Research opportunities are increasingly offered to undergraduate students and recent graduates to help them understand marine and aquatic sciences and offer them an opportunity to consider career options. This session will provide a venue for undergraduate and post- baccalaureate research in a general session that highlights the wide variety of student research. Presenting at meetings is an opportunity for students to engage with the science community and for interested faculty to discuss potential projects. This session is designed for students who are presenting for the first time at a scientific meeting and for mentors looking for possible graduate students. Students are not limited to this session, and we encourage any student who wishes to submit an abstract to a specialized science session in the subject of their research to consider that option as well.
Lead Organizer: David Fields, Bigelow Labs for Ocean Sciences (dfields@bigelow.org)
Co-organizers:
Lisa ROM, NSF (elrom@nsf.gov)
Hayley Schiebel, Suffolk University (hschiebel@suffolk.edu)
Presentations
02:30 PM
EVALUATING SUSTAINABILITY METRICS FOR MAINE RIVER HERRING POPULATIONS IN THE PENOBSCOT RIVER (8713)
Primary Presenter: Aileen McDonald, American University (aileenmcd2@gmail.com)
There are two species of River Herring found throughout Maine watersheds, alewife (Alosa pseudoharengus) and blueback herring (A. aestivalis). River Herring hatch in inland lakes and rivers and migrate to the ocean until they are adults, returning inland annually to their natal waters in the spring to spawn. River Herring play a vital role in the overall function of several ecosystems, providing nutrients in river, lake, and ocean systems, a food source for larger fish, mammals, and birds, and a fishery for humans. For centuries, dams blocked access to rivers and streams in Maine, drastically reducing fish passage which led to population declines. Currently, efforts are underway to remove dams and improve passage. We investigated biological data of River Herring to understand how restoration efforts have changed populations dynamic. We collected 50 River Herring each week of the herring run from 2014 to 2024 at Milford Dam on the Penobscot River to collect biological data including, total length, weight, species, sex, age, and iteroparity. We prorated daily count data based on the biological samples taken. Counts of River Herring have increased from 400 thousand to over 5 million in the past 10 years. Species proportions have shown higher amounts of alewive than blueback herring in each year other than 2016 and 2022. In addition, we found that there were more older River Herring in recent years, with blueback herring up to age seven and alewife up to age eight. We observed a higher number of repeat spawners in recent years. We observed that the timing of the run appears earlier in 2021-2023 than in previous years. Our study is essential for understanding how River Herring species and their ecosystems are affected by improved habitat and fish passage.
02:45 PM
INCREASING DOMAIN ADAPTABILITY FOR MACHINE LEARNING IN UNDERWATER IMAGE PROCESSING (9071)
Primary Presenter: Kira Kaplan, Smith College (kakaplan@smith.edu)
The deep sea is one of the least explored environments on the planet. However, with technology advancing, scientists are finally able to explore these untouched areas with instrumentation such as Remote Autonomous Vehicles (ROVs) and Autonomous Underwater Vehicles (AUVs). Large volumes of data are now being collected in short amounts of time, leading to a backlog of data in need of processing. Employing the human eye to analyze such massive quantities of video data is no longer feasible; we need an efficient alternative. Computer vision, a subfield of machine learning, is one solution to processing visual data. However, marine underwater imagery presents many challenges to machine learning, including class imbalances in datasets, variation in illumination and image quality, and an overall lack of sufficient training data. One resource that is quickly gaining momentum is FathomNet, an open source, underwater image database meant for enabling machine learning in the ocean. However, even the pre-trained models available on FathomNet experience difficulty in detecting marine objects from different domains. This study investigates novel methods for increasing the domain adaptability of the megalodon FathomNet object detector through fine-tuning processes and image enhancement algorithms. The fine-tuned megalodon model showed increased performance compared to the original megalodon model and revealed specific challenges in detecting objects in blurry underwater images. This suggests that underwater image enhancement algorithms may be employed to reduce distortion in images and improve model performance within different domains.
