• 研究论文 •

### 砒砂岩地区降雨与植被耦合关系对侵蚀产沙的影响

1. 1. 黄河中下游数字地理技术教育部重点实验室,开封 475004
2. 河南大学环境与规划学院,开封 475004
• 收稿日期:2015-09-30 修回日期:2015-12-20 出版日期:2016-03-20 发布日期:2016-03-20
• 作者简介:

作者简介:张喜旺(1979- ),男,河南辉县人,博士,副教授,主要从事生态环境遥感和GIS应用等研究.E-mail: zxiwang@163.com

• 基金资助:
国家科技支撑计划项目(2013BAC05B01);河南大学青年科研人才种子基金(ZZJJ20140009);黄河中下游数字地理技术教育部重点实验室开放基金(GTYR2013006)

### Coupling relationship of precipitation and vegetation and its influence for sediment yield in Pisha sandstone area

Xiwang ZHANG1,2(), Fen QIN1,2

1. 1. Key Laboratory of Geospatial Technology for Middle and Lower Yellow River Regions, Ministry of Education, Kaifeng 475004, Henan, China
2. College of Environment & Planning of Henan University, Kaifeng 475004, Henan, China
• Received:2015-09-30 Revised:2015-12-20 Online:2016-03-20 Published:2016-03-20

Abstract:

Pisha sandstone area is a major source of coarse sediment, as well as a region with the most intense soil erosion in the Loess Plateau. Its erosion is difficult to control and manage in the eco-environmental construction. This paper reveals the influence of coupling relationship between precipitation and vegetation on sediment yield by analyzing the relationship among precipitation, vegetation and sediment yield. MODIS 250-m NDVI data and TRMM rainfall data are used to research the distribution patterns and matching relations between rainfall and vegetation during the year. Based on the analytical results, a rainfall and vegetation coupling index RV is constructed to reflect the situation of erosion and sediment yield. The index RV is validated by the measured data of Longmen control hydrological station. This paper further analyzes the impact of rainfall and vegetation matching pattern on erosion and sediment yield in different years. Correlation analysis and multiple regression analysis are carried out between sediment yield and the different combinations of annual rainfall, cumulative NDVI and their temporal distribution parameters (kurtosis and skewness). The results show that, (1) The concentration degree and skew degree of rainfall during the year, and its inter-annual instability are more obvious than NDVI. (2) The coupling index RV is not sensitive for the erosion produced by little precipitation, but can effectively reflect the relative size of the sediment yield in the annual period, especially for the time of the maximum erosion. And the correlation coefficient is 0.84. (3) Rainfall kurtosis and skewness have the highest degree of correlation with annual sediment yield, with correlation coefficient reaching 0.94 and 0.87 respectively, which significantly affect erosion and sediment yield. (4) By comparing the several regression models, we found that rainfall kurtosis and skewness are most important for promoting the model fitting degree of model, indicating its impact on erosion and sediment yield. (5) If rainfall, NDVI and their distributed statistical parameters are simultaneously considered in the modeling process, the model correction determination coefficient can reach 0.9232, having the best fitting degree for the sediment yield. Therefore, the matching pattern of rainfall and vegetation during the year is the dominant factor impacting erosion and sediment yield. And the same annual rainfall will cause significant differences in erosion and sediment yield due to their different distribution patterns during the year. When heavy and continuous rainfall occurs, the weaker the protective effect of vegetation is, the greater the amount of erosion and sediment is. Moreover, during the growth period when vegetation lost its protection capability, even a small rainfall event would produce a greater erosion and sediment yield.