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

Abstract

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.