A numerical experiment of conditional symmetric baroclinic instability (CSBI) is carried out including Kessler's parameterization, in order to show a possible mechanism for formation of meso-β scale rainbands associated with extratropical cyclones and Baiu front. The unstable condition of CSBI is controlled by Richardson number defined by equivalent potential temperature. The flow is unstable for CSBI when the equivalent Richardson number is less than unity. In the numerical model, horizontal and vertical space resolutions are 2 and 0.5km, respectively, in a 250km×10km domain. We assumed an initial zonal flow with vertical shear, which is geostrophically balanced with the meridional gradient of virtual potential temperature. All the derivatives with respect to zonal direction vanish due to zonal symmetric assumption. An unstable condition for CSBI is initially set everywhere in the domain, adjusting the water vapor content included. To set on CSBI, a few percents of supersaturation is initially incorporated at the center portion of the domain together with initial random vertical velocity disturbances. A typical slantwise circulation of CSBI is seen with horizonal scale of about 150km independent of the initial supersaturation area. The total meridional heat (equivalent potential temperature) transport is northward. The total meridional transport of zonal momentum is not clear from the present experiment. The total transport of heat is upward, while the total zonal momentum transport is downward. A slantwise stratiform cloud appears and is sustained along the updraft of CSBI. A mid-tropospheric convective unstable layer is induced above the stratiform cloud due to the differential advection of equivalent potential temperature by the CSBI circulation. Convective cells appear in this layer.Thus, CSBI prepares the favorable environment for the seeder-feeder precipitation mchanism to work efficiently. At the top of the boundary layer, upward velocity exceeds 0.02ms-1 due to the convergence induced by the CSBI, horizontally extending over 30km and lasting for more than 15hrs. A low level jet is formed due mainly to the deceleration of zonal wind at the mid-troposphere, together with a slight acceleration below.
The behaviour of the stationary and travelling planetary-scale waves during northern hemisphere blocking has been investigated. Data from 29 winter seasons from 1950 to 1979 were used. A winter season was defined as all days from 1 December through 28 February. The data were decomposed in a quasi-stationary part, computed as a running time average, and a travelling part, being the difference between the observed value and the time average. For each latitude line, the geopotential height of the first three waves (quasi-stationary as well as travelling) were computed as a function of longitude and time. We have focused our interest on the total contribution from the planetary-scale (the first three) quasi-stationary versus travelling waves. The behaviour of the planetary-scale waves during two major blocking situations is documented. Characteristics of these waves during blocked flow have been computed for the 29 winter seasons. It is found that the stationary component of the height field for the first three waves is almost always greater than the corresponding climatological value in areas with blocked flow, and that the resultant of the height field of the travelling ultra long waves is mostly greater than zero in the same area. These two conditions are almost necessary for the formation of blocks, but they are not sufficient.
Relationships between midlatitude cold surges and tropical convections are investigated utilizing 30-60 day filtered 850mb meridional winds and outgoing longwave radiation (OLR) data during the three Northern Hemisphere winters of 1980-83. The 30-60 day southerly surges over the eastern Indian Ocean off the west Australian coast act as the trigger and intensifier of low-frequency convective systems which systematically propagate eastward across the equatorial Indian Ocean and the western Pacific. In comparison, the northerly surges from the subtropical western Pacific (east of Philippines) tend to enhance equatorial convection between about 150°E and the date line, which is about 20 to 50 degrees of longitude downstream of the northerly blowing longitudes (120°-130°E).
According to the observation data from 1975 to 1979, the "structure change" in the Baiu front in China occurs in late May in each year: the vanishment of the horizontal temperature gradient in the lower layer and the change in the stability for moist convection from stable state to nearly neutral state. The ground surface temperature in North China increases rapidly and its horizontal gradient along -115°E vanishes in China around May in these years. In late May of 1979 the air temperature in the lower layer rises largely in the semiarid region (Takla Makan-Mongol Plateau*Gobi Desert-North China) and the polar frontal zone in the eastern part of Eurasia (90°E-140°E) shifts from -45°N to -60°N. The area where the deep mixed layer due to dry convection with a thickness of more than 200mb (2000m) begins to develop rapidly in late April over the semiarid region from Takla Makan to North China, a month before the "structure change" in the Baiu front in China. The heat budget in the atmosphere over Mongol Plateau suggests that the air temperature rise and the maintenance of high temperature in the lower layer in May are mainly due to the sensible heat supplied from the ground as well as in North China. The intense airmass transformation over that region warms the air parcel to some extent before it reaches North China. This seems to provide the more favorable condition for the change in the thermal structure in the Baiu front in South China. The"structure change" in the Baiu front in China and the northward shift of the polar front in the eastern part of Eurasia seems to be a response to the heating from the ground in the semiarid region from Takla Makan to North China.
