Performance Evaluation of Single and Multi-Objective Calibration in Rainfall-Runoff Modelling

Abstract

Hydrologic modelling is widely used as a tool for making decisions and predictions in planning and managing water resources. Successful application of a hydrologic model depends upon careful calibration and validation using appropriate performance measures in satisfying corresponding performance evaluation criteria. Performance of MIKE 11 NAM rainfall runoff model was assessed using four calibration objectives and their different combinations, as applied to upper Gin catchment, Sri Lanka. The four calibration objectives measured different aspects of hydrograph: good water balance, good overall agreement of the shape of the hydrograph, good agreement for peak flows, and good agreement for low flows. Their numerical performance was measured using four objective functions from which fifteen calibration schemes were formed (four single objective schemes and eleven multi-objective schemes). Using aggregated distance measure, equal weights were assigned to the four objective functions. Shuffled complex evolution algorithm was used to solve the multiple calibration objective problems. Model performance was evaluated using three criteria: Nash-Sutcliffe coefficient (NSE), percent bias (PBIAS) and coefficient of determination (R 2 ). Results revealed significant trade-offs between the objective functions, highlighting that no single calibration objective was able to depict all the aspects of the hydrograph simultaneously. However, multi-objective calibration yielded more accurate and consistent simulations covering different aspects of the hydrograph, simultaneously with overall best performance shown for combination of the four objective functions satisfying all the performance evaluation criteria (NSE=0.56, R 2 = 0.56, PBIAS=17%), compared to the single-objective calibration. Instead of using R 2 alone, use of the corresponding regression slope as a weighing factor of R 2 was recommended following further analysis of simulation results.

Temporal Transferability of Calibrated Hydrological Model Parameters: A Case Study of Gin Catchment, Sri Lanka

“…As recommended by Moriasi et al [26], the PBIAS values obtained in the calibration and verification runs fall within the acceptable range of ±25%. The RSR values for both the runs fall within the acceptable range [5], [6], [18], [31]. Since the optimized parameter values and performance measures fall within the acceptable ranges, MIKE 11 NAM model was able to reasonably simulate the rainfall-runoff generation process in the Gin catchment.…”

“…Hydrological Model This study applied the Nedbør-Afstrømings-Model (NAM), a lumped conceptual module built in MIKE 11, to simulate rainfall-runoff in a continuous mode. The model has been successfully applied by previous researchers for hydrological investigations in many catchments including the catchments in Sri Lanka: rainfall-runoff simulations [15], [16], [17], model performance evaluation studies [18], land use change impact assessments on flood formations [19], and runoff simulations in un-gauged catchments [5], [6].…”

“…Auto calibration was set under the two objective functions: overall volume error [F1(∅)] [Eq. ( 4)] and overall root mean square error [F2(∅)] [18] [Eq. ( 5…”