This project presents the design and development of a quadcopter drone equipped with a swappable payload system for multi-mission operations. Unlike conventional single-purpose drones, the proposed platform enables rapid integration of interchangeable payloads, including a robotic manipulator, camera, lidars, and a magnetometer. A standardized mechanical and electrical interface was developed to support plug-and-play functionality, allowing payloads to be exchanged without modification to the core flight system.

The frame is constructed from lightweight carbon fiber tubes and plates. The UAV is equipped with a flight controller that is responsible for receiving operator commands to control the drone, while payload modules operate through dedicated control and communication links, entirely separate from the flight controller system. Particular emphasis is placed on the robotic arm payload, which enables aerial interaction tasks such as object manipulation and contact-based inspection. Key engineering challenges addressed include payload-induced center of gravity shifts, increased power demand, robotic arm work envelope intersection with propellers’ rotation region and the dynamic coupling between flight stability and the robotic arm’s motion. Design evaluation demonstrates the feasibility of the modular architecture, highlighting trade-offs between payload configuration, flight endurance, and system stability. The proposed system offers a scalable and cost-effective solution for applications in infrastructure inspection, mining, and hazardous environment monitoring, and provides a foundation for future work in autonomous aerial manipulation systems. 

Clear and accessible communication is essential in public-facing documents, particularly in policy-making contexts. Policy writers should prioritize readability to ensure that documents can be understood by their intended audiences. When policies are overly complex, the information they contain may become inaccessible to the communities they are meant to serve. This study evaluates the readability of five New Mexico Tech institutional policies to determine whether they effectively communicate with their target audiences. Two widely used readability metrics, the Flesch Reading Ease and Flesch–Kincaid Grade Level scales, were applied to measure the difficulty of each document. In addition, the policies were examined for linguistic features identified by Maaß (2020) that are known to increase textual complexity, including dense syntax, long sentences, and specialized terminology.

Preliminary analysis indicates that all five policies fall within “difficult” or “very difficult” readability ranges. Features such as lengthy sentences, complex grammatical structures, and administrative or technical jargon appear consistently across the documents and contribute to reduced accessibility. These barriers may disproportionately affect readers such as international students or individuals with disabilities. Overall, the results suggest that adopting plain language principles could significantly improve the accessibility and usability of institutional policies for the New Mexico Tech community. 

Many butterflies are collected and preserved in museums, providing a reference for researchers decades into the future. However, undertaking genetic studies of these preserved specimens poses an interesting challenge, as only the legs of butterflies can be used to gather DNA. Because only a small portion of tissue can be used, it is crucial for researchers to become proficient in DNA extraction with minimal tissue samples. This study investigates whether there are differences in DNA yield between legs taken from museum-prepared butterfly specimens and legs taken from freshly collected butterflies. Museum specimens were obtained from the Museum of Southwestern Biology arthropod collection, while fresh specimens were collected in Socorro, New Mexico, through field sampling. Methods included field collection, specimen preparation, thawing of frozen material, DNA extraction from butterfly legs, and quantification of DNA yield. By comparing DNA extracted from museum-prepared legs and freshly collected legs, this study aims to determine whether museum specimens can provide DNA of comparable quantity for genetic research. Although the research is still ongoing, the goal of this study is to determine whether butterfly legs from museum specimens can be reliably used for DNA extraction without causing significant damage to the preserved specimens. The results may help guide future genetic studies, sampling protocols, and experimental design while supporting the preservation of museum collections, especially rare and endangered butterfly species. 

