Researchers evaluate tropical cyclone risks

Imagine you’re planning an outdoor picnic with your family. You check the weather and it says there’s a 30 percent chance of rain. Do you follow through with your plans? What if it is an 80 percent chance? Is there a particular threshold on which you base your decision?

Researchers at Florida Institute of Technology have been working alongside the 45th Weather Squadron to answer these types of questions as part of evaluating tropical cyclone threats to the space program at Cape Canaveral Air Force Station, Patrick Air Force Base, and Kennedy Space Center.

Professor Dr. Steven Lazarus and research associate Michael Splitt, joined by several graduate students, have spent the last seven years evaluating the National Hurricane Center’s Tropical Cyclone Wind Probability Forecast Product and developing a tool that helps decision-makers better understand and evaluate weather risks.

Understanding the meaning behind weather probabilities and being able to interpret and react to them could help emergency workers and other decision makers, like the 45th Weather Squadron, better prepare for dangerous weather and, ultimately, save lives.

“Our work is more on the end of interpreting the data,” Splitt explained. “Is 30 percent a high number or a low number when you take into account the storm’s location and the uncertainties in the forecast? We are trying to demystify these numbers and give them more actionable meaning.”

The team received a series of three grants from NASA to complete this research, totaling approximately $56,500 over 4 years. The grants were used primarily to support three graduate students, Jaclyn Shafer ‘08, Denis Botambekov ‘11 and Sarah Collins ‘13, all of whom completed their master’s theses as part of the larger research project.

The first two portions of the project focused on taking the already existing WPFP tool issued by the National Hurricane Center – the one used to create the cone-shaped hurricane paths you would commonly see on the news during hurricane season — and evaluating the various yearly improvements, including storm intensity, which is a new component.

“Like a lot of models, they’re not static, as they’re always under development,” Lazarus said. “They run in real time, basically, to issue forecasts for tropical cyclones out in the Atlantic Basin. But, you can imagine, after the season, you have your statistics and your probability forecasts and then you work on model improvements during the off-season and it is ready for the new season.”

While the first two segments of their research focused on evaluating the WPFP tool and examining historic storms, the third, and most recent, portion of this project focused on determining probabilities for any given storm, based on a statistical model they developed. The results of this research are being published in an upcoming issue of “Weather and Forecasting,” a journal distributed by the American Meteorological Society.

All of this research was key in determining thresholds to help the 45th Weather Squadron make weather-related decisions.

One of the notable discoveries to come from assessing this data is the importance of relative distance from the storm.

“It turns out one of the controlling parameters is your distance from the storm, but it depends on how large the storm is,” Splitt said. “For a small storm to have the same relative distance, it’s got to be a lot closer. It turns out the math of the probabilities depends on this parameter.”

By plotting these normalized distances versus the probabilities, Splitt and Lazarus were able to uncover some theoretical curves that they expect to help decision-makers better understand and interpret the data.

“Probabilistic information is interesting because everyone has a different opinion about the ‘magic number’ for when a decision should be made,” Splitt said. “Do all emergency managers deal with these numbers the same way? We haven’t answered all the questions, but we’ve given a theoretical background to help interpret these numbers and lead to better decision-making.”

As they expand on this research, Splitt and Lazarus hope to further clarify this “magic number” so that emergency managers and decision-makers, like those at the 45th Weather Squadron, can continue to improve their processes and ensure the safety of those in areas affected by tropical cyclones.

The gray-shaded circles represent different possible locations of the storm, based on what the error is estimated to be.
The gray-shaded circles represent different possible locations of the storm, based on what the error is estimated to be.


by Ashley Carnifax

Show More
Back to top button