URI_Research_Magazine_Momentum_Spring_2015_Melissa-McCarthy

“Hurricane forecasting is just mathematics, physics and computer science... the solution to those equations is the forecast.” - Isaac Ginis

thousands of computer processors and contains hundreds of thousands of lines of code. Its computational design is unique in that it tracks the eye of a storm as it travels across the ocean. Since the winds of a hurricane are strongest in the eye wall, having a higher resolution for the eye region, as HWRF model allows, is crucial. URI has a proud history of assisting the NHC in improving its forecasts. Through grants from NOAA and the Navy, the University has access to supercomputers for conducting research. Ginis has worked alongside government scientists to make changes to the operational codes of the model. Indeed, after Hurricane Sandy, he met regularly with NCEP scientists for months to address issues in the HWRF that had emerged during the storm.

resolution necessary for accurate hurricane predictions is just a few kilometers wide. But when more grid points are added to the system, it becomes much more difficult and expensive to calculate the equations. Ideally, if Professor Ginis did not have to consider any computing or cost constraints, he would put the grid points as closely together as possible. Indeed, he has a student researching a model with grid points only 30 meters apart. There are, however, strict constraints on how quickly forecasts need to be created for NHC’s operational model. “A five-day forecast needs to be done within one hour,” Ginis says. For the operational requirements of NHC, he, unfortunately, cannot have the resolution he would like. NHC’s operational model is not lacking in dynamism, however. The model is supported by

continues on next page

Spring | 2015 Page 5