基于遥感技术对松花江悬浮泥沙浓度的分析  

Analysis of suspended sediment concentration in Songhua River based on remote sensing technology

作　　者：韩旭 杜崇[1] 刘福全 李瑞 HAN Xu;DU Chong;LIU Fu-Quan;LI Rui(School of Hydraulic and Electric Power,Heilongjiang University,Harbin 150080,China)

机构地区：[1]黑龙江大学水利电力学院,哈尔滨150080

出　　处：《黑龙江大学工程学报》2022年第1期18-24,共7页Journal of Engineering of Heilongjiang University

基　　金：黑龙江省高校基本科研业务费科技创新项目(2018-KYYWF-1570)。

摘　　要：针对目前悬浮泥沙的遥感反演多以低分辨率卫星为主的现状,采用经过重采样后分辨率可达到10 m的Sentinel-2卫星,对松花江哈尔滨段大顶子山水利枢纽上游丰水期和枯水期悬浮泥沙含量进行反演对比研究。选取了B2、B3、B4、B8作为反演模型构建敏感波段,并对这些波段进行比值处理,构建反演模型。结果表明,模型精度较高,在枯水期,平均相对误差为17.42%,均方根误差为0.034 g·L^(-1);在丰水期,平均相对误差为16.02%,均方根误差为0.016 g·L^(-1)。依据构建的反演模型,绘制了水体悬浮泥沙分布图,分析了悬浮泥沙的分布特征和随时间变化的趋势。研究发现,研究区域的悬浮泥沙浓度分布总体是由河中心向岸边逐渐升高的趋势,且丰水期的悬浮泥沙含量比枯水期的悬浮泥沙含量高。The remote sensing inversion of suspended sediment is mainly based on low-resolution satellites.Sentinel-2 satellite can reach 10 m after re-sampling,was used to make a comparative study on the inversion of suspended sediment content in the upper reaches of DaDingZiShan water control project in Harbin section of Songhua River.B2,B3,B4 and B8 were selected as the sensitive bands for inversion model construction,and the ratio processing was performed to construct the inversion model for these bands.The results show that the model has high accuracy.In dry season,the average relative error is 17.42%,the root mean square error is 0.034 g·L^(-1),and in wet season,the average relative error is 16.02%.The mean square error is 0.016 g·L^(-1).According to the constructed inversion model,the distribution map of suspended sediment in water body is drawn,and the distribution characteristics of suspended sediment and its changing trend with time are analyzed.It is found that the concentration distribution of suspended sediment in the study area is gradually increasing from the river center to the shore,and the suspended sediment content in wet season is higher than that in dry season.

关 键 词：遥感技术 悬浮泥沙 松花江 

分 类 号：TV142.3[水利工程—水力学及河流动力学]