Search

Estimating karst water resources at a global scale represents a key step towards enhancing groundwater management. Although recharge estimations at global scale exist, the special karst features leading to different recharge processes, in particular the preferential flow path, have not been adequately accounted for. Representing these special recharge processes is crucial for realistically quantifying recharge in karst regions. In this study, we present a new version of a global karst recharge model and the estimation of its model parameters using multiple observations. The updated model includes a new routine to consider eight land cover types for calculating evapotranspiration, interception and infiltration, as well as a new routine to account for snow and glacier melt. To prepare parameter estimation, six karst landscapes are defined based on the properties of karst grids (spatial resolution of 0.25°). For each of these landscapes, model parameters are found separately using a Monte Carlo uncertainty estimation framework and observations from the international soil moisture network (ISMN) and actual evapotranspiration observations from FLUXNET. With the new calibration, the updated model provides more precise estimates of groundwater recharge in athe karst regions of the world. It will therefore serve as a too to improve water management in karst regions and identify areas potentially suffering from water shortage.

The overall aim of the MEWAC-FEMAR project (Middle East Regional Water Research Cooperation Program – Feasibility of Managed Aquifer Recharge), funded by the German Federal Ministry of Education and Research (BMBF), is to support and enhance a safe and sustainable water supply in the Middle East by Managed Aquifer Recharge (MAR). One target region is Syria, which is facing water insecurity regarding water availability, water supply, and agricultural water needs. In addition, the region is severely affected by war which results in extreme destruction of infrastructure and other resources. The project's objective is to provide a model-based framework for integrated water resources management (IWRM) to identify suitable sites for MAR and to predict their impact on the regional water balance and the local situation. The proposed framework incorporates different methods and model techniques with applications ranging from global to local scales. The model workflow is completely based on open-source tools to ensure sustainable use of the project efforts. Modeling on the global scale is mainly done by data-driven methods like GIS Multi-Criteria Decision Analysis (GIS-MCDA) to answer questions like where are suitable and feasible areas for MAR. Regional scale models aim to describe the hydrological cycle by distributed numerical methods (e.g., MODFLOW-OWHM) to predict and manage MAR under consideration of the regional groundwater situation. Local-scale numerical models can provide detailed and site-specific insights into the flow and transport processes of the unsaturated zone and the underlying aquifer. Specific activities in Syria include e.g., evaluation of wastewater reuse for irrigation. Finally, the current process understanding with regard to MAR methods is critically evaluated and expanded.

Where does water go when it rains? Where are floods generated and how? What controls stream water quality during events? These questions are important to many fields from engineering and flood protection to water and ecosystem management and prediction of impacts of global change. The most elusive processes in the process-ensemble underlying these questions is subsurface stormflow (SSF), the fast event response triggered by lateral subsurface flow. SSF is prevalent and a more important process than generally accounted for because a basic understanding based on systematic studies across scales and sites is still lacking. However, only with systematic studies will it be possible to really advance our understanding by discovering general principles of SSF functioning and to provide protocols and best practices for its assessment, both experimentally and with respect to modelling.In many natural landscapes, SSF, i.e. any subsurface flow that occurs in response to a precipitation event, plays a major role in runoff generation: either by contributing directly to streamflow or by producing saturated areas or return flow, which then is the underlying cause of saturation excess overland flow. Therefore, much of what we see as event response in the hydrograph might be the direct or indirect result of SSF. It is likely that the discharge signal of SSF, including the indirectly triggered response in the stream, is larger than we generally assume. While its importance is probably largest in the headwaters, headwaters make up 70% of the stream network and greatly influence the supply and transport of water and solutes downstream. However, SSF is elusive and poorly accounted for as measurements are difficult for several reasons: the inaccessibility of the subsurface, the large spatial variability and heterogeneity, the variable sources and the fact that it is a threshold-driven process that only occurs during certain events. Thus, systematic studies of SSF are lacking, mainly due to difficulties of quantification.We suggest such a systematic study of SSF in different environments, across scales, and using a well-designed and replicated selection of approaches including novel approaches. This will be followed by a systematic evaluation of methods and possible proxies as well as model intercomparison, evaluation and improvement. Thereby, we will focus on 4 challenges: 1) Development of novel experimental methods,2) Spatial patterns of SSF, 3) Thresholds and cascading effects of SSF, 4) Impacts of SSF.Whereas standard single research projects investigate part of this puzzle at a specific location, this Research Unit provides the unique opportunity of fitting a large number of puzzle pieces together. This Research Unit will have a strong emphasis on experimental work in four contrasting catchments from the low to high mountain ranges (Sauerland, Ore Mountains, Black Forest, Alps) that then directly feeds into a collaborative modelling effort, which in turn influences experimental design in an iterative process.

