تحلیل توزیع‌های آماری در برآورد اثرات تغییر اقلیم بر سیلاب‌های آینده (مطالعه ی موردی: حوضه ی آذرشهر چای)

افزایش گازهای گلخانه‌ای و گرمایش جهانی در اثر تغییرات اقلیمی می‌تواند باعث افزایش احتمال وقوع رخدادهای حدی اقلیمی مانند سیلاب و افزایش فراوانی و شدت آن در بعضی از مناطق کره ی زمین شود. از این رو ضرورت بررسی مقادیر حدی شدت بارش و فراوانی رخداد این کمیت طی دوره‌های گذشته و همچنین تاثیر گرمایش جهانی بر روند آن طی دوره‌های آتی کاملاً احساس می‌شود. در پژوهش حاضر اثر تغییرات اقلیمی بر رواناب حوضه ی آذرشهر با مدل canesm2 تحت سناریوهای انتشار، rcp2.6، rcp4.5 و rcp8.5 با ریزمقیاس گردانی آماری مدل sdsm از طریق مدل هیدرولوژیکی swat بررسی شد. نتایج ارزیابی مدل sdsm با ضریب نشساتکلیف 95/0 بیانگر دقت بالای این مدل در ریزمقاس نمایی داده‌های بزرگ مقیاس است. نتایج مدل اقلیمی حاکی از افزایش دما به میزان 1/0 تا 25/0 درجه ی سانتی گراد و افزایش 4 تا 7 درصدی بارش در دوره ی زمانی 20051976 نسبت به دوره ی 20592030 می‌باشد. به منظور تحلیل فرکانس و شدت سیلاب با استفاده از مدل easy fit مناسب ترین توزیع براساس سه آزمون نکویی برازش انتخاب شد. نتایج بررسی رژیم جریان‌های حداکثر سالانه (فراوانی و شدت) از طریق برازش توزیع‌های احتمالاتی با کمترین میزان خطا برای دوره ی پایه توزیع ویبول، دوره ی آینده rcp2.6 توزیع لوگ پیرسون نوع 3، rcp4.5 لوگ نرمال و rcp8.5 لوگ نرمال به عنوان بهترین توزیع انتخاب شده است. فراوانی و شدت سیلاب نیز افزایش یافته به طوری که، در دوره ی بازگشت 500 ساله افزایش 98 درصدی دبی حداکثر دوره ی آتی rcp8.5 نسبت به دوره ی پایه مشهود شده است.

Statistical Distributions Analysis for Estimating of Climate Change Effects on Future Floods (Case Study: Azarshahrchay Basin)

1IntroductionThe assessment report fifth of the Intergovernmental Panel on Climate Change shows that global warming has led to a change in the water cycle due to increased greenhouse gas emissions. In the present time, with the increase of industrial activities and the neglected environmental issues, the effects of climate change have become more evident and poses this phenomenon as a global difficult. Increasing the probability of occurrence of extreme climatic events such as flood and increasing the frequency and intensity of the effects of climate change. Due to changes in climate and global warming, the probability of heavy rainfall and consequently the risk of flood due to incorrect drainage system and physical and environmental factors have increased. Therefore, the study of the region’s climate is important given the new scenarios and flood frequency analysis with suitable statistical distributions for future planning.2 MethodologyIn the present study, the effects of climate changes on the runoff of Azarshahrchay Basin with CanESM2 model under RCP2.6, RCP4.5 and RCP8.5 release scenarios assessment report fifth (AR5) of the Intergovernmental Panel on Climate Change (IPCC), with Statistical down scaling model (SDSM), for the period 19762005 and 20592030 by the hydrologic model SWAT have been investigated. The accuracy of the simulation was evaluated with three indicators: Root Mean Square Error (RMSE), Coefficient of Determination (R2) and Nash–Sutcliffe Efficiency (NSE). An analysis of the frequency of maximum annual flood for both base and future periods using their probability distribution function (PDF) and the Easyfit model. In this model, 5 types of probability distribution including Normal, Normal Log, Pearson, Log Pearson Type 3 and Weibull were used. The best distribution for each basic and future time series were ranked and selected by using three Chisquare, Kolmogorov–Smirnov, Anderson–Darling tests. In order to study how the maximum flood discharge regime changes in the base and future periods were used two indices: 1) The probability and the return period in the equal flows 2) Intensity of flow in the equal return periods3 ResultsThe obtained factors of the three RMSE, R2, and NSE indicators showed the good performance of the SDSM model in the down scaling the largescale data. Investigating the performance of the SDSM model in the downscale of the Azarshahr station’s climate data with a Coefficient of Determination and Nash–Sutcliffe of 0.99 and 0.98 for temperature for the period 19902001 and 0.86 and 0.83 for precipitation in the period 19762005. The simulation results showed a rise in temperature during the period 20302059 under scenarios and the highest increase was related to RCP8.5 (0.23°c). Also, rainfall at a station increased by 7.44 percent to RCP2.6 and at another station decrease by 7.57 percent to RCP8.5. The performance analysis of the SWAT model indicates a good accuracy of the model in runoff simulation with R2 and Nash 0.6 on average. The results of the 2.1% increase in runoff and the maximum flood peak and the probability of flood events in March and April (late winter and early spring) have been shown by the SWAT model. Results of the study of the regime of maximum annual flows (frequency and intensity) by fitting probabilistic distributions with the lowest error rate for the base distribution period of the Weibull, future period RCP2.6 distribution Log Pearson Type 3, RCP4.5 Log Normal and RCP8.5 Log Normal as best distribution are selected. Also, the frequency and intensity of flood have increased. In the return periods of constant, the maximum discharge increased, and in maximum discharge constant, with increasing return period (1000 years), the discharge rate significantly increased. So, in the 500year return period is expected a 98% increase in maximum discharge RCP8.5 future period than base period. The most critical scenario is RCP8.5 scenario.4 Discussion and conclusionThe results indicate the impact of climate change on the basin in the future period. Therefore, knowing the increase in precipitation intensity, the flood risk increases. The occurrence of terrible floods due to climate change have caused many damages in different parts of the world in recent decades. The results of this study, like other previous studies, confirm that climate change is significant, especially with the increasing frequency of floods, governments, organizations, and educational centers need to take appropriate measures to eliminate or reduce the effects of climate change and adaptation to extreme events such as floods.