We have carried out a combined barotropic-baroclinic instability analysis using a moist three-level primitive equation model with a view to investigate the role of moist convection and non-geostrophic effects on the growth of the monsoon onset vortex and upper tropospheric easterly waves. We have also conducted an idealized study of the growth and energetics of the monsoon onset vortex using a non-linear five-level global spectral model incorporating parameterized cumulus heating. In the linear model we found a lower tropospheric growing mode whose scale length, structure and the growth rate resemble the observed characteristics of the monsoon onset vortex. This mode had maximum amplitude centered around 10°N and had a wavelength of 3500 km, a doubling time of 7 days and a westward phase speed of 1 ms-1. Our analysis revealed two upper tropospheric growing modes. These upper tropospheric modes have scale length and westward phase speed similar to the observed tropical monsoon easterly waves in the upper troposphere. In an idealized numerical experiment, by imposing a south to north heating gradient in a multilevel non-linear global spectral model, we have simulated the monsoon onset type of vortex. Using wave-CISK type of cumulus heating, we investigated the growth of this generated vortex. The growth of the cyclonic vortex was found in the presence of cumulus heating clearly elucidating the role of moist processes in the growth of the monsoon onset type of vortex. The energetics calculations of the simulated vortex revealed that there was a steady increase of AZ, AE, KZ and KE after the cumulus heating was switched on in the model. Barotropic exchange of energy shows some positive exchange initially and later on baroclinic energy conversion in the presence of cumulus heating becomes very prominent and is the main source of energy for the growth of the vortex.
Eight specially developed radiosondes were sent into cumulonimbus clouds developed over Pohnpei, Micronesia. The radiosondes transmitted both the images and the electric charge of precipitation particles during their ascent through the clouds. Micronesian clouds contained numerous ice crystals. Interesting images of precipitation particles included spheroidal raindrops, frisbee-like snowflakes, tetrapod-like hailstones, and spatial crystals with wide planes near the tropopause. Precipitation mechanisms were identified through analysis of the precipitation particle distribution with regard to height. Three different cloud types were identified. Rainbands producing heavy precipitation grew frozen-particle-hailstones in a narrow layer just above the freezing level. Water accumulated heavily in this layer due to the interaction between the warm rain process and the drop freezing process. Two other cloud types seen were isolated cumulonimbus clouds and upper convective clouds. In the former, both warm rain and graupel formation processes worked independently, while graupel formation dominated in the latter cloud. Electric properties of clouds were also briefly described.
A comparison is performed between K-theory and a 2nd-order closure model for the calculated profiles and fluxes within and above a vegetation canopy. Also, the relationship between the bulk transfer coefficients and individual leaf transfer coefficients is presented on the basis of K-theory and the closure model. Well-known differences are detected in the shape of the calculated profiles within the lower canopy layer. However, the differences are not serious in the calculated fluxes above the canopy, resulting in small differences in the values of the bulk transfer coefficients. With the computational simplicity, the results support the use of K-theory as being adequate enough in the prediction of the surface fluxes by a simple model.
A high-resolution (6.0-km grid at 36°N) one-way triple-nested hydrostatic numerical model successfully simulates meso-β-scale vortices developing along the Japan-Sea Polar-airmass Convergence Zone (JPCZ) cloud band. Simulated key features of the vortices include an eye-like" structure with its diameter of several tens of kilometers and surrounding spiral bands of upward motion. The simulated vortices emerge first as bendings in a belt of concentrated positive vorticity on the order of 10-3s-1 along the JPCZ. The bendings become increasingly sharp and eventually their troughs of the vorticity belt wrap round themselves to form meso-P-scale cyclones with a pressure deficit of 2-4 hPa accompanied by a dry eye and spiral bands of upward motion, within 4-8 hours. A warm core characterizes the eye and adjacent moist ascent. Not only agreement in spatial scale and time scale for growth between simple linear theory and the simulation, but energy conversion analysis of the simulated disturbance show barotropic shear instability for the dominant developing mechanism of the meso-β-scale vortices.
In order to get a deeper insight into the stratospheric circulation during sudden warming events, evolution of a circumpolar vortex perturbed by forced Rossby waves is investigated for a wide range of external parameters by using a high-resolution barotropic model in a spherical domain. If the amplitude of the wave forcing is large, the main polar vortex breaks down and the absolute vorticity is mixed irreversibly over the hemisphere. On the other hand, the main vortex migrates off the pole during the forcing period but returns to the pole afterwards without much erosion, if the amplitude of the forcing is small. The evolution of the polar vortex depends not only on the amplitude of wave forcing but also on other parameters such as latitudinal and longitudinal scales of the forcing, intensity of the polar vortex and latitudinal configuration of the vortex. The polar vortex breaks easily, if latitudinal extent of the forcing is wide or zonal wavenumber of the forcing is not 1 but 2. If the initial polar vortex is strong enough, there is a clear separation between the two types of the response at a critical amplitude of the wave forcing, i. e., breakdown or migration of the main polar vortex. Moreover, the evolution is highly dependent on the latitudinal configuration of the initial zonal wind over the globe : If the initial zonal wind is symmetric with respect to the equator, the vortex breaks easily. On the other hand, it is very robust if the zonal wind is anti-symmetric.
