Process-based monitoring and modeling of Karst springs - Linking intrinsic to specific vulnerability.

The presented work illustrates to what extent field investigations as well as monitoring and modeling approaches are necessary to understand the high discharge dynamics and vulnerability of Karst springs. In complex settings the application of 3D geological models is essential for evaluating the vulnerability of Karst systems. They allow deriving information on catchment characteristics, as the geometry of aquifers and aquitards as well as their displacements along faults. A series of Karst springs in northwestern Switzerland were compared and Karst system dynamics with respect to qualitative and quantitative issues were evaluated. The main objective of the studies was to combine information of catchment characteristics and data from novel monitoring systems (physicochemical and microbiological parameters) to assess the intrinsic vulnerability of Karst springs to microbiological contamination with simulated spring discharges derived from numerical modeling (linear storage models). The numerically derived relation of fast and slow groundwater flow components enabled us to relate different sources of groundwater recharge and to characterize the dynamics of the Karst springs. Our study illustrates that comparably simple model-setups were able to reproduce the overall dynamic intrinsic vulnerability of several Karst systems and that one of the most important processes involved was the temporal variation of groundwater recharge (precipitation, evapotranspiration and snow melt). Furthermore, we make a first attempt on how to link intrinsic to specific vulnerability of Karst springs, which involves activities within the catchment area as human impacts from agriculture and settlements. Likewise, by a more detailed representation of system dynamics the influence of surface water, which is impacted by release events from storm sewers, infiltrating into the Karst system, could be considered. Overall, we demonstrate that our approach can be the basis for a more flexible and differentiated management and monitoring of raw-water quality of Karst springs.

Spatiotemporal scales of river-groundwater interaction - The role of local interaction processes and regional groundwater regimes.

Drinking water production in the vicinity of rivers not only requires the consideration of different spatiotemporal scales and settings of river-groundwater interaction processes, but also of local and regional scale groundwater regimes. Selected case studies in combination with field-experiments and the setup of high-resolution groundwater flow models enabled the investigation of the spatiotemporal development of microbial (classical fecal indicator bacteria and total cell counts) and selected organic micropollutants in riverine and regional groundwater for different hydrological settings, including low and high flow conditions. Proxy indicators suitable as surrogates for the diverse contaminations in alluvial aquifers with different settings could be identified. Based on the study results, the basic elements for both groundwater management and river restoration concepts are derived, which include the: (1) compilation and evaluation of the "current state" concerning hydrogeology, microbiology and contamination by organic micropollutants, (2) definition of field-experiments to qualitatively assess variability related to the "current state", and (3) quantitative assessment of groundwater regimes, including variability of groundwater components and inflow areas, by application of high-resolution groundwater flow models. The validity and transferability of the concept and inferred controls (specifically drivers and controls of river-groundwater interaction) are tested by evaluations derived from hydraulic relationships to river sections with comparable settings and regional groundwater flow regimes in general. The results of our investigations illustrate the influence of dynamic hydrologic boundary conditions on river-groundwater interaction and of regional scale groundwater flow regimes on the water composition of riverine groundwater systems. It is demonstrated how to identify river sections and their variations with intensified river-groundwater exchange processes and how to quantify the transient character of the different groundwater components that constitute the raw water quality of drinking water wells near rivers.

Characterization of a managed aquifer recharge system using multiple tracers.

Knowledge about the residence times of artificially infiltrated water into an aquifer and the resulting flow paths is essential to developing groundwater-management schemes. To obtain this knowledge, a variety of tracers can be used to study residence times and gain information about subsurface processes. Although a variety of tracers exists, their interpretation can differ considerably due to subsurface heterogeneity, underlying assumptions, and sampling and analysis limitations. The current study systematically assesses information gained from seven different tracers during a pumping experiment at a site where drinking water is extracted from an aquifer close to contaminated areas and where groundwater is artificially recharged by infiltrating surface water. We demonstrate that the groundwater residence times estimated using dye and heat tracers are comparable when the thermal retardation for the heat tracer is considered. Furthermore, major ions, acesulfame, and stable isotopes (δ2H and δ18O) show that mixing of infiltrated water and groundwater coming from the regional flow path occurred and a vertical stratification of the flow system exist. Based on the concentration patterns of dissolved gases (He, Ar, Kr, N2, and O2) and chlorinated solvents (e.g., tetrachloroethene), three temporal phases are observed in the ratio between infiltrated water and regional groundwater during the pumping experiment. Variability in this ratio is significantly related to changes in the pumping and infiltration rates. During constant pumping rates, more infiltrated water was extracted, which led to a higher dilution of the regional groundwater. An infiltration interruption caused however, the ratio to change and more regional groundwater is extracted, which led to an increase in all concentrations. The obtained results are discussed for each tracer considered and its strengths and limitations are illustrated. Overall, it is demonstrated that aquifer heterogeneity and various subsurface processes necessitate application of multiple tracers to quantify uncertainty when identifying flow processes.

Estimating the spatial distribution of artificial groundwater recharge using multiple tracers.

Stable isotopes of water, organic micropollutants and hydrochemistry data are powerful tools for identifying different water types in areas where knowledge of the spatial distribution of different groundwater is critical for water resource management. An important question is how the assessments change if only one or a subset of these tracers is used. In this study, we estimate spatial artificial infiltration along an infiltration system with stage-discharge relationships and classify different water types based on the mentioned hydrochemistry data for a drinking water production area in Switzerland. Managed aquifer recharge via surface water that feeds into the aquifer creates a hydraulic barrier between contaminated groundwater and drinking water wells. We systematically compare the information from the aforementioned tracers and illustrate differences in distribution and mixing ratios. Despite uncertainties in the mixing ratios, we found that the overall spatial distribution of artificial infiltration is very similar for all the tracers. The highest infiltration occurred in the eastern part of the infiltration system, whereas infiltration in the western part was the lowest. More balanced infiltration within the infiltration system could cause the elevated groundwater mound to be distributed more evenly, preventing the natural inflow of contaminated groundwater. Dedicated to Professor Peter Fritz on the occasion of his 80th birthday.

