Application of Quantile Regression in Analysis if Suspended Sediment Load Rating Curve

Author(s):
Message:
Abstract:
The sediment rating curve is the most widely used method to estimate river suspended sediment load that shows the relation between conditional mean of suspended sediment load and river discharge using Ordinary Least Square (OLS) regression and is be applied to estimate suspended sediment load as a function of the river discharge. The OLS regression model is sensitive to outliers and when its assumptions including assumptions related to the residuals analysis are not satisfied, is not acceptable. Quantile regression is a statistical method that can be used to overcome these limitations in sediment rating curve analysis. In this study, quantile regression method was used to estimate sediment rating curve using data from Alang-Darreh hydrometry station in Golestan province (recorded period years 1987-2001) and the results were compared with the conventional OLS regression method. The results show that application of OLS regression in sediment rating curve analysis led to bias estimation while quantile regression without OLS regression’s limiting assumptions can be appropriately show the effect of river discharge on different quantiles of suspended sediment load distributions especially in upper and lower tail. In addition, it was found that a the magnitude of impact of river discharge belonging to upper, lower and central quantiles of suspended sediment load respectively and with increase in river discharge, the suspended sediment load show more skewness to the right. Moreover, the quantile regression concept is presented as a very important tool to extract the probability density and cumulative distribution functions of suspended sediment load for specific value of river discharge.
Language:
Persian
Published:
Iranian Journal of Irrigation & Drainage, Volume:11 Issue: 2, 2017
Pages:
240 to 250
https://www.magiran.com/p1741652  
سامانه نویسندگان
  • Author
    Meysam Salarijazi
    Associate Professor Water Engineering, Gorgan University, Gorgan, Iran
    Salarijazi، Meysam
اطلاعات نویسنده(گان) توسط ایشان ثبت و تکمیل شده‌است. برای مشاهده مشخصات و فهرست همه مطالب، صفحه رزومه را ببینید.
مقالات دیگری از این نویسنده (گان)