EGU22-6534
https://doi.org/10.5194/egusphere-egu22-6534
EGU General Assembly 2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Influences of temporal variability on the calibration of a hydrologic model  

Tibebe Tigabu1, Paul Wagner1, Balaji Narasimhan2, and Nicola Fohrer1
Tibebe Tigabu et al.
  • 1Kiel University, Institute for Natural Resource Conservation, Hydrology and Water Resources Management, Kiel, Germany (ttigabu@hydrology.uni-kiel.de)
  • 2Indo-German Centre for Sustainability, Indian Institute of Technology Madras, Chennai 600036, India

Abstract: Parametrization is an important step to construct a reasonable hydrologic model for a catchment. However, selecting appropriate model parameters can be challenging, particularly in data scarce regions. Conventionally, hydrologic model parameters are selected through calibration assuming that catchment processes and model parameters are stationary over time. However, the assumption of stationarity may not be valid all the time due to temporal changes in the behaviors of catchments. Therefore, the purpose of this study is to investigate the influence of temporal variability on the SWAT model parametrization using different calibration periods. To this end, we calibrated the SWAT model based on daily and monthly streamflow data in the Adyar catchment, Chennai. Results showed that the SWAT model performance and parameter values differed when the calibration periods were shifted by one year. This is reflected in the KGE (Kling Gupta Efficiency) values that varied between 0.38 to 0.68 for calibration periods of 2004-2007,2005-2008, 2006-2009, 2007-2010, 2008-2011, 2009-2012 and 2010-2013. Likewise, the selection of values for sensitive model parameters varied even though the parameter values were chosen in the same ranges. Moreover, independent model evaluation for wet and dry years showed significantly different performance indices and model parameter values. The model efficiency of wet years (NSE = 0.59 and KGE= 0.68) was by far better than the model efficiency of dry years (NSE = -0.59 and KGE = 0.1). In general, this study provides a good insight into hydrologic model calibration under non-stationarity conditions.

How to cite: Tigabu, T., Wagner, P., Narasimhan, B., and Fohrer, N.: Influences of temporal variability on the calibration of a hydrologic model  , EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-6534, https://doi.org/10.5194/egusphere-egu22-6534, 2022.