Evaluation of C3S multi-model seasonal forecasts of wind and solar energy potential over China

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Provides comparative performance metrics and bias analysis for C3S seasonal forecast models, useful for improving renewable energy forecasting.

🏢実務担当者:Can guide selection of forecast models and bias correction methods for wind and solar energy management.

<title>Abstract</title> <p>Seasonal forecasts of wind and solar resources are essential for integrating variable renewable energy into power grids, yet a systematic evaluation of multi-model seasonal forecasts for both variables over China has remained insufficient. This study provides a comprehensive evaluation of six seasonal forecast models from the Copernicus Climate Change Service (C3S) over China during 2022– 2024. At a lead time of one month, all models exhibit higher skill in simulating solar radiation than wind speed. ECMWF SEAS5 consistently outperforms the other models, showing the smallest bias and lowest root-mean-square error for both variables. Spatially, most models overestimate wind power density, especially in western and northern China, while solar radiation is commonly overestimated in the Sichuan Basin but underestimated in western China by several models. Pronounced seasonal disparities exist as solar radiation is overestimated in summer and autumn but underestimated in spring and winter, whereas wind biases peak in spring and minimize in autumn across most regions. Forecast biases exhibit a strong and systematic dependence on terrain elevation, showing a non-monotonic model-dependent reliance. Lead-time analysis reveals that with increasing lead time, solar forecast biases generally increase, whereas wind biases unexpectedly decrease, suggesting that long-lead wind forecasts could be increasingly dominated by predictable large-scale signals. These findings provide practical guidance for model selection, bias correction, and operational application of C3S seasonal forecasts in wind and solar energy management over China. The results underscore the need for model-specific bias adjustment and regionally tailored strategies to improve the reliability of renewable resource predictions.</p>

• Research Square https://doi.org/10.21203/rs.3.rs-10044389/v1first seen 2026-07-10 04:45:32