SPATIAL VARIABILITY OF CS-137 AND THE INFLUENCE OF SAMPLING ON ESTIMATES OF SEDIMENT REDISTRIBUTION

The principal objectives of this study were to assess the spatial vari ability of Cs-137 within a cultivated field, and to examine the influe nce of grid size and positioning on estimates of sediment redistributi on. A 11 x 11 grid (6.25 ha) was sampled in a gently sloping cultivate d field in central Saskatchewan, Canada. The mean Cs-137 areal activit y in the cultivated field was 1910 Bq m-2, with a coefficient of varia tion (CV) of 28%. Spatial variability within the cultivated field was 55% greater than that in a nearby undisturbed (native) field, and this was assumed to result from the combination of wind erosion, localized water erosion, and possibly tillage redistribution. Lag one serial co rrelation was noted for two cultivated transects, and one downslope tr ansect showed a positive linear relationship between Cs-137 activity a nd distance. Therefore, estimation of an average Cs-137 areal activity based on only one 11-point transect in the cultivated field was inapp ropriate, and a sample size of 32 would be required to estimate the po pulation mean with 95% confidence and an allowable error of +/-10%. In the cultivated field redistribution of Cs-137 was used to assess sedi ment redistribution over the past 35 years. Sample locations in the cu ltivated field with Cs-137 areal activities less than the lower 95% co nfidence limit of the mean for the undisturbed field were considered t o be eroded, while locations with activities greater than the upper 95 % confidence limit were considered to be depositional. The cultivated grid was subdivided into four 6 x 6 grids to determine the influence o f grid size and positioning on sediment redistribution. Statistical te sts indicated that there were no significant differences between grids , and it was concluded that a 6 x 6 grid provided an optimal sample si ze, regardless of position, to assess the relative degradation of the cultivated field.