Deriving probabilistic soil distribution coefficients (K_d). Part 1: General approach to decreasing and describing...

Area of Science:

• Environmental Geochemistry
• Radiochemistry
• Soil Science

Background:

• Accurate radionuclide distribution coefficients (K_d) are crucial for environmental risk assessment.
• Existing K_d datasets often exhibit high variability, limiting their direct application in modeling.
• Soil properties significantly influence radionuclide sorption and mobility.

Purpose of the Study:

• To develop a general approach for deriving probabilistic radionuclide K_d values with reduced uncertainty.
• To identify key soil factors influencing K_d variability, using uranium (U) as a case study.
• To evaluate the utility of non-soil geological materials for augmenting soil K_d datasets.

Main Methods:

• A systematic subsetting approach was applied to a comprehensive K_d dataset.
• Subsets were defined based on soil factors (pH, organic matter, texture) and experimental methods (sorption/desorption).
• Statistical analysis was performed on uranium K_d values and analogue data from geological materials.

Main Results:

• Soil pH and organic matter content were identified as major factors reducing uranium K_d variability.
• The combination of pH and organic matter allowed for K_d confidence intervals as narrow as two orders of magnitude.
• Subsoil and till data were found to be suitable for enhancing soil uranium K_d datasets, unlike gyttja.

Conclusions:

• The proposed method effectively reduces uncertainty in probabilistic K_d estimates by utilizing site-specific geochemical data.
• pH and organic matter are key parameters for refining uranium K_d values in soil.
• Analogue data from specific geological materials can supplement soil K_d data, improving risk modeling reliability.