[M-AG32] Renewable Energy

[MAG32-P12] Automatic cloud cover observation using all-sky imager: Feasibility study and comparison with manual observation

Chen-Jeih Pan¹, Hsin-Chih Lai^2,3, *Shih-Sian Yang² (1.Department of Space Science and Engineering, National Central University, Taiwan, 2.Environmental Research and Information Center, Chang Jung Christian University, Taiwan, 3.Department of Green Energy and Environmental Resources, Chang Jung Christian University, Taiwan)

Keywords:All-sky imager, Cloud cover, Cloud optical thickness, Manual observation, Automatic observation, Solar power

The Taiwan Central Weather Administration (CWA) has scheduled to abrogate manual observation of meteorological parameters at all CWA weather stations in 2026. As a proxy of sunshine and weather conditions, cloud cover is a traditional parameter observed manually by the naked eye. Among those CWA weather stations established over a century, the cloud cover observation usually has an extra 40–80 years period of data compared with pyranometer observation, which plays crucial roles in solar resource assessment, energy balance, and climate change studies. To extend the data availability of cloud cover observation, we tested the feasibility of automatic cloud cover using all-sky imagers (ASIs) in this study. Cloud cover was extracted from images taken by ASIs at five representative CWA stations. The automatic observation done by ASIs was further compared with manual observation, and correlation coefficients greater than 0.8 were found between the two observations at these stations. Besides, we found some factors that may cause an overestimate/underestimate of cloud cover on automatic and manual observation. The present study demonstrates that ASIs are feasible for automatic cloud cover observation. We expect the automatic cloud cover observation will be a suitable successor to manual cloud cover observation at CWA weather stations in the near future.