高光谱遥感水质监测技术研究进展OA

Hyperspectral remote sensing offers contiguous narrow bands and high spectral resolving power,enabling quantitative retrieval of water-quality parameters and,in turn,wide-area,rapid,and dynamic assessment of aquatic environments.This review consolidates recent methodological advances across three interlinked domains.We first reconsider data acquisition across spaceborne,airborne,UAV-borne,and shipborne imaging spectrometers,highlighting sensor traits and their suitability for contrasting optical water types.Then we outline processing workflows that preserve spectral fidelity for modeling,including radiometric and atmospheric correction,mitigation of sun-glint and adjacency effects,and geometric normalization.Finally,we examine retrieval strategies spanning physics-based and semi-analytical formulations,data-driven methods,and hybrids that embed optical priors within learning frameworks.Representative applications are discussed with a focus on aligning tasks with sensors and acknowledging operational constraints.Despite clear momentum,several cross-cutting issues still constrain scale-up:engineering readiness and cross-sensor calibration;domain shift and limited in-situ observations in real waters;generalization and uncertainty quantification;and the continuing need for interpretable models and community standards.Looking ahead,priorities include more robust correction pipelines,physics-informed and transfer-learning designs,fusion with multispectral imagery and synthetic aperture radar,edge or near-real-time processing,and open benchmarks alongside policy support—steps that can move HRS toward routine,large-scale operation in water-quality monitoring and evaluation.