Comparaison de méthodes de conception de systèmes de culture innovants pour la gestion durable des adventices

2020 
Trois demarches de conception de systemes de culture ont ete comparees sur leur contribution a la gestion durable des adventices : (i) conception par expertise d’acteurs multiples ; co-conception par des agriculteurs (ii) d’un systeme innovant lors d’ateliers ponctuel ou (iii) de systemes innovants pour chaque agriculteur d’un groupe sur une demarche de long terme. Les systemes definis ont ete simules avec le modele de dynamique de la flore adventice FLORSYS. Une Analyse en Composantes Principales des resultats de six indicateurs decrivant la nuisibilite des adventices pour la production, la diversite des adventices et l’intensite d’usage d’herbicides a ete realisee. Tout d’abord, la nuisibilite des adventices et l’usage d’herbicides ne sont pas correles. Ensuite, les systemes innovants concus selon la methode (i) conduisent a des changements plus importants et contrastes, a l’inverse de la co-conception par des agriculteurs, particulierement avec des systemes concus pour chaque agriculteur. Enfin, l’evolution de la durabilite a ete evaluee avec le modele d’evaluation multicritere DEXiPM. La durabilite evolue plus frequemment, de maniere positive pour les methodes de conception (i) et (ii). Neanmoins, les seuls systemes innovants combinant reduction significative de la nuisibilite des adventices pour la production et augmentation de la durabilite sont issus du groupe (iii), avec des modifications profondes de la rotation (reduction du ble tendre, abandon de la betterave) envisagees apres 10 ans de conception pas-a-pas. Ces resultats illustrent les differences et complementarites de ces differentes methodes. Abstract : Comparing methods for designing innovative cropping systems aiming at sustainable weed management Three methods of innovative design for cropping systems were compared for their contribution to sustainable weed management: (i) de novo expert design by multiple stakeholders; co-design by farmers (ii) of one cropping system during workshops; (iii) of cropping systems for each farmer member of a DEPHY network group over several years. The weed simulation model FLORSYS was used to evaluate these systems with six indicators, describing weed harmfulness for crop production, weed contribution to biodiversity and herbicide use. A principal component analysis on this dataset illustrated that there is no correlation between herbicide use and weed harmfulness for crop production. De novo expert design led to bigger changes in practices and systems performances, compared to co-design with farmers, especially Cavan N., et al. 190 Innovations Agronomiques 81 (2020), 189-200 when innovative systems are designed for each farmer of a group. The evolution of sustainability of these cropping systems was assessed with the multicriteria assessment model DEXiPM and again more changes, often positive, were found with the design methods (i) and (ii). However, only a few innovative cropping systems designed with method (iii) were able to reduce weed harmfulness for production while enhancing sustainability, thanks to major changes planned in crop sequences (winter wheat reduction and sugarbeet removal), after ten years of step-by-step design. These results illustrate differences and complementarities of these design methods.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []