بسط منحنی‌های هاف برای پنج ایستگاه منتخب در شرق دریاچه ارومیه

نوع مقاله : مقالات پژوهشی

نویسندگان

1 دانشگاه تبریز

2 تبریز

چکیده

در این مطالعه، با استفاده از آمار رگبارهای ثبت شده برای 517 رویداد در 5 ایستگاه عجب­شیر، آذرشهر، بناب، بستان­آباد و لیقوان (واقع درشرق دریاچه ارومیه)، منحنی­های­ هاف بسط داده شد. کل رویدادهای منتخب براساس مدت دوام رگبار در چهار کلاس متمایز به­شرح 1- صفر تا دو ، 2- دو تا شش، 3- شش تا دوازده و 4- بیش از دوازده ساعت دسته­بندی شدند. سپس منحنی­های هاف برای هر دسته به­ازای درصد احتمالات 10 درصد، 20 درصد، ... و 90 درصد رسم گردید. ضمنا منحنی­های هاف برای کل رویدادها (بدون دسته­بندی) در هر ایستگاه نیز تهیه شد. در این مطالعه، از توزیع‌های آماری رایج در هیدرولوژی استفاده شد. همچنین، سه شاخص جدید که نشان­دهنده نسبت درصد عمق بارش از آغاز بارندگی تا پایان 25 درصد، 50 درصد و 75 درصد زمان بارندگی منحنی هاف 50 درصد به منحنی هاف 90 درصد می­باشند، به­صورت S، I و Q تعریف شد. هیتوگراف رگبار طرح در هر ایستگاه برای کل رویدادها (بدون کلاس­بندی) برای منحنی هاف 50 درصد به­دست آمد. مدل ریاضی منحنی­های هاف در فرم مدل لاجستیک بسط و پارامترهای آن تخمین زده شدند. نتایج حاکی از آن است که برای کلاس­های با مدت دوام کمتر از 6 ساعت بخش قابل توجهی از بارش در چارک­های اول یا دوم زمانی رخ می­دهد. درحالیکه در بارش­های با مدت دوام بیش از 6 ساعت بارش­ها با شدت کم آغاز شده و به­تدریج تا آخر بارندگی بر شدت آن­ها افزوده می­شود. همچنین نتایج نشان داد که روند تغییرات مقادیر شاخص­های S، I و Q به­صورت S>I>Q می­باشد. بر اساس نتایج حاصله، مدل لاجستیک قادر به برازش خیلی خوب رگبارها در ایستگاه­های منتخب می­باشد، طوریکه ضریب همبستگی بین مشاهدات و مدل بین 978/0 و 998/0 است که از نظر آماری در سطح 5 درصد معنی­دار بودند.

کلیدواژه‌ها


عنوان مقاله [English]

Development of Huff Curves for the Five Selected Stations in the East of Urmia Lake

نویسندگان [English]

  • saina vakili azar 1
  • Yagob Dinpashoh 2
1 University of Tabriz
2 University of Tabriz
چکیده [English]

Introduction: Water is an important element of all living things. Availability of fresh water in any region is very important. Therefore, understanding the rainfall characteristics is so crucial in water resources management. One of the main tools in analyzing storms is Huff curve. Many investigators used this method for rainfall analysis with different duration. The main aim of this study is plotting and analyzing storms characteristics in the five stations namely Ajabshir, Azarshahr, Bonab, Bostan-Abad and Ligvan.
Materials and Methods: In this study, using the 517 storms in the selected stations (located in the East of Urmia Lake), the Huff curves were extracted. The time period used is from 2001 to 2015. Quality of data was checked carefully prior to analysis. In the first step, the total selected storms were classified into the four distinct classes according to their rainfall duration including i) 0-2, ii) 2-6, iii) 6-12 and more than 12 hours. Then the Huff curves of each category were plotted for different probabilities of 10 percent, 20 percent, … and 90 percent. Analysis conducted for each of the classes, separately.  Moreover, the Huff curves were plotted using the information of all events (i.e. without classification). In this study, some commonly used statistical distributions in hydrology were utilized. The three newly defined indices namely S, I, and Qwere defined and used in the present study. The design storm hyetographs for the selected stations and all the events (without classification) prepared for 50 percent Huff curves. The mathematical model of Huff curves were extracted as the Logistic model. The model parameters were estimated using the Curve Expert software.
Results and Discussion: According to the 50 percent probability for Huff curves, the following results were obtained. For the short- time (0-2 hours) storms, the most proportion of rain received in the first and second quartiles. In the first quartile, between 28 to 44 percent of the total rainfall depth received in the selected stations. In the other words, short storms initiated with high intensity and followed by mild intensity. In the case of 2-6 hours storms class, in the two stations, a large portion of the rain (about 34 up to 39 percent) received in the first quartile. However, in the other two stations about 31 up to 34 percent of total rain received in the second quartile. In the station namely Ligvan (about 28percent of total precipitation depth) received in the third quartile. In some of the stations, and in the case of rainfall duration class of 2-6 hours storms starts with high intensity. However, in some of the other sites rain begin with mild intensity. In addition, for the storms with 6-12 hours duration, three stations can be included in the second quartile, because about 31percent of total precipitation received in this time quartile. However, in the two stations, (about 29 up to 33percent of the precipitation depth) received in the third quartile. In the class of duration 6 to 12 hours, storms begins with mild intensity and the intensity of rain increases as time advances then, finally the intensity of rain decreases till rain ceases. In addition, it can be concluded that for the storms with duration of more than 12 hours, for the station namely Azarshahr a large portion of precipitation (about 35percent of precipitation depth) received in the first quartile. Furthermore, in the two stations about 30 percent of total precipitation received in the second quartile. However, in a station namely Ligvan about 32 percent of total precipitation depth received in the third quartile. In other words, storms with duration of more than 12 hours, different stations had different temporal patterns. Based on 90 percent probability Huff curve, it was found that in the case of short- time storm class, almost in all of the stations, rainfall begins with mild intensity. Then the intensity increases gradually to reach peak in the end of the third quartile. In the 25 percent of remaining time (i.e. the last quartile) the intensity decreased again until the rain terminated. For the rainfall classes of duration more than 2 hours, precipitation reaches to the peak in the last quartile. In the other words, the precipitation begins with low intensity and gradually increases its intensity till the end of rain. In this study, three new indices that represent the ratio of precipitation at 50 to 90 percent probabilities were introduced and the values of these indices were calculated for the selected stations.
Conclusions: It can be concluded that the most portion of rainfall received in first quartile and or second quartile for storms having duration less than 6 hours. Whereas for storms with duration more than 6 hours, rainfall started with low intensity and then the intensity increased through the rainfall duration. The results indicated that at all of the stations and for each of the duration time classes, the order of changing the values of  S, I and Q indices was as S>I>Q. The modeling of the cumulative percent of precipitation as a function of cumulative percent of rainfall duration time performed using the Logistic model for each of the time classes and then its parameters were calculated which are presented in the Table 4. Based on the results, it was found that the Logistic model is able to fit the mentioned curve very well for all of the selected stations. The correlation coefficients estimated between the observed and modeled values were found to be between 0.978 and 0.998 for the sites. The results of this study anticipated to be useful in design of urban drainage structures and rainfall- runoff modeling.

