The evolution process and structure of a polar low of the meso-α-scale, which formed over the Japan Sea on 11 December 1985 are studied. In Part I, the evolution process and the meso-α-scale structure of the polar low are studied. In Part II, the meso-β-scale fine structure of the polar low is studied. An important synoptic-scale setting for the polar low formation is the low-level cyclonic polar air streams over the Japan Sea, which are induced by the development of a synoptic-scale low over the northwestern Pacific near Japan. In association with the passage of a meso-α-scale cold vortex aloft over the Japan Sea a "moist dome" (dome-shaped polar air of moist neutral lapse rate) forms over the Japan Sea. At the same time, a meso-α-scale low (polar low) develops and moves south-eastward. This polar low has a warm core structure in the 900∼700 mb layer and well-organized spiral cloud bands. The cyclonic circulation of the polar low is limited in the lower troposphere. The 24-hour numerical prediction of the present polar low by a meso-scale primitive equation model is examined by comparing with observations. The formation process and the meso-α-scale structure of the polar low are well predicted by the model from the initial state 12 hours before the polar low formation.
The evolution process and structure of a polar low (meso-α-scale low in the polar airmass), which developed on 11∼12 December 1985 are studied in Part I of the present study. The fine structure of the polar low is analyzed in Part II. Radar and dense surface observation data indicate the formation of a meso-β-scale low associated by well organized spiral echo bands within the polar low. The horizontal extension and the life time of the meso-β-scale low is∼100km and several hours, respectively. While the polar low moves southeastward with speed of∼50km/hour, the meso-β-scale low moves eastward with speed of∼30km/hour. Strong gust of∼25m/sec and the surface pressure fall of∼2mb/30min are observed when the center of the meso-β-scale low at its mature stage passes over. The relative vorticity and convergence of the surface winds of∼100×10-6 sec-1 and∼-50×10-6 sec-1 are estimated over the central portion of the low. The thick moist neutral layer and the cyclonic flows induced by the polar low are considered to be a favorable environment for the developing of the meso-β-scale low. It is an important conclusion of Part II that the present polar low has a multi-scale structure, i. e., a meso-β-scale vortex in a meso-α-scale polar low.
Ozone concentration measured at two mountain locations in northern Japan (Teine Mountain in spring 1987 and Hakkoda Mountain in 1983-1987) is used to investigate the 'spring maximum' phenomena of lower tropospheric ozone. Ozone concentration at both mountain locations shows little diurnal variation. At Hakkoda Mountain, ozone concentration is at its maximum in spring. In April and May, the monthly mean concentrations are about 60ppb and 1-hour maximum concentrations are between 90 and 110ppb. In summer, the monthly mean ozone concentration decreases to the level that is observed in winter. In spring, elevated ozone concentrations over 80ppb are accompanied by high temperature under anticyclonic conditions. High positive correlation between ozone concentration and 850mb temperature and negative correlation between ozone concentration and 850 mb relative humidity are found in spring. Whereas, in summer, no relationship is found between ozone concentration and 850mb temperature. We considered that the positive correlation between ozone concentration and temperature in spring results from isentropic downward transport of ozone-rich air in the middle troposphere within a moving anticyclones which successively pass in spring in northern Japan.
A new approach is proposed to solve linear theory of conditionally unstable convection. In the traditional methods (Haque, 1952; Lilly, 1960; Kuo, 1961) the relationship among growth rate, width of the unstable region and static stability parameter is determined by first finding functional forms in unstably and stably stratified regions independently, and then imposing the continuity conditions for pressure and normal velocity at the boundaries between the two regions. In this study, the problem is considered to be a linear steady response of a stably stratified atmosphere to localized heating, where the growth rate is regarded as a Rayleigh damping/Newtonian cooling-type damping rate. This problem of forced motion is linked to the convection problem when the horizontal distribution of the heating rate is required to be proportional to that of the upward motion. Thus linear theory of conditionally unstable convection is reduced to a special case of the general problem of forced motion in a stably stratified atmosphere. From the standpoint of the forcing problem, the formation of the downward motion outside the unstable region is explained by the internal gravity waves propagating succesively in the stably stratified atmosphere.
