A two-layer, quasi-geostrophic, low-order model is constructed to investigate the nonlinear interactions between the zonal flow, topographically forced waves and free baroclinic waves. In Part I free baroclinic waves are excluded to direct our attention to the zonal flow-forced wave interactions. In the conservative case, without external thermal forcing and dissipation, equilibrium solutions are obtained and the resonance condition in the present two-layer model is examined. Multiple flow equilibria are also obtained in the non-conservative case. However, unlike the barotropic model of Charney and DeVore (1979), there do not exist two stable equilibria. A multiplicity of the time-dependent solutions is found in a certain range of the external parameters. There exist two or more stable periodic (or aperiodic) solutions for the same external conditions. The selection of a solution depends on the initial condition. As the external thermal forcing parameter increases or decreases by bits, the period of a stable periodic solution can become doubled, quadrupled, and so on. Finally aperiodic solutions appear through the period-doubling sequence phenomenon.
Nonlinear interactions between the zonal flow, topographically forced waves and free baroclinic waves are investigated by using the two-layer, quasi-geostrophic, low-order model constructed in Part I (Yoden, 1983). An idealized topography is given by a Fourier component with the largest scale permitted in the present model. When the zonal flow is more unstable with respect to a free wave than to the forced wave, there appears a final steady state in which the finite amplitude free wave with a constant phase velocity balances with the marginally stable zonal flow and the forced wave decays out. On the other hand, when the flow is more unstable with respect to the wave component directly coupled with the topography, the flow system has both of the forced and free wave components. All the wave components are coupled with the topography in least severely truncated case in this paper (two meridional modes and three zonal wavenumbers of n, 2n and 3n are permitted). Then the flow system has several types of time-dependent behavior depending on the external parameters such as the external thermal forcing, the frictional dissipation and the static stability: Steady flow with constant forced wave and propagating free wave, periodic or quasi-periodic oscillation and irregular fluctuation. For the external parameters corresponding to the real atmosphere, there appears an irregular fluctuation with large-amplitude waves. Statistical relation between the zonal flow and waves in the irregular fluctuation is investigated over a long time-span. The flow pattern at each time step is classified into one of three categories in terms of the magnitude of the mean zonal flow. Composite fields in three categories are characterized by the zonality in the high-index state and the moderate state and by the meander of the flow in the low-index state. When the flow is in the low-index state, both of the mean value and vertical shear of the zonal flow are small, the stationary waves have larger amplitudes, and the transient waves have smaller amplitudes compared with the high-index and the moderate states. The structure of the stationary waves in the irregular fluctuation is different from that of the forced wave in the equilibrium solutions in Part I.
In order to discuss three-dimensional teleconnection in the zonally asymmetric height field during the Northern Hemisphere winter, the so-called one-point correlation map is used. In the map, the correlation coefficient between 700mb zonally asymmetric height at the base grid point and p-mb (p: pressure) zonally asymmetric height at every other grid point is computed, based on the data set of monthly mean height for 30 winter months (Decembers, Januarys and Februarys for the 10 winters 1969-70 through 1978-79). The discussion is mainly forcused on the case where the base grid point is (45°N, 135°E) at 700mb level over the northern part of Japan. The results show that there is a strong teleconnection in the troposphere between the northern part of Japan and the western part of Barents Sea (or the northern part of Norwegian Sea), and the 700mb zonally asymmetric height over the northern part of Japan is highly correlated in the negative sense with the 200mb zonally asymmetric height over the subtropical western Pacific Ocean. The teleconnection mentioned above is discussed in relation to the standing wave of geopotential height of zonal wave number 1 and this standing wave is examined as the response of a model atmosphere to forcing by topography and stationary heat sources in the atmosphere. The teleconnection with lag time in the zonally asymmetric height field and the zonally symmetric seesaw in the geopotential height field are also discussed.
The energetics characteristics of the Asian winter monsoon over the Aleutian low area are described using twice daily upper-air observations during a three year period. The area shows the typical energetics features observed for occluded cyclones. However, at the tail end of the winter monsoon, they are distinctly different from energetics patterns reported over source regions and the Kuroshio area. The kinetic energy level is significantly lower, and the energy transformation is much less intense. The cross-isobaric generation and dissipation are roughly balanced. The contribution of baroclinic conversion is secondary as the energy source.
