Delineation of discrete conduit networks in karst aquifers via combined analysis of tracer tests and geophysical data

Assessment of the karst network geometry based on field data is an important challenge in the accurate modeling of karst aquifers. In this study, we propose an integrated approach for the identification of effective three-dimensional (3D) discrete karst conduit networks conditioned on tracer tests and geophysical data. The procedure is threefold: i) tracer breakthrough curves (BTCs) are processed via a regularized inversion procedure to determine the minimum number of distinct tracer flow paths between injection and monitoring points, ii) available surface-based geophysical data and borehole-logging measurements are aggregated into a 3D proxy model of aquifer hydraulic properties, and iii) single or multiple tracer flow paths are identified through the application of an alternative shortest path (SP) algorithm to the 3D proxy model. The capability of the proposed approach to adequately capture the geometrical structure of actual karst conduit systems mainly depends on the sensitivity of geophysical signals to karst features, whereas the relative completeness of the identified conduit network depends on the number and spatial configuration of tracer tests. The applicability of the proposed approach is illustrated through a case study at the Hydrogeological Experimental Site (HES) in Poitiers, France.

The genesis and evolution of karstic conduit systems in the Chalk

The Upper Cretaceous Chalk Group is renowned as a major aquifer, but the development of secondary porosity due to karstic conduits is poorly understood. Hydrogeological data and evidence from boreholes, sections, and tracer tests indicate that dissolutional conduits occur throughout the Chalk aquifer. Here, we assess the evidence for Chalk karst, and combine it with theoretical models of dissolution and cave formation to produce a conceptual model for the development of karstic conduits. Dissolution due to the mixing of saturated waters of contrasting chemistry along key lithostratigraphical inception horizons form extensive but isolated conduit networks. These form a significant proportion of the secondary porosity and enhance permeability. They prime the aquifer for the development of more integrated conduit networks formed by focussed recharge of unsaturated surface derived water. However, the porous, well-fractured nature of the Chalk means that the time needed to form large integrated cave systems is often longer than timescales of landscape change. Continued landscape evolution and water table lowering halts conduit development before they can enlarge into cave systems except where geological and geomorphological settings are favourable. Groundwater models need to consider the formation of secondary karst permeability as this has a major influence on groundwater flow.

Optimal Sensor Placement in Hydraulic Conduit Networks: A State-Space Approach

Conduit bursts or leakages present an ongoing problem for hydraulic fluid transport grids, such as oil or water conduit networks. Better monitoring allows for easier identification of burst sites and faster response strategies but heavily relies on sufficient insight in the network’s dynamics, obtained from real-time flow and pressure sensor data. This paper presents a linearized state-space model of hydraulic networks suited for optimal sensor placement. Observability Gramians are used to identify the optimal sensor configuration by maximizing the output energy of network states. This approach does not rely on model simulation of hydraulic burst scenarios or on burst sensitivity matrices, but, instead, it determines optimal sensor placement solely from the model structure, taking into account the pressure dynamics and hydraulics of the network. For a good understanding of the method, it is illustrated by two small water distribution networks. The results show that the best sensor locations for these networks can be accurately determined and explained. A third example is added to demonstrate our method to a more realistic case.

A Karst Probability Map for the Western Mountain Aquifer (Israel & West Bank) using a stochastic modeling approach

<p>Investigating the structure of conduit networks in karst aquifers is a common challenge when working in these complex hydrogeological environments. The network geometry plays an important role in karst flow dynamics, but highly karstified areas are often difficult to characterize by field measurements. Here, we present a methodology that generates karst conduit network geometries reasonably quick without solving complex flow or dissolution equations, and that uses only little input information. The stochastic approach also enables the investigation of the uncertainty of generated networks in the form of a karst probability map.</p><p>The “Stochastic Karst Simulator” (SKS) is a stochastic modeling approach developed by Borghi et al. (2012) to generate a 3D karst conduit network by computing a minimum effort path between the given inlet and outlet points. This study uses such a modeling approach to characterize the karst network geometry of the Western Mountain Aquifer (WMA), a highly karstified and exploited carbonate aquifer located in Israel and the West Bank. The SKS simulations are based on a conceptual model of the aquifer’s karst genesis, to identify the position of karst springs and recharge zones over past geological ages.</p><p>Three different phases of karst formation are identified for the WMA. Phase 1: a paleo-discharge zone exists, located close to the present-day coastline of Israel, phase 2: a period of extreme low sea levels during the Messinian salinity crisis, when paleo-canyons were reactivated along this coastline, and phase 3: the modern-day outlets of the aquifer. The iterative approach of the SKS algorithm accounts for these different phases and creates new conduit pathways by building on ones formed in earlier phases. The algorithm also uses the hydrological model of the study site as soft information, providing knowledge about the internal heterogeneities of the karst formations (e.g. statistical properties of fractures). The resulting karst probability map is compared to the location of the most productive pumping wells in the region, assuming a high yield in groundwater abstraction indicating major karst conduits near the pumped well. </p><p>We demonstrate the method by showing a reconstruction of the karst conduit networks at the WMA model area, an otherwise not available spatial information. The simulations show that the changes in karst spring and recharge locations have a great impact on the geometry and connectivity of the conduit network. Overarching trends in the conduit orientation of the resulting probability map are in keeping with the proposed karst genesis model, resulting in the evolution of a hierarchical network. High karstification is indicated around modern-day springs, also in agreement with the location of numerous pumping wells in that region.</p><p>The SKS algorithm is a useful tool to test different hypotheses of karst genesis and to understand the evolution of karst network geometries. The methodology is numerically efficient, and its inputs can be easily adjusted. Soft information on karst development allows for the generation of a sound hydraulic parameter field, which can be implemented in hydrological models to better understand and manage these aquifer systems.</p>

