Abstract:
Karstaquifers represent an important groundwater resource and they are highly vulnerable with respect to contamination due to fast transport through the karst system and due to the limited attenuation of contaminants in karst terraines. In order to understand such a complex system, consisting of the compartments epikarst, vadose zone and the phreatic saturated zone, and to predict the variations in concentrations of point and non-point source contaminants, numerical groundwater modelling techniques are frequently employed.
This study considers two main aspects: firstly, the selection of an appropriate modelling tool for the simulation and prediction of groundwater flow and transport in a karst aquifer system, the adaptation of the model to the specific problem, the construction, testing as well as validation of the model. Secondly, the provision of data input, required for the modelling of such a complex flow system.
A Double-Continuum Porous Equivalent Model was found to be the most appropriate for the study area. Each of the two flow systems of the aquifer, the fast flow, i.e. conduit system, and the slow, i.e. fissured system, is simulated with a separate continuum, and both are coupled by an exchange term. The model applies to flow from a catchment, that discharges at a spring, or where the flow into a river via alluvial deposits can be quantified. A large number of borehole logs together with information from water level fluctuations and fluvial history enabled the identification of aquifer base and zones of higher and lower hydraulic conductivities.
A soil moisture balance approach together with a water balance for the canopy (accounting for interception loss) produced sufficiently accurate values for groundwater recharge. Important parameters in the recharge calculation are the field capacity the interception capacity and the rapid recharge threshold. Although the rapid recharge component as a total percentage of the annual recharge is of only minor importance, it provides recharge for periods, where otherwise no recharge would have been calculated, although observed in the field. The importance of the influence of the epikarstic zone on the temporal distribution of recharge input and the relative distribution of recharge between fast and slow system needs to be stressed. The analysis of time series of hydraulic and physicochemical parameters yielded storage and recession constants for. the drainage from the subcutaneous zone, which could be devided into a fast and a slow recharge component.
The analysis of spring flow, temporal and spatial variation of groundwater levels, borehole hydraulic tests and also the time series of physico-chemical parameters of spring water allowed the quantification of storage and hydraulic conductivity for both, the slow (fissured) and the fast (conduit) system, for flood as well as drought conditions. The importance of the dependance of the above parameters on the measurement scale due to the heterogeneous distribution of fast and slow flow paths was recognized. The relationship between hydraulic conductivity and the sampled volume could be determined by evaluating the parameters at the catchment, several different borehole and laboratory scales. This procedure allowed the allocation of hydraulic conductivity at the modelling scale. The analysis of the various time series provided an insight into the different processes affecting the recharge and groundwater circulation. The chronology of the different parameter variations helped to understand the sequence of events.
The dominant hydraulic parameters in the flow simulation are the storage coefficient and the hydraulic conductivity of the fissured system, that control the storage and release of groundwater over time. The transport model is applicable to point-source and non-point source input, which also allows the modelling of complex recharge events, where a number of events of different intensity are superimposed. The determining hydraulic parameters in the transport simulation have been identified as the hydraulic conductivity of the fissured system and the transmissivity of the conduit system. The model has been validated and can be employed for predictive purposes, i.e. without any further corrections and calibration, the model could reproduce time variant fluctuations of delta180, a nonpoint-source tracer as well as the breakthrough of a point source tracer.
