A Novel Quality Control Framework for Long‐Term Hourly Sunshine Duration Data in China

ABSTRACT Long‐term sunshine duration (SSD) data are critical for solar energy assessments, climate modelling, and ecosystem studies. This study develops a promising systematic quality control (QC) framework to enhance the reliability of historical hourly SSD data across China during 1951~2023, considering the historical data status and characteristics of SSD and related meteorological elements observed from surface stations. The four‐step QC framework integrates (1) consistency checks with daily SSD, (2) plausible range verification, (3) temporal alignment with theoretical sunrise and sunset times and (4) integrated spatial and cross‐element consistency assessments using cloud cover, ground surface temperature variation and precipitation duration. Analysis reveals 4467 station‐months with erroneous zero SSD values (primarily between 1951 and 2000) and 357,500 days exceeding theoretical sunlight hours (predominantly pre‐1980), highlighting critical data gaps linked to historical digitisation limitations. Spatial consistency checks identified regional anomalies, with over 90% data coverage post‐1960, while cross‐element checks demonstrated strong correlations between sunshine percentage (SSP) and cloud cover. The framework assigns QC flags to classify data quality, achieving a robust validation of the dataset. Results underscore the necessity of multi‐element verification, particularly in addressing sparse station coverage in early decades. Furthermore, the study suggests that QC procedures can be further refined with the availability of high‐resolution hourly cloud cover data, enabling more precise assessments at finer temporal scales. This study not only addresses data quality issues in Chinese SSD records but also provides a transferable framework for improving solar radiation datasets in regions with similar observational challenges.