Assessment of source tracking methods for application in spring water.

For discriminating between human and animal faecal contamination in water, microbial source tracking (MST) approaches using different indicators have been employed. In the current study, a range of 10 such MST indicators described in the scientific literature were comparatively assessed. Bacteriophages infecting host strains of Bacteroides (GA-17, GB-124 and ARABA 84) as well as sorbitol-fermenting bifidobacteria proved useful for indicating human faecal contamination while Rhodococcus coprophilus was associated with animal-derived faecal contamination. These potential source indicators were present in samples of faecal origin, i.e. either in human wastewater or animal waste, from many different regions in Switzerland and therefore showed a geographic stability. In addition, the MST indicators were abundant in surface water and were even sensitive enough to detect faecal contamination in spring water from two study areas in Switzerland. This is the first study that has compared and successfully applied MST methods in spring water.

Improved detection of Rhodococcus coprophilus with a new quantitative PCR assay.

Agricultural practices, such as spreading liquid manure or the utilisation of land as animal pastures, can result in faecal contamination of water resources. Rhodococcus coprophilus is used in microbial source tracking to indicate animal faecal contamination in water. Methods previously described for detecting of R. coprophilus in water were neither sensitive nor specific. Therefore, the aim of this study was to design and validate a new quantitative polymerase chain reaction (qPCR) to improve the detection of R. coprophilus in water. The new PCR assay was based on the R. coprophilus 16S rRNA gene. The validation showed that the new approach was specific and sensitive for deoxyribunucleic acid from target host species. Compared with other PCR assays tested in this study, the detection limit of the new qPCR was between 1 and 3 log lower. The method, including a filtration step, was further validated and successfully used in a field investigation in Switzerland. Our work demonstrated that the new detection method is sensitive and robust to detect R. coprophilus in surface and spring water. Compared with PCR assays that are available in the literature or to the culture-dependent method, the new molecular approach improves the detection of R. coprophilus.

Identification of fecal input sites in spring water by selection and genotyping of multiresistant Escherichia coli.

The localization of fecal input sites is important for water quality management. For this purpose, we have developed a new approach based on a three-step procedure, including a preparatory phase, the screening of multiresistant bacteria using selective agar plates, and a typing phase where selected Escherichia coli isolates are characterized by antibiotic resistance profiles and molecular fingerprinting techniques (pulsed-field gel electrophoresis [PFGE]). These two well-known source tracking methods were combined in order to reduce cost and effort. This approach was successfully applied under field conditions in a study area located in the north-western part of Switzerland. E. coli isolates from spring water and surface water samples collected in this area were screened with selective agar plates. In this way, 21 different groups, each consisting of strains with the same pattern of antibiotic resistance, were found. Of these, four groups were further analyzed using PFGE. Strains with identical PFGE profiles were detected repeatedly, demonstrating the suitability of this method for the localization of fecal input sites over an extended period of time. Identical PFGE patterns of strains detected in water from two different springs were also found in the stream flowing through the study area. These results demonstrated the applicability of the new approach for the examination of incidents of fecal contamination in drinking water. The advantages of the described approach over genotyping methods currently being used to identify sources of fecal contaminants are a reduction in time, costs, and the effort required. Identical isolates could be identified without the construction of large libraries.

Novel Bacteroides host strains for detection of human- and animal-specific bacteriophages in water.

Bacteriophages active against specific Bacteroides host strains were shown to be suitable for detection of human faecal pollution. However, the practical application of this finding is limited because some specific host strains were restricted to certain geographic regions. In this study, novel Bacteroides host strains were isolated that discriminate human and animal faecal pollution in Switzerland. Two strains specific for bacteriophages present in human faecal contamination and three strains specific for bacteriophages indicating animal faecal contamination were evaluated. Bacteriophages infecting human strains were exclusively found in human wastewater, whereas animal strains detected bacteriophages only in animal waste. The newly isolated host strains could be used to determine the source of surface and spring water faecal contamination in field situations. Applying the newly isolated host Bacteroides thetaiotaomicron ARABA 84 for detection of bacteriophages allowed the detection of human faecal contamination in spring water.

Validation of a numerical indicator of microbial contamination for karst springs.

Rapid changes in spring water quality in karst areas due to rapid recharge of bacterially contaminated water are a major concern for drinking water suppliers and users. The main objective of this study was to use field experiments with fecal indicators to verify the vulnerability of a karst spring to pathogens, as determined by using a numerical modeling approach. The groundwater modeling was based on linear storage models that can be used to simulate karst water flow. The vulnerability of the karst groundwater is estimated using such models to calculate criteria that influence the likelihood of spring water being affected by microbial contamination. Specifically, the temporal variation in the vulnerability, depending on rainfall events and overall recharge conditions, can be assessed and quantified using the dynamic vulnerability index (DVI). DVI corresponds to the ratio of conduit to diffuse flow contributions to spring discharge. To evaluate model performance with respect to predicted vulnerability, samples from a spring were analyzed for Escherichia coli, enterococci, Clostridium perfringens, and heterotrophic plate count bacteria during and after several rainfall events. DVI was shown to be an indication of the risk of fecal contamination of spring water with sufficient accuracy to be used in drinking water management. We conclude that numerical models are a useful tool for evaluating the vulnerability of karst systems to pathogens under varying recharge conditions.