Birk, Steffen (2002):

Characterisation of Karst Systems by Simulating Aquifer Genesis and Spring Responses: Model Development and Application to Gypsum Karst.-

Tübinger Geol. Abh., Reihe C, Nr. 60, 122 S., 105 Abb., 8 Tab.; Tübingen, Januar 2002.

Abstract:
Karst aquifers are important groundwater resources, which are highly vulnerable to contamination due to fast transport in solutionally enlarged conduits. Management and protection of karst water resources require an adequate aquifer characterisation at the catchment scale. Due to the heterogeneity and complexity of karst Systems, this is not easily achieved by standard investigation techniques such as pumping tests. Therefore, a process-based numerical modelling tool is developed, designed to support the karst aquifer characterisation using two complementary approaches: Firstly, the simulation of conduit enlargement, which aims at predicting aquifer properties by forward modelling of long-term karst genesis; secondly, the simulation of heat and solute transport processes, which aims at inferring aquifer properties from short-term karst spring response after recharge events.

Karst genesis modelling is applied to a conceptual setting based on field observations from the Western Ukraine, where the major part of known gypsum caves is found. Gypsum layers are typically supplied by artesian flow of aggressive water from insoluble aquifers underneath. Processes and parameters, controlling solutional enlargement of single conduits under artesian conditions, are identified in detailed sensitivity analyses. The development of conduit networks is examined in parameter studies, suggesting that the evolution of maze caves is predetermined by structural preferences such as laterally extended fissure networks beneath a horizon less prone to karstification. Without any structural preferences vertical shafts rather than maze caves are predicted to develop. The structure of the mature conduit system is found to be determined during early karstification, which is characterised by high hydraulic gradients and low flow rates in the gypsum layer.

Short-term karst spring response after recharge events is firstly examined in parameter studies by for-ward modelling. The numerical simulations reveal that different controlling processes of heat and solute transport account for the different behaviour of water temperature and solute concentration frequently observed at karst springs. lt is demonstrated that these differences may be employed to reduce the ambiguity in the aquifer characterisation.

In order to test the feasibility of the corresponding inverse approach, which aims at inferring aquifer properties from the karst spring response, the model is applied to a field site in Southern Germany Urenbrunnen spring, Vöhringen). Data input is provided by both literature and own field work. Several models, which reproduce the results of a combined tracer and recharge test, are calibrated to spring discharges and solute concentrations measured after a recharge event. In order to validate the calibrated models, the measured spring water temperatures are simulated by heat transport modelling. The model application yields information on aquifer properties as well as flow and transport processes at the field site. Advection is identified as the dominant transport process, whereas the dissolution reaction of Gypsum is found to be insignificant in this case.

The application to gypsum aquifers demonstrates that both suggested approaches are suitable for the characterisation of karst systems. Model results, however, are highly sensitive to several input parameters. in particular in karst genesis modelling. Therefore, extensive field work is required to provide reliable data for site-specific model applications. In order to account for uncertainties, it is recommended conduct parameter studies covering possible ranges of the most influential parameters.

Dissertation at the Geowissenschaftliche Fakultät, Universität Tübingen, Center for Applied Geoscience, Sigwartstr. 10, 72076 Tübingen, Germany

Impressum / Datenschutz