In Situ VIS-NIR Spectroscopy for a Basic and Rapid Soil Investigation.

Visible and near-infrared (VIS-NIR) spectroscopy is extensively used in the field of soil science to predict several soil properties, mostly in laboratory conditions. When measured in situ, contact probes are used, and, very often, time-consuming methods are applied to generate better spectra. Unfortunately, spectra obtained by these methods differ greatly from spectra remotely acquired. This study tried to address this issue by measuring reflectance spectra directly with a fibre optic or a 4° lens on bare untouched soils. C, N content and soil texture (sand, silt, and clay) prediction models were established using partial least-square (PLS) and support vector machine (SVM) regression. With spectral pre-processing, some satisfactory models were obtained, i.e., for C content (R2 = 0.57; RMSE = 0.09%) and for N content (R2 = 0.53; RMSE = 0.02%). Some models were improved when using moisture and temperature as auxiliary data for the modelling. Maps of C, N and clay content generated with laboratory and predicted values were presented. Based on this study, VIS-NIR spectra acquired with bare fibre optic and/or a 4° lens could be used to build prediction models in order to obtain basic preliminary information on soil composition at the field scale. The predicting maps seem suitable for a fast but rough field screening.

Diffuse Reflectance Spectroscopy for Black Carbon Screening of Agricultural Soils under Industrial Anthropopressure.

Visible and near-infrared spectroscopy (VIS-NIRS) is a fast and simple method increasingly used in soil science. This study aimed to investigate VIS-NIRS applicability to predict soil black carbon (BC) content and the method's suitability for rapid BC-level screening. Forty-three soil samples were collected in an agricultural area remaining under strong industrial impact. Soil texture, pH, total nitrogen (Ntot) and total carbon (Ctot), soil organic carbon (SOC), soil organic matter (SOM), and BC were analyzed. Samples were divided into three classes according to BC content (low, medium, and high BC content) and scanned in the 350-2500 nm range. A support vector machine (SVM) was used to develop prediction models of soil properties. Partial least-square with SVM (PLS-SVM) was used to classify samples for screening purposes. Prediction models of soil properties were at best satisfactory (Ntot: R2 = 0.76, RMSECV = 0.59 g kg-1, RPIQ = 0.65), due to large kurtosis and data skewness. The RMSECV were large (16.86 g kg-1 for SOC), presumably due to the limited number of samples available and the wide data spread. Given our results, the VIS-NIRS method seems efficient for classifying soil samples from an industrialized area according to BC content level (training accuracy of 77% and validation accuracy of 81%).

Screening Risk Assessment of Agricultural Areas under a High Level of Anthropopressure Based on Chemical Indexes and VIS-NIR Spectroscopy.

Intensive anthropogenic activity may result in uncontrolled release of various pollutants that ultimately accumulate in soils and may adversely affect ecosystems and human health. Hazard screening, prioritisation and subsequent risk assessment are usually performed on a chemical-by-chemical basis and need expensive and time-consuming methods. Therefore, there is a need to look for fast and reliable methods of risk assessment and contamination prediction in soils. One promising technique in this regard is visible and near infrared (VIS-NIR) spectroscopy. The aim of the study was to evaluate potential environmental risk in soils subjected to high level of anthropopressure using VIS-NIR spectroscopy and to calculate several risk indexes for both individual polycyclic aromatic hydrocarbons (PAHs) and their mixture. Results showed that regarding 16PAH concentration, 78% of soil samples were contaminated. Risk assessment using the most conservative approach based on hazard quotients (HQ) for 10 individual PAHs allowed to conclude that 62% of the study area needs further action. Application of concentration addition or response addition models for 16PAHs mixture gave a more realistic assessment and indicates unacceptable risk in 23% and 55% of soils according to toxic units (TUm) and toxic pressure (TPm) approach. Toxic equivalency quotients (TEQ) were below the safe limit for human health protection in 88% of samples from study region. We present here the first attempt at predicting risk indexes using VIS-NIR spectroscopy. The best results were obtained with binary models. The accuracy of binary model can be ordered as follows: TPm (71.6%) < HI (85.1%) < TUm (87.9%) and TEQ (94.6%). Both chemical indexes and VIS-NIR can be successfully applied for first-tier risk assessment.