New Mexico Tech’s next CubeSat mission (NMTSat) will include a coherent beacon transmitter operating at 150 MHz, 400 MHz, and 1067 MHz. The signals will be transmitted to an array of ground stations around the world. The received phase differences between these three frequencies carry information about the total electron content (TEC) of the ionosphere. To enable wide, low-cost deployment of ground receivers, we are prototyping solutions that can be produced, shipped, and installed at scale. This work uses inexpensive SDRs (e.g. RTL-SDRs) within each receiver alongside an embedded computer such as a Raspberry Pi. These low-cost SDRs have limited bandwidth (approximately 3 MHz for the RTL-SDR). To bypass this limitation, we propose two distinct approaches. The first method involves using an on-board frequency generator in combination with frequency mixers to downconvert the beacon signals into frequencies within the 3 MHz RTL-SDR passband, all while preserving the phase difference between the signals that carries the TEC information. The second method involves synchronizing three separate SDR receivers to a common clock, with each receiver tuned to their respective carrier frequencies. Both approaches are presented as candidate architectures for low-cost coherent multi-frequency receivers. 

Precision landing remains a critical challenge in autonomous aerial systems, particularly in environments where traditional vision- or GPS-based methods become unreliable. This work proposes a bioinspired approach to drone landing based on electroreception mechanisms observed in bees. Bees are known to sense electric fields through mechanosensory hairs, enabling them to detect and interact with charged surfaces such as flowers. Motivated by this capability, this project develops an electric field–guided landing framework for unmanned aerial vehicles. A simulated environment is constructed in which a drone navigates toward a charged landing pad using only local electric field measurements. Instead of directly measuring field gradients, the system estimates them from temporal variations in sensed field intensity, mimicking biological sensing constraints. The estimated gradient is then used to guide motion through both classical control and learning-based strategies. In particular, a supervised learning controller is trained to imitate an ideal guidance policy under nominal conditions, enabling improved robustness in the presence of noise, environmental disturbances, and electromagnetic interference. Results from large-scale simulations demonstrate that the proposed approach achieves reliable landing performance under challenging conditions, including sensor noise, control latency, and field distortions. Compared to baseline gradient-following controllers, the learning-based method exhibits smoother trajectories and higher success rates. This work highlights the potential of bioinspired sensing as an alternative paradigm for autonomous navigation and contributes toward the development of resilient landing systems for next-generation aerial robotics. 

How has COVID-19 evolved compared to other viruses? Understanding the evolutionary relationships of coronaviruses is important for monitoring viral adaptation and emergence. In this study, we performed a comparative genomic analysis of the SARS-CoV-2 variants Alpha, Beta, Gamma, Delta, and Omicron alongside SARS-CoV-1, Rhinovirus A, and the human coronaviruses OC43 and 229E. Multiple sequence alignments were used to construct a phylogenetic tree, thereby visualizing the genetic relationships among these viruses. This approach provides a framework for examining patterns of viral evolution and may contribute to ongoing efforts in surveillance, vaccine research, and understanding coronavirus genomics. 

Fused Deposition Modeling (FDM) is widely used for rapid prototyping and functional part fabrication, but the mechanical performance of printed parts remains strongly dependent on slicer settings. This study investigates how different solid-build strategies and extrusion temperatures affect the tensile and flexural behavior of PLA+ components printed on an Ankermake M5. Prior literature consistently showed that 100% infill provides the greatest strength, but it left unresolved whether a fully solid part produced by traditional rectilinear infill performs differently from one produced by increasing vertical perimeter walls until the part is completely solid. To address this, bending and tensile specimens were printed using both approaches and mechanically tested. Bending samples were evaluated using peak load, while tensile samples were converted from failure load to tensile stress for normalized comparison. Results showed that increasing vertical wall count improved bending performance, with the fully solid wall-based configuration exceeding the strength of standard 100% rectilinear infill. Tensile results also demonstrated a strong dependence on extrusion temperature, with performance generally peaking near 210°C due to improved interlayer adhesion. Skin type further influenced results, with normal skin outperforming fuzzy skin in standard infill configurations, while fuzzy skin showed competitive behavior when paired with vertical wall strategies. These findings suggest that for high-performance FDM parts, wall-based solid construction combined with optimized extrusion temperature can provide a stronger overall mechanical profile than conventional 100% infill methods. 

