LAUSR

Abstract Branched broomrape (Phelipanche ramosa (L.) Pomel) is a parasitic plant, which causes severe yield losses in major crops worldwide. Due to its broad host range, including numerous non-parasitic weed species, the persistence of its seeds in the soil, and the poor efficiency of available management techniques, broomrape management is complex. In a previous paper, we developed a broomrape-dynamics model called P hera S ys to support the design of management strategies combining multiple techniques aiming at long-term control of broomrape. Here, the objective is to use this simulation model to (1) check the consistency of simulations vs. literature data, (2) evaluate the potential of cropping systems to manage the combination of branched broomrape and weeds, (3) investigate whether weeds can biologically regulate parasitic plants in agroecosystems. Five contrasting cropping systems including different levers known to influence broomrape dynamics were simulated with different weather series. Four simulation series were run, with or without broomrape as well as with or without weeds, to discriminate the individual effects of weeds and broomrape on crop production as well as the effect of weeds on broomrape dynamics. Simulations with P hera S ys showed that delayed sowing in combination with the use of trap and catch crops are promising for reducing broomrape infestation and yield losses in the long term. Tolerating a temporary and/or low-density weed flora in such cropping systems could improve broomrape management because spring/summer weeds could reduce broomrape seed bank by triggering broomrape germinations that would not reproduce. During cash-crop growth, weed contribution to broomrape infection would be negligible. However, these conclusions are only valid if broomrape-attaching weeds reproduce before broomrape has time to do so, which needs to be checked with field experiments for most weed species.