03:00 PM
BEYOND BIOACCUMULATION: A DEEP DIVE INTO PFAS TRANSFORMATIONS AND BIOACCUMULATION IN MAINE'S BLUE MUSSELS (9676)
Primary Presenter: Ella Webber, Colby College (ekwebb25@colby.edu)
Per- and polyfluoroalkyl substances (PFAS) are persistent contaminants that can accumulate in filter-feeding organisms like blue mussels. Based on data collected during Bigelow's REU program, this study examines PFAS uptake, elimination, and biotransformation. This involved incubating blue mussels in high PFAS-concentration seawater for one week of uptake and one of depuration. Preliminary results demonstrate rapid PFAS accumulation with varying rates among compounds, suggesting differences in tissue saturation limits. Despite mussels accumulating PFOS to levels exceeding Maine CDC consumption guidelines, depuration reduced concentrations to below-recommended levels within a week. Biotransformation was minimal for most PFAS, but EtFOSA was converted to FOSA, a more prevalent compound in environmental samples. These findings challenge using FOSA as a reliable biomarker for PFAS contamination in Maine's mussels. Further investigation aims to understand these complex PFAS dynamics in mussels, which is crucial for assessing human health risks and informing regulatory decisions by employing statistical analysis and comparing environmental trends.
03:15 PM
UNTANGLING IMPACTS OF BACTERIAL INTERACTIONS AND CHANGING ENVIRONMENTAL CONDITIONS ON THE GROWTH AND TOXIN PRODUCTION OF PYRODINIUM BAHAMENSE VAR. BAHAMENSE (9309)
Primary Presenter: Lauren D'Amore, University of South Florida (ldamore715@gmail.com)
The dinoflagellate Pyrodinium bahamense forms harmful algal blooms (HABs) in Florida estuaries. This alga produces saxitoxins, which bioaccumulate and negatively impact ecosystems, public health, and local economies. Complex interacting drivers make it difficult to predict Pyrodinium bloom intensity and impacts. For example, increasing ocean temperatures may increase Pyrodinium growth rates and facilitate range expansions. Changing weather patterns, including extreme storms and droughts, cause large salinity swings alongside high summer temperatures, which may accelerate or inhibit bloom formation. Moreover, bacterial interactions influence toxin production in closely related HAB dinoflagellates. It is unknown how climate change will influence interactions in Pyrodinium phycospheres or the downstream effects on blooms. We designed a factorial experiment to test the effects of temperature, salinity, and bacterial interactions on the growth rate, biomass, and toxin production of Pyrodinium. Pyrodinium cultures co-cultured with low- and high-diversity bacterial communities were grown in low (20 PSU), standard (25 PSU), and high salinity (30 PSU) at standard (24ºC) and high temperature (32ºC). Growth is monitored with imaging flow cytometry to detect changes in cell size and morphology alongside growth rates and total biomass. Saxitoxins will be extracted during exponential growth phase and quantified with HPLC. This project will increase our understanding of how marine heatwaves and more frequent extreme weather will influence HAB occurrence and severity in Florida estuaries.
03:30 PM
Developing Methodologies for Remotely Operated Vehicle Based Photogrammetry to Monitor Coral Reefs in the National Marine Sanctuary of American Samoa (9405)
Primary Presenter: Ella Ashford, Willamette University (ella.m.ashford@gmail.com)
Mesophotic coral ecosystems are a significant component of the coral reef environments in the National Marine Sanctuary of American Samoa (NMSAS). Stretching from 30m to more than 100m deep, these ecosystems are poorly explored as divers must use highly specialized diving techniques to reach these depths. This study explored the viability of using a small Remotely Operated Vehicle (ROV) to create photogrammetry models with the goal of enabling NMSAS to more efficiently assess and monitor mesophotic coral reefs. In this study, shallow water photogrammetry models created from both ROV and diving methodologies were compared to evaluate the feasibility of ROV based photogrammetry. Additionally a NMSAS Standard Operating Procedure was developed inclusive of ROV operations, camera settings, and transect strategies. This exploratory field research indicates that ROV based photogrammetry could be a promising option for mesophotic coral reef monitoring in the sanctuary. Additionally, ROV technology can engage local communities in ocean stewardship through videos, 3D models, and hands-on education initiatives. In conclusion, a small ROV can be utilized as a lower-cost, more efficient, and safer method for both shallow and mesophotic coral reef monitoring in remote locations while also serving as an exciting science communication tool.
SS02A - Undergraduate Research in Marine and Aquatic Sciences
Description
Time: 2:30 PM
Date: 29/3/2025
Room: W201CD