Technological approach to invasive species: the role of remote sensing and predictive modelling

Remote sensing, predictive modelling and machine learning are increasingly central to detecting and monitoring invasive plant species across broad and heterogeneous landscapes. This review synthesises 25 years of remote sensing-based invasive plant species research (2000-2025) to determine how sensor selection, analytical methods and reporting practices have evolved. Records from Web of Science and Scopus were screened using defined eligibility criteria and coded by platform, sensor, species, region, method, impact category, accuracy reporting and linkage with the United Nations Sustainable Development Goals. Bibliometric keyword analysis was used to examine thematic structure and emerging research directions. The evidence shows rapid publication growth after 2015, continued dominance of satellite data, increasing use of uncrewed aerial systems, light detection and ranging, synthetic aperture radar and machine learning classifiers and persistent geographic imbalance. Of 166 retained studies, 149 provided extractable impact information and 76 reported overall accuracy values. The literature indicates a shift from descriptive mapping toward integrated monitoring, risk assessment and decision support frameworks for biodiversity conservation, restoration planning and invasive plant management.