Search

In studies of global warming, increases in tropospheric temperature as well as increases in surface temperature have attracted attention. Simulations of trends in these two temperatures appear to differ from trends in observations by surface sites, radiosondes, and satellites. Moreover, observation errors such as uncertainties in measurement precision and calibration, environmental changes, and the reorganization of network sites hamper the ability to quantify these influences on temperature trends. This paper presents multidecadal (1965–2016) trends in lower-tropospheric temperature for south-central Japan derived from thickness temperature, a measurement based on pressure data from exactly known altitudes at the summit of Mount Fuji (3776 m) and surrounding meteorological sites. The resulting trend is compared with the trends in surface temperature and in the temperature at 700 hPa measured by radiosonde. Although surface temperature increased faster than tropospheric (thickness) temperature in the study area for the 1965–2016 period, tropospheric temperature increased faster than surface temperature after 1985. Additionally, it is found that radiosonde data are not appropriate for determining the temperature trend at constant altitudes because atmospheric warming raises the altitude of the pressure levels.

Individual aerosol particles were collected in the middle troposphere at 5–6 km altitude in the PACE-III (Pacific Atmospheric Chemistry Experiment) campaign, conducted between Melbourne in Australia (38°S) and Sendai in Japan (38°N) in July 1995. Particles collected on electron microscope grids were subsequently examined using a transmission electron microscope equipped with an energy-dispersive X-ray (EDX) analyzer. Based on the EDX analysis and morphological identification, particles were classified into seven categories, including sulfuric acid, sulfate, sea-salt and mineral particles. Within the radius range of 0.1–0.5 μm, S-rich particles (sulfuric acid and sulfate) were dominant (> 60%) in the observation regions. These particles are considered an important background aerosol component. The latitudinal distributions of S-rich particles containing potassium (K) and minerals were studied for all S-rich particles of 0.1–0.5 μm. The percentages of K-containing S-rich particles were > 20% north of 35°N and south of 15 S. S-rich particles with K were likely to have originated from biomass or biofuel burning on the continent and were transported through the middle troposphere. In addition, S-rich particles with minerals were often present in relatively high percentages (> 40%) in the northern mid-latitudes. The present study suggests that 30–70% of S-rich particles were influenced by material of surface origin north of approximately 30°N and south of 25°S, where westerly winds were dominant. The results obtained in this campaign were compared with those in the previous PACE-I campaign (January 1994). The comparison suggested that seasonal shifts of the locations of the ITCZ and wind belts play an important role in the formation of the seasonal difference in the spatial distributions of concentration and mixing properties of S-rich particles in the middle troposphere over the western Pacific.

The Aeolian Dust Experiment on Climate Impact (ADEC) was initiated in April 2000 as a joint five-year Japan–China project. The goal was to understand the impact of aeolian dust on climate via radiative forcing (RF). Field experiments and numerical simulations were conducted from the source regions in northwestern China to the downwind region in Japan in order to understand wind erosion processes temporal and spatial distribution of dust during their long-range transportation chemical, physical, and optical properties of dust and the direct effect of radiative forcing due to dust. For this, three intensive observation periods (IOP) were conducted from April 2002 to April 2004. The in situ and network observation results are summarized as follows: (1) In situ observations of the wind erosion process revealed that the vertical profile of moving sand has a clear size dependency with height and saltation flux and that threshold wind velocity is dependent on soil moisture. Results also demonstrated that saltation flux is strongly dependent on the parent soil size distribution of the desert surface. (2) Both lidar observations and model simulations revealed a multiple dust layer in East Asia. A numerical simulation of a chemical transport model, CFORS, illustrated the elevated dust layer from the Taklimakan Desert and the lower dust layer from the Gobi Desert. The global-scale dust model, MASINGAR, also simulated the dust layer in the middle to upper free troposphere in East Asia, which originated from North Africa and the Middle East during a dust storm in March 2003. Raman lidar observations at Tsukuba, Japan, found the ice cloud associated with the dust layer at an altitude of 6 to 9 km. Analysis from lidar and the radio-sonde observation suggested that the Asian dust acted as ice nuclei at the ice-saturated region. These results suggest the importance of dust's climate impact via the indirect effect of radiative forcing due to the activation of dust into ice nuclei. (3) Studies on the aerosol concentration indicated that size distributions of aerosols in downwind regions have bimodal peaks. One peak was in the submicron range and the other in the supermicron range. The main soluble components of the supermicron peak were Na+, Ca2+, NO3−, and Cl−. In the downwind region in Japan, the dust, sea salt, and a mixture of the two were found to be dominant in coarse particles in the mixed boundary layer. (4) Observation of the optical properties of dust by sky-radiometer, particle shoot absorption photometer (PSAP), and Nephelometer indicated that unpolluted dust at source region has a weaker absorption than originally believed. A sensitivity experiment of direct RF by dust indicated that single scattering albedo is the most important of the optical properties of dust and that the sensitivity of instantaneous RF in the shortwave region at the top of the atmosphere to the refractive index strongly depends on surface albedo. A global scale dust model, MASINGAR, was used for evaluation of direct RF due to dust. The results indicated the global mean RF at the top and the bottom of the atmosphere were − 0.46 and − 2.13 W m− 2 with cloud and were almost half of the RF with cloud-free condition.

