Hydrologic Data Exploration and River Flow Forecasting of a Humid Tropical River Basin Using Artificial Neural Networks

Abstract. Recently Feed-Forward Artificial Neural Networks (FNN) have been gaining popularity for stream flow forecasting. However, despite the promising results presented in recent papers, their use is questionable. In theory, their “universal approximator‿ property guarantees that, if a sufficient number of neurons is selected, good performance of the models for interpolation purposes can be achieved. But the choice of a more complex model does not ensure a better prediction. Models with many parameters have a high capacity to fit the noise and the particularities of the calibration dataset, at the cost of diminishing their generalisation capacity. In support of the principle of model parsimony, a model selection method based on the validation performance of the models, "traditionally" used in the context of conceptual rainfall-runoff modelling, was adapted to the choice of a FFN structure. This method was applied to two different case studies: river flow prediction based on knowledge of upstream flows, and rainfall-runoff modelling. The predictive powers of the neural networks selected are compared to the results obtained with a linear model and a conceptual model (GR4j). In both case studies, the method leads to the selection of neural network structures with a limited number of neurons in the hidden layer (two or three). Moreover, the validation results of the selected FNN and of the linear model are very close. The conceptual model, specifically dedicated to rainfall-runoff modelling, appears to outperform the other two approaches. These conclusions, drawn on specific case studies using a particular evaluation method, add to the debate on the usefulness of Artificial Neural Networks in hydrology. Keywords: forecasting; stream-flow; rainfall-runoff; Artificial Neural Networks

Download Full-text

Assessment of shallow landslide susceptibility using artificial neural networks in Jabonosa River Basin, Venezuela

Engineering Geology ◽

10.1016/j.enggeo.2004.10.004 ◽

2005 ◽

Vol 78 (1-2) ◽

pp. 11-27 ◽

Cited By ~ 255

Author(s):

H. Gómez ◽

T. Kavzoglu

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

River Basin ◽

Landslide Susceptibility ◽

Shallow Landslide ◽

Artificial Neural

Download Full-text

Large-scale Groundwater Simulation using Artificial Neural Networks in the Danube River Basin

10.5194/egusphere-egu2020-8212 ◽

2020 ◽

Author(s):

Illias Landros ◽

Ioannis Trichakis ◽

Emmanouil Varouchakis ◽

George P. Karatzas

Keyword(s):

Neural Networks ◽

Artificial Neural Networks ◽

River Basin ◽

Danube River ◽

Bayesian Regularization ◽

Data Set ◽

The Danube River ◽

Precipitation Temperature ◽

Artificial Neural ◽

Danube River Basin

<p>In recent years, Artificial Neural Networks (ANNs) have proven their merit in being able to simulate the changes in groundwater levels, using as inputs other parameters of the water budget, e.g. precipitation, temperature, etc.. In this study, ANNs have been used to simulate hydraulic head in a large number of wells throughout the Danube River Basin, taking as inputs, precipitation, temperature, and evapotranspiration data in the region. Different ANN architectures have been examined, to minimize the simulation error of the testing data-set. Among the different training algorithms, Levenberg-Marquardt and Bayesian Regularization are used to train the ANNs, while the different activation functions of the neurons that were deployed include tangent sigmoid, logarithmic sigmoid and linear. The initial application comprised of data from 128 wells between 1 January 2000 and 31 October 2014. The best performance was achieved by the algorithm Bayesian Regularization with a error of the order &#160;based on all observation wells. A second application, compared the results of the first one, with the results of an ANN used to simulate a single well. The pros and cons of the two approaches, and the synergies of using both of them is further discussed in order to distinguish the differences, and guide researchers in the field for further applications.</p>

Download Full-text