The Local Ensemble Transform Kalman Filter (LETKF) has proven to be a very

accurate, model-independent data assimilation algorithm that can be

implemented efficiently on highly parallel computer architectures.

This talk will survey some of the latest results with the LETKF,

including its application to an estuarine ocean model and to

bias correction of atmospheric surface pressure observations in the Global

Forecast System. The talk will also discuss some particulars of

the LETKF's computational efficency, including ways in which it

can be readily adapted to highly irregular model grids.