HR: 09:00h
AN: U11A-02 INVITED [PDF]
TI: A Biased View of Data Assimilation
AU: * Anderson, J L
EM: jla@ucar.edu
AF: NCAR, P.O. Box 3000, Boulder, CO 80307-3000 United States
AB:
Modern data assimilation for the atmosphere and ocean
combines information from observations and a prediction model to
produce an estimate of the state of the physical system. Data
assimilation is used not only to generate 'analyses' of the
system state, but also to produce initial conditions for
prediction models. In addition, it can be used to improve both the
observing system as well as the prediction model.
All of the methods in common use can be derived as heuristic
simplifications of an elementary Bayesian filtering algorithm.
An introduction to general filtering will be followed by a
brief description of several prominent assimilation
algorithms. Special attention will be given to ensemble-based
filtering methods which have been developed relatively
recently. A number of interesting research problems, related both
to innovative applications of existing assimilation methods
and to improving these methods will be outlined.
DE: 3210 Modeling
DE: 3260 Inverse theory
DE: 3337 Numerical modeling and data assimilation
DE: 4263 Ocean prediction
SC: U
MN: 2003 Fall Meeting