A FUNCTIONAL MODEL OF RADIOCESIUM TURNOVER IN BROWN TROUT

We describe a mechanistic model to explore the environmental and metabolic determinants of radiocesium kinetics in a freshwater fish, the brown trout (Salmo trutta). Our model, which incorporates submodels for fish growth, food consumption, ambient temperature, and radiocesium turnover, was validated and calibrated to observations in three year-classes of brown trout from a Norwegian lake contaminated with Chernobyl fallout. Observed growth and independent estimates of food consumption matched simulations. Whereas our model successfully described trends and seasonal dynamics of radiocesium in brown trout, it was highly sensitive to the temperature submodel with model uncertainty largely driven by uncertainties in absorption efficiency, feeding rate, and temperature. We used the model to simulate both fallout and steady-state (constant prey radiocesium) conditions. Peak radiocesium in fish after a fallout depended upon time of year, fish size, rate of food consumption and growth, and water temperature regime. Among the modeled lake types (temperate, boreal, and subalpine), peak radioactivity was highest for brown trout living in the warmest lake, feeding and growing at a maximum rate, and for fallout occurring early in spring. In these lakes, increasing growth rate from 80 to 100% of the maximum nearly doubled peak radioactivity. The model predicted that the relationship between body size and fish radioactivity changed from negative to positive within 2 yr after a fallout. At steady state, the model predicted seasonal dynamics in accumulation of radiocesium, with maximum biomagnification in autumn and minimum in early spring. Maximum biomagnification was related to fish size (weak negative slope), growth rate, and temperature regime, and a metamodel was developed for predicting biomagnification.