Artificial Neural Networks (ANNs) are being used increasingly to predict and forecast water resources variables. In this paper, the steps that should be followed in the development of such models are outlined. These include the choice of performance criteria, the division and pre-processing of the available data, the determination of appropriate model inputs and network architecture, optimisation of the connection weights (training) and model validation. The options available to modellers at each of these steps are discussed and the issues that should be considered are highlighted. A review of 43 papers dealing with the use of neural network models for the prediction and forecasting of water resources variables is undertaken in terms of the modelling process adopted. In all but two of the papers reviewed, feedforward networks are used. The vast majority of these networks are trained using the backpropagation algorithm. Issues in relation to the optimal division of the available data, data pre-processing and the choice of appropriate model inputs are seldom considered. In addition, the process of choosing appropriate stopping criteria and optimising network geometry and internal network parameters is generally described poorly or carried out inadequately. All of the above factors can result in non-optimal model performance and an inability to draw meaningful comparisons between different models. Future research efforts should be directed towards the development of guidelines which assist with the development of ANN models and the choice of when ANNs should be used in preference to alternative approaches, the assessment of methods for extracting the knowledge that is contained in the connection weights of trained ANNs and the incorporation of uncertainty into ANN models.

Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applications

In this two-part series, the writers investigate the role of artificial neural networks (ANNs) in hydrology. ANNs are gaining popularity, as is evidenced by the increasing number of papers on this ...

Artificial Neural Networks in Hydrology. I: Preliminary Concepts

This paper forms the second part of the series on application of artificial neural networks (ANNs) in hydrology. The role of ANNs in various branches of hydrology has been examined here. It is foun...

Artificial Neural Networks in Hydrology. II: Hydrologic Applications

This review considers the application of artificial neural networks (ANNs) to rainfall-runoff modelling and flood forecasting. This is an emerging field of research, character- ized by a wide variety of techniques, a diversity of geographical contexts, a general absence of intermodel comparisons, and inconsistent reporting of model skill. This article begins by outlining the basic principles of ANN modelling, common network architectures and training algorithms. The discussion then addresses related themes of the division and preprocessing of data for model calibration/validation; data standardization techniques; and methods of evaluating ANN model performance. A literature survey underlines the need for clear guidance in current modelling practice, as well as the comparison of ANN methods with more conven- tional statistical models. Accordingly, a template is proposed in order to assist the construction of future ANN rainfall-runoff models. Finally, it is suggested that research might focus on the extraction of hydrological 'rules' from ANN weights, and on the development of standard performance measures that penalize unnecessary model complexity.

Hydrological modelling using artificial neural networks

Daily reservoir inflow forecasting using artificial neural networks with stopped training approach

Spatially distributed rainfall patterns can now be detected using a variety of remote–sensing techniques ranging from weather radar to various satellite–based sensors. Conversion of the remote–sensed signal into rainfall rates, and hence into runoff for a given river basin, is a complex and difficult process using traditional approaches. Neural–network models hold the possibility of circumventing these difficulties by training the network to map rainfall patterns into various measures of runoff that may be of interest. To investigate the potential of this approach, a very simple 5 × 5 grid cell synthetic watershed is used to generate runoff from stochastically generated rainfall patterns. A back–propagation neural network is trained to predict the peak discharge and the time of peak resulting from a single rainfall pattern. Additionally, the neural network is trained to map a time series of three rainfall patterns into a continuum of discharges over future time by using a discrete Fourier series fit to the runoff hydrograph.

Neural-Network Models of Rainfall-Runoff Process

In this investigation the practicality of applying a backpropagation neural network to modeling watershed response characteristics is examined. Two separate tests were performed. One test involved testing the ability of a neural network to predict time to peak and peak discharge resulting from unique storms produced with spatially distributed rainfall. The other test involved training a neural network to predict volumetric discharge from a time series of rainfall.

Jason Smith

Papers

Neural-Network Models of Rainfall-Runoff Process

Streamflow Forecasting Using Trainable Neural Networks