Scholarlink

<p>Surface downward solar radiation (SDSR) is a critical factor influencing photovoltaic power generation, yet its seasonal prediction capability of climate models over the eastern China and the sources of their prediction biases remain unclear. This study evaluates the prediction skill of summer SDSR in the eastern China with a 1-month lead time using observational, reanalysis and historical hindcast data from 1993 to 2016. A comprehensive analysis is conducted to investigate the sources of low prediction skill in the eastern China, especially in the middle-lower reaches of Yangtze River basin (MLYRB). Key findings include: (1) Significant inter-model disparities in prediction skills are identified. The ECMWF, CMCC, and multi-model ensemble (MME) demonstrate superior spatial anomaly correlation coefficient (ACC) and temporal correlation coefficient (TCC), though with larger mean bias and root mean square errors (RMSE). In contrast, UKMO, Meteo_France, and NCEP exhibit inferior ACC performance despite the advantages in mean bias and RMSE metrics. (2) Relatively large mean biases and RMSE, but higher TCC display over the northeastern China, whereas greater uncertainties appear across multiple evaluation metrics with notably lower prediction skill over MLYRB. (3) The reduced prediction skill in the eastern China (especially in MLYRB) may attribute to two primary factors: direct impacts from model discrepancies in predicting monsoon circulation intensity and position, and indirect effects arising from substantial model biases in the tropical Pacific sea surface temperature (SST). These findings provide crucial references for seasonal photovoltaic power prediction in the eastern China while establishing theoretical foundations for model bias correction strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Assessing prediction skill and identifying potential bias sources of summer surface downward solar radiation over the Eastern China

  • 期刊论文

  • 发表日期：

    2025年11月10日

查看全文
  • Zouxing Lin,
  • Renhe Zhang,
  • Yifeng Lin

摘要

Surface downward solar radiation (SDSR) is a critical factor influencing photovoltaic power generation, yet its seasonal prediction capability of climate models over the eastern China and the sources of their prediction biases remain unclear. This study evaluates the prediction skill of summer SDSR in the eastern China with a 1-month lead time using observational, reanalysis and historical hindcast data from 1993 to 2016. A comprehensive analysis is conducted to investigate the sources of low prediction skill in the eastern China, especially in the middle-lower reaches of Yangtze River basin (MLYRB). Key findings include: (1) Significant inter-model disparities in prediction skills are identified. The ECMWF, CMCC, and multi-model ensemble (MME) demonstrate superior spatial anomaly correlation coefficient (ACC) and temporal correlation coefficient (TCC), though with larger mean bias and root mean square errors (RMSE). In contrast, UKMO, Meteo_France, and NCEP exhibit inferior ACC performance despite the advantages in mean bias and RMSE metrics. (2) Relatively large mean biases and RMSE, but higher TCC display over the northeastern China, whereas greater uncertainties appear across multiple evaluation metrics with notably lower prediction skill over MLYRB. (3) The reduced prediction skill in the eastern China (especially in MLYRB) may attribute to two primary factors: direct impacts from model discrepancies in predicting monsoon circulation intensity and position, and indirect effects arising from substantial model biases in the tropical Pacific sea surface temperature (SST). These findings provide crucial references for seasonal photovoltaic power prediction in the eastern China while establishing theoretical foundations for model bias correction strategies.