SCIUSI

Stereo Cloud Imaging Using Sky Imagers

30 June 2015 - 31 December 2015

Lead Scientist: Dong Huang

Observatory: SGP (Southern Great Plains)

Stereo-imaging techniques have the potential to provide a complementary and inexpensive way to characterize the properties and behavior of clouds in an extended area (up to 30 by 30 km). The fine spatial resolution and high frame rate of modern sky cameras enable the estimation of cloud height, horizontal motion, morphology, and spatial arrangement in exquisite detail. The stereo-imaging technique could also fill some of the cloud observation gaps of the current ARM sensors. For example, it allows for the tracking of small individual cumulus clouds that can be smaller than a radar beam volume, thereby offering a perspective on cloud spatial arrangements and cloud life cycles. Supported by the DOE SunShot Initiative, BNL has developed a fully automated cloud stereo imaging software package based on which the surface irradiance and power production at utility-scale solar farms are forecasted with lead times of 1 to 15 minutes. In order to increase the utility of the cloud stereo-imaging technique in atmospheric science research, it needs to be evaluated against independent cloud measurements, importantly those from lidars and cloud radars. We propose to deploy two BNL sky imagers in the ARM SGP site from July to August, 2015 to assess the uncertainties of cloud base height and wind velocity retrievals by the stereo-imaging technique. We also plan to bring two narrow field-of-view cameras to the IOP and, together with the wide-angle sky imagers, these cameras will enable seamless examination of cloud spatial structure from 1 m to 10 km scales.