220 TRANSURANIC ELEMENTS IN THE ENVIRONMENT 



proportion of the total aerosol. It was also noted that the concentration of larger particles 

 increased with the wind speed. 



Travis (1975; 1976) has developed a model for the redistribution of wind-eroding 

 soil-contaniinant mixtures using the Gillette and Blifford (1972) and Gillette, Blifford, 

 and Fryrear (1973) relationships for the vertical and horizontal flux. Tliis model assumes 

 that the contaminant is closely associated with the soil aggregates and moves in the same 

 manner as the soil. Since it incorporates only studies from eroding agricultural soils, it 

 should be Umited to these soil conditions. 



Shinn et al. (1976) measured the vertical profile of dust in the atmosphere at two 

 sites and related these profiles to the vertical dust flux by the eddy-correlation method. 

 The two sites were an area at NTS where plutonium contamination had occurred during a 

 series of nonnuclear tests over 20 yr earlier (GMX Area) and in an agricultural field in 

 west Texas. Measurements were made simultaneously of the dust flux by an optical 

 particle detector and the mean wind and temperature profiles. 



It was noted that the mass-concentration particle size distributions at both sites had a 

 maximum at about 4 or 5 iim mass median aerodynamic diameter, and tliis decreased by 

 an order of magnitude at 1 and 10 jLim. This distribution does not agree with the data of 

 Chepil and Woodruff (1957) or Sehmel (1976a), which show significant quantities of 

 particles much greater than 10 lum at elevations up to 30 m above the ground. Shinn et al. 

 (1976) indicate that this could be due to reaggregation after sampling because the 

 sedimentation velocity of such particles would be greater than u^. However, this remains 

 an uncertainty tliat requires investigation. 



The wind-profile measurements showed the roughness length (zq) at the Texas site to 

 be 0.044 cm and at the GMX site to be 2.0 cm. This resulted in a drag coefficient 

 referenced to the wind speed at 2 m (U:^/u2oo) of 0-05 at the Texas site and of 0.10 at 

 the GMX site. It was noted that the dust profiles in this study, as well as in previous ones, 

 fit a power law with exponents of either —0.2 or —0.35. Tliis is due to the fact that all are 

 nearly bare surfaces and the measurements were made in dynamic neutral atmospheric 

 stability conditions. The dust flux calculation was parameterized by several simple 

 relations. 



F = k| (7) 



where F - flux 

 Z = height 



X = dust concentration 

 K - eddy diffusivity 



Since 



K = u^ kz (8) 



where k is the van Karman constant (k = 0.4). Since the dust concentration follows a 

 power-law distribution with height, 



dz z 



