Two methods for retrieving UV index for all cloud conditions from sky imager products or total SW radiation measurements
Résumé
Cloud effects on UV Index (UVI) and total solar radiation (TR) as a function of cloud cover and sunny conditions (from sky images) as well as of solar zenith angle (SZA) are assessed. These analyses are undertaken for a southern-hemisphere mid-latitude site where a 10-years dataset is available. It is confirmed that clouds reduce TR more than UV, in particular for obscured Sun conditions, low cloud fraction (<60%) and large SZA (>60°). Similarly, local short-time enhancement effects are stronger for TR than for UV, mainly for visible Sun conditions, large cloud fraction and large SZA. Two methods to estimate UVI are developed: (1) from sky imaging cloud cover and sunny conditions, and (2) from TR measurements. Both methods may be used in practical applications, although Method 2 shows overall the best performance, as TR allows considering cloud optical properties. The mean absolute (relative) differences of Method 2 estimations with respect to measured values are 0.17 UVI units (6.7%, for 1 min data) and 0.79 Standard Erythemal Dose (SED) units (3.9%, for daily integrations). Method 1 shows less accurate results but it is still suitable to estimate UVI: mean absolute differences are 0.37 UVI units (15%) and 1.6 SED (8.0%). Cloud effects on UV Index (UVI) and total solar radiation (TR) are assessed. The role of cloud cover and sunny conditions (from sky images) and of solar zenith angle is investigated. The analyses are undertaken for a southern-hemisphere mid-latitude site where a 10-years dataset is available. It is confirmed that clouds affect TR more than UVI, both in reducing and enhancing the cloud-free radiation. From the analyses, two methods to estimate UVI are developed: (1) from sky imaging cloud cover and sunny conditions, and (2) from TR measurements. Methods' estimations are compared with measurements. © 2014 The American Society of Photobiology.