بررسی ارتباط الگوی پیوند از دور دریای شمال خزر با بارش‌های ایران (مطالعه موردی: استان اردبیل)

در این پژوهش ارتباط الگوی دریای شمال خزر با بارش های استان اردبیل شامل ایستگاه های اردبیل، خلخال، مشگین شهر و پارس آباد در بازه زمانی 1987 تا 2015 بررسی شد. پس از شناسایی همبستگی بین شاخص ncp و بارش استان اردبیل و بررسی همدیدی فازهای مثبت و منفی الگو، از شبکه عصبی مصنوعی برای مدل سازی این رابطه استفاده گردید. نتایج نشان داد که در مشگین شهر، همبستگی بارش و ncp در فوریه و دسامبر مثبت و معنادار بود، اما در مقیاس فصلی، ارتباطی بین آن ها وجود نداشت. همبستگی بارش های اردبیل با شاخص  ncpمعنادار نبود، ولی در خلخال، بارش ماه های نوامبر و دسامبر با شاخص  ncpهمبستگی مثبت داشت. بیشترین همبستگی بین بارش پاییزه پارس آباد و شاخص ncp بود که همبستگی آن ها مثبت و در سطح 99 درصد معنادار بود. استقرار ناوه بر روی دریای مدیترانه و قرارگیری منطقه موردمطالعه در جلوی ناوه، وضعیت چرخندی فراهم آورده که باعث انتقال رطوبت دریای شمال، مدیترانه، دریای سیاه و اقیانوس اطلس به سمت ایران شده که این وضعیت موجب وقوع بارش های پاییزه در دشت مغان می شود. اما در فاز منفی، با ایجاد پشته ی قوی در ایران، شرایط جوی پایدار در منطقه حاکم شده و موجب وزش هوای گرم و خشک آفریقا به سمت ایران می گردد که دوره خشکسالی منطقه می باشد. ارتباط بین بارش پاییزه پارس آباد با شاخص ncp با استفاده از مدل شبکه عصبی مصنوعی، مدل سازی شد. نتایج نشان داد که ضریب رگرسیونی مدل، 0.98 و مجموع مربع خطاهای آزمون، 2.16 بود که نشان می دهد که مدل از برازش مناسبی برخوردار است.

investigation of the relationship between the north caspian sea pattern tele connection and iran’s precipitation (case study: ardabil province)

introductionthe north caspian sea pattern (ncp) is one of the atmospheric phenomena that originates from the pressure fluctuation at the 500 hectopascals level between the north sea and the caspian sea. for the north caspian sea pattern, a numerical index has been defined, which is calculated based on the geopotential height difference of 500 hectopascals between the north sea and the caspian sea. after the numerical calculation of this index, the output number is positive or negative, and the output of the positive number indicates the negative phase and the output of the negative number indicates the positive phase of this index.materials and methodsto investigate the relationship between the precipitations of ardabil province with the north caspian sea pattern, first the average precipitation, temperature and relative humidity of ardabil, parsabad, khalkhal and meshgin shahr stations during 1987 2015 was prepared from the iri meteorological organization.then, the data of the north caspian sea pattern for the period of 1987 to 2005 were prepared and used from the climate research system of east anglia university. in order to calculate the mentioned index until 2015, statistical equations were used and the numerical amount of ncp index was prepared for the years 1987 to 2015.then the monthly, seasonal and annual relationship of the north caspian sea pattern with the parameters of precipitation, temperature and relative humidity of iran during a period of 29 years (1987 2015) was investigated with pearson correlation at 95 and 99% level. in order to model the relationship between precipitation and the ncp model, the forward perceptron artificial neural network model was used.results and discussionthe results showed that the correlation between monthly, seasonal and annual precipitation in ardabil with the ncp index is not significant, and relative humidity has no significant relationship with the ncp index, but the temperature in february and july had a significant relationship with the ncp index at the 95% level.the results also showed that the relationship between precipitation and ncp index in khalkhal is more than ardabil, which was significant in november and december as well as summer and autumn seasons and the annual average of their relationship was significant at the 95 and 99% level. in november and december (autumn season), the correlation coefficient of precipitation and ncp index was positive, which shows that the amount of precipitation increases as the ncp index becomes positive. the correlation between the ncp index and the precipitation of mashgin shahr in january and december was significant and positive at the 95% level. the highest correlation between precipitation and ncp index was observed in pars abad, especially in autumn, and at the 99% level, their correlation was significant and positive; this means that with the positive ncp index, the amount of autumn precipitation in the moghan plain will also increase. in order to fit the best artificial neural network model to the data, a regression line was used, and in this model, the r coefficient was 0.98 for the test data, 0.98 for the validation, and as overall, for the mentioned neural network model, its coefficient was equal to 0.98%, which indicates the appropriate fit of the model in predicting the amount of autumn precipitation in pars abad.conclusionmoghan plain, which is located in a flat area, and sablan mountains and high altitudes do not have an effect on its climate, are the only tele connection patterns that determine its precipitation regime, and this is the reason why the relationship between autumn rains is positive and significant with the ncp index. the positive phase of the north caspian sea pattern has been associated with the high mediterranean trough, so that the studied area is located in front of the trough and the transfer of moisture from the mediterranean sea, the atlantic ocean and the black sea has caused autumn rains in parsabad. therefore, in spite of the influence of enso and north atlantic oscillation patterns on the precipitation of ardabil province, the north caspian sea pattern also plays an important role in its autumn precipitation, especially in the moghan plain.