Detailed measurements of enhanced solar irradiation at the surface due to the presence of clouds

The impact of cloudiness on the shortwave downwelling (SDR) and longwave downwelling radiation (LDR) at the surface is investigated by means of collocated pyranometer/pyrgeomater radiation measuremenst and full-sky imager observations. The measurementents have been performed in Westerland, a seaside resort on the North Sea island of Sylt, Germany (54.93 N, 8.31 E) during summer 2004 and 2005. A main improvement to previous studies on this subject resulted from the very high temporal resolution of sky images (every 15 seconds) and the radiation (every second) measurements. With this it was possible to observe an excess of solar irradiation compared to clear sky irradiation by more than 500 W m-2

Camera images reveal that largest excess radiation is reached close to overcast situations with cirrus or altocumuls clouds partly obsuring the solar disk, and preferably with cumulus clouds in lower levels. The physical mechanism for largest local solar radiation most likely consists of a combination of diffusion of direct solar radiation at high altitudes and scattering at cloud sides at lower altitudes. These findings support the results of e.g. \cite{SabburgWong00} for UV-B radiation.

Quantification of total cloud amount (TCA) from the camera measurements is rendered difficult due to strong forward scattering at atmospheric aerosols near the direction of the sun and due to scattering at the acrylic display dome that shelters the camera. Various techniques have been applied to minimize systematic overestimation of TCA resulting from these scattering effects. Comparison with observations from the German Weather Service DWD at the nearby airport of Sylt still show an overestimation of TCA by the camera in 47% of all cases. Still, differences between TCAs derived from the cameras pictures and from human observations are whithin 1 octa in 72% and within 2 octa in 85% of all cases.

TCA from LDR measurements according to the APCADA-Algorithm developed by Duerr and Philipona (2004) underestimates the observed cloud cover in 47% of all compared cases. Differences are within 1 octa in 60% and within 2 octa in 74 % of all cases. This is the first time, that the APCADA algorithm is directly validated against both human observations and all-sky images.

Duerr, B., Philipona, R., Automatic cloud amount detection by surface longwave downward radiation measurements, Journal of Geophysical Research. D. Atmospheres. Vol. 109, no. D5. Mar. 2004