Socioeconomic Drought Propagation in Cascading Reservoirs in Southeastern Brazil

Código

I-EBHE0196

Autores

Júlia Mella Massing, Vinícius Bogo Portal Chagas, Pedro Luiz Borges Chaffe

Tema

WG 2.1: Drought in the Anthropocene

Resumo

Brazil?s hydropower generation capacity in recent years was often compromised by prolonged drought events. Drought characterization usually relies on large-scale meteorological and hydrological conditions without properly considering water demands, artificial reservoir storage and resilience. Understanding how socioeconomic droughts are generated under the influence of reservoirs is crucial for effective water and energy generation management, especially for cascading systems. In this work, we investigate the propagation of meteorological droughts into hydrological and socioeconomic droughts in a series of cascading hydropower plants (HPPs) reservoirs of the Paranapanema basin in southeastern Brazil. The Paranapanema basin, with a total area of 105,000 km2, relies on a series of cascading reservoirs for water storage and consequently hydroelectric power generation. We analyze a series of eight HPPs: Jurumirim, Piraju, Chavantes, Ourinhos, Salto Grande, Canoas II, Canoas I, Capivara and Rosana. For each HPP, we analyze daily time series (1990-2020) of water storage, turbined water flow, inflow and outflow as available at the Brazilian National Electric System Operator database. We characterize meteorological droughts with the Standardized Precipitation Index (SPI) and hydrological droughts with the Standardized Streamflow Index (SSI). Regarding socioeconomic droughts, we adopt the Multivariate Standardized Reliability and Resilience Index (MSRRI), which is composed of two covariates: (i) the Inflow-Demand Reliability Index (IDR), which evaluates the reliability of inflow in sustaining the water demands for energy generation; and (ii) the Water Storage Resilience Index (WSR), which evaluates the resilience of water storage in sustaining the water demands. We use turbined water flow to define water demands. All indices are calculated at the monthly time scale. Our results show that, for every reservoir, the MSRRI time series is in general aligned with both SPI and SSI time series, indicating that socioeconomic hydrological droughts are usually followed by socioeconomic droughts, revealing a low resilience of the reservoirs. Our interpretation is that the HPP operations closely follow streamflow predictions as part of the Brazilian Integrated Operation System. However, we found a clear differentiation in how socioeconomic droughts are generated based on the reservoir location, that is, whether it is upstream or downstream. Reservoirs located upstream had socioeconomic droughts linked with inflow-demand deficiency (that is, IDR). On the other hand, reservoirs located downstream in the basin had drought events linked mainly with water storage deficiencies (that is, WSR). These findings suggest that the development of socioeconomic droughts is better understood when incorporating water demands and resilience information, allowing for better water security management.