Progress in the Development of a Localized Particle Filter for Weather Prediction

This event has passed. See the seminar recording here:

Dr. Jonathan Poterjoy

UMD AOSC

Monday January 28, 2019, 12:00-1:00 PM

ESSIC Conference Room 4102, 5825 University Research Ct, College Park, MD 20740

Abstract:

The success of ensemble Kalman filters in oceanography, meteorology, and other fields of geoscience is remarkable, considering the relatively small ensemble sizes permitted by high-performance computing resources available at research and operational centers. Dimension-reduction procedures and methods for treating sampling errors in Gaussian approximated probability densities, via covariance localization, variance inflation, etc., make this achievement possible. As moderate to large ensemble sizes become regularly available for geophysical data assimilation, a natural progression from current Kalman filter-based strategies is to attempt more general (Bayesian) probability density estimation. For the case of numerical weather prediction, advancements of this sort may be necessary to extract information from underused observing systems such as cloud- and precipitation-affected satellite measurements, or better constrain dynamical processes responsible for severe convective storms and tropical cyclones. Each of these applications challenges the underlying assumptions of data assimilation systems currently used operationally. This presentation will summarize recent efforts by one research group to develop Monte Carlo filters that bridge between Gaussian and Bayesian data assimilation methods for numerical weather prediction. Strategies adopted for this purpose rely on sequential importance resampling techniques used for particle filters, but their practical application for high-dimensional systems follow from ideas emerging from decades of ensemble Kalman filtering research. Using a recently developed particle filter method that operates effectively for high-dimensional applications, this presentation will also introduce applications that motivate future development of non-Gaussian filters.   

Bio-sketch:

Prof. Jonathan Poterjoy is in the Department of Atmospheric and Oceanic Science at the University of Maryland. His research focuses on the development and application of advanced data assimilation techniques for studying geophysical problems. Much of this work targets hazardous weather events that present major challenges for environmental prediction, such as tropical cyclones and severe convective storms. Prior to joining the University of Maryland, Jonathan held postdoc fellowships at the NOAA Atlantic Oceanic and Meteorological Laboratory in Miami, FL and the National Center for Atmospheric Research in Boulder, CO. He also worked briefly as a postdoc at the NOAA National Severe Storms Laboratory in Norman, OK. Jonathan holds a PhD in Meteorology from the Pennsylvania State University, and a BS in Meteorology and Applied Mathematics from Millersville University.

Webex info: 

Event number: 732 837 689
Event password: essic
——————————————————-
To join the online event
——————————————————-
1. Click here to join the online event.
Or copy and paste the following link to a browser: 
https://umd.webex.com/umd/onstage/g.php?MTID=e9400859f098ee42ae12fc902663d99b1
2. Click “Join Now”.
——————————————————-
To join the audio conference only
——————————————————-
US Toll: +1-415-655-0002
Global call-in numbers: https://umd.webex.com/umd/globalcallin.php?serviceType=EC&ED=710538302&tollFree=0s
Access code: 732 837 689
——————————————————-
For IT assistance
——————————————————-
Contact Travis Swaim at: tswaim1@umd.edu

Follow ESSIC:

ESSIC homepage: http://essic.umd.edu/

ESSIC seminar calendar: MSQ-4102; http://go.umd.edu/essicseminar

ESSIC twitter: http://twitter.com/ESSICUMD

ESSIC facebook: http://facebook.com/ESSICUMD

ESSIC seminar coordinator: Dr. John Yang, jxyang@umd.edu