i was asked to teach a course in tropical meteorology. i thought i understood how hurricanes worked. i knew what the conventional explanation was. so i started to teach it and i had one of those horrible experiences where you start talking about something and you realize it doesn't really make any sense. and that leads you down a path of trying to understand what does make sense. and that's what we call research. narrator: one of the most important tools emanuel uses in his work is computer modeling. dr. emanuel: most people believe that models are used principally to make forecasts, and of course they are. but a lot of modeling is done not for the purpose of forecasting a particular phenomenon, but for understanding it. narrator: using models emanuel can run simulated hurricanes hundreds of times a day and compare the results with data taken from real hurricanes. when differences are seen between the simulated hurricanes and the real ones, he can pinpoint the cause of those differences in the models and achieve not only better understand