Scholars Junction - Mississippi State University Institutional Repository
Not a member yet
83664 research outputs found
Sort by
Intensifying heat stress impacts cotton flowering and boll development efficiency
No full textCotton-growing regions face heat stress during flowering and boll development, adversely affecting reproductive fitness, yield, and quality. Identifying and incorporating superior physiological and reproductive traits into elite genetic background may improve yield potential under stress. This study quantified the effects of heat stress (36/24 °C, HS) on 25 traits of 16 upland cotton cultivars during the reproductive stage, compared to control conditions (32/24 °C, CNT). Under HS, all cultivars exhibited transpirational cooling, with stomatal conductance increasing by 1-fold and transpiration by 1.6-fold compared to CNT. Under HS, leaf temperature was 3.5 °C higher than CNT plants, and PSII quantum efficiency dropped by 20% compared to CNT. HS also reduced the ratio of reproductive to vegetative dry mass (57%), indicating poor resource partitioning towards reproductive organs. The specific leaf area was increased by 34%, whereas the leaf chlorophyll index dropped by 28% under HS compared with CNT. When exposed to HS, pollen germination was reduced by 71% across cultivars compared to the CNT. While some cultivars maintained similar boll and seed numbers between CNT and HS, a corresponding tolerance was not observed in seed cotton yield. On average, seed cotton yield and lint yield decreased by 19% and 26% under HS, respectively. In contrast, seed yield remained stable, suggesting a disproportionate reduction in intra-boll components. Seed oil content and fiber quality, except for fiber uniformity, displayed susceptibility to HS. Multidimensional responses of cotton cultivars to HS highlight the need to map genetic loci governing heat tolerance. This knowledge is essential for pinpointing effective breeding targets
Impact of climate change stressors—temperature, CO2, and UV-B—on early growth and development of different cover crop species
No full textDifferent cover crop (CC) species may respond differently to the projected climate change scenarios. A study was carried out in a controlled environmental chamber to evaluate early season growth and development of five CC species: cereal rye (Secale cereale L.), triticale (x Triticosecale Wittmack), winter wheat (Triticum aestivum L.), crimson clover (Trifolium incarnatum L.), and mustard (Brassica juncea). Treatments consisted of two levels of carbon dioxide (CO2) (420 and 720 ppm), ultraviolet-B (UV-B) radiation (0 and 10 kJ m−2 day−1), and temperatures (29/21°C and 19/11°C day/night), and their combinations. Root, shoot, and physiological parameters were recorded, and a combined stress response index (CSRI) was derived. Results indicated that higher CO2 (+CO2) had a net positive effect on all five CC species, with CSRI values ranging from 1.0 to 5.1. Conversely, higher UV-B radiation (+UV) had a net negative impact, with CSRI values ranging from −2.9 to −7.6. The most favorable environment for all CC species was the combination of increased fall temperature and elevated CO2 (+T+ CO2). The negative impact of +UV was mitigated in an elevated CO2 and a high temperature environment, mimicking fall temperatures in the US Midsouth. Among the CC species, mustard was the most responsive, with a 151% increase in root and shoot combined dry weight under the +T+ CO2 treatment and an 86% decrease under the +UV treatment. Rye and triticale were the least impacted by the imposed climatic stressors. These results are of particular interest to the agricultural and environmental science community as they offer insights into developing and selecting CC species with adaptable and desirable morphological characteristics in anticipation of a changing climate
Sorting It Out
No full textWe describe a card sorting activity called a Q-sort that teachers can use to foster reflection and discussion
Can We Use Visible-Near Infrared and Mid Infrared Spectroscopy as a Tool for Wetland Soil Identification?
No full textThe wetland delineation process is primarily based on the visual recognition of anaerobic soil indicators by trained individuals, and is a complex and subjective task that is prone to error. Therefore, an objective alternative is needed to identify wetland soil; however, no such method currently exists that is rapid and easy to deploy. Accordingly, the objective was to evaluate soil spectroscopic classification approach as a rapid, deployable alternative by testing its feasibility to differentiate wetland from non-wetland soils. This study used visible-near infrared and mid-infrared (MIR) ranges for this task. A total of 440 wetland and non-wetland soils were sampled across Mississippi followed by obtaining visible/near-infrared and MIR spectra under both fresh and dried conditions. Support Vector Classification (SVC) and Random Forest (RF) methods were then used to classify spectra based on wetland/non-wetland status with a 75 %/25 % calibration and validation split. This split was repeated for 50 iterations to obtain randomized calibration and validation sets for model calibration and achieve average model performance. The average classification accuracy across all models was ∼91 %, with the highest accuracy of 99.6 % achieved on MIR spectra. The accuracy, precision, and recall scores showed similar performances between SVC and RF ranging their values from ∼80 % - 100 %. This study showed the reliability and ease of wetland determinations using spectroscopy as an objective and rapid wetland recognition method, while reducing the need for an expert for determination
Design and validation of an aerothermodynamic digital twin framework for hypersonic glide vehicles
No full textAccurate and rapid prediction of aerothermodynamic loads remains an ongoing challenge for intelligent trajectory design and optimization for hypersonic vehicles. Existing approaches rely on engineering approximations to estimate aerothermal heating and mechanical loading, which are valid only for a limited range of flow conditions and geometric shapes. Direct use of computational fluid dynamics (CFD) simulations to support trajectory optimization is infeasible considering the computational expense associated with calculating a numerical solution to the Navier-Stokes equations, even under the assumption of steady-state flow. To address this shortcoming, I introduce an aerothermodynamic digital twin (AT-DT) framework which enables rapid prediction of aerothermodynamic loads and thermal response along the surface of a hypersonic vehicle. The framework uses a data-driven surrogate model to predict the aerothermodynamic loading anywhere on the vehicle surface as a function of freestream conditions, vehicle attitude, and wall temperature. The surrogate model is coupled with a