Skip to main content Skip to secondary navigation

Undergraduate Research Symposium 2023

Main content start

Friday, October 20, 2023
Hartley Conference Center and Patio
Mitchell Earth Sciences Building

Research presented by students of the SDSS undergraduate research programs:

Program Schedule

2:00-3:00 pm - Select oral presentations in Hartley Conference Center
3:00-5:00 pm - Poster presentations on the Patio
5:00 pm - dinner as part of the SDSS Alumni Awards Reception and Dinner - RSVP Required

Oral Presentations

  • 2:00 Supply Chains and Social Responsibility Assessment in South Indian Fishery Improvement Projects (FIPs)  Aditi Prakash, MUIR
  • 2:15 Understanding Cover Crop Decomposition in The Salinas Valley To Reduce Nitrate Leaching Diego Gutierrez, SESUR
  • 2:30 Designing a Fixture for Precise, Local Electrochemistry on Strained Thin-Film Membranes, Michael Pascal, SUPER
  • 2:45 A GIS-based Multicriteria Decision Analysis Tool for Siting Utility-Scale Solar in the Navajo Nation, Ryne Zen-Zhi Reger, SUPER


Supply Chains and Social Responsibility Assessment in South Indian Fishery Improvement Projects (FIPs)
Aditi Prakash, MUIR

Fishery Improvement Projects (FIPs) are programs designed to guide fisheries in gaining sustainability certification by the Marine Stewardship Council (MSC). These FIPs focus on a fishery and address the human and ecological issues in the area. The focus of this research was to find the current status of labor rights in South Indian fisheries and to document the supply chain involved in two specific FIPs. This is the first step in evaluating FIPs as a sustainability certification process. A desk-based review was conducted utilizing Conservation International’s Social Responsibility Assessment framework, and a supply chain map for the Kerala shrimp industry and the Tamil Nadu blue swimming crab industry was created.
We found that Kerala is usually one of the first to pass legislation to help its people and has seen great improvements in many areas of quality of life. However, this prosperity does not always extend to its fisheries, as they do not experience the rest of the state's high “income, land holdings, housing quality, health conditions, and literacy levels”. Another defining aspect of Kerala’s fisheries sector is the heavy presence of interstate migrant workers, who experience the benefits of Kerala’s higher quality of life, although they remain a disadvantaged group.
Palk Bay (Tamil Nadu) has high biodiversity and a rich history. Until the creation of the International Maritime Boundary Line (IMBL), fishermen from both the Indian and Sri Lankan sides of the bay fished freely. The creation of the IMBL along with the fallout from the Sri Lankan Civil War created a uniquely tense political environment surrounding Palk Bay and fishing rights. Tamil Nadu’s fisheries are governed with a distinct legal pluralism, as many different state departments, local traditionally elected groups like uur panchayats, and stakeholder groups like boat owners associations are all interacting. Tamil Nadu recently instated a four-tier system of co-management to address the many different governing bodies involved and to include local people in decision-making and enforcement formally.
The Social Responsibility Assessment revealed that India has ratified only six of the ten relevant ILO conventions for fisheries, and this absence of protection is seen in legislation beyond the ILO conventions, and in poor enforcement of the protections that do exist. There are many governing bodies involved in fisheries, each with its own set of regulations, which complicates enforcement significantly. In addition, Indian enforcement often lacks the manpower or ability to protect workers or the environment. Further research, such as conducting interviews with local fishermen is the next step in evaluating the effectiveness of FIPs and what social impact they have.

Understanding Cover Crop Decomposition in The Salinas Valley To Reduce Nitrate Leaching
Diego Gutierrez, SESUR
In California’s Salinas Valley, the groundwater is polluted with nitrate due to the high production of leafy green vegetables and the heavy amount of fertilizers applied. To prevent nitrate leaching, the local water board has passed a policy that incentivizes cover crops, or crops grown in the off-season, in this case, to scavenge nitrogen leftover from the previous crop. It is known that cover crops can help prevent nitrate leaching, however, there exists a knowledge gap regarding the effect of the C: N ratio, residual soil nitrogen, and soil temperature on cover crop decomposition. The goal of this study was to understand the decomposition process of cover crops under various climate and crop termination conditions to inform the new policy. This study is a lab incubation at three temperatures (10, 15, and 20C) lasting 112 days with 432 tubes, filled with soil, water, or ammonium nitrate, and cover crops terminated at various times to study the effect of C: N ratios with destructive samples taken bi-weekly. The destructive samples were used to measure the amount of plant available nitrogen (PAN) which is the sum of nitrate-nitrogen plus ammonium-nitrogen, a reflection of nitrogen susceptible to groundwater leaching. The results from the first 57 days of the experiment are presented and show that in cover crops with a lower carbon-to-nitrogen ratio, which would indicate an earlier termination date, more PAN is released. The results also show that in treatments with residual nitrogen, meant to replicate leftover fertilizer or crop residue, the cover crop releases more PAN than without residual nitrogen. Our results can inform policymakers in setting cover crop guidelines that indicate an optimal C: N ratio and the preferable season for cover crop growth and termination for reduction of groundwater nitrate contamination.
Designing a Fixture for Precise, Local Electrochemistry on Strained Thin-Film Membranes
Michael Pascal, SUPER
Electrolysis has presented itself as a more sustainable alternative to traditional chemical and fuel synthesis processes. However, its large-scale viability is limited by the availability of rare-earth catalysts. The efficiency of low-cost catalysts can be improved by imposing certain strain states upon them, though the magnitude of these efficiency improvements has yet to be determined. In this project, a specialty fixture was designed to more flexibly perform electrochemical measurements on strained thin-film membranes. Unlike previous fixtures, this fixture is able to perform repeatable measurements at discrete points along the film (~0.05 mm2 contact points). A custom three-electrode measurement setup was fashioned from a syringe and needle, where electrolysis would be performed at the interface between the thin film, and the electrolyte droplet suspended at the needle’s tip. A three-axis linear rail system supports the entire electrolysis setup, allowing the needle/syringe device to translate independently in 3D space. The new fixture allows for more robust, flexible electrochemistry measurements and enables the ability to measure dynamic and heterogenous strain states on stretched thin-film catalyst membranes.
A GIS-based Multicriteria Decision Analysis Tool for Siting Utility-Scale Solar in the Navajo Nation
Ryne Zen-Zhi Reger, SUPER
Tribal lands hold an estimated 6.5% of the total national utility-scale solar technical potential in the contiguous United States. Current siting practices typically ignore the social and cultural impacts of solar development, which can delay or terminate utility-scale energy projects especially if a project infringes on individual livelihood and local land use practices. Through early engagement with the tribal government of the Navajo Nation, we combined a Geographic Information System with a Multicriteria Decision Analysis (GIS-MCDA) to determine the relative suitability of utility-scale solar development on Navajo tribal lands. This study demonstrates what is required to improve the social and cultural considerations in a decision-support tool and provides guidance on development of a framework for multiple decision-makers at the local community to central government level. An Analytical Hierarchy Process (AHP) was employed to allow for users of the siting tool to evaluate their preferences and relative weights of eight criteria consisting of environmental, social, and economic data. We demonstrate how site suitability varies depending on the objective of the user. The flexibility of the siting tool allows for decision makers from every level—household, chapter, and central government—to engage with the process of siting an energy project. Our siting tool provides accessible siting knowledge that empowers local communities to lead their own energy projects, while also contributing to the clean energy transition and energy sovereignty of the Navajo Nation. Improvements to the tool will require additional tribal consultation, and accurate data that incorporate social and cultural considerations in the siting tool, such as land designations, and cultural and heritage sites. The framework of the tool can be adapted to a web application with extensions to site community-scale solar, and wind energy, to support climate mitigation planning and the energy transition in the Navajo Nation.
