Tracking ideal varieties for agroecological weed management in organic wheat. A simulation study

Abstract

Even though weeds are recognized as one of the main causes of yield losses in organic crop production, breeding for varietal competitiveness is still rare. The objective of the present study was to run simulations with the mechanistic crop-weed model FlorSys to identify (1) variety-intrinsic parameters that make organic wheat (Triticum aestivum L.) more competitive against weeds in organic conditions, and (2) wheat ideotypes, i.e., optimal combinations of wheat-parameter values to maximise yield, weed tolerance and weed suppression. FlorSys is based on an individual-based 3D canopy model, simulating daily crop-weed seed and plant dynamics over the years, depending on the cropping system and pedoclimate. The model includes three winter wheat varieties, and 20 virtual varieties were added by randomly combining variety-parameter values according to a Latin Hypercube Sampling (LHS) plan, respecting parameter ranges and correlations observed in the actual varieties. A global sensitivity analysis was run, using another LHS plan to combine wheat varieties, crop rotations and management techniques in 2800 contrasting organic cropping systems (e.g., undersowing with clover, no-till, type of weed flora) and 400 conventional systems. Average yield potential per variety was highly correlated to yield in the presence of weeds (R² = 0.79) but only slightly to weed tolerance, i.e., the opposite of yield loss due to weeds (R² = 0.10). Classification and regression trees (CART) identified parameters (notably, low pre-emergent seedling loss, fast post-emergence growth rate, late flowering) and their combinations that drive potential yield and competitiveness against weeds. If nitrogen and water were non-limiting after crop emergence, ideotypes that combine late flowering, late grain maturity and winter hardiness with early highly covering and homogeneous plant establishment, possibly to the detriment of root development, were the most relevant. If crop establishment was too late and/or unsatisfactory, a homogeneously tall canopy whose heavy plants presented a high harvest index was essential. No/low-till systems that rely primarily on mechanical weeding or herbicides needed early-emerging varieties with small/thick leaves and small plants per unit biomass to limit crop damage by weeding tools and improve herbicide penetration into the canopy, respectively. These results are suggestions for breeders and farmers to breed and choose wheat varieties for agroecological cropping systems aiming to better regulate weeds by biological interactions. Further research is needed to link model parameters to variables that could be measured routinely during breeding schemes.