Colin has been interested in hurricanes and cyclones since he was a kid. When it came time for graduate school, he knew he wanted to study weather and cyclones. While his department represents only one third of the Atmospheric, Oceanic and Space Sciences program, this small but mighty group has taken their field by storm.
His journey to the University of Michigan was a joint venture. As happens with many graduate students, Colin and his girlfriend were applying to graduate programs at the same time. This complicated their efforts, but Michigan had nationally-ranked programs for both of them. “At the end of the day, it was an easy decision.” he says.
Colin was able to dive headfirst into his research soon after he arrived at U-M. Some of his early coursework involved carry-over concepts from master’s classes he had taken, so he was able to focus more time and energy on beginning to build a body of research. He was able to work with established Ph.D. students on small branches of their projects before beginning his own dissertation.
His dissertation uses variable-resolution models to study tropical cyclones – improving both climate models and weather forecasts. His modelling shows that variable-resolution grids can dramatically lessen the amount of time needed to run a simulation. More broadly, using a global grid with different gridbox sizes, he can focus on regions where cyclones are common, including the Atlantic Ocean, the breeding ground for storms affecting the United States. By focusing on smaller areas, it takes less time to run the simulation. And running simulations is not a quick, casual affair: using supercomputers available only through a partnership with the National Center for Atmospheric Research, a 25-year climate simulation took 6 months to run.
Similar modeling setups hadn’t previously caught on in the atmospheric science field, but his lab has tested for accuracy and found that the framework used at U-M has performed much better than expected in forecasting cyclones. This positive performance shows significant benefit to the field without significant change to the current modelling infrastructure.
Colin finishes his graduate studies this year and is pursuing a postdoctoral or teaching position.