1. Introduction
The radiative forcing of mineral dust can either be positive or negative (Houghton et al. 2001). Previous studies have pointed out that the evaluation of the radiative effect of mineral dust on regional and global climates involves large uncertainties (Sokolik and Toon 1996; Sokolik et al. 2001; Tegen et al. 1996). The accurate estimation of the radiative forcing of aerosols including mineral dust requires the knowledge of their optical properties like complex refractive index in addition to other properties such as size distribution, morphology, and the internal mixture ratio of dust and other components. In particular, the size distribution could vary during long-range transportation due to the differences in the terminal settling velocity, which is particle-size dependent. Liao and Seinfeld (1998) reported that shortwave radiative forcing is extremely sensitive to the mass median diameter in the 1.0–5.0-μm region. Claquin et al. (1998) showed that the knowledge of the mean radius is required in the determination of the amplitude of the shortwave and longwave forcings when large particles are involved. Therefore, the size distribution should be measured accurately and with a sufficiently high resolution.
Several methods exist for measuring the size distribution of mineral dust (J. S. Reid et al. 2003). However, each method has its limitations (Sokolik et al. 2001). The optical particle counter (OPC) can be used to measure the amount of light scattered by individual particles (Chun et al. 2001; Iwasaka et al. 2003; Kim et al. 2004b; Porter and Clarke 1997). It can measure aerosol number concentrations at intervals of several minutes; however, it cannot separate the mineral dust from whole aerosols. The OPC contains several channels and is strongly influenced by the complex refractive index and nonsphericity of the particles, which is fairly significant for mineral dust. A cascade impactor can be used to segregate particles based on their mass-to-drag characteristics (Arimoto et al. 1997; Park et al. 2004; E. A. Reid et al. 2003; Tanaka et al. 1989). By using chemical analyses, the size distribution for each chemical composition can be obtained; however, the size separation is uncertain, and it exhibits a coarse size bin resolution. An electron microscope (EM) can furnish detailed information on the properties of the individual particles, such as its morphology; it can also determine the internal mixing state by elemental analysis if the EM is attached with an energy dispersive X-ray analyzer (EDX; Gao and Anderson 2001; Zhang et al. 2003). At present, however, the reports with regard to systematic measurements are limited due to the need for extensive and elaborate work on the large number of different aerosol particles. By using sky radiance distribution measurements obtained with a skyradiometer (Nakajima et al. 1996) and Aerosol Robotic Network (AERONET) sun photometer (Holben et al. 1998), it is possible to retrieve a column-integrated aerosol size distribution using the optical inversion method (Alexander et al. 2002; Aoki and Fujiyoshi 2003; Dubovik et al. 2002; Kim et al. 2004a; Murayama et al. 2001; Tanaka et al. 1989; Tanré et al. 2001). However, the analysis requires various assumptions; further, clear-sky conditions are essential for performing the measurements. In addition, with the exception of the EM, it is difficult to separate the mineral dust from whole aerosols by using the abovementioned methods, although the individual size distribution of mineral dust is required for the evaluation of its optical properties.
Coulter Multisizer analysis can be used for measuring the soil or sediment size. The Coulter Multisizer is based on the electrical sensing zone (ESZ) or the Coulter principle (Coulter 1956). It is used to determine the number and volume of particles suspended in a conductive liquid. When a particle passes through a small aperture with two electrodes on either side, it induces a change in the resistance between the electrodes. This process results in an electronic pulse proportional to the particle volume. A pulse height analysis yields the size distribution of an ensemble of particles. The collected aerosol particles are suspended in an electrolyte solution. In this study, the used electrolyte solution is for measuring solid materials and is not a near-saturated solution for measuring water-soluble materials. Accordingly, the measurement is possible only for the water-insoluble particles (WIPs). The chemical elements of Asian dust, excluding oxygen, mainly comprise Si, Na, K, Ca, Mg, Al, and Ti, which are all water-insoluble with the exception of Ca (Kanamori et al. 1991). For the Asian dust phenomenon at Yakushima Island during spring 1989, the total mass concentration of the water-insoluble chemical elements was about 4 times that of water-soluble Ca (Kanamori et al. 1991). Therefore, we consider the mineral dust to be composed of water-insoluble aerosol particles and that mineral dust particles can be separated from bulk aerosol particles by the dissolution of the collected particles. In addition, this measurement is not affected by the complex refractive index, shape, and orientation of the particle when it passes through the aperture, whereas these parameters strongly affect the size measurement in other methods.