In addition to overland flow, subsurface stormflow (SSF) can play a major role for runoff generation during certain events. Even though SSF is a well-recognised process, there is a lack of systematic studies on SSF, partly because it is very difficult to quantify. In hydrological modelling, this can lead to an unclear distinction of SSF from other processes. If parameters that describe SSF processes or thresholds are implemented in hydrological models, they are often used as fitting parameters and can contribute to overall model uncertainty. So far, there has been no systematic benchmarking of SSF routines in hydrological models.This presentation discusses how we plan to address this research gap. In order to address the inconsistency of SSF representation in hydrological models, different existing lumped hydrological models will be set up for four study sites located in the Alps, Ore Mountains, Black Forest and Sauerland. In a first step, different models will be calibrated using only basic data like discharge observations, climate data and readily available geodata. Differences in SSF simulations will be detected and quantified and the models will be benchmarked regarding the simulation of SSF dynamics and associated uncertainties. In a next step, we will include new experimental data on SSF derived at the four study sites in the calibration of the lumped hydrological models by a multi-objective calibration and evaluation framework. In order to consider SSF observations collected at scales different than the scale of model application, new SSF metrics will be developed. Testing different combinations of these metrics for model calibration it will be possible to state which SSF proxies can lead to the most productive improvement of SSF simulations.Identifying current weaknesses in SSF representation of current models, and providing directions for improving them by including the most beneficial SSF metrics, this project will show potential for the improvement of SSF simulations through SSF data collection. In a final application of the most reliable SSF simulations to all study sites, we will show the impact of extreme wet or extreme dry conditions on SSF occurrence and SSF volumes.

Modelling solute mixing and transport processes is one of the key steps to effectively managing karstic groundwater resources, particularly under the threat of climate change and risk of contamination. For that reason, a considerable body of literature has been devoted to understanding and describing solute mixing and transport processes in karst aquifers. However, due to the strong multiscale heterogeneity (from microscale to aquifer scale), modelling solute mixing and transport processes in karst aquifers remains a challenging task. This presentation critically reviews the current state of knowledge and fundamental challenges in the modelling of solute mixing and transport processes in karst aquifers, thereby collocating and synthesizing the existing body of knowledge in the literature. To provide a holistic and objective picture of the state-of-the-art of the solute mixing and transport modelling, we performed a bibliometric analysis on the relevant literature for karst groundwater studies (over 2800 scientific papers). Further, with a meta-analysis of scientific papers focusing on the quantitative tracer tests, we evaluated the field-based transport parameters that are typically served for the solute mixing and transport models.The review unveils the fundamental modelling hinges underlying a successful modelling practice for the solute mixing and transport processes in karst, thereby discussing to what extent and in what ways we are dealing with these challenges. The major modelling challenges are defined as follows: (i) Model conceptualization based on data collection and system understanding (e.g., How well is the problem of interest defined? To what extent is the domain of interest described?), ii) Model selection considering the choice of a dominant physicochemical process (e.g., How well is the process of interest represented by a set of governing equations over the problem domain?), iii) Time-variability of solute mixing and transport processes (e.g., To what extent do the parameters represent the process of interest under the different time-scales?), iv) Model parametrization considering the parameter non-uniqueness and transferability (e.g., How realistic are the model parameters? To which extent are they transferable over the same aquifer?), v) Uncertainty quantification in model results (e.g., How robust are the model results? How much are we (un)certain about our model?). Finally, we address potential research directions and knowledge gaps by encouraging the community for building a protocol for solute transport modelling in karst aquifers, as well as providing more transparent and reproducible results.

Managed aquifer recharge (MAR) is a powerful approach to counteracting the negative impacts of overexploited groundwater resources and enhancing groundwater availability in water-scarce regions. One common MAR strategy is aquifer storage and recovery (ASR). Theoretically, the ASR techniques can be adopted for karst systems to store and recover water within the system or to increase the subsurface inflow into the alluvial system from the adjacent upland karst system. However, karst systems present a particular challenge for the application and technical implementation of ASR, due to the associated strong anisotropy and heterogeneity, high runoff dynamics, as well as the underlying uncertainties regarding the available data.This research is oriented to provide a regional scale model of a karst-alluvial system in Lebanon, analyzing the impact of MAR scenarios as well as quantifying boundary inflow to adjacent alluvial systems. For this reason, several model conceptualizations (considering multi-model concepts) and ASR application scenarios are adopted and tested. For transferability, real-world case studies and idealized yet generalizable systems are considered and assessed. The model pre- and postprocessing is entirely script-based and uses open-source tools to ensure sustainable useAn anisotropic fast-marching algorithm is used to create spatially distributed karst channel networks through the Python package pyKasso. In addition, a discrete continuum model is developed (e.g., CfPy), where the stochastic conduit networks are implemented as a large ensemble of the plausible and representative karstic system. The regional karst alluvial model results are also used to make general recommendations for MAR site selection in the karstic study site.

