Study of climate indices of Tamar River basin Golestan Province in terms of climate change using by LARS-WG model

Global warming caused by human activity and climate change is one of the issues attracted that has attention of many climate scientists. The relationship between climate parameters should be used in climate change studies to understand the complex nature of the environment and predict changes in the future. The reliable tool to investigate climate change effects on different systems is using the climate simulations by coupled general circulation of atmosphere and ocean. These models are capable to model the oceanic and atmospheric parameters for a long time period using IPCC scenarios. Due to the low spatial resolution of down scaled phenomena, in general circulation climate models, these models cannot provide accurately approximation of climate conditions of study areas. Therefore, outputs of these models should be down scale to weather station.
The use of statistical methods especially when lower cost and faster assessment of climatic factors is required, have more advantages and capabilities. These models downscale the large scale circulation data by using outputs of GSM models and applying specific scenarios that produce climate data. In this study a type of GCM model as HADCM3 for the period 2046-2065 was used. To simulate climatic parameters in Tamar Basin, the HADCM3 data downscaled using LARS-WG mode under A2 and A1B scenarios. Tamar river basin is located in Golestan Province north-east of Iran that have 1525.3 km² area. There are a few climatology and rain gauges in Tamar river basin. Most of these gauges except Tamar station that have more than 40 years precipitation and temperature data have short inventory period data (15 years rainfall data and 8 temperature data). According to the International organization WMO standards which at least thirty years considered as reference period, therefore, in this study the Tamar climatology data that were recorded for 30 years were used. For this purpose the temperature and rainfall data of Tamar station In a 30-year period (1981-2011) Was extracted. Due to the lack of sunshine data in stations, the Maravehtappe synoptic data, located at 30 km from the centre of the basin, was used. According to the Tamar basin area and variation in hypsometry of basin and also Tamar station located at outlet of basin, the rainfall and temperature data collected in this station cannot present the whole of basin changes. To solve the mentioned problem the temperature data was generalized for the whole of basin using a gradient equation with the differences between altitude of the station and the average altitude of the basin. The rainfall data also after the hydrologic processing, was transfered to the average altitude of basin using gradient equation. So the 30 day data in the month was randomly selected and the minimum and the maximum temperature data based on Tamar, Rebat-e-Ghrabil and Cheshmekhan station that located at the outside of the basin was extracted.  Also the rainfall data of Tamar station with Tangrah, Rebat-e-Gharabil and cheshmekhan that are located at the outside of the basin were used. Then according to the obtained data, the gradient related to 30 days for each year was plotted, and a relationship was obtained. Totally, 2700 gradient relationship for 30 year also for maximum and minimum temperature and rainfall data were generated. Then, 30 gradient relationships for the maximum temperature and the minimum temperature and the rainfall data were selected with the gradient relationship of each year with higher correlation coefficient. Then the gradient relationship for each year and according to the highest percentage of watershed area that was located in the same altitude of centroid of the basin was acquired. Maximum and minimum temperature data for each year were moved to the center of the basin and data corresponding to the height of centroid the basin for log to climate models were obtained. In this study, in order to down scale of the atmospheric general circulation model data HADCM3, the LARS-WG model which is one of the weather generator models was used. To run this model in this research, calibration period was selected between 1981-2011, years then the model was run after preprocessing the input data.In the next step the model was assessed with NSE and RMSE and MAE indices.
Results show that the simulation data for this period are in good agreement with observation data. To evaluate climate fluctuations in the Tamar basin, general circulation model data were down scaled using LARS-WG model according to both A1B and A2 scenarios and thus the daily values of the parameters were generated. The results showed that the average temperature will increase under A2 scenario about 2.48 ° C and under A1B scenario about 2.43 ° C. Meanwhile the maximum temperature change will be higher than the minimum temperature change. From this subject we can conclude that the changes (increases) in the average air temperature in the future will be most affected by the minimum temperature. The results show that 16% increase in precipitation under A2 scenario and 2% rainfall under A1B scenario during 2046-2065 periods. Also, sunshine hours in the study period will be reduced under both scenarios. The results indicate that for the A2 scenario has the highest emissions of carbon dioxide, methane and nitrous oxide, higher temperatures and more rain are expected.