Identifying key components of weed beet management using sensitivity analyses of the GeneSys-Beet model in GM sugar beet

Genetically-modified (GM) sugar beet varieties tolerant to non-selective herbicides would be useful for managing weed beet, an annual form of Beta vulgaris impossible to eliminate with herbicides in sugar beet. However, it is highly probable that the herbicide-tolerance transgene would be transmitted to the weed through pollen flow. It is therefore essential to study how weed beet. particularly Herbicide-Tolerant (HT) populations, develop in cropping systems and how to optimise crop succession and management for controlling these weeds. As multiple interactions and long-term effects make field experiments impractical, we carried out a simulation study with a deterministic and mechanistic model, G ene S ys- B eet , which quantifies weed beet dynamics and gene flow in cropping systems with interactions with climate, soil structure and hydro-thermal conditions. The sensitivity analysis consisted of 250 000 random combinations of input variables to rank cropping system components according to their effect on both total and GM weed beet infestations. Frequency of sugar beet crops, crop succession, manual and mechanical weeding and tillage were identified as the most important variables. Several cultivation techniques must be combined to efficiently control weed beet. Our recommendations are complex, but a delayed return of sugar beet in the rotation. Harvest should be followed as soon as possible by a shallow tilling; tillage should always be as shallow and as early as possible, except before sugar beet where mouldboard ploughing is advisable. If possible, sowing dates should be delayed. Sugar beet should be weeded mechanically and/or manually, aiming at late and efficient, rather than early or frequent operations. Herbicides should be applied whenever possible and target all weed beet stages and genotypes. Set-aside must be cut as frequently and as late as possible.