Performance and complementarity of two systemic models (reservoir and neural networks) used to simulate spring discharge and piezometry for a karst aquifer

Line Kong a Siou (1) , Perrine Fleury (2) , A. Johannet (1) , Valérie Borrell (3) , S. Pistre (3) , Nathalie Dörfliger (2)

1 LGEI - Laboratoire de Génie de l'Environnement Industriel
2 BRGM - Bureau de Recherches Géologiques et Minières
3 HSM - Hydrosciences Montpellier

Karst aquifers can provide previously untapped freshwaterresourcesand have thus generated considerable interest among stakeholders involved in thewater supplysector.Here we compare the capacity of two systemic models to simulate the discharge and piezometry of a karst aquifer.Systemic models have the advantage of allowing the study of heterogeneous, complex karst systems without relying on extensive geographical and meteorological datasets.The effectiveness and complementarityof the two models are evaluated for a range of hydrologic conditions and for three methods to estimate evapotranspiration (Monteith, a priori ET, and effective rainfall).The first model is a reservoir model (referred to as VENSIM, after the software used), which is designed with just one reservoir so as to be as parsimonious as possible. The second model is a neural network (NN) model. The models are designed to simulate the rainfall-runoff and rainfall-water level relations in a karst conduit.The Lezaquifer, a karst aquifer located near the city of Montpellierin southern France and a critical water resource, was chosen tocompare the two models. Simulated discharge and water level were compared after completing model design and calibration.The results suggest that the NN model is more effective at incorporating the nonlinearity of the karst spring for extreme events (extreme low and high water levels), whereas VENSIM provides a better representation of intermediate-amplitude water level fluctuations.VENSIM is sensitive to the method used to estimate evapotranspiration, whereas the NN model is not. Given that the NN model performs better for extreme events, it is better for operational applications (predicting floods or determiningwater pumping height). VENSIM, on the other hand, seems more appropriate for representing the hydrologic state of the basin during intermediate periods, when severaleffectsare at work: rain, evapotranspiration, developmentofvegetation, etc.A proposal forimprovingboth models is also provided.

Mots clés: Water management; Modeling; Karst; Neural networks; Reservoir model

Domaines: Hydrologie

Line Kong a Siou, Perrine Fleury, A. Johannet, Valérie Borrell, S. Pistre, et al.. Performance and complementarity of two systemic models (reservoir and neural networks) used to simulate spring discharge and piezometry for a karst aquifer. Journal of Hydrology, 2014, pp.3178--3192. ⟨10.1016/j.jhydrol.2014.10.041⟩. ⟨hal-02047991⟩