Groundwater resources are generally evaluated by quantifying groundwater recharge and its storage into the aquifer using for example conceptual or numerical models. Recharge modelling provides a preliminary estimation of the renewable part of the groundwater resource. Moreover, results of regional groundwater recharge models may be used as input data for models on a smaller scale, i.e. at the catchment scale dedicated to water management for present and future conditions. It is thus necessary to constrain the recharge models for example by comparing their outputs to historical long-term observations of groundwater flows that can be derived from time series of groundwater levels or spring discharge.Karst systems with their high infiltration rate and preferential flow in enlarged conduit networks, react quickly to climatic events and changes. Thus, they can be used as proxies to evaluate the impact of global change on groundwater resources. Karst systems are present in different climatic regions of Europe, which allows comparing long-term trends of groundwater recharge with spring discharge (similar or opposing trends). In our study, two different regional recharge models covering entire Europe – one calculating potential groundwater recharge using simple soil water balance methods and one calculating groundwater recharge over karst areas only using 1-D physical equations for infiltration – are compared to the measured spring discharge from a large European database of more than 100 monitored karst systems. To identify and highlight changes in dominant recharge processes related to changing climatic or physiographic (land cover / land use) conditions, different variables used in the regional recharge models will be correlated to indices describing dynamics of karst spring discharge. The results of our study will help to understand the impact of climatic and groundwater recharge related influences on various geographic locations and give insights on uncertainties in the model structure of the applied regional groundwater recharge models.

High mountains, which exhibit alpine and subalpine characteristics, represent 15% of the earth’s land area and are estimated to contribute about 17% of global runoff. Depending on hydrogeological setting, a significant amount of catchment water can be stored in high mountainous underground as groundwater, which can contribute substantially to streamflow and represent an important water source. However, high-alpine catchments are often characterized by great geological complexity and highly heterogeneous hydraulic properties. For that reason, proper system characterization, monitoring and modeling remain challenging. In this study, we investigated a geologically complex alpine catchment in the Dolomites (Italian Alps) by combining hydrogeological investigation, hydrological monitoring and numerical modelling. A process based but spatially lumped hydrological model was applied to simulate the continuous measured catchment discharge in a period of three years, which covers a large variation of hydrodynamic conditions. The current model structure couples the sequential hydrogeological units within the studied catchment: (1) the fractured dolomitic rocks as bedrock aquifer and 2) the unconsolidated deposits accumulating on the slopes and at the valley floor as porous aquifer. In order to evaluate the model structure and parameterization in depth, we applied a multi-step evaluation approach considering both parameter sensitivity and uncertainty. The current modelling results demonstrate that the newly developed model can reproduce most discharge behavior of aquifers. The model indicates a dynamic linkage between surface and subsurface storage units during different flow conditions. Besides the matrix and conduit flow in fractured dolomitic aquifer, it highlights the important role of unconsolidated sediments (porous aquifer) to the storage and discharge behavior of the entire groundwater system. Furthermore, with the comprehensive model evaluation we learned the model structure deficit during extreme high flow condition and proposed a more detailed hydrogeological conceptual model to improve the model realism.

Groundwater level forecasting with machine learning has been widely studied due to its generally accurate results and little input data requirements. Furthermore, machine learning models for this purpose are set up and trained in a short time when compared to the effort required for process-based, numerical models. Despite the high performance of models obtained at specific locations, applying the same model architecture to multiple sites across a regional area might lead to contrasting accuracies. Likely causalities of this discrepancy in model performance have been barely examined in previous studies. Here, we investigate the link between model performance and the effects of geospatial site characteristics and time series features. Using precipitation and temperature as predictors, we model groundwater levels at approximately 500 observation wells in Lower Saxony, Germany, using a 1-D convolutional neural network with a fixed architecture and hyperparameters tuned for each time series individually. The performances are evaluated against geospatial and time series features using correlation coefficients. Model performance is negatively influenced at sites near waterworks and densely vegetated areas. Besides, the more complex the time series, the higher the metrics, but autocorrelation reduces the model performance. The new insights evidence that further information is required at certain locations to improve model accuracy due to external impacts.

Since the 1970s, Holling's socio-ecological systems (SES) approach has been a most predominant theoretical force in resilience research in the context of the climate crisis. From Holling's approach, however, two contrasting scientific approaches to resilience have developed, namely, naturalism and constructivism. While naturalist resilience research takes SES as complex systems marked by non-linearity and evolutionary changes, constructivist resilience research focuses on the embeddedness of SES in heterogenous contexts. In naturalist resilience research resilience is defined as a system property, while in constructivist resilience research resilience is politically loaded and historically contingent. The aim of this paper is to review and structure current developments in resilience research in the field of climate change studies, in terms of the approaches, definitions, models and commitments that are typical for naturalism and constructivism; identify the key tension between naturalist and constructivist resilience research in terms of the widely discussed issue of adaptation and transformation, and discuss its implications for sustainable development; and propose a research agenda of topics distilled from the adaptation-transformation tension between naturalist and constructivist resilience research.

The lumped parameter modeling approach has been widely applied in karst hydrology for, among other applications, the understanding of their functioning of the assessment or groundwater availability in a context of global change. Nonetheless, such an approach generally does not account for land-cover land-use (LCLU) changes and their potential impacts on recharge processes. The study focuses on three forests dominated karst catchments: Kerschbaum (Lower Austria), Baget (French Pyrenees) and Oeillal (southern France), and investigates how LCLU changes in a lumped parameter modeling approach can affect both the internal fluxes and the model performance. The active subspace method is used to perform sensitivity analysis of model parameters, and to quantify parameter uncertainty. We show that the consideration of a semi-distributed recharge constitutes a relevant approach to capture the impact of LCLU changes on flow dynamics, but also introduces more uncertainty in the modeling approach. This approach may thus allow identifying the trade-off between modeling approach complexity and its performance. Finally, it gives new insight for the assessment of LCLU changes impacts on karst groundwater resource.

