Over much more than 100 years the Human Observer has made cloud cover determinations at many places on earth while just looking at the sky. These data are a valuable source of information about trends in cloudiness on long time scales. Clouds reflect solar radiation and influence the upwelling radiation from the earth-atmosphere system. Hence, changes in cloud cover and cloud composition influence our climate. Therefore, it is of great value to obtain long and consistent records of cloud cover which can then be used to track changes in cloudiness and establish a link with climate variability and climate change. In the past decade the Observer has been replaced more and more by automatic instruments. They are cheaper than the Observer and can operate 24 hours, 7 days a week. And this is a great advantage when you want to study in detail the diurnal variability of clouds.

In the last 20 years a number of instruments have been developed to determine cloud cover. The question we asked ourselves was which of these instruments and techniques would best be able to replace the Observer. In general, there are two types of instruments available. The first type of instrument records either long- or shortwave radiation coming from the entire sky above the instrument and interprets this radiation in terms of cloud cover by means of some calculations. This type of instrument is a so-called passive instrument. The second type of instrument sends out a light or a radar pulse in the vertical direction and measures how long it takes before the pulse reflected by the cloud above is detected back at the instrument. This type of instrument is a so-called active instrument. The active instruments have the advantage that they can determine the height of the clouds very accurately, but a disadvantage is that they only observe the sky directly overhead.

From the analysis of one year of collocated data it turns out that the passive instruments, [and more specifically the NubiScope] are best able to reproduce the observations of the Observer. The NubiScope scans the entire sky every 6 minutes and measures the infrared radiation emanating from 36*23 (azimuth*zenith) points of the sky. From these observations a cloud cover value can be calculated. However, the disadvantage of this instrument is that you never exactly know what the altitude of the cloud is. Therefore, it is proposed to develop a new instrument to determine cloud cover that combines the advantages of the active and passive instruments.

Boers, R., M. J. de Haij, W. M. F. Wauben, H. Klein Baltink, L. H. van Ulft, M. Savenije, and C. N. Long (2010), Optimized fractional cloudiness determination from five ground-based remote sensing techniques, J. Geophys. Res., 115, D24116, doi:10.1029/2010JD014661

Figure 1. Cloud fraction histogram for all instruments and the reference algorithm. The Observer data combined from Rotterdam and De Bilt are showed in the purple line with error bar. The error bar denotes the absolute maximum and absolute minimum values for the last 30 year climate record [1971 – 2000]. NUB stands for NubiScope, TSI for Total Sky Imager, APC for APCADA [all three passive], CLD for Cloudnet, LD4 for LD40 [both active], Ref for Reference and OBS for the Observer (mean of Rotterdam and De Bilt)