Next Article in Journal
A Weight of Evidence (WOE) Approach to Assess Environmental Hazard of Marine Sediments from Adriatic Offshore Platform Area
Previous Article in Journal
Editorial—Modelling of Floods in Urban Areas
Article

Uncertainty Estimation in Hydrogeological Forecasting with Neural Networks: Impact of Spatial Distribution of Rainfalls and Random Initialization of the Model

1
HydroSciences Montpellier, Univ. Montpellier, IMT Mines Ales, IRD, CNRS, 30100 Ales, France
2
AQUASYS, 2 Rue de Nantes, 44710 Port-Saint-Père, France
*
Authors to whom correspondence should be addressed.
Academic Editor: Renato Morbidelli
Water 2021, 13(12), 1690; https://doi.org/10.3390/w13121690
Received: 28 April 2021 / Revised: 7 June 2021 / Accepted: 14 June 2021 / Published: 18 June 2021
(This article belongs to the Section Hydrology)
Neural networks are used to forecast hydrogeological risks, such as droughts and floods. However, uncertainties generated by these models are difficult to assess, possibly leading to a low use of these solutions by water managers. These uncertainties are the result of three sources: input data, model architecture and parameters and their initialization. The aim of the study is, first, to calibrate a model to predict Champagne chalk groundwater level at Vailly (Grand-Est, France), and, second, to estimate related uncertainties, linked both to the spatial distribution of rainfalls and to the parameter initialization. The parameter uncertainties are assessed following a previous methodology, using nine mixed probability density functions (pdf), thus creating models of correctness. Spatial distribution of rainfall uncertainty is generated by swapping three rainfall inputs and then observing dispersion of 27 model outputs. This uncertainty is incorporated into models of correctness. We show that, in this case study, an ensemble model of 40 different initializations is sufficient to estimate parameter uncertainty while preserving quality. Logistic, Gumbel and Raised Cosine laws fit the distribution of increasing and decreasing groundwater levels well, which then allows the establishment of models of correctness. These models of correctness provide a confidence interval associated with the forecasts, with an arbitrary degree of confidence chosen by the user. These methodologies have proved to have significant advantages: the rigorous design of the neural network model has allowed the realisation of models able to generalize outside of the range of the data used for training. Furthermore, it is possible to flexibly choose the confidence index according to the hydrological configuration (e.g., recession or rising water table). View Full-Text
Keywords: neural networks; uncertainty; hydrogeology; probability; probability density function; model; rainfall neural networks; uncertainty; hydrogeology; probability; probability density function; model; rainfall
Show Figures

Figure 1

MDPI and ACS Style

Akil, N.; Artigue, G.; Savary, M.; Johannet, A.; Vinches, M. Uncertainty Estimation in Hydrogeological Forecasting with Neural Networks: Impact of Spatial Distribution of Rainfalls and Random Initialization of the Model. Water 2021, 13, 1690. https://doi.org/10.3390/w13121690

AMA Style

Akil N, Artigue G, Savary M, Johannet A, Vinches M. Uncertainty Estimation in Hydrogeological Forecasting with Neural Networks: Impact of Spatial Distribution of Rainfalls and Random Initialization of the Model. Water. 2021; 13(12):1690. https://doi.org/10.3390/w13121690

Chicago/Turabian Style

Akil, Nicolas, Guillaume Artigue, Michaël Savary, Anne Johannet, and Marc Vinches. 2021. "Uncertainty Estimation in Hydrogeological Forecasting with Neural Networks: Impact of Spatial Distribution of Rainfalls and Random Initialization of the Model" Water 13, no. 12: 1690. https://doi.org/10.3390/w13121690

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop