A comparison of techniques used to estimate the amount of resuspended soil on plant surfaces.

The objectives of this study were to compare four common techniques used to estimate soil mass loadings on plant surfaces and to assess the need to account for particle-size distributions of both the soil tracer and contaminant of concern within the soil. Soil loadings (g soil kg-1 dried plant) from split samples collected in a pasture near Chernobyl were estimated using soil tracers of plutonium analyzed via alpha spectroscopy (mean +/- standard error; 1.0 +/- 0.2), titanium analyzed with an inductive coupled plasma spectrometer; (3.6 +/- 0.6), and neutron activation analysis for scandium (8.1 +/- 1.6), as well as simply washing the soil off the vegetation (34.1 +/- 5.6) Differences were significant at p < 0.001. We also found that soil loading estimates from any one technique varied by a factor of 10 depending on the soil particle size used in the calculations. This was because soil loadings decreased when smaller-sized soil fractions dominated the resuspension process. However, the percent of the plant's total contamination attributable to soil loading increased with smaller soil particles. Smaller soil particles apparently contribute less to the mass of soil loading (g soil kg-1 dry plant), but more to the total plant contamination (Bq) because of the higher concentration of contaminant found in the smaller-sized soil fractions. Differences in mass loading estimates due to the technique chosen (a factor of 10), or due to differences in elemental concentration as a result of the soil particle size used in the calculation (also a factor of 10), were greater than the natural variability observed in the field (2.5).