Variance in Impact of Experiential Learning in Environmental Education
Aadya Joshi, MUIR
This study aimed to explore the impact of different teaching techniques when teaching about biodiversity. The purpose was to understand if experiential learning (active learning) was more useful when teaching environmental education to middle school children. The study took two middle school grades (Grade 7 and Grade 8) at a residential school in India, divided the grade in half and taught the same materials in two different formats to the different halves. This was done by giving the students an initial quiz, running the lecture/activity portion of the class and then giving a final quiz to understand how much they learned after. The study showed that Grade 7 seemed to have responded better to passive learning, while Grade 8 responded to active learning.
Ten-Earth PlanktoScope System to Observe Population Dynamics in Complex Ecosystems
Abby Cummings, CEE-VPUE
Marine phytoplankton are responsible for nearly half of the planet’s primary production — producing oxygen and fixing billions of metric tons of carbon dioxide each year — and are the basis of the marine food chain. In the midst of anthropogenic climate change, it is imperative to understand wide scale plankton dynamics in order to see how they are responding to changes in ocean temperatures, and what implications that may have for the planet at large. Efforts to research the complex ecological dynamics of plankton across oceans have been met by a lack of specialized instrumentation.
PlanktoScope is an open-source tool that provides a platform for high-throughput imaging of plankton. Its portability, ease of use, and low cost make it scalable for plankton biodiversity monitoring. PlanktoScope can be controlled by any Wifi-Enabled device, making it an accessible and practical tool for wide scale observation.
We configured ten of these PlanktoScopes into what we called the ten-earth system to monitor experimental, aquatic ecosystems collected at Jasper Ridge, and accrue a total of 400,000 images for analysis.
Biology of Arctic Sea Ice: Investigating Microalgae Community Composition with Varying Sea Ice Characteristics
Ali Palm, SESUR
Single-celled photosynthetic primary producers generate most of the fixed carbon that ultimately supports Arctic Ocean food webs and drives numerous biogeochemical cycles. The particulate organic carbon (POC) produced by primary producers supports both the pelagic and benthic ecosystems on shallow Arctic shelves, each of which supports unique assemblages of fish, birds, and mammals that humans rely on for food. In the sea-ice-covered Arctic Ocean, the first algal blooms during the Spring of each year are the sea-ice microalgae, which become the foundation of the Arctic Ocean food web. Specifically in the Chukchi Sea, high rates of net primary production (NPP) supports large populations of zooplankton, fish, seabirds, and marine mammals (Ershova et al. 2015, Kuletz et al. 2015, Logerwell et al. 2015, De Robertis et al. 2017, Moore & Kuletz 2019) and result in large export fluxes of organic matter that sustains rich benthic communities (Grebmeier et al. 1988, 2006, 2015, Lalande et al. 2007). Therefore changes in NPP of the Chukchi Sea could alter biogeochemical cycling and have complex effects on food quality and community structure of other organisms (Lyon & Mock, 2014). It is hypothesized that sea ice loss will lead to changes in the connection between the foundational sea-ice algae and benthic communities through alterations in timing, size and community composition of sea-ice microalgal blooms.
Using the Chukchi Sea as a case study, we looked at physical sea ice measurements such as ice thickness, snow depth, vertical profiles of temperature, salinity, brine volume and Chl a as well as microalgal community characteristics such as species composition, cell size and quantity at 12 different sea ice stations in the Chukchi Sea. Through further analysis of the collected data, we hope to better understand the physiological characteristics of sea-ice microalgae that are caused by different sea-ice characteristics. As sea-ice algae are the basis of the Arctic food web, in order to better understand how a changing Arctic and losses in sea-ice will affect the Arctic ecosystem and species that rely on it, it is needed to understand how changing sea-ice characteristics are related to microalgae community composition. Understanding this will help give an idea of the response that sea-ice algae have to changing environmental conditions.
State Estimation Techniques of Plasma Chemistry for Semiconductor Manufacturing
Alison Fajardo, SUPER
Plasma is the fourth state of matter. This highly energetic and ionized matter forms in the presence of high temperature or intense energy sources. Understanding the collective behavior between free electrons and ions, known as plasma physics, is important for engineering and science. Its use extends to semiconductor manufacturing, spacecraft propulsion, fusion energy, and more. Approaches through experimentation often find difficulty measuring plasma properties. To address this, using theory to create simulations through computational modeling is key. Data-driven computational modeling allows an understanding of physical and chemical processes. In the Plasma Dynamics Modeling Laboratory (PDML), optical emission spectroscopy data is used to track the behavior of ion and radical species in collisional and radiative processes. Taking the extended Kalman Filter (EKF) coupled with a 0-D plasma global model, the quantity and quality of data required to get reliable state estimation for parameters that are difficult to measure will be assessed. The recursive mathematical algorithm, EKF, is used to estimate the state of dynamics systems based on previous state estimates and experimental measurements.
Effect of Optical and Thermal Properties of Heterolayer Stacks on the Performance of Flash-Annealed Ferroelectric Memories for Energy-Efficient Computing
Avni Vat, SUPER
High-performance computing and data intensive application spaces are accelerating computing energy use at an unsustainable rate with current projections for energy usage reaching 21% of global electrical energy demand by 2035. Developing low switching energy and scalable non-volatile memory like Hafnium-Zirconium (HZO) ferroelectric field effect transistors (FeFETs) embedded into the CMOS back-end-of-line (BEOL) reduces energy consumption at the device level and through reduction in distances over which data is transferred.  Flash lamp annealing, a highly non-equilibrium millisecond thermal processing technique, is a promising way to crystallize HZO into the ferroelectric orthorhombic phase without damaging underlying BEOL metallization and dielectric layers.  This project aims to develop a quantitative understanding of the temperature variation with time and depth within the heterolayer stack by investigating the optical and thermal properties of the layers beneath the layer of interest. Optical data in the UV and visible range of thin-films of common BEOL materials before and after annealing was used to develop a time-dependent model of the temperature variation as a function of depth. Experimental spectra indicate changes in the absorptivity of HZO/TiN/HZO after crystallization of the stack due to changes in surface roughness possibly resulting from grain growth. This variation in absorptivity before and after annealing suggests that the optical properties are temperature-dependent, a notable observation in the context of temperature modelling. Improved accuracy from the updated experimental absorptivity values helps in determining ideal conditions for the HZO layer to crystallize in the ferroelectric orthorhombic phase.
