An examination of soil texture and water content estimates from nuclear magnetic resonance
Description
TitleAn examination of soil texture and water content estimates from nuclear magnetic resonance
Date Created2021
Other Date2021-05 (degree)
Extent1 online resource (x, 104 pages)
DescriptionAn accurate description of soils is critical to agricultural practices because water and nutrients accessible to crops are stored in shallow soils at ~1-2 m depth. Traditional methods used for soil characterizations involve invasive, sparse, and time-consuming direct soil sampling techniques. Geophysical methods have shown promise to infer the physical properties of soils non-invasively and spatially. Nuclear magnetic resonance (NMR), the only geophysical method that directly measures the water content, has been increasingly used in soil science. This thesis focuses on exploring the use of NMR to characterize soil properties in agricultural practices.
The first laboratory study investigates the effect of clay, silt, and sand content in synthetic samples on NMR relaxation times. NMR measurements are collected on mixtures of 1%-60% kaolinite clay, 5%-85% glass beads, and 8%-94% quartz sand by mass. The effect of silt and sand content can be seen in soils with low clay content (<10%) but do not have an impact on soils with high clay content (>10%). The clay content dominates the NMR response for soils with >10% clay. The strong effect of the clay content on the NMR responses is due to the high specific surface area and the distribution of clay throughout the soils.
The second laboratory study explores the use of NMR as an alternative method to estimate soil texture of water-saturated soil samples collected in three agriculture sites in Moline, IL, USA. NMR presents a possible alternative method to determine the soil texture as other commonly used methods, such as textural analysis or laser diffraction, are time-consuming or need to be calibrated. To determine the soil texture using NMR, NMR transverse relaxation time (T2) distributions are separated into three regions using two T2 cutoff times, which are further used to estimate the clay, silt, and sand fractions. The NMR-derived soil texture is compared to that estimated by the sieve-pipette and laser diffraction method. NMR data with site-specific T2 cutoff times obtained from quantitative comparison with the sieve-pipette method offers a better estimation of soil texture than the laser diffraction method for soils with low magnetic susceptibility.
While the NMR method has been widely used to characterize saturated systems, its use in partially saturated systems (e.g., agriculture sites) is still being developed. The third study examines the use of a small-diameter NMR logging probe to estimate soil water content during two experiments in June and September 2016 at an agriculture site in central California. The NMR data can monitor the changes in water content from the two experiments at each borehole; however, at each experiment, NMR water content is lower than the water content derived by the reference time-domain reflectometry method because of the presence of soils with high magnetic susceptibility. A proposed calibration method, which used a linear model to account for the magnetic susceptibility, fail to accurately capture the total NMR estimated water content. Other mechanistic models are needed for more accurately calibrating.
NotePh.D.
NoteIncludes bibliographical references
Genretheses, ETD doctoral
LanguageEnglish
CollectionGraduate School - Newark Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.