Long-term reduction in (137)Cs concentration in food crops on coral atolls resulting from potassium treatment.

Bikini Island was contaminated on March 1, 1954 by the Bravo detonation (U.S. nuclear test series, Castle) at Bikini Atoll. About 90% of the estimated dose from nuclear fallout to potential island residents is from cesium-137 ((137)Cs) transferred from soil to plants that are consumed by residents. Thus, radioecology research efforts have been focused on removing (137)Cs from soil and/or reducing its uptake into vegetation. Most effective was addition of potassium (K) to soil that reduces (137)Cs concentration in fruits to 3-5% of pretreatment concentrations. Initial observations indicated this low concentration continued for some time after K was last applied. Long-term studies were designed to evaluate this persistence in more detail because it is very important to provide assurance to returning populations that (137)Cs concentrations in food (and, therefore, radiation dose) will remain low for extended periods, even if K is not applied annually or biennially. Potassium applied at 300, 660, 1260, and 2070 kg ha(-1) lead to a (137)Cs concentration in drinking-coconut meat that is 34, 22, 10, and about 4% of original concentration, respectively. Concentration of (137)Cs remains low 8-10 y after K is last applied. An explanation for this unexpected result is discussed.