Aircraft observations were carried out in the middle troposphere at 5–6 km altitude over Australia in the PACE-II (Pacific Atmospheric Chemistry Experiment) campaign in October 1994. Concentrations of particles with radii larger than 4 nm and 0.15 μm were measured with a condensation nucleus (CN) counter and an optical counter, respectively. Individual aerosol particles were collected on electron microscopic grids using a jet impactor and subsequently examined using a transmission electron microscope coupled with an energy-dispersive X-ray (EDX) analyzer. High concentrations of CN (⩾4 nm) and large particles (0.15–0.5 μm radius) were found over the areas in northern Australia between Alice Springs and Darwin. Samples collected in this region had a large fraction of submicrometer soot-containing particles that had both K and S with a median S/(S+K) weight ratio of 0.4. The hygroscopic properties of particles with radii between 0.03 and 0.8 μm were examined with the dialysis (extraction) of water-soluble material. The sample collected in the area with high particle concentrations showed that the abundance of internally mixed soot particles was 88% of particles in the radius range of 0.1–0.35 μm. Approximately 50% of these mixed particles with radii ⩾0.2 μm were covered with semi-transparent material (probably organic material). Present research suggested that these particles were aged biomass burning particles originating from northern Australia and had been modified by chemical reactions and probably by coagulation during long-range transport. These soot particles were found to be present as internally mixed particles, with a soluble fraction that would make them efficient cloud condensation nuclei (CCN).

According to records of meteorological stations of the Japan Meteorological Agency (JMA), the occurrence of Kosa dust events increased during 2000-2002. This was verified by the analyses of aerosol optical depth at 500 nm (τ500) and Angstrom exponent (α) obtained on a minutely basis by sun photometers in JMA from 1998 through 2002. The significant aerosol event (τ500 > 0.3) was classified into “a significant large particle event” (α ≤ 1.0) and “a significant small particle event” (α > 1.0). “Significant large particle event days” in Ryori were clearly more frequent from 2000 to 2002 compared with 1998 and 1999, and corresponded with the frequencies of Kosa dust events in China and Japan. In Yonagunijima, Japan, however, “Significant large particle event days” were not corresponded with the frequencies of Kosa dust events in China. Moreover, the occurrence distribution of the Angstrom exponent (α), an index of particle size distribution, in Yonagunijima showed that typical Kosa events were rare, while significant small particle events occurred frequently. This indicates the dominant influence of relatively small particles on the large spring aerosol optical depth in Yonagunijima, rather than large particles such as Kosa dust. Trajectory analysis suggested that these small particles originated from urban pollution in the east coast of the Asian continent and Taiwan and/or smoke from biomass burning in Southeast Asia.

