Abstract
To sustain a stable food supply without losses from various hazards, such as insects, diseases, or abnormal climate conditions induced by global change, application of genetically modified (GM) organisms is one potential tool for agriculture. However, GM crops do not exist in traditional agricultural environments; thus, the risks resulting from the application of GM crops to agriculture must be made clear. Crops, such as rice belonging to the Gramineae family, can be cross-pollinated by wind. This enables non-GM crops to be pollinated by GM crops. Prior to the cultivation of GM crops, the degree of cross-pollination must be estimated. This should take into consideration meteorological conditions, because of their importance in pollen flow. We constructed a system model to calculate the cross-pollination distribution by using data on the geographical distribution of GM donor and non-GM recipient crop fields, meteorological elements, and flowering data. The system consists of a main program to calculate cross-pollination rates, a program to include the isolation distance, and a sub-program to average the maps of the cross-pollination rates in recipient fields. The system can predict the regional cross-pollination rate by incorporating the isolation distance and differences in the flowering period. The system was applied to three areas around Tsukuba City, Ibaraki Prefecture, with different distributions of paddy fields and hypothetical donor (ratio 30%) and recipient (ratio 70%) fields. The general trend was that the cross-pollination rates were lower in areas of clustered donor fields. The calculated cross-pollination rate can mostly be explained by the length of the border between the donor and recipient fields. This is because rice pollen is dispersed only within short distances from donor fields. In order to clarify the influence of meteorological conditions on the variation in the cross-pollination rate, 10 years of simulations were performed. The cross-pollination rate varied by about a factor of three (0.03-0.09% for one of the simulated fields) during the simulated 10 years.
