NG31B-3804:
Observational Non-Gaussian Humidity-Temperature 1DVAR Satellite Retrievals in the West Pacific
Wednesday, 17 December 2014
John Michael Forsythe1, Anton Kliewer1, Steven J Fletcher2 and Andrew S Jones1, (1)Cooperative Institute for Research in the Atmosphere, Fort Collins, CO, United States, (2)Colorado State University, Cooperative Institute for Research in the Atmosphere, Fort Collins, CO, United States
Abstract:
The Cooperative Institute for Research in the Atmosphere (CIRA) 1-Dimensional Optimal Estimator (C1DOE) is a 1DVAR retrieval system that assumes, like in many variational and ensemble based assimilation systems, that the background and observational errors follow a Gaussian distribution. Current research indicates that this assumption may not always be true and that water vapor mixing ratio can follow a lognormal distribution. To account for this probability distribution C1DOE has been updated to include both a lognormal transform and a lognormal formulation of the cost function. Verification of an improvement over the Gaussian assumption is given by comparing the water vapor retrievals with the Microwave Surface and Precipitation Products System’s (MSPPS) total precipitable water product and with coincident radiosondes.