Ultraviolet (UV) radiation plays an important role in our atmosphere and has major effects on surface conditions for life on Earth. In the stratosphere, UV radiation has the chance to split a diatomic oxygen molecule, which can then react with surrounding oxygen to form an ozone molecule. These ozone molecules absorb most of this harmful UV radiation, protecting life on the surface. The UV Index is a measure of how much harmful radiation is able to penetrate this ozone layer and reach the surface, and is a number typically ranging from 1–10. To measure these quantities, we mounted a UV radiometer on the roof of Workman Center to measure the UV radiation reaching the surface in 305, 320, 340, and 380 nm bands. Then, from these measurements, we obtain an estimate for column ozone over Workman Center, which is calibrated to TEMPO satellite measurements. We also obtain a UV index measurement. From these results, we find that peak UV index during the day is inversely related to the measured column ozone. Additionally, we find that column ozone is typically constant throughout the day, although some days exhibit increased variability. 

Food waste in various university systems represents significant environmental, economic, and institutional challenges. This study examines and evaluates the scale and impacts of food waste at New Mexico Institute of Mining and Technology, using established research on higher education dining systems to estimate campus-level effects. It is hypothesized that New Mexico Tech generates substantial amounts of food waste through overproduction and student plate waste, resulting in measurable environmental and financial consequences that could be remediated by implementing campus composting programs. To evaluate this hypothesis, the study applies findings from peer-reviewed literature on institutional food waste that includes measured estimates of plate waste per meal as well as annual per-student waste estimates. This approach provides a research-based estimate of total annual food waste by the university and its various associated impacts. Additionally, findings show that composting has been shown to reduce overall carbon footprints and improve resource efficiency in waste management systems. Findings suggest that even moderate levels of food waste per meal can accumulate into large annual totals, potentially exceeding 200,000 pounds of organic waste on campus. The integration of composting programs could substantially mitigate these impacts by reducing landfill dependence and associated emissions. This study concludes that food waste is a significant and under addressed issue at New Mexico Tech University, with implications for greenhouse gas emissions, resource efficiency, and operational costs. Addressing this issue presents an opportunity for the university to improve its sustainability practices while supporting potential educational and research-supported objectives through implementing campus composting programs. 

Freshwater ecosystems constitute approximately 0.1% of the Earth’s surface yet remain among the most threatened environments worldwide. Within these systems live yeasts that are largely overlooked compared to bacterial and macroinvertebrate communities commonly used to assess freshwater health. Aquatic yeasts contribute to nutrient cycling and organic matter decomposition and are increasingly recognized as members of insect-associated microbiomes in freshwater habitats, particularly mosquitoes, although their ecological roles in mosquito biology remain poorly understood. Using fruit-baiting and culture-based approaches, we documented 185 fungal isolates, including a large number of yeasts (57% of isolates), from an urban freshwater stream in west-central Illinois. We further examined interactions between these yeasts and Aedes mosquitoes by testing the influence of representative yeast taxa on mosquito oviposition behavior and egg hatch rates. Yeast communities were dominated by Ascomycota, with Meyerozyma representing 41% of all isolates, including both pigmented and non-pigmented yeast forms. Yeast exposure altered mosquito oviposition patterns and affected egg hatching success relative to controls. Additionally, 16S rDNA sequencing revealed that pigmented Meyerozyma guilliermondii isolates were dominated by distinct bacterial communities compared to non-pigmented isolates, and we predict these assemblages influence microbial interactions within freshwater habitats. Our results demonstrate that urban freshwater systems harbor a rich and understudied yeast community with measurable effects on mosquito reproductive behavior and early life-stage survival. These findings highlight freshwater yeasts as ecologically meaningful but overlooked contributors to aquatic ecosystems and mosquito life cycles, with important implications for vector ecology and public health. 