Individual aerosol particles were collected in the middle troposphere, at 5–6 km altitude, over the western Pacific Ocean in the Pacific Atmospheric Chemistry Experiment campaign, between 38°S and 38°N, in January 1994. Particles were collected on electron microscope grids, over approximately 1.5° latitude bands using a 1-mm diameter jet impactor and were examined with an energy-dispersive X-ray (EDX) analyzer and transmission electron microscope. Based on the EDX analysis and morphological identification, particles were classified into various types, including sea salt, mineral, sulfate, and sulfuric acid. For the radius range of 0.1–1 μm, sulfuric acid and ammonium sulfate were found across most of the observed regions and are considered important background aerosols. As well, sea salt particles were abundant over the tropical regions (present at up to 67%), in association with vertical transport by convective clouds. North of 20°N, mineral particles were dominant, suggesting transport of Asian dust–storm particles from China in the westerly air stream. Evidence for the modification of mineral particles, prior to their collection in the free troposphere, is also presented and discussed.

Aerosol-induced atmospheric optical phenomena were analyzed at four sun photometer sites in Japan, viz., Ryori (39°02´N, 141°50´E), Tsukuba (36°03´N, 140°08´E), Yonagunijima (24°28´N, 123deg;01´E), and Minamitorishima (24°18´N, 153°58´E), over the period MarchHApril 2002 to elucidate the causes of such phenomena, and explain regional differences. Using observed optical properties, the phenomena were classified into three kinds of aerosol events: Type A, Type B, and Type C, which, respectively, correspond to relatively clear conditions, typical Kosa conditions, and haze conditions. The frequency of occurrence of these aerosol events was found to be significantly different at the four observation sites in Japan during the observation period. Type B events were found to occur rather frequently in Ryori in northern Japan. The severe Kosa events observed at Ryori on April 10, 2002, which constitute a case study, passed over Northeast China where dust storms have rarely been reported before 2000. The smallest mean aerosol optical depth at 500 nm (τ500) and Angstrom exponent (α) were observed in Minamitorishima. However, according to other studies, the mean τ500 in Minamitorishima was larger than those in the eastern Pacific, which showed that the influence of the Asian outflow on atmospheric optical properties was still significant even 3,000 km away from the Asian continent. The mean τ500 and α, as well as the frequency of occurrence of Type C events, were the largest at our observation site in Yonagunijima. Based on the trajectory analysis obtained during the observation period, the high frequency of Type C events at Yonagunijima was ascribed to the haze from biomass burning in Southeast Asia and/or urban pollution in coastal areas in Southeast China. The frequency of Type C events in Ryori and Tsukuba indicated that haze is not an uncommon occurrence, not only in southern Japan, but also in central and northern Japan, and that, along with Kosa events, they may affect atmospheric optical properties over Japan in the spring.

Optical thickness of the Asian dust aerosols over the ocean near Japan was retrieved from the visible data of the Geostationary Meteorological Satellite (GMS-5) from April 7 to 22, 2002 using the method by Masuda et al. [Remote Sens. Environ. 82 (2002) 238] in which the nonsphericity of dust particles was taken into account. The retrieved optical thicknesses were then compared with those derived from the sunphotometer and the skyradiometer measurements. The mean and the standard deviation of the optical thickness difference, GMS minus radiometers, were −0.03 and 0.17, respectively, for 85 samples where the optical thickness ranges from 0.25 to 1.07.

