Hydropower under climate uncertainty: Characterizing the usable capacity of Brazilian, Colombian and Peruvian power plants under climate scenarios

Abstract

Abstract Hydropower may be a low-carbon option to increase power generation in developing countries, but these countries are some of the most vulnerable to climate change. Climate change can affect hydropower generation through changes in the timing and magnitude of precipitation, rising temperatures, and glacier mass changes. Evaluating climate impacts on hydropower generally requires detailed local input data and hydrological models, which may not be available in many developing nations. Nevertheless, the need to understand the impacts is essential for the developing world. Here we present a modeling framework that relies on remotely sensed and global gridded datasets forced by an ensemble of 21 general circulation models (GCMs) under two representative concentration pathways (RCPs) to evaluate climate-induced impacts on hydropower through the 21st century. We include 134 hydropower plants (>100\u202fMW), representing 42% of hydropower installed capacity in South America, across five regions of Brazil, Colombia, and Peru. Our results suggest the median monthly usable capacity would increase for Colombia (+2.6% to +8.4% for RCP 4.5 and 8.5, respectively) and Peru (+6.7% to +9.3% for RCP 4.5 and 8.5, respectively) by 2100 relative to the late 20th century. For Brazil, we observe a mix of reductions and increases in usable capacity. While our results suggest potential reductions for the dry season usable capacity in the Parana, Paraguay, and Southeast Atlantic regions of Brazil, we also observe slight increases in usable capacity during the rainy months for all its regions. These results can help inform future planning decisions and potential interconnections between the three countries. Additionally, the proposed framework can contribute to an increased capability to evaluate climate-induced risks to power systems in developing countries, where data and computation resources can be limited.