A common way to characterize catchments is to use catchment descriptors that summarize important physical aspects of a given catchment, often by aggregating large geospatial datasets into a single number. Such descriptors aim at extracting information that can inform us about different aspects of catchment functioning, help us to infer dominant hydrological processes, identify similarity among different sites, and transfer information across them. In this study we analyze a large sample of research articles indexed in Scopus, that were returned from the search words “catchment characteristic”, “catchment descriptor”, “catchment attribute”, “catchment indicator” or “catchment property”, to identify current practices of catchment characterization in hydrological science and related disciplines.We particularly focus on analyzing the variety of data sources on which catchment descriptors are usually based (e.g., digital elevation, land use, lithographic and soil texture maps), on identifying how the datasets are aggregated into catchment descriptors, and on exploring how the value of those descriptors is assessed.Based on this large sample of studies that cover diverse research areas (e.g., water quantity, water quality, lake research, aquatic ecosystems), different types of studies (e.g., data-based analysis, hydrological and statistical modeling, field studies) and various application purposes (e.g., descriptive site comparison, catchment clustering/classification, quantitative driver/control identification, regionalization), we provide a categorized overview of practices that are currently used for catchment characterization. This overview will provide guidance for future studies by summarizing the status quo and its strengths and limitations, and by providing suggestions for future research.

Due to EC’s easy recordability and the existence of a strong correlation between EC (electrical conductivity) and discharge in certain catchments, EC is a potential predictor of discharge. This potential has not yet to be widely addressed. In this paper, we investigate the feasibility of using EC as a proxy for long-term discharge monitoring in a small karst catchment where EC always shows a negative correlation with the spring discharge. Given their complex relationship, a special machine learning architecture, LSTM (Long Short Term Memory), was used to handle the mapping from EC to discharge. LSTM results indicate that the spring discharge can be predicted well with EC, particularly in storms when the dilution dominates the EC dynamic; however, the prediction may have relatively large uncertainties in the small or middle recharge events. A small number of discharge observations are sufficient to obtain a robust LSTM for the long-term discharge prediction from EC, indicating the practicality of recording EC in ungauged catchments for indirect discharge monitoring. Our study also highlights that the random or fixed-interval discharge measurement strategy, which covers various climate conditions, is more informative for LSTM to give robust predictions than other strategies. While our study is implemented in a karst catchment, the method may be also suitable for non-karst catchments where there is a strong correlation between EC and discharge.

In a context of climate, environmental, ecological and socio-economical changes, understanding and predicting the response of hydrological systems on regional to global spatial scales, and on infra-seasonal to multidecadal time-scales, are major topics that must be considered to tackle the challenge of water resource management sustainability. In this context, a number of strongly-linked key issues need to be addressed by the scientific community, including: (i) identifying climate drivers of hydrological variations, (ii) understanding the multi-frequency characteristics of hydroclimate variability, including evolution of extremes (meteorological/hydrological event scale to long-term natural/internal climate- or anthropogenic-driven variations and trends), (iii) assessing the influence of local- to regional-scale basin properties on hydrological system response to climate variability and change, (iv) identifying the evolving contribution of anthropogenic water use in observed hydrological variations. Based on pan-European collaborations, activities of the EURO-FRIEND “Large-scale variations in hydrological characteristics” group aim at generating new findings to improve our understanding of hydrological systems behavior sensu lato (i.e. surface and sub-surface) on large spatial and temporal scales (i.e continental – multidecadal). Through selected examples, this contribution emphasizes recent research developments in characterizing and modeling of climate-hydrology linkages at different temporal and spatial scales, as well as recent insights on climate-hydrology scaling characteristics (i.e. long-term persistence, dependance of processes, of hydrological behaviors, of large-scale climate/hydrology linkages on time-/spatial scales), long-term hydrometeorological reconstructions, and large-scale hydrological model refinement taking into account spatial heterogeneity of watershed physical characteristics.

Water resources from high-alpine karst aquifers are used for drinking, hydropower generation and artificial snowmaking. Therefore, understanding of their storage and flow dynamics is crucial for a sustainable water management. However, high-alpine karst areas are characterized by a great geological complexity due to the presence of mountain block fractured and karst aquifers interdigitating with the heterogeneous valley floor porous aquifers. For that reason, hydrogeological characterization and model prediction remains a big challenge. In this work, we investigated a geologically complex alpine catchment in the Dolomites (Italian Alps) by using experimental data and a reservoir numerical model to simulate three years of stream discharge. The structure of the model is based on experimental knowledge of the catchment and on previous studies and investigations. It (1) includes snow dynamics and accounts for hydrogeological heterogeneities, (2) separately considers karstic conduit and matrix flow in a dolomitic aquifer and flow through the porous deposits accumulating on the slopes and at the valley floor in an unconsolidated aquifer (non-karst), and (3) takes into account the groundwater transfer between the two aquifers. In the frame of a multi-step model evaluation, we used a Regional Sensitivity Analysis with three performance measures including observations of catchment discharge, karst spring discharge and unconsolidated aquifer spring discharge to assess the realism of model simulations. We show that the newly developed model reliably reproduces the hydrogeological variability of the catchment, even during strongly different hydroclimatic conditions. Analyzing its simulated storage dynamics, we can show that despite its moderate storage, the porous aquifer contributes most to catchment discharge, while the largest storage of the system is the matrix of the dolomite aquifer that recharges the unconsolidated aquifer together with discharge from the karstic conduits. A clear seasonality of groundwater storage in the karst matrix and of unconsolidated aquifer discharge indicates a strong sensitivity of this complex aquifer system to climatic variability.

