Mid-infrared (MIR) spectroscopy has emerged as a rapid measurement technique that has the potential to complement if not substitute laboratory analysis of soil properties. Mid-infrared spectroscopy has been successfully used for estimating dynamic soil properties (DSP): moisture, organic carbon, cation exchange capacity, electrical conductivity, and pH. However, there are scanty or no reports on hydrological soil properties (HSP): infiltration, soil hydraulic conductivity, water retention, and available water capacity. In addition, the available conventional methods of measuring hydrological soil properties are labor-intensive and time-consuming. The goal of this project is to enable Natural Resources Conservation Service (NRCS) field offices in Texas and Mississippi to utilize MIR spectroscopy to derive DSPs and HSPs in office without performing laborious and costly conventional field or laboratory measurements.

Spectroscopy is a rapid and non-destructive technique which we can rely on sensing multiple properties of soil simultaneously. Unique spectral signatures govern the accuracy and reliability and we have the utmost potential to use this technology to develop a “single sensor” to measure different constituents of the soil. Use of different regions of spectra: UV, VisNIR, and Mid infra-red, to estimate different properties have been demonstrated in literature for dry ground soils, mostly as distinct efforts utilizing a single spectral region at once.

Among the aforementioned spectral regions, VisNIR reflectance spectroscopy is the most widely investigated for soil characterization and the UV range is the least. MIR range produces accurate predictions compared to the other two, however several limitations are existing as barriers preventing bringing the technology to field-level applications. This project will enable the use of integrated UV-VisNIR-MIR spectroscopic techniques for in-situ soil sensor development. The research seeks to answer the question of ‘whether integrated UV-VisNIR-MIR spectroscopy can be used for field soil sensing? If so, how?’

Wetlands are a vital part of the ecosystem, providing a haven for unique flora and fauna as well as serving as a sink for floodwaters, nutrients, and pollutants. In the United States, actions affecting wetlands are regulated under the Clean Water Act. To confirm if an area is a wetland, a certified professional must observe the location for wetland indicators and make an assessment. This leads to the need for training to make a determination, and decisions may be biased due to subjectivity. Spectroscopy, particularly near and mid-infrared, has been widely used to characterize soils and has the potential to be used for determining wetland status through soil indicators. This project will investigate the accuracy of classification models built on soil spectra from the UV to the mid-infrared range with the goal of providing objective results on wetland status without the need for extensive training.

Spectroscopy is a rapidly developing technology that is currently permeating many scientific domains and applications. It has shown promise in diverse applications including soil chemical and physical property estimation, plant tissue nutrient measurement, and detection of plant diseases. However, the use of this technique to early detect viral infections of sweet potato is not established yet. The goal of this project is to implement spectroscopy-based techniques to identify spectral signatures that are sensitive to sweet potato potyviruses in greenhouse conditions enabling the detection of infections before the onset of visual symptoms.

Impact damage to sweet potatoes is assessed on research farms of MSU Pontotoc Experiment Station. An impact recording device (IRD) is inserted in the product flow and used to measure the impact. Survey experiments are conducted for selected commercial sweet potato packing lines in Houston and Vardaman of Mississippi to assess impacts imparted to sweet potato roots due to packing equipment. Characteristics (e.g., peak acceleration, velocity change) of the impacts produced on the packing lines from the entrance to the outlet will be measured using the IRD.

The use of near and mid-infrared spectroscopy for analyzing the chemical composition of plants, soils, and other materials is emerging as a cheap and convenient alternative to the traditional method of wet lab analysis. However, most forms of IR spectroscopy require samples to be finely ground, which is time-consuming, especially when analyzing large numbers of samples. While the amount of grinding needed for soils is well-documented, there is a lack of research on the grinding needed for plant materials. This research seeks to identify the particle size of various leaf samples that both optimizes spectral quality and the amount of time spent processing samples.

Published article: https://www.nature.com/articles/s41598-023-33558-5