The Multisizer-3 Coulter Counter (Beckman Coulter, Inc.)—the latest instrument based on the Coulter principle—can provide a high-resolution size measurement with a maximum of 300 channels for an arbitrary size range; the maximum size range is 2%–60% of the aperture diameter. Furthermore, since the number concentration is determined at each bin in the solution, the quantitative size distribution with a high resolution and the concentration unit can be obtained by using volumes of electrolyte solution and sampled air. On the other hand, when the smallest aperture diameter of 20 μm is used, the Coulter Multisizer can analyze the particle size within a diameter range of 0.4–12 μm; therefore, the measurable particle size should be greater than 0.4 μm in diameter. McTainsh et al. (1997) introduced particle size analyses for sediments and soil samples using a Coulter Multisizer. In this study, we apply the Coulter Multisizer method to the measurements of the size distribution of Asian dust and demonstrate the temporal and spatial variation in the WIP size distribution in Japan during the Asian dust phenomena in order to estimate the optical properties for the evaluation of the radiative forcing and for the satellite remote sensing. Furthermore, we improve the analytical method and show the results. We then summarize the characteristics of the Coulter Multisizer method.
2. Methods
a. Sampling and size distribution analyses
Filter sampling was conducted at Nagasaki (32°45′N, 129°52′E), Okayama (34°39′N, 133°54′E), Kofu (35°39′N, 138°34′E), and Tokyo (35°41′N, 139°45′E), Japan, in 2003 during the spring—the time of the year when the Asian dust phenomena typically occur (Fig. 1). Aerosols were collected on Nuclepore filters with a pore size of 0.4 μm in diameter (Whatman) through a tube with an inner diameter of 10 mm in open air over a sampling period of 1 day. The tube was used due to the constraint of setting our own sampling instruments at each sampling site, although using the tube resulted in sampling losses. At Kofu, when the Asian dust phenomenon occurred, the sampling periods were shortened in order to obtain a higher time resolution. The details of the sampling conditions at each site are listed in Table 1. Blank samples were periodically obtained for a qualitative check of the measurement. Simultaneously, the number concentrations of the aerosol particles were measured at regular intervals of 10 min using optical particle counters at all the sites, excluding Okayama, in order to monitor the abundance of Asian dust and the variation in the total aerosol abundance. At Okayama, the measurements of the suspended particulate matter (SPM) concentrations from the Minamigata Motor Vehicle Exhaust Gas Monitoring Station, which is located 2 km south of the Okayama University, were used instead of the OPC measurements.
Elemental carbon (EC) is a component of water-insoluble aerosol particles and it sometimes aggregates with Asian dust (e.g., Quinn et al. 2004). However, in the Coulter Multisizer method, it is essentially impossible to separate EC aerosols from dust aerosols. Therefore, the result includes not only Asian dust but also EC aerosol size distribution, although the EC aerosol size is generally considerably smaller than that of dust.
b. Calibration of transportation losses in the sampling tubes
c. Estimation of particles broken by ultrasonification
The collected aerosol particles were extracted into the electrolyte by ultrasonification. Although ultrasonic oscillations facilitate effective extraction, the fracturing of mineral dust particles and aggregation due to the powerful ultrasonic oscillations leads to an underestimation of the WIP size distribution. The extent of this underestimation was estimated.
The ultrasonic effect on sizing was evaluated by comparing the size distribution processed by ultrasonification with the original one. The test sample must be collected in a manner such that the fracturing of the particles is avoided. The aerosol particles collected on a filter were required to undergo treatment during the extraction process. The particles collected on a hard plate can be suspended in the electrolyte solution simply by flushing with a wash bottle and do not require processing. Therefore, a cascade impactor (Tokyo Dylec, LP-20RPA) was used for aerosol particle sampling. In this case, no grease was coated on the impaction plates because the cascade impactor is not intended for aerosol sampling by classification according to aerodynamic size. Therefore, the adhesion of the collected particles with grease on the impaction plate is avoided. Sampling was conducted at Kofu on 3 April 2004, during the Asian dust phenomenon. The aerosol particles from each impaction plate were suspended in the electrolyte solution without ultrasonification and then the original size distribution was measured. After the suspended solution was treated by ultrasonification for 3 min, the resulting size distribution was measured. The solution was then subjected to further ultrasonification for 3 min.
