Downscaling the atmospheric general circulation model's data and its application in simulating the climatic parameters (Case study: Guilan province)

Climatological parameters are among the most important ecological capability evaluation criterias, on other hand, climate changes which has intensified in the past decades and has the consequences such as; increases in climatic threshold phenomena like occurrence of horrific floodings, destructive hurricanes, abnormal and sudden colds and heats, untimely rainfalls and heavy snows, widespread droughts and etc., has shown the necessity of studding the impacts of climate change on various parts of economy and social more than before. But due to low spatial and temporal resolution of global climate models, regional experts must increase the resolution of this model's data using downscaling methods. This study has evaluated the power and precision of LARS-WG model in climatic data generating and future climate forecasting for Guilan province of Iran. Accordingly the daily data collected from synoptic stations in Guilan, with a minimum of 15 years of daily data, between 1995 to 2009 has been used. The parameters used included; rainfall, minimum temperature, maximum temperature and solar radiation. The results show that, the highest calculated value of Mean Absolute Error (MAE) for rainfall modeled data was 14.48 in Astara station, and the highest bias calculated value for rainfall parameter was -4.35 and in Astara station too. The model precision for modeling the minimum and maximum temperature was desired for modeling this parameters, and the highest MAE and bias values for minimum temperature were 0.17 and 0.065 in sequence and both in Anzali station. Also, the highest MAE and bias calculated values of maximum temperature were 0.26 and 0.23 in sequence and both in Rasht station. And in LARS-WG precision in modeling the solar radiation parameter, the Rasht station had the highest calculated MAE and bias values, with the amount of 0.31 and 0.08 in sequence. Results analysis show that the LARS-WG model, has the proper power and precision for climate modeling and data generating in Guilan province of Iran.