MSU Libraries Digital Repository (Michigan State University)
Not a member yet
106711 research outputs found
Sort by
EVALUATING THE SUCCESS OF A KINETIC MODEL TO PREDICT CHROMATOGRAMS OF IGNITABLE LIQUIDS UNDER DIFFERENT EVAPORATION MODES AND IN THE PRESENCE OF PASSIVE-HEADSPACE EXTRACTION
No full textThesis (M.S.)--Michigan State University. Forensic Science - Master of Science, 2025In forensic fire debris analysis, extracted samples must be compared to an extensive reference collection that must include liquids evaporated to various levels. The evaporation of these ignitable liquids is currently done in-house, in a process that may be time-consuming and with methods that vary from laboratory to laboratory. As such, our laboratory previously developed a kinetic model to predict chromatograms of evaporated ignitable liquids to which samples may be compared. However, development and validation of this model were accomplished under different evaporation conditions and without the use of passive headspace extraction. Therefore, the objectives in this work were to evaluate the effect of evaporation mode and extraction on the predictive success of the kinetic model. To do this, gasoline samples were evaporated to varying evaporation levels under three evaporation modes: 1) in a graduated cylinder with sparging and agitation, 2) without agitation, and 3) in a petri dish with no sparging or agitation to simulate a thin film. Samples were then analyzed with gas chromatography \u2013 mass spectrometry (GC-MS). Additionally, gasoline and paint thinner were evaporated in a graduated cylinder with sparging and agitation, then analyzed both with and without passive-headspace extraction. Chromatograms of the experimentally evaporated ignitable liquids were compared to chromatograms predicted by the kinetic model using Pearson product-moment correlation (PPMC) coefficients and mean absolute percent error (MAPE). Overall, the model performed well, with strong correlation between experimental and predicted chromatograms regardless of evaporation mode and the presence of extraction.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
SHOULD I STAY OR SHOULD I GO? EVALUATING BEAVER RESPONSES TO FLOW DEVICES
No full textThesis (M.S.)--Michigan State University. Fisheries and Wildlife - Master of Science, 2025North American Beaver (Castor canadensis) build dams that cause flooding, frequently leading to conflict with humans. In areas with conflict, beavers are often removed lethally or to a lesser extent translocated to mitigate the conflict. However, recent interests in retaining ecosystem services provided by beavers (e.g., wetland wildlife habitat) has resulted in advancements of non-lethal management techniques, including use of flow devices. Whereas effectiveness of flow devices for controlling water levels has been evaluated, no studies have quantified how beavers respond to installation of these devices. Ideally, flow devices control flooding and thereby reduce human conflict, and the beavers remain resident to provide ecosystem services. We live-captured and radio-tagged beavers at five sites in southern Michigan on low gradient, generally sandy and silty bottomed areas with water flow. We monitored beaver behavior and locations using trail cameras and radiotelemetry in response to flow device installation. We found that all beaver colonies remained active after installation of flow devices. Initial 5-day behavioral responses of beavers varied, with 2 colonies increasing nightly detections and maintenance of dams while others avoided the area for up to 55 days, suggesting a level of neophobia. However, after initial investigation of flow device and varied short-term responses, 4 of 5 beaver colonies attempted to prevent outflow by plugging the outlet or building another dam downstream requiring flow device maintenance or modification. Downstream water conditions including turbidity, bedload, and temperature did not change following flow device installation. Our results support the utility of flow devices to prevent flooding while keeping beaver resident. We also show that beaver behavior varies considerably and that practitioners installing flow devices need to monitor, maintain, and modify flow devices to ensure continued functionality.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
EXPLORING CULTURALLY-CENTERED ICTS FOR MENTAL HEALTH SUPPORT : UNDERSTANDING THE ROLE OF ISLAMIC CULTURE AND RELIGIOSITY IN MENTAL HEALTH HELP-SEEKING AMONG MUSLIMS IN THE U.S.
