The structure of the multiple vortices formed in small scale atmospheric vortices is studied in a laboratory experiment. A guide vane type simulator is shown to be capable of producing the multiple vortices for large swirl ratios. The number of subsidiary vortices increases with the swirl ratio. A condition with 3 subsidiary vortices is examined closely with an improved smoke-wire and a two component hot-wire anemometer. Core sizes, velocity distributions and other properties of the subsidiary vortices are determined by the measurement. A subsidiary vortex rotates around the center of the parent vortex keeping its identity. But the ratio of the translational velocity of the subsidiary vortex to mean tangential velocity is about 0.9 in the layers far from the ground, and about 0.7 near the ground because of the vertical shear in the mean field. The cospectra between the two horizontal velocity components indicate that in the upper layers the subsidiary vortex transports momentum from the maximum velocity region of the parent vortex toward the center. The transport process is more complex near the ground.
An inertio-gravity wave critical layer is defined as the region in which the conventional WKB-type dispersion relation is mathematically invalid (Yamanaka and Tanaka, 1984b). This layer is bounded by a turning level at which wavefront parallels the basic isopycnic surface and by a level inside which geostrophic adjustment for the wave momentum can hardly take place. The valve-like critical 'level' absorption (Grimshaw, 1975) is explained by an anisotropic wavefront revolution due to the baroclinicity of the basic field. The induced zonal-mean field, correct to the second order of wave amplitude, resembles an inertial oscillation or a symmetric isopycnical motion. The net absorption rate of the critical 'layer', which coincides with that of the non-inertial waves (Booker and Bretherton, 1967), can be explained by a resonant energy conversion from the wave to the zonal-mean inertial oscillation. The geostrophic flow deceleration results from a kind of resonant redistribution of mean kinetic energy so as to maintain the zonal-mean isopycnical motion; this leads to a final equilibrium state in which Richardson number of the mean zonal flow in the critical layer is unity. Although the non-inertial approximation holds outside the layer with a basic Richardson number larger than unity, above-mentioned features are important to discuss the gravity-wave stress and quasi-horizontal diffusion in the strato-sphere.
Linear responses of the stationary planetary scale disturbances to subtropical and trop-ical heat sources are investigated for summer northern hemisphere. The model with high vertical resolution is used. Its top is 80km. Zonal mean flows are assumed using FGGEIIIB data by ECMWF. If there are heat sources at 30°N and 10°N, the solutions agree well with observations. The wave of zonal wave number 1 propagates to 60°N. It has two maxima around 30°N and 60°N and minimum at 40°N. The wave of wave number 5 can propagate to 45°N and has only one maximum there. The results suggest that Indian Monsoon activities and heat sources over the Tibetan Plateau can affect on the stationary patterns in middle and high latitudes through poleward propagation of stationary Rossby waves even in summer. On the other hand, the meridional profiles of zonal flows significantly affect on whether the waves forced by heating at ITCZ can propagate northward or not.
A performance study of the tropospheric general circulation with the MRI (Meteorological Research Institute)•CM is presented for January. The MRI•CM is basically identical to the UCLA•GCM (Arakawa and Mintz, 1974; Arakawa and Lamb, 1977) with minor changes to both the dynamical and physical processes of the model. The resolution of the model is 5° and 4° in longitudinal and latitudinal directions, respectively. The top of the model is located at 100mb and the atmosphere is divided into 5 layers. Both the sea surface temperature and the sea ice distributions are specified based on the climatological data, while other variables are determined within the model either prognostically or diagnostically. The model has succeeded in reproducing basic characteristic features of the general circulation. Simulations are especially good in the tropics except that the mixing ratio of water vapor near the surface is underestimated in the model. Overall characteristics of the southern hemisphere are also simulated well. On the other hand some systematic disagree-ments between the model and the climatology are found in the vicinity of the Tibetan Plateau*. Northerly is too strong near the surface along the eastern coast of China, and easterly is also too strong in the southern periphery of the Plateau. This anticyclonic flow is cold and dry, and thus enhances evaporation in the Bay of Bengal and precipitation over the equatorial Indian Ocean. It is suggested that the effects of small scale mountain ranges should not be underscored because they determine low level flows and thus the heating distribution through the air mass transformation process over the warm ocean. Another notable defect is the poor simulation of the Aleutian low both in its position and intensity. This is consistent with the too low static stability in high latitudes due to the excessively cold temperature in the upper part of the model in high latitudes. The maximum decrease in the static stability is found in the area from Alaska to the north-western part of Canada, in agreement with the extensions of low pressure area in that direction. Arakawa and Schubert's (1974) theory is adopted in parameterizing penetrative cumulus convection. Thermal and dynamical roles of the parameterization are studied through budget analyses of heat and zonal momentum in low latitudes.