A balloon observation was carried out on 22 August in 1993 from Xianghe Scientific Balloon Base (39.75°N, 117.0°E) near Beijing in China. Individual aerosol particles in the five samples collected in the troposphere and lower stratosphere were analyzed by using a transmission electron microscope equipped with an energy-dispersive X-ray (EDX) analyzer. Types of particles were classified by the quantitative EDX analysis and particle morphology. Following results were obtained by the analyses of aerosol particles in the radius range of 0.1–0.5 μm: (1) Sulfate particles were dominant (80%) in aerosol particles collected between 4 and 6 km altitude. (2) Sulfuric acid particles were present in 74% of particles at ∼8 km altitude, 91% at 11 km, 95% at 17 km and 88% at 21.2 km. (3) “S-rich” particles with K were collected both in the troposphere and lower stratosphere. It was considered that the particles containing K found at ∼5, ∼8 km altitude could originate from burning processes in the continent including the Tibetan plateau and be transported to the middle troposphere. (4) Sulfuric acid particles with Fe were present in 20–30% of sulfuric acid particles in the lower stratosphere. (5) Particles mainly composed of minerals were present in 6, 11% of particles at ∼5, ∼8 km, indicating the vertical transport to the upper troposphere. (6) Mineral particles which contain sulfuric acid and sulfate suggest the formation of sulfuric acid and sulfate on mineral particles by heterogeneous processes in the troposphere. (7) Sea-salt particles with and without minerals were collected in the troposphere and lower stratosphere, suggesting the vertical transport by convective clouds.

Aerosol and precipitation samples were obtained at the summit of Mt. Fuji, the highest peak (3776m a.s.l.) in Japan, in the summers of 1997, 1998, and 1999. The mountaintop might be affected by valley wind during the afternoon, but is located in the free troposphere during the morning. The temporal variations of chemical species in the aerosol and precipitation samples correspond with meteorological conditions. The SO4 2− in the aerosol and precipitation exhibits high concentration with low temperature air mass, indicating the influence of long range transport from the Asian Continent. The contribution of the free troposphere to the chemical species obtained at the summit is estimated to be at least 30% during the summer season.

1995年11月15日に富士山頂でトロポポーズフォールディングを観測した。このフォールディングは山頂で強い波状のオゾン変動を示したので、プランドルの混合長理論を用いて山頂高度でのオゾン波動中の鉛直混合距離をケーススタディとして見積もった。その結果、山頂での継続時間は18分と短かったものの、最大でオゾン振幅10.1ppbvという波が観測された。この波は、245.1mの鉛直混合距離に相当する。フォールディングの間中観測された他の波は、それぞれ鉛直混合距離が45.1、72.0、70.8、102.5mと見積もられた。観測したフォールディングはこれまで富士山頂で観測したフォールディングの中で珍しい波状構造を持っていた。そのためこのフォールディングの値をフォールディング一般の典型値とする事は出来ない。また、この見積もりはケーススタディであり、また多くの仮定を含んでいることにも注意しなければならない。しかし、東アジア域でのフォールディングの直接観測はほとんど無いので、この結果は意義があると考えられる。この解析に加えて、もしリモートセンシング法を含む何らかの手段でフォールディング中の波動の鉛直擾乱速度が判れば、今後富士山頂の個々のフォールディングに対して成層圏から対流圏へのオゾンフラックス量の見積もりが出来る可能性がある。

Ozone and CO were simultaneously measured for 17 days at the summit of Mt. Fuji (3776 m above sea level) in summer 1997. A classification of air mass origins based on backward trajectory analysis showed that both ozone and CO exhibit minimum concentrations in the air mass originating in southeast Asia, probably due to photochemical destruction during transport over the ocean. Both ozone and CO show maximum concentrations in the air mass originating in northeast Asia, reflecting background concentrations. Air mass groups transported long range from the Pacific, southeast Asia, and northeast Asia exhibit positive correlation between ozone and CO, while the air mass group staying around Japan that was thought to include fresh anthropogenic pollution negatively correlated. Diurnal variation of ozone in the air mass from the Pacific and southeast Asia (relatively clean and humid air) shows that ozone is likely to be photochemically destroyed in the air uplifted by valley winds at the summit in the daytime. Both diurnal variations of CO and ozone show evening enhancements in the case that the air at the summit passed over the Nagoya urban and industrial area before arriving at the summit. Thus, the polluted air is likely to be lifted up by convective plumes in the Nagoya area around noon, being transported with diffusion to Mt. Fuji in the evening.

