by Jessica McCann

When Julia Muccino was a little girl, she dreamed of growing up and studying finite element modeling of environmental fluid dynamics and data assimilation. No, not really.

Like most kids, she had a variety of interests throughout junior high and high school. She was an only child who was serious about academics. Muccino loved writing and English, and also excelled in math and science. She played flute and saxophone in the marching band. She chose to study engineering in college and eventually grew into a highly specialized career.

“It wasn’t as though I was 7 years old and knew I wanted to do computational fluid mechanics,” says Muccino, an associate professor of civil and environmental engineering at Arizona State University. “I’m generally just a curious person. So it was more the people along the way who influenced me and introduced me to new things throughout my education.”

Today, Muccino works with the ADvanced CIRCulation model for oceanic, coastal, and estuarine waters. ADCIRC is a complicated numerical model scientists use to predict water circulation along coastlines and where sea tides meet the mouths of rivers.

The model includes information such as coastline shape, contours of the ocean bottom, and atmospheric wind and pressure, and approximations to the laws of physics. Researchers use the model to get information such as the height and speed of water at certain points throughout a region. It is used by scientists at Louisiana State University to simulate hurricanes. After the devastation caused by Hurricane Katrina, the model has gained much more attention.

Independently, scientists can use instrumentation to measure the height and speed of water at the same points. But such observational data won’t be in perfect agreement with model simulations. Errors are inherent in both direct observation and simulation. In a field of research known as data assimilation, models and observations can be combined in sophisticated ways.

“The result, ideally, is a melding of the best features of both,” the ASU professor explains.

Muccino earned a bachelor’s degree in civil engineering at Villanova University, and master’s and doctorate degrees at the University of Notre Dame. She was first introduced to ocean sciences during two summers at the Institute of Ocean Sciences in Victoria, British Columbia while working on her Ph.D. Muccino then completed a research fellow position at The Flinders University of South Australia, which deepened her interest in ocean sciences. She arrived at ASU in 1997.

Muccino represents ASU as part of a seven-university consortium. Funded by the National Science Foundation (NSF), the group is working to further develop a modular data assimilation system known as Inverse Ocean Modeling (IOM). The system software uses information technology to speed the development of this interface between numerical models and observational data. The key? Let the computers to do more of the computational work.

“Computers are very patient,” says Muccino. “They don’t mind doing millions of computations to produce a single number.”

Using IOM software, the model user can input the unique (or model-dependent) components of the algorithm, as well as the physical ocean data. An algorithm is a logical sequence of steps for solving a problem, often written out as a flow chart. Those steps can be translated into a computer program.

The software then takes over by generating the advanced, yet more universal, code needed to complete the system. This not only speeds development of the model, it minimizes the likelihood of user mistakes when building such a complex algorithm.

The IOM team made the software relatively easy to use. But Muccino says they fear it may just as easily be misused. The system is still sophisticated and complex, and it takes a reasonable amount of knowledge to use it correctly. As a result, outreach and education has been a significant part of the project.

“The main objective is to make the data assimilation algorithm more accessible to the scientific and engineering communities,” says Muccino. “But, at the same time, we don’t want people to inadvertently misuse the software because they don't know how and why it works.”

Muccino works not only on the interface between ADCIRC and IOM, she is key in the project’s educational efforts. With the help of a consultant, she designed and built the IOM website. The site serves as a repository for the software, a place for research exchange, and an educational tool for the techniques used in IOM. Her ultimate goal is to make the general process of implementing data assimilation with a model a more attainable goal.

“The goal for me has never been to become an oceanographer. I want to learn about these techniques and bring them from the science community to the engineering community. I want to invite and help people to do the same thing with their engineering models,” she explains.

Muccino credits the NSF Research Experience for Undergraduates (REU) program at the University of Notre Dame with feeding her inquisitiveness. The program helped her to see her education in a new light.

“For the first time, I saw that academics weren’t just about homework and tests,” she says. “It was fun to see what it could be like to actually explore an idea, not know how it was going to turn out, and then find that it was okay if it didn’t work out the way you thought it would.”

Muccino strives to incorporate such enlightenment in her own teaching. The approach has earned her several distinctions. She was named Outstanding Engineering Educator of the Year by the Arizona chapter of the National Society of Professional Engineers, and Teacher of the Year by ASU’s student chapter of the American Society of Civil Engineers, among other honors.

For the past year, Muccino has been on hiatus from teaching so she could focus her efforts on the IOM project. While she admits she has no idea where her own career path may eventually lead, she has a clear vision about where her research is headed.

Muccino sees a day when professionals from many different geosciences and engineering disciplines will be able to use the software to create their own data assimilation models. More importantly, they’ll be able to do the work quickly and with relative confidence that they’re doing it correctly.

Data assimilation research is supported by the National Science Foundation. For more information, contact Julia Muccino, Ph.D., Ira A. Fulton School of Engineering, 480.965.0598. Send e-mal to Julia.Muccino@asu.edu