Hydrology and Earth System Sciences www.hydrol-earth-syst-sci.net 1027-5606 1607-7938 11 5 2007 10.5194/hess-11-1563-2007 http://www.hydrol-earth-syst-sci.net/11/1563/2007/ http://www.hydrol-earth-syst-sci.net/11/1563/2007/hess-11-1563-2007.html http://www.hydrol-earth-syst-sci.net/11/1563/2007/hess-11-1563-2007.pdf 1563 1579 2007-09-20 Neural network modelling of non-linear hydrological relationships R. J. Abrahart bob.abrahart@nottingham.ac.uk L. M. See School of Geography, University of Nottingham, Nottingham NG7 2RD, UK School of Geography, University of Leeds, Leeds, LS2 9JT, UK Two recent studies have suggested that neural network modelling offers no worthwhile improvements in comparison to the application of weighted linear transfer functions for capturing the non-linear nature of hydrological relationships. The potential of an artificial neural network to perform simple non-linear hydrological transformations under controlled conditions is examined in this paper. Eight neural network models were developed: four full or partial emulations of a recognised non-linear hydrological rainfall-runoff model; four solutions developed on an identical set of inputs and a calculated runoff coefficient output. The use of different input combinations enabled the competencies of solutions developed on a reduced number of parameters to be assessed. The selected hydrological model had a limited number of inputs and contained no temporal component. The modelling process was based on a set of random inputs that had a uniform distribution and spanned a modest range of possibilities. The initial cloning operations permitted a direct comparison to be performed with the equation-based relationship. It also provided more general information about the power of a neural network to replicate mathematical equations and model modest non-linear relationships. The second group of experiments explored a different relationship that is of hydrological interest; the target surface contained a stronger set of non-linear properties and was more challenging. Linear modelling comparisons were performed against traditional least squares multiple linear regression solutions developed on identical datasets. The reported results demonstrate that neural networks are capable of modelling non-linear hydrological processes and are therefore appropriate tools for hydrological modelling. Abrahart, R. J.: Neural networks and the problem of accumulated error: an embedded solution that offers new opportunities for modelling and testing, in: Hydroinformatics'98: Proceedings Third International Conference on Hydroinformatics, edited by: Babovic, V. and Larsen, C. L., Copenhagen, Denmark, 24–26 August 1998, A.A. Balkema Publishers, Rotterdam, The Netherlands, 2, 725–731, 1998. Abrahart, R. J., Heppenstall, A. J., and See, L. M.: Timing error correction procedure applied to neural network rainfall-runoff modelling, Hydrol. Sci. J., 52(3), 414–431, 2007. Abrahart, R. J., Kneale, P. E. and See, L. M. (Eds.): Neural Networks for Hydrological Modelling, A. A. Balkema Publishers, Rotterdam, The Netherlands, 2004. Abrahart, R. J. and See, L. M.: Neural network emulation of a rainfall-runoff model, Hydrol. Earth Syst. Sci. Discuss., 4, 287–327, 2007. Abrahart, R. J., See, L., and Kneale, P. E.: Investigating the role of saliency analysis with a neural network rainfall-runoff model, Computers and Geosciences, 27(8), 921–928, 2001. Abrahart, R. J. and White, S.: Modelling Sediment Transfer in Malawi: Comparing Backpropagation Neural Network Solutions Against a Multiple Linear Regression Benchmark Using Small Data Sets, Phys. Chem. Earth (B), 26(1), 19–24, 2001. Alippi, C.: Selecting Accurate, Robust, and Minimal Feedforward Neural Networks, IEEE Transactions on Circuits and Systems – I: Fundamental Theory and Applications, 49(12), 1799–1810, 2002. Adams, R., Dunn, S. M., Lunn, R., Mackay, R., and O'Callaghan, J. R.: Assessing the Performance of the NELUP Hydrological Models for River Basin Planning, J. Environ. Planning Manage., 38(1), 53–76, 1995. ASCE: Artificial Neural Networks in Hydrology. I. Preliminary Concepts, J. Hydrol. Eng., 5(2), 115–123, 2000a. ASCE: Artificial Neural Networks in Hydrology. II. Hydrologic Applications, J. Hydrol. Eng., 5(2), 124–137, 2000b. Bhattacharya, B., Price, R. K., and Solomatine, D. P.: Data-driven modelling in the context of sediment transport, Phys. Chem. Earth, 30(4–5), 297–302, 2005. Beven, K. J.: Rainfall-Runoff Modelling: The Primer, John Wiley & Sons Ltd., Chichester, England, 2001. Birikundavy, S., Labib, R., Trung, H. T., and Rousselle, J.: Performance of Neural Networks in Daily Streamflow Forecasting, J. Hydrol. Eng., 7(5), 392–398, 2002. Campolo, M., Soldati, A., and Andreussi, P.: Artificial neural network approach to flood forecasting in the River Arno, Hydrol. Sci. J., 48(3), 381–398, 2003. Chen, X., Chen, Y. D., and Xu, C.: A distributed monthly hydrological model for integrating spatial variations of basin topography and rainfall, Hydrol. Processes, 21(2), 242–252, 2007. Cheng, C. T., Ou, C. P., and Chau, K. W.: Combining a fuzzy optimal model with a genetic algorithm to solve multi-objective rainfall-runoff model calibration, J. Hydrol., 268(1–4), 72–86, 2002. Cheng, C.