Estimating historical rice methane inventory via pesticide records and remote sensing
Bhumikorn Kongtaveelert, MUIR
Rice agriculture has been identified as a major source of methane to the atmosphere, contributing roughly 10% of the global methane budget, due to anoxic soil conditions during extended inundation. There is increasing interest in modifying rice management practices to reduce GHG emissions. As a result, there is a need to better monitor the timing and duration of rice field flooding and drainage at high spatial and temporal resolution, during both the growing and winter seasons. Here, we analyze the potential and limitations of using a combination of synthetic aperture radar (e.g. Sentinel-1), multispectral imaging (e.g. Sentinel-2), and remotely sensed thermal signals to map current irrigation management choices during rice production, such as in California’s Central Valley. In addition, we use public records of pesticide applications in rice fields (e.g. from California Department of Pesticide Regulation), to infer flooding and drainage, as certain pesticides can only be applied when fields are drained and others require fields to remain flooded for set durations. Leveraging this combination of data sources, the project aims to improve upon classification methods of flooded and dry periods within United States rice cultivation regions to provide high resolution data on flooding dynamics. This has the potential to increase the accuracy of models of rice methane emissions to better inform policy decisions.
Using Pressure to Design Better Materials for Energy Applications
Camille Slagle, SUPER
Sr9/8TiS3 (STS) is a transition metal perovskite chalcogenide favorable for semiconductor and photonic applications.  Perovskite refers to the mineral structure ABX3, and a chalcogenide refers to the X3 as Selenium or Sulfur, in our case. The application of external pressure has proven to be a tool to tune structural and electronic properties without modifying chemical composition; thus, mechanical compression represents an effective tool to study how the properties of TMPCs can be optimized. We used a diamond anvil cell (DACs) to increase the pressure and Raman spectroscopy to study its effects on the sample’s vibrational spectrum. Pressure was increased at a constant rate and a Raman measurement was taken every few GPa up to 48.8 GPa. We observed peak splitting as the pressure increased, which we attribute to a Ti-S stretching mode. Learning more about how STS reacts and how it changes with pressure allows us to learn more about how Tuning structural and electronic properties by applying an external pressure allows for optimization of its material properties, and results can be used to understand applications of these materials in photovoltaics and lighting applications, as well as further inquiry into sustainable and green energy sources to combat rising climate and energy costs and demands.
The Generation of Toxic Hexavalent Chromium In California Soils During Wildfires
Charles Lamb, SESUR
Wildfires are increasing in frequency and severity globally. Among the hazardous toxic metals transported during wildfires is chromium (Cr), a metal that is prevalent in California’s soils. Under extreme temperatures, Cr(III) can oxidize to toxic Cr(VI), which is a carcinogen. Soil chemistry and mineralogy can control the formation of Cr(VI) during wildfires, which is poorly understood. We collected serpentine, greenstone, chert, and sandstone-derived soils from chaparral ecosystems at the Jasper Ridge Biological Preserve (San Mateo County, CA) and burned them at 200, 400, 600, and 800 ¬∞C for 2 h in a muffle furnace to simulate varying wildfire severities. We characterized samples using X-ray fluorescence spectrometry (XRF), UV-vis spectroscopy, scanning electron microscopy (SEM), and synchrotron X-ray absorption near edge spectroscopy (XANES). Total Cr concentrations were highest in the serpentine soils, followed by greenstone, chert, and sandstone. Total Cr(VI) and exchangeable Cr(VI) concentrations progressively increased up to 600 ¬∞C and then decreased at 800 ¬∞C in all samples. The results showed that the fraction of chromite (FeCr2O4) mineral slightly decreased up to 600 ¬∞C (63 ¬± 1.5  %) but increased to  91 ¬± 0.5  % at 800 ¬∞C. Chromite formation consumes the available Cr(III), explaining the lower formation of Cr(VI) at 800 ¬∞C. Exchangeable Cr(VI) percentage over total Cr at 600‚ÑÉ was positively correlated with total Fe concentration in the samples. Energy dispersive X-ray spectroscopy (EDS) of SEM images showed that Cr is mostly associated with Fe either in the structure of chromite or with Fe oxy-hydroxide minerals, suggesting that Fe minerals control the formation of Cr(VI) in soils. This information can inform targeted policies and interventions for mitigating respiratory health risks to communities disproportionately impacted by wildfires.
Inside The Black Box: Scientific Machine Learning for Infectious Disease Modeling          
Claire Morton, MUIR
The use of neural networks for parameter estimation in infectious disease models (Scientific Machine Learning, or SciML) is a recent development that may be promising but has not been rigorously evaluated. We use simulations to critically evaluate the ability of SciML to infer model parameters across three use cases. First, we test whether SciML can be used to learn social mixing patterns in a structured population. We find that, while SciML accurately detects the number of groups in simulated populations, it fails to learn the correct mixing parameters of the groups, especially for noisy input data. Second, we evaluate the ability of SciML to learn a time-varying quarantine rate function in a population. We find that SciML does not identify the optimal rate. In both of these cases, SciML fits input data well while finding incorrect underlying parameters, suggesting that SciML can be misleading when fitting epidemic data. Finally, we attempt to estimate the frequency of asymptotic cases in a simulated disease using only data on the symptomatic hospitalized population. We observe that SciML makes less accurate estimations compared to several existing methods. Overall, this research suggests that SciML is not as suitable for potential applications as previously thought. New studies seeking to rigorously use SciML for epidemic modeling should run their models on simulated data to test whether models can accurately infer underlying dynamics.
Soil Fungal Competition in Alaskan Boreal Forests: All’s Fair in Love and Spore
Ellie Fajer, SESUR
Soil fungi drive the accumulation and decomposition of soil organic matter, affecting a major carbon sink. However, predicting how soil carbon dynamics are impacted by biotic interactions, such as competition between diverse fungal groups, remains poorly understood. While some studies have supported the “Gadgil Effect,” which suggests competition between ectomycorrhizal and saprotrophic fungi reduces decomposition rates, others have found increases in decay. Further, significant knowledge gaps remain around how competition varies among fungal species and across climatological gradients. Using soil samples collected across a latitudinal gradient of Alaskan boreal forests, we studied the outcomes of competition between lineages of saprotrophic and ectomycorrhizal fungi. DNA sequencing results showed how climatic conditions and tree-level variables affect fungal community composition. We further used a series of network analyses to infer how competition is contingent on fungal physiology and phylogeny. Ultimately, we seek to better understand how fungal competition can affect soil carbon dynamics to refine carbon modeling and support carbon sequestration efforts.
Filling the Sustainable Development Goal 14 'Life Below Water' Funding Gap: Blended Finance as a Dark Horse, Microfinancing Barriers, & Metrics for Measuring Financial Resilience
Emily Zhao, MUIR
The author’s research takes sustainable finance, meaning the taking of environmental, social, and governance (ESG) considerations into account when making investment decisions in the financial sector, and applies it to marine systems and its actors. Currently, the United Nation’s Sustainable Development Goal (SDG) 14 ‘Life Below Water’ is the least funded SDG of all 17 goals, with a financial gap of $149.02 billion needed per year to implement SDG14. The author’s work revolved around how to fill the 85.39% SDG14 funding gap, resulting in three deliverables:
1. An internal memorandum to the Stanford Center for Ocean Solutions (COS) on the definition, potential, and barriers to applying blended finance. Key takeaways include the need for multilateral development banks (MDBs) to use more risk-mitigation products like guarantees and subordinated debt in their activities, credit rating agency reforms to increase leniency on banks with development mandates, increased transparency in blended finance transactions and scaling up of deal sizes, and improved understanding on financial returns for investments in blue nature-based solutions.
