FLOW-TRANSPORT AND NITRATE-TRANSPORT MODE LING IN THE BERNER-SEELAND,SWITZERLAND

Citation
D. Biaggi et al., FLOW-TRANSPORT AND NITRATE-TRANSPORT MODE LING IN THE BERNER-SEELAND,SWITZERLAND, Eclogae Geologicae Helvetiae, 87(2), 1994, pp. 429-438
Citations number
9
Categorie Soggetti
Geology
ISSN journal
00129402
Volume
87
Issue
2
Year of publication
1994
Pages
429 - 438
Database
ISI
SICI code
0012-9402(1994)87:2<429:FANMLI>2.0.ZU;2-B
Abstract
The Bernese Seeland is one of the major gravel aquifers in the Alpine foreland basin. Due to quantify and quality problems the groundwater m anagement authorities requested the conceptualization and implementati on of a regional hydrogeological model. The surface area of the aquife r is 71 km(2) with a dense observation network: 140 observation wells (18 of them are instrumented with limnigraphs), 106 sites for measurin g the surface elevation of rivers and channels, 55 locations for disch arge measurements. In this area, hydraulic heads and infiltration rate s were measured during 24-28 September, 1992 and 22-23 May, 1993. Appr oximately steady-state conditions were maintained by keeping the disch arge of the river Aare constant. The flow model has been calibrated us ing steady-state data. A maximum likelihood approach is used to estima te model parameters. Model parameters (i.e. hydraulic conductivity) ar e obtained by automatic estimation based on the model response (i.e. h eads, in- and exfiltration rates) and prior information. Prior informa tion consists mainly of hydraulic conductivities from flowmeter measur ements and the groundwater recharge calculated from SMA climate data. The steady-state calibration results in 41 hydraulic conductivity zone s, which reproduce sufficiently the head measurements as well as infil tration rates within an error limit. In addition, the interpretation o f measured flow rate data (e.g. river infiltration) improved the ident ifiability of the conductivity set significantly. The calibrated flow model is applied to modeling the nitrate concentrations in the aquifer . Dispersivities are estimated from field tracer tests and from interp reting flowmeter data. During a first step it is assumed that the nitr ates behave like conservative tracers without sorption, decay or any c hemical reactions (i.e. redox reactions from nitrate to nitrite). The comparison of measured and calculated concentrations results in a good agreement and the regional concentration distribution is reproduced s atisfactorily. The flow and transport models are used both to forecast the aquifer's response to future operations (production and injection ) and to quantify the regional nitrate flux (agricultural activity) in to the groundwater.