تحلیل وقوع روزهای عادی و یخبندان با روش زنجیره مارکف مرتبه اول (مطالعه موردی: ایستگاه‌های همدید همدان و ملایر)

نوع مقاله: مقاله پژوهشی

چکیده

در این مطالعه احتمالات پیشامدهای متوالی روزهای عادی و یخبندان برای ایستگاه­های همدید همدان و ملایر، با استفاده از زنجیره مارکف مرتبه اول دو حالته مورد تحلیل قرار گرفت. بدین­منظور، از آمار کمینه دمای روزانه ماه­های اکتبر تا مارس سال­های 2000 تا 2014 استفاده شد. داده­ها بر اساس ماتریس شمارش تغییر حالت روزهای یخبندان و عادی مرتب شده و احتمال مربوطه با توجه به روش درست­نمایی بیشینه محاسبه گردید. احتمال ساده و تداوم روزهای یخبندان و عادی، محاسبه شد. نتایج نشان داد که ایستگاه­های همدان و ملایر، به­ترتیب تعداد 860 و 1199 دو روز عادی متوالی و تعداد 235 و 246 روز عادی بعد وقوع یخبندان را به­خود اختصاص داده­اند. بر همین اساس، روزهای یخبندان بعد از روز عادی در این ایستگاه­ها به­ترتیب 235 و 246 بودند. همچنین تعداد یخبندان­های متوالی دو روزه در ایستگاه­های مذکور به­ترتیب 1400 و 1039 روز بود. درصد احتمال وقوع دوره یخبندان متوالی ایستگاه­های همدان و ملایر به­طور میانگین به­ترتیب 06/81 و 73/76 بوده است. درصد احتمال وقوع روزهای یخبندان در این ایستگاه­ها به­طور میانگین 11/60 و 99/46 بوده است. همچنین درصد احتمال تداوم پی­درپی یخبندان 2 تا 5 روزه نشان داد که مقدار متوسط این احتمال در ایستگاه همدان به­ترتیب، برابر با 9/50، 73/43، 67/37 و 78/32 و در ایستگاه ملایر به­ترتیب، برابر با 82/37، 72/30، 25/25 و 99/20 بوده است. از آنجایی که دماهای پایین تاثیرات مخربی بر تولیدات کشاورزی دارند، از این نتایج جهت پیش­بینی و جلوگیری از خسارات احتمالی می­توان استفاده کرد.

کلیدواژه‌ها


عنوان مقاله [English]

Analysis of Frost-Free and Freeze days based on First-Order Markov Chain Probability Model (Case Study: of Hamedan and Malayer Synoptic Stations)

چکیده [English]

This study tries to analyze the consecutive frost-free and freeze day’s occurance probabilities, in Hamedan and Malayer Synoptic stations using first-order two-state Markov Chain method. The analysis was performed for the minimum daily temperature data during the October to March in the period of 2000-2014. The data were arranged according to frequency matrix of the freeze and frost-free days mode change and the elements of probability matrix were calculated by the maximum likelihood method. In addition, the initial and consecutive probabilities of the frost-free and freeze periods were also calculated. Based on the results, the number of observed periods with 2 consecutive frost-free days at Hamedan and Malayer synoptic stations were 860 and 1199, respectively. In the same order the occurrences numbers of frost-free days after freezing days at the same stations were also 235 and 246, respectively. Accordingly, 235 and 246 freezing days were occurred after frost-free event and also, 1400 and 1039 periods of 2 consecutive freeze days were detected at Hamedan and Malayer stations, respectively. The average probability percentages of the consecutive freeze periods were 81.06 and 76.73 for the Hamedan and Malayer stations, respectively. The average probability percent of freeze days were 60.11 and 46.99 in the studied stations, respectively. Also, probability percentages for freezing duration of 2-5 days showed that the average probability percentages of the consecutive days were equal to 50.9, 43.73, 37.67 and 32.78 in Hamadan station and 37.82, 30.72, 25.25 and 20.99 in Malayer station, respectively. So the results can be used to predict and prevent potential damage, since the low temperatures have a detrimental effect on agricultural production.

کلیدواژه‌ها [English]

  • Hamedan
  • Malayer
  • Markov chain
  • Period freeze
  • Period frost-free
Boer R, Campbell LC and Fletcher DJ, 1993. Characteristics of Frost in a Major Wheat– growing Region of Australia. Australian Journal of Agricultural Research 44(8): 1731-1743.
Box GEP, Hunter SJ and Hunter WG,  2005. Statistics for Experimenters. John Wily & Sons. Inc., U.S.A. 633 P.
Continas JV, 2000. A climatology of freezing rain in Great Lake region of North America. Monthly Weather Review 128: 3574- 3588.
Daniel S, 1985. StatisticalMethods in the Atmospheric Sciences. Dep of Soil, Crop and Atmospheric Sciences, ITHACA, Cornell Univ, New Yourk 453 P.
Dash PR, 2012. A markov chain modelling of daily precipitation occurrences of Odisha. International Journal of Advanced Computer and Mathematical Sciences 3(4): 482-486.
Hoaglin DC, Mosteller F and Tukey JW, 2011. Exploring Data Table, Trends, and Shapes, John Wily & Sons. Inc., U.S.A.
Khalili A, 1997. Integrated Water Plan of Iran. Meteorological Studies, Ministry of power. Iran.
Madelin M and Beltrando G, 2005. Spatial interpolation– based mapping of the spring frost hazard in the Champagne Vineyards. Meteorological Applications 12: 51-56.
Muller GV, Nunez MN and Seluchi ME, 2000. Relationship between ENSO cycles and frost events within the Pampa Humeda Region. International Journal of Climatology 20(13): 1619-1637.
Selvaraj S and Selvis T, 2010. Stochastic modelling of daily rainfall at Aduthurai. International Journal of Advanced Computer and Mathematical Sciences 1(1):52-57.
Srinivasareddy GV, Bhaskar RS, Purohit R C and Chittora A K, 2008.  Markov chain model probability of dry, wet weeks and statistical analysis of weekly rainfall for agricultural planning at Bangalore.  Karnataka Journal Agric 21 (1): 12-16.
Tait A and Zheng X, 2003. Mapping frost occurrence using satellite data. Journal of Applied Meteorology 42( 2): 193- 203.
Watkins SC, 1991. The Annual period of freezing temperatures in Central England 1850-1989. International Journal of Climatology 11 (8): 889-896.
Waylon PR and Leboutillier DW, 1998. The statistical properties of freeze date variables and length of the growing season. Journal of Climate 2:1314- 1328.