lightweight, one-dimensional thermal solver to analyze heat soak into the structure. The AT-DT model is validated using experimental data from the HIFiRE-5b flight test. The results indicate the framework is able to accurately predict aerothermodynamic loads and changes in the vehicle wall temperature along the reentry trajectory. Using an AT-DT, thermal analysis of a trajectory can be performed in a matter of seconds, compared to hours or days for traditional conjugate heat transfer (CHT) analysis techniques. This novel framework is coupled with a gradient-based trajectory optimization algorithm that leverages modern GPU computing hardware and open-source deep learning libraries. Unlike many gradient-based algorithms which use finite differencing to estimate the derivatives of the loss functions and constraints, the proposed methodology leverages automatic differentiation to rapidly calculate gradients with a high degree of accuracy. This architecture is used to determine a series of active maneuvers which minimize heating at specific locations on the vehicle while still achieving desired terminal location constraints. Unlike many trajectory optimization approaches, this framework can be used to determine optimal trajectories based on the design characteristics and limitations of individual vehicle components, enabling a more sophisticated approach to maximizing performance and ensuring survivability of hypersonic glide vehicles
Advanced EO/IR sensor data analysis: DTW-based methods for track simulation, clutter reduction, and classification
No full textSince their initial use by John Johnson at the U.S. Army Night Vision Laboratory (later renamed the U.S. Army Night Vision and Electronic Sensors Directorate, NVESD) for target identification, recognition, and prediction, Electro-Optical and Infrared (EO/IR) sensors have become widely employed in surveillance, intelligence gathering, geospatial monitoring, and military operations. With a projected market valuation of $12.9 billion by 2031, EO/IR sensors play a crucial role in addressing global military challenges and advancing Unmanned Aerial Vehicles (UAVs). This is particularly evident in the commercial and civilian UAV sectors, where Beyond Visual Line of Sight (BVLOS) research has become essential. To support these advancements, the Federal Aviation Administration (FAA) is actively funding research aimed at enabling BVLOS flight for commercial and civilian applications. The objective of this research is to enhance EO/IR sensor data analytics by employing Dynamic Time Warping (DTW)-based approaches. Specifically, this research (1) Evaluates EO/IR sensor data and assesses decluttering techniques (2) Compares and contrasts various time series and sensor trajectory classification techniques (3) simulates sensor signals (4) utilizes machine learning methods for sensor classification
Netflix and Kill: the role of true crime consumption in the development of acquired capability for suicide
No full textThe Interpersonal-Psychological Theory of Suicide (IPTS) identifies acquired capability as a key factor contributing to suicidal behavior. Previous research has shown that violent media exposure can increase suicidal risk through desensitization, imitation, and trauma activation. However, little is known about the role of true crime consumption (TCC), a popular form of violent media engagement, in relation to IPTS constructs. This study examined whether TCC predicts acquired capability for suicide, perceived burdensomeness, and thwarted belongingness. A total of 215 participants were retained after data cleaning. Hypotheses predicted that higher TCC would be associated with higher acquired capability (H1), that TCC would not significantly relate to burdensomeness or belongingness (H2), and that curiosity-driven engagement with true crime would predict lower fear of death (H3). Correlational and multiple regression analyses found no significant associations between TCC and acquired capability, thwarted belongingness, or perceived burdensomeness. Additionally, curiosity did not predict differences in fear of death. These findings suggest that TCC, unlike other forms of violent media, may not meaningfully contribute to IPTS risk factors. Limitations included sample size, methodological constraints, and lack of a validated TCC measure. Results underscore the need for refined methodologies and targeted measures in future research to clarify the potential relationship between true crime engagement and suicide risk
Toward approachable reinforcement learning: Using dataflow application and large language models for human-understandable policies
No full textOptimal decision-making under uncertainty is a challenge in everything from board games to business scenario planning. With visual programming systems, end-users can utilize reinforcement learning (RL) agents to identify an optimal policy in their own stochastic environments. However, while the RL agent can demonstrate that an optimal policy exists, RL polices tend to be opaque and difficult to interpret. This thesis investigates how Large Reasoning Models (LRMs) can generate human-understandable rules from RL policies. I compare multiple approaches and models, varying the format of the RL policy provided to the LRM as well as the prompting strategy. My results show that the efficacy of each approach depends on the environment. In low stochasticity environments, LRMs reason more effectively by observing RL agent episode trajectories, whereas in high-stochasticity environment, LRMs reason more effectively by reviewing an RL agent’s policy itself. This work contributes to bridging the gap between RL’s computational power and the need for transparent, human-simulatable decision rules
Trace mineral supplementation strategies for beef cattle during preconditioning and feedlot receiving
No full textBeef cattle routinely undergo a variety of stressors within the preconditioning and feedlot receiving phases which may impair animal performance and health. Trace mineral supplementation, either through dietary or injectable sources, is proposed to mitigate these effects through metabolic and immune system support. Two systematic reviews and meta-analyses were conducted to evaluate the efficacy of two different oral supplementation methods, organic (OTM) and inorganic (INR), and injectable trace mineral (ITM) solutions on average daily gain (ADG) and morbidity. Across 20 studies, OTM supplementation increased ADG (P = 0.01) when compared to INR, but did not affect morbidity (P = 0.92). In a separate analysis of 16 studies, ITM administration did not influence ADG (P = 0.21) or morbidity (P = 0.20). Collectively, these findings suggest that while morbidity outcomes remain unchanged, targeted trace mineral supplementation strategies through OTM sources can support animal performance during periods of increased stress