No full textThesis (Ph.D.)--Michigan State University. Information and Media - Doctor of Philosophy, 2025This dissertation examines mental health help-seeking practices among Muslims in the United States. Mental health disparities persist among racial, ethnic, and religious minorities in the United States, with Muslims living in the U.S. (MLUS) facing unique challenges shaped by intersecting factors such as Islamophobia, discrimination, and acculturation stress. Despite a high prevalence of mental health concerns, MLUS underutilize professional psychological services, a pattern influenced by cultural stigma, religious beliefs, and systemic barriers including lack of culturally competent care, financial constraints, and institutional mistrust.Guided by the Culture Centered Approach (CCA), this dissertation amplifies the voices of MLUS to explore how Islamic culture and religiosity shape perceptions of mental health, help-seeking behaviors, and engagement with information and communication technologies (ICTs) as potential support tools. Using a qualitative, hermeneutic phenomenological design, the study draws on semi structured interviews with 22 Muslim adults in Michigan from diverse backgrounds. Key findings reveal that participants distinguish between cultural stigma and Islamic teachings, with stigma primarily rooted in ethnic or national cultures rather than religion. Islamic beliefs and practices serve as both protective coping mechanisms and frameworks for meaning making yet can also create feelings of guilt or inadequacy when religious coping alone is insufficient. Structural barriers, such as affordability, language, lack of culturally or religiously matched providers, and concerns about confidentiality further complicate both help-seeking and care. Participants demonstrate agency by seeking faith-aligned mental health content, utilizing digital tools for privacy and validation, and proposing culturally resonant solutions. ICTs, including teletherapy, mental health apps, and social media, are valued for their privacy, accessibility, and ability to provide judgment-free support, but concerns about data security and the lack of human connection remain. Participants advocate for faith-integrated digital interventions, mosque based mental health services, culturally competent providers, and leveraging community resources like zakat for mental health support. As defined by Muslim aid, \u201cZakat is a mandatory charity required of all adult Muslims who meet the threshold. It is the third pillar of Islam and refers to the obligatory payment of Zakat in Islam, distinguishing it from voluntary acts of generosity like Sadaqah. The Zakat definition encompasses the act of purifying one\u2019s wealth by giving a portion of it to those in need, as prescribed by Islamic law\u201d. This study advances the CCA by foregrounding the intersection of Islamic culture, religiosity, and structural realities in shaping mental health help-seeking and care experiences among MLUS. It highlights the need for culturally grounded, community-informed mental health practices and digital solutions that are responsive to the diverse needs of Muslim communities in the U.S. The findings have practical implications for practitioners, policymakers, and technology developers seeking to improve mental health equity and access for religious minoritiesDescription based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
THE EXTENT OF TREES OUTSIDE FORESTS USING SENTINEL-2 AND CANOPY HEIGHT DATA : THE CENTRAL DRY ZONE OF MYANMAR
No full textThesis (M.S.)--Michigan State University. Forestry - Master of Science, 2025Trees Outside Forests (TOF) provide important ecosystem services such as carbon sequestration, regulating temperature, and supporting rural livelihoods. They are especially important in places with low forest cover, such as in Myanmar\u2019s Central Dry Zone (CDZ). However, there is no baseline data for TOF in this region, and they are overlooked in national forest monitoring. The primary objective of this study was to create the first map of TOF extent in Chauk and Nyaung\u2010U Townships in Myanmar\u2019s CDZ and to check if adding canopy height data improves tree cover mapping. Sentinel\u20102 satellite images were used to measure fractional vegetation cover from the Normalized Difference Vegetation Index (NDVI). These data were combined with a global canopy height map to separate trees from crops and other vegetation. Forest areas were removed using the JRC Global Forest Cover dataset to create a TOF extent map. Very High Resolution (VHR) satellite imagery was used to assess accuracy. The results revealed that tree cover was about 14% of Chauk Township area and 20% of Nyaung-U Township area. Total TOF extent in the study area was 35,827.54 ha (15% of study area) and approximately 82% of total tree cover of the study area. It highlights the significance of TOF in both townships. Regarding tree cover mapping, adding canopy height data greatly improved accuracy: R2 increased from 0.19 to 0.79, RMSE% decreased from 241.6% to 36.86%, and bias changed from strong overestimation (+225.5%) to slight underestimation (-21.53%). In addition, canopy height data integration improved overall accuracy from 45% to 80%. This study presents the first spatially detailed extent of TOF in Myanmar. This study also provides a low\u2010cost and repeatable way to map TOF extent and provides baseline data of TOF to support land use planning, agroforestry programs, and climate change strategies in Myanmar.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