The dissolution of carbonate rock ‘karstification’ creates pronounced surface and subsurface heterogeneity and results in complex flow and transport dynamics. Consequently, water resources managers face significant challenges keeping calm when dealing with karst water resources especially in times of environmental change. My lecture not only will provide an overview of the peculiarities of karst hydrology but it will also offer some approaches that facilitate the assessment of environmental changes on karst water resources. Using two case studies, one at the plot scale and the other at the scale of an entire continent, I will contrast the opportunities and challenges of dealing with karst across different scales and climatic regions. In particular, I will elaborate (1) how understanding on dominant karst processes can be obtained, (2) how this understanding can be incorporated into karst specific modelling approaches, and (3) how karst models developed at different scales can be used for water management and water governance. The presentation will conclude with some thoughts to facilitate less furious implementations of karst approaches for everyone.

In Ethiopia, more than 80% of big freshwater lakes are located in the Rift Valley Lake Basin, which is serving for multipurpose water use of over 30 million people. The basin is one of the most densely populated regions in Ethiopia and it covers an area of 53,035 km2. However, most of the catchments recharging these lakes are ungauged and their water balance is not well quantified, and hence, limiting the development of appropriate water resource management strategies. Prediction for ungauged catchments has demonstrated its effectiveness in hydro-climatic data-rich regions. However, these approaches are not well evaluated in the climatic data-limited condition and the consecutive uncertainty emerging in the small catchments is not adequately quantified. In this study, we use the HBV model to simulate streamflow using global precipitation and potential evapotranspiration products as forcings. We develop and apply a Monte-Carlo scheme to calibrate the model and quantify uncertainty at 16 catchments in the basin where gauging stations are available. Out of these, we use 14 best catchments to derive the best regional regression model by correlating the best calibration parameters, the best validation parameters, and parameters that give the most stable predictions with catchment attributes that are available throughout the basin. A weighting scheme in the regional regression accounts for parameter uncertainty in the calibration. A spatial cross-valuation that is applied 14 times always leaving out one of the gauged catchments provides 14 regional regression functions that express uncertainty regionalization. It also shows that the regionalization procedure that uses the best validation parameters for regionalization provides the most robust results. We then subsequently apply the 14 spatial regression functions of the cross-validation to the remaining 35 ungauged catchments in the Rift Valley Lake Basin to provide regional water balance estimations including quantification of regionalization uncertainty. With these results, our study provides a new procedure to use global precipitation and evapotranspiration products to predict and evaluate streamflow simulation for hydro-climatically data scares regions considering uncertainty. It, therefore, enhances the confidence in the understanding of water balance in those regions and will support the planning and development of appropriate water resource management strategies. Keywords: Parameters Estimation, Uncertainties, Ungauged Catchment, Weighted Regression, Water Balance

Continental- to global-scale hydrologic and land surface models increasingly include representations of the groundwater system, driven by crucial Earth science and sustainability problems. These models are essential for examining, communicating, and understanding the dynamic interactions between the Earth System above and below the land surface as well as the opportunities and limits of groundwater resources. A key question for this nascent and rapidly developing field is how to evaluate the realism and performance of such large-scale groundwater models given limitations in data availability and commensurability. Our objective is to provide clear recommendations for improving the evaluation of groundwater representation in continental- to global-scale models. We identify three evaluation approaches, including comparing model outputs with available observations of groundwater levels or other state or flux variables (observation-based evaluation); comparing several models with each other with or without reference to actual observations (model-based evaluation); and comparing model behavior with expert expectations of hydrologic behaviors that we expect to see in particular regions or at particular times (expert-based evaluation). Based on current and evolving practices in model evaluation as well as innovations in observations, machine learning and expert elicitation, we argue that combining observation-, model-, and expert-based model evaluation approaches may significantly improve the realism of groundwater representation in large-scale models, and thus our quantification, understanding, and prediction of crucial Earth science and sustainability problems. We encourage greater community-level communication and cooperation on these challenges, including among global hydrology and land surface modelers, local to regional hydrogeologists, and hydrologists focused on model development and evaluation.

