Spectroscopy offers new opportunities for soil analysis

A research from the University of Katowice opens new perspectives in the field of soil analysis. Through spectroscopy, researchers were able to determine the presence and concentration of key elements in soil

by Matteo Cavallito

 

Soil analysis, a crucial strategy for proper planning of agricultural planning strategies, does not have to go through time-consuming laboratory surveys. In fact, a quicker but no less effective technique can be used to obtain reliable soil profiling: near-infrared reflectance spectroscopy. This is suggested by a research from the University of Silesia in Katowice, Poland, published in the journal Talanta.

Spectroscopy is a technique for delineating the chemical composition of a substance from analysis of the variation in the spectrum of light passing through it. The investigation led to the application of this method on samples of a soil type classified as Haplic Luvisol. This is described by the European Soil Atlas as “a well-developed soil whose main pedological characteristic is the formation of the argic sub surface horizon through the destruction and movement (eluviation and translocation) of silicate clay from the surface topsoil layer.”

The study

Using also partial regression statistical models, researchers were able to predict the concentrations of cadmium, copper, lead, nickel, chromium, zinc, manganese and iron in soil samples. “A total of 234 soil samples were investigated, and their reflectance spectra were recorded in the spectral range of 1100–2500 nm,” the study states.

Further, “The optimal spectral preprocessing was selected among 56 different scenarios considering the root mean squared error of prediction (RMSEP). The partial robust M-regression method (PRM) was used to handle the outlying samples.”

PRM is a statistical method that complements others such as partial least squares regression (PLSR) and, according to the magazine Spectroscopy – is particularly useful in weighting so-called outlier samples that can adversely affect the performance of other models.

Results

The study provided reliable detection of the presence of the elements. “The results of this study indicated that the spectra of Haplic Luvisol soil samples, collected between 1100 and 2500 nm, contained much relevant information which can be used for monitoring Cd, Cu, Pb, Ni, Cr, Zn, Mn, and Fe at low concentrations.”

In particular, the authors say, “tThe most promising models were obtained for estimating the amount of Cu (using PRM) and Pb (using the classic Partial Least Squares (PLS)), leading to RMSEP expressed as a percentage of the response range, equal to 9.63% and 11.5%, respectively.” In addition, “favorable PLS models were also built for Zn, Mn, and Fe.”

New frontiers for soil analysis

The research thus opens up new perspectives in the promising field of soil analysis and, in particular, the measurement of concentrations of its key elements. Indeed, data-driven detection techniques – the basis of so-called precision agriculture – make it possible to monitor soil composition in real time, the US journal further says. Especially using remote monitoring technologies like drones, aircraft, balloons or satellites.

Significantly, the scientists finally remark, the analytical methods used in the study proved effective on a type of soil that is widespread in the Planet. “Considering the significant representation of Haplic Luvisol soils worldwide (ca. 5% of the total area) and in Poland (up to 44%) and their arable potential,” the study states, “the results are helpful in the international context.”