Abstract:
اﻟﮕﻮﻫﺎی ﮐﻼن ﻣﻘﯿﺎس ﺟﻮی-ﻗﯿﺎﻧﻮﺳﯽ ﻣﺘﻐﯿﺮی ﻣﻨﺎﺳﺐ ﺑﺮای ﭘﯿﺶﺑﯿﻨﯽ ﻋﻨﺎﺻﺮ اﻗﻠﯿﻤﯽ ﺑﻪ ﺧﺼﻮص روزﻫﺎی ﺑﺮﻓﯽ ﻫﺴﺘﻨﺪ. ﺑﺎرش ﻫﺎی ﺑﺮﻓﯽ ﺳﺒﮏ و ﻧﯿﻤﻪﺳﻨﮕﯿﻦ ﺗﺎ ﺳﻨﮕﯿﻦ اﺳﺘﺎن اردﺑﯿﻞ، ﻋﻼوه ﺑﺮ ﺗﺄﺛﯿﺮﭘﺬﯾﺮی از ﻋﻮاﻣﻞ ﻣﺤﻠﯽ، ﺑﺎ ﭘﺪﯾﺪهﻫﺎی ﮐﻼن ﻣﻘﯿﺎس ﮔﺮدشﻫﺎی ﺟﻮی-اﻗﯿﺎﻧﻮﺳﯽ ﻧﯿﺰ در ارﺗﺒﺎط ﻫﺴﺘﻨﺪ. در اﯾﻦ ﭘﮋوﻫﺶ، وﯾﮋﮔﯽﻫﺎی آﻣﺎری روزﻫﺎی ﺑﺮﻓﯽ اﯾﺴﺘﮕﺎه ﻫﺎی ﺳﯿﻨﻮﭘﺘﯿﮏ اﺳﺘﺎن اردﺑﯿﻞ ﻣﻮرد ﻣﻄﺎﻟﻌﻪ ﻗﺮار ﮔﺮﻓﺖ. ﺑﺮای ﻣﻘﺎﯾﺴﻪ ﻣﯿﺎﻧﮕﯿﻦ دوره ﻫﺎی روزﻫﺎی ﺑﺮﻓﯽ اﯾﺴﺘﮕﺎه ﻫﺎی ﻣﻮرد ﺑﺮرﺳﯽ، از آزﻣﻮن ﺗﯽ دو ﻧﻤﻮﻧﻪ ای ﻣﺴﺘﻘﻞ اﺳﺘﻔﺎده ﺷﺪ. در اﯾﻦ ﻣﻄﺎﻟﻌﻪ، ﻋﻼوه ﺑﺮ ﺗﺤﻠﯿﻞﻫﺎی ﺗﻮﺻﯿﻔﯽ، از روش ﻫﻤﺒﺴﺘﮕﯽ اﺳﭙﯿﺮﻣﻦ، ﺗﺤﻠﯿﻞ روﻧﺪ ﺧﻄﯽ و ﭘﻠﯽﻧﻮﻣﯿﺎل درﺟﻪی ﺷﺶ و ﺗﺤﻠﯿﻞ رﮔﺮﺳﯿﻮن ﭼﻨﺪﮔﺎﻧﻪ ﺑﻪ روش ﭘﺲروﻧﺪه ﺑﺮای ﺗﻮﺟﯿﻪ درﺻﺪ ﺗﻐﯿﯿﺮات ﺗﺒﯿﯿﻦ ﺷﺪهی روزﻫﺎی ﺑﺮﻓﯽ اﺳﺘﺎن اردﺑﯿﻞ ﺗﻮﺳﻂ 27 اﻟﮕﻮی ﮐﻼن ﻣﻘﯿﺎس ﮔﺮدشﻫﺎی ﺟﻮی–اﻗﯿﺎﻧﻮﺳﯽ اﻗﯿﺎﻧﻮسﻫﺎی آرام و اﻃﻠﺲ اﺳﺘﻔﺎده ﺷﺪ. ﻧﺘﺎﯾﺞ روﻧﺪ ﺧﻄﯽ ﺗﻐﯿﯿﺮات روزﻫﺎی ﺑﺮﻓﯽ اﯾﺴﺘﮕﺎ اردﺑﯿﻞ، ﺣﺎﮐﯽ از اﻓﺰاﯾﺶ آرام ﺗﻌﺪاد روزﻫﺎی ﺑﺮﻓﯽ اﯾﺴﺘﮕﺎه ﻣﻮرد ﻣﻄﺎﻟﻌﻪ در ﻃﻮل دوره ی آﻣﺎری اﺳﺖ. روزﻫﺎی ﺑﺮﻓﯽ اﻏﻠﺐ اﯾﺴﺘﮕﺎه ﻫﺎی ﻣﻮرد ﻣﻄﺎﻟﻌﻪ، دارای ﻫﻤﺒﺴﺘﮕﯽ ﻣﻌﻨﯽ دار در ﺳﻄﺢ ﺧﻄﺎی 1 و 5 درﺻﺪ ﺑﺎ ﯾﮑﺪﯾﮕﺮ ﺑﻮدﻧﺪ و اﯾﻦ مساﻟﻪ، ﺑﯿﺎن ﮐﻨﻨﺪهی ﻓﺮاﮔﯿﺮی ﺑﺎرش ﻫﺎی ﺑﺮﻓﯽ در ﺳﻄﺢ اﯾﺴﺘﮕﺎه ﻫﺎی اﺳﺘﺎن اردﺑﯿﻞ اﺳﺖ. ﻧﺘﺎﯾﺞ ﻧﺸﺎن داد ﮐﻪ در ﺑﯿﻦ اﯾﺴﺘﮕﺎه ﻫﺎی ﻣﻮرد ﻣﻄﺎﻟﻌﻪ، اﯾﺴﺘﮕﺎه ﺧﻠﺨﺎل ﻫﻤﺒﺴﺘﮕﯽ ﺑﯿﺶﺗﺮ و ﻣﻌﻨﯽ دارﺗﺮی ﺑﺎ اﻟﮕﻮﻫﺎی ﺟﻮی-اﻗﯿﺎﻧﻮﺳﯽ اﻗﯿﺎﻧﻮس آرام دارد. ﻧﺘﺎﯾﺞ آزﻣﻮن ﺗﯽ دو ﻧﻤﻮﻧﻪ ﻣﺴﺘﻘﻞ ﻧﺸﺎن داد ﮐﻪ اﺧﺘﻼف ﻣﯿﺎﻧﮕﯿﻦ روزﻫﺎی ﺑﺮﻓﯽ در ﺗﻤﺎﻣﯽ اﯾﺴﺘﮕﺎه ﻫﺎی ﻣﻮرد ﺑﺮرﺳﯽ در دو دوره ﻣﻄﺎﻟﻌﺎﺗﯽ اﺧﺘﻼف ﭼﻨﺪاﻧﯽ ﺑﺎ ﯾﮑﺪﯾﮕﺮ ﻧﺪارﻧﺪ. ﭘﺮاﮐﻨﺶ ﻣﻘﺎدﯾﺮ ﻫﻤﺒﺴﺘﮕﯽ روزﻫﺎی ﺑﺮﻓﯽ اﯾﺴﺘﮕﺎه ﻫﺎی رد ﻣﻄﺎﻟﻌﻪ ﺑﺎ اﻟﮕﻮﻫﺎی ﺟﻮی-اﻗﯿﺎﻧﻮﺳﯽ اﻗﯿﺎﻧﻮس اﻃﻠﺲ و آرام ﻧﺸﺎن داد ﮐﻪ در اﻏﻠﺐ اﻟﮕﻮﻫﺎی ﻣﻮرد ﺑﺮرﺳﯽ، ﻣﯿﺰا ﺒﺴﺘﮕﯽ روزﻫﺎی ﺑﺮﻓﯽ ﺑﺎ اﻟﮕﻮﻫﺎی ﻣﻨﺘﺨﺐ ﻣﻮرد ﻣﻄﺎﻟﻌﻪ از ﺟﻨﻮب ﺑﻪ ﺷﻤﺎل اﺳﺘﺎن اﻓﺰاﯾﺶ ﻣﯽﯾﺎﺑﺪ.
Introduction
