Water Resources Management

, Volume 27, Issue 7, pp 2103–2124 | Cite as

Comparison of Artificial Neural Network Methods with L-moments for Estimating Flood Flow at Ungauged Sites: the Case of East Mediterranean River Basin, Turkey

  • Neslihan Seckin
  • Murat CobanerEmail author
  • Recep Yurtal
  • Tefaruk Haktanir


A regional flood frequency analysis based on the index flood method is applied using probability distributions commonly utilized for this purpose. The distribution parameters are calculated by the method of L-moments with the data of the annual flood peaks series recorded at gauging sections of 13 unregulated natural streams in the East Mediterranean River Basin in Turkey. The artificial neural networks (ANNs) models of (1) the multi-layer perceptrons (MLP) neural networks, (2) radial basis function based neural networks (RBNN), and (3) generalized regression neural networks (GRNN) are developed as alternatives to the L-moments method. Multiple-linear and multiple-nonlinear regression models (MLR and MNLR) are also used in the study. The L-moments analysis on these 13 annual flood peaks series indicates that the East Mediterranean River Basin is hydrologically homogeneous as a whole. Among the tried distributions which are the Generalized Logistic, Generalized Extreme Vaules, Generalized Normal, Pearson Type III, Wakeby, and Generalized Pareto, the Generalized Logistic and Generalized Extreme Values distributions pass the Z statistic goodness-of-fit test of the L-moments method for the East Mediterranean River Basin, the former performing yet better than the latter. Hence, as the outcome of the L-moments method applied by the Generalized Logistic distribution, two equations are developed to estimate flood peaks of any return periods for any un-gauged site in the study region. The ANNs, MLR and MNLR models are trained and tested using the data of these 13 gauged sites. The results show that the predicting performance of the MLP model is superior to the others. The application of the MLP model is performed by a special Matlab code, which yields logarithm of the flood peak, Ln(QT), versus a desired return period, T.


L-moments Regional flood frequency Artificial neural networks Ungauged catchments 


