كليدواژه :
رﮔﺮﺳﯿﻮن ﮐﻮاﻧﺘﺎﯾﻞ , رﮔﺮﺳﯿﻮن ﺧﻄﯽ ﻣﻌﻤﻮﻟﯽ , روﻧﺪ , ﻣﺘﻐﯿﺮﻫﺎي ﺳﯿﻼب
چكيده فارسي :
سابقه و هدف: بررسي روند تغييرات سيلاب حوضهها در اغلب موارد تنها بر اساس تحليل روند متغير دبي اوج سيلاب با استفاده از آزمونهاي رايج پارامتري و ناپارامتري (رگرسيون خطي معمولي، من-كندال، سن و...) است. در كنار محدوديتهاي اوليه اين روشها معمولاً به برآورد ميانگين يا ميانه شرطي ميپردازند و كوانتايلهاي مختلف را در نظر نميگيرند در حاليكه برآورد دامنه مناسبي از كوانتايلهاي شرطي منجر به درك بسيار مناسبي از الگوي تغييرات ميشود. هدف اين تحقيق كاربرد روش رگرسيون كوانتايل براي برآورد روند زماني (كوانتايلهاي شرطي) متغيرهاي دبي اوج، حجم و تداوم سيلاب ميباشد كه اين تحليل منجر به درك مناسبتري از تغييرات متغيرهاي حداكثر سيلابهاي سالانه ميشود.
مواد و روشها:در گام اول سري زماني حداكثر سيلاب سالانه ايستگاه هيدرومتري تله زنگ در جنوب غربي ايران با طول دوره آماري 55 سال مدنظر قرار گرفت و سري زماني دبي اوج، حجم و تداوم حداكثر سيلاب سالانه استخراج گرديد. در گام بعدي با استفاده از رگرسيون خطي معمولي تحليل روند سريهاي متغيرهاي حداكثر سيلاب سالانه انجام شد و كارايي مدل رگرسيون خطي با استفاده از معيارهاي دقت برازش، آزمون معني داري و تحليل باقيماندهها مورد بررسي قرار گرفت. سپس با در نظر گرفتن بازه (95/0-05/0 با گام 01/0) خطوط رگرسيون كوانتايل براي تحليل روند متغيرهاي حداكثر سيلاب سالانه برآورد شد و معيارهاي دقت برازش و معني داري آماري براي اين خطوط تعيين گرديد. با در نظر گرفتن كوانتايل هاي منتخب 95/0، 85/0، 75/0، 65/0، 55/0، 45/0، 35/0، 25/0، 15/0 و 05/0 نمودار خطوط رگرسيون كوانتايل براي متغيرهاي سيلاب ترسيم شد.
يافتهها: نتايج رگرسيون خطي معمولي بيانگر روند مثبت براي متغيرهاي سيلاب است اما تحليلهاي تكميلي نشان داد اين روش نميتواند ابزار مناسبي براي تحليل روند متغيرهاي سيلاب در اين تحقيق باشد. كاربرد رگرسيون كوانتايل در مقايسه با رگرسيون خطي معمولي منجر به دسترسي به طيف وسيعي از شيب خطوط روند شده است. براي هر سه متغير مورد بررسي 15% شيب خطوط رگرسيون كوانتايل بيشتر از شيب برآورد شده با روش رگرسيون خطي و در ساير موارد كمتر از ان بوده است. بررسي خطوط رگرسيون كوانتايل نشان ميدهد خطوط رگرسيون كوانتايل براي متغير حجم سيلاب در كوانتايلهاي كران بالايي و براي متغيرهاي دبي اوج و تداوم سيلاب در كوانتايلهاي كران بالايي و بازه مياني از نظر آماري معني دار بودهاند و در كران پاييني كوانتايلها تعداد معدودي از رابطه هاي خطي قابل پذيرش شدهاند به طوريكه براي متغيرهاي دبي اوج، حجم و تداوم سيلاب به ترتيب 59%، 31% و 73/0 موارد خطوط رگرسيون كوانتايل در سطح 05/0 از نظر آماري معني دار بودهاند.دقت برازش خطوط رگرسيون كوانتايل در كران بالايي و بازه مياني كوانتايلها بيشتر از كران پاييني ميباشد.
