A41I-0193
Inflation Factors for Satellite XCO2 Retrieval Errors
Thursday, 17 December 2015
Poster Hall (Moscone South)
Brad Weir1, Lesley E Ott1, Abhishek Chatterjee2, Krzysztof Wargan3, Jon Nielsen1 and Steven Pawson1, (1)NASA Goddard Space Flight Center, Global Modeling and Assimilation Office, Greenbelt, MD, United States, (2)NASA Goddard Space Flight Center, Greenbelt, MD, United States, (3)Science Systems and Applications, Inc., Lanham, MD, United States
Abstract:
In order to make inferences about measurements and their relationship to model data it is essential to understand their uncertainty. Satellite retrievals of trace gas mixing ratios usually quantify uncertainty by using a retrieval error covariance matrix. This work evaluates the consistency of the reported error covariance of retrievals of XCO2 from the Atmospheric CO2 Observations from Space (ACOS) algorithm applied to the Greenhouse Gases Observing Satellite (GOSAT) and Orbiting Carbon Observatory 2 (OCO-2). The approach approximates inflation factors using a posteriori diagnostics derived from the state estimation of fields of CO2 mixing ratios in the NASA Goddard Earth Observing System 5 (GEOS-5) model. A significant benefit of applying these diagnostics to the state estimation problem, as opposed to the inversion problem, is that they are able to separate retrieval errors from transport errors. The estimated retrieval error inflation factors for ACOS V3.4 are the greatest over Northern Africa and Australia. This presentation shows that over Northern Africa, the need for inflation is most likely due to the bias correction, and over Australia, the need for inflation is most likely due to uncertainty in the aerosol parameterization. The impact of the inflation is then assessed by comparing the results of the state estimation with and without inflation against in situ observations.
