Author ORCID Identifier

https://orcid.org/0000-0001-9783-4295

Year of Publication

2021

Degree Name

Master of Science in Biosystems and Agricultural Engineering (MSBiosyAgE)

Document Type

Master's Thesis

College

Agriculture; Engineering

Department/School/Program

Biosystems and Agricultural Engineering

First Advisor

Dr. Joseph Dvorak

Abstract

Alfalfa is a popularly grown crop because of its value as a nutritious feed source for livestock. The efficient production of an alfalfa crop relies on the monitoring of certain parameters, like height, quality, and yield. Traditionally, producers have used manual measurements of alfalfa plant height to estimate the nutritive quality and yield of a growing alfalfa crop. Manual measurements of plant height are often labor intensive and provide low resolution data that is not acceptable for full field scale assessment of growing alfalfa. The two studies presented in this thesis offer detailed insight into the rapid and accurate monitoring of alfalfa with LiDAR and UAV technologies. The first study explores the use of a simple single beam LiDAR sensor to accurately estimate the average canopy height and yield of an alfalfa crop. Predictive models of alfalfa canopy height were developed and evaluated to find the optimal LiDAR derived measurements to use. The resulting measurements were then used to build predictive models of yield, and the best yield model was determined. The best models of canopy height and yield both incorporated the 95th percentile of LiDAR derived canopy height as a single explanatory variable. The second study assesses the field conditions, flight parameters, and general statistical descriptors that should be considered for the stable collection and application of UAV derived canopy height information. Data taken from different alfalfa fields at different flight parameters with different statistical processing were all compared. General canopy height distribution statistics from UAV flights flown at or below 50 m with nadir and oblique camera angles over thick stands of alfalfa were determined to be reliable for the detection and application of the alfalfa canopy surface. Using these determined methods, predictive models of canopy height and yield were generated and compared. The best model of average canopy height used the 50th percentile of UAV derived canopy height from an UAV flight at 30 m in a nadir imaging configuration. The best model of yield used the 95th percentile from an UAV flight at 50 m in an oblique imaging configuration.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2021.340

Share

COinS