Fungi Contaminated by Radionuclides: Critical Review of Approaches to Modeling

INTRODUCTION Data derived from sampling at areas contaminated by the Chernobyl fallout have demonstrated the organic layer of soil to be a major pool of radiocesium in forest ecosystems (Linkov and Schell, 1999, Ruhm et al., 1996). Fungi can directly bind or precipitate radionuclides; they also can indirectly affect radionuclide speciation and mobility in forest soils (Gadd, 1996). Fungi are the greatest living biomass in the decomposing organic layers of forest soil. They are the primary sources of enzymes necessary to degrade the litter. Fungi are one of the most important components of forest ecosystems, since they determine to a large extent the fate and transport processes of radionuclides in forests (Linkov and Von Stackelberg, 1999; Riesen et al., 1999; Linkov and Schell, 1999). Understanding radionuclide accumulation by fungi is also essential for the development of remediation technologies for contaminated areas. Phytoremediation is emerging as an attractive alternative to high-cost traditional cleaning methods for large areas polluted with low to moderate levels of heavy metals and radionuclides. Recent studies show that phytoremediation of soils contaminated by Cs may be feasible (Last et al., 1997). Mycorrhizal fungi can maximize radionuclide accumulation and removal from contaminated soil by plants (Entry et al., 1996). This paper is not intended to provide a comprehensive review of biological processes, but rather to critically review existing modeling techniques. In the recent International Atomic Energy Agency model-model and model-data intercomparisons (BIOMASS, 1999) the radionuclide accumulation in fungi was found to be the most difficult to predict. Developing reliable radioecological models to describe the radionuclide accumulation in fungi requires an evaluation of many uncertain parameters and variables. These parameters have been found to be especially uncertain as compared to other forest compartments. Many parameters are represented by aggregated values averaged over time, space and species. The uncertainty of these parameters should be considered as an integral part of the modeling process. Excessive model complexity will result in a limited utility of the model, since many uncertain model parameters decrease the overall reliability of model predictions. On the other hand, using over-simplistic models in situations when a better understanding of the ecosystem has already been achieved, might also lead to unrealistic model results that could possibly contradict available experimental data.