Statistical model for managing the transfer of radiocesium to rice plants under various soil chemical properties: I. Model selection
2019
Yagasaki, Y. (Hama Agricultural Regeneration Research Centre, Fukushima Agricultural Technology Centre (Japan)) | Saito, T. | Niitsuma, K. | Sato, M. | Ota, T.
We generated statistical models to estimate radiocesium (sup(137)Cs) concentrations in rice grains according to assessments of soil chemical properties. The models were constructed by analyzing sup(137)Cs concentrations in soil and rice grains from paddy fields in Fukushima Prefecture collected by the Ministry of Agriculture, Forestry and Fisheries in 2012-2015 from a total of 290 fields, and data from two experimental fields with elevated risks of radiocesium transfer. A model with two explanatory variables, soil exchangeable sup(137)Cs and soil exchangeable potassium (K), well explained variations in sup(137)Cs concentrations in rice grains on log-transformed scales (R**2=0.68, P<0.001, n=309). Moreover, partial regression coefficients indicated that at a constant concentration of soil exchangeable K, sup(137)Cs concentrations increased in rice grains with increasing soil exchangeable sup(137)Cs concentrations. In addition, the effects of soil exchangeable sup(137)Cs concentrations on rice grain sup(137)Cs concentrations were nearly equal to that of soil exchangeable K. The trend in the sup(137)Cs concentration in rice grains estimated with the model for several different scenarios of soil exchangeable sup(137)Cs concentration agreed well with observed trends. These results show that the present model can be used to manage risks of sup(137)Cs transfer to rice.
اظهر المزيد [+] اقل [-]