Keywords: holistic simulation, computational fluid dynamic, one-way coupling, sand transfer, desertification, fence.

Recently, concern with environmental problems increased and one of the typical examples is desertification. The speed of desertification in the world is 1,900m²/s and the total area becomes about 6 Mha every year. It is a critical problem for the people living in and around deserts. Even in Japan, a great amount of yellow sand from the Taklamakan Desert and the Gobi Desert in China causes health damage to the people of Japan. There is an urgent need to stop or suppress desertification.

Traditionally, tree planting and windbreak-fence setting have been carried out to protect farms and towns from sand movement. However, these are ad hoc remedies, and their effects are not clear and are limited to the local area. This comes from the fact that the physics of sand transfer has not been clarified yet, especially based on fluid dynamics. To clarify the physics, field experiments in deserts are necessary. However, since the environment of a desert includes too many different conditions, it is difficult and of time consuming to conduct field experiments. In the sense, it is expected that numerical simulations i.e. computational fluid dynamics (CFD) are useful to stop or suppress desertification. However, the numerical procedure and the physical model have not yet been established. For example, it is very difficult to estimate the formation and movement of a dune with spatial-temporal large scale. Many papers which deal with sand transfer can be found in the literature. For example, Zhang et al. [1] and Herrmann and Sauermann [2] conducted the numerical simulation of various dunes, but their results were not satisfactory in quality. Furthermore, although a number of physical models for sand transfer have been proposed, still it is not clear which model is the best. Therefore, in order to stop desertification, it is necessary to develop a numerical procedure to correctly predict sand transfer.

In the present investigation, we develop a holistic simulation code to predict sand transfer, focusing on the saltation process. In the code, turbulent flow field, sand particle movement by saltation, and modification of the ground surface geometry are iteratively computed. First, the code developed is verified with the temporal change of a dune shape by a constant wind. Second, it is applied to the flow around a dune with a fence which is arranged to suppress sand transfer. Investigating the numerical results, it is confirmed that the most active sand transfer occurs between the reattachment point and the dune crest, a higher fence makes a new and small dune downstream of the initial dune, and it is difficult to stop sand transfer from one fence to another.

1

R. Zhang, T. Kawamura, M. Kan, "Numerical Simulation of Formation and Movement of Various Sand Dunes", New Developments in Computational Fluid Dynamics, 165-174, 2005. doi:10.1007/3-540-31261-7_15

2

H.J. Herrman, G. Sauermann, "The Shape of Dunes", Physica A, 283, 24-30, 2000. doi:10.1016/S0378-4371(00)00124-2

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £140 +P&P)