Karst and fractured aquifers are characterized by a duality of infiltration and flow. Flow and transport modelling in these systems is highly challenging and requires a thorough characterization of the subsurface and extensive collection of data. This talk provides a short review on the evolution of modeling in karst aquifers, to later zoom to one experimental poorly studied site in Mount Lebanon: the Nahr El Kalb surface water/ aquifer system composed of limestone and dolostones of Jurassic to Cenomanian age. A high-resolution monitoring is taking place since 2014 to characterize the subsurface, understand spring responses, conceptualize flow and transport in complex systems, and simulate flow in variably heterogeneous systems of Mount-Lebanon. Different methods have been applied for the characterization of flow and transport in different types of karst systems with variable heterogeneities, hydrodynamic conditions, and climatic input: 1) snow-governed versus rain governed springs, 2) fissured karst aquifers, 3) highly complex heterogenous karst with little knowledge of the subsurface, and 4) highly complex karst with a cave access. Such utilized methods include times series analysis, tracer experiments, micropollutants sampling campaigns, stable isotope studies in addition to the use of stochastic geomodelling to infer from surface characteristics information about preferential flow in the subsurface. Selected distributed and lumped flow models will be presented such as 1) distributed integrated hydrological model, 2) semi distributed linear reservoir model, 3) 2-D dual continuum, 4) discrete fractured network models, and 5) convolution neural network models (CNN), that have been tailored to simulate flow in spring catchment areas to account for the degree of karstification and varying hydrodynamic responses. In this talk we illustrate a modeling approach can be scaled up to the region for a better sustainable groundwater management in poorly studied aquifer systems.

Karst and fractured aquifers are characterized by a duality of infiltration and flow. Flow and transport modelling in these systems is highly challenging and requires a thorough characterization of the subsurface and extensive collection of data.  This talk provides a short review on the evolution of modeling in karst aquifers, to later zoom to one experimental poorly studied site in Mount Lebanon: the Nahr El Kalb surface water/ aquifer system composed of limestone and dolostones of Jurassic to Cenomanian age. A high-resolution monitoring is taking place since 2014 to characterize the subsurface, understand spring responses, conceptualize flow and transport in complex systems, and simulate flow in variably heterogeneous systems of Mount-Lebanon. Different methods have been applied for the characterization of flow and transport in different types of karst systems with variable heterogeneities, hydrodynamic conditions, and climatic input: 1) snow-governed versus rain governed springs, 2) fissured karst aquifers, 3) highly complex heterogenous karst with little knowledge of the subsurface, and  4) highly complex karst with a cave access. Such utilized methods include times series analysis, tracer experiments, micropollutants sampling campaigns, stable isotope studies in addition to the use of stochastic geomodelling to infer from surface characteristics information about preferential flow in the subsurface. Selected distributed and lumped flow models will be presented such as 1) distributed integrated hydrological model, 2) semi distributed linear reservoir model, 3) 2-D dual continuum, 4) discrete fractured network models, and 5) convolution neural network models (CNN), that have been tailored to simulate flow in spring catchment areas to account for the degree of karstification and varying hydrodynamic responses. In this talk we illustrate a modeling approach can be scaled up to the region for a better sustainable groundwater management in poorly studied aquifer systems.