Regionalization Approach for Extreme Flood Analysis Using L-moments

سال انتشار: 1390
نوع سند: مقاله ژورنالی
زبان: انگلیسی
مشاهده: 69

فایل این مقاله در 14 صفحه با فرمت PDF قابل دریافت می باشد

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

JR_JASTMO-13-7_018

تاریخ نمایه سازی: 1 آذر 1402

چکیده مقاله:

Flood frequency analysis is faced with the problem of data and information limitation in arid and semi-arid regions. Particularly in these regions, the length of records is usually too short to ensure reliable quantile estimates. More than ۷۵% of Iran is located in arid and semi-arid regions and despite the low annual precipitation, often large floods occur. One way to provide more information is to use many records from a region with similar flood behaviour, rather than only at-site data. This research is aimed to delineate homogeneous regions in the study area for further hydrological studies. Estimating regionalized parameters and identification of the best-fit distributions are the other specific objectives of the research. Several watershed attributes in relation to flood were characterized, among which the main characteristics were found by factor analysis. Later, preliminary identification of homogeneous regions was carried out using cluster analysis and region-of-influence approaches. The homogeneity test was done by H-statistic, a testing method based on L-moments. The results of this test showed that a subdivision of selected watersheds into homogenous groups is necessary. Therefore, three homogenous regions were formed. The Z-statistic based on L-moments and L-moment ratio diagrams were applied for identification of the best-fit distribution in each homogenous region. In the regionalization procedure five three-parameter distributions i.e. Generalized Logistic (GLO), Generalized Extreme Value (GEV), Generalized Pareto (GPA), three-parameter Lognormal (LN۳), and Pearson type III (PE۳) were fitted to the three homogeneous regions and the best-fit distributions were identified using L-moments approach. The results of goodness-of-fit analysis for the three regions indicates that the GEV, LN۳ for the regions (۱) and (۲), and GLO and GEV distributions for the region (۳) give acceptably close fits to the regional average L-moments. In general, the GEV distribution could be adopted as the appropriate distribution for the study area.

نویسندگان

H. Malekinezhad

Natural Resources Faculty, Yazd University, Yazd, Islamic Republic of Iran.

H. P. Nachtnebel

Institute of Water Management, Hydrology and Hydraulic Engineering, BOKU University, Vienna, Austria.

A. Klik

Institute of Hydraulics and Rural Water Management, BOKU University, Vienna, Austria.