高村 民雄

The Changbai Mountain Natural Reserve (2000 km(2)), north-east China, is a very important ecosystem representing the temperate biosphere. The cover types were derived by using multitemporal Landsat TM imagery, which was modified with DEM data on the relationship between vegetation distribution and elevation. It was classified into 20 groups by supervised classification. By comparing the results of the classification of different band combinations, bands 4 and 5 of an image from 18 July 1997 and band 3 of an image from 22 October 1997 were used to make a false colour image for the final output, a vegetation map, which showed the best in terms of classification accuracy. The overall accuracy by individual images was less than 70%, while that of the multitemporal classification was higher than 80%. Further, on the basis of the relationship of vegetation distribution and elevation, the accuracy of multitemporal classification was raised from 85.8 to 89.5% by using DEM. Bands 4 and 5 showed a high ability for discriminating cover types. Images acquired in late spring and midsummer were recognized better than other seasons for cover type identification. NDVI and band ratio of B4/B3 proved useful for cover type discrimination, but were not superior to the original spectral bands. Other band ratios like B5/B4 and B7/B5 were less important for improving classification accuracy. The changes of spectral reflectance and NDVI with season were also analysed with 10 images ranging from 1984 to 1997. Seperability of images in terms of classification accuracy was high in late spring and summer, and decreased towards winter. There were five vegetation zones on the mountain, from the base to the peak: deciduous forest zone, mixed forest zone, conifer forest zone, birch forest zone and tundra zone. Spruce-fir conifer dominated forest was the most dominant vegetation (33%), followed by mixed forest (26%), Korean pine forest (8%) and mountain birch forest (5%).

The Changbai Mountain Natural Reserve (2000 km(2)), north-east China, is a very important ecosystem representing the temperate biosphere. The cover types were derived by using multitemporal Landsat TM imagery, which was modified with DEM data on the relationship between vegetation distribution and elevation. It was classified into 20 groups by supervised classification. By comparing the results of the classification of different band combinations, bands 4 and 5 of an image from 18 July 1997 and band 3 of an image from 22 October 1997 were used to make a false colour image for the final output, a vegetation map, which showed the best in terms of classification accuracy. The overall accuracy by individual images was less than 70%, while that of the multitemporal classification was higher than 80%. Further, on the basis of the relationship of vegetation distribution and elevation, the accuracy of multitemporal classification was raised from 85.8 to 89.5% by using DEM. Bands 4 and 5 showed a high ability for discriminating cover types. Images acquired in late spring and midsummer were recognized better than other seasons for cover type identification. NDVI and band ratio of B4/B3 proved useful for cover type discrimination, but were not superior to the original spectral bands. Other band ratios like B5/B4 and B7/B5 were less important for improving classification accuracy. The changes of spectral reflectance and NDVI with season were also analysed with 10 images ranging from 1984 to 1997. Seperability of images in terms of classification accuracy was high in late spring and summer, and decreased towards winter. There were five vegetation zones on the mountain, from the base to the peak: deciduous forest zone, mixed forest zone, conifer forest zone, birch forest zone and tundra zone. Spruce-fir conifer dominated forest was the most dominant vegetation (33%), followed by mixed forest (26%), Korean pine forest (8%) and mountain birch forest (5%).

[1] A Mie lidar was used to make observations of Asian dust over Hefei (31.90degreesN, 117.16degreesE) in spring 2000. This paper presents main features of vertical distribution and temporal variation of Asian dust extinction coefficient at 532-nm wavelength. It was found that the Asian dust events contributed significantly to very large aerosol extinction coefficients in the boundary layer or middle troposphere. Extinction coefficient value as large as 0.7 km(-1) at 3 km above ground level (AGL) has been observed. The Asian dust extinction coefficient profiles showed very big changes from early morning to evening and also appeared substantial amount of temporal variation during the nighttime. For the severe Asian dust days the depth of boundary layer rose up to around 4 km AGL, but only 1similar to2 km AGL for normal days. The seasonal average aerosol extinction coefficient profiles also showed that larger aerosol extinction coefficients from 1- to 10-km altitude range were observed in the springtime rather than in any other season. Meanwhile, solar extinction and radiation measurements indicated that the Asian dust particles were composed of larger size particles and they produced significantly perturbation in solar radiation on the Earth surface.

