IMAGe logo
   Contact Us | Visit Us
advanced  

Data Assimilation Research Section Turbulence Numerics Team Geophysical Statistics Project
SAMSI logo

The Institute for Mathematics Applied to Geosciences
Theme for 2007: Statistics for Numerical Models

 IMAGe logo

Workshop II. Application of Random Matrices Theory and Methods

7-9 May 2007; Boulder, CO


Agenda

VENUE: FOOTHILLS LAB BLDG 2, ROOM 1001

Monday, 7 May 2007

Statistical issues in interpreting geophysical data and numerical output


8:00 - 8:30   Registration and Coffee
8:30 - 9:00   Welcome and introductory remarks
Doug Nychka, NCAR/IMAGe
9:00 - 10:00   Spatial patterns of probabilistic temperature change projections
Presentation
Reinhard Furrer, Colorado School of Mines
10:00 - 10:15   Break
10:15 - 11:15   The use of covariance matrices in dimension reduction for space-time data
Presentation
Ian Jolliffe, University of Exeter
11:15 - 12:15   Statistics for detecting climate change
Presentation
Serge Guillas, Georgia Institute of Technology
12:15 - 1:30   Lunch
1:30 - 2:30   Spatial analysis to quantify numerical model bias and dependence: How many climate models are there?
Presentation
Mikyoung Jun, Texas A&M University
2:30 - 3:30   High-dimensional eigen-analysis and spiked models
Presentation
Debashis Paul, University of California, Davis
3:30 - 3:45   Break
3:45 - 4:45   Nonstationary covariance models
Montserrat Fuentes, North Carolina State University
5:30 - 7:00   Mixer at the Millennium Harvest House

Tuesday, 8 May 2007

Morning Session: Data assimilation and dynamical systems

8:30-9:00   Coffee
9:00 - 9:45   Models for very large covariance matrices in atmospheric and oceanic sciences
Presentation
Chris Snyder, NCAR/MMM/IMAGe
9:45 - 10:30   Methods for dealing with spurious covariances arising from small samples in ensemble data assimilation
Presentation
Jeff Whitaker, ERSL, Physical Sciences Division - NOAA
10:30 - 10:45   Break
10:45 - 11:30   A space-time Kalman filter for combining satellite radiance data with a sediment transport model
Jonathan Stroud, University of Pennsylvania
11:30 - 1:30   Lunch

Afternoon Session: Statistical results for dealing with large covariance structures

1:30 - 2:20   Methods of estimating covariance matrices, their inverses and eigenstructures which take advantage of sparsity
Presentation
Peter Bickel, University of California, Berkeley
2:20 - 3:10   Random matrix techniques for estimation of a D dimensional covariance matrix from S samples when D > S
Steve Simon, Alcatel-Lucent, Bell Labs
3:10 - 3:30   Break
3:30 - 4:15   Sparse estimation of large covariance matrices via a hierarchical Lasso penalty
Ji Zhu, University of Michigan

Wednesday, 9 May 2007

Statistics for modeling spatial and space/time covariances

8:30 - 9:00   Coffee
9:00 - 9:45   Paleoclimate reconstructions
Presentation
Caspar Ammann, NCAR/Climate and Global Dynamics Division
9:45 - 10:30   Progress and problems in estimating climate variables and covariances from incomplete data sets
Presentation
Tapio Schneider, California Institute of Technology
10:30 - 10:45   Break
10:45 - 11:30   Why EOFs?
Presentation
Joe Tribbia, NCAR/Climate and Global Dynamics Division
11:30 - 1:00   Lunch
1:00 - 2:00   Use of reduced-rank covariance estimates for objective analyses of historical data sets
Presentation
Alexey Kaplan, Lamont-Doherty Earth Observatory of Columbia University
2:00 - 3:00   Covariance tapering for likelihood based estimation in large spatial data sets
Cari Kaufman, SAMSI/NCAR
3:00 - 3:20   Break
3:20 - 4:20   Multiresolution models for nonstationary covariances
Presentation
Doug Nychka, NCAR/IMAGe
4:20 - 4:40   Closing remarks
Doug Nychka, NCAR/IMAGe

WORKSHOP ADJOURNS


UCAR : NCAR : CISL : IMAGe : [  CMG   DAReS   GSP   RISC   TNT  ]
For problems with the page, contact Tim Hoar - thoar 'at' ucar 'dot' edu

© 2010, UCAR | Privacy Policy | Terms of Use | Contact UCAR | Visit UCAR | Sponsored by NSF Logo