تاثیر شکست سری زمانی داده‌های بارش بر تغییرات مشخصه‌های خشکسالی (مطالعه موردی شهرهای تبریز و اراک)

خشکسالی یک مخاطره اقلیمی است که در همه‌ی مناطق آب و هوائی رخ می‌دهد. متغیرهای مختلف هواشناسی از جمله بارندگی، دما و رطوبت‌نسبی نقش مهمی در وقوع، شدت و تداوم خشکسالی دارند. تغییر اقلیم و گرمایش جهانی می‌تواند سبب بروز ناهمگنی و ایجاد شکست در داده‌های سری زمانی متغیرهای هواشناسی شود. این ناهمگنی می‌تواند سبب تغییراتی در مشخصه‌های خشکسالی شود. هدف اصلی این پژوهش بررسی همگنی سری زمانی بارش به‌عنوان مهم‌ترین متغیر موثر بر خشکسالی هواشناسی، در تعدادی ایستگاه سینوپتیک در طول دوره آماری 1957 الی 2016 در مناطق مختلف کشور و بررسی تغییرات مشخصه‌های خشکسالی در دوره‌های قبل و بعد از نقطه شکست می‌باشد. پس از بررسی سری‌های زمانی مقادیر بارش سالانه مشخص گردید که فقط ایستگاه‌های تبریز و اراک دارای نقطه شکست می‌باشند. از این‌رو در ادامه، وضعیت خشکسالی بر اساس شاخص‌ها‌ی spi، spei، rdi و erdi در ایستگاه‌های مذکور در دو دوره‌ی قبل و بعد از نقطه شکست پایش و مشخصه‌های خشکسالی بر مبنای مدل زنجیره مارکف و ماتریس احتمال انتقال و هم‌چنین ویژگی‌های دوره‌های خشکسالی بر مبنای تئوری ران تعیین گردید. نتایج حاکی از کمتر شدن اطمینان‌پذیری و بیشتر شدن آسیب‌پذیری نسبت به خشکسالی در دوره دوم در مقایسه با دوره اول است. ضمن آن‌که، متوسط و حداکثر تداوم دوره‌های خشکسالی نیز در همه‌ی موارد در دوره دوم نسبت به دوره اول بیشتر شده است. به‌طور خلاصه تمامی مشخصه‌های خشکسالی در دوره دوم نسبت به دوره اول دارای شدت، مدت و فراوانی بالاتری هستند. به‌طور کلی نتایج تغییرات مشخصه‌های خشکسالی در دوره بعد از شکست نسبت به دوره قبل از آن می‌تواند ناشی از گرمایش جهانی و در نتیجه افزایش تبخیر-تعرق و ایجاد و یا تشدید اثرات خشکسالی باشد. ضمن آن­که بر اساس نتایج حاصل از شاخص erdi در هر دو ایستگاه و در هر دو مقیاس زمانی، شرایط رطوبتی نسبت به سایر شاخص‌ها کمی خشک‌تر شده است. به عبارت دیگر می‌توان بیان نمود که تا حدودی بارش موثر در سال‌های اخیر نسبت به سال‌های اولیه دوره مورد بررسی کاهش یافته است. پیشنهاد می‌شود مشابه با این پژوهش در مورد همه ایستگاه‌های سینوپتیک کشور که دارای دوره آماری طولانی می‌باشند، انجام شده تا اثر گرمایش جهانی و در نتیجه تغییرات بارش و دما بر وقوع خشکسالی نمایان‌تر شود.

The Impact of Breakpoints in Rainfall Time Series on Drought Characteristics Changes (Case Study: Tabriz and Arak Stations, Iran)

Introduction: Drought is a climatic phenomenon and an integral part of climate fluctuations that occurs periodically and intermittently throughout the world and across all climates. However, the magnitude of this natural hazard in arid and semiarid regions, such as most parts of Iran, is more acute due to the high sensitivity and weakness of these areas, and its effects may persist for years after the occurrence of drought. Drought is a multifaceted phenomenon as precipitation, temperature, evaporation, wind and relative humidity play important roles in the drought characteristics such as occurrence, severity, and magnitude. Climate change and global warming, and in some cases displacement of meteorological stations cause heterogeneity in time series of meteorological data. Therefore, the purpose of this study was to investigate the homogeneity and break point in precipitation time series data and the effects of a break point in drought characteristics in some synoptic stations in Iran. Materials and Methods: In this study, homogeneity of rainfall time series data at two time scales of annual (water year) and plant growth periods in some selected synoptic stations of Iran with different climatic conditions was investigated. For this purpose, four tests including Standardized Normal Homogeneity test (SNH), Buishand’s Range test (BHR), Buishand’s U test (BUR) and Petite’s test were applied and the break points were determined. Then, at the stations with break points in the precipitation data series, the drought severity values were determined using four indices of SPI, SPEI, RDI and eRDI, for two periods, (before and after of the break points). Then drought characteristics based on Markov Chain Model and Transition probability matrix including vulnerability, reliability, reversibility and stationary of three condition of droughts (dry, normal and/or wet condition) were determined for the two time scales periods (annual and plant growth periods). Then, the differences between the characteristics for the two periods were investigated. Also, the characteristics of droughtfree time intervals for the two periods based on Run’s theory were determined and compared. Results and Discussion: Based on the homogeneity tests, precipitation data of Arak and Tabriz stations for two scales of annual and plant growth periods have break points. According to the results, in the most cases, the second period’s reversibility was lower than the first period. Reliability and vulnerability also decreased and increased in all cases in the second period, respectively, compared with the first period. In most cases, there was an increase in stationary of drought in the second period relative to the first period. The rate of change in the probability of survival of the normal and wet condition in both periods was increasing and in some cases decreasing. Regarding the results of Run’s theory at the growth periods scale, the average and maximum duration of drought periods increased in all cases in the second period. The minimum, average and maximum severity of drought periods also increased in most cases in the second period. The minimum, average, and maximum values increased in most cases in the second period. On an annual basis, the number of drought periods in most cases has increased in the second period. The average and maximum duration of drought periods increased in all cases in the second period. The minimum, average, and maximum severity of drought periods also increased almost in all cases in the second period. Minimum, average, and maximum of drought magnitude values increased in most cases in the second period with respect to the first one. The minimum, average and maximum values of the droughtfree durations (interval time without drought conditions) in most cases were lower in the second period. At the annual scale, the minimum duration of drought was one year in all cases and no change was found between the time slices. The average duration in most cases was lower in the second period. Conclusion: The results show that the rainfall data of Arak and Tabriz stations have break points in the time scales of plant growth period and annual  periods. The reliability was decreasing while the vulnerability of drought was increasing in the second period, indicating an increase in drought occurrence in recent decades. Moreover, the probability of drought stability (stationary) in the second period increased in most cases. The average and maximum duration of drought periods also increased in the second period. The minimum, average, and maximum drought severity, and the minimum, average, and maximum of magnitude of drought periods were higher during the second period. In most cases, the minimum, average, and maximum of severity and magnitude of droughtfree time intervals were lower in the second period. In general, difference in the characteristics of drought before and after of precipitation break point can be due to increased evapotranspiration, as a result of global warming, intensifying the effects of drought. Moreover, based on the results of the eRDI index, the climatic conditions became drier in both stations and time periods. In other words, it can be stated that the effective rainfall has decreased to some extent in recent years compared to the early years of the study period. Further studies are needed to assess the changes in drought characteristics in all synoptic stations in the country having longterm data.