[1] A Mie lidar was used to make observations of Asian dust over Hefei (31.90degreesN, 117.16degreesE) in spring 2000. This paper presents main features of vertical distribution and temporal variation of Asian dust extinction coefficient at 532-nm wavelength. It was found that the Asian dust events contributed significantly to very large aerosol extinction coefficients in the boundary layer or middle troposphere. Extinction coefficient value as large as 0.7 km(-1) at 3 km above ground level (AGL) has been observed. The Asian dust extinction coefficient profiles showed very big changes from early morning to evening and also appeared substantial amount of temporal variation during the nighttime. For the severe Asian dust days the depth of boundary layer rose up to around 4 km AGL, but only 1similar to2 km AGL for normal days. The seasonal average aerosol extinction coefficient profiles also showed that larger aerosol extinction coefficients from 1- to 10-km altitude range were observed in the springtime rather than in any other season. Meanwhile, solar extinction and radiation measurements indicated that the Asian dust particles were composed of larger size particles and they produced significantly perturbation in solar radiation on the Earth surface.

This article discusses common features of photosynthetically active radiation (PAR) in the Mongolian grain farm region during the period of 1986 to 1996. The quantitative characteristics of solar radiation, such as direct, diffuse and global radiation are evaluated in the study through the long term trend. The mean pattern of seasonal variation of PAR shows that the yearly minimum global (direct+ diffuse) PAR is 68.9±7.6MJ/(m2 month)in December while the maximum is 307.5±10.3MJ/(m2 month) in May. The yearly summation is 2.22±0.07GJ/(m2yr)and is very stable compared with monthly variations. © 2001, The Society of Agricultural Meteorology of Japan. All rights reserved.

We propose a method to determine the extinction-to-backscattering ratio (S1 parameter) in the troposphere from multiwavelength (355, 532, 756, and 1064 nm) lidar data. In our approach, reference profiles are prepared by using the wavelength dependence of the extinction coefficient as derived either from the sun photometer data or from the Mie calculation. By comparing these reference profiles with the profiles calculated using the conventional Fernald inversion method, the S1 parameter is determined for each wavelength. When a reasonable range is covered at 532 nm, this method makes it possible to determine the S1 parameters for shorter or longer wavelengths for which full range observation cannot be attained, due presumably to small laser power or limited detector efficiency. In addition, information about the vertical uniformity of aerosol properties can be derived from the S1 dependence of the difference between the reference and retrieved profiles.

Ground-based, optical monitoring of the NO2 column density and aerosol optical thickness is described. The instrument consists of a solar radiation spectrometer and a conventional sunphotometer, both mounted on a sun-tracker and operated automatically. From daytime measurements in Chiba during the winter of 1998, variations of NO2 and aerosol are analyzed. Because of the capability of simultaneous, real time measurement, this method is particularly suitable for air pollution studies in city areas.

Ground-based, optical monitoring of the NO2 column density and aerosol optical thickness is described. The instrument consists of a solar radiation spectrometer and a conventional sunphotometer, both mounted on a sun-tracker and operated automatically. From daytime measurements in Chiba during the winter of 1998, variations of NO2 and aerosol are analyzed. Because of the capability of simultaneous, real time measurement, this method is particularly suitable for air pollution studies in city areas.

Vertical profiles of backscattering coefficients, optical thicknesses, and columnar size distributions of aerosols were obtained by simultaneous measurements with lidar, a sunphotometer, and an aureolemeter in Tsukuba, Japan, from November 1991 to December 1992. Several conspicuous characteristics were found in the relationship between aerosol size distribution and stratification. In summer an accumulation mode is dominant, and aerosols were heavily loaded in the planetary boundary layer. Turbid atmospheres with an abundance of large particles are observed in the middle troposphere in the spring. In autumn and winter the troposphere is clear so that columnar aerosol size distributions reflect stratospheric aerosols. During the observation period, volcanic aerosols that are due to the Mt. Pinatubo eruption were being loaded in the stratosphere. The mode radius in the volume size distribution of the stratospheric aerosol was observed to increase from 0.45 mu m in November 1991 to 0.6 mu m in October 1992, and decreased after October 1992. Total aerosol loading in the stratosphere was estimated to be maximum in the spring of 1992, minimum in the autumn of 1992, and increased again after the autumn of 1992. (C) 1998 Optical Society of America.

Vertical profiles of backscattering coefficients, optical thicknesses, and columnar size distributions of aerosols were obtained by simultaneous measurements with lidar, a sunphotometer, and an aureolemeter in Tsukuba, Japan, from November 1991 to December 1992. Several conspicuous characteristics were found in the relationship between aerosol size distribution and stratification. In summer an accumulation mode is dominant, and aerosols were heavily loaded in the planetary boundary layer. Turbid atmospheres with an abundance of large particles are observed in the middle troposphere in the spring. In autumn and winter the troposphere is clear so that columnar aerosol size distributions reflect stratospheric aerosols. During the observation period, volcanic aerosols that are due to the Mt. Pinatubo eruption were being loaded in the stratosphere. The mode radius in the volume size distribution of the stratospheric aerosol was observed to increase from 0.45 mu m in November 1991 to 0.6 mu m in October 1992, and decreased after October 1992. Total aerosol loading in the stratosphere was estimated to be maximum in the spring of 1992, minimum in the autumn of 1992, and increased again after the autumn of 1992. (C) 1998 Optical Society of America.

