Evaluation of four hydrological models to simulate daily streamflow time series in a tropical river basin of Brazil

The Electric Energy Company of Parana (COPEL GeT), the Meteorological System of Parana (SIMEPAR) and RHAMA Consulting company are undertaking the research project PD-6491- 0503/2018 for the development of a hydrometeorological seasonal forecasting for Brazilian reservoirs. The project, sponsored by the Brazilian Electricity Regulatory Agency (ANEEL) under its research and development program, aims the forecasting of streamflow, at temporal scales ranging from 1 to 270 days, at hydro power enterprises, which are integrated by the National Power System Operator (ONS) through its Interconnected System (SIN). With the aim of implement a seasonal forecasting system using different hydrological modeling approaches, it is necessary first to validate the use of different hydrological models during the calibration and validation stages. This work evaluates the performance of four conceptual hydrological models to represent daily streamflow regimes at four hydropower plants located in the Teles Pires river basin (Brazil). The adopted models included the GR4J, HYMOD, HBV, and the SMAP. The calibration of the parameters for each hydrological model was performed using the SCE-UA method and a triangular weighting function was adopted for routing the hydrograph between sub-watersheds. The evaluation of each model was elaborated by the comparison of the observed and simulated streamflow time series during the calibration (2010-2016) and the validation period (2016-2019) using deterministic metrics and statistical tests. A post-processing procedure based on the quantile-quantile method was applied in order to correct the simulated data and reduce the bias with respect the observed data. In general, the results show that the SMAP model present a better performance to simulate the daily streamflow regimes at the simulated hydropower plants, with Nash-Sutcliffe coefficient (NSE) greater than 0.65, and NSElog values greater than 0.8. In addition, the bias correct procedure shows a significant improvement in the adjust of the simulated data to represent the periodic streamflow regimes in the selected river basin.