Large-scale hydrological models have been widely used for water resources management, such as studying human impacts (e.g., pumping and irrigation) on groundwater. Currently, most of these models do not explicitly include karst features for the recharge and groundwater simulations. However, the geological properties in karst regions substantially differ from non-karst areas, which makes recharge and groundwater flow behaviors distinctly different between the two types of systems. Due to challenges of combining karstic and non-karstic processes, of simulating inter-catchment groundwater flow, and of parameterizing karstification over large areas in karst regions, global karst groundwater flow models currently do not exist. In this study, we propose a general approach to integrate karstic and non-karstic processes and a hierarchical approach to confine the karstic groundwater flow parameters over large domains. First, we selected six karstic catchments (with different catchment sizes and climates) with adequate observations to test the combination of karstic and non-karstic simulations at the aquifer and catchment scale. We show that using system signatures helps to identify the necessary model structures and to integrate karstic and non-karstic processes. Second, we defined an Inter-catchment Groundwater Flow index (IGF) to quantitatively address groundwater flow crossing topographic boundaries. Third, we classify the level of karstification based on spring and catchment properties and evaluate different strategies for parameterization of karstic groundwater flow processes at varying degrees of karstification. Overall, our study provides a solid basis for a continental-scale karstic groundwater flow model, complementary to current global scale hydrologic modeling efforts where this process is still missing.

Groundwater transit time refers to the travel time of a water molecule through an aquifer from recharge at the water table to discharge at a surface water body (e.g., river). Analysing transit times provides a primary way to understand the overall transport characteristics of a hydrological system and is of interest in many aspects of environmental management. For example, studying transit time distribution can facilitate the mitigation of pollutant transport risks and ecosystem restoration. Hydrogeological heterogeneity of an aquifer is a major controlling factor for groundwater flow paths and transit time distributions. In this study, we investigate the impacts of spatial variability of hydrogeological properties on transit times by combining measurements and a new semi-analytical numerical modelling scheme. Passive tracer transport data from several catchments in Europe are obtained from open databases. Groundwater transit time in these catchments are inferred from both tracer transport data and numerical modelling. Comparing the results in different catchments provides a comprehensive way of understanding the impact of hydrogeological heterogeneity on groundwater transit time.

Model assessment is a crucial part of hydrological modelling studies. The traditional intuitive approach is to judge model performance to explore its effectiveness and informativeness about the system reality. However, this approach does not necessarily guarantee that one captures system hydrological functioning from the model output. Here, we proposed a novel model assessment strategy that provides a direction to constrain model output space with the concept of model functionality. In the study, we used StorAge Selection (SAS) function approach as a process diagnostic tool to explore the model output space which is simultaneously informative on the system hydrological behaviour and functioning. To do that, the SAS model was fed by a karst-dedicated hydrological model (VarKarst) simulations to model the karst aquifer d18O transport and young water fraction of system discharge, Fyw. Model functionality was assessed by a new model verification metric, named Exceedance Probability Ranked Score, EPRS. The eligible clusters from the model output were then served in order to examine the model parameter space. Our findings provide direction to indicate that using young water fraction, Fyw as a process-diagnostic metric leads to an improvement in model realism while carrying a physically realistic model parameter set throughout the model output space.

Groundwater pollution threatens human and ecosystem health in many areas around the globe. Shortcuts to the groundwater through enlarged cracks and fissures, often referred to as concentrated recharge, are known to transmit short-lived pollutants into carbonate aquifers endangering water quality of around a quarter of the world population. However, the large-scale impact of concentrated recharge on water quality remains poorly understood. Here we apply a continental-scale model to quantify for the first time the danger of groundwater contamination by degradable pollutants through concentrated recharge in carbonate rock regions. We show that concentrated recharge is the primary reason for the rapid transport of contaminants to the groundwater, increasing the percentage of non-degraded pollutants from

Lumped karst hydrological models often suffer from the over-simplified structures. In recent days, hydrochemical has been used as an auxiliary information to define realistic karst model structures. For karst aquifers, the hydrochemical dynamics of karst springs contain important information about the internal behavior of the karst aquifer that can be conducive to model identification and calibration. In this presentation, electrical conductivity (EC), as a substitution for Ca2+ concentration of the spring discharge, is evaluated for its potential for model structure identification and reduction of simulation uncertainty. A new framework to integrate EC in karst modeling is developed at a small, well-instrumented karst catchment near Guilin City, China. A set of different hydrologic models were enabled to consider the linear dissolution process of Ca2+ and its transport to investigate the models’ abilities to reproduce the behavior of spring discharge and EC. We found that most hydrologic models obtain similar performances concerning spring discharge, however, their performances in simulating spring EC show obvious differences. The combination of observed spring discharge and EC could identify a more realistic model structure that was in accordance with the observed perception of this karst aquifer’s functioning. Using the identified “most realistic” model, we use sensitivity analysis to show that spring EC only improves the identifiability of one hydrologic parameter mostly due to too complex EC dynamics during recharge events and therefore has limited potential to reduce discharge simulation uncertainty. Consequently, our new framework to include EC in karst models opens new doors for more realistic simulation but an explicit treatment of uncertainties remains necessary due to EC’s limited potential to reduce simulation uncertainty. The next step of our work is to integrate appropriate nonlinear dissolution process of carbonate rock in the model to further improve the simulation of spring EC (Ca2+).

