<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
<title>Auburn University</title>
<link href="https://etd.auburn.edu:443" rel="alternate"/>
<subtitle>The DSpace digital repository system captures, stores, indexes, preserves, and distributes digital research material.</subtitle>
<id xmlns="http://apache.org/cocoon/i18n/2.1">https://etd.auburn.edu:443</id>
<updated>2026-11-17T07:34:02Z</updated>
<dc:date>2026-11-17T07:34:02Z</dc:date>
<entry>
<title>A Remote Sensing-Based Framework for Detecting and Monitoring Brown Spot Needle Blight Symptoms in Pine Forests</title>
<link href="https://etd.auburn.edu/handle/10415/10668" rel="alternate"/>
<author>
<name>Singh, Swati</name>
</author>
<id>https://etd.auburn.edu/handle/10415/10668</id>
<updated>2026-11-13T20:08:23Z</updated>
<published>2026-11-13T00:00:00Z</published>
<summary type="text">A Remote Sensing-Based Framework for Detecting and Monitoring Brown Spot Needle Blight Symptoms in Pine Forests
Singh, Swati
Brown spot needle blight (BSNB), caused by Lecanosticta acicola, threatens loblolly pine (Pinus taeda L.) forests in the southeastern United States. This dissertation evaluated remote sensing and machine learning approaches for assessing BSNB symptoms across spatial scales and examined their consistency across plots and years. A literature review (Chapter 2) was followed by five empirical studies (Chapters 3-7), progressing from individual crowns to satellite imagery, and a synthesis of the findings (Chapter 8). The review identified limited remote sensing research on BSNB despite advances in applications to other needle diseases. The empirical studies combined field crown ratings from 14 plots in Alabama with unmanned aerial vehicle (UAV) multispectral imagery, structure-from-motion (SfM) point clouds, light detection and ranging (LiDAR) data, and Sentinel-2 satellite imagery. UAV multispectral classification achieved approximately 95% overall accuracy within individual plots. When all pixels from each reference polygon were kept in the same data subset, the mean macro-average F1 score (macro-F1) was 0.80. SfM point clouds generated from a low-cost red-green-blue (RGB) camera accurately estimated canopy height but missed many trees in closed canopies. Integrated UAV RGB-LiDAR models achieved 90.6-99.7% within-plot overall accuracy, with canopy height and cover among the strongest predictors of BSNB severity. However, UAV multispectral model performance declined when applied to new plots (macro-F1 = 0.64) and a new year (macro-F1 = 0.37). UAV severity maps were aggregated to 10 m cells to derive disease fraction, which served as the reference for Sentinel-2 regression. Sentinel-2 predictions of disease fraction were less accurate than predictions from training-mean baseline. These findings demonstrate that UAV imagery can support BSNB symptom assessment within stands, but model transfer remains limited across stands and years and from 3 cm UAV imagery to 10 m satellite imagery. Broader operational monitoring will require validation on independent sites and years, repeated seasonal observations to evaluate sensor suitability, and standardized field and processing workflows to improve model transferability across stand conditions.
</summary>
<dc:date>2026-11-13T00:00:00Z</dc:date>
</entry>
<entry>
<title>Downed Timber Research: Demonstration of a felling head for a rubber-tired feller-buncher specializing in timber harvesting of hurricane damaged sites</title>
<link href="https://etd.auburn.edu/handle/10415/10667" rel="alternate"/>
<author>
<name>Walker, James</name>
</author>
<id>https://etd.auburn.edu/handle/10415/10667</id>
<updated>2026-11-13T18:58:28Z</updated>
<published>2026-11-13T00:00:00Z</published>
<summary type="text">Downed Timber Research: Demonstration of a felling head for a rubber-tired feller-buncher specializing in timber harvesting of hurricane damaged sites
Walker, James
This research project is with the Downed Timber Research project which is a bold yet obtainable idea which benefits areas devastated by hurricanes. There are major issues with the current logging operations in the Southeast with regards to harvesting storm damaged timber because the salvage recovery is difficult and slow with harvesting systems common in the region. The Downed Timber Research project proposes a new method for harvesting storm damaged timber by integrating ideas from the forest industry on ground-based equipment that is commonly used by logging crews in the Southeast. In this research project, a new kind of attachable felling head was designed, fabricated, tested, and analyzed. This felling head attaches to a rubber-tired feller-buncher and features a boom that can extend a felling head grapple saw eighteen feet from the base of the machine. The grapple saw can then grab, cut, and accumulate downed or standing timber for a grapple skidder to skid to the logging deck. The mechanical integrity of individual components and dynamic (tipping) simulations of the machine system were tested in engineering software and the results of each show the machine is suitable for in-woods use. A prototype was manufactured to confirm the tests and was comprised of a John Deere 843L feller-buncher with a modified John Deere FD45 felling head mounted with a Barko 80 XLE boom attached to a Waratah FL85 dangle head grapple saw installed on the far end of the boom. Field testing demonstrated that the prototype performed within the specifications of the evaluated conditions and operated as intended, supporting the need for further testing and design development. It is expected that this machine will improve the efficiency of downed timber harvesting operation for southeastern logging crews from the ability to harvest standing and down stems in any orientation.
