Hydro power can provide a source of dispatchable low-carbon electricity and a storage solution in a climate dependent energy mix with high shares of wind and solar production. Therefore, understanding the effect climate has on hydro power generation is critical to ensure a stable energy supply, particularly at a continental scale. Here we introduce a framework to model the effect of climate factors on hydro power generation at the country level based on a machine learning method, the random forest model. Using climate data from the ECMWF ERA5 reanalysis product, this model is used to produce a publicly accessible hydro power dataset from 1979 to present for twelve European countries. By using multiple lagged values of temperature and precipitation, the model accurately reproduces hydro power generation: depending on the country considered, the correlation between actual and simulated values vary between 0.87 and 0.99 for run-of-river generation and between 0.53 and 0.91 for reservoir-based generation, and the normalised mean absolute error decrease compared to previous method optimal lag by average 9.9% and 4.9%, respectively. While the approach used here is unlikely suitable for detailed hydro power production simulations at the basin scale, the model has also been tested at a sub-country scale to test its flexibility, using the six Italian power bidding zones. It is shown that its accuracy degrades only marginally. Overall, the framework and results presented could provide a useful reference for applications such as pan-European (continental) hydro power planning and for system adequacy and extreme events assessments.

HO Linh;  DUBUS Laurent;  DE FELICE Matteo;  TROCCOLI Alberto; 

2020-10-13

MDPI

JRC119878

1996-1073 (online),   

https://publications.jrc.ec.europa.eu/repository/handle/JRC119878,   

10.3390/en13071786 (online),