d. Improvements in the sampling and analysis methods
The sampling and analysis methods were improved in order to avoid the possibility of bias in the transportation losses in the sampling tube and particle breaking and aggregation dispersion by ultrasonification. Aerosol particles were guided to the filter folder directly without a tube. The electrolyte solution and extraction procedure were modified on the basis of the method used for a clay sample provided by Beckman Coulter, Inc. The aerosol particles collected on Nuclepore filters were soaked in 90-mL Na3PO4 solution for 2 days and washed using a magnetic stirrer, instead of an ultrasonic cleaner, for 30 min. A 70-μm-diameter pore size aperture, whose measurable size range is 1.4–42 μm in diameter, was used to measure particles around 10 μm in diameter with a good signal-to-noise ratio. Aerosol sampling was conducted at Kofu during spring 2004 using the improved method. The WIP size distribution was analyzed using the improved procedure.
e. Estimation of extraction efficiency from a filter
It is possible that the extraction efficiency by using a magnetic stirrer in the improved method is low because the agitation is not as powerful as that due to ultrasonic oscillations, although particle breaking and aggregation dispersion might be reduced. Therefore, the extraction efficiency was evaluated.
The extraction efficiency was determined by comparing the known particle number on the Nuclepore filter with the particle number measured using the improved analysis method for the filter. The filtered sample was prepared by the filtration of a standard suspension of Asian dust and ion-exchanged water for washing the electrolyte. The particle number on the filtered sample can be calculated from the filtrated volume of the standard suspension and its concentration. The standard suspension of Asian dust was prepared from a melting snow sample that includes deposited Asian dust; the sampling was performed in Sapporo, Japan, on 12 March 2004, by settling coarse particles greater than 20 μm and diluting the measured concentration. The particle number concentration with each size bin was measured using the Coulter Multisizer.
3. Results
a. Size distributions of Asian dust particles
Figure 2 shows the results for Nagasaki during 11–14 April 2003. The number concentration of aerosol particles in the ranges >1 and >2 μm increased from 0600 Japan standard time (JST: UTC + 0900) on 12 April; subsequently, the concentration in the range >5 μm increased after 0800 JST (period 1). The WIP size distribution shows a slight peak in this period. All the concentrations in each size range stabilized from 1600 JST 12 April to 0600 JST 13 April; consequently, all the counts increased (period 2). The maximum of the number concentrations in the range >1 μm were recorded at 1400 JST (period 3). Furthermore, the peak height of the WIP size distribution recorded the maximum value at this time. The volume mode diameter and standard deviation were 1.99 μm and 2.26, respectively. In the following period, the concentrations in the range >1 μm decreased gradually (period 4), and the peak value of the WIP size distribution varied by two-thirds that of the maximum with Dm = 2.04 μm and σ = 2.04 in period 3.
The results for Okayama during 11–14 April 2003, are shown in Fig. 3. The SPM concentration increased sharply at 1600 JST 12 April. The concentration decreased gradually after increasing to approximately 0.11 mg m−3. The spike of the WIP size distribution appeared to a slight degree at around Dm = 3.6 μm (period 1). After 1000 JST 13 April, the SPM concentration began to increase again (period 2). In this period, the WIP size distribution exhibited a broad lognormal distribution with spikes. The size of the larger spike in period 2 of around 3–4 μm was consistent with that of the spike in period 1. Hereafter, the SPM concentration constantly ranged around 0.06–0.08 mg m−3, and the peak value of the WIP size distribution exhibited a maximum (period 3). The mode diameter was 1.69 μm and the standard deviation was 1.98. In the following period (period 4), the SPM concentration partly decreased to around 0.06 mg m−3, and the mode and width of the WIP size distribution increased slightly to Dm = 1.90 μm and σ = 2.30, respectively; this occurred despite a decrease of approximately 20% in the peak height.