Dissolution of CaCO3 in calcareous soils is mainly governed by CO2 which forms a weak but ubiquitous acid in the aqueous phase. Soil CO2 concentrations are generally higher than atmospheric concentrations due to the CO2 production in the soil. It is generally assumed, that it is mainly the CO2 concentration in the soil and the discharge that control the re-location of CaCO3, and thus the further formation of soil and karst. In most cases soil and karst systems are considered to be static and that the CaCO3 dissolution process is a steady state process. However, we know that soil CO2 concentrations can be highly dynamic and are affected by soil temperature and soil moisture. Our objective was to investigate whether this steady state assumption regarding carbonate dissolution and transport can be applied or whether we have to consider the dynamics and interaction of soil CO2 and dissolution of CaCO3 in the aqueous phase.We report on insights from a 3 year field study in a calcareous soil in which soil CO2 concentrations and its response to soil moisture and precipitation were investigated. Low intensity precipitation resulted in slow increase in soil CO2 concentration, since increased soil water content blocks formerly air-filled pores. Intense precipitation events were followed by fast infiltration and probably preferential flow. Intense precipitation also resulted in temporary drops in soil CO2. These drops can be explained by a relative under-saturation of the soil solution at a certain depth. The soil solution is mixed with infiltrating rain water, which is still equilibrated with the lower atmospheric CO2 concentrations and thus drawing CO2 from the surrounding soil air. These mechanisms should results in a much stronger dissolution of local CaCO3 and net transport of dissolved CaCO3.A following laboratory experiment on mesocosms of natural soil and restructured soil was used to test and reproduce the observed CO2 patterns as well as dissolution and transport of carbonate due to precipitation events. These experiments also showed that higher intensity of precipitation results in stronger drops in soil CO2 concentration and higher transport rates of dissolved CaCO3. Hydrus1D was used to model soil CO2 dynamics and dissolution of CaCO3 in the aqueous phase for the measured scenarios. The observed general pattern of the “drops” of soil CO2 could be easily reproduced confirming the assumption of CO2 undersaturated soil water right after the precipitation events. The natural soil mesocosm showed comparable patterns in all precipitations experiments. The restructured soil mesocosm showed a high mobilization and drainage during the first precipitations experiments which then fast declined to the level of the natural soil mesocosm. We interpret this as fast dissolution and washing off of carbonates attached to the macropore surfaces in which preferential flow occurs.We conclude that dynamics and interaction of soil CO2 and dissolution of CaCO3 in the aqueous phase are highly dynamic and affected by preferential flow. It seems that general patterns can be reproduced using Hydrus 1D, with the hydrological parametrization as a major challenge.This research was financially supported by DFG (MA 5826/2-1).

About 50% of the Slovene drinking water demand are covered by karst aquifers. Consequently, appropriate protection and sustainable management of these aquifers are essential. Due to the hydrologic complexity and variability of karst systems, predicting potential contaminations and aquifers’ responses to changes in climate conditions, remains a challenge for karst research. In this study, the dynamics of potential solute contaminant transport in the Unica springs catchment, located in the southwest of Slovenia, are characterized and simulated with a semi-distributed karst model. The catchment encompasses autogenic and allogenic recharge across an area of about 820 km². The autogenic karst massive hosts one of Europe’s largest confluences of subterranean river branches and is strongly karstified. To take into account the temporal and spatial characteristics of the catchment, the model is linked with a GIS-based approach to assess spatiotemporal karst vulnerability. The validation of the model is performed by discharge observations and tracer experiments. The resulting simulations enable us to identify hot spots and hot moments of high contamination risks. By simulating solute contaminant transport during extremely dry and extremely wet years, we show that contaminant transport slows down under dry conditions. This can be explained by fast flow pathways not being activated or hydraulically connected under low flow conditions, which results in a retention and dilution of the contaminant in the aquifer. Our new approach improves the understanding of potential contaminants' transport behavior in a large complex karst system and justifies the consideration of spatiotemporal hydrologic variability in solute transport forecasting. It proposes a good basis for a better evaluation, management and protection of water resources in karst areas.

Excess Nitrogen (N) deposition from industrial, domestic and agricultural sources has led to increased nitrate leaching, increased gaseous N emissions, the loss of biological diversity, and has affected C sequestration in forest ecosystems. Nitrate leaching affects the purity of karst water resources, which contribute around 50 % to Austria’s drinking water supply. Here we present the first comprehensive evaluation of a 26 years record of dissolved inorganic N (DIN) concentrations and fluxes from a karst catchment in the Austrian Alps (LTER Zöbelboden), which was not affected by local N sources but solely by long-range N deposition (20-25 kg N ha-1 y-1 total N deposition). We inferred from soil chemical and microbial data as well as nitrate leaching, that the forest ecosystems in the catchment are likely saturated with respect to nitrogen. Consequently, 60-70% of the atmospheric N input was lost via leaching of NO3- to the karst aquifer or emission of N2O to the atmosphere. However, due to high dilution DIN concentrations in the runoff rarely exceed 2 mg N l-1. An exception were periods of forest disturbances. A number of strong storms (2007-2008) caused some major windthrows as well as single tree damages (5-10% of the catchment). Runoff concentrations of DIN showed clear responses to the disturbances with an increase (~ 1 mg N l-1) until 2008/09 and a decreased again in 2010/11 to pre-disturbance levels. Apart from disturbances, drought years led to an increase in NO3- in the soil water in the following years. We observed the subsequent changes of the dynamics of DIN in runoff with a high-resolution water probe during 2018 and 2019. This data shows that the severity of the drought and the magnitude of the first rewetting event after a period of drought drives the size of the flush of DIN. It is likely that N deposition will lower with legislated emission reductions and that the currently N leaky ecosystems may immobilize more N when climate is becoming warmer in the future. However, we hypothesize that the karst aquifer will still receive DIN rich runoff water due to long-term lags in the recovery of a closed N cycle and because of expected climate events such as storms and droughts.

