Student Seminar Series: Fei He, Tropical cyclones and their simulation to varying initial conditions in AGCMs


Add to Cal
  • Speaker: Fei He, PhD student, Department of Atmospheric, Oceanic, and Space Sciences, University of Michigan
  • Host Department: Statistics
  • Date: 11/06/2013
  • Time: 1:00 PM

  • Location: 4464 East Hall

  • Description:

    Abstract:  Tropical cyclones, also popularly known as hurricanes or typhoons, are one of the most lethal, destructive and expensive natural disasters. A single storm in Bangladesh in 1970 killed nearly half a million people. During the 2004 and 2005 hurricane seasons, United States suffers a financial loss of over $150 billion. Improving our understanding in tropical cyclone and its simulation in numerical models can offer us both societal and economic benefits. This presentation includes two sections. The first part is on the fundamental understandings of tropical cyclones, including their 3-dimensional structure, the sustaining mechanism, their genesis, death, transfiguration, societal and economic damages, and simulations in atmospheric global climate models (AGCMs). The second part is on the use of the newly developed statistical tool to help us understand the representation of tropical cyclones in AGCMs and assess their sensitivities to varying initial conditions. It is a new framework of uncertainty quantification (UQ) technique with Latin Hypercube Sampling (LHS) method to generate samplings and The Expanded Multivariate Adaptive Regression Splines (EMARS) to characterize response functions. Our work suggests it is a powerful tool that has the potential to be widely used for other weather systems such as front system or extra-tropical cyclones.