Applying the bootstrap technique for studying soil redistribution by caesium-137 measurements at basin scale

The use of the bootstrap technique to estimate the reference level of Cs-13 7 in an uneroded site is tested. The analysis is developed using Cs-137 mea surements made in a small experimental Sicilian basin. In the reference are a the Cs-137 activity is normally distributed with a known sample mean valu e, in equal to 94.4 mBq cm(-2). The influence of Cs-137 reference site samp ling was determined generating samples having a fixed size, N and six diffe rent values of the sample coefficient of variation, CV, by a Monte Carlo te chnique. Then, for each size N, the probability distribution of the mean mu of the sequences generated by Monte Carlo technique is defined. The soil r edistribution is determined, both at morphological unit and basin scale, us ing the proportional method of Martz & de Jong for calculating the net soil loss. The analysis showed that the spatial distribution of the net soil lo ss E-i and the basin value E-b are independent of the sample size, N, and t he coefficient of variation, CV, of the samples drawn from the reference ar ea, if the bootstrap technique is applied for estimating the mean mu(mu) to use as reference value. The soil redistribution is also examined using as reference value the quantiles mu(2.5), mu(25), mu(75), mu(97.5) correspondi ng to a frequency F(mu) equal to 2.5, 25, 75 and 97.5%, respectively. In co nclusion, the analysis established. that a robust estimate of the reference value can be obtained even in fields where a small number of samples was d rawn (high CV of the Cs-137 activity of the field samples), using the boots trap technique for generating sequences of reference values having known me an value in (the mean value of the Cs-137 activity of the drawn field sampl es) and large sample size (N = 50).