The predictability of the operational ECMWF forecast model in the tropics has been studied using the archived data for 1983 and 1984. The predictability of the very large scale quasi-stationary motions for the time scale of longer than 30 days and the transient motions for the time scale of 3-10 days are examined separately. The very large scale motion is predictable only up to 2 days in the tropics. This is very short compared to about 6 days in the northern hemisphere mid-latitudes and about 4 days in the southern hemisphere. The largest part of the error for 7 day forecasts is found to be due to the systematic error; about 70-90% of the total error can be explained in this way. The systematic error has a structure similar to the gravest internal symmetric Rossby mode of zonal wavenumber 1, but it grows without propagation. Several forecast experiments have been performed to isolate the cause of the systematic error. It has been shown that the error is most sensitive to the convective heating distribution in the tropics, indicating a major weakness of the convective parameterization in the model. The predictability of the transient disturbances in the tropics is examined using the time filtering technique. The results indicate that the disturbances are predictable up to or beyond 4 days over the central Atlantic and over the central Pacific. However, the forecast skill is low over the far eastern parts of the Pacific ocean where the disturbance develop into tropical storms. This is probably due to the resolution of the model being too coarse for predicting tropical storm development. The forecast is also not skillful in the Indian Ocean where satellite wind data is not available.
The interaction between small-scale convections and large-scale flow in the airmass transformation processes is investigated by making use of a numerical model. In the model a uni-directional large-scale wind is assumed. Fine resolution is used in the cross-stream and vertical directions in order to treat explicitly the convections transporting heat, mois-ture and momentum in the transformed layer, while coarse resolution is used in the down-stream direction to represent the slow evolution of the large-scale flow. The average profiles of potential temperature, specific humidity and velocity indicate that the transformed layer forms a moist convectively mixed layer characterized by a multiple-layer structure. When the sea surface temperature rises in the downstream direction, a pressure drop in that direction is produced by the convective heating, and the acceleration due to the pressure gradient force is balanced mainly with the deceleration due to the frictional force induced by the convective momentum flux. Because the excess of the pressure gradient force over the frictional force produces a mean downward motion, the development of the mixed layer is suppressed. Because the suppression is brought about by the convective heating which develops the mixed layer, the process can be called a negative feedback process.
The lower-frequency variations in the winter monsoon system have been examined using a l0-day low-pass filter. Results indicate that the lower tropospheric north-south thermal gradient over eastern Asia, the Siberian high pressure system, and the eastern Asia subtropical jet stream varied coherently with a time scale of 10-20 days. Composite study reveals that the 200-mb streamfunction variations on the 10-20 day time scale are longsynoptic-scale disturbances propagating slowly eastward. The cyclonic disturbances weaken as they move eastward south of the Tibetan Plateau and begin to strengthen when the lower tropospheric thermal troughs from Siberia move southward. The irrotational circulations on the 10-20 day time scale are closely related to the streamfunction variation with a quarter-wave phase shift. The tropical-midlatitude interactions in the winter monsoon system are seen to be the interactions between these long-synoptic-scale waves and the local monsoon circulations based on the phase relationships among various elements studied. The divergent circulations of the propagating waves trigger the cold air outbreak over eastern Asia when the wave trough is situated south of the Tibetan Plateau. The propagating waves are seen to intensify over the ocean suggesting a transfer of energy from the monsoon system to the waves.
The large temperature variation in summer over the northeastern part of Japan (Tohoku District) is studied with special emphasis on the occurrence of anomalously cold summer. The result of EOF analysis of the monthly mean temperature for 1951-1980 and that of the daily mean temperature for 1976-1980 over Japan shows that the lst EOF component accounts for -70% of the total variance. The spatial function of the 1st EOF component indicates the same sign over the whole analyzed domain with small value in the southwestern Japan and the large value over the Pacific side of Tohoku District. Some statistical relations between the time variation function of the 1st EOF component and some meteorological parameters expressing the large-scale pressure (height) field are examined. The large-scale situation in the anomalously cold summer is also compared with that in the anomalously hot summer. It is concluded that the cold summer in Tohoku District caused by the southwestward intrusion of the polar maritime airmass which is formed over the Okhotsk Sea, the Bering Sea and the northern Pacific. The high correla-tion among the surface air temperature, the surface sea water temperature and the sunshine duration are also pointed out. It is the very interesting and important fact that the anomalously cold summer occurs over the localized area (the Pacific side of Tohoku District), despite the fact that the cold summer is caused by the large-scale easterly flow from the polar maritime airmass. The structure of the cold easterly flow which causes the localized Tohoku cold summer will be studied in the Part 2 of the present study.
