REVIEW OF RESUSPENSION MODELS 215 



velocity for initial movement has a minimum at about 0.15 mm (150 idm) with a friction 

 velocity of about 0.15 m/sec; i.e., grains smaller than this size and grains larger than this 

 size require higher friction velocities to initiate any movement. The stabiUty of the finer 

 grains is illustrated by the simple experiment of Bagnold (1943), in which he placed a pile 

 of talcum powder on a smooth surface and exposed it to winds. Tlie layer was stable at 

 relatively liigh wind speeds. However, a few grains of larger particles sprinkled on the pile 

 resulted in rapid dispersion at a wind speed much lower than would serve to disperse the 

 particles without tliis added factor. It is believed that the relative stability of the small 

 particles is due to the fact that they do not protrude above the laminar layer; thus no 

 drag force is exerted on them. The minimum in the curve of velocity required to institute 

 movement and particle size, then, is due to the balance between the increasing drag force 

 as the particle increases in size and the increasing downwind force from gravity as the 

 particle becomes larger. Above about 0.15 to 0.2 mm, the threshold velocity required to 

 initiate movement increases as the square root of the product of the particle diameter and 

 its specific gravity (Bagnold, 1943; Chepil, 1945a). 



As a result of this threshold friction velocity, it is apparent that direct aero- 

 dynamic pickup of small particles from the soil is unlikely. Instead, the process of 

 saltation is the key to producing the suspended fraction because the impact of these 

 particles as they strike the ground provides the energy to propel the smaller particles 

 above the laminar layer into the wind stream where they are transported by eddies in the 

 wind. Tlius, in the talcum powder example, the sand particles sprinkled on the talcum 

 powder served the function of the saltating particles. In a field, knolls, ridges, sand 

 pockets, or other areas most exposed to the wind and/or containing the easily erodible 

 grains start to erode at a lower velocity than the rest of the field. Once the erosion starts, 

 it spreads downwind, and the bombarding action of the particles in saltation causes 

 movement in other parts of the field that normally would not be eroded (Chepil, 1945a). 

 The threshold velocity of the field is therefore the threshold of the most exposed or most 

 erosive spots in the field. Since the avalanching effect of saltation increases down the field 

 in the direction of the wind, the length of the field is also a factor in the degree of 

 erosion. 



An important consequence of the role of sahation in the production of resuspension 

 is that there will be no dust, or particles flowing in suspension, unless the wind speeds are 

 great enough to produce sahation under the conditions of the tleld. This places a 

 threshold condition on the wind speeds required to resuspend particles from the ground. 



Bagnold (1943) measured the rate of soil flow for desert sands and found that these 

 rates could be described by an equation of the form of 



q = Cu|^ (4) 



where q = rate of soil flow (grams per centimeter width per second) 

 p - density of the air 

 g = acceleration of gravity 

 C = constant that differs for differing soils and forms of erosion 



Bagnold concluded that on the desert sands the tlow in suspension was small, about y2o^/' 

 of the total flow, as compared with saltation and surface creep. Chepil (1945a; 1945b) 

 made similar measurements on agricultural soils in a wind tunnel. His results indicated 