AbstracL Air samples at 8-13 km were collected regularly using a commercial airliner to obtain long-term measurements of carbon monoxide (CO) mixing ratio in the upper troposphere over the western Pacific between Australia and Japan during April 1993 - December 1997. The measurements in 1997 clearly reveal an anomalous CO increase during September to November in the Southern Hemisphere, with a maximum of 320-380 ppb around 20°S in October. Tropical biomass burning, not urban/industrial emissions, was the main source for the enhanced CO in 1997. A similar southern-spring increase due to biomass burning was observed in previous years. The peaks showed a large interannual variation associated with the El Nino/Southern Oscillation (ENSO) events. The largest CO spring peak appeared during the strong El Nino event in 1997, while the weak La Nina year of 1996 was marked by a largely suppressed CO spring peak. The outgoing longwave radiation (OLR) anomaly is largest during the El Nino events indicating that the events cause a longer drought in the tropics and significantly influence the enlargement of biomass burning in tropical Southeast Asia. Thus the most likely cause for the ENSO-cycle CO variability is a year-to-year change of biomass-burning emissions mainly from Southeast Asia. The appearance of the CO spring peak in the southern subtropics is discussed on the basis of the possible long-range transport of biomass-burning CO from Southeast Asia to the upper troposphere over the western South Pacific.

During the PACE-5 campaign over Australia and Indonesia in October 1997, we used an aircraft to measure carbon monoxide (CO) and hydrogen (H2). Latitudinal distributions of CO and H2 clearly showed a large increase from northern Australia to Kalimantan in Indonesia. Elevated CO levels over northern Australia were observed only in the smoke plumes of savanna fires. A thick smoke haze from forest fires over Kalimantan contained very high CO mixing ratios of 3 to 9 ppm. These enhanced CO mixing ratios correlated well with increased concentrations of H2, nitrogen oxides (NOx), and aerosols. Emission ratios from biomass burning in Kalimantan ranged 0.06 0.1 for H2/CO (ppb/ppb), 0.0002 to 0.0005 for NOx/CO (ppb/ppb), and 0.43 to 1.0 for number of aerosols/CO (cm−3/ppb). These values were much lower than emission ratios in northern Australia. This difference suggests that the biomass burning in Indonesia was intense and that, due to a strong El Nino event, an unique composition of trace gases was formed in the smoke haze.

The 1997 El Nino unfolded as one of the most sever El Nino Southern Oscillation (ENSO) events in this century and it coincided with massive biomass burning in the equatorial western Pacific region. To assess the influence on the atmosphere, aircraft observations of trace gases and aerosol were conducted over Kalimantan in Indonesia and Australia. Over Kalimantan in Indonesia, high concentrations of O3, NOx, CO, and aerosols were observed during the flight. Although the aerosol and NOx decreased with altitude, the O3 had the maximum concentration (80.5 ppbv) in the middle layer of the smoke haze and recorded very low concentrations (∼20 ppbv) in the lower smoke layer. This feature was not observed in the Australian smoke. We proposed several hypotheses for the low O3 concentration at low levels over Kalimantan. The most likely are lack of solar radiation and losses at the surface of aerosol particles.

Air samples at 8.5–13 km were collected regularly using a commercial airliner between Australia and Japan, and they were measured for CO mixing ratios to obtain time series data from April 1993 to July 1996. When averaged over 12 latitudinal bands between 30°N and 30°S, two overall features emerge from these data. First, CO levels in the upper troposphere decreased in all latitudinal bands. Second, the seasonal cycle showed significant differences between the northern and southern hemispheres. In the southern hemisphere a strong maximum in the CO mixing ratio (up to around 90 ppb) was found in October-November. The most likely source for this enhanced CO is tropical biomass burning. Methane oxidation and transport of industrial CO from the northern hemisphere were estimated as relatively minor sources during the austral spring. Air mass trajectories indicate that an extremely high CO level of ∼130 ppb observed in November 1994 between 10° and 20°S was due to enhanced biomass burning in Southeast Asia and/or northern Australia. On the other hand, air mass trajectories for the 20°-30°S region indicate that CO-rich air from biomass burnings over southern Africa or South America was transported across the South Indian Ocean within ∼1 week by the strong westerly winds around the subtropical jet. Thus it is concluded that a rapid horizontal transport coupled with deep convection plays an important role in the appearance of the CO spring peak in the upper troposphere over the western South Pacific.