Using the infrared (IR) irradiance data observed by GMS-1 geostationary satellite, it was tried to quantitize the deep convective activity in a manner which is physically more interpretable than the conventional areal average. This was partly accomplished by defining the intensity index of the deep convective clouds within the 1° longitude-latitude square mesh in the manner which takes the horizontal variation of the observed temperature and the vertical distribution of the atmospheric temperature into account. Diurnal variation of the deep convective activity was investigated by using the abovementioned intensity index for the period during northern hemispheric winter (December 1978January 1979) and summer (July-August 1979). During northern winter, the amplitude of the diurnal variation is large over the Indonesian region and the northern Australia. Within this area it was found that there exist a distinct contrast in the phase of diurnal cycle between the land and the adjacent ocean. Composite analysis has revealed that the land shows the suppressed convective activity in the morning hours with the minimum around 9 o'clock local time. The convective activity is rapidly enhanced in the afternoon and reaches the maximum at around 18 o'clock local time. In contrast, the adjacent ocean clearly shows the enhancement of the convective clouds in the local morning and the suppressed convective activity in the afternoon and night. During northern summer the intense diurnal variation was observed over the southern part of Tibetan Plateau. The amplitude is also large over the western Pacific to the east of Philippines. Composited diurnal cycle shows that there exists a contrast between the continent and the ocean, being similar to the one observed during northern winter. The continent and the large islands generally show the minimum convective activity around 9o'clock in the local morning and the maximum enhancement aound 18 o'clock in the evening. In contrast the oceanic area like western Pacific shows the maximum enhancement of the deep convective clouds around 6- to 9 o'clock local time in the morning and the suppressed activity in the afternoon. In addition to this land-ocean contrast, the evidences which suggest the regional effect on the diurnal variation are also discussed.
Geostationary Meteorological Satellite (GMS) observations of diurnal variations of cirrus canopy and eye diameter in mature northwest Pacific typhoons are presented. The variations in cirrus canopy are quantitatively determined by the areal distribution of equivalent black body temperatures (TBB) (10.5-12.5μm, window channel) in the ranges: TBB_??_-70°C, -70°C, <TBB_??_-50°C, -50°TBB_??_30 and -30°TBB_??_0°C. The coldest region has a sharp maximum at 0600-0730LST and a minimum at 1800-2100LST. The maxima in the warmer cloud tops occur some hour later at 1500-1800LST and 2100LST, respectively, as rising the TBB thresholds. Satellite observed eye diameter exhibits a wedge-shaped morning minimum and a plateau-shaped afternoon-evening maximum. The time lagged diurnal variation of areal extent of warmer cirrus clouds, and the out of phase variation of eye diameter compared to the coldest cirrus clouds, is explained as a result of outward and inward expansion of cirrus following an early morning peak in eyewall and spiral band convections. When a typhoon passes over or near to large islands, the total cirrus canopy is reduced in size and a second afternoon peak in areal coverage develops.
The flow characteristics of the small scale atmospheric vortices are investigated based on a laboratory experiment for various types of ambient circulation. The following results are obtained when the ambient circulation is supplied in the upper layers: 1) The radial distributions of velocity are found to have two maxima, which may correspond to the cores of the tornado and the mesocyclone respectively, 2) the vortex breakdown occurs at small inflow angles when Reynolds number is large, and the swirl angles near the core are larger than those when the ambient circulation is supplied throughout the whole convergence layer, and 3) lower layers have steeper radial gradient of the horizontal velocity compared to the upper layers right outside of the core.
In the pre-GARP experiments in 1969 and 1970, we measured mean profiles of wind speed, air temperature and humidity over the open sea. Many abnormal wind profiles were obtained under the conditions of high wind speeds and swell wave. The wind profiles did not have a logarithmic shape, but had a jet-like flow at about twice as high as wave height. All of the wind profiles were observed over the open sea about 200-300km away from the land by the use of R/V Hakuho-Maru. Wind speeds were measured by five small threecup anemometers installed along the pole 9.6m away from the bow, and wave heights were observed by the gyroscope, accelerometer and sonic wave gauge. Overspeeding of the cup anemometers, ship movements and thermal stratification of the air layer were taken into account for obtaining the wind profiles. Our conclusions are as follows: As a mean wind speed increases on the open sea with significant swell waves, a departure from a logarithmic wind profile becomes larger, and obvious two peaks in the profile are recognized corresponding to the wave height distribution with two maxima.
