Modelling the effects of cropping systems on the seed bank dynamics and the emergence of weed beet

Weed beet (Beta vulgaris) is a serious problem in sugar beet fields in many European countries and in the USA. This weed is the progeny either of accidental hybrids between sugar beet (ssp. vulgaris) and wild beet (ssp. maritima), or of bolted sugar beet plants in the case of varieties with low bolting resistance. Because of its proximity to the crop, the weed cannot be eradicated by herbicides in sugar beet crops. With the advent of genetically modified (GM) sugar beet varieties tolerant to non-selective herbicides, weed beet could also become tolerant to these herbicides because the sugar beet and weed beet are interfertile. It is therefore crucial to evaluate and develop cropping systems for managing weed beet. Consequently, we need models quantifying the effects of cropping systems on weed beet dynamics. Because of the seed longevity, the seed bank of weed beet constitutes a key step for these dynamics. The objective of the present work was to develop a model quantifying the effects of tillage, in interaction with soil climate and structure, on the seed bank dynamics and the emergence of weed beet. The model was based on sub-models predicting (a) soil environment (climate, structure) resulting from the cropping system and weather, (b) vertical soil seed distribution after tillage, depending on the tool, the characteristics of the tillage implement and the soil structure; and (c) seed mortality, dormancy, germination and pre-emergent growth depending on season, soil environment, seed depth and age. Seed mortality occurs only during autumn; seed dormancy increases during summer and autumn and decreases during winter; it also increases with seed depth. Germination is triggered by rain or tillage and driven by hydrothermal time; pre-emergent shoot elongation increases with thermal time; pre-emergent seedling mortality increases with soil clod size and seed depth. The sub-models for soil environment and seed movements were based on existing models; the weed beet sub-model was developed from our previously published studies and completed here with additional experiments. Simulations were carried out to show how crop rotation and tillage influence weed beet seed bank and emergence in different crops. (Résumé d'auteur)