The detailed analysis on the strucure of the polar maritime airmass is made for two typical cases of summertime cold spell (11-20 August 1976 and 25 July-4 August 1980) over Tohoku District. The summertime cold weather over Tohoku District is caused by the westward cold air outbreak from the polar maritime airmass formed over the Okhotsk Sea, the Bering Sea and the northern Pacific. The thin (the surface to about 1 km) layer of the cold easterly wind shows the character of the mixed layer and is capped by stable layer. The large thermal gradient (∂T/∂x--2K/100km), and the thermal advection (∨•∇T-0.2K/hour) are restricted within the mixed layer. The amount of total heat energy supplyed from the sea to the atmosphere over the sea adjacent to Tohoku District is estimated to be -1001y/day. The polar maritime air can reach Tohoku District without the full airmass transformation, because the sea water is cold in the Pacific adjacent Tohoku District. The analysis shows the influence of topography and the ground temperature on the air temperature and cloud (sunshine duration) in the cold air. It is infered that the orographically induced downward motion in the west side (lee-side for the easterly wind) of the mountain ranges in Tohoku District disperses the cloud formed in the upper part of the mixed layer, and causes the temperature rise due to the insolation over the Japan Sea side of Tohoku District. The thin cold air is rapidly warmed over the ground and/or sea surface with the higher temperature. This explains that the summertime cold and cloudy weather is localized over the Pacific side of Tohoku District.
The regional variations of the spectra of the surface winds in Japan are investigated from the viewpoint of spectrum climatology. For 206 stations in Japan, the spectra of the surface winds over periods from 2 hours to a few years are analyzed using 5 years of 10-min averaged, hourly data. Characteristics of the spectra can be used as a climatic indicator. The shape of the spectrum at short periods (less than 10 days) depends on the local topography, and that at long periods (more than 10 days) is related to the geographical location. Most of the spectra are classified into three categories: The first is a strong wind type with a major synoptic peak, the second a strong wind type with a pronounced annual peak, the third a weak wind type with a major diurnal peak. But the spectra on the Nansei Islands (24-31°N, 123-130°E) are quite different from those on the main islands of Japan. The spectra on the Nansei Islands show two marked features: Firstly, the 40-60 day periodicity is predominant. Secondly, the synoptic portion of the spectra has a wide range of periods from 2 to 20 days. These features are related to the climate on the Nansei Islands, which is maritime and subtropical. The geographical distribution of the 40-60 day fluctuations is in good agreement with that of the zonal mean components of the surface winds. The 40-60 day fluctuations are predominant during summer half year and are associated with strong winds.
A case study is made of an exceptionally heavy rainstorm that hit the northwestern coastal area of Kyushu, Japan, on 23 July 1982. The 5hr rainfall accumulation was as high as 412mm at the city of Nagasaki. The precipitation occurred along the warm front as-sociated with a medium scale disturbance that developed along the Baiu front. The atmos- phere in the pre-storm period was very moist through virtually the entire troposphere and its static stability was conditionally unstable with a lifted index of -1.5°C, thus exhibiting characteristics more similar to the tropical atmosphere than the typical pre-storm environment over the midwestern United States in spring seasons. Prior to the development of heavy precipitation over the Nagasaki area, an intensive rainband was propagating with a speed of 40km hr-1 eastsoutheastward over northwestern Kyushu. Its structure is found to be different from the structure of typical tropical squall clusters in that the heaviest rainfalll occurred in the middle of the rainband without a well defined leading edge. A dramatic change occurred in the rainband when its southern tip reached the Nagasaki area; it stopped propagating and stayed there for the next 5hr. Its line structure changed to a blob structure at the meso-β scale (-100km). The timing of the precipitation development at Nagasaki coincided with the peak period of the incursion of very moist air associated with the southwesterly low-level jet. During the period of heavy precipitation, a new cloud cluster formed over the sea 300km west of Nagasaki. It kept developing and propagating eastward with a speed of 60km hr-1, and eventually merged with the rainstorm over Nagasaki. These observations strongly suggest that the Nagasaki rainstorm was trapped orographically. Several other heavy precipitation events are cited which also occurred in association with medium scale disturbances along the Baiu front and in which the precipitation developed over the coast areas upwind of the low-level jets.
