On the generation of an optimized fractional cloudiness time series using a multi-sensor approach

Five active and passive remote sensing techniques that observe clouds are operated nearly continuously at the Cabauw Experimental Site for Atmospheric Research [CESAR]. The instruments involved are: 2 ceilometers, including the sensor that is used operationally by KNMI for generating the automated cloud observations; a cloud radar; an infrared pyrgeometer; a scanning pyrometer and a total sky imager. The total cloud cover has been determined for each of these techniques at 10-minute intervals and their characteristics have been investigated. Next an algorithm was designed to derive a continuous and optimized record of fractional cloudiness that takes into account the strength and weaknesses of the individual observational techniques. The algorithm accounts for breaks in instrumental availability and assigns an error to the data. The fractional cloudiness characteristics of the individual instruments and the synthetic algorithm have also been evaluated against a climatology of human cloud observations.
The paper presents details and the challenges encountered in the evaluation of the cloudiness obtained with the individual instruments and the construction of the optimized fractional cloudiness series. Particularly the lack of a reference is a serious issue. The measure of uncertainty of the resulting series and the relative contribution of each instrument will be discussed.