Soil organic matter (SOM) and pH are essential soil fertility indictors of paddy soil in the middle-lower Yangtze Plain. Rapid, non-destructive and accurate determination of SOM and pH is vital to preventing soil degradation caused by inappropriate land management practices. Visible-near infrared (vis-NIR) spectroscopy with multivariate calibration can be used to effectively estimate soil properties. In this study, 523 soil samples were collected from paddy fields in the Yangtze Plain, China. Four machine learning approaches—partial least squares regression (PLSR), least squares-support vector machines (LS-SVM), extreme learning machines (ELM) and the Cubist regression model (Cubist)—were used to compare the prediction accuracy based on vis-NIR full bands and bands reduced using the genetic algorithm (GA). The coefficient of determination (R²), root mean square error (RMSE), and ratio of performance to inter-quartile distance (RPIQ) were used to assess the prediction accuracy. The ELM with GA reduced bands was the best model for SOM (SOM: R² = 0.81, RMSE = 5.17, RPIQ = 2.87) and pH (R² = 0.76, RMSE = 0.43, RPIQ = 2.15). The performance of the LS-SVM for pH prediction did not differ significantly between the model with GA (R² = 0.75, RMSE = 0.44, RPIQ = 2.08) and without GA (R² = 0.74, RMSE = 0.45, RPIQ = 2.07). Although a slight increase was observed when ELM were used for prediction of SOM and pH using reduced bands (SOM: R² = 0.81, RMSE = 5.17, RPIQ = 2.87; pH: R² = 0.76, RMSE = 0.43, RPIQ = 2.15) compared with full bands (R² = 0.81, RMSE = 5.18, RPIQ = 2.83; pH: R² = 0.76, RMSE = 0.45, RPIQ = 2.07), the number of wavelengths was greatly reduced (SOM: 201 to 44; pH: 201 to 32). Thus, the ELM coupled with reduced bands by GA is recommended for prediction of properties of paddy soil (SOM and pH) in the middle-lower Yangtze Plain.

As a precious soil resource, black soils in Northeast China are currently facing severe land degradation. Visible and near-infrared spectroscopy (vis-NIR, 350–2500 nm) and mid-infrared spectroscopy (MIR, 2500–25,000 nm) have shown great potential to predict soil properties. However, there is still limited research on using MIR in situ. The aim of this study was to explore the feasibility of in situ MIR for the prediction of soil total nitrogen (TN) and total phosphorus (TP) and to compare its performance with the use of laboratory MIR, as well as the use of in situ and laboratory vis-NIR. A total of 450 samples from 90 soil profiles, along with their in situ and laboratory spectra of MIR and vis-NIR, were collected in a field with ten different tillage and management practices in a typical black soil area of Northeast China. Partial least square regression (PLSR), random forest (RF) and multivariate adaptive regression splines (MARS) were used to generate the calibrations between the spectra and the two properties. The results showed that both MIR and vis-NIR were able to predict the TN whether in laboratory or in situ conditions, but neither of them could predict the TP quantitatively since there was no sensitive band on both spectra regarding the TP. The prediction accuracy of the TN with laboratory spectra was higher than that with in situ spectra, for both vis-NIR and MIR. The optimal prediction accuracy of the TN with laboratory MIR (RMSE = 0.11 g/kg, RPD = 3.12) was higher than that of laboratory vis-NIR (RMSE = 0.14 g/kg, RPD = 2.45). The optimal prediction accuracy of in situ MIR (RMSE = 0.20 g/kg, RPD = 1.80) was lower than that of in situ vis-NIR (RMSE = 0.16 g/kg, RPD = 2.14). The prediction performance of the spectra followed laboratory MIR > laboratory vis-NIR > in situ vis-NIR > in situ MIR. The performance of in situ MIR was relatively poor, mainly due to the fact that MIR was more influenced by soil moisture. This study verified the feasibility of in situ MIR for soil property prediction and provided an approach for obtaining rapid soil information and a reference for soil research and management in black soil areas.

Robust soil organic matter (SOM) mapping is required by farms, but their generation requires a large number of samples to be chemically analyzed, which is cost prohibitive. Recently, research has shown that visible and near-infrared (vis-NIR) reflectance spectroscopy is a fast and accurate technique for estimating SOM in a cost-effective manner. However, few studies have focused on using vis-NIR spectroscopy as a covariate to improve the accuracy of spatial modeling. In this study, our objective was to compare the mapping accuracy from a spatial model using kriging methods with and without the covariate of vis-NIR spectroscopy. We split the 261 samples into a calibration set (104) for building the spectral predictive model, a test set for generating the vis-NIR augmented set from the prediction of the fitted spectral predictive model (131), and a validation set (26) for evaluating map accuracy. We used two datasets (235 samples) for Kriging: a laboratory-based dataset (Ld, observations from calibration and test datasets) and a laboratory-based dataset with vis-NIR augmented predictions (Au.p, observations from calibration and predictions from test dataset), a laboratory-based dataset with vis-NIR spectra as the covariance (Ld.co) and augmented dataset with predictions using vis-NIR with vis-NIR spectra for the covariance (Au.p.co). The first one to seven accumulated principal components of vis-NIR spectra were used as the covariates when we used the measurement of Ld.co and Au.p.co. The map accuracy was evaluated by the validation set for the four datasets using Kriging. The results indicated that adding vis-NIR spectra as covariates had great potential in improving the map accuracy using kriging, and much higher accuracies were observed for Ld.p.co (RMSE of 5.51 g kg⁻¹) and Au.p.co (RMSE of 5.66 g kg⁻¹) than without using vis-NIR spectra as covariates for Ld (RMSE of 7.12 g kg⁻¹) and Au.p (RMSE of 7.69 g kg⁻¹). With a similar model performance to Ld.p.co, Au.p.co can reduce the cost of laboratory analysis for 60% of soil samples, demonstrating its advantage in cost-efficiency for spatial modeling of soil information. Therefore, we conclude that vis-NIR spectra can be used as a cost-effective technique to obtain augmented data to improve fine-resolution spatial mapping of soil information.