Mean radius (r) and moments around r, total concentration of droplets, liquid water content (W), terminal fall speed and visual range (Vm), have been computed for a sample of 239 droplet size distri-butions of valley, advection and radiation fog. The scattering, absorption and extinction coefficients, the albedo for single scattering, the phase function P(cosθ) at several scattering angles θ, the coeffi-cients χn, χm connected with the Legendre series expansion and the δ-M approximation of the phase function have also been computed at seventy-four wavelengths from 0.35 to 90μm. Data elaboration has basically consisted in the determination of the standard statistical parameters (histogram, mean, value, dispersion, skewness and kurtosis) for each of the above quantities and each of the various fog groups considered. Results concerning the grouping of fog spectra according to the fog type show that all the quantities are widely dispersed within each fog group and, in addition, the respective histograms partially overlap to each other; similar results also emerge by grouping the spectra according to their values of W and Vm, so that it appears unreliable to consider mean values as characteristic of a given fog group. Finally, a description of the mean physical and optical properties of the whole sample is presented, with the spectra all grouped together.
The wave propagation properties of short internal gravity waves (wave length λ_??_10km) are very different from those of long ones (λ-100km) owing to nonhydrostatic effects. Considering this observation, we parameterize the orographic gravity wave drag (GWD) in two ways. The major difference between two schemes is in the vertical partitioning of drag forcing, i, e., one weighs mainly in the stratosphere (type A) and the other in the troposphere (type B). We apply them to a global numerical weather prediction model and study their impacts on medium-range forecasts. These two schemes individually reduce systematic forecast errors and their combination achieves the best forecast skill. In the troposphere, the impacts of these two schemes, including time evolutions, are very similar to each other. Hence, the troposphere is considered to be insensitive to the vertical partitioning of GWD at least within a medium-range time scale. In the case of the type A scheme, most of the drag forcing given to the lower-stratospheric mean-flow is rapidly transferred downward and contributes to change in the tropospheric circulations. In the stratosphere, the impacts of the type A scheme is much larger than those of the type B, especially in a short-range time scale. The stratospheric impacts of the type B scheme gradually increase after a few days, possibly corresponding to a time scale required for the vertical propagation of Rossby waves excited in the lower troposphere. The improvement of forecasts is evident in the zonal-mean fields. Although the type A scheme almost eliminates the westerly errors in the stratosphere, there still remain westerly errors in the troposphere. The type B scheme effectively reduces the remaining tropospheric errors. This may justify a need of tropospheric drag forcing, like the type B scheme.
The effects of orographic gravity wave drag (GWD) on zonal-mean fields of medium-range forecasts are analyzed by means of the transformed Eulerian-mean (TEM) method. Results show that the geostrophic adjustment to GWD behaves very differently between the stratosphere and troposphere.In the troposphere, both the tropospheric and stratospheric GWDs significantly change Eliassen-Palm (EP) flux divergence due to large-scale (model-resolvable) waves. The change in the EP flux divergence is much larger than the net change of tonal wind and GWDs themselves and it is almost balanced with the change in the Coriolis acceleration term due to meridional flows. Among wave activities, transient gravity waves resolved in the model are considered to play important roles in the vertical redistribution of additional wave moments due to GWD. In the stratosphere, the stratospheric GWD induces a hemispheric single cell meridional circulation. The vertical motions in this cell cause significant temperature changes. The Coriolis acceleration due to GWD-induced meridional flows is almost balanced with the GWD itself. In contrast with the troposphere, EP flux divergence is less affected by GWD. From the view of Lagrangian-mean meridional circulation, diabatic heating in the tropics and wave, mean-flow interaction due to planetary-scale waves have been recognized mainly to drive a hemispheric single cell circulation (the so-called Brewer-Dobson circulation) in the lower-stratosphere. Our results indicate that the stratospheric GWD contributes to the maintenance of the single cell circulation as well. Especially in the midlatitudes of the northern hemisphere, the GWD can be regarded as an important forcing and considerably enhances lower-stratospheric downward motions in the polar side of the subtropical jet stream. It might affect significantly the transport of trace constituents in the stratosphere.
We have presented a numerical model of non-migrating tides assuming a differential heating local-ized only on land caused by upward heat flux in the planetary boundary layer. The amplitude of the diurnal temperature oscillation corresponding to the heat source is assumed to be 3K at the ground at the equator, and to decay exponentially with a scale height of 1km. Horizontal and vertical structures of non-migrating tides are solved by using classical tidal theory, in which a land-sea distribution is expanded into a complex Fourier series at each latitude with longitudinal wavenumbers ranging from -20 to 20, and further expressed by about 40 Hough modes for each zonal wavenumber. We have found that the horizontal distribution of the heat source has more significant effects on the behavior of non-migrating tides than the vertical distribution. The model pressure variations of non-migrating tides correlate well with the land-sea distribution at the ground, and agree excellently with observations. They are confined to a region equatorward of 30° latitude, while they spread longitudinally with increasing altitude, which is attributed to the imbalance between the westward and eastward propagating components. Horizontal wind velocities of non-migrating tides are about 2 to 3m/s near 20km altitude, and have vertical wavelengths as short as 2 to 5km, which is fairly consistent with ground-based radar observations.