The regional variations of the spectra of the surface winds in Japan are investigated from the viewpoint of spectrum climatology. For the 206 stations in Japan, the spectra of the surface winds over the period from 2 hours to a few years are analyzed using 5 years of 10min averaged, hourly data. The diurnal variation and synoptic fluctuations depend on the local topography. Their geographical distributions show a regional contrast. The diurnal variation is dominant at the inland stations. On the other hand, the synoptic fluctuations are dominant at the coastal stations. The 10-20 day fluctuations are systematically dominant on the Nansei Islands (24-31°N, 123-130°E). The geographical distribution of the 10-20 day fluctuations coincides with that of the 40-60 day fluctuations and that of the zonal components of the surface winds. This feature is related to the climate on the Nansei Islands, which is subtropical and maritime. The geographical distribution of the annual variation does not show a clear pattern, except for the Nansei Islands where the annual variation is small. This is because the annual variation depends both on the local topography and latitude. Models of the spectrum of the surface winds accounting for the relationship between the mean wind speed and the latitudinal distribution are derived from the results. The models show that the short-term fluctuations (less than 10 days) depend on the mean wind speed, that is, the local topography and that the long-term fluctuations (more than 10 days) are affected by the latitude. Characteristics of the spectrum can be used as a climatic indicator. The models may have application to the same climatic regions as Japan.
Based on the statistical consideration for air-sea sensible and latent heat and momentum fluxes estimated from the data obtained at OWS-T (29°N, 135°E) from June 1950 to November 1953, the relationship between the standard error of monthly mean fluxes and the number of the data used in estimation was examined. The standard errors varying seasonally for the number of the data were presented, which were given by the statistical theory, i.e., the standard deviation of the population divided by the square root of the number of the sampled data. These results were also confirmed by the computer simulation for randomly sampled and small-in-number data, and they holded even though the distribtuion of the population deviates from the normal one. From the actual number of the marine meteorological data compiled by the Japan Meteorological Agency from 1961 to 1985, it could be expected that in the sea south of Japan, momentum fluxes can be estimated with accuracies of 0.15 and 0.1 dyn⋅cm-2 in winter and summer, respectively, the sensible heat fluxes 5 and 1 W⋅m-2 and the latent heat fluxes 15 and 5 W⋅m-2.
The origins of lightning are determined through a thunderstorm electricity model (Takahashi, 1984) with inclusion of a discharge process. Lightning origins are found to vary greatly with microphysical parameters. The calculations successfully simulate the following: the initial appearance of lightning at the cloud top followed by low level lightning; two layers of origin in maritime thunderstorm lightning as opposed to only one layer of origin in continental thunderstorms; the maintenance of a negative charge source at -10°C and shallow level lightning due to interactions between the main cloud cell and the induced second cloud cell in maritime thunderstorms.
An ethylene-ozone Chemiluminescent instrument has been modified so that it can control the sample and ethylene flow rates irrespective of the ambient pressure and temperature. Through detailed laboratory tests the following characteristics have been found. (1) The sensitivity of the instrument is highly dependent on the sample mass flow rate as ex- pected from the chemiluminescent reaction scheme. (2) The change in sensitivity as a function of the ethylene mass flow rate is much less. (3) The error associated with the stability of the zero level and the standard deviation from a linear relationship is less than 1 ppbv. (4) The change in the sensitivity for pressures between 1000 and 500mb is insignificant. A larger reduction in sensitivity at lower pressures necessitates corrections based on the calibration curve obtained. It amounts to about 10% at 400mb under optimum flow conditions. Aircraft observations of ozone up to the altitude of 6km have been made. The flow control system has proved to maintain the flow rates constant to within 3% on board the aircraft. Ozone mixing ratios obtained by this experiment are also reported.
A special sonde for measuring cloud particles has been developed. The sonde has a small TV camera, and is called a Cloud Particle Video Sonde (CPVS). The CPVS transmits images of cloud particles from 7μm to 1000μm diameter over a 1.6 GHz microwave link to a ground station. Observations have shown that the CPVS provides highly quality images of cloud particles at altitudes up to 12km MSL. Using the collection efficiency of the CPVS for water droplets determined from a laboratory experiment, the number concentration of cloud particles and the variation of their size distribution with height can be calculated.