…To Seek Out New Data and New Algorithms – Mario Juric

Mario Juric, a self-described “data driven astrophysicist with interest in extracting knowledge from large survey datasets” has joined LSST as Project Scientist for Data Management. Growing up, Mario wanted to explore space and “boldly go where no one has gone before.” However, he soon realized that none of the world’s space agencies were interested in a 6'8" astronaut and decided to become an astrophysicist. “Alas, somewhere along the way I got my first computer and caught the love-to-think-about-code-and-algorithms bug,” Mario explained. “I’ve spent the rest of my career trying to reconcile my interests in computing and astronomy, until I got to my current position contributing to the astronomical data generating machine of the next decade – LSST”.

As Data Management Project Scientist, Mario’s primary responsibility is to ensure that LSST software, infrastructure, and data products are capable of delivering the science envisioned for LSST. One of the biggest challenges in his new role will be to design a data analysis system that can robustly and automatically operate on orders of magnitude more data than experienced by current surveys.

“I always try to take up projects that I’ll be excited about every single day, and that push the boundaries of what is currently possible. To become a part of this extremely talented team that is hard at work creating a modern, efficient, data processing stack for LSST was hard to say no to – so much so that I’ve cut short my stay as a Hubble Fellow at Harvard University to join LSST this January.”

A native of Croatia, Mario earned a B.Sc. in Physics at the University of Zagreb, followed by a Ph.D. in astrophysical sciences from Princeton University. Prior to his two years at Harvard, Mario spent three years as a postdoctoral fellow at the Institute for Advanced Study at Princeton. Most recently, he has worked on understanding Milky Way structure using SDSS and PanSTARRS PS1 surveys. The LSST, he says, will enlarge by two orders of magnitude the volume of the Galaxy able to be explored. For example, “by enabling researchers to trace the structure of the Galactic stellar halo out to the virial radius, including stellar streams and dwarf galaxies, the LSST promises to reveal much about how the Galaxy formed and evolved to its present state.”

Mario’s research lives at the intersection of computationally intensive and observational astrophysics. “I’m especially intrigued by the problem (and promise!) of the extraction of knowledge, of learning, from large datasets. In my career, I’ve worked on asteroid surveys, with the Sloan Digital Sky Survey, and most recently with PanSTARRS. LSST has the potential to leave an even greater mark on astronomy than these surveys had. This brings with it a number of extremely exciting challenges, the most basic being: how do we convert all this wonderful data to useful information? Also, how do we create a code-base on which both LSST and future surveys will be able to stand? And how do we transform the community from the usual theorist/observer mental frame of reference, to one of computer-assisted knowledge discoverers?”

To facilitate that transformation, Mario advises the current and upcoming generation of students and future scientists to “be broad in your interests. Learn information theory. Use Python. Know C++,” because “while LSST will be a giant leap forward in all the key science areas used to design it, it is the unexpected discoveries that will prove most fascinating. Only time will tell what those will be.”

Article written by Robert McKercher and Mario Juric