Verification of statistical cloudiness estimations for Europe
Blanka Bartók, Zoltán Imecs
Abstract. Verification of statistical cloudiness estimations for Europe. The climate forcing induced by cloud cover consists one of the main doubtful aspect of climate change predictions. In the case of cloudiness even the sign of the trends are not cohesive in a given region. In this sense further investigation regarding the behavior of cloudiness are indicated. In this study a statistical estimation of total cloudiness is elaborated using the method of instrumental variables. For this analyze surface-observed monthly mean cloudiness data was applied for the period of 1973-1996. In the second part of the study the verification of results is established using an independent satellite retrieved data series for the period of 2005-2011. Based on verification can be conclude that the applied statistical estimation is able to reproduce the measured values with an RMSE 7, 3%, the difference between the measured and predicted changes of cloudiness is 1.44%, found a stronger decrease of cloudiness in real data as the estimation had indicate. The main differences between the observed and predicted value is evident in the distribution of the frequencies showing a shifting towards the lower values in observed data but not recognized in the estimated values. In the geographical distribution of estimations errors sign a difference is detected between the water surfaces and continental regions.
Keywords: cloudiness, statistical estimation, verification, climate change