Tropospheric ozone events observed at the summit of Mount Fuji (3776 rn above sea level) were analyzed as case studies. The ozone, intruding from the lower stratosphere by the cutoff low or the tropopause fold over the Asian continent, is transported to the middle troposphere over Japan even in summer. The subtropical jet, which is intensified over Japan, also contributes to the summer intrusion from the stratosphere. A long stratospheric streamer, as described previously by Appenzeller et al. ( 1996), brings about a persistent ozone enhancement at the summit of Mount Fuji. The large variation of summer ozone over Japan is attributable to the alternate overspreading of these ozone-rich stratospheic air masses and the ozone-depleted subtropical maritime air mass. In contrast, winter ozone variation is relatively small at the summit of Mount Fuji. The steep potential temperature gradient between 500 hPa and 300 hPa in the winter cutoff low, which restrains vertical mixing in the upper troposphere, possibly causes less influence of the stratospheric air mass on the middle troposphere in winter, since mixing processes around a cutoff low play an important role in air mass exchange between the stratosphere and the troposphere.

A new fast-response ozone photometer which uses ultraviolet (UV)-absorption is described. Its three UV-absorption cells each cycle among three modes: reference, sample and transition modes. The three cells are differentially phased, enabling the ozone concentration to be determined in a cell in one second without interruption. The system is effective for measurements requiring a fast response, such as measurements on aircraft. We made a trial instrument that adopts this system and confirmed its theoretically predicted operation. The trial instrument showed a significantly faster response than earlier types of ozone photometers. This system may be applicable to other gas measurements that uses UV-absorption.

The latitudinal distribution of the tropospheric ozone over the western Pacific Ocean has been observed during two aircraft experiments, namely the 1993 Pre-PACE and the 1994 PACE-I experiments. The ozone concentration in the northern middle latitudes ( > 20°N) was high (40–50 ppbv) and stable in both the experiments, however, the air originating from the Pacific Ocean showed lower concentration of ozone compared to the one that came from the Eurasian Continent. A sharp gradient of the ozone concentration was observed around 10–25°N in these experiments. To the north of this gradient, the air mass with high ozone and low water vapor concentration came from the Eurasian Continent. South of the gradient, the air mass with low ozone and high water vapor concentration came from the tropical Pacific. Extremely low ozone concentrations (as low as 20 ppbv) were observed around the equator during the field experiments. The reason is probably that the ozone was photochemically destroyed by the high water vapor concentration in the boundary layer near the equator and this air mass was lifted up by the Walker circulation. The air mass which came from the middle of Australia showed high concentration of ozone (∼ 50 ppbv) around 20°S on the flight course; on the other hand, the air mass which came from the ocean showed low concentration of ozone around 24°S. The air mass with high concentration of ozone may be affected by mixing with the boundary layer in which photochemical ozone production occurs.

Number concentration and size distribution of aerosol particles were measured on board aircraft during the PACE (Pacific Atmospheric Chemistry Experiment) campaign from Australia to Japan in January 1994. The spatial distribution of condensation nuclei (CN) (r ≥ 4 nm) at 5-6 km altitude showed large variabilities in concentrations from 10 2 to 10 3 mg -1 ; that is, the concentrations were low (70-500 mg -1 ) in the intertropical convergence zone, high (400-1500 mg -1 ) in the subtropical high-pressure area, and low again in the higher latitudes. An apparent opposite tendency was present between CN and large particle (r≥0.15 μm) concentrations. The size distributions in the subtropical region exhibited high number concentrations of very fine particles (r < 0.02 μm). Together with the horizontal observation, vertical observations of aerosols were carried out over some areas. In the subtropical area (Saipan), CN concentration increased with altitude in contrast to the large particle concentration. Also most of the particles collected at 6 km altitude over Saipan contained sulfuric acid. These results are consistent with the results of Clarke (1993, J. geophys. Res. 98,20,633-20,647)that new particle formation is favored in the upper troposphere.

