Determining the feasibility of photonic noses to improve in-field pest monitoring

点击翻译按钮获取中文摘要

The 'green revolution' has allowed significant crop production increases over the past sixty years. Greater innovation is required, however, to sustainably feed a human population predicted to reach 9.7 billion by 2050. This increase in population size will place significant pressure on the agricultural sector to achieve higher crop yields. Reducing crop losses within existing production systems will facilitate food security without increasing resource use. Invertebrate pests and plant pathogens reduce crop yields by up to 23 % so their control presents a means to offset food security concerns and provide wider economic stability. Conventional agricultural production systems depend on synthetic pesticides to minimise crop losses to pests and pathogens. There is, however, increasing pressure for growers to adopt integrated pest management (IPM) principles to reduce synthetic pesticide use as they are associated with negative impacts on human and environmental health. IPM reduces pesticide use and enhances crop yields by minimising invertebrate pest and plant pathogen build-up rather than curing infestations and is underpinned by monitoring, but this is generally too unreliable and expensive to implement effectively. Development of an automated, real-time monitoring platform will offset these issues and reduce crop losses. Plants emit low levels of volatile organic compounds (VOCs). Changes in VOC emissions occur when plants are subjected to stressful environment. Electronic noses are a technology that is potentially suitable for in-field plant health monitoring. These systems characterise the overall plant VOC profile to create a digital fingerprint that is compared to a 'normal' fingerprint to determine changes. Most electronic noses use sensors that suffer from sensitivity issues and aging effects, making them ineffective tools under field conditions. Within this project we intend to build an interdisciplinary community of agricultural science, optical sensing and machine learning experts to develop a novel plant health monitoring platform that enhances agricultural production through localised pest and disease monitoring that can be targeted with appropriate control measures.