Rapid assessment of soil aggregate stability (AS) is vital for a better understanding of the processes of soil aggregate breakdown, which is necessary for effective soil erosion control planning. This study explored the possibility of quantifying the mean weight diameter (MWD) of soil aggregates after applying the three disruptive treatments of fast wetting (FW), slow wetting (SW) and mechanical breakdown (MB), representing the dominant aggregate breakdown mechanisms, using a handheld mid-infrared (MIR) spectrophotometer on soils of the Belgian loam belt. Partial least squares regression (PLSR) was used to build calibration models for the three treatment methods, using a calibration set accounting for 70% (75% in the case of MB) of the samples, which were validated by using a validation set of 30% (25% in the case of MB) of the total samples. Results showed very good prediction accuracies with coefficient of determination (R2) values of 0.73, 0.85 and 0.79, root mean square error of prediction (RMSEP) values of 0.29, 0.46 and 0.26 mm, ratio of performance to interquartile range (RPIQ) values of 2.16, 4.98 and 3.64 and residual prediction deviation (RPD) values of 1.99, 2.60, and 2.22, for FW-MWD, SW-MWD, and MB-MWD, respectively. The best prediction was obtained for SW-MWD. The results of this study suggest that, for soils where soil organic carbon is a relevant indicator of aggregate strength, the three AS indices can be estimated to a high accuracy with spectra information in the MIR region of the electromagnetic spectrum.
Keywords: Aggregate stability; Mid infrared spectroscopy; Partial least squares regression; Soil erosion.
Copyright © 2020 Elsevier B.V. All rights reserved.