Holden, Claire and McAinsh, Martin and Taylor, Jane (2023) ATR-FTIR Spectroscopy-Linked Chemometrics : A Novel Approach to the Analysis and Control of the Invasive Species Japanese Knotweed. PhD thesis, Lancaster University.
![[thumbnail of 2023HoldenPhD]](https://eprints.lancs.ac.uk/style/images/fileicons/text.png)
Abstract
Japanese knotweed (Reynoutria japonica), an invasive plant species, causes negative environmental and socio-economic impacts. A female clone in the United Kingdom, its extensive rhizome system enables rapid vegetative spread. Plasticity permits this species to occupy a broad geographic range and survive harsh abiotic conditions. It is notoriously difficult to control with traditional management strategies, which include repetitive herbicide application and costly carbon-intensive rhizome excavation. This problem is complicated by crossbreeding with the closely related species, Giant knotweed (Reynoutria sachalinensis), to give the more vigorous hybrid, Bohemian knotweed (Fallopia x Bohemica) which produces viable seed. These species, hybrids, and backcrosses form a morphologically similar complex known as Japanese knotweed ‘sensu lato’ and are often misidentified. The research herein explores the opportunities offered by advances in the application of attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy-linked chemometrics within plant sciences, for the identification and control of knotweed, to enhance our understanding of knotweed biology, and the potential of this technique. ATR-FTIR spectral profiles of Japanese knotweed leaf material and xylem sap samples, which include important biological absorptions due to lipids, proteins, carbohydrates, and nucleic acids, were used to: identify plants from different growing regions highlighting the plasticity of this clonal species; differentiate between related species and hybrids; and predict key physiological characteristics such as hormone concentrations and root water potential. Technical advances were made for the application of ATR-FTIR spectroscopy to plant science, including definition of the environmental factors that exert the most significant influence on spectral profiles, evaluation of sample preparation techniques, and identification of key wavenumbers for prediction of hormone concentrations and abiotic stress. The presented results cement the position of concatenated mid-infrared spectroscopy and machine learning as a powerful approach for the study of plant biology, extending its reach beyond the field of crop science to demonstrate a potential for the discrimination between and control of invasive plant species.