Hyperspectral Modeling of Soil Organic Matter Based on Characteristic Wavelength in East China

Abstract

Soil organic matter (SOM) is a key index of soil fertility. Visible and near-infrared (VNIR, 350–2500 nm) reflectance spectroscopy is an effective method for modeling SOM content. Characteristic wavelength screening and spectral transformation may improve the performance of SOM prediction. This study aimed to explore the optimal combination of characteristic wavelength selection and spectral transformation for hyperspectral modeling of SOM. A total of 219 topsoil (0–20 cm) samples were collected from two soil types in the East China. VNIR reflectance spectra were measured in the laboratory. Firstly, after spectral transformation (inverse-log reflectance (LR), continuum removal (CR) and first-order derivative reflectance (FDR)) of VNIR spectra, characteristic wavelengths were selected by competitive adaptive reweighted sampling (CARS) and uninformative variables elimination (UVE) algorithms. Secondly, the SOM prediction models were constructed based on the partial least squares regression (PLSR), random forest (RF) and support vector regression (SVR) methods using the full spectra and selected wavelengths, respectively. Finally, optimal SOM prediction models were selected for two soil types. The results were as follows: (1) The CARS algorithm screened 40–125 characteristic wavelengths from the full spectra. The UVE algorithm screened 105–884 characteristic wavelengths. (2) For two soil types and full spectra, CARS and UVE improved the SOM modeling precision based on the PLSR and SVR methods. The coefficient of determination (R2) value in the validation of the CARS-PLSR (PLSR model combined with CARS) and CARS-SVR (SVR model combined CARS) models ranged from 0.69 to 0.95, and the relative percent deviation (RPD) value ranged from 1.74 to 4.31. Lin’s concordance correlation coefficient (LCCC) values ranged from 0.83 to 0.97. The UVE-PLSR and UVE-SVR models showed moderate precision. (3) The PLSR and SVR modeling accuracies of Paddy soil were better than those for Shajiang black soil. RF models performed worse for both soil types, with the R2 values of validation ranging from 0.22 to 0.68 and RPD values ranging from 1.01 to 1.60. (4) For Paddy soil, the optimal SOM prediction models (highest R2 and RPD, lowest root mean square error (RMSE)) were CR-CARS-PLSR (R2 and RMSE: 0.97 and 1.21 g/kg in calibration sets, 0.95 and 1.72 g/kg in validation sets, RPD: 4.31) and CR-CARS-SVR (R2 and RMSE: 0.98 and 1.04 g/kg in calibration sets, 0.91 and 2.24 g/kg in validation sets, RPD: 3.37). For Shajiang black soil, the optimal SOM prediction models were LR-CARS-PLSR (R2 and RMSE: 0.95 and 0.93 g/kg in calibration sets, 0.86 and 1.44 g/kg in validation sets, RPD: 2.62) and FDR-CARS-SVR (R2 and RMSE: 0.99 and 0.45 g/kg in calibration sets, 0.83 and 1.58 g/kg in validation sets, RPD: 2.38). The results suggested that the CARS algorithm combined CR and FDR can significantly improve the modeling accuracy of SOM content.

Improved multivariate modeling for soil organic matter content estimation using hyperspectral indexes and characteristic bands

“…From the above results, some useful conclusions can be drawn: (1) for the same spectral transformation, RF models (for example, model CARS-R-RF) had lower accuracy, while the SVR and PLSR models (for example, model CARS-R-SVR and CARS-R-PLSR) have higher accuracy; (2) for the same modeling methods, SOM predicted models using R (model CARS-R-SVR) had lower accuracy, models using FDR and CR (model CARS-FDR-SVR and CARS-CR-SVR) had better accuracy. Overall, this study showed that CR and FDR transformation could improved modeling accuracy, and CARS algorithm could improve the SVR and PLSR modeling accuracy, that was consistent with most research results [ 13 , 17 , 18 , 34 ].…”

“…DOA was calculated using the reflectance of soil at 550, 600 and 650nm. SOM content was negatively correlated with DOA [ 24 , 27 , 45 ]. If the soil has rich SOM content, the reflectance spectral curve of soil shows a flat state in the range of 550–650 nm with low DOA value.…”

“…Zhao et al . [ 27 ] selected the optimal RI, DI and NDI calculated from continous VNIR spectra and establish the relatively accurate SOM inversion model. Compared with the predicted model using full spectral data, the modeling accuracy based on spectral index was acceptable, and the modeling variables were fewer.…”

“…CARS algorithm selects characteristics bands with high absolute regression coefficient values in the PLSR model [ 32 ]. Some studies reported that CARS could compressed the number of original spectral wavelengths to lower than 16% [ 30 , 34 ]. CARS algorithm was an effective way to reduce the number of inputs and improve the PLSR, random forest (RF) and support vector regression (SVR) modeling accuracy of SOM [ 28 , 34 , 35 ].…”

“…SVR and RF are capable to model complex, non-linear and linear relationships between variables [ 11 ]. Many studies found that SVR model had the robust and accurate prediction, and the RF model performed relatively weak in spectral modeling SOM [ 11 , 13 , 34 , 41 ]. However, other studies showed that RF model or RF model combined with CARS accurately predicted SOM [ 30 , 42 ].…”