Analysis of potential radiocaesium soil contamination: A case study using two different simulation models

An agricultural area near to a nuclear power plant in Northern Italy was used to evaluate the radiocaesium leaching predictions obtained by employing two models which differently approach the solute sorption process (linear equilibrium sorption or diffusion-controlled sorption kinetics). Assessments were made through both statistical tests and radiological risk map comparison. Twenty-eight soil samples were taken based on a wide regular grid and analysed for routine physicochemical soil parameters. Specific sorption parameters were calculated by empirical equations derived from regression analyses of literature data. The point data obtained by deterministic simulations were then interpolated by means of geostatistical techniques (i.e. ordinary and disjunctive kriging) to accomplish maps of the 137Cs concentration in the examined topsoils. The statistical analyses (t-test and F-test) did not show any significant difference between the two simulation sets. On the contrary, graphical information derived from maps macroscopically pointed out that the dynamic model based on the nonequilibrium approach predicted a higher 137Cs content leaching into both soil types in agreement with the available field data on the local radiocaesium migration into soil profiles.