A model for the free convective internal boundary layer (IBLt) is derived by replacing the time axis in the mixed layer theory with the downwind distance axis. Summers' (1964) original model for the heat island is modified to the coastal atmosphere, taking into consideration the horizontal variation of the surface heat flux and vertical variation of the temperature lapse rate of the stable marine atmosphere. Also, a model for the spatial variation of turbulent quantities in the IBLt is obtained. Estimation of the potential temperature within the IBLt agrees well with observations. Estimated trends of turbulent quantities agree with observations as a first order estimation.
A numerical model was proposed to simulate air flow and gas diffusion over a two-dimensional ridge in neutral and stable stabilities (bult Richardson number Rib=0.0, 0.7, 1.8). This model is based on the finite difference method and can precisely calculate the viscous convective flow over a no-slip oblique boundary. It was found from the calculated results that a reverse flow appeared upstream of the ridge and a standing vortex disappeared downstream when bulk Richardson number Rib became greater than about 1.0, and that the gas concentration increased upstream of the ridge in stable condition. Next, a wind tunnel experiment was performed in neutral and stable stabilities (Rib=0.0, 0.7 and 1.8) in order to verify the calculated results. The experimental results showed that the flow patterns and concentrations significantly change with stability, in particular high concentration appeared upstream of the ridge due to the reverse flow in stable conditions. Finally, the experimental results were compared with the calculated results and the good agreement was obtained between them.
Previously, the author (1976) investigated statistical relations between meteorological elements and the oxidant concentration in Hiroshima Prefecture using the data of warm season in 1974. In this previous paper, correlations between several meteorological elements and the daily maximum value of the oxidant concentration in Hiroshima Prefecture were examined, where the selected meteorological predictors were the wind direction and speed at 950mb level, the static stability of lower atmosphere and the maximum surface air temperature during the daytime. The present paper reports the extended analysis with new supplementary predictors, the washout effect of rainfall and the maximum value of the oxidant concentration measured at 0900JST in Hiroshima Prefecture. The data of warm seasons from 1976 to 1978 are used in addition to those of 1974. The steadiness of the above correlations is tested by the data of 1975. Here, the present investigation is aimed at the prediction of the daily maximum oxidant concentration in Hiroshima Prefecture whose day is same as that of the prediction publication in the morning. In conclusion, it is suggested that an improved prediction scheme of the oxidant concentration in Hiroshima Prefecture can be obtained on the basis of the present analysis.
For regional weather forecasting and spatial meteorological/environmental studies, the effectiveness of the data collection is reduced if one or more of these concurrent observations are missing or not recorded. By processing the available meteorological data for the Kingdom of Saudi Arabia, it is found that the large amount of information on meteorological variables is either partly missing or not recorded for a time period. It is, therefore, necessary to simulate such missing records using appropriate estimation model. The estimation models cited in the literature are reviewed and Kalman filter estimation model is selected for simulation purpose. Shannon's entropy concept is used to estimate the parameters of the model. Methodology is then applied to simulate missing records on temperature, pressure and humidity. The estimation performance of the model is evaluated by computing mean and variance of both observed and simulated values as well as on the basis of statistical theory of error variance.
The upper-air data were examined for a few cases of heavy rainstorms over Japan, in expection of finding internal gravity waves generated over the storm areas. Among the cases, there was one case where variations of temperature, relative humidity and horizontal wind velocity in the troposphere showed features of internal gravity waves in the downstream of heavy rainfall (∼150mm/day) area. The horizontal and vertical wave lengths of the variations were ∼400km and ∼3km, respectively. The amplitudes of temperature, humidity and wind velocity variations were estimated as ∼2K, ∼20% and 4m/sec respectively. But, no wave-like variation was detected in geopotential height. The wave-like variations were not found from the observational data 12 hours later.