Karst systems are important for drinking water supply. Future climate projections indicate increasing temperature and a higher frequency of strong weather events. Both will influence the availability and quality of water provided from karst regions. Forest disturbances such as windthrow can disrupt ecosystem cycles and cause pronounced nutrient losses from the ecosystems. In this study, we consider the time period before and after the wind disturbance period (2007/08) to identify impacts on DIN (dissolved inorganic nitrogen) and DOC (dissolved organic carbon) with a process-based flow and solute transport simulation model. When calibrated and validated before the disturbance, the model disregards the forest disturbance and its consequences on DIN and DOC production and leaching. It can therefore be used as a baseline for the undisturbed system and as a tool for the quantification of additional nutrient production. Our results indicate that the forest disturbance by windthrow results in a significant increase of DIN production lasting ∼ 3.7 years and exceeding the pre-disturbance average by 2.7 kg ha−1 a−1 corresponding to an increase of 53 %. There were no significant changes in DOC concentrations. With simulated transit time distributions we show that the impact on DIN travels through the hydrological system within some months. However, a small fraction of the system outflow ( 1 year.

Karst aquifers are an important source of drinking water in many regions of the world. Karst areas are highly permeable and produce large amounts of groundwater recharge, while surface runoff is typically negligible. As a result, recharge in these systems may have a different sensitivity to climate and land cover changes compared to other less permeable systems. However, little effort has been directed toward assessing the impact of climate and land cover change in karst areas at large-scales. In this study, we address this gap by (1) introducing the first large-scale hydrological model including an explicit representation of both karst and land cover properties, and by (2) analysing the model's recharge production behaviour. To achieve these points, we first improve the evapotranspiration estimation of a previous large-scale karst recharge model (VarKarst). The new model (V2Karst V1.0) includes a parsimonious representation of relevant ET processes for climate and land cover change impact studies. We demonstrate the plausibility of V2Karst simulations at carbonate rock FLUXNET sites using soft rules and global sensitivity analysis. Then, we use virtual experiments with synthetic data to assess the sensitivity of simulated recharge to precipitation characteristics and land cover. Results reveal how both vegetation and soil parameters control the model behaviour, and they suggest that simulated recharge is sensitive to both precipitation (overall amount and temporal distribution) and land cover. Large-scale assessment of future karst groundwater recharge should therefore consider the combined impact of changes in land cover and precipitation properties, if it is to produce realistic projections of future change impacts.

Chalk aquifers are an important source of drinking water in the UK. Understanding and predicting groundwater levels is therefore important for effective water management of this resource. Chalk is known for its high porosity and, due to its dissolvability, exposed to karstification and strong subsurface heterogeneity. To cope with the karstic heterogeneity and limited data availability, specialised modelling approaches are required that balance model complexity and data availability. In this study we present a novel approach to simulate groundwater level frequency distributions with a semi-distributed karst model that represents subsurface heterogeneity by distribution functions. Simulated groundwater storages are transferred into groundwater levels using evidence from different observations wells. Using a newly developed percentile approach we can simulate the number of days exceeding or falling below selected groundwater level percentiles. Firstly, we evaluate the performance of the model to simulate three groundwater time series by a spilt sample test and parameter identifiability analysis. Secondly, we apply a split sample test on the simulated groundwater level percentiles to explore the performance in predicting groundwater level exceedances. We show that the model provides robust simulations of discharge and groundwater levels at 3 observation wells at a test site in chalk dominated catchment in Southwest England. The second split sample test also indicates that percentile approach is able to reliably predict groundwater level exceedances across all considered time scales up to their 75th percentile. However, when looking at the 90th percentile, it only provides acceptable predictions for the longest available time scale and it fails when the 95th percentile of groundwater exceedance levels is considered. Modifying the historic forcings of our model according to expected future climate changes, we create simple climate scenarios and we show that the projected climate changes may lead to generally lower groundwater levels and a reduction of exceedances of high groundwater level percentiles.

In karst systems, surface near dissolution carbonate rock results in a high spatial and temporal variability of groundwater recharge. To adequately represent the dominating recharge processes in hydrological models is still a challenge, especially in data scare regions. In this study, we developed a recharge model that is based on a perceptual model of the epikarst. It represents epikarst heterogeneity as a set of system property distributions to produce not only a single recharge time series, but a variety of time series representing the spatial recharge variability. We tested the new model with a unique set of spatially distributed flow and tracer observations in a karstic cave at Mt. Carmel, Israel. We transformed the spatial variability into statistical variables and apply an iterative calibration strategy in which more and more data was added to the calibration. Thereby, we could show that the model is only able to produce realistic results when the information about the spatial variability of the observations was included into the model calibration. We could also show that tracer information improves the model performance if data about the variability is not included.