Horticulture: Smart trap for improved early detection of vine weevil to enable successful application of integrated pest management

Vine weevil is an economically important pest of UK soft-fruit and ornamental crops that causes millions of pounds worth of damage every year. Controlling vine weevil utilises many elements of integrated pest management (IPM) but growers continue to experience economic losses due to this species. These losses can be largely attributed to a lack of an effective (determined by the sensitivity and reliability of the monitoring tool) monitoring tools leading to ineffective or inappropriate control applications being undertaken. This project will develop a semiochemical-based 'smart' monitoring tool to enable growers to make informed pest management decisions and unlock the full potential of current IPM compatible technologies. This is critical as growers have already switched from using synthetic chemical insecticides and instead rely on the use of crop hygiene and biological controls, but these approaches require monitoring to determine efficacy as well as to correctly time applications.

The novelty of this project is that it is the first to bring together improved knowledge of vine weevil visual and chemical ecology with a prototype 'smart' monitoring tool. With these tools it will be possible to design and test a novel commercial vine weevil monitoring tool. Key to improving the sensitivity and reliability of the monitoring tool will be the development of an effective lure. The ability to automatically and reliably detect the presence of vine weevils within crops means that growers can reduce labour costs whilst at the same time receiving real-time information on the pest status of a crop.

Our proposed work plan will be realised through the completion of four research objectives, 1. Develop an effective vine weevil lure; 2. Test monitoring tools with and without the addition of the vine weevil lure under semi-field conditions; 3. Test the best combination of monitoring tool design and vine weevil lure under commercial cropping conditions (soft-fruit and/or ornamentals); 4. Communicate results from the project through agronomist-led site visits.

Currently the presence of vine weevil adults is often determined through visual plant assessments. However, this approach is time consuming, does not produce timely data and may damage crops so more effective monitoring methods are urgently needed. Most research in this area has focused on the use of artificial refuges or traps for adults, here referred to as monitoring tools, but these tools do not provide growers with timely or reliably data on the pests status of their crops. Existing monitoring tools exploit the negative phototaxis behaviour exhibited by adult vine weevil, which causes individuals to seek shelter during daylight. Designs such as grooved wooden boards placed on the ground, corrugated cardboard wrapped around plant stems or rolls placed on the ground are often used. A purpose-designed vine weevil monitoring tool is available from ChemTica (Costa Rica) and works on the same principle. A key reason why these monitoring tools do not work effectively is that little consideration has been given to the design and a semiochemical lure has not been developed for this pest. Recent work by members of this project team has demonstrated the importance of size, colour, shape and entrance number in determining in monitoring tool efficacy. Furthermore, it has been demonstrated that vine weevils aggregate and exhibit attraction to conspecifics, suggesting that potential insect derived semiochemical sources could be found in aggregation or trail-following pheromones. There is then the potential to optimise the design of a vine weevil monitoring tool and to develop and effective semiochemical lure in order to provide growers with a highly sensitive and reliable vine weevil monitoring tool. The monitoring tool can be further enhanced by incorporating the machine learning model that exploits key morphological features associated with vine weevil (e.g., extended snout and elbowed antennae) to classify individuals with an accuracy of 85 %.