  1. Abolverdi J, Khalili D (2010) Development of regional rainfall annual maxima for Southwestern Iran by L-moments. Water Resour Manag 24:2501–2526CrossRefGoogle Scholar
  2. Acreman MC, Sinclair CD (1986) Classification of drainage basins according to their physical characteristics; an application for flood frequency analysis in Scotland. J Hydrol 84:365–380CrossRefGoogle Scholar
  3. Atiem IA, Harmancioglu N (2006) Assessment of regional floods using L-moments approach: the case of the River Nile. Water Resour Manag 20:723–747CrossRefGoogle Scholar
  4. Berthet HG (1994) Station-year approach: tool for estimation of design floods. J Water Resour Plan Manag ASCE 120(2):135–160CrossRefGoogle Scholar
  5. Bobee B, Rasmussen PF (1995) Recent advances in flood frequency analysis. U.S. National Report to International Union of Geodesy and Geophysics 1991–1994. Rev Geophys 33(supp):1111–1116Google Scholar
  6. Broomhead D, Lowe D (1988) Multivariable functional interpolation and adaptive networks. Complex Syst 2:321–355Google Scholar
  7. Burn DH (1990) Evaluation of regional flood frequency analysis with a region of influence approach. Water Resour Res 26(10):2257–2265CrossRefGoogle Scholar
  8. Chapra SC, Canale RP (2002) Numerical Methods for Engineers, 4th edn. McGraw-Hill, New YorkGoogle Scholar
  9. Cigizoglu HK, Kisi O (2005) Flow prediction by three back-propagation techniques using k-fold partitioning of neural network training data. Nord Hydrol 36(1):49–64Google Scholar
  10. Cressie N (1993) Statistics for spatial data, revised edition. Wiley Interscience, New YorkGoogle Scholar
  11. Cunnane C (1989) Distributions for Flood Frequency Analysis. WMO Operational Hydrology Report No.33. World Meteorological Organization, GenevaGoogle Scholar
  12. Cybenco G (1989) Approximation by superposition of a sigmoidal function. Math Control Signals Syst 2:303–314CrossRefGoogle Scholar
  13. Dalrymple T (1960) Flood frequency analyses. US Geological Survey Water Supply Paper no. 1543-A:11–51Google Scholar
  14. Dawson WC, Wilby R (1998) An artificial neural network approach to rainfall-runoff modelling. Hydrol Sci J 43(1):47–66CrossRefGoogle Scholar
  15. Dawson CW, Abrahart RJ, Shamseldin AY, Wilby RL (2006) Flood estimation at ungauged sites using artificial neural Networks. J Hydrol 319:391–409CrossRefGoogle Scholar
  16. Duch W, Jankovski N (1999) Survey of neural transfer functions. Neural Comput Surv 2:163–212Google Scholar
  17. El-Bakyr MY (2003) Feed forward neural networks modeling for K-P interactions. Chaos, Solitons Fractals 18(5):995–1000Google Scholar
  18. Ellouze M, Abida H (2008) Regional flood frequency analysis in Tunisia: identification of regional distributions. Water Resour Manag 22(8):943–957CrossRefGoogle Scholar
  19. French MN, Krajewski WF, Cuykendall RR (1992) Rainfall Forecasting in space and time using neural network. J Hydrol 137:1–31CrossRefGoogle Scholar
  20. Gahegan M, German G, West G (1999) Improving neural network performance on the classification of complex geographic datasets. Geogr Syst 1:3–22CrossRefGoogle Scholar
  21. Goh ATC (1995) Back-propagation neural networks for modeling complex systems. Artif Intell Eng 9(3):143–151Google Scholar
  22. Govindaraju RS, Rao AR (2000) Artificial Neural Networks in Hydrology. Kluwer Academy, Norwell, 329 pCrossRefGoogle Scholar
  23. Greenwood JA, Landwehr JM et al (1979) Probability weighted moments: definition and relation to parameters of several distributions expressible in inverse form. Water Resour Res 15(5):1049–1054CrossRefGoogle Scholar
  24. Grehys G (1996) Presentation and review of some methods for regional flood frequency analysis. J Hydrol 186:63–84CrossRefGoogle Scholar
  25. Hagan MT, Menhaj MB (1994) Training feed forward networks with the Marquardt algorithm. IEEE Trans Neural Netw 6:861–867Google Scholar
  26. Haykin S (1998) Neural Networks - A Comprehensive Foundation, 2nd edn. Prentice-Hall, Upper Saddle River, pp 26–32Google Scholar
  27. Hornik K, Stinchcombe M, White H (1989) Multilayer feed forward networks are universal approximators. Neural Netw 2:359–366CrossRefGoogle Scholar
  28. Hosking JRM (1986) The Theory of Probability Weighted Moments. Research Rep. RC 12210. IBM Research Division, Yorktown Heights, 160 ppGoogle Scholar
  29. Hosking JRM (1990) L-moments: analysis and estimation of distributions using lineer combinations of order statistics. J R Stat Soc 52(2):105–124Google Scholar
  30. Hosking JRM (1991) Approximations for use in Constructing L-moments Ratio Diagrams. Res Report, RC-16635, vol 3. IBM Res Division, New YorkGoogle Scholar
  31. Hosking JRM, Wallis JR (1993) Some statistics useful in regional frequency analysis. Water Resour Res 29(2):271–281CrossRefGoogle Scholar
  32. Hosking JRM, Wallis JR (1997) Regional Frequency Analysis - an Approach based on L-moments. Cambridge University Pres, New YorkCrossRefGoogle Scholar
  33. Hussain Z, Pasha GR (2009) Regional flood frequency analysis of the seven sites of Punjab, Pakistan, using L-moments. Water Resour Manag 23:1917–1933CrossRefGoogle Scholar
  34. Jaiswal RK, Goel NK, Singh P, Thomas T (2003) L-moment based flood frequency modelling. Inst Eng (India) 84:6–10Google Scholar
  35. Jingyi Z, Hall MJ (2004) Regional flood frequency analysis for the Gan-Ming River basin in China. J Hydrol 296:98–117CrossRefGoogle Scholar
  36. Kanellopoulos I, Wilkinson GG (1997) Strategies and best practice for neural network image classification. Int J Remote Sens 18(4):711–725CrossRefGoogle Scholar
  37. Karim MDA, Chowdhury JU (1995) A comparison of four distributions used in flood frequency analysis in Bangladesh. Hydrol Sci J 40(1):55–66CrossRefGoogle Scholar