2. A slide deck on challenges and opportunities in microfinancing local innovation in blue foods, informing a World Economic Forum workshop with blue food investors at the 2023 SDG Summit in New York. Key challenges include limited capacity, onerous eligibility requirements and opaque application processes, and high risk for investors due to the fragmented nature of SSF operations and inadequate financial incentives.
3. A briefing on the author’s proposed metrics for measuring household financial resiliency for small-scale fishers (SSF). The financial resiliency’s framework has six dimensions: economic resources, financial inclusion, financial literacy, social capital, financial stress, and external/environmental conditions.
Estimating Plantar Pressure Distribution During Walking Using Footstep-Induced Structural Vibrations

Gait health monitoring assesses walking patterns through parameters such as stride length, joint kinematics, and gait stability to aid early diagnosis (e.g. muscular dystrophy), fall risk prediction, and physical rehabilitation (e.g. stroke, brain injury). In-home gait monitoring devices provide long-term, continuous, and convenient gait monitoring. Structural vibration-based gait monitoring is particularly robust to visual obstruction and privacy-friendly, compared to other in-home gait monitoring methods.
Gait stability, defined as walking balance and coordination, is a gait parameter that can be evaluated through plantar pressure distribution. Pressure asymmetry between feet may indicate muscular imbalance, joint dysfunction, or injury. Spatial and kinematic pressure parameters can be reliably extracted from in-sole pressure sensors or floor mats, but these methods greatly limit the range of in-home data collection. There is little research on the extraction of plantar pressure distribution from footstep-induced vibration data. Therefore, this project aims to expand upon existing gait health monitoring data predict center of pressure, loading force, and foot-off force from footstep-induced floor vibration features to determine plantar pressure distribution.
Gait monitoring data was recorded for sixteen subjects who exhibit a variety of gait abnormalities, including patterns consistent with hamstring, hip, knee and ankle injuries. A random forest regressor machine learning model was designed to predict pressure ground truths from vibration features. This model predicts pressure parameters based on vibration features with 93% accuracy (16 subjects), suggesting that footstep-induced floor vibration is a robust indicator of gait stability for a variety of gaits. Future work should expand upon modeling accuracy by considering between-person and floor-type variability, as well as a wider range of gait abnormalities.
Understanding Public Perceptions of Affordable Housing Using Eye-tracking
Huilan Huang, CEE-VPUE
California is facing a shortage of affordable housing statewide. One reason for this is the ability for citizens to voice their concerns at council meetings in order to prevent new affordable housing developments from being built in their communities. In order to further understand these attitudes, this study builds upon an existing survey which investigates how various labels may affect the ways proposed affordable housing developments are perceived by the public. The goal of the study is to assess how the designations affordable, historic, and sustainable impact conscious and unconscious evaluations of proposed housing developments. Findings from the study could be used to help inform local officials as they attempt to obtain the passage of more affordable housing.
In this study, eye-tracking software was embedded into an online survey in order to assess participants’ subconscious perceptions of affordable housing. The survey was internally piloted amongst members of the Stanford community. Preliminary results suggest that from the developments designated as sustainable, participants tended to gaze at the facade of the building more often when the development they were provided was labeled affordable, rather than sustainable or no label at all. This trend is observed less among the developments designated as historic. Limitations of this study include the small, limited sample size (N = 24) and some of the participants being aware of the manipulation that took place. Future work may include running the study with a larger sample size that is more representative of the population of California.
Deciphering the Role of Redox Conditions on Groundwater Arsenic Contamination in the San Luis Valley, Colorado
Jaden Southern, SESUR
Arsenic (As) is extremely hazardous to human health and a major threat to groundwater quality. Groundwater As contamination has been linked to myriad adverse health outcomes including cardiovascular disease, diabetes, and various forms of cancer. Many communities like that of the San Luis Valley, Colorado, USA, rely heavily on groundwater services for drinking water, and groundwater As contamination within the region has been linked to these negative health outcomes. While naturally occurring in soil and sediments, the primary drivers behind As release into groundwater can vary depending on the presence of myriad environmental triggers. Many naturally occurring factors can influence As levels in an aquifer including pH, redox conditions, and source rock mineralogy. Thus, further understanding of local drivers behind As release is essential to informing regional water management policy and decisions. In this study, I sought to determine the potential role of redox conditions on As mobilization in San Luis Valley groundwater. Determination of redox potential was done through regression and spatial analyses of well depth, percent clay composition, and dissolved As concentrations throughout the area. As measurements were derived from 496 community-provided water samples collected across the San Luis Valley. Clay occurrences were extrapolated from a regional well log database and matched to the location of water samples using geospatial kriging methods. Through extensive statistical and spatial analyses, I gained insight into the redox conditions of San Luis Valley groundwater and possible areas of concern for As mobilization. Significant trends were found between greater depth, higher fines percentages, and higher As levels. Further, preliminary exploration of spatial relationships show a qualitative correlation between higher percentages of clay in aquifer soil, deeper wells, and higher As levels. My results can guide further research into the effects of both redox conditions and clay sources on As concentrations in regions dependent on groundwater for drinking.
A Probabilistic Analysis of Sea-Level Rise Impact on Earthquake Liquefaction on the Island of Alameda
Jaelen Sobers, CEE-VPUE
Earthquake liquefaction occurs when an earthquake causes susceptible and sufficiently saturated soils to act as a liquid. The island of Alameda has been identified as an area currently at risk of earthquake liquefaction due to its proximity to two faults (the Hayward and San Andreas faults) while also having a large portion of its area being composed of artificial fill (young, loose soil). Moreover, liquefaction also relies on the amount of water in the soil. Considering the current expectations of sea-level rise and the nature of Alameda as an island, the impact of sea-level rise on soil saturation is expected to amplify the occurrence and severity of earthquake liquefaction. The severity of earthquake liquefaction is often quantized by an index called the Liquefaction Potential Index (LPI). As a result, this study aims to utilize probabilistic analysis to determine the expected LPIs at various points across the island at three different sea-level rise scenarios given that an earthquake does occur. Through analysing the patterns and differences in expected LPIs at these scenarios, we aim to identify how the hazard changes with sea-level rise.
Remote Sensing of Sugarcane Areas Globally
Jessie Kong, SESUR
Sugarcane is a crop grown in tropical and subtropical regions throughout the world. Because of its use in the production of sugar and ethanol, it is important to track its growth and land use. As more and more land is used for growing sugarcane, concerns rise about food and water security. Knowing where sugarcane is grown and how much is being produced is important to be able to appropriately allocate water to it. However, crop labels in many countries are sparse and government data is often outdated and inaccurate. Land surveys have proved to be particularly challenging because each sugarcane producing region has a different growth cycle [1]. The Global Ecosystem Dynamics Investigation (GEDI) is a NASA managed sensor on the International Space Station that develops 3D maps of forest canopy heights and biomass density. We propose that GEDI data can be used for crop detection based on height in agricultural regions that produce sugarcane. From our previous research, GEDI data has been shown to create relatively differentiations between different crop types. For this 10 week project, our research objectives were:
- Collect, organize, and analyze existing sugarcane data from governments of the top 10 sugarcane producing countries and other agricultural organizations.
