Abstract
In mechanical weed control research, there is a tendency to prioritise practical application over the establishment of a solid theoretical foundation. This emphasis can lead to a fragmented knowledge base and relatively modest research progress. The widespread use of ANOVA exacerbates this issue by primarily focusing on the differences between treatments, neglecting an exploration of the mechanisms behind the observed results. In response to these limitations and in line with rapid advances in sensor and robotics technology, this paper presents a proposal to establish a theoretical framework for mechanical weed control. The framework aims to emphasise understanding of the underlying mechanisms of mechanical weed control, rather than solely focusing on their observable outcomes. As part of this proposal, a simulation model known as HarrowSim based on regression parameters is presented. Simulation runs show the importance of factors such as treatment intensity, selectivity, crop tolerance and weed pressure in relation to crop yield response to post-emergence weed harrowing. While currently parameterised for post-emergence weed harrowing, it is argued that HarrowSim is relevant for all post-emergence mechanical weed control methods, and is useful for teaching purposes and for inspiring future research. While the theoretical framework and HarrowSim have their weaknesses, these can also be perceived as opportunities since they can help focus attention on key issues for future research.
Weed Research, EarlyView. Read More