Interannual climate variability strongly influences renewable energy availability, making it a critical factor for achieving UN Sustainable Development Goals (SDGs). However, our knowledge about the potential solar and wind energy production in tropical South America and its relation to ocean‐atmospheric modes of variability is limited; modes such as El Niño/Southern Oscillation (ENSO), the Atlantic Meridional Mode (AMM), among others. Therefore, we investigate the influence of these modes on solar and wind energy. We apply partial correlations and composite analyses to reanalysis and satellite data to identify the processes connecting large‐scale ocean‐atmospheric variability to seasonal anomalies in renewable power generation. Our study identifies three energy hubs as regions with high climatological mean energy availability: The north Caribbean (NC), eastern Brazil (EB) and western Perú/Bolivia (WPB). ENSO influences the sea level pressure (SLP) gradients, generating wind anomalies that directly affect the wind capacity factor (CF). ENSO also affects the solar CF through reduced atmospheric moisture transport and convergence, which results in fewer clouds leading to higher‐than‐average surface radiation or by atmospheric subsidence. ENSO impacts the NC and EB hubs, with weaker effects in the WPB hub. The AMM is associated with cross‐equatorial wind anomalies that modulate wind CF, as well as moisture convergence and cloud cover, thereby influencing solar CF. Wind CF in the NC and EB hubs is inversely modulated by the AMM, weakening winds and reducing radiation over the NC and strengthening winds and increasing radiation on the EB. The Atlantic equatorial El Niño mode (Atl3) exerts minor effects, with anomalies confined to the equatorial Atlantic. Overall, we find limited complementarity between solar and wind energy at interannual time‐scale. Our results provide insights for forecasting energy production and managing energy storage for periods of low renewable energy availability.
Gardeazabal et al. (Sun,) studied this question.