb zonal wind propagates eastward with uniform phase speed. On the other hand, the QBO mode of SST does not propagate with uniform phase speed and a marked phase difference is observed at the area around 150-160°E, where the zonal wind anomalies at 700mb level have the largest amplitudes. Thus, the QBO mode of tropical SST does not always propagate in parallel with that of the tropical troposphere and it is suggested that the tropical tropospheric QBO mode propagating eastward from the Indian Ocean brings about the remarkable phase shift of SST anomalies around 150-160°E. Since the SWT anomalies located in the main thermocline fluctuate in phase with those in the near surface layer (0-50m depth), it is inferred that these variations, which are influenced by the accumulation and release of warm water east of the Philippines, result from the dynamic response of the ocean to the wind stress with the QBO variability.
Relationships between the global sea surface temperature (SST) and the surface air temperature over Japan (TsJ) are examined by computing lagged cross-correlations based on the monthly mean data for 1970-1984. High correlation is found between the TsJ for December and the preceding SST over the central equatorial Pacific around 0°N, 160°W. Since the SST anomaly in this area persists for a long period, this relationship holds significant back to the summer (the SST leading the TsJ by half a year). For the TsJ in January or February, there is no significant relationship with the SST in the tropics. Therefore this relationship cannot be obtained by using three-month mean data. There is a high correlation between the TsJ for January or February and the SST over the North Pacific around 40°N, 150°W for a 0 to 2 months' lag with a lead in the SST. The corresponding circulation anomaly for December differs from that for January or February, although both are associated with an anomalous southerly wind and warm air advection near Japan. It is also found that the TsJ and the associated circulation pattern for March are highly correlated with the following El Nino event. During summer there is a high correlation between the SST over the western Pacific and the TsJ for July.
A multi-beam measurement of winds in the troposphere and lower stratosphere was performed in January 1986 by the MU radar. Twelve beams were used to scan the wind field zonally through the zenith angle from -30° to 30°. Analysing the observation data with high horizontal resolution, we have investigated the small horizontal structure of the wind for the first time. An analysis was made of remarkably strong oscillations of the vertical wind component with a large vertical scale which appeared in the lower stratosphere. Assuming the local wind homogeneity over any neighboring two beam positions in the zonal scan, we make a horizontal time section, which shows that the oscillations are zonally monochromatic. The line-of-sight velocities are fitted to a wavelike structure at each height by the least-squares method, and it is found that the zonal wavelength is 5-30km. According to the phase difference between the zonal and vertical components, it is likely that the oscillations are due to internal gravity waves. The zonal phase veolcity is very small (1-5m/s) compared with the background wind (≈50m/s), indicating that the waves are almost at rest relative to the ground, like a topographic wave. Another interesting feature is that the zonal wavelength changes with height, suggesting that the oscillations are not simply due to a monochromatic gravity wave, but have a complicated three-dimensional structure.
The Rayleigh lidar with a XeF excimer laser provides accurate temperature profiles in an altitude range from 30 to 70km. The power spectral density of the temperature fluctuation as a function of the vertical wavenumber is determined with these data. Both the stratospheric and mesospheric spectra can be obtained simultaneously. The slopes of the observed spectra are between -2.5 and -3.0, and almost coincide with the published values obtained by the wind observations. The observed spectral density in the upper stratosphere is smaller than that in the lower mesosphere, while the saturation spectrum theory predicts that the saturation spectrum in the upper stratosphere is larger than the lower mesospheric one. Difference between the observed and theoretical spectra are discussed.
The sea surface temperature (SST) in the tropical western Pacific displays a marked intraseasonal (30-60 days) variability and fluctuates, exhibiting a coupling with tropical convective activity on the same time scale with a phase difference of 10-20 days. The SSTs are above normal to the east of the eastward-propagating 30-60 day mode disturbances. An air-sea feedback system appears to be rather significant for the 30-60 day oscillation. It was also found that the air-sea coupling, with a phase difference on the 30-60 day time scale, was weak over the tropical western Pacific during the northern summer of 1981, the year preceding the ENSO event of 1982/83.