摘要
[目的]为科学判别流域尺度土壤侵蚀类型并给出相应的发生概率。[方法]构建基于深度学习(deep learning,DL)的松花江流域土壤侵蚀模数(erosion modulu,EM)计算模型,并计算不同类型的侵蚀模数。以降雨、气温、风速3个侵蚀模数影响因子为随机变量,利用数值模拟和高斯核密度估计法(gaussian kernel density estimation,GKDE)构建EM概率评价方法,给出不同土壤侵蚀强度组合的发生概率。[结果]EM计算模型验证期的R^(2)均>0.86。流域内平均每年有74.47%发生微度水蚀与微度风蚀;12.86%的面积发生轻度及以上水蚀与微度风蚀;12.56%的面积发生轻度及以上风蚀与微度水蚀;0.11%的面积水蚀强度与风蚀强度均在轻度及以上。36个典型像元中,发生微度水蚀与微度风蚀的平均概率为57.45%;发生微度水蚀与轻度风蚀的平均概率为30.26%;发生微度水蚀与中度风蚀的平均概率为8.03%;发生轻度水蚀与微度风蚀的平均概率为2.11%;发生微度水蚀与重度风蚀的平均概率为2.08%;发生其余强度组合的平均概率在0.05%以下。[结论]构建的松花江流域EM计算模型精度较高,揭示了松花江流域土壤侵蚀类型空间分布特征,并给出不同土壤侵蚀强度组合的发生概率,为松花江流域土壤侵蚀治理提供依据。
[Objective]To scientifically identify the types of soil erosion at the watershed scale and give the corresponding probability of occurrence.[Methods]This study constructs a deep learning(DL)-based model for calculating soil erosion modulus in the Songhua River Basin and calculates different types of soil erosion modulus.Using three erosion modulus influencing factors,namely rainfall,air temperature and wind speed,as random variables,numerical simulation and Gaussian Kernel Density Estimation(GKDE)were used to construct the EM probability evaluation method,which gives the probability of occurrence of different combinations of soil erosion intensities.[Results]The R^(2)of the validation period of the EM computational models were all>0.86;74.47%of the average annual occurrence of slight water erosion and slight wind erosion in the watershed;12.86%of the area of slight and above water erosion and slight wind erosion;12.56%of the area of slight and above wind erosion and slight water erosion;0.11%of the area of water erosion strength and wind erosion intensity are both slight and above.36 typical image elements of the 36 typical images,the average probability of occurrence of slight water erosion and slight wind erosion is 57.45%;the average probability of occurrence of slight water erosion and mild wind erosion is 30.26%;the average probability of occurrence of slight water erosion and moderate wind erosion is 8.03%;the average probability of occurrence of mild water erosion and slight wind erosion is 2.11%;the average probability of occurrence of slight water erosion and severe wind erosion is 2.08%;and the evaluations of occurrence of the remaining combinations of probabilities were all below 0.05%.[Conclusion]The calculation model of erosion modulus in the Songhua River basin constructed in this study has high accuracy,reveals the spatial distribution characteristics of soil erosion types in the Songhua River basin,and gives the probability of occurrence of different combinations of the intensity of the two types of erosion,which provides a basis for the management of soil erosion in the Songhua River basin.
作者
邢贞相
王嘉麒
张鸿雪
宋健
王轶男
段维义
宫铭
黄昌丽
XING Zhenxiang;WANG Jiaqi;ZHANG Hongxue;SONG Jian;WANG Yinan;DUAN Weiyi;GONG Ming;HUANG Changli(School of Water Conservancy and Civil Engineering,Northeast Agricultural University,Harbin 150030,China;Institute of Geographic Sciences and Natural Resources Research,Beijing 100101,China;Heilongjiang Qiqihar Ecological Environment Testing Center,Qiqihar,Heilongjiang 161005,China)
出处
《水土保持学报》
CSCD
北大核心
2024年第5期116-128,共13页
Journal of Soil and Water Conservation
基金
国家自然科学基金项目(51979038,52209008)。
关键词
松花江流域
土壤侵蚀
深度学习算法
高斯核密度估计法
概率评价
Songhua River basin
soil erosion
deep learning
gaussian kernel density estimation method
probability evaluation