Abstract:We constructed a small covered karst system with corresponding numerical models and improved the traditional exponential function to decompose spring recession hydrographs. The results indicate that the slow flow dominated by the porous media plays a critical role in the whole recession process. There is a shift in the contribution of the matrix to karst discharge systems with increasing hydraulic conductivity, and the changing process of the water storage rate of fractured rocks with spring hydrographs contains three stages. The quantitative analysis of karst spring hydrographs aids in characterizing the hydrodynamic properties of aquifer systems. However, the relationships between spring hydrographs and hydrologic functions of different media lack in‐depth research. This study detected the rainfall‐discharge processes of covered karst systems using two combined discrete‐continuum models (a coupled continuum‐pipe model and a hybrid model) developed based on laboratory experiments. An improved exponential function was used to decompose spring recession hydrographs with a great performance, and the results indicate that the slow drainage controlled by the porous medium dominates the whole recession process. There is a shift in the contribution of the matrix to karst discharge systems with the increase of its hydraulic conductivity. Although spring hydrographs have a great dependency on the hydraulic process of fracture networks, the variation in permeability of the matrix significantly changes the storage capacity of fractured rocks and spring discharge behaviours. The changing process of water storage rate of the matrix and fractures with spring hydrographs reveals three stages: a period of steep rise and fall, a period with slight fluctuations, and a gentle descent stage. The parameter sensitivity analysis in both models shows that the covering porous layer has a strong effect on the patterns of karst spring hydrographs, and the specific yield is more sensitive than hydraulic conductivity. While these simulations were conducted at the laboratory scale, the findings provide insights into practical applications of spring protection and water resources management.

Hybrid modeling of karstic springs: Error correction of conceptual reservoir models with machine learning

A Hybrid Framework Based on Lstm for Predicting Karst Spring Discharge Using Historical Data

Improving the simulations of the hydrological model in the karst catchment by integrating the conceptual model with machine learning models

A data-driven approach for modelling Karst spring discharge using transfer function noise models

Karst spring discharge modeling based on deep learning using spatially distributed input data

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

A hybrid self-adaptive DWT-WaveNet-LSTM deep learning architecture for karst spring forecasting

Hybrid Data-Driven Models for Hydrological Simulation and Projection on the Catchment Scale

Comparison of artificial neural networks and reservoir models for simulating karst spring discharge on five test sites in the Alpine and Mediterranean regions

Modeling the Hydrological Behavior of a Karst Spring Using a Nonlinear Reservoir-Pipe Model

Enhancing physically-based hydrological modeling with an ensemble of machine-learning reservoir operation modules under heavy human regulation using easily accessible data

A Hybrid, Non‐Stationary Stochastic Watershed Model (SWM) for Uncertain Hydrologic Simulations Under Climate Change

Taking into Account both Explicit Conduits and the Unsaturated Zone in Karst Reservoir Hybrid Models: Impact on the Outlet Hydrograph

Hourly rainfall-runoff modelling by combining the conceptual model with machine learning models in mostly karst Ljubljanica River catchment in Slovenia

Enhancing Streamflow Simulation Using Hybridized Machine Learning Models in a Semi-Arid Basin of the Chinese Loess Plateau

Integrating field work and large-scale modeling to improve assessment of karst water resources

A novel hybrid of support vector regression and metaheuristic algorithms for groundwater spring potential mapping

Advancing Streamflow Prediction in Data-Scarce Regions Through Vegetation-Constrained Distributed Hybrid Ecohydrological Models

Modelling Spring Discharge and Solute Transport in Conduits by Coupling CFPv2 to an Epikarst Reservoir for a Karst Aquifer

Multi-model approach in a variable spatial framework for streamflow simulation

Identification of karst spring hydrographs using laboratory and numerical simulations considering combined discrete‐continuum approaches