where Dep. is the number (or mass) of particles deposited per unit surface per unit time, kz is the turbulent diffusion coefficient, z is the height, while N and vs are the particle concentration and sedimentation velocity, respectively. Under average conditions (see previous chapter) the concentration of aerosol particles decreases with increasing altitude. It should be mentioned, however, that this decrease is not necessarily observed near the surface (e.g. in the first 1-2 m). In this case turbulent diffusion may be an effective removal process, especially for smaller particles for which the second term at the right-hand side of equation [5.1] is negligible.

Impaction is a function of several variables. According to Ranz and Wong (1952) the impaction efficiency (defined in Subsection 4.1.2) depends on the parameter ijj:

18 ndc where A is the Stokes-Cunningham correction factor (see Subsection 4.1.1), pp is the particle density, dp the particle diameter, vh the horizontal wind speed, fx the dynamic viscosity, and dc is the characteristic size of the collector (e.g. the diameter of a cylindrical wooden branch). Since the impaction efficiency increases with increasing i/f, particle removal by impaction is more effective if the wind speed is high, the particle diameter is great and the collector size is small. We must stress, however, that t;,, is the laminar flow speed. In a turbulent stream the impaction mechanism is obviously more complicated.

The dry deposition of aerosol particles (D') is generally measured by horizontal microscopic slide or so-called dustfall cans and jars. However, the results of such measurements, wide-spread in local pollution studies (Corn, 1976), have to be interpreted with caution because of the disturbance of the laminar and turbulent flow regime by the collector. Furthermore, the laminar layer covering the collector surface may be very different from that over soil and vegetation. In any case, if we also measure the particle concentration N, a parameter with the dimension of velocity can be defined (Junge, 1963):

This parameter is termed the deposition velocity.

Concerning the estimation of different dry removal processes let us consider Fig. 41, reported by Hidy (1973), but due to Chamberlain (1960). Curve A in this figure refers to experimental deposition velocities, measured over flat surfaces roughened by grass. On the other hand, line B represents the sedimentation velocity calculated by equation [4.1], It can be seen that curve A approximates line B in the range of very large particles with significant sedimentation velocities (</p>10 //m). With decreasing particle size the deviation is more and more important. It is believed that this phenomenon is caused by turbulent diffusion. There is some indication that curve A begins to increase with decreasing size in the range where Brownian motion becomes important.

To estimate the significance of impaction by trees we refer to the results of Neuburger and his co-workers (see Hidy, 1973), based on particle concentration measurements carried out inside and outside a forest area. Neuburger's data show that more than 80 % of pollen (coarse particles) is removed by trees while the corresponding value for Aitken particles is 34 %. Further research is needed, however, to confirm these results.

Was this article helpful?

## Post a comment