Handout (1.5 MB)
First, the ultra-sonic anemometer-thermometer data during the sand storm periods in 2000-2004 have been analyzed. These data are wind velocities with 10Hz frequency, and obtained at three layers of the 325m meteorological tower in Beijing. It is revealed that the air motion during sand storm can be divided into three parts, the turbulent small eddies, the gusty wind which is a kind of coherent structure, and large scale basic flow. Their time scales are 0-1 min., 1-10 min. and larger than 10min. respectively. All they transport horizontal momentum downwards to the ground surface and contribute to the soil erosion and sand-dust emission, hence the friction velocity should consist of their combination. However, in the lower part of atmospheric boundary layer, the descending component of basic flow suppresses the dust particles keeping them within the bottom layers. But the gusty wind has coherent structure, and due to the gusty wind not turbulence, the dust can entrain from lower to upper levels.
Second, we use Lagrangian Stochastic Model to simulate the particle trajectory during sand storm period. The random walk of particles is a Markov process, or a Brownian motion called Wiener type. There are three classical categories of particle motion determined by the diameter of particles: Creep with d>500µm, saltation with 70µm