Tens of millions of people live in areas that are at risk for flooding due to climate change, sea level rise, and melting of glaciers. UMBC’s Maryam Rahnemoonfar and a team of researchers are using data science, machine learning, and artificial intelligence (AI) to analyze enormous volumes of climate data, and Arctic and Antarctic observations in ways that could help populations prepare for and respond to these risks.
Rahnemoonfar, associate professor of information systems, is the principal investigator on a new five-year, $13 million grant from the National Science Foundation’s Harnessing the Data Revolution (HDR) Big Idea program. Through the grant, she will launch and direct iHARP, the NSF HDR Institute for Harnessing Data and Model Revolution in the Polar Regions.
“It is so exciting to be selected as one of the five HDR institutes in the nation, however, this comes with huge responsibility,” she says. “We are the first data science and machine learning institute in the world that is dedicated to research in polar regions.”
Data science meets climate science
Climate scientists rely on data that are incredibly challenging to disentangle, Rahnemoonfar explains. AI offers solutions to analyzing these large datasets, providing sophisticated models that make the most use of the quantity and quality of data available.Maryam Rahnemoonfar, right, discussing the imaging iHARP research.
The researchers involved with this grant will reduce uncertainties in projecting sea level rise by combining physics modeling, machine learning techniques, and data analysis. The results of their work will inform policymaking to address national and global priorities related to the climate crisis. Further, Rahnemoonfar notes, the team will investigate novel data science techniques that can be applied to other disciplines encountering challenges related to complex data.
Tackling challenges together
This project builds on research from Rahnemoonfar’s prior collaborative project, Intelligent Solutions for Navigating Big Data from the Arctic and Antarctic, supported by a 2018 NSF grant and a 2019 Amazon Machine Learning Research Award. The team will continue this work on a larger scale.
“I am proud of our strong convergence team with many members who are leaders in their fields, including our domain scientists and data scientists, and I am thrilled to lead this effort,” says Rahnemoonfar. She is working with co-PIs Jianwu Wang, associate professor of information systems at UMBC; Mathieu Morligehm at Dartmouth College; Shashi Shekhar at the University of Minnesota; and Jan Lenaerts at the University of Colorado Boulder.
Project collaborators have expertise in computer and information science, electrical engineering, civil and environmental engineering, Earth science, atmospheric science, oceanic science, mathematics, statistics, physics, geology, glaciology, and data science. An interdisciplinary approach, supported by the nimble nature of information systems, is essential to the team’s success, says Vandana Janeja, chair and professor of IS at UMBC.
“Dr. Rahnemoonfar’s ground-breaking climate change research involves a highly multidisciplinary team within IS and across multiple institutions,” says Janeja. “This grant is an example of our faculty’s bold research agenda and strong teams. This type of work has both led to wide community impacts and also translated into curriculum innovations with strong student engagement.”
In addition to working with UMBC faculty, Rahnemoonfar will lead a group of researchers from institutions across the country, as well as government and industry researchers. These partners include the University of Colorado Boulder, Dartmouth College, University of Minnesota, University of Alaska Fairbanks, Bowie State University, Amherst College, University of Texas at Austin, NASA Universities Space Research Association, NASA Jet Propulsion Laboratory, NVIDIA, IBM, and Amazon.
The grant will also involve dozens of undergraduate and graduate students and postdoctoral researchers. Rahnemoonfar and her team will engage students from underrepresented groups by working with UMBC’s Center for Women in Technology and similar programs, as well as the Baltimore’s Women in Machine Learning and Data Science chapter, which she currently leads.Jianwu Wang (left), Maryam Rahnemoonfar, and Masoud Yari.
Additionally, the project will support educational and outreach activities, with an eye toward workforce development. This includes programming for K-12 and college students, and lectures and training opportunities for data science and domain science professionals. Students will work alongside partners including NASA, Amazon, and IBM, and will have internship opportunities with federal and industry partners.
The research teams also plan to develop museum exhibits to help public audiences access their work.
Impact of multidisciplinary teams
“This major multi-institutional NSF award reflects so much hard work and ingenuity. National programs like iHARP, with its focus on some of the great challenges our society is facing, add to UMBC’s growing reputation for innovation and excellence in public impact research,” says Karl V. Steiner, vice president for research at UMBC.
“The College is experiencing extraordinary research growth. This is made possible by both the development of multidisciplinary teams and our increased focus on leadership development,” says Keith J Bowman, dean of the College of Engineering and Information Technology. “This project team, and others recently funded or pending, benefit from the tremendous faculty talent we have recruited in the last several years.”
The solutions that are developed through this work will have applications beyond environmental issues. Rahnemoonfar anticipates the team’s research will impact the future of medicine, autonomous driving, and remote sensing, and that the students working on the project will become the next generation of experts addressing these global issues.
In addition to Rahnemoonfar and Wang, the UMBC team also includes Janeja; Aryya Gangopadhyay, professor of information systems; Masoud Yari, research assistant professor in information systems; Karen Chen, assistant professor of information systems; Osman Gani, assistant professor of information systems; and Don Engel, associate vice president for research development, and assistant professor of computer science and electrical engineering.
Banner image: Some of the iHARP research team. All photos by Marlayna Demond ’11 for UMBC.