Monitoring invasive non-native plant species using hyperspectral remote sensing data: a case study on the West Highland Way in Scotland

Abstract

This study investigates the application of hyperspectral remote sensing for monitoring Rhododendron ponticum, an invasive non-native species, along the West Highland Way railway in Scotland. The primary objectives were to evaluate the effectiveness of aerial hyperspectral imagery in detecting rhododendron at a regional scale, assess the consistency of classification model performance over increasing scale, and identify key factors affecting the spectral separability of rhododendron from other vegetation. The results demonstrated that hyperspectral imagery can effectively identify Rhododendron, though the accuracy of all models was highly impacted by the heterogeneity of the surrounding vegetation, highlighting the challenges of spectra separability due to the similarity between rhododendron and other green vegetation species. The findings underscore the potential of hyperspectral remote sensing as a scalable, cost-effective tool for invasive species monitoring, while also demonstrating a need for further research into integrating temporal datasets to enhance detection precision and reduce commission errors. This work contributes to the broader field of environmental monitoring and invasive species management, offering insights into the use of advanced remote sensing technologies for biodiversity conservation in the future.