Precision weed control: a spatially explicit individual based model

Abstract

Yield reduction through competition between crop and weeds is a major challenge for farmers. Therefore, most fields in Europe are treated with herbicides or mechanical methods for weed control nowadays. Modern techniques like high precision GPS, image recognition and robotics enable autonomous and site specific weed control. In the future, it might be possible to manage weeds on a plant-by-plant basis. This would provide great opportunities for the conservation of weed diversity. The implementation of such management schemes requires the understanding of the competition process in a spatial context. We present a spatially explicit individual-based model suitable to simulate the effect of precise weed control measures for specific plants. Growth models in form of differential equations are set up for each individual plant. Competition is formulated dependent on the spatial configuration of both weeds and crops. First simulation results with a parametrisation for wheat and Viola arvensis are presented. Without any control measures, crop growth curves show a large variability, whereas an efficient weed control (efficacy 95%) results in nearly uniform growth curves. This was corroborated in an experiment with wheat and V. arvensis for the same spatial configuration as in the model. Further development of the model will include multispecies situations as well as long-term population dynamics of the different species.

Keywords

• competition
• individual based
• plant-plant interaction
• simulation model
• spatial explicit