Tuesday, 28 September 2010
ABC Pre-Function (Westin Annapolis)
GDAS weather model winds are produced operationally by NCEP, daily at 6-hour intervals. These wind products are used extensively in weather forecasting as well as atmospheric and oceanographic research. In addition, the GDAS wind products are often used for satellite wind validation. Past global analyses of GDAS winds to QuikScat, SSM/I, AMSR-E and TMI winds has found standard deviations at or below 1.5 m/s with small biases. In this work, we look more extensively at the wind difference patterns and use ASCAT, QuikSCAT, SSM/I and WindSat wind speeds and directions to determine which bias regions can be attributed to which products. We highlight the wind patterns of specific regions and discuss why these regions of greater difference exist. The satellite data used will include the new WindSat product created by Meissner and Wentz. This wind product is produced using a blending of the standard radiometer physical retrieval algorithm for non-raining atmospheres and the new wind-through-rain algorithm that uses C-band and X-band frequencies to find combinations of different channels sufficiently sensitive to wind speed but less sensitive to rain. The QuikScat data are from a new geophysical model function that has been derived using over 10 years of data by Ricciardulli and Meissner. In addition to exploring differences of each satellite wind with the model winds, we also look at C-band and Ku-band wind differences by using ASCAT and QuikSCAT data comparisons. We will present wind differences over time to highlight the changes in the NCEP model made throughout the years.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner