A Look at vis-NIR and NIR Spectroscopy in Characterizing Soil Fertility

According to a recent study published in Soil Use and Management, visible near-infrared (vis-NIR) spectroscopy and NIR spectroscopy can be used to conduct in situ soil fertility assessments (1).

Soil fertility is important for the flourishing of crops. It is defined as the ability of soil to sustain and optimize crop yield (2). Farmers often enhance the fertility of their soil through the addition of inorganic and organic fertilizers (2). Traditional soil fertility assessments typically require extensive sampling and laboratory analysis, which can be time-consuming and expensive (1). However, vis-NIR spectroscopy is thought to be a potential alternative method for soil fertility analysis because of its ability to deliver high spatial and temporal resolution data (1).

Farmer hands hold soil earth sunset. Agriculture. Engineer checks soil fertility with argon. Business ricks employee land agro company. Farmer hands pouring earth sunset. Modern agro farm eco. Generated with AI. Image Credit: © Ibad - stock.adobe.com

Luca Bragazza of Agroscope and his team in Switzerland demonstrated in a recent study that vis-NIR spectroscopy can be used for this application. To test the feasibility of the technique, the researchers tested different methods of acquiring soil spectra, focusing on several parameters, including texture, pH, organic carbon content, cation exchange capacity (CEC), and major nutrients (1).

The research involved collecting soil samples from 134 sampling points using an Edelman auger to a depth of 20 cm (1). The research team collected the vis-NIR spectra from three positions: the raw soil surface; the cutaway side of the soil sample; and the cleaned and smoothed soil surface (1).

Regarding the instrumentation, the researchers used two portable spectrometers, a micro-electric mechanical systems (MEMS)-based spectrometer and a research grade vis-NIR spectrometer to evaluate the soil parameters (1). The researchers developed partial least squares regression (PLSR) calibration models for the selected soil parameters, evaluating model performance based on several metrics, including the ratio of performance to interquartile range (RPIQ), R2, root mean squared error (RMSE), and Lin's concordance correlation coefficient (CCC) (1).

Although the study showed that vis-NIR spectroscopy could predict some soil parameters accurately, both spectrometers used in the study struggled to accurately predict total and exchangeable calcium, potassium, phosphorus, and total magnesium (1).

Among the different scanning positions tested, the cutaway side of the soil core collected with the Edelman auger provided the best results. This method allowed for more accurate and consistent measurements, making it the recommended best practice for in-situ soil vis–NIR scans (1). The research-grade spectrometer generally delivered better performance indicators for most parameters, yet the MEMS-based spectrometer also provided satisfactory predictions, demonstrating its potential for field applications where portability and cost are critical considerations (1).

The ability to rapidly assess soil fertility in the field using vis–NIR spectroscopy has profound implications for agricultural management (3). Farmers and agronomists can obtain immediate feedback on soil health and properties, which provide them more information to better improve crop productivity (3). This technology can also aid in sustainable land management practices by allowing for continuous monitoring of soil conditions, reducing the need for excessive fertilizer use, and minimizing environmental impact (1).

Bragazza and his team suggest that further research should focus on improving the predictive capabilities of vis–NIR spectroscopy for the soil parameters that were not satisfactorily predicted in this study (1). Enhancing the calibration models and exploring additional spectral pretreatment methods could lead to more accurate and comprehensive soil fertility assessments.

References

(1) Metzger, K.; Liebisch, F.; Herrera, J. M.; et al. The Use of Visible and Near-Infrared Spectroscopy for In-Situ Characterization of Agricultural Soil Fertility: A Proposition of Best Practice by Comparing Scanning Positions and Spectrometers. Soil Use and Management 2024, 40 (1), e12952. DOI: 10.1111/sum.12952

(2) International Atomic Energy Agency, Improving Soil Fertility. IAEA.org. Available at: https://www.iaea.org/topics/improving-soil-fertility (accessed 2024-06-13).

(3) Recena, R.; Fernandez-Cabanas, V. M.; Delgado, A. Soil Fertility Assessment by Vis-NIR Spectroscopy: Predicting Soil Functioning Rather than Availability Indices. Geoderma 2019, 337, 368–374. DOI: 10.1016/j.geodermaa.2018.09.049