Rising textbook costs and their affordability create financial barriers that directly affect student success and decrease student equity going against NMT’s vision of equity and academic excellence. 30% of first-generation college students are not purchasing required textbooks due to costs, and they are twice as likely to both struggle in classes and not enroll in them due to textbook costs, and with textbook prices rising at twice the rate of inflation, this is an issue worth paying for to uphold equity standards and to aid New Mexico Tech students. We investigate the causes of rising textbook costs, their pricing trends and we discuss some of the less-than-ideal environment for academic publishing and textbook publishing specifically. This research investigates open-source alternatives to paid textbooks with the goal of increasing equity and affordability for students as well as making college classes more accessible for those who would not be able to afford it otherwise.

The solution we investigate is OpenCourseWare initiated by The Massachusetts Institute of Technology (MIT) and the Open Learning Initiative which was started by Carnegie Mellon University in 2002. 

Purpose: Stress is a major concern in the cybersecurity field with over half of all cyberdefense professionals reporting severe stress in a year. Stress has a negative impact on both job performance and the health and retention of cyberdefense staff. If stress can be unobtrusively measured, then individuals and organizations can intervene with proven stress management strategies, to maximize performance and protect their workforce.

Hypothesis: Self-report and physiological stress assessment can reliably quantify stress caused by cybersecurity tasks.

Methods: A computer based cybersecurity challenge (game) was developed to induce mild cognitive stress. In-game measures of performance were augmented with measures of cognitive and physiological stress. Perceived cognitive stress was measured using a custom self report that included the NASA task load index (TLX). Physiological stress was measured using wearable technology to record heart rate variability (HRV) and galvanic skin response (GSR).

Results: The expected outcomes demonstrated that under a stress inducing environment, task performance will be degraded and participants will demonstrate increased stress measured by self report and physiology. We also expect performance and perceived stress to be greater in participants who began the test with higher levels of physical and mental stress.

Discussion: We expect to find that stress can be measured, is cumulative and affected by pre-task stressors, and can be managed using proven interventions. This will allow us to extend this practice to real world situations and to affect policies among cyberdefense employers, such as encouraging proper rest, diet and exercise and assisting staff with external stressors. 

Additive manufacturing, in this case 3D printing, is the process of depositing material in successive layers to construct complex objects from 3D model data. It's often used for prototyping and recreation, as it can be very precise with minimal investment. However, the process is prone to error, and in large scale production these errors occur often enough to waste a lot of time and resources. With the use of our GuidedFlow model, a normalizing flow based AI model that takes image data and then makes estimates at when and where errors are likely to arise, we can accurately and efficiently check for these errors throughout the entire printing process. Through our training on 537 images and 129 videos across 5 different printed objects, we have attained an average accuracy of 97.4%, which is higher than the models we tested against. As such, it's safe to conclude that our GuidedFlow model is an important step in the evolution of error assessment in 3D printing.

The NMTSat ground station team is focused on receiving mission data from our LEO space weather satellite. The ground station is divided into three elements, the tracking mount, the RF reception, and the control system. The control system is also to be utilized by the Ultra High Frequency (UHF) command center to exchange housekeeping data and other non mission data information. The UHF command center is a separate but highly related project. The tracking mount is led by Josiah Martinez. It must be able to support a receiver dish of up to 3 meters under different wind loading conditions and be able to accurately point to and track the satellite across the sky. The X-Band RF receiver chain consists of the electronics which receives and processes the bit stream from the satellite, and is led by Ethan Lazar. The goal is a capability of 100 Mbps mission data from the satellite using low-cost hardware, ideally under $5k of electronics. The back end control system will be responsible for commanding the tracking mount, planning satellite passes, and archiving received data, initially under operator control, with plans for automation later. This is the responsibility of Andrew Sieber. 

