Towards Automated Monitoring of Woody Plant Encroachment on Floodplains: A Remote Sensing and Hydrology Framework

Woody plant encroachment, the expansion of shrubs and trees into historically herbaceous ecosystems, is accelerating globally across rangelands, savannas, and floodplains, with consequences for biodiversity and hydrology. We present a vision for an automated monitoring framework that fuses multi-source remote sensing data with GeoAI on open platforms (e.g., Open Data Cube/Digital Earth Platforms) to examine floodplain encroachment caused by altered hydrological regimes. The framework combines optical, synthetic aperture radar, structural and hydrological data, applying spatiotemporal analytics to map vegetation change and link it to hydrological drivers. A key feature is the linkage between encroachment dynamics and hydrology: woody expansion trajectories are statistically related to indicators of flow regime, flooding, soil moisture, and groundwater. Using lagged and change-point analyses, the framework can detect thresholds and explain observed patterns without complex hydrological models. Decision-support outputs include maps attributing vegetation changes to drivers and early warnings or triggers for management, such as recommendations for environmental watering. The approach is open, reproducible, and scalable, moving beyond mapping where encroachment occurs to explaining why, and enabling earlier interventions.