The large, synoptic and meso scale features of the Baiu front that occurred during July 1982 are described, based mainly on GMS-IR digital data.The periods of synoptic and meso scales were determined by spectral analysis of the cloud amounts in the Baiu front.Attention was focused on three subjects:(1) the large-scale (5-day mean) features during the active phase of the Baiu front, (2) the difference of synoptic-scale (having a period of about six days) and mesoscale (a period of a few days to one day) variations in cloud features occurring over the continent (the western part of the Baiu front) from those occurring over the NW Pacific (the eastern part) and (3) the difference in frontal activity between 00GMT and 12GMT. During the active phase characterized by large cloud amounts in the 5-day mean, large-scale baro-clinicity, low-level southwest to west-southwest wind found south of the frontal zone, moisture flux convergence and the horizontal gradient of equivalent potential temperature (θe) in the lower layer of the frontal zone are strong, compared with those found during the inactive phase. The large-scale features around the frontal zone over the continent differ from those over the .NW Pacific.These regional differences in the large-scale features are reflected in the synoptic and meso scale cloud features in the frontal zone. Over the continent where the frontal zone is characterized by a strong low-level wind and an air mass of high θe, the synoptic-scale and mesoscale variations are prominent.The mesoscale cloud clusters that develop with a one-day period have diameters of -500km and consist of convective clouds. Over the continent, the one-day period variations of clouds are related to the underlying topography; mesoscale cloud clusters tend to develop in the basin areas around 00GMT and in the high altitude regions around 12GMT.There was hardly any meander of the frontal zone with a synoptic-scale period. Over the NW Pacific where the Baiu front is characterized by baroclinicity and an air mass of rather lower θe, the frontal zone meanders with a synoptic-scale period.The mesoscale cloud clusters that compose the synoptic-scale cloud systems are relatively large (-100Okm) and develop having periods of a few days.The mesoscale clusters consist of areas of narrow convective and wide stratiform clouds. In the frontal zone over Japan, both synoptic and meso scale variations show transitional features from those over the continent to those over the NW Pacific. Over the continent, the Baiu frontal cloud zone shifts north about 200km around 00GMT.The shift is associated with the increase in the low-level moisture flux south of the frontal zone, the increases in the low-level moisture flux convergence and middle-level moisture content in the frontal cloud zone. In Part II of this study, the structure of the Baiu front and features of sysnoptic and meso scale Baiu frontal disturbances will be clarified by analysis of relative vorticity fields. c 1989, Meteorological Society of Japan
The features of polar/comma-cloud lows in the vicinity of the Japan Islands in winter 1986/87 (December, January and February) are studied. In the present paper, the polar/comma-cloud low (abbreviated as p/c low) is defined as a mesa α-scale low that forms in polar air stream north of the main polar frontal zone, and forms spiral or comma shaped cloud system of 200-700km extent in its formation and mature stages. This study shows the following features; The p/c lows develop most frequently in midwinter over the Japan Sea (45-35°N) and the northwestern Pacific Ocean (55-40°N), where strong low-level baroclinicity is maintained by surface fluxes in the polar airmass between the continent area and the relatively warm sea area. They rarely appear over the Asian Continent, the Yellow Sea and the East China Sea. The p/c lows form in the low-level cyclonic flow of polar air induced by developed synoptic-scale lows. The p/c lows develop in the vicinity of meso-α-scale cold vortices in the middle troposphere embedded in the synoptic-scale trough or cut-off low. The p/c lows develop in a zone 500-1000km north of the major polar frontal zone, where the baroclinicity in the lower'middle troposphere is strong. The characteristics of p/c lows in these regions are compared with characteristics of p/c lows in the northeastern Atlantic Ocean and the northeastern Pacific Ocean.
Observations of the strong and persistent glacier wind were made on the 40km-long San Rafael Glacier (46°41'S, 73°51'W) in the Patagonia Northern Icefield, Southern Chile, South America. From observations near the glacier terminus, glacier wind characteristics in the warm summer season were revealed to be as follows. The wind blows at a frequency of 80 to 90% during the summer season. In the strongest and also most frequent case, the thickness is more than 100m and maximum wind speed is 5m/s . The strong and persistent glacier wind is due to the large scale of this glacier. The main regulating factor for the day to day variation in the occurrence of the glacier wind is the upper air wind speed. When the upper wind is strong, the glacier wind is suppressed and the depth of the glacier wind is shallow. The factor determining the diurnal variation of this wind is the temperature of the ambient air outside the influence of the glacier. There exists a periodicity of 1 to 3 hours in the wind speed of the glacier wind on developed days. The continuance of this wind system after it leaves the glacier is limited to a short distance. Analyzing wind data on glaciers in various regions of the earth, glacier size seems to affect the surface wind speed, probably due to the existence of these glacier winds.