Underground mine pillars are subjected to sustained and cyclic stresses that can induce time-dependent deformation, progressive damage, and eventual instability. These responses are associated with creep, fatigue, and alternating creep-fatigue loading, yet their relative effects on failure time, damage accumulation at failure, and crack-evolution stages remain poorly understood. This study compares sandstone behavior under creep, fatigue, and alternating creep-fatigue loading to determine which condition causes the earliest failure, which results in the greatest damage at failure, and whether cracking stages can be identified and classified using unsupervised clustering. Sandstone specimens were tested under these loading conditions with continuous acoustic emission (AE) monitoring. Mechanical response, AE parameters, and the slope of the frequency–magnitude distribution (b-value) were analyzed. Subsequently, AE waveform spectrograms and convolutional neural network (CNN) autoencoders were used to extract latent-space features, and K-means clustering was applied to identify fracture patterns. Results show that fatigue caused the earliest failure and produced the highest AE activity, indicating the most intense rock damage. K-means clustering also revealed three distinct clusters for each loading condition, interpreted as initial compaction, subcritical crack growth, and critical crack growth, enabling objective discrimination of damage states. These findings demonstrate that the loading path strongly controls time-dependent damage evolution and microseismic response in rock, underscoring the importance of incorporating creep-fatigue interactions into long-term stability assessments of underground excavations. 

The Large Magellanic Cloud (LMC) and the Small Magellanic Cloud (SMC) are interacting irregular dwarf galaxies that are orbiting the Milky Way at distances of ~50 kpc and ~63 kpc, respectively. Particularly, the LMC is very actively forming stars, making the giant molecular clouds (GMCs) that host this star-formation prime targets for ALMA observations. Many GMCs across these systems have been studied for their star-formation activity, energetics, and metallicity, yielding a catalogue of spectral line observations. This study gathers all available ALMA observations of 23 GMCs at 1pc spatial resolution, and characterizes the detected spectral lines relative to 13CO(1-0) as a proxy for total molecular content. Particular attention is given to CS(2-1) line as a dense gas tracer in the nitrogen-depleted clouds as it compares to more usual extra-galactic star-formation tracers and C18O as a tracer of hot cores in extreme dense environments with optically thick 13CO.

This work contributes to the ongoing effort to bridge star-formation studies on galactic and extra-galactic scales. All line emission is analyzed at both the cloud-scale–where all emission is averaged across the cloud–and at parsec scales–where dendrogram analysis identifies substructures within individual clouds. We find how the correlation between star-formation rates and dense gas emission vary when observed on different physical scales and speculate on what that means for analysis in other extragalactic contexts. 

Very low frequency (VLF) radio waves, ranging from near DC to 30kHz, serve as natural probes of the Earth's ionosphere and magnetosphere. Phenomena such as whistler waves, electromagnetic pulses generated by lightning that travel along the Earth’s magnetic field, can provide information about electron density in the plasmasphere and can be detected passively at ground level. This project presents the design and implementation of a VLF ground receiver system intended for continuous monitoring and analysis of naturally occurring VLF signals. The receiver starts with a shielded loop antenna coupled to a low-noise preamplifier, feeding into a LabJack T7 data acquisition device for digitization at variable sampling rates to record the full VLF band. Signal processing is performed in Python, utilizing fast Fourier transforms (FFT), power spectral density (PSD) estimation, and short-time Fourier transform (STFT) spectrograms to identify and visualize VLF phenomena, including whistlers, sferics, and power-line harmonics. Preliminary results demonstrate clear intake and detection of events and the generation of spectrograms consistent with desired results. The system's portability and limited amount of hardware make it suitable for quick and easy deployment. This work contributes to ongoing efforts in low-budget space weather monitoring and provides a platform for undergraduate research in ionospheric physics and signal processing, as well as a proof of concept for a type of data that we hope to receive from the cube satellite project. 

