Abstract:Recently, differentiable modeling techniques have emerged as a promising approach to bidirectionally integrating neural networks and hydrologic models, achieving performance levels close to deep learning models while preserving the ability to output physical states and fluxes. However, there remains a lack of systematic exploration into the performance and physical interpretability of hybrid models that use neural networks to replace the runoff generation and routing processes in regionalized modeling. This research developed 12 regionalized hybrid models based on a differentiable parameter learning (DPL) framework, utilizing the Hydrologiska Byråns Vattenbalansavdelning (HBV) model as the foundational backbone. These hybrid models incorporate neural networks to replace the various physical processes within the runoff generation and routing modules. The publicly available CAMELS dataset is employed to evaluate the performance and interpretability of these hybrid models. The results show that while the median Nash-Sutcliffe efficiency (NSE) and Kling-Gupta efficiency (KGE) coefficients for all hybrid models are lower than those of the purely data-driven regionalized long short-term memory neural network (LSTM) model (median NSE: 0.742, median KGE: 0.762), the best-performing hybrid model (median NSE: 0.731, median KGE: 0.761) approaches the LSTM model and has better physical interpretability. Embedding neural networks does not inherently guarantee improved performance and may, in some cases, even result in reduced performance. The degree of performance enhancement is not significantly correlated with the number of embedded neural networks. Compared to replacing the runoff generation process, substituting the routing process with neural networks yields more substantial performance improvements and enables the learning of different routing patterns based on the catchment's static attributes. This study underscores the importance of reasonably balancing the location, complexity, and quantity of embedded neural networks to achieve a trade-off between model performance and interpretability in hybrid modeling. These insights contribute to advancing regionalized hybrid modeling development.

A Surrogate Model for the Variable Infiltration Capacity Model Using Deep Learning Artificial Neural Network

A two-stage surrogate model based on ANN and AdaBoost for multi-objective parameter optimization of the Variable Infiltration Capacity model

A Stepwise Surrogate Model for Parameter Calibration of the Variable Infiltration Capacity Model: the Case of the Upper Brahmaputra, Tibet Plateau

Seasonal streamflow forecasting by surrogate modeling in the Yarlung Zangbo River Basin, China

A Hybrid Forecasting Model to Simulate the Runoff of the Upper Heihe River

Comprehensive Analysis for Long-Term Hydrological Simulation by Deep Learning Techniques and Remote Sensing

Runoff simulation driven by multi-source satellite data based on hydrological mechanism algorithm and deep learning network

Integrating hybrid runoff generation mechanism into variable infiltration capacity model to facilitate hydrological simulations

Machine Learning Improvement of Streamflow Simulation by Utilizing Remote Sensing Data and Potential Application in Guiding Reservoir Operation

Research on Runoff Simulations Using Deep-Learning Methods

Coupling SWAT and LSTM for Improving Daily Streamflow Simulation in a Humid and Semi-humid River Basin

Comprehensive comparison of LSTM and VIC model in river ecohydrological regimes alteration attribution: A case study in Laohahe basin, China

Improved monthly runoff time series prediction using the CABES-LSTM mixture model based on CEEMDAN-VMD decomposition

Application, interpretability and prediction of machine learning method combined with LSTM and LightGBM-a case study for runoff simulation in an arid area

Short-term streamflow time series prediction model by machine learning tool based on data preprocessing technique and swarm intelligence algorithm

Modeling and simulating of reservoir operation using the artificial neural network, support vector regression, deep learning algorithm

Superior performance of hybrid model in ungauged basins for real-time hourly water level forecasting – A case study on the Lancang-Mekong mainstream

Machine learning surrogates for efficient hydrologic modeling: Insights from stochastic simulations of managed aquifer recharge

Coupling Deep Learning and Physically Based Hydrological Models for Monthly Streamflow Predictions

Development of a surrogate method of groundwater modeling using gated recurrent unit to improve the efficiency of parameter auto-calibration and global sensitivity analysis

Exploring the performance and interpretability of hybrid hydrologic model coupling physical mechanisms and deep learning