Extended Abstract (924K)P1.5
NOAA Polar Orbiting Satellite Observing System Experiments using the NCEP GDAS
James A. Jung, CIMSS, Madison, WI; and T. H. Zapotocny, J. F. Le Marshall, and R. Treadon
Through collaboration with National Center for Environmental Prediction (NCEP) personnel, as well as substantial National Oceanographic and Atmospheric Administration (NOAA) hardware support, the Joint Center for Satellite Data Assimilation (JCSDA) has obtained the capability to run Observing System impact studies of all data types used in the Global Data Assimilation system (GDAS). Such studies allow a better understanding of how to utilize current and future in-situ and remotely sensed data types in present-day three-dimensional data assimilation systems. These studies have the ability to examine the forecast impact of existing data types, and simulate the forecast impact if one or more satellites (data sources) fail for an extended period of time. A final very important advantage of these studies is that they are completed at the operational resolution (T254L64 for this study) with the operational version of the model to produce as realistic scenario as possible.
Results in this poster will investigate the scenario of having one (NOAA-15) and two (NOAA-15 and NOAA-16) of the NOAA polar orbiting satellites fail. A primary goal is to determine how detrimental such a failure would be to NCEP's operational weather forecast quality. The removal experiment includes eliminating NOAA-15 AMSU data and NOAA-16 AMSU and HIRS data from the operational stream during 45 day time periods of each extreme season. NOAA-17 remains on in both experiments.
Results demonstrate that the day 0-7 anomaly correlations are substantially better in the extratropics of each hemisphere with the inclusion of all three NOAA satellites. In fact, a nearly linear decrease in anomaly correlation is noticed for each satellite removed. A second very apparent benefit of including data from three NOAA satellites is a more accurate tropical cyclone prediction. For the time periods investigated, the forecasted tropical cyclone position was more accurately reproduced 75% of the time with data from three NOAA satellites. These positive impacts from multiple NOAA satellites are shown even when including the full complement of other satellite and non-satellite based observations.
Poster Session 1, Data impact tests and OSSEs
Monday, 15 January 2007, 2:30 PM-4:00 PM, 212B
Previous paper Next paper