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

Paper No. 3
Presentation Time: 8:35 AM

IDENTIFICATION OF THE TRANSPORT PROPERTIES OF A CHALK KARST AQUIFER IN WESTERN PARIS BASIN BY MEANS OF UNIVARIATE CLUSTERING


ABSTRACT WITHDRAWN

, matthieu.fournier@etu.univ-rouen.fr

The karst hydrologist has at his disposal a set of methods to explore and study karst aquifers. Among these methods, there are the hydrodynamic methods like spring hydrographs, sorted discharge distribution of the annual hydrograph and time series analyses. Hydrodynamic methods are often used on the annual hydrograph of electrical conductivity, water discharge and even turbidity. Time series analyses have proven to be useful in improving understanding of karst systems. However with time series analysis, we lose the time evolution which have with hydrographs. Moreover, we cannot compare the variations between electrical conductivity, turbidity and water discharge which allow to identify the transport properties within the karst system (direct transfer of ground water or surface water, deposition of suspended matter, resuspension of intrakarstic sediments). To answer these problems, we propose the use of univariate clustering with Fisher algorithm to optimally partition observations in homogeneous clusters based on their description using a single quantitative variable. Applied to electrical conductivity, turbidity and water discharge, this method allow to define water types (weak, mean or strong values for each one of three variables). The comparison of their time variations allows to identify the transport properties and the periods where they occur. We used this method to a spring of a chalk karst aquifer located in western Paris basin (France) during the 2002 hydrologic cycle. The results show that the annual volume of spring water is decomposed of 47.5% of ground water, 23.5% of flood water by direct transfer of surface water, 14.4% of spring water results from deposition of suspended matter and 14.6% of spring water results from resuspension of intrakarstic sediments. The released of ground water occurs during low water and the released of flood water occurs during recharge and storm events in summer time. The released of spring water resulting from deposition occurs during weak rain event intensity from March at June. The released of spring water resulting from resuspension occurs from April to September. To conclude, this method applied to electrical conductivity, turbidity and water discharge, allows to identify the transport properties on the annual hydrologic cycle and their occurence period.