CAREER: (TIERRA) Tracking Island Evolution and Resilience through Remote Analyses of Atolls

Atolls, low-lying coral islands surrounding inner lagoons, are critical ecosystems that support biodiversity and human communities, but face escalating threats from sea-level rise, ocean acidification, intensified storms, and human activities. This research project seeks to better understand how atolls respond to these challenges by studying changes in their size and shape over time.

By analyzing satellite imagery and modeling the potential response of these islands to changing climates, the project aims to identify atolls most at risk of land loss and provide insights for conservation and community adaptation strategies. This project will provide a detailed and reproducible approach to studying atoll island change and resilience.

An essential component of the project is its commitment to education and diversity, offering hands-on research, training, and mentoring opportunities for undergraduate students, particularly those from underrepresented backgrounds. These efforts will not only generate actionable knowledge for policymakers and conservationists but also train a new generation of scientists dedicated to tackling environmental challenges.

The outcomes will contribute to safeguarding these fragile islands and the communities and species that rely on them. By leveraging modern remote sensing technologies and numerical modeling, it will enhance our understanding of how these unique landforms respond to both natural and human driven impacts. This research will not only advance scientific knowledge but also foster a new generation of diverse and skilled Earth and Environmental scientists equipped to tackle future environmental challenges.

This project integrates remote sensing, machine learning, and numerical modeling to study atoll systems at both global and local scales, with the goal of understanding their resilience to climate change and human impacts. Utilizing Landsat's extensive satellite imagery archive and high-resolution Planet Labs imagery, the research will enhance remote sensing accuracy and expand the spatial coverage of a global database of atoll morphometrics while expanding the datasets to include wave and storm climate data per atoll.

By employing machine learning techniques, such as convolutional neural networks, the project will automate landcover classification and atoll mapping of satellite imagery. Fieldwork at Glover’s Reef Atoll in Belize will ground-truth these methods, ensuring robust modeling and accurate predictions. The study addresses key questions about the drivers of atoll evolution—natural processes like coral growth and sediment dynamics versus external influences like wave climate and human activity—and will inform predictive models of atoll responses to environmental challenges.

In parallel, the project advances environmental science education by embedding computational thinking, data science, and computational proficiency into a revamped Environmental Computation curriculum at Colby College. By designing courses that blend computer science and environmental studies, including data science, AI, and field-based methods, the project fosters diversity and accessibility in the earth sciences, empowering students from all backgrounds to tackle complex environmental problems.

This initiative aligns with the NSF’s mission by addressing pressing environmental challenges while cultivating a new generation of interdisciplinary scientists equipped with computational and analytical skills critical for advancing research and conservation. This project is jointly funded by the Geomorphology and Land-use Dynamics Program (GLD), the Earth Sciences Division (EAR), and the Established Program to Stimulate Competitive Research (EPSCoR).

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.