نتيجه گيري: كاربرد رگرسيون كوانتايل ميتواند بدون تأثير از محدوديتهاي روشهاي متداول تحليل روند متغيرهاي سيلاب منجر به دسترسي به طيف وسيعتري از نتايج كاربردي تحليل روند شود. همچنين تفاوت مشخصي بين شيب روند متغيرهاي سيلاب براي كوانتايلهاي مختلف (بخصوص كوانتايلهاي كران بالا) در مقايسه با شيب برآورد شده توسط رگرسيون خطي معمولي وجود دارد بنابراين روش رگرسيون خطي معمولي نميتواند ابزاري مناسب براي بررسي روند رويدادهاي حدي باشد. نتايج نشان ميدهد روند متغيرهاي حدي سيلاب به مراتب بيشتر از روند برآورد شده با رگرسيون خطي معمولي ميباشد و به عبارتي رگرسيون خطي در اين تحقيق منجر به كم برآوردي شيب روند افزايشي متغيرهاي سيلاب شده است. همچنين تحليل چند متغيره روند سيلاب با استفاده از رگرسيون كوانتايل مشخص ميكند به دليل وجود روند قابل توجه در شرايط حدي براي هر سه متغير سيلاب، تغييرات در پتانسيل خطر سيلاب به مراتب بيشتر از نتايج به دست آمده با استفاده از تحليل تك متغيره ميباشد
چكيده لاتين :
Background and Objectives: Investigation of the basin floods in most cases is only based on flood peak trend analysis using conventional parametric or non-parametric (ordinary linear regression (OLR), Mann-Kendall, Sen) tests. In addition to the primary restrictions, these methods usually are provided to estimate the conditional mean or median and do not consider different quantiles while assessing the appropriate domain of conditional quantiles leads to a very good understanding of trend pattern. The objective of this study is using quantile regression (QR) to estimate the time trend (conditional quantiles) of flood variables including peak, volume and duration that result in better understanding of variables of annual maximum floods (AMF).
located in southwestern Iran was considered and the time series of AMF peak flow, volume and duration were extracted.In the next step, trend analysis of AMF variables time series performed using OLR and their efficiency were investigated using fitting precision criteria, statistical significant test and residuals analysis. Then, QR lines were estimated for AMF variables trend analysis considering (0.05-0.95 with 0.01 steps) and their fitting precision criteria and statistical significant test were determined. Considering selected quantiles0.05, 0.15, 0.25, 0.35, 0.45, 0.55, 0.65, 0.75, 0.85 and 0.95 QR lines were plotted for AMF variables.
Results: The OLR results indicated positive trends for AMF variables but complementary analysis showed that this method cannot be a suitable analysis for AMF variables trend analysis in this research. The QR application resulted in wide range of line slopes in comparison with OLR method. For all three variables 15% of estimated line slopes using QR were more than their estimation by OLR. Investigation of QR lines indicated statistical significant regression lines of AMF volume were related to upper bound quantiles while for AMF peak and duration were related to quantiles mid bound plus upper bound and there were a few acceptable QR lines for lower bound for all three variables so that for AMF peak, volume and duration 59%, 31% and 73% of QR lines were statistical significant considering 0.05 significance level. The fitting precisions of QR lines of upper and mid bounds were more than lower bound.
Conclusion: The quantile regression can be used without affecting the limitations of conventional methods for AMF variables trend analysis to access a wider range of applied trend analysis. Also there are certain differences between AMF variables trend slopes (especially for upper bound quantiles) in comparison with those estimated with OLR therefore the OLR method could not be a useful tool for trend assessment of extreme events. The results show trend of extreme flood variables are significantly more than those estimated by OLR and in other words the OLR led to underestimation of AMF variables increasing trend slope. Moreover, multiple variables flood trend analysis using QR revealed that considering significant trends for three flood variables, the flood potential risk are significantly more than those estimated using single variable analysis.