Aircraft measurements of aerosol sizes, intensities of direct-solar and circumsolar (aureole) radiations, and upward and downward fluxes of solar radiation were carried out over Nagoya, a typical urban area in Japan, using an optical particle counter, an aureolemeter and spectral pyranometers, respectively. Vertical profiles of optical thicknesses and volume spectra of aerosols have been successfully retrieved by inverting measured aureole intensities. The obtained values have been utilized to estimate the absorption indices of aerosols from the downward flux measurements. The results are summarized as follows: 1) The concentration and the vertical stratification of tropospheric aerosols vary considerably day by day. 2) Bimodal volume spectra of aerosols with radii smaller and larger than r∼0.5μm generally prevail in the troposphere. The former is more predominant than the latter in the haze layer, and viceversa above that layer. 3) Estimated values of the imaginary index of refraction of tropospheric columnar aerosols are within the range of 0.005∼0.014 in the visible region and 0.008∼0.020 in the near infrared region. Corresponding values for the haze layer are slightly larger, i. e. 0.007∼0.018 and 0.011∼0.028 in the visible and near infrared regions, respectively.
In order to comprehend the Lagrangian-mean circulations and wave-mean flow interactions in the mid-latitude troposphere, the Eady baroclinic mode is examined using the transformed Eulerian-mean method in pressure-isentrope hybrid vertical coordinates (p†-TEM) proposed by Iwasaki (1989). The intersection of isentropes with an upper or lower boundary is shown to be essential for the formation of Lagrangian-mean circulation and wave-mean flow interactions of the Eady mode. The regions of isentropes intersecting the lower or upper boundary are called boundary isentrope layers. Based on the p†-TEM, the linear Eady mode forms a direct circulation with poleward and equatorward flows in upper and lower boundary isentrope layers, respectively. The Eliassen Palm (EP) flux is converted from the Coriolis acceleration of mean flows within the lower boundary layer and absorbed in the mean flows within the upper boundary layer. Taking into consideration the β effect (latitudinal variation of the Coriolis parameter), the poleward mean-meridional flows and EP flux convergence zones are expected to exist in the whole troposphere except for the lower boundary layer. These results account well for the p†-TEM analysis of a GCM (NCAR CCM1) run.
Weakly non-linear theory of steady hydrostatic mountain waves in 2-layered stratified Boussinesq fluid of infinite depth is presented. Weakly non-linear effects (second-order correction) on drag, downslope wind and the steepening or flattening of the streamline are examined, and are found to be very sensitive to the depth of the lower layer, D, ι2/l1 (ι1≡N1/U and ι1≡N1/U; N1 and N2: BruntVaisala frequencies of the lower layer and upper layer, respectively; U: horizontal wind) and terrain shape. Drag obtained from linear theory is invariant under the change of π in ι1D, while that obtained from weakly non-linear theory is no more invariant under the change of π in ι1D. The theory gives an estimate of the applicability range of linear theory. The theory is found to be in good agreement at least in a qualitative sense with non-linear numerical solutions for some cases.
By using a simple energy balance climate model and observed surface air temperature data, we investigate several internal and external causes of the globally or hemispherically averaged recent climatic change. Among the external causes, the increase of atmospheric CO2 concentration brings a warming trend of the surface air temperature, which may be detected in the next thirty years with a statistical confidence limit of more than 78%; the volcanic activity is responsible for the decade-to- decade temperature change and the effect of the anthropogenic aerosols may have contributed to the recent temperature changes. Also, we examine the effects of multiplicative and additive stochastic forcings as internal causes. Although Nicolis (1988) noted the increase of the temperature variance by the multiplicative stochastic forcing, we could not find such a monotonic increase in the variance of the temperature observed in the last hundred years. The additive stochastic forcing may contribute to a year-to-year temperature variation, and its magnitude corresponds to 0.22-0.28% of the input of the solar radiation at the top of the atmosphere.
An exact solution of the infrared radiative transfer through multi-homogeneous layers has been developed in the framework of a random band model for the Lorentz lines. Comparing with this solution the accuracy of the Curtis-Godson and Godson methods, which are currently used in calculating the radiative transfer through the inhomogeneous atmosphere, has been investigated. It has been shown that the error of the sequential procedure of Godson is variable depending on whether the transmittance is calculated upward or downward direction of the atmosphere. A new sequential procedure has been proposed using the analytical exact solution for two layers. It is shown that this new sequential procedure is of better accuracy than the methods of Curtis-Godson and Godson for the radiative transfer calculation of water vapor, carbon dioxide and ozone.