Typical signatures of the transition zone of cumulus cloud shadows in solar radiation

Abstract

Investigating solar radiation and its variability due to clouds and aerosol is critical for efficient and reliable solar energy systems. During broken cloud conditions, reflections at cloud edges and changing aerosol properties in the vicinity of clouds affect the surface solar radiation significantly. In these situations, the distinction between clouds and clear skies with aerosol is not always well defined[1]. As seen from the surface, this region exists around cloud core shadows and is called the transition zone. Here, a unique dataset of observations from a dense pyranometer network is used to detect and investigate signatures of shortwave broadband transmittance in the transition zone. The TROPOS pyranometer network consists of up to 100 individual stations. Data of one campaign is used for this study: 60 stations were distributed over an area of about 6 km² during the S2VSR[2] measurement campaign in 2023 at the ARM Southern Great Planes (SGP) site in Oklahoma, USA. The surface solar irradiance is measured at each station with a time resolution of 10 Hz. The transition zone is detected and characterized by applying a modified clear sky detection algorithm[3] to the data. An additional component of our analysis is the determination of the cloud motion. This vector is determined using the Farneback optical flow algorithm[4] on a cloud shadow mask calculated from the “Clouds Optically Gridded by Stereo” (COGS) product[5]. The study aims to quantify the small-scale effects of the transition zone on surface solar irradiance and potential photo-voltaic yield. This information is valuable for photo-voltaic site planning and provides scientifically relevant insights into the interaction between clouds, aerosol and solar radiation. [1] e.g., Calbó et al. 2017, https://doi.org/10.1016/j.atmosres.2017.06.010 [2] https://www.arm.gov/research/campaigns/sgp2023s2vsr [3] Bright et al. 2020, https://doi.org/10.1016/j.rser.2020.109706 [4] Farneback 2000, https://doi.org/10.1109/ICPR.2000.905291. [5] Romps & Oktem et al. 2018, https://doi.org/10.1175/bams-d-18-0029.1