The meso-α-scale low development over the northeastern Japan Sea on 13-14 January 1986 is studied. The meso-α-scale low developed in the NW-quadrant of a parent large-scale low, which developed over the eastern coast of the Asian Continent. The local low-level frontogenesis occurred as a result of differential thermal advection between the polar air stream from the Continent to the W-SW of the large-scale-low and the warm air stream of Pacific origin secluded near the center of the large-scale low. The meso-α-scale low developed in the zone of the frontogenesis under the influence of a cold vortex aloft. The thick moist layer, which indicated the upward motion, over the NE side of the meso-α-scale low was seen in the warm secluded flow of the large-scale low, while the dry layer, which indicated the subsidence, over the SW side of the meso-α-scale low was seen in the cold polar air stream. This was a favorable feature for the baroclinic development of the meso-α-scale low. The meso-α-scale low reached its mature stage over the Japan Sea. In this stage, the surface analysis shows a distinct warm core in the center of the cyclonic circulation. The warm core disappeared when the meso-α-scale low weakened as it passed over the cold landmass of Hokkaido. The tip of the secluded warm air flow of the parent large-scale low spiralled asymptotically into the meso-α-scale low. The confluence of this warm air flow and the cold outflow from the island of Hokkaido caused intense snowfalls around the meso-α-scale low.
The effects of explicit cloud water are examined by adding the mixing ratios of cloud ice and cloud liquid water to the prognostic variables in the UCLA GCM. A simple cloud model and the published radiation schemes are incorporated in the model. Besides a grid-scale condensation process, the other source of cloud ice and cloud liquid water is the detrainment from the cumulus. The detrained cloud ice and cloud liquid water are obtained by the microphysical calculation in the Arakawa-Schubert cumulus scheme. The results are compared with the observations, concerned with cloudiness, planetary albedo, OLR and the dependence of cloud water content on temperature. The dissipation process of cloud ice with use of a realistic ice fallspeed produces realistic simulations. Cloud ice detrained from the cumulus simulates anvil clouds in the tropics. These clouds coexist with those due to grid-scale condensation, such as thin cirrus. Even at low temperatures, anvil clouds have dense cloud ice content which is one to two order larger than that of clouds due to grid-scale condensation. Anvil clouds produce low OLR, high planetary albedo and strong shortwave absorption in the tropical convective regions. Grid-scale condensation is active in moist environment due to the sublimation of the detrained cloud ice, and produces a net positive heating rate in the upper troposphere. The sublimation and melting of cloud ice, and the evaporation of rain produce a net cooling rate in the lower troposphere.
The climatological features of temporal and spatial variations of large-scale circulation and lower cold air (LCA) in the winter northern hemisphere are investigated using 9-year mean temperature and geopotential height data, from 1980 to 1988. The large-scale circulation pattern abruptly changes in late December (the 70th pentad) and mid-February (the 9th pentad), and shows a more meandering pattern in the period between these two pentads. We define this period as the climatological 'midwinter' in the northern hemisphere. During midwinter, LCA regions lying around the eastern part of Siberia and the Northwest Territories of Canada exhibit periodical and phase-locked variations with periods of about 25-35 days. The phase-locked migration patterns of three major LCAs in the eastern part of Siberia, Northwest Territories of Canada, and Scandinavian Peninsula in midwinter are investigated using the composite maps of 850 hPa temperature anomaly. The LCA over the Scandinavian Peninsula migrates from the high cyclonic activity area in the North Atlantic to the north side of the Tibetan Plateau. The LCA in Siberia is formed periodically with large temperature variation and migrates to the Central Pacific. The LCA over the North American Continent spreads in the meridional direction. The phaselocking of LCA movement is closely related to the appearances of some natural seasons and singularities in Japan.
A global atmospheric transport model was developed using a simple semi-Lagrangian method and the inter-hemispheric exchange was investigated. The time evolution of the mixing ratio of an artificial tracer at grid points with resolution of 2.5°horizontal and 14 vertical pressure levels is obtained using three-dimensional wind fields analyzed at the European Centre for Medium Range Weather Forecasts. A steep horizontal gradient zone was observed in the equatorial region of the lower troposphere and indicates a boundary between the hemispheres. The latitude of this boundary changed with longitude and the seasons ; it was located north of the equator, on the average. The boundary coincided with the Inter Tropical Convergence Zone in the Pacific throughout the year and shifted to 30°N in Asia in the northern summer and to 30°S in South America in the northern winter. Monthly flux of mass across this zone was estimated. The maximum mass exchange from south to north occurred in December and the minimum in May ; the maximum from north to south was in May and the minimum in October. The interhemispheric exchange time was about one year. The results of this experiment suggest that sources and sinks of atmospheric constituents along the Indo-China peninsula may affect air quality in the Southern Hemisphere during the northern summer while such sources and sinks in the Amazon Basin may affect air quality in the Northern Hemisphere during the northern winter.
