Doctoral Dissertations

Keywords and Phrases

GPR; Machine Learning; Multispectral; Soil Properties; Thermal; UAV

Abstract

"Soil properties are critical for agricultural management. Geophysical techniques such as ground penetrating radar (GPR) and electromagnetic induction have been used to measure soil properties for agricultural applications. However, their measurements are only acquired along select traverses within a field, and the data processing can be time-consuming, the instruments can also be expensive. Unmanned aerial vehicles (UAVs) are a recent advancement used for precision agriculture. Compared with geophysical data, UAVs have higher resolution and wider coverage and can quickly collect data across the whole field. The operation is also relatively simpler with a more economical acquisition cost. This research investigates the potential for field-scale estimation of soil properties using UAV multispectral and thermal data at two study areas in Missouri, USA. Three estimated soil properties were soil water content (SWC) obtained using GPR, and electrical conductivity (EC) and magnetic susceptibility (MS) measured by electromagnetic induction. Various combinations of UAV-based data were utilized to predict soil properties using the random forest model. This research is the first to correlate geophysical measurements to UAV-based data. Results showed there are correlations between UAV data and estimated soil properties, especially in wetter soils where predictions were generally more accurate. Combining thermal data with multispectral data, including reflectance bands and vegetation indices, yields the best predictions. This research demonstrated the applicability and feasibility of emerging UAV technologies in agriculture, especially regarding soil properties estimation" -- Abstract, p. iv

Advisor(s)

Grote, Katherine

Committee Member(s)

Rogers, J. David
Liu, Kelly H.
Smith, Ryan G.
Nelson, Kelly A.

Department(s)

Geosciences and Geological and Petroleum Engineering

Degree Name

Ph. D. in Geological Engineering

Publisher

Missouri University of Science and Technology

Publication Date

Summer 2024

Pagination

viii, 146 pages

Note about bibliography

Includes_bibliographical_references_(pages 46, 89 & 132-143)

Rights

©2024 Yunyi Guan , All Rights Reserved

Document Type

Dissertation - Open Access

File Type

text

Language

English

Thesis Number

T 12424

Electronic OCLC #

1460027305

Share

 
COinS