- Create maps showing land used for sugarcane production.
- Compare government datasets with our own results.
By using GEDI to differentiate tall and short crops across crop cycles, we were able to compare government sugarcane land use statistics with GEDI developed sugarcane maps.
Digital and Physical Prototypes for Biophilic Illusions Project
Jonah Blaydes-Greenberg, CEE-VPUE
This summer, I had the privilege of working on the Biophilic Illusions Digital and Physical Prototyping Projects at the Billington Lab. This multidisciplinary research endeavor aimed to seamlessly blend technology and nature within workspace design, enhancing the overall user experience and well-being.
In the digital realm of the project, my responsibilities centered on the evaluation of technology acceptance. To accomplish this, I helped craft and refine a comprehensive Qualtrics Survey tailored to gauge users' acceptance of various modalities integrated into workspaces. The survey was designed to adhere to the Unified Theory of Acceptance and Use of Technology 2 (UTAUT 2) model, ensuring robust and insightful data collection. After collecting survey responses, I cleaned and analyzed the data, laying the foundation for in-depth insights into technology acceptance trends. Furthermore, I conducted regression models, providing statistical evidence to support decision-making processes that we ultimately used in informing our conclusions.
In the physical component of the project, the primary objective was to replicate qualities of nature within indoor spaces. To achieve this, the project used non-rhythmic variation, multi-sensory stimulation, and real-time information in order to convey accurate depictions of the outdoors, inside. This involved the collection of real-time wind data, including both direction and speed, via a wind detection device, which was then transmitted to an ESP 32 (an Arduino) via cloud technology. My role was first collecting this data, and then extending to manipulating this data to make it compatible with our image processing program, ensuring that it could effectively contribute to the creation of immersive natural illusions. Additionally, my job was to update and refine the image processing program to enhance the quality and realism of the nature-inspired visual effects.
My summer experience at the Billington Lab was both enriching and transformative, allowing me to delve into the realms of technology acceptance and nature integration within workspace design. Through digital survey creation, data analysis, and regression modeling, as well as hands-on work with real-time data and image processing, I  was able to learn more about the advancement of innovative approaches to enhancing the workplace environment, with the ultimate goal of improving worker well-being.
Sustainable Agriculture: Pantoea As A Biological Control Agent
Keona Blanks, MUIR
We aimed to study the effect of flower-colonizing bacteria on seed development in buckwheat. In preliminary experiments conducted in Nagano, Japan, we found a large increase in the probability of buckwheat seed development when their flowers were inoculated with Pantoea bacteria. This increase is likely due to the bacteria’s ability to enter floral tissues from nectar and grow there, thereby suppressing fungal pathogens that would otherwise prevent the seed from properly maturing. Given these findings, we pursued the possibility of using Pantoea bacteria as biological control agents to control fungal pathogens and improve buckwheat yield. We investigated the effectiveness of spraying three strains of Pantoea onto buckwheat inflorescences to increase seed production. We found that Pantoea marginally increased seed development in thrum flowers. Spraying rather than pipetting the bacteria treatment resulted in discrepancies between pin and thrum flower types due to anatomical differences. Pantoea thus has not proven to effectively increase seed development in Buckwheat. More evidence is needed for biological control agents to suppress fungal infections in crops.
Sit, Reflect, And Cool Down: Using Mirrors to Fight Climate Change          
Kyle Haslett
According to, the Earth’s temperature has been rising by an average of .14 degrees Fahrenheit per decade since 1880, for a total of about 2 degrees, with the 10 warmest years on record having been recorded since 2010. Harvard’s Dr. Ye Tao saw this as a pressing issue, and developed the Mirrors for Earth Energy Rebalancing (MEER) framework.
The basis of the framework centers the conversation about climate change around the temperature solely, not other factors like rising levels of CO2.
This summer I worked with Harvard researcher Erik Shriner to develop tools to evaluate the local benefits of using mirrors in an attempt to incentivize urban areas. The local benefits MEER considered are 1. A reduction of energy needed to cool your home or building with the addition of mirrors on the roof, and 2. Placing mirrors on bodies of water to increase water retention, and mitigate the rising water temperature. Water retention is an important issue.
According to an article published in Nature Communications, while artificial lakes and reservoirs account for 5% of global lake storage capacity, they are responsible for 16% of the evaporation volume. This places the annual reservoir evaporation loss at 20% of the global annual consumptive water use.
Not Presenting: Evaluating The Role of Nature-Based Coastal Vulnerability Solutions For Low-Income Residents In The Galveston Bay Area
Laia Bent, MUIR
The Galveston Bay Area (GBA) in southeast Texas has experienced many significant hurricanes since the turn of the century, prompting significant investment in coastal protection infrastructure. Traditional engineered solutions like floodwalls, dams, and levees require large upfront investments and long-term maintenance funds. Nature-based solutions (NBS), including habitat restoration and conservation, offer a lower cost and more resilient coastal protection option which may better serve low-income communities. In this work, we identified four habitats in the GBA that offer significant protection from coastal hazards and could therefore be utilized for NBS: coastal forests, marshes, dunes, and seagrasses. We first identified areas in the GBA that are at high risk for coastal hazards using the NatCap InVEST Coastal Vulnerability Model, which took the areas of each of the four protection habitats as inputs, generating a hazard exposure index and habitat protection index for each point along the shoreline. Using these indices, two types of risk areas were identified. Replace areas included points identified as having high exposure and low habitat protection, indicating that they would benefit from having their current habitat replaced with a higher protection habitat. Protect areas included points identified as having high exposure, high habitat protection, and high habitat loss, indicating that they would benefit from having their current high-protection habitat conserved. Both replace and protect points were then filtered for proximity to low-income areas and high-population areas, to find high-risk areas where NBS implementation would be most beneficial. The final replace/protect areas were then used to modify the four coastal protective habitat areas accordingly, and the modified habitats were re-run through the InVEST model to simulate whether NBS would impact exposure. Thus far, simulations where protect areas are conserved have shown significant impact on exposure, and further simulations will be done with input from local GBA organizations.
Life Cycle Impacts of Concrete Floors in Sirajganj, Bangladesh
Lauren Owens, CEE-VPUE
Previous observational studies have shown that houses with soil flooring in Bangladesh have higher incidences of childhood disease due to the survivability and removability of pathogens. A proposed method of reducing the disease incidence rate is the installation of concrete floors in these homes. This project, conducted in conjunction with the Stanford Department of Epidemiology, investigates the survivability and removability of Escherichia coli on various types of concrete flooring surfaces. The tested concrete flooring materials include standard concrete with ordinary Portland cement, standard mortar with ordinary Portland cement, and concrete with Portland cement and fly ash as a supplementary cementitious material. Along with investigating the ability of these concrete materials to prevent hosting bacteria, the environmental impact of the large-scale construction of these floors must also be considered. Concrete production accounts for almost 10% of global carbon emissions, so criteria for the determination of the optimal concrete material must also include the sustainability of the material. Using life cycle assessment software SimaPro, we created life cycle assessments for each type of flooring material to investigate total greenhouse gas emissions, use of energy resources, and production of solid waste. These life-cycle assessments account for byproducts of shipping materials to Sirajganj, the experiment location; byproducts of material production; and byproducts of flooring production.