</summary>
<dc:date>2026-11-13T00:00:00Z</dc:date>
</entry>
<entry>
<title>Transformational Leadership Mitigates Race-Based Biases in Social Pain and Support Gaps at Work</title>
<link href="https://etd.auburn.edu/handle/10415/10666" rel="alternate"/>
<author>
<name>Lehman, Isabella</name>
</author>
<id>https://etd.auburn.edu/handle/10415/10666</id>
<updated>2026-11-12T20:22:35Z</updated>
<published>2026-11-12T00:00:00Z</published>
<summary type="text">Transformational Leadership Mitigates Race-Based Biases in Social Pain and Support Gaps at Work
Lehman, Isabella
Black employees experience subtle workplace bias through the invalidation of their emotional pain, an experience termed social pain minimization (SPM). Building on race-based pain bias research (e.g., Deska et al., 2020), this work investigates whether managers exhibit similar biases when judging hypothetical Black--relative to White--employees. I then explore whether transformational leadership (TL) can buffer against the minimization of Black workers’ pain, offering an encouraging leadership-based pathway for reducing these workplace inequities. In Study 1, managers judged Black individuals as less distressed and in need of fewer coping resources than White individuals. Study 2 extended these findings, demonstrating that TL moderated pain judgments for hypothetical employees, such that more transformational leaders made more equitable judgments of hypothetical Black workers’ pain. These pain judgments mediated TL’s effect on employee support judgments. Study 3 extended this work to employees’ reported workplace experiences using a longitudinal design. It was discovered that greater perceived TL at Time 1 predicted lower SPM at Time 2. However, SPM at Time 2 did not prospectively predict perceived supervisor support or organizational commitment at Time 3 after accounting for prior levels of these outcomes. TL nevertheless remained positive. Across three complementary studies, these findings demonstrate race-based disparities in judgments of social pain, identify transformational leadership as a potential correlate of more equitable pain judgments and lower subsequent experiences of SPM, and clarify the longitudinal relationships among leadership, pain minimization, and workplace outcomes.
</summary>
<dc:date>2026-11-12T00:00:00Z</dc:date>
</entry>
<entry>
<title>Microbial Dynamics in Forage-Based Beef Systems</title>
<link href="https://etd.auburn.edu/handle/10415/10665" rel="alternate"/>
<author>
<name>Barrett, Tessa</name>
</author>
<id>https://etd.auburn.edu/handle/10415/10665</id>
<updated>2026-11-09T20:55:29Z</updated>
<published>2026-11-09T00:00:00Z</published>
<summary type="text">Microbial Dynamics in Forage-Based Beef Systems
Barrett, Tessa
Warm-season annual forbs such as sunn hemp, soybean, and lablab are valued forages in Southeastern beef systems for their nitrogen-fixing capacity and secondary metabolites which can alter rumen microbial communities. However, specifics around microbial shifts, persistence, and transfer along the rumen-feces-soil continuum when supplemented with forbs remains unclear. Fistulated steers supplemented with sunn hemp, soybean, or lablab supplied rumen fluid and feces to evaluate rumen microbiome changes and microbial succession in fecal degradation plots over 112 days using 16S rRNA amplicon sequencing. Forb supplementation altered specific rumen phyla but did not affect overall diversity or community structure across treatments. In fecal degradation plots, treatment did not differentiate microbial communities (P &gt; 0.73). Conversely, sample type was the dominant driver of microbial variation (P &lt; 0.01), with feces transitioning from a rumen-like anaerobic community toward an aerobic, soil-associated profile over time.
</summary>
<dc:date>2026-11-09T00:00:00Z</dc:date>
</entry>
</feed>
