Lecture

Speaker: Corey Scheip, Ph.D., L.G. (WA), P.G. (NC), Data Science and Earth Observation Team Lead, Senior Geoscientist, BGC Engineering

Hosted by: Dr. Karl Wegmann, Professor and Associate Head of MEAS Dept., and CGA Faculty Fellow

NOTE this event is co-organized by NC State’s Department of Marine, Earth and Atmospheric Sciences.

Summary: It’s easier than ever to access geospatial data and spin up a machine learning model. Knowing what the results actually mean, and whether anyone can use them, is an entirely different problem. This talk examines the state of geospatial ML in the applied geosciences and focuses on landslide susceptibility and inventory mapping for two examples. Susceptibility modeling is unusually easy to do badly: DEMs, soil surveys, and landslide databases are easy to download, and a respectable-looking AUC is a few lines of AI-generated code away. But are the results useful? Models for mapping landslides with lidar, on the other hand, are a different beast. They are more complex to create, but done with care, they can 100x the speed of human mappers and augment the way we’ve been mapping landslides for decades. In 2026, with modern AI-driven tooling, the promise for both applications is real. However, most of the promise doesn’t come from the models themselves – it comes from the underlying data and decisions that the human expert made along the way.

About the speaker: Corey Scheip, Ph.D., L.G. (NC, WA) leads BGC Engineering’s Data Science and Earth Observation team, applying geospatial data science, high-performance computing, and machine learning to applied geoscience problems. His work largely focuses on remote sensing data like lidar, optical satellite imagery, and radar, to characterize how natural hazards affect the built and natural environments.

Studio

Speaker: Dr. Caren Cooper, Professor, Dept. of Forestry and Environmental Resources, and CGA Faculty Fellow, North Carolina State University

Hosted by: Dr. Georgina Sanchez, CGA’s Director of Research Engagement

Summary: Do you acquire more data than you’re equipped to analyze? Could you build better AI models if you had more human annotations? Can you imagine new research questions to answer if you had hundreds or thousands of people helping annotate data? A major challenge with using sensors and remote technologies to acquire data is dealing with the flood of information. With aerial imagery and camera trap photos, researchers often face a bottleneck in processing data rather than in acquiring it. Computers and AI can help, but a core need that remains is the human ability of pattern recognition. The Zooniverse platform enables researchers to accomplish what would otherwise not be possible or practical. In this studio, we’ll distinguish between research questions that can be advanced through conventional methods and those that can be advanced with Zooniverse. We’ll walk through how to build a project on the Zooniverse platform, including how to upload data, create volunteer workflows, set rules for consensus levels, and feed results from one workflow into more refined workflows. We’ll also discuss NC State’s Citizen Science Campus and opportunities to pilot test new projects through a Project Incubator.

NOTE: Participants are encouraged to bring a laptop to walk through a project on the Zooniverse platform.

About the speaker: Caren Cooper, professor of public science, leads the Citizen Science Campus program, including the Project Incubator and a graduate certificate in Participatory Sciences. Her research has involved participatory projects about soundscapes, water, and birds. Her work advances the field of participatory sciences and includes the development of the Data Ethics Toolkit for Participatory Sciences and the IDEAL Participatory Sciences program.

Lecture

Speaker: Dr. Ted Wong, Data Scientist and Research Associate, Urban Analytics, Ross Center for Sustainable Cities, World Resources Institute

Hosted by: John Vogler, Research Scholar, CGA

Summary: World Resources Institute is a large environmental NGO which (among other things) provides geospatial data and analyses to local governments to support sustainable urban planning. This talk will examine how WRI builds and deploys data products supporting city-level planning and decision-making for extreme heat, urban biodiversity, and decarbonization. We will discuss theories of change, user research, project-level business models, strategies for mainstreaming, representation of uncertainty, and of course all the technical considerations behind our data products.

About the speaker: Ted Wong is a data scientist on the Urban Analytics team at WRI Ross Center for Sustainable Cities. He develops data products, analytical tools, and training resources to support governments and communities in equitable, data-backed planning processes for decarbonization and climate resilience. Ted is an ecologist by training, and before coming to WRI he was a faculty member at Bryn Mawr College and the Community College of Philadelphia. He teaches an occasional GIS class for social policy students at the University of Pennsylvania. Ted has a B.S. and Ph.D. in Biological Sciences from Stanford University, and a Master of Forest Science degree from the Yale School of the Environment.

Panel | FRIDAY, OCT 23, 3:00 pm – 4:30 pm | 1216 Jordan Hall Addition

Co-Organizers: Center for Geospatial Analytics and the Climate and Sustainability Academy

Moderator: Mary Smith, Investigative Reporter, CBS 17

Panelists:

> Katherine Anarde, Assistant Professor of Coastal Engineering in NC State’s Department of Civil, Construction, and Environmental Engineering

> Gian Tavares, Director, State Resilience Office, NC Department of Environmental Quality

> Steven Koller, Insurance Research and Policy Engagement Manager at Insurance for Good and Research Fellow at Harvard’s Joint Center for Housing Studies

Summary: Current approaches for assessing flood risk and protecting people and infrastructure in the United States are rooted in regulatory frameworks established in the late 1960s. Floodplain mapping, insurance requirements and building standards continue to reflect assumptions and benchmarks developed more than half a century ago, despite rapidly changing climate, land use and exposure. This panel brings together experts from different domains to critically examine whether these legacy systems remain adequate for today’s and tomorrow’s challenges. We will explore strategies that challenge the status quo and consider what is needed to more effectively manage flood risk in a changing environment. Taking a forward-looking, interdisciplinary perspective — spanning insurance, economics, hazard modeling, communication, governance and policy — the discussion will explore where and when damage is expected, who will be most affected and how risks can be managed more equitably, proactively and effectively.

TBA

Speaker: TBA

Hosted by: the Geospatial Graduate Student Organization

Summary: Forthcoming.

About the speaker: Forthcoming.

Symposium | 12:30 pm – 1:30 pm | Hunt Library Teaching and Visualization Lab (room 4505)

The MGIST Digital Symposium is an opportunity for our graduating professional master’s students to present their Capstone projects to the geospatial community at NC State and beyond. This includes student peers, alumni, faculty and staff from across the university, and our community and industry partners.

The technology in Hunt Library will allow students to present digital posters of their Capstone projects on a large, immersive projection screen in a social atmosphere with networking and open discussion. The event is capped off with an award to the winner of the poster competition!

Schedule TBA