2005 Salt Lake City Annual Meeting (October 16–19, 2005)

Paper No. 9
Presentation Time: 10:15 AM


BIRK, Steffen1, GEYER, Tobias2, LIEDL, Rudolf1 and SAUTER, Martin2, (1)Center for Applied Geoscience, Univ of Tübingen, Sigwartstr. 10, 72076 Tübingen, Germany, (2)Geoscience Center Göttingen, Univ of Göttingen, Goldschmidtstr. 3, 37077 Göttingen, Germany, martin.sauter@geo.uni-goettingen.de

Karst aquifers can be conceptualized as dual flow systems comprising localized and often turbulent flow in solution conduits and Darcian flow in the fissured porous rock. Conventional borehole methods, such as slug tests and pumping tests, appear to be appropriate to obtain estimates of hydraulic parameters of the fissured system. Boreholes in karst catchments, however, are typically rare, as the unsaturated zone is often very thick and thus drilling very costly. Moreover, borehole tests are only of limited use for the characterization of the conduit system, because boreholes are usually not directly connected to karst conduits. In contrast, surface karst features, such as sinkholes and karst springs, are likely to be connected to the conduit system. Thus, tracer testing between sinkhole and karst spring provides one promising method for the quantitative characterization of conduit properties. Even without injection of artificial tracers, water discharging from a karst spring carries an imprint of the spring catchment, which makes it possible to infer properties of both fissured system and conduit system from hydraulic and physico-chemical spring responses to recharge events. A framework is presented, which combines various characterization techniques, such that the different methods efficiently complement one another. The coupled continuum-pipe flow model CAVE represents a keystone within this concept. Depending on data type and data availability, the model can be used to infer information about aquifer properties by inverse modeling, e.g. of hydrograph recession or tracer transport, or by forward modeling, e.g., of spring responses to recharge. It is demonstrated that modeling results are likely to be ambiguous but can be constrained if additional information, such as independent estimates of fissured system properties from pumping test analysis, is available.