Lash model using hydrological response units (HRU) for simulation of the climate change impacts on the hydrology of the upper-middle Grande River Basin, southeast Brazil

Abstract

In Brazil, approximately 70% of all electricity is produced through hydropower plants (HPs), which makes the country dependent on the hydrological regime. Therefore, the power generation capacity can be compromised due to climate change and its impacts over the hydrological regime. In this sense, the hydrological modeling represents an important tool in the context of water resources management. It can be used in the assessment of the impacts, especially associated with climate change. In this context, this study aims: i) to enhance the structure of Lavras Simulation of Hydrology (LASH) with the introduction of the concept of hydrological responses units (HRUs), which has potential to increase the ability of this model to simulate runoff and other hydrological variables, such as soil water storage; ii) to apply this improved version of the model to simulate the impacts on the hydrological conditions and the potential for electric energy generation of the Furnas hydroelectric plant (FHP). LASH model was calibrated and validated using this new approach for the Grande river basin defined at FHP. Soil water storage maps were developed made for each sub-basin discretized by the model. For simulation of the hydrological impacts from climate change, it was considered the Representative Concentration Pathways (RCPs) 4.5 and 8.5 premises simulated by the regional climate models (RCMs) Eta-HadGEM-ES, Eta-MIROC5 and Eta-CanESM2, between the years 2007 and 2099. The hypothetical average annual energy production for FHP was estimated through the Power Duration Curves (PDCs). The results of the Nash-Sutcliffe statistical coefficient (NS), in a daily step, were 0.86 and 0.77, respectively, for the calibration and validation, which allow qualifying the model as high performance for hydrological simulation. Soil water storage behavior was consistent with the geomorphological, climatic and land use characteristics. The results related to the hydrological impacts from climate change indicate a reduction in the monthly average streamflow over the analyzed period for all the RCM (Eta-HadGEM2- ES, Eta-MIROC5 and Eta-CanESM2) and RCPs. The more critical projections were those based on the Eta-HadGEM2-ES model, which resulted in a decrease in the annual energy production of up to -53% over the 21st century in FHP.