-T., Zhao, M.-Y., Chau, K. W., and Wu, X.-Y.: Using genetic algorithm and TOPSIS for Xinanjiang model calibration with a single procedure, J. Hydrol., 316(1–4), 129–140, 2006. Curry, B. and Morgan, P. H.: Neural networks, linear functions and neglected non-linearity, Computational Management Science, 1, 15–29, 2003. Dawson, C. W., Abrahart, R. J., and See, L. M.: HydroTest: a web-based toolbox of evaluation metrics for the standardised assessment of hydrological forecasts, Environ. Modell. Software, 22(4), 1034–1052, 2007. Dawson, C. W. and Wilby, R. L.: Hydrological modelling using artificial neural networks, Progress in Physical Geography, 25(1), 80–108, 2001. Dümenil, L. and Todini, E.: A rainfall-runoff scheme for use in the Hamburg climate model, in: Advances in Theoretical Hydrology – A tribute to James Dooge, edited by: O'Kane, J. P., Elsevier, Amsterdam, The Netherlands, 129-157, 1992. Gaume, E. and Gosset, R.: Over-parameterisation, a major obstacle to the use of artificial neural networks in hydrology?, Hydrol. Earth Syst. Sci., 7(5), 693–706, 2003. Govindaraju, R. S. and Rao, A. R. (Eds.): Artificial Neural Networks in Hydrology, Kluwer Academic Publishers, Dordrecht, The Netherlands, 2000. Han, D., Kwong, T., and Li, S.: Uncertainties in real-time flood forecasting with neural networks, Hydrol. Processes, 21(2), 223–228, 2007. Hettiarachchi, P., Hall, M. J., and Minns, A. W.: The extrapolation of artificial neural networks for the modelling of rainfall-runoff relationships, J. Hydroinformatics, 7(4), 291–296, 2005. Hillel, D.: Modeling in soil physics: A critical review. In: Future Developments in Soil Science Research, Soil Society of America, Madison, Wisconsin, USA, 35–42, 1986. Hu, C., Guo, S., Xiong, L., and Dingzhi Peng, D.: A modified Xinanjiang model and its application in northern China, Nordic Hydrology, 36(2), 175–192, 2005. Hsu, C. Y., Lee, H. Y., and Lin, Y. L.: Estimation of sediment yields in a watershed area using ANN method, in: River Basin Management II (Transactions of the Wessex Institute, Ecology and the Environment Vol. 60, Progress in Water Resources Vol. 7), edited by: Brebbia, C. A., WIT Press, Southampton, UK, 461–470, 2003. Huang, W., Xu, B., and Chad-Hilton, A.: Forecasting flows in Apalachicola River using neural networks, Hydrol. Processes, 18(13), 2545–2564, 2004. Jayawardena, A. W. and Zhou, M. C.: A modified spatial soil moisture storage capacity distribution curve for the Xinanjiang Model, J. Hydrol., 227(1–4), 93–113, 2000. Kim, B., Choi, W., and Kim, H.: Using neural networks with a linear output neuron to model plasmaetch processes, Proceedings IEEE International Symposium on Industrial Electronics, 1, 441–445, 2001, doi:10.1109/ISIE.2001.931830 Kisi, O.: Multi-layer perceptrons with Levenberg-Marquardt training algorithm for suspended sediment concentration prediction and estimation, Hydrol. Sci. J., 49(3), 1025–1040, 2004. Kisi, O.: Suspended sediment estimation using neuro-fuzzy and neural network approaches, Hydrol. Sci. J., 50(4), 683–696, 2005. Lin, C.A., Wen, L., Lu, G., Wu, Z., Zhang, J., Yang, Y., Zhu, Y., and Tong, L.: Atmospheric-hydrological modeling of severe precipitation and floods in the Huaihe River Basin, China, J. Hydrol., 330(1–2), 249–259, 2006. Liong, S. Y. and Chan, W. T.: Runoff Volume Estimates with Neural Networks, in: Neural Networks and Combinatorial Optimization in Civil and Structural Engineering, edited by: Topping, B. H. V. and Khan, A. I., Proceedings Third International Conference on the Application of Artificial Intelligence to Civil and Structural Engineering, Edinburgh, UK, 17–19 August 1993, Civil-Comp Press, Stirling, UK, 67–70, 1993. Loke, E., Warnaars, E. A., Jacobsen, P., Nelen, F., and Almeida, M. D.: Artificial neural networks as a tool in urban storm drainage, Water Sci. Technol., 36(8–9), 101–109, 1997. Maier, H. R. and Dandy, G. C.: Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applications, Environ. Modell. Software, 