There is an ever growing contamination of PFAS due to industrial and factory means. These pollutants have since leaked into every aspect of our lives, our bodies themselves, the water we drink, and products we consume. PFAS or polyfluoroalkyl substances are forever chemicals, as they do not break down in the environment and they continue to accumulate. As they accumulate we begin to see the adverse effects, with current research they have been deemed as carcinogenic and toxic to wildlife. The goal of this review is the testing of contamination in flowing water that originates from underground cave karst systems near and around Ruidoso, NM. This however proves a challenge, since water is being collected through open access means it faces extreme matrix and contamination challenges. A standard to work against this is created in accordance with EPA 1633, and for this study we will be using a SPE machine followed by the LCMS. The SPE machine, or solid phase extraction machine, uses an analyte to bind with the PFAS rather than bonding it to the water. With this analyte bond it is then taken to an LCMS machine, or liquid chromatography mass spectroscopy, to test for the analyte and PFAS spikes. From the spot of collection, there were traces of contamination from paper food packaging, water-repellant coatings, to even firefighting foams. These sources represent recent, traceable contributors to PFAS pollution, and their impacts are expected to continue accumulating over time. 

Materials failure rarely begins when we first see a crack. Long before visible fracture appears, the material is already changing through subtle deformation, strain localization, and microdamage that often escape conventional observation. This research aims to detect that hidden evolution early enough to matter. Using laboratory data from prismatic rock specimens and loaded in compression to failure, the study develops an Artificial Intelligence (AI) driven framework that combines Convolutional Neural Networks (CNN) with digital image correlation (DIC) to identify the earliest detectable signatures of damage before collapse becomes obvious. Images recorded throughout loading are converted into displacement and strain fields, allowing the model to learn how the rock progressively departs from its intact state. The goal is not simply to observe cracks after they appear, but to determine how early AI can recognize the onset of instability from weak, spatially distributed patterns that precede visible failure. By tracking the temporal evolution of these patterns, the framework seeks to reveal where damage first localizes, how it intensifies, and when the transition from distributed deformation to unstable cracking begins. The expected outcome is a physically meaningful and practically useful early warning indicator for rock damage, grounded in full field image analysis rather than subjective visual inspection alone. By enabling AI to see damage before the human eye can clearly recognize it, this work opens a new direction for safer and smarter structures, and earlier intervention in engineering systems. 

The NMT Space Weather Explorer incorporates a multi-frequency beacon system to probe the electron density in the Earth’s ionosphere. The beacon transmitter will be inspired by the design used in the Coherent Electromagnetic Radio Tomography (CERTO) beacon instrument. The signal will be transmitted by a multi-frequency crossed-dipole antenna with three sets of reflectors. The dipole must be 1 meter across and extend 50 centimeters outward from the body of the satellite. Traps make the antenna resonant at 150, 400, and 1067 MHz. This antenna must be folded into a small volume at the bottom end of the 6U CubeSat. This will be a significant miniaturization of the design compared to previous missions with similar antennas, such as DMSP, CASSIOPE, and C/NOFS. We are exploring various combinations of deployable mechanisms including shape memory alloys, tape springs, and 3D printed collapsible structures to minimize stored volume. We are exploring the possibility of using NiTi shape memory alloy wires as antennas as well as for actuating folding mechanisms. 

Robotic-assisted surgery has significantly advanced modern medical procedures by enabling trained professionals to perform minimally invasive operations that reduce trauma and accelerate patient recovery. However, current surgical robotic platforms are extremely expensive, limiting their availability to large, well-funded medical institutions. This financial barrier restricts global access to advanced surgical technologies and reduces opportunities

for research and development in under-resourced arise. This study investigates whether a cost-effective robotic-assisted surgical system can be developed using open-source hardware and software frameworks without compromising precision, reliability, or safety. The motivation for this research is to expand accessibility to robotic surgical tools and promote innovation in educational and research institutions with limited funding. The project follows a systems engineering and mechatronics-based design framework. Its primary objective is to demonstrate stable communication between robotic hardware and a ROSS 2-based control architecture, enabling a more accessible development platform. Early testing shows consistent repeatability and accurate end-effector positioning within controlled environments. The system integrates a JetCobot 7 robotic arm with a Jetson microcomputer and an onboard vision system. Operating on Ubuntu with ROSS 2, the platform supports motion planning, inverse kinematics, and real-time control. A virtual reality interface enables intuitive user input, while software constraints ensure operational safety. Experimental evaluation focuses on positional accuracy, repeatability, latency, and system stability. Results are expected to demonstrate a functional, low-cost prototype capable of precise manipulation tasks. This research suggests that affordable robotic-assisted surgical systems are feasible, with future work focusing on safety, improved precision, and surgical simulations. 

