引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 157次   下载 99 本文二维码信息
码上扫一扫!
分享到: 微信 更多
基于随机森林算法的土壤侵蚀影响因子研究——以赣江上游流域为例
朱青, 国佳欣, 郭熙, 韩逸, 刘士余, 陈蕾, 罗海玲
江西农业大学 国土资源与环境学院 江西省鄱阳湖流域农业资源与生态重点实验室, 江西 南昌 330045
摘要:
[目的] 分析影响赣江上游流域土壤侵蚀的主要因素,为该区水土流失治理与科学管理提供科学依据。[方法] 基于2015年Landsat 8遥感影像、MODIS NDVI数据、数字高程模型(DEM)、土壤类型和降雨数据,采用RUSLE模型和随机森林算法对赣江上游流域土壤侵蚀及其影响因子进行定量化分析。[结果] 2015年赣江上游流域土壤侵蚀强度由东南向西北逐渐加剧,总体上处于轻度侵蚀水平,土壤侵蚀总量为3.45×107 t/a,平均土壤侵蚀模数为1 046.38 t/(km2·a),比南方红壤丘陵区土壤允许流失量[500 t/(km2·a)]高出2倍之多;子流域9,11,15平均土壤侵蚀模数分别为1 672.66,1 715.83和1 565.36 t/(km2·a),处于中度侵蚀级别,为研究区重点防治区域;其余子流域均为轻度侵蚀级别。[结论] 各子流域的土壤侵蚀受植被覆盖与管理因子(C)和坡长坡度因子(LS)影响较大,两者重要程度分别在30%和20%以上,土壤可蚀性因子(K)和降雨侵蚀力因子(R)的重要程度偏低,均未超过10%。其中子流域9,11,21主要受LS因子影响,其余子流域均受C因子主控。
关键词:  土壤侵蚀  RUSLE模型  随机森林  影响因子  赣江上游流域
DOI:10.13961/j.cnki.stbctb.2020.02.009
分类号:S157.1
基金项目:国家自然科学基金项目"不同尺度上植被空间分布格局的水土保持效应研究"(31960331)
Research on Influencing Factors of Soil Erosion Based on Random Forest Algorithm—A Case Study in Upper Reaches of Ganjiang River Basin
Zhu Qing, Guo Jiaxin, Guo Xi, Han Yi, Liu Shiyu, Chen Lei, Luo Hailing
College of Land Resource and Environment, Jiangxi Agricultural University, Key Laboratory of Poyang Lake Watershed Agricultural Resources and Ecology of Jiangxi Province, Nanchang, Jaingxi 330045, China
Abstract:
[Objective] The main factors affecting soil erosion in upper reaches of the Ganjiang River basin were analyzed, in order to provide a reference for local soil erosion control and scientific management.[Methods] The data involved in this paper including Landsat 8 remote sensing image in 2015, MODIS NDVI, digital elevation model (DEM), soil type and rainfall. The RUSLE model and the random forest algorithm were used to quantitatively analyze soil erosion and its influencing factors in the upper reaches of the Ganjiang River basin.[Results] In 2015, the soil erosion intensity was gradually increased from southeast to northwest in the upper reaches of the Ganjiang River basin, and the soil was at a mild erosion level in general. The total amount of the soil erosion was 3.45×107 t/a. The average soil erosion modulus was 1 046.38 t/(km2·a), which was about two times higher than the allowable amount of soil erosion 500 t/(km2·a) in the red soil hilly region of Southern China. The average soil erosion moduli of sub-basins 9, 11 and 15 were 1 672.66, 1 715.83 and 1 565.36 t/(km2·a) respectively, which were at a moderate erosion level. These sub-basins were the key regions which need to be prevented and controlled in the study area. The rest of the sub-basins were under a mild erosion level.[Conclusion] Soil erosion in every sub-basin was greatly affected by the vegetation cover and management factor (C) and the slope length and slope (LS), and the importance of which was 30% and 20% respectively. The importance of soil erodibility factor (K) and the rainfall erosive force factor (R) were less than 10%. Among all sub-basins, sub-basin 9, 11, and 21 were mainly affected by LS factor, and the rest were mainly controlled by C factor.
Key words:  soil erosion  RUSLE model  random forest  influence factor  upper reaches of Ganjiang River basin