DOMAIN ADAPTIVE 3D OBJECT DETECTION FOR AUTONOMOUS VEHICLES
No full textThesis (Ph.D.)--Michigan State University. Electrical and Computer Engineering - Doctor of Philosophy, 2025Object detection using LiDAR point clouds relies on a large amount of human-annotated samples when training the underlying detectors' deep neural networks. However, generating 3D bounding box annotation for a large-scale dataset could be costly and time-consuming. Alternatively, Unsupervised Domain Adaptation (UDA) enables a given object detector to operate on a novel new data, with unlabeled training dataset, by transferring the knowledge learned from training labeled source domain data to the new unlabeled target domain. Pseudo-label strategies, which involve training the 3D object detector using target-domain predicted bounding boxes generated by a pre-trained model, are commonly used in UDA. However, these pseudo-labels often introduce noise, negatively impacting performance. Furthermore, current UDA 3D object detection methods tend to overlook the inherent geometric information in 3D point clouds, as well as valuable information from other modalities, such as images. In this dissertation, our objective is to address three key limitations and shortcomings of state-of-the-art unsupervised domain adaptation approaches for point-cloud based 3D object detection. We mitigate these limitations and shortcomings by developing three novel frameworks that: (a) Denoise pseudo labels generated by pre-trained models; (b) Extract and exploit 3D geometric information; and (c) Employ and inject 2D image information into 3D point clouds to improve LiDAR-based object detection in the target domain. Under our first contribution, we developed a new framework, DALI, which addresses two common forms of pseudo-label noise: distribution-level (global level) and instance-level (object level) types of distortions. In particular, DALI addresses the domain gab between different datasets that exhibit object sizes' variations, which results in target-domain noisy pseudo labels. In that context, DALI introduces a post-training size normalization strategy that reduces distribution-level noise by identifying an optimal unbiased scale after training. To mitigate instance-level noise, we build libraries of 3D models and LiDAR sensors and propose two pseudo point cloud generation methods \u2014 ray-constrained and constraint-free - that simulate LiDAR scanning to produce cleaner pseudo point clouds. To explore the benefits of geometric information in 3D point cloud, we developed our second contribution, ENGIN. Here, we propose an unsupervised geometric size estimation method that infers average object dimensions from accumulated point clouds of predicted boxes. We then design a geometric diffusion network to generate denoised, augmented point clouds for target-domain pseudo labels. ENGIN also introduces a BEV domain classifier with a domain embedding strategy to enhance joint training on source and target samples.To take advantage of valuable information from other modalities, such as images, in our third contribution, CMOS, we propose a framework for unsupervised domain adaptation in 3D object detection by leveraging cross-modality semantic information from images. The method jointly trains a 2D segmentor and a 3D detector using only 3D bounding box labels. Key components include bidirectional semantic injection (3D-to-2D and 2D-to-3D), semantic-consistent augmentation for aligned data transformations, and mutual-modality filtering to reduce false positives. We evaluated DALI, ENGIN, and CMOS on widely used public datasets, including KITTI, nuScenes, and Waymo. The superior experimental results demonstrate the effectiveness of the proposed frameworks.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
WHITE-TAILED DEER RESPONSES TO ALTERNATIVE SILVICULTURAL APPROACHES FOR PROMOTING DIVERSITY AND SUSTAINABILITY OF MICHIGAN\u2019S NORTHERN HARDWOOD FORESTS
No full textThesis (Ph.D.)--Michigan State University. Fisheries and Wildlife - Doctor of Philosophy, 2025This study evaluated wildlife responses to varying silvicultural treatments in northern hardwood forests, with a primary focus on the effects of different treatments on white-tailed deer herbivory of desirable tree regeneration. I quantified how white-tailed deer use responded to silvicultural treatments across the northern hardwood region of Michigan and how physical barriers potentially alter deer use and behavior. Chapter 1 provides a general introduction to white-tailed deer, silviculture, and northern hardwood forests, with a focus on Michigan. Chapter 2 provides a review of current and historical literature, with a particular emphasis on eastern deciduous forests in North America. Within this review, I extracted data from past research, summarized research across North American wildlife species, and examined the impacts that different silviculture had on herbivory and granivory. Part of my analysis involved investigating the effects of white-tailed deer herbivory on fast- and slow-growing trees, as well as comparing silviculture methods. I found that herbivory has been studied more extensively than granivory, and it reduces the recruitment of woody species in eastern deciduous forests in North America. However, silviculture can