Deposition velocity of PM2.5 sulfate in the summer above a deciduous forest in central Japan

Abstract

In order to increase knowledge of aerosol dry deposition for the regional assessment of acid deposition and transboundary air pollution in East Asia, an experimental study on PM2.5 sulfate deposition was implemented in the early summer of 2009. The experimental field was located in a deciduous forest at the foot of Mt. Asama, central Japan. Aerosol fluxes were obtained using the aerodynamic gradient method. Three aerosol samplers were placed on an experimental tower at 21, 24 and 27 m above the ground surface, and collected PM2.5 on filters for chemical analysis. Vertical concentration differences between 21 m and 27 m of PM2.5 sulfate were detected significantly when the concentration exceeded 1 μg m −3. Mean deposition velocity was estimated to be 0.9 ± 1.0 cm s −1 in the daytime and 0.3 ± 0.3 cm s −1 in the nighttime. In the case that a height-dependent correction in the roughness sub-layer was taken into account, the deposition velocities increased more, especially in daytime. Higher deposition velocities in the daytime were associated with larger friction velocities and unstable conditions. The deposition velocities observed in this study were in agreement with other experimental results found in the literature. On the other hand, they were higher than those calculated by theoretical models. Two empirical parameterizations (Wesely, M.L., Cook, D.R., Hart, R.L., 1985. Measurement and parameterization of particulate sulfur dry deposition over grass. Journal of Geophysical Research 90, 2131–2143; Ruijgrok, W., Tieben, H., Eisinga, P., 1997. The dry deposition of particles to a forest canopy: a comparison of model and experimental results. Atmospheric Environment 31, 399–415) were validated by the observations. The general trend of higher daytime and lower nighttime deposition velocities was similar among the observation and the two parameterizations. The large variability found in the measurement was not reproduced by the parameterizations, because it is attributable to random error from the differences between the samplers. The observations were in accordance with the parameterization of Ruijgrok et al. (1997) for a forest, although much larger than that of Wesely et al. (1985) for grasslands. This indicates the large difference in aerosol deposition velocities between forests and grasslands.