A rainstorm occurred over the central part of Japan (-36°N/14°E) within Typhoon 8124 (Gay) for 22-23 October 1981 and it was studied mainly using Doppler radar data. The main purposes of the present study are to clarify the structure of the rainstorm and to know whether it was a typhoon spiral band or another type of precipitation system. The typhoon was under transformation into an extratropical cyclone in the southeastern portion of a large-scale trough. Satellite and radar data show that this rainstorm occurred on the southeastern edge of a wide cloud band to the north of the typhoon center. The most outstanding feature of this rainstorm found by Doppler radar was the existence of a slant axis of strong wind from lower levels on the southeastern side to upper levels on the northwestern side. This means the existence of a mesoscale slant updraft. Below the axis of the slantwise updraft, convective-scale vertical motion was embeded in the mesoscale updraft. Above the axis, convective-scale vertical motion was generally small. The middle-level air intruded into the northwestern portion of the rainstorm in the southern part. It is suggested that the intruded air was cooled by evaporation of precipitation particles and formed a mesoscale downdraft. Although a pronounced surface convergence line was as-sociated with the rainstorm, its effect on the rainstorm was subsidiary except in the southern part. The structure of the rainstorm was partially similar to those of a typhoon spiral band, eyewall clouds and a squall line in the middle latitudes. However, this structure is considered as a characteristic feature of a rainstorm which occurs to the north of a typhoon in extratropical transition. The result and interpretation of a mesoscale budget of condensed water in the form of precipitation particles are also shown to study the relative importance of production and transport of precipitation particles in the rainstorm.
The two-stage numerical model developed by Kimura (1983) is extended to three-dimen-sional and used to study the relation between local winds and photochemical air pollution on the Kanto Plain. The first stage is a three-dimensional, timedependent local winds model which gives the wind velocity and vertical diffusion coefficient. The second stage is a photochemical air pollution model which uses the results of the first stage as input data. The calculated results are compared with observed surface wind velocity and pollutant concentrations when the large scale wind are light (≤3m/s). Good agreement between calculated and observed wind fields is obtained. The calculated concentration of ozone also agrees with the observed oxidant concentration, which is expected to be almost equal to ozone concentration. The simulation also suggests that the mountainous area of Central Japan is very important for the wind system over the Kanto Plain, forcing pollutants from the Tokyo metropolitan area to be more often transported westward than eastward. Previous air pollution models have needed a lot of meteorological data but some of these, such as upper level wind velocities, are difficult and expensive to obtain. The two stage model will be very useful for preliminary estimation of photochemical air pollution effects where detailed meteorological data are not available.
The data from several rural stations of the air quality monitoring network in Japan are useful for representation of the tropospheric ozone on the ground surface. The seasonal variation of the tropospheric ozone is investigated by using these data along with those from ozonesonde soundings at Sapporo, Tateno and Kagoshima. They show a regular pattern of a summer minimum as well as a spring maximum which is a common feature of the tropospheric ozone at middle latitudes in North America and Europe. The summer minimum implies that photochemical production is not a principal factor to control the tropospheric ozone, and it may be attributed to the weak transport of ozone from the stratos-phere, being associated with the fact that the tropical marine air mass prevails over Japan in summer.
This paper is mainly concerned with power spectra and scales of turbulence in strong winds in the range of 10-60m/s. The spectra and the scales were calculated with the method of maximum entropy spectral analysis. The spectra for both longitudinal and ver-tical velocities are represented well by the algebraic expression suggested by Fichtl and McVehil (1970) and Busch (1973). But the factor controlling peakedness and the peak wave number vary with mean wind speed for the longitudinal velocities. The peak wave number is not proportional to the height above the ground for the vertical velocities. A few kind of scales are dealt with and compared with each other. As the integral scales are overes-timated due to underestimates of high-frequency components of wind fluctuations, the meas-ured values were corrected. The ratio of integral scales to scales connected with spectral peaks agrees well with the theoretical result obtained from the algebraic expression of spectrum. The scales of the longitudinal velocities for over-sea winds increased with the mean wind velocities and were larger than those for over-land winds. The above experi-mental result shows that a similarity theory given as a function of a parameter nz/U does not hold true. The scales of the vertical velocity for a rough terrain were not proportional to height. It also means that the similarity theory is not consistent with the present data for the vertical velocity.