Electronic Theses and Dissertations Archive

Date

2026

Document Type

Thesis

Degree Name

Master of Science

Department

Electrical & Computer Engineering

Committee Chair

Eddie Jacobs

Committee Member

Alfredo Ramirez

Committee Member

Mohammadreza Davoodi

Abstract

Nitrogen (N) deficiency is a significant contributor to crop stress and yield loss in cotton, yet early detection remains challenging because conventional vegetation indices such as NDVI only resolve physiological changes after prolonged stress has produced outward structural symptoms. Solar-induced chlorophyll fluorescence (SIF), estimated via the improved Fraunhofer Line Depth (iFLD) method, offers a functional indicator of photosynthetic activity that responds to stress before structural changes emerge. This thesis presents two studies investigating the use of a single-channel line-scan hyperspectral camera for proximal SIF retrieval and its application to early N-stress detection in cotton. Study I demonstrated that an in-scene reflectance panel can replace a dedicated spectroradiometer as the irradiance source for iFLD retrieval, performing comparably to or better than the dual-channel spectrometer method. Study II applied this validated method to cotton plants under six nitrogen treatment levels over 20 days, finding that SIF/PAR detected treatment differences as early as Day 5 ($\Delta = +2.37$ between N-deficient and N-saturated treatments) while NDVI remained insensitive ($\Delta = -0.40$). SIF/PAR also maintained a moderate to strong monotonic relationship with the nitrogen gradient across the full dataset. These results demonstrate that a single-channel hyperspectral line-scanner with in-scene reflectance targets provides a simplified, lower-cost pathway for iFLD-based SIF retrieval with sufficient sensitivity for early nitrogen stress detection in cotton.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to ProQuest/Clarivate.

Notes

Open Access

Download

Share

COinS