基于Sentinel-2A遥感影像土壤有机质含量的反演研究  

Inversion of Soil Organic Matter Content Based on Sentinel-2A Remote Sensing Image

作　　者：马驰[1] MA Chi(Liaoning College of Communication,Shenyang,Liaoning 110122)

机构地区：[1]辽宁省交通高等专科学校,辽宁沈阳110122

出　　处：《北方园艺》2020年第2期94-100,共7页Northern Horticulture

基　　金：国家自然科学基金资助项目(41371332);中国地质调查局资助项目(1212010911084);辽宁省交通高等专科学校资助项目(lnccybky201910)。

摘　　要：利用Sentinel-2A遥感影像为数据源,结合德惠市、农安县土壤实地采样的有机质含量实验室化验值,反演研究区表层土壤有机质含量。首先,选取10个分辨率为10 m(20 m)的Sentinel-2A可见光、近红外及短波红外波段,并将遥感影像进行辐射校正与大气校正,消除大气对传感器成像的影响;然后,将各波段反射率与土壤有机质含量进行相关性分析,获得有机质敏感波段;最后,利用多元回归分析的方法建立研究区土壤有机质含量的反演模型,反演研究区表层土壤有机质含量。结果表明:Sentinel-2A遥感影像在可见光与近红外波段与土壤有机质含量具有良好的相关性,且在第6波段达到峰值,为r=-0.817;将反射率进行适当的数学变换后可以有效改善与有机质含量的相关性,其中,指数变换与有机质含量的相关性较好,相关系数为r=-0.867;利用多元回归分析方法建立的研究区土壤有机质含量反演模型SOM=26.31/R2-117.56eR4-9.74R8-1.48+145.89具有较高的模型精度与较好的模型稳定性,模型的决定系数达到R2=0.917,均方根误差为5.86 g·kg-1。In this study,we utilized the remote sensing image of Sentinel-2 A and combine the laboratory test value of the organic matter content in soil samples from Dehui city and Nong′an County to retrieved the soil organic matter content in the surface layer of the area.Firstly,select ten Sentinel-2 A visible,near-infrared and short-wave infrared bands with a resolution of 10 m(20 m),and carried out radiation calibration and atmospheric calibration of the remote sensing images to eliminate the influence of the atmosphere on sensor imaging.Then,obtained the sensitive bands of organic matter through analyzing the correlation of the reflectance of each wave band and soil organic matter content.Finally,the inversion model of soil organic matter content was established by multiple regression analysis to research the soil organic matter content in the surface layer of the area.The results showed that the Sentinel-2 A remote sensing images had a good correlation with the soil organic matter content in the visible and near-infrared bands,and reached a peak value of r=-0.817 in the sixth band.The correlation between the reflectivity and the organic matter content could be effectively improved after the appropriate mathematical transformation of the reflectivity.In addition,it was found that the excellent correlation was exhibited between the exponential transformation and the organic matter content,in which the correlation coefficient was r=-0.867.Moreover,the inversion model SOM=26.31/R2-117.56 eR4-9.74 R8-1.48+145.89 of soil organic matter content in the study area established by multiple regression analysis method possess the advantage of high accuracy and better stability,in which determination coefficient was up to R2=0.917,and the root mean square error was 5.86 g·kg-1.

关 键 词：Sentinel-2A 遥感 有机质 反演 回归分析 

分 类 号：S153.6[农业科学—土壤学] TP79[农业科学—农业基础科学]