+ - Volume 4 ‎(2018)
+ - Volume 3 ‎(2015)
+ - Volume 2 ‎(2014)
+ - Volume 1 ‎(2013)

Application of GLUE Methodology for Estimating the Parameters of the Rainfall-Runoff Model
Journal: Journal of River Engineering
Issue: Journal of River Engineering (Volume: 2, Issue: 8)
Author: Dorsa Darikandeh , Abolfazl Akbarpour , Mohsen Pourreza Bilondi , Seyyed Reza Hashemi
Keywords : Distributed , GLUE , Uncertainty analysis , AFFDEF

Parameter uncertainty of the distributed AFFDEF model was studied using‌ GLUE‌ (Generalized Likelihood Uncertainty Estimation) methodology which linked to the model in MATLAB software. Two historical floods with hourly time step for calibration periods were selected. Simulation runs were done to indicate convergence to a stationary posterior distribution. Finally posterior parameters were created with Latin Hyper Cubic Sampling. It was identified that some observations were not covered in the parameter uncertainty bounds, which suggest that more accurate predictions might be obtained if the model structure and the measured data are improved. ‌The results for calibration period showed that observational discharge values especially peak values bracketed well within 95% confidence interval. Regarding rising and recessing limb as a result of initial conditions and uncertainties originate from base flow separation methods, they were predicted out of the confidence interval. For evaluating the uncertainty degree, P-95CI, ARIL and MNS criteria were used. Also for determining the best-fitting between the observed discharge and simulated discharge, SSE criterion was used.

Article File : [Download (526.9 kB)] ‎1,084 downloads so far
Mahdi Moharrampour
Prof. Abdolkarim Behnia