Mapping woody invasive species over large geographic regions using broadband multispectral sensors: A systematic review of progress, challenges, and opportunities

Woody invasive plants are a significant threat to biodiversity and ecosystem functioning, especially in large and heterogeneous landscapes. This paper provides a systematic review of mapping these species across large geographic regions using remote sensing, particularly freely available broadband multispectral data from sensors such as Sentinel‑2 and Landsat 8. A total of 26 peer‑reviewed studies published between 2015 and 2026 were analysed using the PRISMA guidelines to assess progress, challenges, and promising opportunities in woody invasive species mapping. Results showed that Sentinel‑2 dominates the reviewed papers, owing to its red‑edge bands and frequent revisit periods. Imagery from different times of the year significantly improved species discrimination by capturing seasonal variability. Machine learning algorithms such as support vector machines (SVM) and random forest (RF) consistently achieved high classification accuracies ranging between 80% and 90%. However, persistent challenges remain, including spectral confusion between invasive and native vegetation, uneven field data representation across study areas, and challenges in mapping resolution and validation data availability. While multisource data integration and cloud‑based platforms such as Google Earth Engine (GEE) enhance scalability and computational efficiency, inconsistency in methodological approaches limits transferability across regions. This review highlights the need for improved integration of sensors and algorithms and standardised validation frameworks to support accurate, scalable, and cost‑effective mapping of woody invasive species.