Figure 4 shows the results obtained at Kofu during 11–17 April 2003. The number concentrations in the ranges >2 and >5 μm began to increase from 0000 JST 13 April (period 2). In this period, two sharp WIP size distributions appeared at around Dm = 2.2 and 3.8 μm. In the following period (period 3), the concentrations in the ranges >0.3 and >0.5 μm exhibited an inverse correlation with the other ranges, and a WIP size distribution with Dm = 2.10 μm was observed. However, after 1700 JST, all the range counts increased (period 4). The mode diameter was smaller by 1.71 μm than that observed in period 3. Thereafter, the peak height of the WIP size distribution increased to the maximum with Dm = 1.60 μm and σ = 2.08 (period 5). In the following period, the concentrations in the range >1 μm began to decrease (period 6). The volume mode diameter decreased further below 1.40 μm. Although this trend continued in period 6, only the count in the range >5 μm began to increase after 1530 JST 15 April (period 7). Two sharp WIP size distributions appeared in a manner similar to that observed in period 2.
The results for Tokyo are shown in Fig. 5. The number concentrations in the ranges >0.3 and >0.5 μm increased after 1600 JST 12 April, although the concentrations in the larger ranges decreased (period 1); this resulted in the observation of a slight WIP size distribution. Thereafter, the concentrations in the range >1 μm increased sharply at 2030 JST 13 April, following which the larger range counts recorded their maximum (period 2). Furthermore, the peak of the WIP size distribution showed the highest value with Dm = 1.95 μm and σ = 2.14. Around 1330 JST 14 April, all the concentrations decreased; however, an increase in the count was observed except for those in the range >5 μm (period 3). The peak value of the WIP size distribution decreased by −23%. The mode diameter increased slightly from Dm = 1.95 to 2.24 μm. In the following period, the increase in the concentrations continued, except for those in the >5 μm range (period 4). However, the peak of the WIP size distribution decreased. The distribution exhibited a slight decrease: Dm = 1.57 μm.
b. Error estimation for sizing by ultrasonification
Figure 6 shows the results of the fourth and fifth stages of the cascade impactor of the ultrasonification test for only the first 3 min because the results for the entire 6-min test period were the same as those for the initial 3 min. The results greater than one-fourth of each maximum were fitted with lognormal distributions due to the irregularities of the smaller values. In the fourth stage, the mode diameter changed from 2.97 to 2.53 μm by ultrasonification. In addition, the volume, which was smaller than 1.2 μm, increased to more than twice its original value. This indicates that due to the broken particles and dispersed aggregates resulting from ultrasonification, the number of smaller particles increased. In the fifth stage, the mode diameter also changed from 1.94 to 1.56 μm; furthermore, the ratio of decrease in the geometrical mode diameter was 15%–20%.
c. Size distributions measured using the improved method
Figure 7 shows the results of two Asian dust events measured using the improved method at Kofu during spring 2004. The size distribution was detected in the range up to 10 μm with smaller scattering than the results measured using the previous method. In an earlier event during 30 March–2 April, the mode diameter of the WIP size distribution appeared at Dm = 2.66 μm (period 2). Thereafter, the mode diameter increased to Dm = 3.11 μm as the peak height increased (period 3). Then, the mode diameter decreased with a decrease in the peak height (period 4). The standard deviation ranged from 1.6 to 1.7. A sharp WIP size distribution appeared with a broad distribution in period 1 in a manner similar to period 2 in Fig. 4. In a subsequent event around 17 April, the WIP size distribution appeared at a mode diameter as large as around 5 μm (periods 5 and 6). Thereafter, the peak height increased, but the mode diameter decreased to around 4.3 μm. In this event, the standard deviations exceeded 2 and the size distributions were broader than that in the earlier event.
d. Extraction efficiency by agitation with a magnetic stirrer
Figure 8 shows the result of the extraction efficiency. The error bar indicates the standard deviation of the results of the three samples. The value exceeded around 70% in the measured size range and was the lowest around 3 μm. In the size range >8 μm, the efficiency was almost 100%; however, a few values exceeded 100% and the error bars were relatively high. Because the concentration of the standard suspension of Asian dust in the range >8 μm was one or two orders of magnitude lower than the smaller range concentration, the measuring error was significant in this size range.
4. Discussion
The measured volume-size distributions of water-insoluble particles considered as Asian dust particles obeyed a lognormal distribution; furthermore, the OPC and SPM results indicated that the air mass was indeed Asian dust.
