Climate Change Assessment by using LARS-WG Model in Gilan Province (Iran)

Most of the warming over the last 50 years has been caused by emissions of carbon dioxide (CO2) and other greenhouse gases due to human activities. Observed changes in the climate due to increasing greenhouse-gas concentrations have made it essential to investigate these changes. The General Circulation Model (GCM) is the most current method of investigating climate change studies. Although they are imperfect and uncertain, these models are a key to understanding climate change. However, it still has serious difficulties in reproducing daily precipitation and temperature despite an increasing ability of GCMs to successfully model the presentday climate. Even when daily GCMs output is available, the coarse spatial resolution of GCMs and large uncertainty in their output on a daily scale, particularly for precipitation, indicates that the output is not appropriate for direct use with processbased models and the analysis of extreme events. Output of GCMs requires application of various downscaling techniques. One of the downscaling techniques to create daily site-specific climate scenarios is to makes use of a stochastic weather generator. Recently, weather generators have been used in climate change studies to produce daily site-specific scenarios of future climate. Two important reasons for using LARS-WG model include the provision of a means of simulating synthetic weather time-series with certain statistical properties which are long enough to be used in an assessment of risk in hydrological or agricultural applications and in providing the means of extending the simulation of weather timeseries to unobserved locations. In fact, LARS-WG has been used in various studies, including the assessment of the impacts of climate change which can be divided into three distinct steps: calibration model, validation model, and the generation of synthetic weather data. Changes in the climate variables are studied in Gilan Province located in the north of Iran. The output of two GCM models was compared with a stochastic weather generator. In this study LARS-WG and suitable GCM model were used to produce a climate change scenario.
Material and The study area is Gilan Province, which is situated in the north of Iran and located in the South of the Caspian Sea and has about extent areas of approximately14600 kilometers. The performance of the LARS-WG stochastic weather generator model was statistically evaluated by comparing the synthesized data with climatology period at 8 selected synoptic stations, based on 2 GCMs models (MPEH5, HADCM3) and 2 scenarios (A2, B1). In this study, it has shown the period of base data, including precipitation, minimum and maximum temperatures and solar radiation from 1992 to 2010. Firstly, LARS-WG model was performed based on the historical climate data obtained from 1992-2010 to verify the model. The model was performed after assessing the model ability in each station for all 4 states (2 GCMs models based on 2 scenarios). Then, the results were compared and the best model was chosen to evaluate the climate change in the study area.
Model validation is one of the most important steps of the entire process. The objective was to assess the performance of the model in simulating the climate at the chosen site to determinate whether or not it is suitable for using. Firstly, LARS-WG model was performed based on the historical climate data obtained from 1992-2010 to verify the model. A large number of years of simulated daily weather data were generated and were compared with the observed data through t-test. The monthly mean correlation of the precipitation, minimum and maximum temperature, and solar radiation was accepted at the 0.05 confidence level. Then, to select a suitable GCM model, the LARS-WG stochastic weather generator model considered for MPEH5 and HADCM3 models in A2 and B1 scenarios was compared with the mean of all the models. Of all these 4 states, MPEH5 model based on A2 scenario, which has the least difference with the mean of the models, was selected and used to predict the future climate. Finally, the produced data based on the selected model within the period of 2011-2030 was compared with the observed data within the period of 1992-2010 to evaluate the trend of changes between the two periods.
Research results have shown that the mean of precipitation in Gilan Province has decreased during 2011-2030. The mean of precipitation was estimated 15.2 mm. Likwise, precipitation has decreased in the most parts of the study area. The maximum decrease in precipitation has been related to Astara. In the south and west of Gilan Province, including Langerood, Amlash, Ramsar, Roodbar, and parts of Siahkal, Precipitation has increased, whereas the maximum increase has been in the boundary of Gilan with Mazandaran. Precipitation in other months has decreased except in February, March, August and November while the maximum decrease has been in September. The minimum temperature of the study area will increase in Anzali station most with 0.5 ºC. Generally, the mean of minimum temperature in study area has decreased within the period of 2011- 2030 having 0.4 ºC. Most changes have been in winter and spring when the minimum temperature has increased in these periods. The greatest increase happened in May with 1.9 ºC while the maximum temperature of Gilan Province was 0.4 ºC and the most changes have been in the east of the study area. Most changes have happened in April and May when the maximum temperature has increased in these months about 1.2 ºC. This has decreased wet day length (days with precipitation more than 0.1 mm) while the greatest decrease has happened in Rasht. This has been the minimum decrease in the mountainous areas, whereas it has been decreased the wet day length in the plain of Gilan more than mountainous areas. The amount of decrease in wet day length has been 11 days. The greatest decrease has happened in October with 4 days. This has increased the mean of the dry day length (days with precipitation less than 0.1 mm) that the amount of it has been 12 days. Most changes have happened in Rasht and the difference between the two periods has decreased toward the areas. This has also increased dry day length except in April, whereas the greatest increase is in October for 4 days. It has increased the mean of hot day length (days with maximum temperature more than 30 ºC), three days when the greatest increase has happened in Rasht. Except Anzali and Astara, hot day length has decreased in other places. The greatest increase has been in July for 2 days. This will decrease within the period of 2011- 2030; the mean of frost day length of Gilan Province (days with minimum temperature equal or less than 0 ºC) will decrease for 5 days as compared with the base period. The greatest decrease has happened in Astara, Rasht, Roodbar and the south of Talesh, whereas the fewest changes have been in Anzali. Despite the decrease in the average number of frost days in the whole province, in November and December the number of hot days increases. The number of frost days in other months will decrease while the greatest decrease will be in October for 3 days.