Excessive discharges of nitrogen and phosphorus nutrients lead to eutrophication in coastal waters. Optical remote sensing retrieval of the key eutrophication indicators, namely dissolved inorganic nitrogen concentration (DIN), soluble reactive phosphate concentration (SRP), and chemical oxygen demand (COD), remains challenging due to lack of distinct spectral features. Although machine learning (ML) has shown the potential, the retrieval accuracy is limited, and the interpretability is insufficient in terms of the black-box characteristics. To address these limitations, based on robust and explainable ML algorithms, we constructed models for retrieving DIN, SRP, and COD over coastal waters of Northern South China Sea (NSCS), which is experiencing prominent eutrophication. Retrieval models based on classification and regression trees (CART) ML algorithms were developed using 4038 groups of observations and quasi-synchronous satellite images. A comparison of CART algorithms, including Random Forest, Gradient Boosting Decision Tree, and eXtreme Gradient Boosting (XGBoost), indicated the highest retrieval accuracy of XGBoost for DIN (R = 0.88, MRE = 24.39 %), SRP (R = 0.92, MRE = 33.27 %), and COD (R = 0.75, MRE = 18.58 %) for validation dataset. On the basis of spectral remote sensing reflectance, further inputs of ocean physio-chemical properties, spatio-temporal information, and inherent optical properties may reduce retrieval errors by 30.16 %, 19.85 %, and 3.95 %, respectively, and their combined use reduced errors by 54.71 %. Besides, explainable ML analysis characterized the contribution of input features and enhanced the transparency of ML black-box models. Based on the proposed models, 27,278 satellite images and spatio-temporal reconstruction method, 1-km resolution gap-free daily DIN, SRP, and COD products were constructed from 2002 to 2022 for the coastal waters of NSCS. Under the influence of urbanization and river discharge, nitrogen and phosphorus concentrations in this area were found to have increased by 6.09 % and 11.04 %, respectively, over the past 21 years, with the fastest rise in the Pearl River Estuary, where the eutrophic water area had shown an increase rate of approximately 112.66 km/yr. The proposed robust and explainable ML retrieval models may support ocean environment management and water quality monitoring by providing key eutrophication indicators products over coastal waters.

中文翻译：

---

基于可解释机器学习的沿海水域关键富营养化指标的鲁棒遥感检索

氮磷营养物质过量排放导致近岸海域富营养化。由于缺乏明显的光谱特征，关键富营养化指标（即溶解性无机氮浓度（DIN）、可溶性活性磷酸盐浓度（SRP）和化学需氧量（COD））的光学遥感反演仍然具有挑战性。尽管机器学习（ML）已经展现出潜力，但检索精度有限，并且在黑盒特性方面可解释性不足。为了解决这些局限性，基于稳健且可解释的机器学习算法，我们构建了反演南海北部 (NSCS) 沿海水域 DIN、SRP 和 COD 的模型，该海域的富营养化状况十分严重。使用 4038 组观测数据和准同步卫星图像开发了基于分类和回归树 (CART) ML 算法的检索模型。对随机森林、梯度提升决策树和极限梯度提升 (XGBoost) 等 CART 算法的比较表明，XGBoost 对于 DIN (R = 0.88，MRE = 24.39 %)、SRP (R = 0.92，MRE) 的检索精度最高= 33.27 %)，验证数据集的 COD (R = 0.75，MRE = 18.58 %)。在光谱遥感反射率的基础上，进一步输入海洋理化特性、时空信息和固有光学特性，可分别减少反演误差30.16%、19.85%和3.95%，综合使用可减少误差增加了 54.71%。此外，可解释的机器学习分析表征了输入特征的贡献，并增强了机器学习黑盒模型的透明度。基于所提出的模型，2002年至2022年期间，构建了27278幅卫星影像和时空重建方法，以及1公里分辨率的无间隙日DIN、SRP和COD产品。受城市化和河流排放的影响，近21年来，该地区氮、磷浓度分别增加了6.09%和11.04%，其中上升最快的是珠江口，那里的水体富营养化。面积显示出约112.66平方公里/年的增长率。所提出的稳健且可解释的机器学习检索模型可以通过提供沿海水域的关键富营养化指标产品来支持海洋环境管理和水质监测。