Vertical distributions of the tropospheric ozone over Wakasa Bay, Enshu-nada and some other places in Japan were measured by a series of aircraft measurements during 1987-1991. The backward trajectories of air parcels with the ozone peaks and potential vorticity distributions around them were then analyzed to study their origin. This result showed that the ozone peaks over Japan sometimes originated from the tropopause folding which occurred in the west of Japan. Two ozone peaks had different altitudes and concentrations over Enshu-nada on 21 January 1989 which originated from different tropopause foldings. They seemed to diffuse during the transport which resulted in a stratified structure forming over Japan. The two similar ozone peaks, which originated from the identical tropopause folding and were observed 300 km away on 21 January 1989, belonged to the identical ozone layer based on the potential temperature inclination. Two different types of air masses were simultaneously observed at different points over Japan on 8 August 1990. One was the continental air mass which showed a high ozone and a low water vapor concentration being transported from the northeast of China. The other was the maritime air mass which showed a low ozone and a high water vapor concentration staying over the sea for a few days that didn't come across the stratospheric air mass and pass over the metropolitan area. However, the ozone and water vapor showed a positive correlation even when the maritime air mass passed over the metropolitan or industrial areas, thus showing that the concentrations of the ozone and vapor were both high. The vertical distribution of ozone on 27 April 1991 over Tsukuba was almost uniform and the concentration was approximately 70 ppbv. This suggested that the air masses came from the folding area whose vertical inclination of potential vorticity was small and was transported shortly after this event. Not only intrusion from the stratosphere but diffusion during the transport was also important for the ozone distribution over Japan based on the relationship between the concentration of ozone and the distance of transport.

Since August of 1992, tropospheric ozone has been measured at the top of Mt. Fuji to investigate the background ozone feature of middle troposphere over Japan. During one year, the maximum concentration of monthly mean ozone at Mt. Fuji was 59.3 ppbv in May and minimum was 30.9 ppbv in August. The secondary minimum was recorded in January. The seasonal behavior showed a bi-modal trend. The diurnal amplitude of ozone in summer showed slight effect from the mountain and valley winds, but the amplitude around 2 ppbv is a lot smaller compared to other mountain data. The amplitude of other seasons did not show any significant variations. The effect of the mountain and valley winds were small at the top of Mt. Fuji because of the steep and windy summit. Day-to-day summertime variations varied wildly and often recorded low concentrations of ozone less than 20 ppbv. This low concentration of ozone sometimes continued for 3–5 days. This summer ozone minimum of the lower free troposphere was due to horizontal advection of the ozone poor air. In wintertime, the deviation of ozone concentration was very small. High peaks which lasted only a few hours were observed in spring. Analyzing the cross section chart on the 14th May 1993, the high peak of ozone on this day must have come from the stratosphere along the isentropic surface.

Observation on tropospheric aerosol particles was carried out on 27 April 1991 over Tsukuba, Japan. Aerosol particles collected in the upper troposphere at 7.5 km altitude were studied by electron microscopy. Soot-containing particles were collected in a large number fraction (0.53) of the submicron particles. Such a high proportion of soot particles in the upper troposphere has not been reported in previous research. From the X-ray analysis, vanadium was also detected in some of the particles, indicating the origin of oil combustion. The oil-well fires in Kuwait are likely to be the origin of these soot particles. Calculated trajectories of the air suggested that these particles over Japan had circled the globe.

