Hydrograph estimation with fuzzy chain model

Guclu Y. S., SEN Z.

JOURNAL OF HYDROLOGY, vol.538, pp.587-597, 2016 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 538
  • Publication Date: 2016
  • Doi Number: 10.1016/j.jhydrol.2016.04.057
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.587-597
  • Keywords: Fuzzy, Hydrograph, Peak discharge, SCS, Snyder, Clark, RAINFALL, TIME
  • Istanbul Technical University Affiliated: Yes


Hydrograph peak discharge estimation is gaining more significance with unprecedented urbanization developments. Most of the existing models do not yield reliable peak discharge estimations for small basins although they provide acceptable results for medium and large ones. In this study, fuzzy chain model (FCM) is suggested by considering the necessary adjustments based on some measurements over a small basin, Ayamama basin, within Istanbul City, Turkey. FCM is based on Mamdani and the Adaptive Neuro Fuzzy Inference Systems (ANFIS) methodologies, which yield peak discharge estimation. The suggested model is compared with two well-known approaches, namely, Soil Conservation Service (SCS)Snyder and SCS-Clark methodologies. In all the methods, the hydrographs are obtained through the use of dimensionless unit hydrograph concept. After the necessary modeling, computation, verification and adaptation stages comparatively better hydrographs are obtained by FCM. The mean square error for the FCM is many folds smaller than the other methodologies, which proves outperformance of the suggested methodology. (C) 2016 Elsevier B.V. All rights reserved.