Yongchao Ma

Turbidity is a crucial and reliable indicator that is extensively utilized in water quality monitoring through remote sensing technology. The development of accurate and applicable models for turbidity estimation is essential. While many existing studies rely on uniform models based on statistical regression or traditional machine learning techniques, the application of deep learning models for turbidity estimation remains limited. This study proposed deep learning models for turbidity estimation based on optical classification of inland waters using Sentinel-2 data. Specifically, the fuzzy c-means (FCM) clustering method was employed to classify optical water types (OWTs) based on their spectral reflectance characteristics. A weighted sum of the turbidity prediction results was generated by the OWT-based convolutional neural network-random forest (CNN-RF) model, with weights derived from the FCM membership degrees. Turbidity for four typical waters was mapped by the proposed method using Sentinel-2 images. The FCM method efficiently classified waters into three OWTs. The OWT-based weighted CNN-RF model demonstrated strong robustness and generalization performance, achieving a high prediction accuracy (R² = 0.900, RMSE = 11.698 NTU). The turbidity maps preserved the spatial continuity of the turbidity distribution and accurately reflected water quality conditions. These findings facilitate the application of deep learning models based on optical classification in turbidity estimation and enhance the capabilities of remote sensing for water quality monitoring.

Turbidity is a crucial indicator for evaluating water quality. This study obtained the long-term spatial distribution of water turbidity across Northeast China from 1985 to 2023. A combination of the geographically and temporally weighted regression (GTWR) model, the Lindeman, Merenda, and Gold (LMG) method, and statistical data analysis methods were employed to quantify the spatiotemporal impacts of driving factors on turbidity changes. The stepwise regression model was able to credibly estimate turbidity, achieving a low RMSE of 18.432 Nephelometric Turbidity Units (NTU). Temporal variations in turbidity showed that 69.90% of lakes exhibited a decreasing trend. Spatial variations revealed that lakes with significantly increased turbidity were predominantly concentrated in the Songnen and Sanjiang Plains, whereas lakes with lower turbidity were situated in the Eastern Mountains regions and Liaohe Plain. Temporal changes were closely associated with socioeconomic development and anthropogenic interventions implemented by governments on the aquatic environment. Vegetation coverage, precipitation, and elevation demonstrated significant contributions (exceeding 16.39%) to turbidity variations in the Lesser Khingan and Eastern Mountains regions, where natural factors played a more dominant role. In contrast, cropland area, wind speed, and impervious surface area showed higher contribution rates of above 14.00% in the Songnen, Sanjiang, and Liaohe Plains, where anthropogenic factors were dominant. These findings provide valuable insights for informed decision-making in water environmental management in Northeast China and facilitate the aquatic ecosystem sustainability under human activities and climate change.

Wetland mapping plays a crucial role in monitoring wetland ecosystems, water resource management, and habitat suitability assessment. Wetland classification remains significantly challenging due to the diverse types, intricate spatial patterns, and highly dynamic nature. This study proposed a dynamic hybrid method that integrated feature selection and object-oriented ensemble model construction to improve wetland mapping using Sentinel-1 and Sentinel-2 data. The proposed feature selection approach integrates the ReliefF and recursive feature elimination (RFE) algorithms with a feature evaluation criterion based on Shapley additive explanations (SHAP) values, aiming to optimize the feature set composed of various variables. During the construction of ensemble models (i.e., RF, XGBoost, and LightGBM) with features selected by RFE, hyperparameter tuning is subsequently conducted using Bayesian optimization (BO), ensuring that the selected optimal features and hyperparameters significantly enhance the accuracy and performance of the classifiers. The accuracy assessment demonstrates that the BO-LightGBM model with ReliefF-RFE-SHAP-selected features achieves superior performance to the RF and XGBoost models, achieving the highest overall accuracy of 89.4% and a kappa coefficient of 0.875. The object-oriented classification maps accurately depict the spatial distribution patterns of different wetland types. Furthermore, SHAP values offer global and local interpretations of the model to better understand the contribution of various features to wetland classification. The proposed dynamic hybrid method offers an effective tool for wetland mapping and contributes to wetland environmental monitoring and management.