help mitigate the negative effects of herbivory and increase recruitment. Chapter 3 examines the response of white-tailed deer to alternative silviculture approaches as part of a landscape-scale manipulative study conducted across Michigan. The efficacy of these silviculture approaches in regenerating desirable, diverse woody species in the presence of deer is critical to sustaining northern hardwood forests. I quantified deer use in the research sites and the surrounding landscape. I found that deer use across silviculture treatments initially increased but returned to pre-treatment levels by three years post harvest in three out of four overstory treatment types. In Chapter 4, I further examined how one understory treatment, leaving tree tops, influenced deer movements and browsing impacts throughout the northern hardwood sites. This chapter evaluated how leaving tops could act as a natural exclusion to deer herbivory, using snow tracking, exclosures, and tree plantings. I found that leaving tree tops can deter deer movement, protect planted palatable tree species, and is equivalent to exclosures in protecting single saplings and recruits in managed northern hardwood forests. In conclusion, Chapter 5 combines these findings, which, taken together, indicate that deer use across the northern hardwood forest landscape can be influenced by forestry practices, allowing diverse, healthy northern hardwood forests to regenerate and recruit even in the presence of eastern North America's largest and most abundant herbivore.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
The Fitness\u2013Sleep Interaction on Hippocampal-Dependent Memory : Evidence for the Utility of Multidimensional Sleep Health Metrics
No full textThesis (Ph.D.)--Michigan State University. Kinesiology - Doctor of Philosophy, 2025Health behaviors and their related attributes can influence structural and functional dynamics of the hippocampus, potentially modulating relational memory systems that depend on its function. While higher aerobic fitness is reliably associated with enhanced hippocampal integrity and relational memory, these relationships are not well characterized in young adults. Sleep health appears to influence hippocampal structure in a comparable manner; however, its independent and interactive contributions to relational memory function remain poorly understood, in part due to a lack of consensus on how to appropriately define sleep health. Accordingly, the present study examined how aerobic fitness and sleep health jointly relate to hippocampal-dependent memory in healthy young adults, exploring how these associations may vary across distinct operationalizations of sleep health. A sample of 210 young adults (18 \u2013 25 years old) completed a maximal oxygen consumption (VO2 max) test to index aerobic fitness and a spatial reconstruction task known to engage hippocampal-dependent relational memory. Sleep health was measured via retrospective self-report using the Pittsburgh Sleep Quality Index (PSQI), from which three sleep health metrics were derived: sleep duration, overall sleep quality (i.e., PSQI global score), and a composite score based on the SATED framework, intended to better capture the multidimensional nature of sleep health. As expected, higher aerobic fitness was associated with better task performance; however, sleep health metrics were only indirectly related to task performance through self-paced encoding duration. Notably, a significant interaction emerged between aerobic fitness and the SATED composite, such that poorer sleep health attenuated the fitness-memory association. These findings underscore the utility of multidimensional sleep health metrics in clarifying the relationship between aerobic fitness and hippocampal-dependent processes, offering critical guidance for both study design and intervention development.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
THE ROLES OF CHLORIDE AND HYDROXYL GROUP DENSITY IN THE MEDIATION OF HYDROGEN BOND NETWORKS AND THE FORMATION OF MICROENVIRONMENTS IN DEEP EUTECTIC SOLVENTS
No full textThesis (Ph.D.)--Michigan State University. Chemistry - Doctor of Philosophy, 2025ABSTRACTThere is a pressing need for greener solvents in academic and industrial spheres. Many solvents are volatile organic compounds (VOCs), which are lethal, persistent pollutants. The simplest method to address this is to substitute these solvents with greener ones. Deep eutectic solvents (DESs) are relatively new (ca. 2003), green solvents composed of at least two components which, on mixing, produce a free-flowing fluid below the melting point of either constituent. DESs have low volatility and low flammability relative to their VOC counterparts, with the added advantage of tunable polarities.DESs are thought to form due to hydrogen bonding interactions between constituents. DES components are designated as hydrogen bond acceptors (HBAs) and hydrogen bond donors (HBDs). DESs offer incredible versatility, as many organic and inorganic compounds are capable of hydrogen bond acceptance or donation. Prior to directly replacing VOCs with DESs, interactions between constituents must be well-categorized to predict solvent interactions, allowing for tailoring DESs to specific needs. We present an investigation of the molecular interactions between charged (anionic, cationic) and neutral domains of three choline chloride (ChCl)-based Type III DESs using fluorescence