Atmospheric-oceanic macro scale phenomena have caused much of the worldchr('39')s climate change. Atmospheric-oceanic macro scale patterns are a suitable variable for predicting climatic elements, especially snowy days. In Ardebil province, light, semi-heavy and heavy snowfalls, in addition to local factors, are also related with Large-scale Atmospheric-oceanic circulation patterns. Forecasts for snow days in Ardabil province can provide by determinatiom of relationship between snowfalls and macro-scale phenomena of atmospheric-oceanic circulations in this province.
Materials & Methods
In this research, by snow days data of synoptic stations of Ardabil province as sample of climate of North West of Iran, statistical characteristics of snow days of this province was studied. The statistical period used in Ardebil, Khalkhal, Meshkinshahr and Pars Abad synoptic stations were 42, 31, 22 and 33 years from 1976, 1987, 1996 and 1985 to 2017, respectively. Independent two-sample T-test was used to compare the mean snow days of the stations under study. In this research, In addition to descriptive analysis, spearman correlation, order 6 polynomial and linear trend, multiple regression analysis based on backward method for determine of variability of snow days in Ardabil province by using 27 Large-scale Atmospheric-oceanic circulation pattern of Atlantic and Pacific oceans. Indicators of macro-scale atmospheric-ocean circulation phenomena used in this study were obtained from https://www.psl.noaa.gov/data/climateindices which include: SOI, SENSO, EOF , MEI, NINA1, NINA3, NINA4, NINA3.4, ONI, PWP, TNI, EPO, NOI, NP, PDO, PNA, WP, AO, AMM, AMON, ATLTRI, CAR, NTA, TNA, TSA, WHWP and NAO.