An Analysis of The Correlation Between Land Snail Size Extremes, Landmass Area, And Number Of Species
Lauren Taylor, SESUR
The question of why the largest and smallest species within a higher taxon have the maximum and minimum body sizes that they do is complex, with many factors potentially at play. Two factors that may influence size extremes on a given landmass are the size of the landmass and the number of species on that landmass. Previous work on this topic has focused on large vertebrates, which may not be representative of animals more broadly, and also failed to account for the fact that landmass area and number of species are correlated to each other. I analyzed the relationship between land snail size extremes, landmass area, and number of species in order to expand this work to smaller invertebrates. I used regression analysis to analyze these relationships in order to account for the correlation between landmass area and number of species. This work will contribute to the overall understanding of what factors determine body size.
Greenhouse gas emissions from contaminated rice paddies: from environmental geochemistry to environmental Justice
Natalia Armenta, SESUR
Rice paddies are one of the largest sources of atmospheric methane, leading to increased global warming. It is critical that we are able to develop an understanding and ability to manage production within rice cropping systems that reduces the amount of methane emissions. A primary means to limit methane emissions is by restricting methanogen (the organisms responsible for methane production) activity either through microbial competition or by restricting their food sources. An exciting opportunity arises by managing iron cycling within rice paddies, particularly through the formation of small, rust-like particles known as Ferrihydrite. Through this research we looked at two different types of rice irrigation methods, one being alternate wetting and drying and second being continuously flooded. Through the exploration of these two irrigation methods under different climate conditions We were able to analyze the nano behaviors of Ferrihydrite with its correlation to arsenic accumulation. Ferrihydrite is an iron oxide nanoparticle widespread in soils, sediment, and water. In aqueous environments such as rice paddies, ferrihydrite shows a high affinity for interaction with organic matter that serves as a food source for methanogens. Natural groundwater used for rice irrigation can contain nutrients and containments such as arsenic, phosphate, and nitrate. These nutrients/contaminates compete with organic compounds for adsorption on the surface of ferrihydrite, affecting the production of greenhouse gases. Through the analysis of Ferrihydrite, plant physiology, organic compounds, and contaminants we are able to understand how these reactions control greenhouse gas emissions within rice paddies.
Designing Data Collection Platforms for Digital Wellbeing Support
Niklas Vainio, CEE-VPUE
This poster covers two related projects. The first project, entitled 'Hybrid Physical + Digital Spaces' (HPDS), aims to investigate how the design of interior spaces impacts occupant wellbeing. In service of this goal, we developed a platform that synthesizes wellbeing data from various sources, in order to provide a quantitative understanding of how users respond to different room designs. The platform was successfully deployed in a 4-week pilot study, the conclusions of which are detailed and analyzed. The second project, named Prisma (Personal Reflection and Insight Support Machine), aims to improve how users interpret and act upon their wellbeing data. The system takes the skeleton of the HPDS platform, and adds a LLM-based conversational interface, whereby users can ask natural language questions about their data and habits. The design of this system, as well as future goals for the platform are described in detail.
Protecting our Reefs, One Reef-Safe Sunscreen at a Time
Piper Fleming, MUIR
Sunscreens are widely acknowledged to harm reefs and marine environments, thus motivating the need for effective yet reef-safe sunscreens. Previous clinical studies have shown that filamentous phage Pf, a virus that infects the bacterium Pseudomonas, absorbs UV rays; this experiment establishes its lack of toxicity upon the larvae of Dendraster Excentricus, a species of sand dollar. We exposed embryos to concentrations of 50 and 100 µL/L of both sunscreen and Pf, and measured the percentage of abnormally developed embryos at 0, 3, and 30 hours after fertilization. While currently marketed sunscreens had increasingly deleterious effects with concentration, we observed no relationship between any exposure and toxicity for Pf. As such, biological sunscreens based on ingredients like the Pf phage may be the answer to simultaneously protecting our skin from UV rays as well as keeping harmful chemicals out of both our bodies and our waterways.
Identifying Pathways to Distributive Equity In California MPA Management To Build Adaptive Capacity Amongst Vulnerable And Underserved Coastal Communities
Plengrhambhai Snidvongs Kruesopon, SESUR
Global change is accelerating at a rapid pace, immensely altering the marine environment. Hence, it is increasingly essential to have established policies that mitigate the environmental and socioeconomic factors that impact vulnerable coastal communities. However, many of the current technocratic approaches that support climate adaptation have narrowly focused on the assurance of optimal resource allocation, neglecting the range of social and cultural dynamics that influence the livelihoods, values, rights, and needs of disadvantaged populations. As marine social-ecological systems evolve and transform, it is essential that these nuanced human dimensions are accounted for, otherwise, these interventions risk reinforcing systemic inequalities and injustices that are associated with an uneven climate vulnerability and adaptive capacity.
Thus, this project explores the intersection of ocean ecosystems, marine protected areas (MPAs), ocean access, and climate change in order to ensure that the benefits of healthy and sustainable ocean ecosystems are equitably distributed throughout society. Through this project, the research team intends to develop policy recommendations and culturally relevant guidance for MPA managers in order to both increase long-term ocean benefits and reduce the impacts of climate change, building the resilience of diverse communities.
This research initiative is a two-year state-funded ocean equity project, endorsed by the California Ocean Protection Council. At the Hopkins Marine Station, I collaborated with marine scientists from the National Oceanic and Atmospheric Administration (NOAA), and the MPA Collaborative Network. I was tasked with developing a valuation framework that encapsulated the dynamic and diverse relationships of socio-ecological systems. I also designed and conducted semi-structured interviews by engaging with underrepresented coastal communities, documenting shifting patterns of access, activity, and benefits observed by these disadvantaged populations over time.
Earth’s History in a Grain of Sand: Shape and Textural Analyses of Fluvial Sand Grains in Experimental and Natural Settings
Raisha Abubo, SESUR
As sand is transported across Earth’s landscapes, individual grains are uniquely molded and textured by the various environments they traverse. Thus, grain shapes and textures from lithified sandstones provide insight into Earth’s past, helping us understand how landscapes have transformed over time. Most surface texture analyses have been conducted on quartz grains, as quartz is relatively durable, able to retain surficial textures, and is found in most environments on Earth. However, such analyses become more challenging as we look further back in time, as quartz becomes more susceptible to alteration (e.g., recrystallization) after becoming part of the rock record. Recent work has shown that zircon grains hold similar textures, which is important because zircon is much more durable and is much more resistant to recrystallization. However, much is still unknown about the role of transport distance on zircon surface texture development, a problem that our research hopes to solve. My goal for this project was to simulate natural river sediment transport and grain abrasion in the lab using abrasion mills, then compare these experimental grain shapes and textures to the natural textures and shapes of sand grains sampled along the length of the Mississippi River. Grains textures were imaged using a Secondary Electron Microscope (SEM), and shape and size distributions were quantified using a particle analyzer (CamSizer X2). By quantifying the role of transport distance on zircon surface micro-textures and grain shapes, our goal is to develop a new tool for paleoenvironmental reconstruction that can be applied to older rocks on Earth and on other planets.