The physical meaning of transport properties evaluated from BTC in karst conduit networks using transfer functions

<p>The Transfer Function (TF) approach, applying the Laplace transform, is known to be effective in interpreting tracer BreakThrough Curves (BTCs) in karst systems. Although this approach has several advantages over the classical Advection Dispersion Equation (ADE), the parameters of the TF are difficult to interpret directly in terms of transport properties, e.g., flow velocity and dispersion coefficient.</p><p>We present two approaches to relate the TF parameters to those of the ADE parameters. The first uses a consistency analysis, the other uses an asymptotic analysis in the Laplace space. The TF parameters can be transformed into equivalent ADE parameter groups that have an apparent physical meaning about the transport process. We further provide guidelines for choosing the suitable fitting models for artificial tracer tests and offer some suggestions for utilizing the TF approach in BTCs interpretation.</p>

A model ensemble generator to explore structural uncertainty in karst systems with unmapped conduits

Karst aquifers are characterized by high-conductivity conduits embedded in a low-conductivity fractured matrix, resulting in extreme heterogeneity and variable groundwater flow behavior. The conduit network controls groundwater flow, but is often unmapped, making it difficult to apply numerical models to predict system behavior. This paper presents a multi-model ensemble method to represent structural and conceptual uncertainty inherent in simulation of systems with limited spatial information, and to guide data collection. The study tests the new method by applying it to a well-mapped, geologically complex long-term study site: the Gottesacker alpine karst system (Austria/Germany). The ensemble generation process, linking existing tools, consists of three steps: creating 3D geologic models using GemPy (a Python package), generating multiple conduit networks constrained by the geology using the Stochastic Karst Simulator (a MATLAB script), and, finally, running multiple flow simulations through each network using the Storm Water Management Model (C-based software) to reject nonbehavioral models based on the fit of the simulated spring discharge to the observed discharge. This approach captures a diversity of plausible system configurations and behaviors using minimal initial data. The ensemble can then be used to explore the importance of hydraulic flow parameters, and to guide additional data collection. For the ensemble generated in this study, the network structure was more determinant of flow behavior than the hydraulic parameters, but multiple different structures yielded similar fits to the observed flow behavior. This suggests that while modeling multiple network structures is important, additional types of data are needed to discriminate between networks.

Groundwater Circulation in Fractured and Karstic Aquifers of the Umbria-Marche Apennine

The Umbria-Marche Apennine has a large number of springs that drain water stored in carbonate formations. Spring groundwater constitutes a crucial freshwater resource for many countries, regions, and cities around the world. This study aimed to understand the hydrological mechanisms behind groundwater circulation and their relationship to the structural and stratigraphic settings of specific aquifers. Recession analysis and time series analysis were applied to the daily discharge of six springs monitored over eight years. Both analyses indicated the presence of two types of aquifers: aquifer with unimodal behavior and aquifer with bimodal behavior. The first are characterized by two hydrodynamic sub-regimes, in which fracture networks control the baseflow and conduit networks control the quickflow. In contrast, other springs present only one hydrodynamic sub-regime related to fracture network drainage. Time series analysis confirms the results of recession analysis, showing a large memory effect and a large response time, implying the dominance of the baseflow sub-regime. These results indicate that the Maiolica Formation is characterized by a high degree of fracturation and slight karstification, which control infiltration and percolation, whereas the Calcare Massiccio Formation regulates groundwater circulation in the deeper zones of the aquifer, characterized by a high degree of karstification through moderately developed conduit networks.

Numerical modeling and sensitivity analysis of seawater intrusion in a dual-permeability coastal karst aquifer with conduit networks

Long-distance seawater intrusion has been widely observed through the subsurface conduit system in coastal karst aquifers as a source of groundwater contaminant. In this study, seawater intrusion in a dual-permeability karst aquifer with conduit networks is studied by the two-dimensional density-dependent flow and transport SEAWAT model. Local and global sensitivity analyses are used to evaluate the impacts of boundary conditions and hydrological characteristics on modeling seawater intrusion in a karst aquifer, including hydraulic conductivity, effective porosity, specific storage, and dispersivity of the conduit network and of the porous medium. The local sensitivity analysis evaluates the parameters' sensitivities for modeling seawater intrusion, specifically in the Woodville Karst Plain (WKP). A more comprehensive interpretation of parameter sensitivities, including the nonlinear relationship between simulations and parameters, and/or parameter interactions, is addressed in the global sensitivity analysis. The conduit parameters and boundary conditions are important to the simulations in the porous medium because of the dynamical exchanges between the two systems. The sensitivity study indicates that salinity and head simulations in the karst features, such as the conduit system and submarine springs, are critical for understanding seawater intrusion in a coastal karst aquifer. The evaluation of hydraulic conductivity sensitivity in the continuum SEAWAT model may be biased since the conduit flow velocity is not accurately calculated by Darcy's equation as a function of head difference and hydraulic conductivity. In addition, dispersivity is no longer an important parameter in an advection-dominated karst aquifer with a conduit system, compared to the sensitivity results in a porous medium aquifer. In the end, the extents of seawater intrusion are quantitatively evaluated and measured under different scenarios with the variabilities of important parameters identified from sensitivity results, including salinity at the submarine spring with rainfall recharge, sea level rise, and a longer simulation time under an extended low rainfall period.