Evaluation of soil structural quality using VIS–NIR spectra

Abstract

Application of visible (VIS) and near-infrared (NIR) spectroscopy for prediction of soil properties may offer a cost and time effective approach for evaluation of soil structural quality. Spectral data are often suitable for estimation of biochemical soil quality indicators such as soil organic carbon (SOC), total nitrogen and microbial biomass, while contradictory results have been reported for prediction of soil physical properties that are directly associated with soil structure. The aims of this study were to relate soil structural quality to overall indicators of soil quality, and to assess the efficiency of spectral data for the evaluation of soil structural quality. The study was conducted using 40 sites in Ireland under arable (n=20) and grassland (n=20) management systems. At each site five subplots were selected for soil sampling and twenty-one chemical, biological and physical properties were measured using standard methods as indicators of soil quality. The visual evaluation of soil structure (VESS) was performed to evaluate and classify soil structural quality. Soil properties that were significantly different (P<0.05) between soil structural quality classes were considered for further analysis, and principal component analysis was used to determine the key indicators as a minimum data set (MDS). VIS and NIR spectra were then measured and partial least-squares regression used to predict soil quality indicators associated with soil structural quality. SOC, penetration resistance, magnesium (Mg), aggregate size distribution and CN ratio comprised the MDS. An excellent model was achieved for SOC (RPD>4, R2=0.94; RMSE=0.42). A good model was obtained for Mg, CN (RPD from 2 to 2.5, R2≥0.7), and moderate capability for prediction of aggregate size distribution, and penetration resistance (RPD from 1.5 to 1.99, R2≥0.64). Soil structural quality classes were found to be associated with a number of biochemical and physical soil properties. Soil spectra produced acceptable models for predicting relevant soil structural indicators, and the mean soil spectra were different between soil structural classes. Therefore, a combination of spectroscopic and chemometric techniques can be applied as a practical, rapid, low cost and quantitative approach for evaluating soil structural quality under arable and grassland management systems in Ireland.