Exchange between a river and groundwater, assessed with hydrochemical data

We describe the chemical composition of groundwater from an alluvial granular aquifer in a valley fill flood plain (River Thur Valley). The river flows along this valley and is mostly downwelling on its way, indirectly through an unsaturated zone in the upstream part, and directly through the water-saturated bed in the downstream part. River Thur has been channelized with barriers for more than a century. In 1992, the authorities started to restore a section of River Thur with riverbed enlargements. The land use in the flood plain and the seasonal and climatic conditions (e.g., hot dry summer 2003) result in alterations of the natural geochemical composition of the river water. This groundwater is partly to mainly recharged by bank filtration. Several wells exist near the river that draw groundwater for drinking. In some of these wells, the groundwater has a very short residence time in the subsurface of days to weeks. Bed enlargements and other operations for an enhancement of the exchange of water between the river and groundwater increase the contamination risk of the nearby wells. During bank filtration, the groundwater changes gradually its composition, with increasing distance from the river and with depth in the aquifer. From today's changes of the water quality during riverbank filtration, we tried to extrapolate to the groundwater quality that may arise from future river restorations. Today the groundwater body consists of a mixture of groundwater from the seepage of precipitation and from riverbank filtration. The main difference between river water and groundwater results from the microbial activity in riverbed and bank materials. This activity leads to a consumption of O<sub>2</sub> and to a higher partial pressure of CO<sub>2</sub> in the groundwater. Criteria for the distinction of different groundwater compositions are the distance of a well from the river and the subsurface residence time of the groundwater to reach this well.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.