The first talk (Berger) will illustrate the impact of Atmospheric Motion Vectors (AMVs) on global numerical models, focusing on two primary topics. Topic #1 involves testing a new quality indicator: the expected error. This has shown promise in estimating data quality and may improve the quality control of assimilated AMVs. Topic #2 demonstrates the utility of assimilating GOES Rapid-Scan AMVs to improve forecasts of high-impact weather. These AMVs are more frequent and at a higher horizontal resolution than the operationally assimilated AMVs. The improvement of the NOGAPS track forecast of Hurricane Katrina due to the assimilation of GOES Rapid-Scan AMVs will be shown, in addition to further preliminary results using the GFS and a baseline AMV data denial experiment. The potential for using the expected error in these Rapid-Scan Experiments will also be discussed.

The benefit to operational forecasting from GOES Rapid-Scan AMVs may be optimized if the AMVs are "targeted" towards a certain region at a certain time. The second talk (Majumdar) will describe the utility and performance of the Ensemble Transform Kalman Filter (ETKF) targeted observing strategy. The method has been extended into the medium-range, and it has recently been found to be effective at predicting the influence of dropwindsonde observations on 3-5 day forecasts of winter weather. Concurrently, the method has been extended into the tropics. Guidance for 2006 Atlantic and NW Pacific tropical cyclones will be shown, and perspectives on the utility of the ETKF for identifying preferential regions for Rapid-Scan AMVs will be given.