Open AccessEvaluation of a sky/cloud formula for estimating UV‐B irradiance under cloudy skies
Author(s) -
Sabburg J.,
Wong J.
Publication year - 2000
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2000jd900530
Subject(s) - sky , irradiance , cloud cover , solar zenith angle , zenith , cloud height , brightness , cloud fraction , diffuse sky radiation , sky brightness , environmental science , solar irradiance , cloud top , cirrus , atmospheric sciences , meteorology , remote sensing , cloud computing , physics , astronomy , satellite , optics , geography , computer science , scattering , operating system
A new sky/cloud formula is evaluated using results collected over 1 year from an integrated sky camera and irradiance measurement system. Sky properties were recorded every 6 min, simultaneously alongside solar radiation measurements at Toowoomba, Australia (27.6°S, 151.9°E), over the solar zenith angle range of 4.2° to 64.3°. This is the first sky formula that incorporates a complement of parameters based on sky properties around the Sun. The sky properties included cloud cover, cloud brokenness, cloud brightness variation, the angle of maximum cloud cover and aureole brightness. The sky formula was developed by correlating the ratio of the measured UV‐B irradiance to the modeled clear sky UV‐B irradiance for the same atmospheric conditions, with each of the five sky properties listed above. Variation of the UV‐B ratio for each individual sky property is examined. This sky formula, used in conjunction with a clear sky semiempirical UV‐B model, was compared to three published cloud formulas based on cloud cover alone. Using a data set restricted to cases of cloud obscuring the solar disk, the new formula was found to have the highest r 2 value of 0.82. Further comparison with a fourth published formula using a subset of data containing predominately cirrus cloud type, produced an explained variance of 0.91 between the measured and the modeled data.