NSF Award
2411065
Images are a powerful tool for observing and understanding the natural world. Ground-based imagery, such as from time-lapse and trail cameras, captures short- and long-term environmental processes that cannot be measured in other ways and has the potential to contribute visual information to multiple fields. Applications include understanding a changing water cycle, developing new technologies to monitor streamflow and river depth, and measuring changes to vegetation. These data can then be used in computer simulations that are enabled by artificial intelligence and machine learning (AI/ML) to model and predict different conditions and scenarios. Imagery is well-suited for environmental monitoring that integrates data from different sources, where variables extracted from imagery complement data derived from other sensors located near cameras. However, there are significant barriers to capturing quality imagery and extracting scientifically useful information from imagery. Computer software and training resources will be developed to lower those barriers, allowing people with different levels of technical expertise and backgrounds to advance science using image-based methods.<br/><br/>This project will develop a robust scientific community and accompanying cyberinfrastructure (CI) for using ground-based imagery to study environmental processes. This CI will complement existing remote sensing capabilities using satellite or airborne imagery, where tools such as ArcGIS and QGIS have opened new measurement capabilities and pathways to scientific discovery for a wide range of users. Ground-based imagery has potential for enabling new ecohydrological discoveries, and well-designed CI can empower people with the skills and tools needed for impactful and reproducible science. The specific goals of this Geoinformatics project are to develop: (1) Open-source software (GaugeCam Remote Image Manager Educational – Artificial Intelligence; GRIME-AI) that streamlines and documents reproducible workflows, (2) Benchmark data products (including data from PhenoCam archives) that promote method development and data standards, and (3) Training resources for broadened participation in the emerging scientific community that uses ground-based fixed cameras in ecohydrological research. The CI, available through GaugeCam.org and other public repositories, will be inviting and educational for a broad range of users, including those who may not currently have a strong STEM or data science identity. The project will focus on building community across disciplines through training for new users and increasing the ease of scientific discovery.<br/><br/>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.
