Significance of Impact Factors upon Erosion and Sediment Yield on Northern Shaanxi Loess Plateau Based on BP Neural Network

ZHAO Ming-wei; TANG Guo-an; LI Fa-yuan; YUAN Bao-yin; LU Zhong-chen

您当前的位置：

首页 >

文章列表页 >

Significance of Impact Factors upon Erosion and Sediment Yield on Northern Shaanxi Loess Plateau Based on BP Neural Network

- Significance of Impact Factors upon Erosion and Sediment Yield on Northern Shaanxi Loess Plateau Based on BP Neural Network
- Bulletin of Soiland Water Conservation Vol. 31, Issue 1, Pages: 5-9(2012)
- 作者机构：
  
  1. 南京师范大学虚拟地理环境教育部重点实验室,江苏,南京,210046
  2. 中国科学院地质与地球物理研究所,北京,100029
  3. 中国科学院生态环境研究中心,北京,100085
- 作者简介：
- 基金信息：
- DOI：
  CLC：
- Published：2012
- 稿件说明：
移动端阅览
ZHAO Ming-wei, TANG Guo-an, LI Fa-yuan, et al. Significance of Impact Factors upon Erosion and Sediment Yield on Northern Shaanxi Loess Plateau Based on BP Neural Network[J]. Bulletin of Soiland Water Conservation, 2012, 31(1): 5-9.
DOI：

ZHAO Ming-wei, TANG Guo-an, LI Fa-yuan, et al. Significance of Impact Factors upon Erosion and Sediment Yield on Northern Shaanxi Loess Plateau Based on BP Neural Network[J]. Bulletin of Soiland Water Conservation, 2012, 31(1): 5-9. DOI：

摘要

黄土高原地区的侵蚀产沙与影响因子的关系特征一直是研究热点。以陕北黄土高原23个小流域为实验样区

采用BP神经网络方法

将6个影响侵蚀产沙的因子作为输入变量

侵蚀产沙模数作为输出变量

通过输入变量与隐含层之间、隐含层与输出变量之间的权重矩阵关系

构建关系模型。实验结果显示

该方法可以有效地区分不同因子对侵蚀输沙模数影响的显著性;6个因子对侵蚀产沙影响的显著性由高到低依次为:岩土抗蚀性>蚕食度>沟谷密度>年均降雨量>NDVI指数>粉砂黏土含量。最后

随机选择3个小流域作为检验样本

采用BP神经网络进行预测

验证了该模型的有效性。该研究可望完善小流域侵蚀产沙分析方法。

Abstract

The relations of erosion and sediment yield with their impact factors on the Loess Plateau of China have been a research focus.A relation model based on BP neural network model is constructed by taking 23 small watersheds on the Northern Shaanxi Loess Plateau as test areas.In the relation model

six impact factors are selected as input variables and erosion and sediment yield modulus

as output variable.The weighted matrix is employed to express the interface for input variables and hidden layers and the interface for hidden layers and output variable.Results show that the model can effectively distinguish the correlativity between the six impact factors and erosion—sediment yield modulus.From strong to weak

the six impact factors can be ordered as: soil anti-erodibilitynibble degreegully densityaverage annual precipitationNDVIthe ratio of silt to clay.Finally

the validity of the relation mode is verified by randomly selecting 3 small watersheds and employing BP neural network model.This study may be helpful to improve the methodology of the analyses of erosion and sediment yield in a watershed.

关键词

Keywords

references

Views

1497

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Precise measurement and caculation of carbon sink on check dams in Loess Plateau——A case study at Gaoxigou small watershed， Yulin City， Shaanxi Province

Mechanism of sand-layer sliding induced by water content changes on sand covered loess slopes

Dynamics and driving factors of soil organic carbon sequestration during vegetation restoration on Loess Plateau

Change of climate comfortability and relationship with vegetation cover on Loess Plateau

A Discussion on Trees and Forest Suitability to Sites on the Loess Plateau

Related Author

Huo Shaofeng

Zhao Mengfan

Wang Jian

Li Rongrong

Zhou Mijing

Liu Jiabin

Lü Bo

Han Qilong

Related Institution

College of Soil and Water Conservation Science and Engineering， Northwest A&F University， Yangling

College of Forestry， Northwest A&F University， Yangling

Nanjing University of Information Science & Technology， School of Hydrology and Water;Resources

Key Laboratory of Hydrometeorological Disaster;Mechanism and Early Warning of Ministry of Water Resources

Department of Biological and Food Engineering， Lyuliang University， Lüliang

⁰