The results obtained for Nagasaki, Okayama, and Kofu showed that the WIP size distribution gradually increased in a manner similar to a lognormal distribution. In contrast, a WIP size distribution appeared abruptly at Tokyo. The phenomena were the same as the results of the OPC number concentrations in the range >1 μm. At Nagasaki, there was a slight variation in the mode diameter and standard deviations. At Kofu, both these parameters exhibited a smooth decrease, while they increased at Okayama. At Tokyo, however, they first increased and then decreased. Despite identical or almost similar sampling periods, the results obtained at each sampling site differed from each other. This indicates a high variability in the WIP size distributions for the same Asian dust air mass. Assuming that the maximum WIP size distribution was typical for the air mass at each site, the volume mode diameter decreases from west to east, except for Tokyo. This indicates that large Asian dust particles descend before the small ones in the airmass transport. At Kofu, where the sampling periods were shorter than those at the other sites, two sharp peaks appeared both before and after the main Asian dust event. The significance of these results is not understood at present; however, they are extremely interesting. High-resolution size measurements obtained by using the Coulter Multisizer can assist in understanding these results.
The volume mode diameter of the WIP measured using the Coulter Multisizer ranged from 1.4 to 2.2 μm. Observations carried out with the skyradiometer showed that the coarse volume mode diameter is usually around 4 μm at several sites in Japan (Murayama et al. 2001). Aoki and Fujiyoshi (2003) also demonstrated that the volume size distributions exhibit peaks corresponding to diameters in the range of 4–6 μm at Sapporo. Kim et al. (2004a) indicated that the size distribution shows coarse mode diameters in the ranges of 4–6 or 8–10 μm at Anmyon, South Korea. Tanaka et al. (1989) reported that the volume size distribution obtained by using the cascade impactor (Andersen sampler) shows a sharp peak at around a diameter of 4 μm. If the WIP measured by our method were Asian dust particles, the measured size distributions would be smaller than those obtained from previous studies. The reasons can be classified into theoretically and experimentally induced factors.
One of the theoretical factors is the possibility of the internal mixing of Asian dust with other components. Zhang et al. (2003) showed that approximately 60%–85% of the dust particles were internally mixed with sea salt. Although the OPC, cascade impactor, and optical inversion method using a radiometer measure the internally mixed aerosol particles, the Coulter Multisizer method only measures the water-insoluble components of the particle. Therefore, our results, as compared with those that employed other methods, showed smaller particle size distributions. In addition, the combination of the Coulter Multisizer and other methods listed above can be used to derive the quantitative information of the internal mixture.
The experimentally induced factors are considered to be the dispersion of the aggregations and breaking of the particles by ultrasonification during the extraction of aerosol particles from a filter. In particular, the dispersion of the aggregations occurred only during the immersion of the aggregated aerosol particles into the electrolyte solution due to its transience (Leys et al. 2005). Nevertheless, it was demonstrated that ultrasonification results in a 15%–20% size reduction.
The improved method can eliminate the ultrasonification effect as well as calibrate the transportation losses in the tube. By comparing the results measured using the improved method with that of the previous method, the concentration in the range >3 μm increased and the scattering of its value decreased. The mode diameter measured using the previous method as 1.4–2.2 μm increased to as much as 2.6–3.1 or 4.3–5.6 μm in the two Asian dust events after the improvements, although the phenomenona measured using the improved and previous methods were distinct. On the other hand, the number size distributions obtained by using the OPC are compared with those obtained by using the Coulter Multisizer for both the previous and improved methods, although the OPC measures whole aerosol particles, including the water-soluble components (Fig. 9). It is clear that the consistency of the size distributions after the improvements was considerably better than that before the improvements in the size range from 2 to 5 μm, whose concentrations correlate to the components from Asian dust; however, there is the possibility of an internal mixture of Asian dust with other components and the uncertainty of OPC counts in this range because of the nonsphericity of the particles and the difference in the refractive index between Asian dust and latex particles used in the calibration. These results show that the improved method using the Coulter Multisizer can measure the size distribution of Asian dust in the range from 1 to 10 μm with high resolution.
If the size reduction by ultrasonification was −20%, the mode diameter measured using the previous method was estimated within the range of 1.8–2.8 μm. This size range is consistent with the smaller mode diameter measured using the improved method in 2004.