[1] Since 1993, atmospheric carbon dioxide (CO2) has been continuously observed by the Japan Meteorological Agency at Minamitorishima station (24°18′N, 153°58′E), located about 2000 km off the Asian continent in the western North Pacific. The long-term record shows high-frequency measurements with interesting episodic events with extremely low CO2 mixing ratios 5–10 ppm below the background seasonal cycle. These extremely low CO2 (ELC) events occur several times each year, primarily in July, August, and September, although the number of events varies from year to year. The origins of air masses associated with the ELC events were defined by backward trajectory analyses as well as chemical characterizations using simultaneous observations of other trace gases (CO, CH4, and O3). The results indicate that the air masses with extremely low CO2 were influenced by active biospheric uptake in summer over different continental sink regions in Siberia, northern Asia, and Southeast Asia due to rapid long-range transport driven by strong northerly or southerly winds. The spatial scale of the widespread low-CO2 distribution for the ELC events in 2001 was captured by a simulation experiment using a three-dimensional chemical transport model. It clearly revealed that the Intertropical Convergence Zone around 20°N in the western North Pacific during summer blocked further southward intrusion of ELC events through the lower troposphere.

[1] We report here airborne measurements of atmospheric CO2 over the western North Pacific during the March–April 2001 Transport and Chemical Evolution over the Pacific (TRACE-P) mission. The CO2 spatial distributions were notably influenced by cyclogenesis-triggered transport of regionally polluted continental air masses. Examination of the CO2 to C2H2/CO ratio indicated rapid outflow of combustion-related emissions in the free troposphere below 8 km. Although the highest CO2 mixing ratios were measured within the Pacific Rim region, enhancements were also observed further east over the open ocean at locations far removed from surface sources. Near the Asian continent, discrete plumes encountered within the planetary boundary layer contained up to 393 ppmv of CO2. Coincident enhancements in the mixing ratios of C2Cl4, C2H2, and C2H4 measured concurrently revealed combustion and industrial sources. To elucidate the source distributions of CO2, an emissions database for Asia was examined in conjunction with the chemistry and 5-day backward trajectories that revealed the WNW/W sector of northeast Asia was a major contributor to these pollution events. Comparisons of NOAA/CMDL and JMA surface data with measurements obtained aloft showed a strong latitudinal gradient that peaked between 35° and 40°N. We estimated a net CO2 flux from the Asian continent of approximately 13.93 Tg C day−1 for late winter/early spring with the majority of the export (79%) occurring in the lower free troposphere (2–8 km). The apportionment of the flux between anthropogenic and biospheric sources was estimated at 6.37 Tg C day−1 and 7.56 Tg C day−1, respectively.

Aircraft observation under the Pacific Atmospheric Chemistry Experiment (PACE) program was performed from February 13 to 21, 2000 to examine in detail the distributions of CO 2 in the free troposphere between 5 and 11 km. Continuous measurements of CO 2 mixing ratios were made using an on-board measuring system over the northern North Pacific between Nagoya, Japan and Anchorage, Alaska, and the western North Pacific between Nagoya and Saipan. Other trace gases, such as CO and O 3 , were also observed using continuous measuring systems at the same time. CO 2 over the northern Pacific (35°N and higher) showed highly variable mixing ratios, ranging from 374 ppm in the upper troposphere to 366 ppm in the lowermost stratosphere. This highly variable distribution of CO 2 was quite similar to that of CO, but the relationship between CO 2 and O 3 showed a strong negative correlation. These results indicated that the exchange process between the stratosphere and the troposphere significantly influences the large CO 2 variation. On the other hand, the CO 2 over the western North Pacific to the south of Japan showed no significant variation in the upper troposphere at 11 km but a relatively larger variability at 5 km. The CO 2 enhancement at lower altitudes coincided with the CO elevation due to the intrusion of a polluted air mass. Trajectory analysis indicated that the Asian continental outflow perturbed the CO 2 distributions over the western Pacific. Very low mixing ratios of O 3 of less than 20 ppb were distributed in the latitude band of 15–30°N at 11 km, reflecting the effects of transport from the equatorial region. DOI: 10.1111/j.1600-0889.2004.00088.x