Meet a RC researcher- Eric Wolf
After earning his undergraduate degree at the University of Maryland, Eric Wolf came to the University of Colorado, Boulder as a graduate student. He is now a second-year post-doctoral student in the Department of Atmospheric and Oceanic Sciences. During graduate school, Wolf began using three-dimensional general circulation models to simulate the Earth’s climate. He was able to observe differences in climate between epochs in the Earth’s history and predict its future. A model from the National Center on Atmospheric Research provided the data for Wolf to perform supercomputer simulations of the Earth’s climate. The same model is used to predict the effects of anthropogenic climate change.
After he completed his Ph.D., Wolf’s research transitioned to focus on the new topic of extra-solar planets. These planets, which orbit other stars besides our Sun, have recently gained prominence in the popular science news; they sparked Wolf’s interest because they were an exciting new area of study, and the topic flowed naturally from his graduate work. Because data on the climates of these distant, recently-discovered planets is sparse, Wolf uses supercomputer simulations to draw conclusions. He says, “My role as a modeler is to bridge the gap between ideas and theory in modeling and what we might be able to observe.” He begins with approximations based on a computer model of the Earth’s climate, and then changes the code to account for the mass, radius and orbit of the extra-solar planet. Wolf compares this process to making a series of base substitutions in DNA, saying, “There are a few hundred thousand lines of code in the climate model. I go in and change a couple percent, and that changes the ability and the scientific purpose of the model.”
The ideal goal of Wolf’s research is to find a habitable zone, a place in the orbit around a star where a planet could sustain life. “Too close to the star, the planet is too hot; farther away, it’s too cold,” Wolf says. “This research is driven by the imagination of man, to say, ‘What if we could find another Earth somewhere?’” Although a significant amount of time, work and scientific advancement stands between Wolf and that goal, enormous progress has been achieved in climate modeling over the past two decades. Early models were one-dimensional, showing a single column of data from the planet’s surface to its atmosphere. Wolf says, “It’s only been in the last two or three years that computing availability has let us attack these problems with the three-dimensional models, which gave us a much better and more complete result.”
Diagram of the habitable zone for a planet
Wolf has used Janus since its inception, first for graduate school and now for his post-doctoral research. Access to the supercomputer has shaped his work. Wolf explains, “With the level of computing that’s now available, I’m able to expand my paradigm. In 2009, when one simulation on a desktop took ten days, I could only put five or six simulations in a paper. Having the resources of Janus has allowed me to really expand my studies and do things I never thought I would be able to do, technology-wise or time-wise.” Wolf is now able to use high-resolution models that require 180-240 processors. Access to increased computational speed and power provides him with much more flexibility in model development.
Research Computing helps Wolf in his research by maintaining Janus and addressing any issues he encounters with the supercomputer. He says, “Research computing has always been extremely helpful, as far as issues that have come up. My role is fairly simple: I use the machine. There’s a lot going into maintenance and upkeep that I don’t see. They keep all the wheels greased so I’m free to submit my jobs and analyze my data without a care in the world, but I know behind the scenes, they’re doing a lot of work.”