Earth’s climate system is experiencing unprecedented change as human-made greenhouse gas emissions continue to perturb the global energy balance. Understanding and forecasting the nature of this change, and its impact on human welfare, is both a profound scientific problem and an urgent societal need. Embedded in that scientific task is a technological challenge. New observational technologies are bringing in a flood of new information. Applying data science to this immense stream of information allows science to more deeply explore aspects of climate. However, this astonishing volume of data creates a different challenge: the need for tools that can scale to the size of our ever-expanding datasets.
Unraveling and interpreting that data is of particular fascination to Ryan Abernathey. The physical oceanographer is an associate professor at Columbia University’s Department of Earth and Environmental Sciences who also leads the Ocean Transport Group at Lamont-Doherty Earth Observatory. His research focuses on the role of ocean circulation in the climate system, particularly mesoscale ocean dynamics — the processes that occur at horizontal scales of less than 100 kilometers. A computer modeler as well as a physical oceanographer, Abernathey uses satellite data, computer models, and supercomputing clusters to study the impacts of mesoscale turbulence on the larger circulation of heat, water, and nutrients in the global oceans.
This week, The Oceanography Society named Abernathey among three recipients of its very first Early Career Award. The award recognizes individuals who have demonstrated extraordinary scientific excellence and possess the potential to shape the future of oceanography. The Early Career Award also recognizes individuals who have made significant contributions toward educating and mentoring in the ocean sciences community and/or who have a record of outstanding outreach and/or science communication beyond the scientific community. Abernathey is creating a unique impact In these areas. Below, he discusses the award, his work, the role of big data, and what it all means to future research.
Video: A visualization created by Ryan Abernathey shows circulation within the Antarctic Circumpolar Current.
Q: Congratulations, Ryan. You say your work has two parallel threads; how would you describe your objectives?
A: The central mission for our research group is to understand ocean transport, or how stuff moves around in the ocean. By ‘stuff’ we mean, first and foremost, just the water itself, the ocean currents and the ways those currents transport things we care about. For example, the way they help heat enter the ocean as part of global warming. This matters a lot for the climate and ocean ecosystems. The way we do that is by using two main tools: satellite observations and high-resolution simulations or models. What both of these tools allow us to do is see small-scale ocean processes with more clarity so we can understand them better. And that leads to the data and computing side of our work. We need to see these small-scale processes better. That means we need high quality images with more detail, but that amounts to a much bigger files. These satellite observations/high-resolution images create a whole lot of data to deal with.
Q: Why is it important to get a better understanding of the role of small-scale ocean processes in the ocean?
A: A specific example is phytoplankton. These tiny organisms are the lungs of the ocean; they consume CO2, photosynthesize, and breath out oxygen. But they also need nutrients in order to grow. There is growing evidence that the supply of nutrients from small-scale ocean features, like eddies and fronts, is a really important source of nutrients for these organisms. But the global climate models we use to project future climate change are too coarse to properly represent these features, which means those projections may be missing something. By studying these processes in detail, we can get a sense of what might be missing.
Q: How have you dealt with the “problem” of having so much data to process that it can overwhelm available computational systems?
A: I’ve discovered I just love building tools for working with data and putting them into the hands of as many people as possible, and seeing those people use those tools to do their own research. That’s really satisfying to me, personally. This is not necessarily the most common activity of a scientist. Typically, researchers are expected to produce more and more papers detailing their scientific findings, so this focus on building tools has really been a pivot in my career. It’s been incredibly satisfying. It’s really kind of a community effort.
Q: Community is a big focus for you. For instance, the work you did to bring about and now lead “Pangeo: An Open Source Big Data Climate Science Platform.” Why is creating open source code so important to you?
A: I just feel that it’s a place I can contribute and I like doing it and it’s going to have a real, broad impact. I think a lot of people recognize the challenge of working with these really large data sets but the unique thing our project brings to the table is a vision for what to do about it, an idea of what the future infrastructure for data and computing will look like for oceanography. Participating in data-intensive research requires a lot of expensive infrastructure, and that is exclusionary. So, there’s also a sort of democratizing aspect to what we’re trying to do to make it possible for anyone, at any institution anywhere in the world, to do this data and computationally intensive research.
Q: Clearly, the award takes into account your specific approach to science. Was that important to you?
A: I’m glad the award did cite my work on open software and tools because it’s something that’s traditionally undervalued by the academic reward system. The fact that it can be recognized is a sign of progress. It’s not just about publishing papers. I’m pleased that this output of mine is recognized. That is indicative of a cultural evolution in the incentive structure in academia.
Q: What is most exciting to you about your work?
A: I love the data. I genuinely love looking at ocean data sets. Particularly, really large complex and beautiful ones that reveal these turbulent ocean processes. On a very aesthetic level, I just love to look at and work with ocean data. It’s sort of a unifying thread throughout all this work. The day-to-day motivation is about truth and beauty and these more abstract scientific ideals.
