LES analysis on impact of the forest canopy arrangement on dry deposition

Nakayama, Hiromasa   ; Kadowaki, Masanao   ; Yoshida, Toshiya   

Dry deposition is a phenomenon that gaseous and particulate substances deposit onto ground surfaces by atmospheric turbulence and/or gravity settling. Because forest areas have larger surface areas than those over grass or bare soil, air pollutants are efficiently deposited onto them. Especially, in the upstream edge of forest canopy, total amount of dry deposition becomes highly large by the effects of active advective and turbulence transport and large dry deposition velocity. In general, meso-scale meteorological simulation (MMS) models predict deposited amount at each computational grid based on several tens kinds of land use categories. However, real forest areas consist of various scale of forest patches. Such inhomogeneous forest canopy arrangements are not considered in sub-grid scale in MMS models. There are various methods of investigating dry deposition under the influence of forest area arrangements in detail, e.g., observations and computational fluid dynamics (CFD). It is well known that observations are a rational tool to provide reliable data. However, those are time-consuming, costly, and have a limitation of capturing detailed spatial distributions. On the other hand, the CFD technique has been recognized as a helpful tool with the rapid development of computational technology. In this study, we conducted LESs of turbulent flows, dispersion and dry deposition for different arrangements of two-dimensional forest canopy under a condition of the same area, and investigated the influence of forest canopy arrangements on dry deposition. It is found that the total amount of dry deposition shows a tendency to become larger as the forest canopy is arranged more sparsely and the total area of the canopy edge becomes larger. Our simulation results imply that it is expected to improve computational accuracy by considering the forest canopy arrangements in sub-grid scale when predicting dry deposition by MMS models.