Separating the signal from the noise: promoting alaskan students' inquiry with geographically relevant seismic data and machine learning techniques

The National Science Foundation's Division of Research on Learning in Formal and Informal Settings awarded $889,766 to Concord Consortium Inc. on October 1, 2025, under the STEM Education program (CFDA 47.076) to develop and deliver the SeismicML curriculum project. Through September 30, 2028, Concord Consortium will create a one-week seismology curriculum centered on an innovative block programming interface called DATAFLOW, designed to engage middle school students in Anchorage, Alaska, in authentic geoscience investigations using real seismic data. Key deliverables include the development of evidence-based teaching strategies, a comprehensive curriculum module integrating machine learning (ML) and computational science practices, scientific-grade seismometers for school installation, and block programming tools enabling students to analyze real-time seismic data patterns and conduct computational investigations of their community's natural and human-caused seismic events.

The project utilizes a design-based research approach conducted over two cycles, employing mixed-methods research to evaluate how authentic computer science experiences and ML techniques enhance student motivation and geoscience learning outcomes. Concord Consortium collaborates with two sub-awardees: Kent State University, where Dr. Bridget Mulvey leads research on classroom discourse, teacher interviews, and qualitative data analysis; and the University of Washington, which develops and integrates PyTorch-based deep learning models to detect and discriminate seismic event types, curates and annotates Pacific Northwest seismological datasets, and contributes to educational module development and dissemination through co-authored publications.

Abstract

This project will contribute to the earth science education community's understanding of how engaging students in authentic computer science experiences, including innovative methods such as machine learning, can deepen students' motivation and learning of geoscience concepts. The seismicml project will engage middle school students in anchorage, alaska, in authentic investigations of their community's natural and human-caused seismic events using practices of professional geoscientists. Through a partnership among teachers, geoscientists, educational researchers, technology and curriculum developers, and science administrators, the project will create a one-week seismology curriculum centered around an innovative block programming interface called dataflow. Within the curriculum, students will (1) explore the occurrence of earthquakes in the community by installing scientific grade seismometers in their school, (2) use machine learning to identify and classify seismic events, (3) create data visualizations of seismic events registered at their school, and (4) construct block programs that import real-time seismic data to find patterns in seismic events over different time periods and across different regions. The project will produce evidence-based teaching strategies that promote students' ability to conduct authentic computational science investigations. The goal of the seismicml project is to engage alaskan middle school students in contextualized inquiry investigations with local seismic data to help them understand applications of computer science and machine learning in modern science. Two cycles of design-based research will be conducted to develop the seismicml curriculum and dataflow program. A mixed methods research design will be applied to answer the following research questions: to what extent does using the computationally integrated seismic curriculum build students' computational practices and geoscience content knowledge? What are the novel affordances of integrating geographically relevant data, geoscientific concepts, and authentic computer science and machine learning practices for engaging middle school students in meaningful seismic investigations? Is student engagement with an authentic computationally integrated earth science curriculum associated with improved attitudes, perceived relevance, and science learning outcomes? What types of teacher, curricular, and computation-related supports are necessary to engage students in computationally integrated seismic investigations? Data sources include recordings of classroom discourse, pre- and post-surveys, embedded assessments, dataflow snapshots, and teacher interviews. Project research will generate knowledge about curriculum design and teaching strategies that promote students' engagement in computation-mediated science practices authentic to professional seismologists' work. By demonstrating the effectiveness of embedding computer science and machine learning into specific disciplinary middle school courses, this project will produce a replicable pedagogical model for including machine learning in other stem contexts, including algebra, physics, and career and technical courses. All project materials will be made available for free through open-source and open-content licensing to all future learners, teachers, and researchers beyond the participants outlined in the project. Research findings will be disseminated at conferences and in research and practitioner journals. The project is supported by the computer science for all (csforall) program, which aims to provide all U.S. Students with the opportunity to participate in computer science and computational thinking education in their schools. 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.- subawards are planned for this award.