P4.3
Limb Adjustments of AMSU-A Observations
Mitchell D. Goldberg, NOAA/NESDIS, Camp Springs, MD; and D. S. Crosby and L. Zhou
The AMSU-A is a cross-track scanning instrument measuring outgoing radiation in 15 spectral regions (23.8, 31.4, 89 GHz , and 12 channels between 50.3 and 58 GHz) at 30 beam positions for each of its approximately 800 scan lines per orbit. A feature of a cross-track sounder is the variation of the measurement along the scan line due to the change in the optical path length between the earth and the satellite. This feature is called the limb effect and the variation can be as much as 30 K for the 23.8 GHz window channel and 15 K for the atmospheric temperature channels (53 - 58 GHz). Limb adjusting the measurement to a fixed view angle is a common practice at NOAA/NESDIS and is important for a number of applications. A time series of brightness temperatures require that the data are limb adjusted prior to averaging, otherwise the data averaged will be associated with different atmospheric weighting functions. Regression algorithms that depend on collocated observations of satellite data and radiosonde data are simplified if the satellite data are normalized to a fixed angle, since otherwise there would be difficulty in achieving a reasonable sample size for each beam position. Physical retrieval algorithms often retrieve along the optical path, but generally some sort of beam position dependent bias adjustment is needed to remove spot- to-spot systematic biases. If the limb adjustment procedure produce accurate limb adjusted data, then a physical retrieval would benefit since there would be no need for spot dependent bias adjustments.
The limb adjusted AMSU-A observations are currently used for deriving atmospheric temperature profiles and over the ocean - total precipitable water and cloud liquid water. An archive of AMSU-A limb adjusted observations and retrievals are available from July 1998 to present. At the conference, we will report on the limb adjustment and retrieval methodology as well as validation.
Poster Session 4, At Tribute to Lewis Kaplan: Part II
Wednesday, 12 January 2000, 9:30 AM-11:15 AM
Previous paper Next paper