The extraction efficiency of the collected aerosol particles on a Nuclepore filter into an electrolyte solution was measured by using a filter in order to simulate a sample of the gathered aerosol particles by using a standard suspension of known concentration. It is recognized that the efficiency ranged from approximately 70% to 100%. However, it is possible that the physical properties of the atmospheric aerosol particles, particularly charging, differ from that of the particles once suspended in the electrolyte solution. Consequently, the estimated extraction efficiency includes an uncertainty.
The dispersion of the aggregated particles is unavoidable as long as they are suspended in a solution. However, Leys et al. (2005) show the estimation method of aggregation using a statistical curve-fitting procedure. This method is applicable to the size distribution measured using the Coulter Multisizer because of its high resolution.
5. Conclusions
In this study, we applied the Coulter Multisizer method to measure the size distribution of Asian dust. This instrument can perform high-resolution measurements of size distributions. Aerosol filter samplings were conducted at Nagasaki, Okayama, Kofu, and Tokyo, Japan, during the Asian dust season in 2003. The fit between the measured volume size distributions during the Asian dust phenomena and lognormal distributions is excellent. The measured size distributions varied for different sampling sites. This indicates a high variability in the WIP size distributions for the same Asian dust air mass. Assuming that the maximum WIP size distribution was typical for the air mass at each site, the volume mode diameter decreases from west to east, except for Tokyo.
The volume mode diameter of the WIP measured using the Coulter Multisizer ranged from 1.4 to 2.2 μm. The size distributions were smaller than those obtained in previous studies on Asian dust. The reasons for this can be classified into theoretical and experimentally induced factors. The former, which involves the possibility of the internal mixing of Asian dust with other water-soluble components, yields smaller size distributions because the Coulter Multisizer method can measure only the water-insoluble components of the particle. The experimentally induced factors include the breaking of particles and dispersion of aggregations by ultrasonification. Ultrasonification results in a 15%–20% size reduction. The sampling and analysis methods were improved due to the bias possibilities as a result of transportation losses in the sampling tube and particle breaking and aggregation dispersion by ultrasonification. The aerosol particles were guided to the filter folder directly without a tube. The aerosol particles collected on the Nuclepore filter were soaked for 2 days and washed using a magnetic stirrer, instead of an ultrasonic cleaner. Aerosol sampling was conducted at Kofu in spring 2004 using the improved method. The size distribution was detected as approximately 10 μm with smaller scattering. The mode diameter measured using the previous method (1.4–2.2 μm) increased to as much as 2.6–3.1 or 4.3–5.6 μm in the two Asian dust events after the improvements. In addition, the number size distribution obtained using the OPC was consistent with that obtained using the Coulter Multisizer in the improved method in the size range from 2 to 5 μm. These results show that the improved method using the Coulter Multisizer can measure the size distribution of Asian dust in the range from 1 to 10 μm. The advantages and disadvantages of the Coulter Multisizer method are as follows:
higher resolution than other methods;
measurement of only the water-insoluble particles in the Asian dust, and not the whole aerosol particles;
representation as a concentration unit;
possibility of dispersion of aggregations due to soaking in solution; and
measurable lower size limit of 0.4 μm in diameter.
The Coulter Multisizer method can furnish detailed information regarding the spatial and temporal variations in the mineral dust size distribution. In addition, a combination of the Coulter Multisizer and other methods can be used to obtain quantitative information regarding the internal mixture. Therefore, the optical properties of Asian dust can be estimated with a considerably higher accuracy.
Acknowledgments
The authors thank K. Yamaguchi, H. Yoshimura, and S. Ozawa for assistance in the analysis. We also thank Dr. T. Aoki for providing the snow sample and Okayama prefecture for providing the SPM data. We are also grateful to two anonymous reviewers for helpful comments on the earlier drafts of this manuscript. This study was supported by a Grant-in-Aid for Scientific Research on Priority Areas under Grant 14048228 from the Ministry of Education, Culture, Sports, Science and Technology, Japan.
REFERENCES
Alexander, S., and Coauthors, 2002: Atmospheric aerosol optical properties in the Persian Gulf. J. Atmos. Sci., 59 , 620–634.
