Calibration of forest

Area of Science:

• Environmental Science
• Ecosystem Dynamics
• Radiological Sciences

Background:

• Accurate prediction of cesium-137 (137Cs) environmental fate is crucial for managing radioactively contaminated forests.
• Understanding 137Cs concentrations in various tree parts and soil compartments is essential for effective forest management strategies.

Purpose of the Study:

• To calibrate the Forest RothC and Cs (FoRothCs) model for predicting 137Cs dynamics in forest ecosystems.
• To assess the utility of approximate Bayesian computation (ABC) for model calibration using multi-compartment 137Cs concentration data.

Main Methods:

• The Forest RothC and Cs (FoRothCs) model was calibrated using six years of observational data from four Fukushima Prefecture forest sites.
• An approximate Bayesian computation (ABC) technique was employed, using environmental decay constants of five compartments (leaf, branch, stem, litter, soil) as summary statistics.
• Inferred parameters related to 137Cs transfer processes were compared with existing literature values.

Main Results:

• The ABC technique successfully reconciled FoRothCs model outputs with observed 137Cs concentration trends across all study sites and compartments.
• Estimated model parameters, including root uptake rates, align with values reported in scientific literature.
• Model calibration using ABC significantly reduced prediction uncertainty for 137Cs dynamics.

Conclusions:

• Approximate Bayesian computation (ABC) is an effective method for calibrating forest ecosystem models like FoRothCs.
• Calibrating with multi-compartment 137Cs concentration data improves the accuracy of predicting radiocesium dynamics in forests.
• This approach enhances the reliability of models used for managing radioactively contaminated forest environments.