Probabilistic estimation of suspended sediment load using quantile regression
碩士 === 國立成功大學 === 水利及海洋工程學系 === 102 === Inevitable scatter existed between suspended sediment discharge is caused by variation in sediment supply. However, The sediment rating curve only describes the average relation between streamflow and suspended sediment. This study aims to construct a probabil...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2014
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/06154391033629460624 |
| Summary: | 碩士 === 國立成功大學 === 水利及海洋工程學系 === 102 === Inevitable scatter existed between suspended sediment discharge is caused by variation in sediment supply. However, The sediment rating curve only describes the average relation between streamflow and suspended sediment. This study aims to construct a probabilistic framework for estimation of annual suspended sediment loads. Firstly, quantile regression is employed to establish the nonlinear relationship between measured data for various quantiles. The results of various quantiles are integrated into a sediment conditional probability distribution function for a given streamflow. The obtain conditional distribution functions of different streamflow are used in this study to estimate annual suspended sediment loads in terms of probabilistic description by convolution theorem. The stations of Lan-Yang bridge, Tung-Tou, Lao-Nung and Jen-Shou bridge are selected in this study. The results show that the proposed approach offers more information than the traditional sediment rating curve for annual sediment estimation.
|
|---|