Aoki, K., and Fujiyoshi Y. , 2003: Sky radiometer measurements of aerosol optical properties over Sapporo, Japan. J. Meteor. Soc. Japan, 81 , 493–513.
Arimoto, R., Ray B. J. , Lewis N. F. , Tomza U. , and Duce R. A. , 1997: Mass-particle size distributions of atmospheric dust and the dry deposition of dust to the remote ocean. J. Geophys. Res., 102 , 15867–15874.
Chun, Y., Kim J. , Cheon Choi J. , On Boo K. , Nam Oh S. , and Lee M. , 2001: Characteristic number size distribution of aerosol during Asian dust period in Korea. Atmos. Environ., 35 , 2715–2721.
Claquin, T., Schulz M. , Balkanski Y. , and Boucher O. , 1998: Uncertainties in assessing radiative forcing by mineral dust. Tellus, 50B , 491–505.
Coulter, W. H., 1956: High speed automatic bloodcell counter and cell size analyser. Natl. Elec. Conf. Proc., 12 , 1034–1042.
Dubovik, O., Holben B. , Eck T. F. , Smirnov A. , Kaufman Y. J. , King M. D. , Tanré D. , and Slutsker I. , 2002: Variability of absorption and optical properties of key aerosol types observed in worldwide locations. J. Atmos. Sci., 59 , 590–608.
Gao, Y., and Anderson J. R. , 2001: Characteristics of Chinese aerosols determined by individual-particle analysis. J. Geophys. Res., 106 , 18037–18045.
Hinds, W. C., 1999: Aerosol Technology: Properties, Behavior, and Measurement of Airborne Particles. 2d ed. Wiley-Interscience, 483 pp.
Holben, B. N. E., and Coauthors, 1998: AERONET—A federated instrument network and data archive for aerosol characterization. Remote Sens. Environ., 66 , 1–16.
Houghton, J. T., Ding Y. , Griggs D. J. , Noguer M. , van der Linden P. J. , Dai X. , Maskell K. , and Johnson C. A. , 2001: Climate Change 2001: The Scientific Basis. Cambridge University Press, 944 pp.
Iwasaka, Y., and Coauthors, 2003: Nature of atmospheric aerosols over the desert areas in the Asian continent: Chemical state and number concentration of particles measured at Dunhuang, China. Water Air Soil Pollut. Focus, 3 , 129–145.
Kanamori, S., Kanamori Y. , Nishikawa M. , and Mizoguchi T. , 1991: Chemical outline of Kosa. Hydrospheric-Atmospheric Sciences Kosa (in Japanese), Hydrospheric Atmospheric Research Center, Eds., Kokon-Shoin Publisher, 124–156.
Kim, D., Sohn B. , Nakajima T. , Takamura T. , Takemura T. , Choi B. , and Yoon S. , 2004a: Aerosol optical properties over east Asia determined from ground-based sky radiation measurements. J. Geophys. Res., 109 .D02209, doi:10.1029/2003JD003387.
Kim, Y. S., and Coauthors, 2004b: Dust particles in the free atmosphere over desert areas on the Asian continent: Measurements from summer 2001 to summer 2002 with balloon-borne optical particle counter and lidar, Dunhuang, China. J. Geophys. Res., 109 .D19S26, doi:10.1029/2002JD003269.
Leys, J., McTainsh G. , Koen T. , Mooney B. , and Strong C. , 2005: Testing a statistical curve-fitting procedure for quantifying sediment populations within multi-modal particle-size distributions. Earth Surf. Processes Landforms, 30 , 579–590.
Liao, H., and Seinfeld J. H. , 1998: Radiative forcing by mineral dust aerosols: Sensitivity to key variables. J. Geophys. Res., 103 , 31637–31646.
McTainsh, G. H., Lynch A. W. , and Hales R. , 1997: Particle-size analysis of aeolian dusts, soils and sediments in very small quantities using a Coulter Multisizer. Earth Surf. Processes Landforms, 22 , 1207–1216.
Murayama, T., and Coauthors, 2001: Ground-based network observation of Asian dust events of April 1998 in east Asia. J. Geophys. Res., 106 , 18345–18360.
