Observation and evaluation of sunshine duration changes in turkey and determination of relations with some meteorological parameters

In this study, sunshine duration is estimated by using ground-based observations of some climatic parameters such as cloudiness, relative humidity, wind speed, precipitation, evaporation, pressure, minimum temperature and maximum temperature that are obtained from the 72 meteorological stations which are distributed homogeneous geograghically in Turkey, for the period of 1960-2010. In the estimation of sunshine duration, stepwise regression analysis, which is one of the multiple linear regression methods, has been preferred and used to obtain for the most appropirate model. Meteorological parameters that were used for the estimation of the sunshine duration were different in climatic stations. According to the regression analysis, cloudiness is the most important parameter in the estimation of sunshine duration. Also, for observation of the variation of the sunshine duration, trend analysis has been used. Before the trend analysis, Kruskal-Wallis and Mann-Whitney homogeneity tests have been applied to test the reliability of the data. Mann-Kendall trend analysis method has been used for the stations passed the homogeneity tests. Ultimately, for the period the 1960-1990, many stations have shown significant negative and for period 1990-2010, some stations have shown significant positive trends. By using Geographic Information Systems (GIS) imaging programme monthly, seasonal and yearly sunshine duration maps have been obtained.