Identification and quantification of the effects of flow regime and matrix-conduit interaction in the characterization of karst aquifers
von Markus Giese
Datum der mündl. Prüfung:2017-05-03
Erschienen:2018-03-13
Betreuer:Prof. Dr. Martin Sauter
Gutachter:Prof. Dr. Martin Sauter
Gutachter:Prof. Dr. Rudolf Liedl
Dateien
Name:20180312_Diss_Giese_publ_sub.pdf
Size:4.49Mb
Format:PDF
Zusammenfassung
Englisch
In addition to alluvial aquifers, karst aquifers are counted among the most important groundwater resources. Besides the high benefit of karstic springs, they also pose risks to the user or people living close to those springs. The highly permeable structures within the low-permeable hard rock, which are the result of chemical solution processes, can result in high contaminant concentration and high flow velocity towards the sources. In addition to the high susceptibility to water quality, karst sources have a high potential for damage caused by flooding in the downstream area of the sources. The characterization of karst aquifers is usually based on the evaluation of karst spring hydrographs as an integral signal of the entire catchment area. The spatial distribution of karst-specific heterogeneities and thus their interactions are excluded. A further disadvantage with the analyses of spring discharge is the usually unknown input signal into the system. Due to the application of large-scale pumping tests in karst aquifers, i.e. the intensive and long-term water abstraction from the conduit system, the system can be triggered by a clearly defined input signal. The aim of this thesis is the systematic analysis of pumping tests with regard to the heterogeneities and hydraulic properties for the large number of different conceptual models of karst systems. The work focuses on the changes of local and regional flow patterns caused by different hydraulic properties on different scales of the karstified catchment, discrete and diffuse boundary conditions as well as the interactions between the conduit system and the hard rock matrix. For this purpose, a discrete conduit-continuum model is employed, which was adapted according to the intended use. The analysis is divided into several sections, focusing on different phases and processes of a pumping test. The first section of a pumping test is dominated by internal boundary conditions caused by fast-responding storage and interactions with the adjacent, low permeable environment. Two model parameters are converted into two dimensionless parameters, which are frequently used for the interpretation of classical pump tests, in order to approximate the drawdown behavior of an analytical solution. The conceptual process description associated with the analytical solution allows the two dimensionless parameters to create a characterization scheme for the saturated zone of a karst aquifer. By combining the numerical model parameters with conceptual models, a further conceptual model can be defined in which the exchange flow between the conduit system and the uniformly karstified matrix is limited by the hydraulic interface properties and the resulting effects. Following the period affected by the interface interactions, the flow period is dominated by conduit flow. Scientific sources make a distinction between laminar and turbulent pipe/conduit flow. In general, in karst aquifers there is a lack of information regarding the conduit properties, especially the diameter or the roughness of the conduit, which are required for determining the current state of flow. Applying a defined pump rate, these tube-specific parameters remain as the residual members. Those can be used to calculate quantitative differences, i.e. head losses along the flow direction, between laminar and turbulent flow. These head losses influence the conduit flow behavior as well as the flow pattern on a regional scale. The results of the analysis show that the errors caused by applying laminar flow equations are very small in mature karst systems, whereas the application of turbulent flow equations is demanded for less developed karst systems. For large-scale pumping tests in karst aquifers, such effects superpose both temporally and spatially, which must be considered for the interpretation of drawdown curves. Therefore, a holistic approach is needed that can detect the influences of heterogeneity on the drawdown behavior. The starting point for this analysis is a large-scale pumping test at the Cent Fonts catchment (Languedoc, France). The measured drawdown curve shows that flow regimes during water abstraction in karst aquifers differ from the idealized solutions. The application of the flow dimension concept is an advanced analysis tool without further research demand. The effects of different boundary conditions on the propagation of the cone of depression and thus on the shape of the drawdown curve can be investigated for idealized catchments by applying the flow dimension approach. The results show that the flow behavior during the pumping test is strongly influenced by a radial flow component and is not dominated by a high permeability conduit as previously assumed.
Keywords: Karst; Pumping test; Flow dimension; Turbulent flow; Laminar flow