Theoretical consideration of the katabatic wind occurring on melting snow and ice surfaces, which is usually called a glacier wind, is made and is compared with observational results. The source of cooling is the constant 0°C surface; this is different from the usual katabatic wind which occurs as a result of radiation deficit at the surface. The katabatic wind model of Manins and Sawford (1979), which is a two-layer model incorporating entrainment at the top of the katabatic layer, will be applied to glacier wind. The needed parameters are simplified from the result of measurements of vertical profile of air temperature and wind speed in the present case of katabatic wind on melting snow and ice surfaces. The calculation result shows that on relatively large snow and ice masses, wind speed and thickness are∝s, and sensible heat flux at the surface ∝s-s2 where s is the distance from the upper end of the snow or ice mass. This is different from the result obtained for katabatic wind occurring from a radiational deficit, for which the wind speed is usually ∝√s and thickness is ∝s. The effect of stability of the ambient atmosphere is such that when the stability is low, wind develops faster along the slope. However, there is no destruction of wind due to adiabatic warming of the katabatic wind layer. This modified model can explain the observed wind speed, depth and surface sensible heat flux at a certain site on snow and ice masses quantitatively and can explain the areal development qualitatively.
The influence of changing soil moisture and surface albedo on climate is studied with an atmospheric general circulation model (MRI•GCM-I). In a control run (C run), a standard bucket model with a spatially uniform water holding capacity of 15g cm-2 is used for the ground hydrology. The surface albedo over bare land is specified as 0.14 and that over snow-covered land varies from 0.7 to 0.85 depending on the height of the ground surface. In the second run (AW run), the surface albedo and ground wetness are specified as the climate values. In the AW run, there is no snow-albedo feedback and the ground wetness is not predicted. Two additional runs are performed for separating the effect of albedo from that of ground wetness and for examining the effect of snow albedo. The results show that the atmospheric circulation in the Northern Hemisphere is very sensitive to the surface albedo specification in winter. In the C run, the Siberian high is stronger and extended southeastward compared with that in the AW run due to snow-albedo feedback in C. The Aleutian low in the C run is deeper and shifts eastward. On the other hand, the Icelandic low is deeper in the AW run than in the C run. It is found that snow albedo is an important factor which controls the northern summer climate. When snow albedo is low, snow melts earlier in spring and land becomes dry and warm in summer over the Eurasian arid region. Thus the summer precipitation is reduced there. Over the southeast Asia summer monsoon region in the model, however, moisture flux convergence is enhanced and the summer precipitation is increased. The increased precipitation makes soil wet, which in turn increases precipitation. In short, snow albedo affects summer climate significantly through the interaction of atmosphere and ground hydrological processes. Albedo of snow-free land also has a significant effect on climate particularly in low latitudes. High/low albedo produces less/much precipitation.
The chemical composition and physical state of individual aerosol particles were examined for the aerosol collected on board an aircraft. The collection of aerosol particles was carried out at altitudes between 2.0 and 7.6km including the vicinity of tropopause folding and above tropopause over the Japan Sea on 14 February, 1984. The molecular state of sulfate particles was determined by vapor-deposited calcium thin-film method (pre-coating) in order to avoid the possible chemical modification of samples by ammonia after collection. Nitrate and sulfate in particles were detected by use of the vapor-deposited nitron thin-film method and barium chloride thin-film method, respectively. Whether stratospheric aerosol particles are frozen or supercooled was examined by comparing the morphology of collected particles with that of frozen sulfuric acid particles produced in the laboratory. Sulfuric acid particles were predominant in the stratosphere without any serious ammoniation. These sulfuric acid particles are much larger than those in the lower troposphere. No nitrate-containing particles and few particles of tropospheric origin were found in the stratosphere. The fact that nitrate can not be detected in the particles suggests that most of the nitric acid is not dissolved in particles but exists in the gas phase in the stratosphere. Stratospheric sulfuric acid particles were not frozen, but supercooled even at -48°C. Stratospheric aerosols were constituted of sulfuric acid particles, of which the chemical composition was much more uniform in space than that of tropospheric aerosols. We can safely state that supercooled sulfuric acid particles are predominant in the mid-latitude stratosphere. On the other hand, aerosols of several different origins coexisted in the troposphere. Sulfuric acid particles were mostly predominant in the middle and upper troposphere, although their ammoniated particles also existed. Acidity of the sulfate-containing particles increased with increasing altitude between 2.0 and 5.5km. It was observed that stratospheric sulfuric acid particles coexisted withparticles of lower tropospheric origin (probably ammonium sulfate particles) in the tropopause folding. The observation was carried out during the period when there was a slight possibility that the effect of volcanic eruption of E1 chichon in 1982 on the stratospheric aerosol layer still remained. However, it is not possible to determine the degree of any volcanic effect from the present study.