Regional soil organic carbon prediction model based on a discrete wavelet analysis of hyperspectral satellite data

被引:163
作者
Meng, Xiangtian [1 ]
Bao, Yilin [1 ]
Liu, Jiangui [3 ]
Liu, Huanjun [1 ,2 ]
Zhang, Xinle [1 ]
Zhang, Yu [4 ]
Wang, Peng [4 ]
Tang, Haitao [1 ]
Kong, Fanchang [1 ]
机构
[1] Northeast Agr Univ, Sch Publ Adminstrat & Law, Harbin 150030, Peoples R China
[2] Chinese Acad Sci, Northeast Inst Geog & Agroecol, Changchun 130012, Peoples R China
[3] Agr & Agri Food Canada, Eastern Cereal & Oilseed Res Ctr, 960 Carling Ave, Ottawa, ON K1A 0C6, Canada
[4] Heilongjiang Prov Natl Def Sci & Technol Inst, Harbin 150030, Peoples R China
基金
国家重点研发计划; 中国国家自然科学基金;
关键词
Soil organic carbon; Hyperspectral satellite data; Discrete wavelet analysis; Spectral index; Mapping; REFLECTANCE SPECTROSCOPY; MATTER CONTENT; NEURAL-NETWORKS; IMAGE-ANALYSIS; MOISTURE; NITROGEN; VNIR; REGRESSION; SPECTRA; SYSTEM;
D O I
10.1016/j.jag.2020.102111
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
Most studies have the achieved rapid and accurate determination of soil organic carbon (SOC) using laboratory spectroscopy; however, it remains difficult to map the spatial distribution of SOC. To predict and map SOC at a regional scale, we obtained fourteen hyperspectral images from the Gaofen-5 (GF-5) satellite and decomposed and reconstructed the original reflectance (OR) and the first derivative reflectance (FDR) using discrete wavelet transform (DWT) at different scales. At these different scales, as inputs, we selected the 3 optimal bands with the highest weight coefficient using principal component analysis and chose the normalized difference index (NDI), ratio index (RI) and difference index (DI) with the strongest correlation with the SOC content using a contour map method. These inputs were then used to build regional-scale SOC prediction models using random forest (RF), support vector machine (SVM) and back-propagation neural network (BPNN) algorithms. The results indicated that: 1) at a low decomposition scale, DWT can effectively eliminate the noise in satellite hyperspectral data, and the FDR combined with DWT can improve the SOC prediction accuracy significantly; 2) the method of selecting inputs using principal component analysis and a contour map can eliminate the redundancy of hyperspectral data while retaining the physical meaning of the inputs. For the model with the highest prediction accuracy, the inputs were all derived from the wavelength range of SOC variations; 3) the differences in prediction accuracy among the different prediction models are small; and 4) the SOC prediction accuracy using hyperspectral satellite data is greatly improved compared with that of previous SOC prediction studies using multispectral satellite data. This study provides a highly robust and accurate method for predicting and mapping regional SOC contents.
引用
收藏
页数:15
相关论文
共 71 条
[1]   Improving within-genus tree species discrimination using the discrete wavelet transform applied to airborne hyperspectral data [J].
Banskota, Asim ;
Wynne, Randolph H. ;
Kayastha, Nilam .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2011, 32 (13) :3551-3563
[2]  
Batchelor WD, 1997, T ASAE, V40, P247, DOI 10.13031/2013.21237
[3]   Using Imaging Spectroscopy to study soil properties [J].
Ben-Dor, E. ;
Chabrillat, S. ;
Dematte, J. A. M. ;
Taylor, G. R. ;
Hill, J. ;
Whiting, M. L. ;
Sommer, S. .
REMOTE SENSING OF ENVIRONMENT, 2009, 113 :S38-S55
[4]   Scale-location specific soil spatial variability: A comparison of continuous wavelet transform and Hilbert-Huang transform [J].
Biswas, Asim .
CATENA, 2018, 160 :24-31
[5]   Wavelet decomposition of hyperspectral data: a novel approach to quantifying pigment concentrations in vegetation [J].
Blackburn, G. A. .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2007, 28 (12) :2831-2855
[6]   Retrieval of chlorophyll concentration from leaf reflectance spectra using wavelet analysis [J].
Blackburn, George Alan ;
Ferwerda, Jelle Garke .
REMOTE SENSING OF ENVIRONMENT, 2008, 112 (04) :1614-1632
[7]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[8]   Global soil characterization with VNIR diffuse reflectance spectroscopy [J].
Brown, David J. ;
Shepherd, Keith D. ;
Walsh, Markus G. ;
Mays, M. Dewayne ;
Reinsch, Thomas G. .
GEODERMA, 2006, 132 (3-4) :273-290
[9]   A high resolution map of soil types and physical properties for Cyprus: A digital soil mapping optimization [J].
Camera, Corrado ;
Zomeni, Zomenia ;
Noller, Jay S. ;
Zissimos, Andreas M. ;
Christoforou, Irene C. ;
Bruggeman, Adriana .
GEODERMA, 2017, 285 :35-49
[10]   Geophysical and Hyperspectral Data Fusion Techniques for In-Field Estimation of Soil Properties [J].
Casa, Raffaele ;
Castaldi, Fabio ;
Pascucci, Simone ;
Basso, Bruno ;
Pignatti, Stefano .
VADOSE ZONE JOURNAL, 2013, 12 (04)