Nakajima, T., Tonna G. , Rao R. , Boi P. , Kaufman Y. , and Holben B. , 1996: Use of sky brightness measurements from ground for remote sensing of particulate polydispersions. Appl. Opt., 35 , 2672–2685.
Park, S. H., Song C. B. , Kim M. C. , Kwon S. B. , and Lee K. W. , 2004: Study on size distribution of total aerosol and water-soluble ions during an Asian dust storm event at Jeju Island, Korea. Environ. Monit. Assess., 93 , 157–183.
Porter, J. N., and Clarke A. D. , 1997: Aerosol size distribution models based on in situ measurements. J. Geophys. Res., 102 , 6035–6046.
Quinn, P. K., and Coauthors, 2004: Aerosol optical properties measured on board the Ronald H. Brown during ACE-Asia as a function of aerosol chemical composition and source region. J. Geophys. Res., 109 .D19S01, doi:10.1029/2003JD004010.
Reid, E. A., Reid J. S. , Meier M. M. , Dunlap M. R. , Cliff S. S. , Broumas A. , Perry K. , and Maring H. , 2003: Characterization of African dust transported to Puerto Rico by individual particle and size segregated bulk analysis. J. Geophys. Res., 108 .8591, doi:10.1029/2002JD002935.
Reid, J. S., and Coauthors, 2003: Comparison of size and morphological measurements of coarse mode dust particles from Africa. J. Geophys. Res., 108 .8593, doi:10.1029/2002JD002485.
Sokolik, I. N., and Toon O. B. , 1996: Direct radiative forcing by anthropogenic airborne mineral aerosols. Nature, 381 , 681–683.
Sokolik, I. N., and Coauthors, 2001: Introduction to special section: Outstanding problems in quantifying the radiative impacts of mineral dust. J. Geophys. Res., 106 , 18015–18028.
Tanaka, M., Nakajima T. , Shiobara M. , Yamano M. , and Arao K. , 1989: Aerosol optical characteristics in the Yellow Sand Events observed in May, 1982 at Nagasaki-Part I Observations. J. Meteor. Soc. Japan, 67 , 267–278.
Tanré, D., and Coauthors, 2001: Climatology of dust aerosol size distribution and optical properties derived from remotely sensed data in the solar spectrum. J. Geophys. Res., 106 , 18205–18218.
Tegen, I., Lacis A. A. , and Fung I. , 1996: The influence on climate forcing of mineral aerosols from disturbed soils. Nature, 380 , 419–422.
Uematsu, M., Yoshikawa A. , Muraki H. , Arao K. , and Uno I. , 2002: Transport of mineral and anthropogenic aerosols during a Kosa event over East Asia. J. Geophys. Res., 107 .4059, doi:10.1029/2001JD000333.
Zhang, D., Iwasaka Y. , Shi G. , Zang J. , Matsuki A. , and Trochkine D. , 2003: Mixture state and size of Asian dust particles collected at southwestern Japan in spring 2000. J. Geophys. Res., 108 .4760, doi:10.1029/2003JD003869.
Sampling sites in Japan.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Measured size distribution of (top) WIPs and (bottom) time series of the aerosol number concentration using OPC in Nagasaki during April 2003. The sampling periods for the WIP size distribution measurement are shown in the bottom panel.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Measured size distribution of (top) WIPs and (bottom) time series of the SPM concentration in Okayama during April 2003.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Measured size distribution of (top) WIPs and (bottom) time series of aerosol number concentration using OPC in Kofu during April 2003.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Measured size distribution of (top) WIPs and (bottom) time series of aerosol number concentration using OPC in Tokyo during April 2003.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Test sample size distribution collected using a cascade impactor before and after ultrasonic treatment. The curves are lognormal distributions fitted with results greater than one-fourth of each maximum.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Size distribution of WIPs measured using the improved method and time series of aerosol number concentration using OPC in Kofu during the spring of 2004.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Extraction efficiency of collected aerosol particles on a Nuclepore filter into electrolyte solution stirred with a magnetic stirrer.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Comparison between the number size distributions measured using OPC and the Coulter Multisizer method on both (left) the previous and (right) improved methods at Kofu. The maximum measurable size for the OPC is assumed to be a diameter of 10 μm.
Citation: Journal of Atmospheric and Oceanic Technology 24, 2; 10.1175/JTECH1965.1
Filter sampling conditions at each site.