Synoptic sounding and surface data are used to calculate heat sources and moisture sinks over China during the pre-Meiyu and Meiyu periods of 1987, 1988 and 1989. The horizontal distributions of the vertically integrated heat source <Q1> and moisture sink <Q2> show that during the pre-Meiyu, the maximum heating is located in southern China in a band parallel to the coastline with a secondary maximum over the Yangtze Valley. When the Meiyu sets in, there is an increase in the heating in the Yangtze region ; however, a band of heating continues over southern China during the Meiyu. In 1988 there is a prominent third band of heating over northern China. There is pronounced diurnal variation in rainfall during both the pre-Meiyu and Meiyu periods due to land/sea and mountain/valley breezes. At night, maximum rainfall occurs at lower elevations in the prominent basins of interior China. During the day, rainfall maxima shift to the region between the coast and the coastal mountain ranges and other areas favored by upslope flow. The vertical profiles of the heat source Q1 and moisture sink Q2 averaged over subregions of China show that during both the pre-Meiyu and Meiyu, the precipitation type over the Yangtze region is a mixture of convective and stratiform, whereas over southern China deep cumulus convection predominates. The heating and drying in southern China occur throughout the whole troposphere, but low-level moistening occurs in the Yangtze region. The pattern in southern China resembles that observed in the deep tropics, whereas that in the Yangtze region is similar to that observed with midlatitude mesoscale convective systems where low-level evaporation is important.
Ozone profiles obtained with the DIAL (Differential Absorption Lidar) system at the National Institute for Environmental Studies (NIES), Tsukuba, Japan (36°N, 140°E) were compared with data provided by the satellite sensor SAGE II. The SAGE II data were selected based on criteria of spatial and temporal differences between the DIAL and the SAGE II measurements : five degrees in latitude and fifteen degrees in longitude, within a latitudinal band from 31° to 41°N, and within one, three and five days after or before the DIAL measurements. Results show very good agreement for the individual and the zonal-mean profiles. The average mean difference between the DIAL and the SAGE II measurements over the altitudes 15-50 km was about 10%.
Wang and Mitsuta (1990) reported that a downward water vapor flux was observed in the daytime over the Gobi desert. The present author pointed out the importance of mean flow contribution to total water vapor flux due to air density fluctuation. The correction factor was estimated to be about -22% to reduce the downward flux.
The spectral aerosol-optical-thicknesses (AOTs) of vertical air columns were measured by a ground-based multi-channel sunphotometer at Tsukuba (36.05°N, 140.13°E), Japan, during January through April of 1991 and 1992. The AOTs measured during the 1992 season were much larger, by as much as 0.1 to 0.2, than those in the 1991 season. This difference was mainly attributed to the enhancement of stratospheric aerosols caused by the Mt. Pinatubo eruptions in June 1991. The stratospheric AOTs exhibited a maximum during February 1992, which exceeded the maximum value observed in December 1982 for the El Chichon volcanic dust. The aerosol size distributions in the optically active size range were retrieved from the spectral. A fairly narrow mono-modal size distribution, having a mode radius of about 0. 6μm and an effective variance of less than 0.05, was retrieved for the Pinatubo volcanic aerosols. This mono-modal size distribution is similar to that estimated from Bishop's ring simulations (Asano, 1992).
This note presents observational results and their mechanistic understanding of the long-term variation of Kelvin wave activity in the equatorial lower stratosphere. Analyses are made for 15 years (1974-1988), using the rawinsonde station data at Singapore (1°N, 104°E). Occasionally, clear fluctuations with periods near two weeks can be observed in the zonal wind and temperature fields, suggesting the prominent existence of Kelvin waves. The variability of Kelvin wave activity (measured by the power spectral density integrated over periods in the range around two weeks) is clearly related to the quasi-biennial oscillation (QBO) of the zonal wind in the lower stratosphere. Around the 30 mb level, the Kelvin wave activity is vigorous when the wind regime of the QBO rapidly changes from easterlies to westerlies. In order to understand the observational results, we perform a simplified mechanistic calculation following an idea that the momentum flux due to the Kelvin wave is locally constant and suffers damping which is inversely proportional to the vertical group velocity. In this calculation the observed monthly-mean zonal wind data are used for estimating Kelvin wave amplitudes on a monthly basis. Results of the calculation represent well the long-term variation of Kelvin wave activity associated with the QBO.