By James Crowley
These are exciting times, as we witness the growth of applied mathematics and its increasing relevance to so many sectors of the economy and to our daily lives. And yet a note of regret crept in when we learned in recent weeks that we will soon lose a staunch ally in supporting applied and computational math: the IMA at the University of Minnesota.
I am struck by a recent flood of stories on how mathematics and computing are making new inroads into increasing productivity in the economy. One of the latest to cross my screen notes that “the single greatest instrument of change in today’s business world, the one that is creating major uncertainties for an ever-growing universe of companies, is the advancement of mathematical algorithms and their related sophisticated software.”† The article states that “To some degree, every company will have to become a math house” in order to exploit the efficiencies of algorithms for data analysis.
While it is true that the growth of data science is phenomenal, and that statistics is now reported to be the fastest growing STEM major, we should not forget the important role played by other parts of our discipline. I am reminded of one class of models we (at least in this part of the world) hear about frequently at this time of year: weather prediction models.
Here in Philadelphia, a large overnight snow storm was predicted in early February; the city responded with advanced cancellation of public transportation and many flights, only to wake up to a light dusting rather than the predicted foot-plus of snow. The essentials of the storm were predicted remarkably well—the front would race across Pennsylvania, form a low-pressure cell off the coast, pull in moisture from the south, and move north toward New England, all of which happened. But the details—in particular, the formation of the low-pressure cell about 100 miles to the east of the predicted location—were off by enough to change the weather completely.
Forecasts will improve. They rely on models—the PDEs that describe the physics of the atmosphere––as well as on data. The models are solved by numerical methods run on high-performance computers. Improvements will come, yes, from better data, but they alone won’t result in better predictions. Improvements in the accuracy of numerical codes are also required, and they in turn rely on better models, greater resolution of the codes, and faster computing. This is but one example. Even as we embrace the surge of interest in fields like data science, we know that modeling, analysis, and computing will continue to play important roles.
It is exciting to live at the nexus of advances in data science, modeling, algorithms, and high-performance computing. At the end of January, I had the opportunity to share some of that excitement with the House Science Committee’s Subcommittee on Energy. I had been invited to testify on the value of research supported by the Department of Energy’s Office of Science, in particular by its Advanced Scientific Computing Research (ASCR) program.
Testifying for a congressional committee is an interesting experience. As one of four witnesses, I was humbled to share the table with Norman Augustine, former CEO of Lockheed Martin; Roscoe Giles of Boston University; and Dave Turek, vice president for technical computing at IBM. Each of us had five minutes to speak, facing colored lights that warned us as our time was ending (“Red light, green light . . . all around the town”). A gallery of folks sitting behind us took notes and tweeted as we spoke. Following our testimony, each of the seven members of Congress in attendance had five minutes either to speak or to use for questions and answers.
SIAM submitted written testimony, which was vetted through our Committee on Science Policy. In truth, the written testimony is far more detailed than anything that could be communicated in a short oral presentation. The oral testimony and the accompanying written document are part of a process. We participate in the hope that the information we provide will support ASCR’s important research programs, especially those in applied mathematics and computer science.
And this takes me back to the Institute for Mathematics and its Applications. We received the IMA’s Year in Review report for 2014 this week. In the introduction, current IMA director Fadil Santosa noted that NSF funding for the IMA in its present form, as one of eight mathematics institutes, will end in two years. This means that the IMA, long a valued institute and resource for many in the SIAM community, will change dramatically or possibly even cease to exist. While we recognize that NSF’s Division of Mathematical Sciences needs to review and revise its portfolio, we will greatly miss an institute that has served the applied and industrial math community since 1982. As Santosa wrote in his introduction,
“The IMA has been a major force in applied mathematics. . . . The IMA has been an enabler of interdisciplinary collaborations involving mathematicians and has forever changed the culture of mathematics research. It has also led the way in this country in developing a field now recognized as industrial mathematics.”
Tim Kelley, the editor-in-chief of SIAM Review, isn’t a person who panics easily, but he came close to that state when told that Bob would be stepping down. The reason, of course, is that Bob did a marvelous job as editor of the book reviews, and Tim realized that it would be difficult to find anyone who came even close to combining Bob’s passion for books and his understanding of the SIAM community.
As his colleague Mark Kot noted,
“When I ask Bob if he has received any exciting new books, he usually pulls a new book from one of his stacks, mentions an interesting connection, and tells an interesting story about the author. And then, more often than not, he says something really nice about the book. He continues to find new gold in countless old topics.”
On behalf of the entire SIAM community: Thanks, Bob!
*An alternative interpretation of the acronym SIAM that dates back about twenty years but seems, periodic setbacks notwithstanding, at least as appropriate as ever.