Tropospheric aerosols have been observed for the period from November 1990 to April 1992 with a lidar, a sun photometer, and an optical particle counter. Variations of aerosol optical thickness derived from the lidar and the sun photometer data and measurements are presented. The simultaneous measurements of these instruments also allowed us to estimate the extinction-to-backscatter ratio (S1), which ranged from 20 to 70. Comparison of optical thicknesses derived from both instruments clearly shows the effect of Mt. Pinatubo's eruption and the temporal variation of optical thickness in the stratosphere over 12 km. The possible range of the complex refractive index for the columnar mean aerosols can be deduced from the probable range of S1 derived by the use of an S1 diagram as a function of complex refractive index (m). The imaginary part of m can be estimated provided that the real part of m is known.

The longwave upward radiation was calculated for an urban canopy by using a Monte Carlo model. The effects of the urban geometry were examined in terms of the fractional roof area, the height of the buildings and the emissivity. The urban canopy consists of identically sized buildings and the ground surfaces. The model allows for the temperature differences between the buildings and the ground surface and for multiple reflections in the canyon. The Monte Carlo results show that neglect of the geometric effects causes significant errors in calculated upward radiation: calculations with area-weighting of the radiation emitted from flat homogeneous surfaces are not appropriate. The upward flux is a nonlinear function of the fractional roof area, which may be approximated by a function of the square or cube of the fractional roof area. Neglect of the reflections by non-black surfaces (emissivity < 1) underestimates the upward flux by a few percent for a canopy of emissivity = 0.9. Radiation effects due to multiple reflections in the canyon are parameterized by use of the view factor and the fractional roof area. The parameterization scheme yields accurate results.

The spectral reflectance of the surface in an urbanized area was estimated through airborne measurements of the spectral upward flux of visible radiation in the range 475-750 nm. Atmospheric effects due to Rayleigh and Mie scattering were accounted for by using optical parameters to solve the radiative transfer equation. The values for these parameters were derived from measurements of the particle number concentration and size distribution. The results clearly show a difference in reflectance between urban and suburban areas. The difference in spectral reflectance decreases from the suburban to the urban area. In a metropolitan area, the surface reflectance generally decreases with urban development, and the global upward flux of visible radiation has a similar tendency. This trend supports the idea of a decrease in reflectance due to the modification of the surface structure.

The spectral reflectance of the surface in an urbanized area was estimated through airborne measurements of the spectral upward flux of visible radiation in the range 475-750 nm. Atmospheric effects due to Rayleigh and Mie scattering were accounted for by using optical parameters to solve the radiative transfer equation. The values for these parameters were derived from measurements of the particle number concentration and size distribution. The results clearly show a difference in reflectance between urban and suburban areas. The difference in spectral reflectance decreases from the suburban to the urban area. In a metropolitan area, the surface reflectance generally decreases with urban development, and the global upward flux of visible radiation has a similar tendency. This trend supports the idea of a decrease in reflectance due to the modification of the surface structure. © 1992 Kluwer Academic Publishers.

Simultaneous measurements of aerosols by use of an airborne optical particle counter (OPC) and a ground-based lidar were performed during January, 1986, over the Tsukuba area. The measurements were taken over the middle and lower troposphere in order to obtain both spatial and size distributions, to infer an extinction-to-backscatter ratio (S1) and a complex index of refraction (m). Bimodal distributions were commonly found in the aerosol volume size distributions from the OPC measurements over the flight range (310-4270 m). Although the spectral shape of the accumulation mode exhibited a slight difference between altitudes, it was reasonable to use a constant S1 estimated from the aerosol size distribution near the surface in analyzing the lidar signal. This was true except for the cases when a heavy dust layer was found in the higher troposphere, which may have originated from a Kosa (Asian dust) event. Assuming spatial homogeneity of aerosol optical properties, the mean value of S1 and its range of variation was inferred from the lidar signal along with additional information on optical thickness. Moreover, the tropospheric aerosol refractive index was estimated from the relationship between S1 and m based on the OPC data. In this experiment, two heavy-dust layers were found, one aloft near a height of 5000 m and the other just above the ground surface. In order to determine the ranges of optical parameters for the aerosols in both layers, it was necessary to treat each level separately. Since no in situ data was available for the upper dust layer, S1 for the upper layer was assumed while the data was analyzed for the lower layer to estimate the range of S1 there. The mean extinction-to-backscatter ratio and the imaginary part of the complex refractive index for the total air column of the lower dust layer (less than 4300 m) were estimated respectively as ranging from 32 to 66, and less than 0.04, assuming the real part as 1.55+/-0.03.