HR: 0800h
AN: H41E-0832    [Abstracts]
TI: GEOPHYSICS IN HYDROGEOLOGICAL INVERSE PROBLEM: HERO OR VILLAIN?
AU: * Alcolea, A
EM: andres.alcolea@unine.ch
AF: CHYN, CHYN. Centre for Hydrogeology of Neuchatel. rue Emile Argand, 11, Neuchatel, 2009, Switzerland
AU: Renard, P
EM: philippe.renard@unine.ch
AF: CHYN, CHYN. Centre for Hydrogeology of Neuchatel. rue Emile Argand, 11, Neuchatel, 2009, Switzerland
AU: Mariethoz, G
EM: gregoire.mariethoz@unine.ch
AF: CHYN, CHYN. Centre for Hydrogeology of Neuchatel. rue Emile Argand, 11, Neuchatel, 2009, Switzerland
AB: Geostatistical inverse problem is a powerful tool to aid decision-makers in aquifer management. During the last few years, existing inverse problem codes have been updated in order to accommodate "non- traditional" types of observations (i.e. heads or concentrations). The potential of exhaustive geophysical data has been shown to be well-suited for aquifer characterization. However, limited attention has been devoted to the use of this information in real field hydrogeological inverse problems. In this work, we present an application of inverse problem to the management of coastal aquifers including different types of data (heads, resistivities and prior information on transmissivity and storage coefficient). Spatial variability is characterized using the regularized pilot points method. The procedure is as follows. First, we obtain a characterization of the transmissivity and storage coefficient fields from calibration data. Second, this characterization is used to design a pumping network by means of a genetic algorithm. Several constraints apply, such as operational costs or environmental side effects. Three cases are presented, depending on the calibration data sets: (1) only resistivities (no calibration is performed), (2) heads and prior information of model parameters, and (3) all of them altogether (resistivities are used as external drift). Results show that, by themselves, resistivity or head data sets (and prior information) do not suffice to obtain a reliable characterization of the system. However, the consideration of all data at the same time leads to the best characterization of the system among the ones tested.
DE: 1816 Estimation and forecasting
DE: 1829 Groundwater hydrology
DE: 1835 Hydrogeophysics
DE: 1847 Modeling
DE: 1869 Stochastic hydrology
SC: Hydrology [H]
MN: 2007 Fall Meeting