A Study on the Coastal Water Environment Health Assessment of the Yellow River Estuary Based on Remote Sensing Inversion Data

The assessment of aquatic environmental health plays a vital role in the sustainable protection and management of coastal ecosystems, particularly in the Yellow River Estuary–one of China’s most representative estuarine systems. To address the limitations of existing health assessment studies, which are often constrained by point-based observations lacking spatial continuity and comprehensiveness, this study integrates multiple remotely sensed surface data to perform a comprehensive health assessment of the nearshore waters of the Yellow River Estuary. An evaluation index system was first developed based on the National Seawater Quality Standards and the specific water quality characteristics of the region. Subsequently, long-term retrievals of key water quality parameters were conducted using Sentinel-2 imagery and in situ measurements from 2016 to 2023, employing the QAA-RF (Quasi-Analytical Algorithm based on Random Forest) algorithm. The Analytic Hierarchy Process (AHP) was used to determine the relative weights of each water quality factor. Through weighted integration, spatially continuous water environmental health assessment datasets were generated, enabling seasonal and annual evaluations and spatiotemporal analyses over the study period. The results indicate that the aquatic environmental health of the nearshore waters exhibits a clear spatial gradient, with poorer water quality in areas closer to the estuary and gradual improvement farther offshore. Seasonal variations are also evident, with poorer water quality observed in spring and winter–reflected by higher proportions of Inferior to Category IV and Category IV water quality (4.07% and 4.65% in spring; 1.12% and 3.71% in winter)–and better conditions in summer and autumn (0.51% and 1.42% in summer; 0.81% and 2.38% in autumn). On an annual scale, the overall aquatic environmental health of the Yellow River Estuary’s nearshore waters remains relatively stable. This study provides a novel, spatially explicit framework for evaluating coastal water environmental health using remote sensing and machine learning approaches. By overcoming the limitations of traditional point-based assessments, it offers valuable insights and a scalable methodology for the continuous monitoring and sustainable management of estuarine and coastal ecosystems.