Estimation of Clay and Soil Organic Carbon Using Visible and Near‐Infrared Spectroscopy and Unground Samples

Abstract

Core Ideas The spectra of unground samples can be used to predict soil organic C (SOC) and clay. Vis‐NIR spectroscopy models for ground samples can predict SOC and clay of unground samples. Using unground samples will save time and labor when estimating SOC and clay. Visible and near‐infrared spectroscopy (Vis‐NIR) can accurately predict soil organic C (SOC) and clay from the spectra of air‐dried and ground (DG) samples. However, grinding generally requires a lot of time and labor. In this study, air‐dried and unground (DU) samples were used to exploit the time and accuracy advantages of Vis‐NIR. The reflectance of 117 samples using DG and DU pretreatments was measured in the laboratory. Five spectral pretreatments were used: no‐pretreatment (NP), Savitzky–Golay (SG) smoothing, first derivative (FD), standard normal variate transformation, and multiplicative scatter correction (MSC). When calibrations and validations used the same sample pretreatment (DG or DU), good predictions (R2 > 0.75; RPD > 2.0) could be obtained for both SOC and clay. There were no sample × spectral pretreatments interaction effects on the prediction accuracy, whereas the main effects of the pretreatments were significant for clay but not for SOC. However, when calibrations derived from the DG samples were applied to the DU samples, the prediction accuracy decreased compared with that of the DG samples regardless of the SOC and clay. In this case, good predictions could be acquired for clay using the FD spectra and satisfactory predictions for SOC (0.60 < R2 < 0.75; 1.4 < RPD < 2.0) were obtained from the NP, SG, and MSC spectra. These results indicate that it is viable to predict SOC and clay content using DU samples with acceptable accuracy while saving a lot of time and labor compared with that needed for DG samples.

Prediction of multiple soil fertility parameters using VisNIR spectroscopy and PXRF spectrometry

“…In total, 117 soil samples and their VisNIR reflectance spectra were acquired by Wang and Pan (2016). The soil samples were collected at a depth of 0–20 cm from agricultural fields just after the corn harvest in Fengqiu County near the Yellow River in Henan, China (Figure 1).…”

“…Furthermore, to remove random noise from the spectra, Savitzky–Golay (SG) smoothing with a window size of 9 was used. Due to the good performance of the SG spectra in predicting the SOC (Wang & Pan, 2016), no other pretreatments were applied.…”

“…Twenty spectra (four directions by rotating the sample tray 90˚, each with five spectra) were averaged as the soil spectrum for each sample. For more details on the soil samples and VisNIR spectral measurements, refer to the paper by Wang and Pan (2016).…”

“…Moreover, to overcome extreme results from one single random split, 20 random splits were performed, resulting in 20 calibration datasets and 20 validation datasets. Calibration and validation were performed on each group of calibration and validation datasets, and the mean values of the performance parameters were used to evaluate the prediction accuracy of the soil properties (Wang & Pan, 2016). Partial least squares regression and RF were applied to calibrate the soil properties and the VisNIR and PXRF data.…”

“…Whether these available nutrients can be predicted by VisNIR spectroscopy and PXRF has yet to be studied. Furthermore, whether all multiple soil fertility parameters can be estimated by VisNIR spectroscopy and PXRF is still unknown because researchers rarely study so many soil fertility parameters in one study, and most focus on only some of the parameters (Andrade, Faria, et al., 2020; Wang & Pan, 2016; Wang et al., 2015). In this study, multiple soil fertility parameters—namely, SOC, available N, total P, available P, total K and available K—were studied systematically; since SOC and total N are significantly correlated (Deng et al., 2020; Wang et al., 2015), total N was not studied redundantly.…”