Discussion of Results
The results showed that there is an average of about 35 days of snow at Ardabil station on an annual scale. The number of snow days in January, February, March, December, as well as annual scale, has an almost normal distribution. In Ardebil station, the relative normality of the distribution of the number of snow days on an annual scale and its similarity to December to March is due to the high share of snow days in these months in the composition of the number of annual snow days in this station. The coefficient of variation of most stations is also higher in May and October than in other months. Result of linear trend of snow days variability in Ardabil synoptic station shows a slow increase of snow days of this station. Snow days in most of the stations under study had significant correlations at 1% and 5% error levels by each there, which reflects the prevalence of snowfall at Ardabil province stations. In Khalkhal, Meshkinshahr and Parsabad stations, annual decrease in number of snow days is observed in annual scale. The 6th-order polynomial model predicted snow days in Ardabil station rather than other models. T-test results of two independent samples for snow days of the studied stations showed that significant value in Levenechr('39')s test was greater than 0.05, so the assumption of equality of variances cannot be rejected. The significant value of T test for equality of means indicates that in Ardebil and Khalkhal stations, the assumption of equality is confirmed, but in Khalkhal and Parsabad stations the assumption is rejected. At Parsabad station, the number of snowy days was significantly inversely correlated with the Atlantic oceanenic-atmospheric patterns. Results showed that the snowy days in synoptic stations of Ardabil province have a higher and more significant correlation with atmospheric-oceanic patterns of the Pacific Ocean than atmospheric-oceanic patterns of the Atlantic Ocean. At Ardabil, Khalkhal, Meshkinshahr and Parsabad stations, 48%, 90%, 99% and 49% of the changes in snowy days were explained by the atmospheric-oceanic patterns of the Atlantic Ocean, respectively. The results showed that in Ardabil station, NINO3.4, ONI, TNI and NINO1+2 patterns, in Meshkinshahr station, NP pattern and in Parsabad station, NEI, PNA, NP, NINO3.4 and NINO1+2 patterns, explained the highest percentage of changes by the number of snowy days of these stations with atmospheric-oceanic patterns of the Pacific Ocean. The results also showed that in Ardabil station, WHWP, NTA and AMM patterns, in Khalkhal station, WHWP, TSA and AMM patterns, AMO, CAR and TNA, in Meshkinshahr station, WHWP, ATLTRI, AMM and TNA patterns and in Parsabad station, The AMO pattern explained the highest percentage of changes by the number of snowy days of these stations with atmospheric-oceanic patterns of the Atlantic Ocean
Conclusions
Among synoptic station of Ardabil provinec, Khalkhal station has more significant correlation with Atmospheric-ocemic patterns of Pacific ocean. In Ardabil provinec, Atmosphoric-oceanic patterns of Atlantic ocean, determine variability of snowdays of Khalkhal and Meshkinshahr synoptic stotions more than Ardabil and Parsabad synoptic stations. The results of T-test of two independent samples showed that the difference of mean of snow days in all the studied stations in the two study periods were not significantly different from each other and it can not be claimed that the number of snow days in the studied stations has been affected by climate change. Distribution of correlation values of snow days in the studied stations with Atlantic and Pacific atmospheric patterns showed that in most of the studied models, correlation of snow days with selected patterns increased from south to north of Ardabil province.
Machine summary:
در این مطالعه ، علاوه بر تحلیل های توصیفی، از روش همبستگی اسپیرمن ، تحلیل روند خطی و پلینومیال درجه ی شش و تحلیل رگرسیون چندگانه به روش پس رونده برای توجیه درصد تغییرات تبیین شده ی روزهای برفی استان اردبیل توسط ٢٧ الگوی کلان مقیاس گردش های جوّی–اقیانوسی اقیانوس های آرام و اطلس استفاده شد.
پژوهشگران دیگری نظیر (٢٠١١) Alizadeh et al، (٢٠١٢) Almazroui et al، (٢٠١٣) Gelcer et al،2013( )Ahmadi ،2013( )Farajzadeh Asl et al ،2014( )Sobhani et al ،2017( )Heydari and Khoshakhlag، (٢٠١٧) Goodarzi et al، (٢٠١٨) Chehre-Ara et al، (٢٠١٨) Sinan et al نیز به بررسی ارتباط بین شاخص های پیوند از دورها با تغییرات تعدادی از عناصر اقلیمی در مناطق مختلف جغرافیایی پرداختند.
جدول ٤- مقادیر همبستگی روزهای برفی ایستگاه های مورد مطالعه با الگوهای جوّی–اقیانوسی اقیانوس آرام Table 4- Correlation values of snowy days of the studied stations with atmospheric-oceanic patterns of the Pacific Ocean (به تصویر صفحه رجوع شود) TNI *همبستگی در سطح خطای ٥ درصد معنیدار است .
جدول ٥- مقادیر همبستگی روزهای برفی ایستگاه های مورد مطالعه با الگوهای جوّی-اقیانوسی اقیانوس اطلس (به تصویر صفحه رجوع شود) *همبستگی در سطح خطای ٥ درصد معنیدار است .
در ایستگاه های اردبیل ، خلخال و مشکین شهر، الگوهای جوّی-اقیانوسی اقیانوس اطلس و سایر الگوهای پیوند از دور مورد مطالعه ، همبستگی معنیداری با تعداد روزهای برفی این ایستگاه ها ندارند.