15(1), 101–124, 2000. Moore, R. J.: The probability-distributed principle and runoff production at point and basin scales, Hydrol. Sci. J., 30(2), 273–297, 1985. Moore, R. J.: The PDM rainfall-runoff model, Hydrol. Earth Syst. Sci., 11(1), 483–499, 2007. Mulvaney, T. J.: On the use of self-registering rain and flood gauges in making observations of the relations of rainfall and flood discharges in a given catchment, Proceedings of the Institution of Civil Engineers in Ireland, 4, 19–31, 1851. Nagy, H. M., Watanabe, K., and Hirano, M.: Prediction of Sediment Load Concentration in Rivers using Artificial Neural Network Model, J. Hydraul. Eng., 128(3), 588–595, 2002. Rao, Z. and Alvarruiz, F.: Use of an artificial neural network to capture the domain knowledge of a conventional hydraulic simulation model, J. Hydroinformatics, 9(1), 15–24, 2007. Riad, S., Mania, J., Bouchaou, L., and Najjar, Y.: Predicting catchment flow in a semi-arid region via an artificial neural network technique, Hydrol. Processes, 18(13), 2387–2393, 2004. Rogers, L. L. and Dowla, F. U.: Optimization of groundwater remediation using artificial neural networks with parallel solute transport modeling, Water Resour. Res., 30(2), 457–481, 1994. Sarle, W. S. (Ed.): Neural Network FAQ, Part 3 of 7: Generalisation [periodic posting to the Usenet newsgroup comp.ai.neural-nets]. (ftp://ftp.sas.com/pub/neural/FAQ.html) 1997 [Last accessed: 8 November 2006] Senbeta, D. A., Shamseldin, A. Y., and O'Connor, K. M.: Modification of the probability-distributed interacting storage capacity model, J. Hydrol., 224(3), 149–168, 1999. Senthil Kumar, A. R., Sudheer, K. P., Jain, S. K., and Agarwal, P. K.: Rainfall-runoff modelling using artificial neural networks: comparison of network types, Hydrol. Processes, 19(3), 1277–1291, 2005. Shamseldin, A. Y.: Artificial neural network model for flood forecasting in a developing country, J. Hydroinformatics, in press, 2007. Singh, A. K., Deo, M. C., and Kumar, V. S.: Prediction of littoral drift with artificial neural networks, Hydrol. Earth Syst. Sci. Discuss., 4(4), 2497–2519, 2007. Solomatine, D. P. and Avila Torres, L. A.: Neural network approximation of a hydrodynamics model in optimizing reservoir operation, in: Hydroinformatics 96: Proceedings Second International Conference on Hydroinformatics, edited by: Müller, A., Zurich, Switzerland, 9–13 September 1996, A.A. Balkema Publishers, Rotterdam, The Netherlands, 1, 201–206, 1996. Su, B., Kazama, S., Lu, M., and Sawamoto, M.: Development of a distributed hydrological model and its application to soil erosion simulation in a forested catchment during storm period, Hydrol. Processes, 17(14), 2811–2823, 2003. Todini, E.: The ARNO rainfall-runoff model, J. Hydrol., 175(1–4), 339–382, 1996. Todini, E.: The ARNO model, in: Mathematical Models of Large Watershed Hydrology, edited by: Singh, V. P. and Frevert, D. K., Water Resources Publications LLC, Highlands Ranch, CO, USA, 687–716, 2002. Toth, E., Brath, A., and Montanari, A.: Comparison of short-term rainfall prediction models for real-time flood forecasting, J. Hydrol., 239(1–4), 132–147, 2000. Varoonchotikul, P.: Flood forecasting using artificial neural networks, A.A. Balkema Publishers, Rotterdam, The Netherlands, 2003. Wilby, R. L., Abrahart, R. J., and Dawson, C. W.: Detection of conceptual model rainfall-runoff processes inside an artificial neural network, Hydrol. Sci. J., 48(2), 163–181, 2003. Wood, E. F., Lettenmaier, D. P., and Zartarian, V. G.: A land-surface hydrology parameterisation with subgrid variablility for general circulation models, J. Geophys. Res., 97(D3), 2717–2728, 1992. Yitian, L. and Gu, R. R.: Modeling Flow and Sediment Transport in a River System Using an Artificial Neural Network, Environ. Manage., 31(1), 122–134, 2003. Zhang, Q., Smith, D. W., and Baxter, C. W.: Introduction, J. Environ. Eng. Sci., 3, S1, iii-iv, 2004. Zhao, R.-J.: The Xinanjiang model applied in China, J. Hydrol., 135(1–4), 371–381, 1992. Zhao, R.-J. and Liu, X.-R.: The Xinanjiang Model, in: Computer Models of Watershed Hydrology, edited by: Singh, V. P., Water Resources Publications, Fort Collins, USA, 215-232, 1995. Zhao, R.-J., Zhang, Y.-L., Fang, L.-R., Liu, X.-R., and Zhang, Q.-S.: The Xinanjiang model, In: Hydrological Forecasting: Proceedings of the Symposium on the Application of Recent Developments in Hydrological Forecasting to the Operation of Water Resource Systems, Oxford, April 1980, International Association of Hydrological Sciences Press, Wallingford, UK, IAHS Publication No. 129, 351–356, 1980.