Accurate river water level estimation is essential for flood prevention and water resource management, but conventional methods struggle in complex plain river-network regions where water levels are influenced by precipitation, sluice regulation, human activities, and agricultural irrigation, creating strong nonlinearity and non-stationarity. Hydrological stations provide direct measurements but suffer from sparse spatial coverage, high operational costs, and limited real-time capability. Satellite remote sensing offers an alternative, yet individual sensors have drawbacks: conventional altimeters have sparse cross-track spacing, Synthetic Aperture Radar (SAR) has limited temporal resolution, and the Cyclone Global Navigation Satellite System (CYGNSS) suffers from coarse spatial resolution and observational noise. These limitations underscore the need for advanced multi-source fusion techniques.
In a study published in Satellite Navigation on October 12, 2026, researchers from Nanjing University of Posts and Telecommunications, Henan Polytechnic University, Anhui University, and the Nanjing Institute of Geography and Limnology of the Chinese Academy of Sciences introduced the Physics-Guided Noise-Enhanced Adaptive Fusion Dual-Stream Network (PGNAF-DSNet) for water level estimation over the Jiangsu plain in China. The model integrates Sentinel-1 SAR and CYGNSS observations through physics-guided feature construction, dual-stream temporal encoding, and adaptive feature fusion. On a strictly chronological test set, PGNAF-DSNet achieved a root mean square error (SRMSE) of 0.2388 meters and a correlation coefficient (R) of 0.9889, reducing prediction errors by approximately 50% in regions significantly affected by human activities such as sluice regulation and agricultural irrigation.
The study's key innovations include the development of physics-guided fused features: the Polarization Fusion Surface Water Index (PFSWI), which multiplies Sentinel-1 backscatter differences by CYGNSS reflectivity, and the river width coefficient of variation (Width_cv), embedding physical priors into the model. A selective Gaussian noise enhancement strategy, optimized using the Tree-structured Parzen Estimator (TPE) algorithm with σ = 0.04157, is applied only to channels susceptible to observational errors, acting as a regularizer to improve robustness. The dual-stream architecture employs parallel Transformer and long short-term memory (LSTM) encoders to capture global and local temporal patterns, respectively, before fusion through an Adaptive Feature Fusion (AFF) module that dynamically weights the two branches at the channel level. Ablation experiments confirmed each component's contribution, and SHAP analysis indicated that Width_cv and PFSWI both exhibit high feature importance.
The authors said the study's core contribution lies in moving beyond simple data-level fusion. Previous approaches mainly used CYGNSS to fill temporal gaps in SAR observations without fully exploiting cross-modal feature interactions. By constructing physics-guided fused features and adopting a dual-stream architecture, the model captures both global trends and local dynamics of water level changes in complex river networks. The selective noise enhancement strategy further improves robustness against observational uncertainty and human-induced disturbances.
PGNAF-DSNet offers a practical framework for water level monitoring in plain river-network regions, with potential applications in flood early warning, water resource management, and river-network regulation. The model maintained stable performance at T+6 forecasts (R ≥ 0.95), demonstrating promise for operational forecasting. The framework could also be extended to incorporate additional satellite observations, such as Surface Water and Ocean Topography (SWOT) water surface elevations, and adapted to other river systems through transfer learning. The full study is available at 10.1186/s43020-026-00219-7.
