Abstract
This study aims to estimate and evaluate the characteristic behavior of sunshine duration for long-term records. Sunshine duration and other climate variables such as cloudiness, precipitation, relative humidity, etc., have been measured in meteorological stations for a long time all over the world. But in some cases, such as missing data or unavailable station, the estimation of sunshine duration play a crucial role. Statistical models can be used to predict the sunshine duration over climate variables. To evaluate the behavior of sunshine duration, several climate variables were analyzed for different time scales. The data used in this study were collected from a ground-based meteorological station. In the first, all data were arranged according to different time scales as monthly, seasonal, and annual average values. Prediction models were constructed for each time scale. This study used multiple linear regression (MLR) to build the models and the Pearson correlation analysis to determine the relations between the climate elements. The created models for estimating sunshine duration were validated as well. According to the results, MLR can be utilized and recommended for the prediction of the sunshine duration over climate variables.