کلیدواژه‌ها [English]

  • Design storm
  • Huff Curves
  • Rainfall temporal distribution
  • Urmia Lake
1. Awadallah A.G., and Younan N.S. 2012. Conservative design rainfall distribution for application in arid regions with sparse data. Journal of Arid Environments, 79: 66-75.
2. Azli M., and Rao R. 2010. Development of Huff curves for Peninsular Malaysia. Journal of Hydrology, 388:77-84.
3. Bustami R.A., Rosli N.A., Adam J.H., and Li K.P. 2012. Development of temporal rainfall pattern for Southern Region of Sarawak. UNIMA E. Journal of Civil Engineering, 3:17-23.
4. Ewea A., Elfeki A., Bahrawi J., and AL-Amri N. 2016. Sensitivity analysis of runoff hydrographs due to temporal rainfall patterns in Makkah Al-Mukkramah region, Saudi Arabia. Arabian Journal of Geosciences, 9:424-435.
5. Ghahraman B. 2009. Short-term rain analysis in Khorasan. Iranian Journal of Watershed Science and Engineering, 2(4): 66-73 (In Persian with English abstract).
6. Ghassabi Z., Kamali G.A., Meshkati A., Hajam S., and Javaheri N. 2016. Time distribution of heavy rainfall events in south west of Iran. Journal of Atmospheric and Solar Terrestrial Physics, 145: 53-60.
7. Hatami Yazd A., Ghahraman B., and Khodashenas S.R. 2009. Regionalization of Rainfall Temporal Pattern in Iran. Iran-Watershed Management Science and engineering, 3(6):66-70 (In Persian with English abstract).
8. HuffF.A. 1967. Time distribution of rainfall in heavy storms. Water Resources Research, 3:1007-1019.
9. Jiang P., Yu Z., Gautam M.R., Yuan F., and Acharya K. 2016. Changes of storm properties in the United States:Observations and multi model ensemble projections. Global and Planetary Change, 142:41-52.
10. Karimi V., Solaimani K., Habibnejad Roshan M., and Shahed K. 2014. Comparison of Some Rainfall Temporal Pattern Determination for Urban Flood Estimation (Case Study: Babolsar). Irrigation Engineering and Water, 13:102-112. (In Persian with English abstract).
11. Khaksafidi A., Noura N., Biroudian N., and Najafi Nejad A. 2011.Rainfall Temporal Distribution Patterns in Sistan & Balouchestan Province (Iran). Journal of Soil and Water Protection Research, 17(1):45-61. (In Persian with English abstract).
12. NiyogiD., Lei M., Kishtawal C.H., and Schmid P. 2017. Urbanization impacts on the summerheavy rainfall Climatology over the Eastern United States. Earth Interactions, 21(5):1-17.
13. Todisco F. 2016. The internal structure of erosive and non-erosive storm events for interpretation rainfall simulation. Journal of Hydrology, 519:3651-3663.
14. Wang W., Yin S., Xie Y., Liu B., and Liu Y. 2016. Effects of four storm patterns on soil loss from five soils undernatural rainfall. Catena, 141:56-65.
15. Wu S.J., Yang J.C., and Tung Y.K. 2006. Identification and stochastic generation of representativerainfall temporal patterns in Hong Kong territory. Stochastic Environmental Research and Risk Assessment, 20:171–183.