Please join us Friday, November 4, at 1 PM via ZOOM to hear Dr. Julie Bessac speak on...
TItle: Statistical methods for modeling and prediction of space-time environmental data
Contact the colloquium coordinators (Kristen Roland or Nadun Dissanayake) for the ZOOM information.
Abstract: We will discuss the context and challenges of statistical modeling for multidimensional processes, and in particular, but not restricted to, environmental data. The methods presented aim at characterizing, predicting and simulating complex phenomena by reproducing target quantities such as probabilistic distributions, extremes, space-time dependence, and interaction among variables, as well as multiscale aspects of processes at stake. We focus on several applications of such statistical models: 1) data fusion with Gaussian processes for wind speed space-time prediction, 2) stochastic enhancement of subgrid-scale variability of unresolved wind-related quantities in weather and climate models and 3) the modeling of the bulk and both tails of temperature distribution used in power-grid long-term planning.
Julie Bessac received the B.Sc. degree in fundamental Mathematics and the M.S. degree in Probability and Statistics, respectively in 2008 and 2011 from the University of Rennes 1, France. She received the Ph.D. degree in 2014 in applied Mathematics from the University of Rennes 1, France. Between 2014 and 2017, she was a post-doctoral appointee in the Mathematics and Computer Science Division at Argonne National Laboratory, Argonne, IL. Since 2017, she has been a Computational Statistician at Argonne National Laboratory. Her research focuses on the statistical modeling, forecasting and uncertainty quantification for diverse applications, as for instance geophysical processes and their applications to energy systems.