Geothermal Contributions to Groundwater Arsenic Contamination in the San Luis Valley, CO, USA
Rebecca Liebson, SESUR
The San Luis Valley (SLV), CO, USA is an arid basin highly dependent on groundwater resources for drinking water and agriculture. However, heterogeneously distributed arsenic (As) concentrations throughout the valley threaten safe supply. Variations in groundwater pH and redox conditions can drive As mobilization, but they do not fully explain As concentrations throughout the SLV. Geothermal activity can also result in elevated As concentrations through hot fluid extraction from reservoir rocks followed by ascent and mixing with groundwater, frequently resulting in co-occurrence of contaminants such as lithium (Li), boron (B), and fluoride (F). The SLV is located within the Rio Grande rift and is geothermally active, but it is unclear to what degree this activity influences groundwater As concentrations. Using statistical approaches, we investigated the potential contributions of geothermal activity to regional As contamination. We spatially integrated results of extensive community-engaged groundwater sampling (e.g., As, Li, B) and known local geothermal features (e.g., hot springs, thermal wells). We observed a positive association between dissolved As and Li and B concentrations. Further, a number of groundwater wells with As concentrations greater than 20 μg/L were distributed proximal to geothermal features (e.g., < 20 km). We developed a random forest regression model incorporating 18 major and trace element concentrations, pH, averaged subsurface clay content, and distance to the nearest geothermal feature with r2=0.94 for all training data and r2=0.54 for testing data. We found that potassium (K) concentration was the most important predictor variable, followed by tungsten (W), B, magnesium (Mg), and Li. The relatively high importance of Li and B points to the potential for geothermal influence on As concentrations, operating alongside other processes. Our findings contribute to an improved understanding of environmental controls on groundwater As concentrations within the SLV and aid in our spatial predictions of As contamination.
Understanding Multidecadal Changes in Greenland’s Internal Layering using Archival Radargrams
Reese Dobson, SESUR
Though mass loss from ice sheets is a key causal factor of sea level rise, there is significant uncertainty in model-based projections of future sea level. One key reason for this uncertainty is the spatial and temporal sparsity of subsurface data available. With the recent availability of archival data, there is a new opportunity to expand our understanding of multidecadal subglacial processes in some regions of the Greenland Ice Sheet. For example, recent studies have found that the North Eastern Greenland Ice Sheet (NEGIS) may contribute to sea level rise more significantly than previously modeled. As Greenland’s largest region of fast-flowing ice, or ice stream, NEGIS covers around 12% of the ice sheet. The volume of ice it contains when melted equates to a 1.1-m sea-level-rise.
Here I present the comparison of the recently digitized archival data with modern radar sounding data to evaluate potential signatures of multidecadal changes in subglacial conditions between the data sets. Specifically I explore the internal layer evolution in the area around NEGIS between 1978/79 and the 2010s by adapting and applying the continuity index developed in earlier work. This multi-decadal insight into the behavior of the Greenland Ice Sheet in this critical region has the potential to advance our understanding of its system dynamics and improve models for the resulting global sea level rise.
Local Traditions and External Drivers of Change Influencing Patterns of Marine Resource Subsistence use across the California Coastline
Roya Meykadeh, MUIR
This project explores the interdisciplinary nature of ocean sustainability and ecosystem management. While significant effort has been made to engage commercial and tourism operators in collaborative coastal management, the values and priorities of stakeholders from smaller, less economically powerful communities have received substantially less attention. By investigating the intersection of ocean ecosystems, marine protected areas (MPAs), ocean access, and processes of social and ecological change for diverse and overlooked populations across the California Coast, this research supports the development of management strategies designed to enhance efficiency in policy regulations, equity in MPA management, and the resilience of diverse coastal communities. The main user group surveyed in this study is subsistence and recreational fishermen. These individuals spend substantial time engaging with this ecosystem, and are thus highly conscious of ecological fluctuations.
Specific focuses for our qualitative research included differences between recreational and commercial users in regard to policy implications and means by which users engage with the coastline, the cyclical nature of policy implementation in regard to species protection and MPA placement, and shifting patterns of ocean access and use associated with climate change. Our study develops a valuation framework to explore human relationships with nature and the environment as a primer before discussion of such topics. We explore differing and conflicting motives and values across user groups, finding that the relative economic strength of the commercial sector impacts policy to their advantage.
Our findings revealed a cyclicality to policy implementation in regard to species protection and abundance, and many interviewees observed that changes in regard to fishery management, regulations, and access are driven by a combination of climate change and mismanagement. We found that marine policy must focus on ecosystem management strategies and adopt more timely, dynamic, and humanistic approaches to ecological and recreational coastal management.
Computer Modeling of Plant Water Use and Carbon Uptake
Sabrina Ahmed, SESUR
Plants play a pivotal role in combating climate change by capturing carbon dioxide emissions. Their ability to sequester carbon, however, depends on environmental conditions like water availability and temperature, making it crucial to understand how vegetation will continue to offset human emissions in the face of global warming.
Stomatal pores, located on the leaf surfaces, have significant influence over terrestrial water and carbon fluxes, regulating gas exchange. Understanding and accurately modeling stomatal behavior is essential for predicting climate change’s impacts on ecosystems. Existing stomatal optimization models, however, are largely inaccurate, with errors up to 30%.
We hypothesize that one reason for such errors is a steady-state assumption, or that the rate of water and carbon moving in and out of the plant are at equilibrium. In reality, these factors generally vary with time.
We developed a new model that allows for variable water flow rates, eliminating the steady-state assumption. To test its accuracy, I compared the model’s predictions of leaf water potential, a measure of the water pressure in the leaf that helps drive a pressure gradient and allows water to move from the soil up to the leaf, with measured data from the Desert Botanical Garden in Phoenix, AZ.
To accomplish this, I determined which parameters were the most influential and determined that the maximum rate of carboxylation, xylem water potential at a 50% decline in xylem conductance, and leaf water potential at cavitation, exerted the greatest influence on leaf water potential. I created a particle swarm optimization algorithm to determine parameter values that yielded predictions that best aligned with measured data. Using the values determined by the algorithm has improved the accuracy of our model. This advancement provides crucial insights into the role of plants in carbon and water cycles, strengthening our comprehension of their role in mitigating climate change.
Optimization of Thermal Energy Use on Microgrids
Theodore Hickenlooper, CEE-VPUE
Islanded microgrids are essential for ensuring reliable and resilient energy supply in remote and isolated communities. The efficient utilization of available energy resources in such microgrids is crucial to minimize environmental impact and improve sustainability. This research poster presents a novel approach to enhance energy efficiency within islanded microgrids by addressing the power consumption of heat pumps, a commonly used technology for space heating and cooling. The primary focus of this study is the integration of additional thermal mass into the building structures of islanded microgrids. The concept revolves around the utilization of materials with high thermal mass properties, such as waterbricks or concrete, to store and release thermal energy. By strategically incorporating these materials, the team aims to reduce the power draw of heat pumps during operation, thereby optimizing energy consumption and overall microgrid performance. The research methodology involves modeling/simulation techniques developed both on site and via a partnership with the Naval Postgraduate School. These models are used to assess the impact of increased thermal mass on heat pump efficiency and energy consumption, as well as experiments designed to gather data on how this plays out in the real world. The simulation results are based on real-world scenarios and climate conditions, enabling a comprehensive evaluation of the proposed approach’s  effectiveness.
