Zusammenfassung
Tools for estimating pollen dispersal and the resulting gene flow are necessary to assess the risk of gene flow from genetically modified ( GM) to conventional fields, and to quantify the effectiveness of measures that may prevent such gene. ow. A mechanistic simulation model is presented and used to simulate pollen dispersal by wind in different agricultural scenarios over realistic pollination ...
Zusammenfassung
Tools for estimating pollen dispersal and the resulting gene flow are necessary to assess the risk of gene flow from genetically modified ( GM) to conventional fields, and to quantify the effectiveness of measures that may prevent such gene. ow. A mechanistic simulation model is presented and used to simulate pollen dispersal by wind in different agricultural scenarios over realistic pollination periods. The relative importance of landscape-related variables such as isolation distance, topography, spatial configuration of the. elds, GM field size and barrier, and environmental variation are examined in order to. nd ways to minimize gene. ow and to detect possible risk factors. The simulations demonstrated a large variation in pollen dispersal and in the predicted amount of contamination between different pollination periods. This was largely due to variation in vertical wind. As this variation in wind conditions is difficult to control through management measures, it should be carefully considered when estimating the risk of gene. ow from GM crops. On average, the predicted level of gene. ow decreased with increasing isolation distance and with increasing depth of the conventional. eld, and increased with increasing GM field size. Therefore, at a national scale and over the long term these landscape properties should be accounted for when setting regulations for controlling gene. ow. However, at the level of an individual. eld the level of gene flow may be dominated by uncontrollable variation. Due to the sensitivity of pollen dispersal to the wind, we conclude that gene. ow cannot be summarized only by the mean contamination; information about the frequency of extreme events should also be considered. The modeling approach described in this paper offers a way to predict and compare pollen dispersal and gene. ow in varying environmental conditions, and to assess the effectiveness of different management measures.
Altmetric