from 4 x 1073 cm for the smooth surface to 1 m for a forest. Roughness height is calculated from the tog-linear velocity profile and is the height at which the extrapolated velocity profile reaches zero velocity. Ranges of measured average resuspension rates are correlated as a function of measured or estimated surface roughness heights in Figure 12 (Sehmel, 1975b). Resuspension rates range seven orders of magnitude from 10-!0 to 1073 fraction resuspended/sec. The practical significance of these numbers 10" The initia] resuspension rate correlation indicates that resuspension rates decrease as surface roughness increases, at least for the three smaller we roughness heights. T I T 1 T can be made apparent by noting that a year is 3.2 x 10? sec. NOTE: 1 YEAR = 3.2 x 107 sec However, measured resuspension rates for DDT sprayed on 4 a forest are two orders of magnitude greater than rates for the desert soil (Orgill et_al., 1976). This is an unexpected and unexplained increase in resuspension rates. A possible explanation of the increase might be increased -4 10 resuspension caused by tree movement in the wind. Also, various other grass differences in controlling variables and experimental factors may have influ- enced results. Since the data are so extremely limited, this apparent corre- lation should be used with extreme caution until] correlations are developed based upon several physical parameters instead of only zg. Nevertheless, this correlation does give some justification for estimating resuspension rates until better correlations are developed. ¢ - 3 io? éSo - y— = 10 One resuspension modeling concept is that deposited pollutant particles lose Got tictes when resuspended (Sehmel, 1975a). If the physics of soil resuspension were understood more adequately, it might be anticipated that the soil particle resuspersiton models could be used as a Simple basis for developing predictive contaminant particle resuspension models. Consequently, soil resuspension was investigated by experimentally determining rates at which airborne dust concentrations in the respirable size range increased with increasing wind speed. Experiments were conducted as a function of wind speed on the Hanford reservation at three sites. Increases in airborne particle number distributions as a function of wind speed are shown in Figure 13 for particle diameters of 1.1, 1.9, and 3.6 um. In each case, data points are plotted at the lower value of each experimental wind speed increment. Wind speed increments are indicated by the horizontal lines extending to the right from each data point. Average airborne particle number concentrations increased rapidly and reproducibly with increases in wind speed. In the left portion of the figure, concentrations increased as the 2.9 power of wind speed for wind speeds between 4,5 and 9.4 m/sec during sampling periods beginning on March 3 and May 6. Concentrations in the summer months from May 6 to September 18, 1974 were greater than for the spring months from March 4 toApril 25, 1974, but were less than for late winter months from January 16 to February 8, 1974, 2.9m 1D ALUMINUM TUBE ‘\ \\ 6 | AIRBORNE SOIL CONCENTRATIONS AS A FUNCTION OF WIND SPEED their identity as discrete particles, becoming attached to host soil par- 10 pm uranine FROM NX \. \ \ \. 5 £ 310° J G - ‘ \ s. e 10 < " oor 0 103 \ \\ \ \ Nv sf Mo TRACER FROM ———+$ 30 -11 DOT FROM FOREST-——> ,/ \ Zn$ FROM ‘ ASPHALT SURFACE / / rd ; / / DESERT SOIL i 102 l | 1 ig! l 10 L 10 ROUGHNESS HEIGHT, Zo, cm FIGURE 12. Initial Correlation of Wind Caused Resuspension Rates 202 203