CHARACTERIZATION OF KARST SYSTEMS BY JOINT INVERSION OF SPRING FLOW AND TRANSPORT SIGNATURES FROM A SPATIOTEMPORALLY SMALL-SCALE CONTROLLED TEST
We focus on the simultaneous simulation of flow, heat, and solute signatures of Freiheit karst spring in Minnesota, USA, in response to a spatiotemporally small-scale hydraulic and transport test. Adopting the multi-model concept to address conceptual uncertainty, seven distinctive model variants of karst conduit configuration and zonation were conceptualized. Spring flow and transport signatures were simulated via joint inversion by PEST, considering 1D advective(convective)-dispersive turbulent flow processes along with heat conduction in conduit rock surroundings and the assumption of conservative chloride transport. Calibration of each model variant was performed by minimizing the objective function of weighted least-squares of the residuals. Subsequently, we compare calibrated models in terms of different measures, such as calibration performance and model selection criteria, as well as parameter uncertainties, sensitivities, identifiabilities, and uncertainty variance reduction.
The overall modeling results reveal successful application of a discrete-continuum approach for joint inversion of measured flow and transport signatures from a real-world small-scale controlled test. This result reapproves the reliability of the modeling approaches for flow and transport simulation of karst systems, even at a small scale. Inspecting model outcomes, some insightful results regarding the conduit volume sizes and significance of immobile conduit-associated-drainable storages in karst hydrodynamic modeling were further noted. Moreover, modeling results give an outlook for required CFPv2 improvements, such as concentrated recharge delineation along vertical shafts, linking phreatic conduits up with sinkholes and considering different conduit cross-section shapes.