Unraveling the Early Cretaceous tectonic transition of the northernmost Magallanes Austral Basin (Southern Patagonia)
Thomas Saito, SESUR
The Magallanes Austral Basin (MAB; Southern Patagonia) provides researchers an opportunity to study how sediment is from onshore sources to offshore sinks. We used U-Pb Detrital Zircon Geochronology to understand the provenance and tectonic history of the northern terminus of the basin during the Early Cretaceous. These results provide information regarding the paleogeographic history of the basin during a critical tectonic transition from net-extensional to net-compressional regimes. In particular, we demonstrate that the onset of the fold-and-thrust belt is recorded diachronously in a southward younging direction consistent with observations from the deep-water counterparts to these strata. Furthermore, we note that the signature of fold-and-thrust belt development is cogenetic with a Jurassic age peak typically associated with sourcing from synrift distal sources to the north. We speculate that these ages are more likely indicative of recycling of Aptian-Albian strata along the NE margin of the basin that were uplifted and redeposited during fold-and-thrust belt deformation.
Developing Tools for Prioritising Species Reintroduction in Fragmented Ecosystems
Varun Shirhatti, SESUR
‘Defaunation’ is the process of large-scale animal biodiversity loss; it originates from local declines in populations that compound upon themselves to result in species extinctions. A major impact of defaunation is the loss of species interactions like pollination and seed dispersal, which reduce ecosystem resilience and cause the loss of ecosystem services; reintroducing seed-dispersing species is therefore a prime method of restoring resilience to defaunated ecosystems. One possible response to defaunation is trophic rewilding, the reintroduction of animal species with the goal of restoring trophic interactions to increase ecosystem resilience. However, prioritizing species and areas for reintroductions is critical when dealing with limited conservation resources. In this project, we focus on the Brazilian Atlantic Forest, a highly fragmented and defaunated biodiversity hotspot, to create a framework for identifying which species and sites are best-suited to each other for reintroduction efforts. This was achieved by collecting data on the space required to support viable populations of 14 seed-dispersing mammal species, then comparing that data to the sizes of environmental protected areas within the Serra do Mar region of the Atlantic Forest (extracted using GIS). One factor that proved to be extremely significant was the connectivity between adjacent protected areas – many of the protected areas in the area of interest were too small to support viable populations of the species being considered without accounting for connectivity. The final product was a suitability matrix that ranked the suitability of different sites as reintroduction locations for each species; two versions of the matrix (with and without connectivity) were created and compared. The inclusion of connectivity resulted in a much higher number of suitable reintroduction sites for nearly all the species of interest, suggesting that the presence of large, contiguous protected areas is extremely important for the success of species reintroductions in tropical environments.
Creating Data Stories for a Sustainable Future: From Personal to Global
Yash Narayan, SUPER
Informed, skilled communication and advocacy efficacious youth will be instrumental in realizing a sustainable and carbon-free future. In this research, we explore the recent advances in generative AI for creating an interactive platform that guides energy sustainability decision-making in the family that is both highly accessible to youth and provides actionable, easy-to-understand insights. This AI platform, EnergyAdvisor, can seamlessly digest household data provided by family members, such as hourly smart meter data, energy appliance and activity journaled data, and private human communications among family members, and use this information to generate a dialogue with users that can augment human capability to make energy-related decisions. The tool uses established concepts in power system engineering, energy literacy and education, and motivational theory and embeds this within a chat-based AI system. We believe that youth, supported by the advancements of large language models (LLM), can help bridge the gap between these critical energy insights and their appropriate application within their homes to promote sustainable and carbon-free futures. Ultimately our goal is to deploy it on a larger sample of youth, with the goal of embedding it within data literacy and sustainability program curricula globally. 
What Drives Intermodel Spread in The Southern Ocean’s Response To Standardized Wind Forcing?
Yuchen Li, SESUR
Uncertainty in the simulated response of the Southern Ocean (SO) to observed poleward strengthening of the westerlies (hereafter wind forcing) is likely responsible for persistent model biases in SO sea surface temperatures and sea ice. The wind-forced response has been proposed to follow a two-timescale mechanism, where northward Ekman advection causes fast surface cooling and anomalous upwelling causes slow warming. We attempt to explain significant intermodel spread in the response to a standardized step-like wind forcing in the latest generation of IPCC-class climate models. Uncertainty in the long-term response is related to the background stratification of the upper SO, which indeed varies greatly among models. However, differences in stratification account for less than 30% of the spread in the long-term response, suggesting important roles for parameterized eddy-induced overturning, deep convection, and changes in stratification. We diagnose these intermodel differences and quantify their contribution to the overall spread using a set of standardized wind stress perturbation experiments. Quantifying these sources of uncertainty will help better constrain the forced response of the SO to changing winds.
Investigating the Role of Sterols in Sterol-Producing Bacteria         
Yvette Soto-Hernandez, SESUR
In the world of microbiology, there is much room for discovery, considering the vast number of existing living microorganisms. For this project, I took on the task of investigating the function of sterols (type of lipid) in bacteria. Previously, we believed that only eukaryotes produced sterols and that bacteria did not, but instead hopanoids, to perform the similar function of stabilizing the cell membrane, lipid raft mobility, etc. But recently, sterols were discovered in some myxobacterial species – including Enhygromyxa salina and Minicystis rosea - and methanotrophs, like Methylococcus capsulatus. Considering that this discovery went against everything that was textbook bacteria, this left microbiologist with more questions about bacterial evolutionary history and their behavior, like the one that I’m aiming to answer; what is the function of sterols in bacteria? To begin answering this question, I took two different approaches. First, I grew up two myxobacteria, E. salina and M. rosea, in several different, potentially stressful conditions: at a variety of temperatures; acidic and basic conditions; with diverse food substances; and in anaerobic conditions. I then extracted lipids from these cultures and quantified their relative concentrations using gas chromatography mass spectrometry. By altering growth conditions, I observed large variations in the total sterol concentrations in both bacteria, suggesting sterol may play a role in adapting to these stressful conditions. Some conditions also caused a shift in the sterols produced; when fed live bacteria, both E. salina and M. rosea shifted biosynthesis toward primarily cholesterol production. This variation in the amounts and types of sterols produced provides a foundation to begin to decode the role sterols play in bacteria. The other approach I took involved deleting C-24 sterol reductase and novel sterol reductase MCA0111 from the M. capsulatus genome. These are genes that encode for enzymes with a role in the sterol-biosynthesis pathway; by deleting these genes we can confirm whether these sterol modifications are necessary for these bacteria to live. All in all, this research is only the beginning of discovering what function sterols have in bacteria. But answering this question is ultimately worth it in the biological and evolutionary context. Identifying functions of these sterols in sterol-producing bacteria could potentially lead us to discovering additional, overlooked functions that sterols carry out in eukaryotic species and provide insight into more ancestral and ancient functions for these important lipids.