  38. Kavzoğlu T (2001) An investigation of the design and use of feed-forward artificial neural networks in the classification of remotely sensed images, Ph.D. dissertation, University of Nottingham, Nottingham, United Kingdom, 308 pGoogle Scholar
  39. Kim B, Kim S, Kim K (2003) Modelling of plasma etching using a generalized regression neural network. Vacuum 71(4):497–503Google Scholar
  40. Kjeldsen TR, Smithers JC, Schulze RE (2002) Regional flood frequency analysis in the KwaZulu-Natal province, South Africa, using the index-flood method. J Hydrol 255:194–211CrossRefGoogle Scholar
  41. Kumar R, Chatterjee C, Panigrihy N, Patwary BC, Singh RD (2003) Development of regional flood formulae using L-moments for gauged and ungauged catchments of North Brahmaputra River system. Inst Eng (India) 84:57–63Google Scholar
  42. Lee GC, Chang SH (2003) Radial basis function networks applied to DNBR calculation in digital core protection systems. Ann Nucl Energ 30:1561–1572CrossRefGoogle Scholar
  43. Madsen H, Pearson CP, Rosbjerg D (1997) Comparison of annual maximum series and partial duration series mthods for modeling extreme hydrologic events 2. Regional modeling. Water Resour Res 33(4):759–769CrossRefGoogle Scholar
  44. Maier HR (1995) A review of artificial neural networks. Research Report No. R131. School of Civil and Environmental Engineering, The University of Adelaide, South AustraliaGoogle Scholar
  45. Maier HR, Dandy GC (2000) Neural networks for the prediction and forecasting water resources variables: A review of modeling issues and applications. Environ Modell Softw 15:101–124Google Scholar
  46. Meigh JR, Farquharson FAK, Sutcliffe JV (1997) A worldwide comparison of regional flood estimation methods and climate. Hydrol Sci J 42(2):225–244CrossRefGoogle Scholar
  47. Minns AW, Hall MJ (1996) Artificial neural networks as rainfall-runoff models. Hydrol Sci J 41(3):399–417CrossRefGoogle Scholar
  48. Mkhandi S, Kachroo S (1997) Regional flood frequency analysis for Southern Africa, Southern African FRIEND, Technical documents in Hydrology No.15, Unesco, ParisGoogle Scholar
  49. Noto LV, Loggia G (2009) Use of L-moments approach for regional flood frequency analysis in Sicily, Italy. Water Resour Manag 23:2207–2229CrossRefGoogle Scholar
  50. Ouarda TBMJ, Ba KM, Diaz-Delgado C, Carsteanu A, Chokmani K, Gingras H, Quentin E, Trujillo E, Bobee B (2008) Intercomparison of regional flood frequency estimation methods at ungauged sites for a Mexican case study. J Hydrol 348:40–58CrossRefGoogle Scholar
  51. Paola JD (1994) Neural network classification of multispectral imagery, MSc. dissertation, The University of Arizona, Tucson, Arizona, 169 pGoogle Scholar
  52. Parida BP, Kachroo RK, Shrestha DB (1998) Regional flood frequency analysis of Mahi-Sabarmati Basin (Subzone 3-a) using index flood procedure with L-moments. Water Resour Manag 12:1–12CrossRefGoogle Scholar
  53. Rao AR, Hamed KH (1997) Regional frequency analysis of Wabash River flood data by L-moments. J Hydrol Eng 2(4):169–179Google Scholar
  54. Saf B (2009) Regional flood frequency analysis using L-moments for the West Mediterranean Region of Turkey. Water Resour Manag 23(3):531–551CrossRefGoogle Scholar
  55. Sankarasubramanian A, Srinivasan K (1999) Investigation and comparison of sampling properties of L-moments and conventional moments. J Hydrol 218:13–34CrossRefGoogle Scholar
  56. Shu C, Ouarda TBMJ (2008) Regional flood frequency analysis at ungauged sites using tha adaptive neuro-fuzzy inference system. J Hydrol 349:31–43CrossRefGoogle Scholar
  57. Specht DF (1991) A general regression neural network. IEEE Trans Neural Netw 2(6):568–576CrossRefGoogle Scholar
  58. State Water Works, SWW (1994) The annual maximum flow of Turkish Rivers. The General Directorate of State Water Works Publication, Ankara, TurkeyGoogle Scholar
  59. Stedinger JR, Tasker GD (1985) Regional hydrologic analysis 1. Ordinary, weighted and generalized least squares compared. Water Resour Res 21(9):1421–1432CrossRefGoogle Scholar
  60. Stedinger JR, Vogel RM, Georgiou EF (1993) Frequency Analysis of Extreme Events, Chapter 18. In: Maidment DJ (ed) Handbook of Hydrology. McGraw-Hill, NewYorkGoogle Scholar
  61. Tsoukalas LH, Uhrig RE (1997) Fuzzy and neural approaches in engineering. Wiley, NewYorkGoogle Scholar
  62. Vogel RM, Fennesy MN (1993) L moment diagrams should replace product moment diagrams. Water Resour Res 29(6):1745–1752CrossRefGoogle Scholar
  63. Vogel RM, McMahon TA, Chiew FHS (1993a) Floodflow frequency model selection in Australia. J Hydrol 146:421–449CrossRefGoogle Scholar
  64. Vogel RM, Wilbert OT Jr, McMahon TA (1993b) Flood-flow frequency model selection in Southwestern United States. J Water Resour Plan Manag ASCE 119(3):353–366CrossRefGoogle Scholar
  65. Wiltshire SW (1986) Identification of homogeneous regions for flood frequency analysis. J Hydrol 84:287–302CrossRefGoogle Scholar
  66. Yue S, Wang CY (2004) Possible regional probability distribution type of Canadian annual streamflow by L-moments. Water Resour Manag 18:425–438CrossRefGoogle Scholar
  67. Zrinji Z, Burn DH (1994) Flood frequency analysis for ungauged sites using a region of influence approach. J Hydrol 153:1–21CrossRefGoogle Scholar
  68. Zrinji Z, Burn DH (1996) Regional flood frequency with hierarchical region of influence. J Water Resour Plan Manag ASCE 122(4):245–252CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  • Neslihan Seckin
    • 1
  • Murat Cobaner
    • 2
    Email author
  • Recep Yurtal
    • 1
  • Tefaruk Haktanir
    • 2
  1. 1.Civil Engineering DepartmentCukurova UniversityAdanaTurkey
  2. 2.Civil Engineering DepartmentErciyes UniversityKayseriTurkey

Personalised recommendations