Visible and near-infrared (VIS–NIR) spectroscopy has been extensively applied to estimate soil organic matter (SOM) in the laboratory. However, if field/moist VIS–NIR spectra can be directly applied to estimate SOM, then much of the time and labor would be avoided. Spectral derivative plays an important role in eliminating unwanted interference and optimizing the estimation model. Nonetheless, the conventional integer order derivatives (i.e., the first and second derivatives) may neglect some detailed information related to SOM. Besides, the full-spectrum generally contains redundant spectral variables, which would affect the model accuracy. This study aimed to investigate different combinations of fractional order derivative (FOD) and spectral variable selection techniques (i.e., competitive adaptive reweighted sampling (CARS), elastic net (ENET) and genetic algorithm (GA)) to optimize the VIS–NIR spectral model of moist soil. Ninety-one soil samples were collected from Central China, with their SOM contents and reflectance spectra measured. Support vector machine (SVM) was applied to estimate SOM. Results indicated that moist spectra differed greatly from dried ground spectra. With increasing order of derivative, the spectral resolution improved gradually, but the spectral strength decreased simultaneously. FOD could provide a better tool to counterbalance the contradiction between spectral resolution and spectral strength. In full-spectrum SVM models, the most accurate estimation was achieved by SVM model based on 1.5-order derivative spectra, with validation R² = 0.79 and ratio of the performance to deviation (RPD) = 2.20. Of all models studied (different combinations of FOD and variable selection techniques), the highest validation model accuracy for SOM was achieved when applying 1.5 derivative spectra and GA method (validation R² = 0.88 and RPD = 2.89). Among the three variable selection techniques, overall, the GA method yielded the optimal predictability. However, due to its long computation time, one alternative was to use CARS method. The results of this study confirm that a suitable combination of FOD and variable selection can effectively improve the model performance of SOM in moist soil.

Soil organic matter (SOM) is an important parameter of soil fertility, and visible and near-infrared (VIS–NIR) spectroscopy combined with multivariate modeling techniques have provided new possibilities to estimate SOM. However, the spectral signal is strongly influenced by soil moisture (SM) in the field. Interest in using spectral classification to predict soils in the moist conditions to minimize the influence of SM is growing. The objective of this study was to investigate the transferability of two approaches, SM–based cluster method with known SM (classifying the VIS–NIR spectra into different SM clusters to develop models separately), the normalized soil moisture index (NSMI)–based cluster method with unknown SM (utilizing NSMI to indicate the SM and establish models separately), to predict SOM directly in moist soil spectra. One hundred and twenty one soil samples were collected from Central China, and eight SM levels were obtained for each sample through rewetting experiments. Their reflectance spectra and SOM concentrations were measured in the laboratory. Partial least square-support vector machine (PLS-SVM) was employed to construct SOM prediction models. Specifically, prediction models were developed for NSMI–based clusters with unknown SM data. The models were assessed through three statistics in the processes of calibration and validation: the coefficient of determination (R²), root mean square error (RMSE) and the ratio of the performance to deviation (RPD). Results showed that the variable SM led to reduced VIS–NIR reflectance nonlinearly across the entire spectral range. NSMI was an effective spectral index to indicate the SM. Classifying the VIS–NIR spectra into different SM clusters in known SM states could improve the performance of PLS-SVM models to acceptable prediction accuracies (R²_cv = 0.69–0.77, RPD = 1.79–2.08). The estimation of SOM, when using the NSMI–based cluster method with unknown SM (RPD = 1.95–2.04), was similar to the use of the SM–based cluster method with known SM (RPD = 1.79–2.08). The predictive results (RPD = 1.87–2.06) demonstrated that the NSMI-–based cluster method has potential for application outside the laboratory for SOM prediction without knowing the SM explicitly, and this method is also easy to carry out and only requires spectral information.