anisotropy decay measurements. We measure the composition-dependent rotational diffusion dynamics of chromophores, perylene (neutral), oxazine 725 (cationic), disodium fluorescein (anionic), and rose bengal (anionic) in a series of ChCl:ethylene glycol (ChCl:EG), ChCl:glycerol (ChCl:Gly) and ChCl:1,3-propanediol (ChCl:PD) deep eutectic solvents (DESs). We find that the rotational diffusion dynamics of these chromophores do not correlate with the bulk viscosities of these DES compositions. We specifically address the roles of the anionic component of the HBA in these mixtures (ChCl) and the effects of hydroxyl group density on the HBD (ethylene glycol, glycerol, 1,3-propanediol). We find that the Cl- moiety is pivotal in solvent structure mediation, as the domain containing Cl- demonstrates the strongest degree of interaction. Our results indicate that greater quantities of -OH groups result in the formation of strong hydrogen bonding networks, which favor self-association below ~15 mol% ChCl. Increasing the length of the aliphatic chain between -OH groups results in stronger hydrogen bonding interactions, which likely originate from increased van der Waal\u2019s (VDW) interactions. Our measurements confirm these solvents are all dynamically heterogeneous, and indicate the existence of microenvironments distinct from that predicted by the bulk.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
THE IMPACT OF FULL-COST INSURANCE FOR WHEAT ON PESTICIDE USE : EVIDENCE FROM RURAL CHINA
No full textThesis (M.S.)--Michigan State University. Agricultural, Food and Resource Economics - Master of Science, 2025In China, full-cost insurance has become the dominant form of crop insurance for staple grains since 2024. This type of insurance covers both direct input expenses and indirect costs. Despite its rapid expansion, empirical evidence on its impact on agricultural input use, particularly pesticide application, remains limited.This study examines how full-cost insurance for wheat influences pesticide use among farmers in Jiangsu Province, China. Exploiting a quasi-natural experiment implemented in multiple cities, we analyze an unbalanced panel of 687 wheat-producing households surveyed from 2019 to 2022. Using a Difference-in-Differences strategy, we find that full-cost insurance significantly reduces pesticide use by approximately 9.4%, reflecting risk-reducing behavior. Complementary analysis of pesticide efficiency further shows that full-cost insurance lowers the marginal productivity of pesticide use, providing evidence of moral hazard.We also investigate the role of structural constraints. Using a Difference-in-Difference-in-Difference approach, we find that pesticide reductions are smaller among credit-constrained households, while unconstrained farmers reduce more, consistent with stronger risk-reducing effects. Additional heterogeneity checks reveal that policy impacts are stronger among larger farms, while households recently affected by pest or weed disasters are less responsive.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references
EXPLORING THE EFFECTS OF CLIMATE CHANGE ON STREAM TEMPERATURE UNDER THREE FUTURE CLIMATE SCENARIOS
No full textThesis (M.S.)--Michigan State University. Geological Sciences - Master of Science, 2025The Laurentian Great Lakes Basin is one of the largest freshwater systems in the world and is home to a variety of valuable habitats. Among these is fish habitat provided by streams throughout the basin, which are typically categorized based on temperature. Of key interest are streams classified as cold water that provide habitat for fish such as trout and salmon which are fed in large part by groundwater seeps that can create localized cold water patches and serve to moderate temperatures at larger scales. However, warming summer temperatures caused by climate change threaten this cold water habitat. Stream temperature models can help to identify the level of mitigation needed and where within a stream network management may be required. Here, a stream temperature model is used to evaluate this potential future warming for one such cold water stream in Michigan\u2019s Lower Peninsula, the White River. This stream was selected due to its position near the southern extent of Michigan\u2019s cold water streams. The primary research question addressed was; what is the status of cold water fish habitat in the White River watershed and where this may change with future warming. Chapter 1 explores the types of existing stream temperature models and the known relationships between groundwater and stream temperature. Chapter 2 first constructs a stream temperature model for the present day, and then compares future conditions under three different climate pathways to the present day. Results from Chapter 1 suggest that an energy balance model would be the most effective for future stream temperature conditions. This review also points to gaps in the existing literature on groundwater temperature. Present day conditions modeled in Chapter 2 suggest that only 3% of the stream network exceeded suitable temperatures. Future conditions point to a loss of cold water habitat under all three scenarios. Under the low emission scenario, cold water remained the dominant habitat type, while under medium (high) emission this switched to cool (warm) water being dominant.Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).Includes bibliographical references