东篱科研大数据发现系统（DRDS）

位置：成果数据库 > 期刊 > 期刊详情页

间隔覆盖条件下塿土坡面侵蚀产沙模拟研究

ISSN号：1009-2242
期刊名称：《水土保持学报》
时间：0
分类：S157.1[农业科学—土壤学;农业科学—农业基础科学]
作者机构：[1]山东师范大学人口资源环境学院,济南250014, [2]鲁东大学地理与规划学院,山东烟台264025, [3]山东省水利科学研究院,济南250014
相关基金：国家自然科学基金项目（40801100,50179035,50609022）; 黄土高原土壤侵蚀与旱地农业国家重点实验室基金项目（10501-177,10501-220）

作者：崔素芳[1], 潘英华[2], 吴泉源[1], 张振华[2], 张保祥[3], 郝春红[2], 任尚岗[2]

关键词：间隔覆盖, 覆盖面积比, 临界坡度, 坡面产沙, BP神经网络, interval coverage, coverage area ratio, critical slope, sediment yield on slope, BP neural network

中文摘要：

针对西北旱农地区坡耕地常出现的水土流失及干旱缺水问题,采用间隔覆盖技术,利用室内模拟降雨试验,选取对坡面产沙影响的主要因素坡度、降雨强度、覆盖面积比,探讨各因素对坡面产沙的作用效应;利用BP神经网络模型和多元线性回归模型分别预测坡面产沙量。结果表明：坡度在8.74%-32.50%之间,累积产沙量随着坡度的增加而增大,但在36.40%时反而减少,说明在32.50%-36.40%之间存在一个临界坡度;降雨强度对侵蚀产沙的影响主要表现为,降雨强度越大,累积产沙量和泥沙浓度越大,产沙速率越快。从模拟精度来看,间隔覆盖法在降雨强度较小的情况下应用效果会更好;在降雨强度为60mm/h,坡度为17.60%的地区,最好采用50∶150的覆盖面积比;采用BP神经网络模型拟合产沙量的效果明显高于多元线性回归模型。

英文摘要：

Using of thin plastic film to cover slope surfaces could lead to slope runoff and soil erosion in dry-land area in northwest China.Interval coverage strategies were investigated using indoor artificial rainfall.Three main factors（slope,rainfall intensity,and coverage area ratio）were selected to analyze the effect of sediment yield for a Lou soil surface.BP neural network model and multiple linear regression model were used to predict the sediment yield and to obtain a theoretical foundation for practical application and evaluation.The results indicated that accumulated sediment yield increased with increasing slope during 8.74%~32.50%,but decreased in the slope of 36.40%,it showed that it existed a critical slope in 32.50% ~36.40%.The effect of rainfall intensity on sediment yield was mainly as follows：The higher the rainfall intensity,the accumulated sediment yield and sediment concentration were the greater,the rate of sediment was the faster.The simulation results demonstrated that an interval coverage strategy was better under low rainfall intensity.When the rainfall intensity was 60mm/h and slope was 17.60%in the area,the best use of coverage area ratio was 50∶150.It indicated that the effect of fitting the sediment yield of the BP neural network model was higher than that of the multiple linear regression model.

同期刊论文项目