Stars are so distant that even the largest single telescopes typically cannot resolve their surfaces, causing them to appear as point-like sources of light. Optical interferometry overcomes this limitation by combining light from multiple telescopes to effectively create a much larger “virtual telescope,” allowing astronomers to probe fine details of stellar structure. In this work, we analyze observations of the symbiotic star system CH Cyg using computational image reconstruction methods. We construct a model of the star’s surface with the PMOIRED software to describe how brightness changes from the center to the edge of the star. We will use the power law for the limb darkened disk as it is a simple two parameter law. To evaluate the reliability of the model, we apply a bootstrapping technique to estimate confidence intervals on the fit between the model and the observed data. We then use our best fit model as a starting point for image reconstruction using OITOOLS and compare the reduced χ2 to the PMOIRED model. Together, these methods demonstrate how interferometry and computational modeling can reveal detailed stellar structure while providing quantitative measures of confidence in the results. 

Respiratory syncytial virus (RSV) is the leading cause of bronchiolitis and pneumonia among infants under one year of age in the United States, and is increasingly recognized as a significant contributor to respiratory illness in elderly people. In this study, we fit a gamma-distributed delay model of RSV viral dynamics to clinical viral load data from eight subjects challenged with the A2 strain using an ODE nonlinear mixed-effects (NLME) framework in Monolix. The model employs the linear chain trick with 60 eclipse and 60 infected cell compartments (123 ODEs total) to capture gamma-distributed delays in cell infection and viral production. Three fits were performed: (i) all eight subjects, (ii) excluding Patient 7 (no detectable viral replication), and (iii) excluding both Patients 4 and 7 (suspected outliers). Removing Patient 7 substantially increased the estimated population viral production rate (35,071 → 96,934 virions/cell/day) and reduced inter-individual variability, confirming its influence as an outlier. Our NLME estimates are compared against individual Nelder-Mead fits from González-Parra and Dobrovolny (2015) on the same dataset. When excluding Patient 7, we obtained the most biologically interpretable population estimates. These results offer further insight into RSV dynamics and provide a framework for future studies examining antiviral effects on viral kinetics. 

The area that is now the northern Chihuahuan Desert was historically a grassland ecosystem, and it is experiencing an encroachment of shrubs, which likely lead to desertification. Culling the numbers of these shrubs is being researched as a solution to reversing this process. Tebuthiron is a herbicide that targets woody plants but is not harmful to grasses, so it has been proposed as a potential solution to this problem. However, not much is known about the effects of TBH on soil microbial functions, including the ones that would allow restoration processes to function i.e. nitrogen and phosphorus cycling. No studies to date have reported a full greenhouse gas (GHG) account of TBH application. We collected soil and litter from the field from under the three main shrubs that are encroaching - Creosote (Larrea tridentata), Mesquite (Prosopis glandulosa), and One-Seed Juniper (Juniperus monosperma). This was done to determine the effects of TBH on nitrogen and phosphorus transformation rates, as well as the GHGs associated with microbial activity. Statistical analyses will be presented on the results of this experiment. This should highlight the effect of TBH on soil biogeochemistry in relation to both the different shrub species and the presence of litter in the experiment. 

Food insecurity has become an increasingly recognized issue in colleges, with direct consequences to academic performance and wellbeing of students. This study looks into the impacts of food insecurity for college students and the potential prevalence of the issue at New Mexico Tech. To evaluate the hypothesis that New Mexico Tech is at risk, past studies and surveys were analyzed, specifically looking into New Mexico populations and institutions. The findings show that food insecurity is a larger issue in New Mexico in comparison to the rest of the country, and is associated with increased stress and a lower academic performance. As an institution in rural New Mexico, it is fair to say that New Mexico Tech also faces similar struggles and risks.