Joanna Joiner (PI)
NASA Goddard Space Flight Center
Joanna.Joiner@nasa.gov

Using Infrared Satellite Data for Assimilation and Model Evaluation

This is a two part proposal that involves the use of infrared data from NASA Earth Observing System (EOS) satellites for data assimilation and model evaluation.  The first part of the proposal continues a series of assimilation experiments utilizing radiances from the Atmospheric InfraRed Sounder (AIRS) within the fvSSI data assimilation system developed at the Global Modeling and Assimilation Office (GMAO).  The fvSSI consists of the finite volume General Circulation Model (fvGCM) coupled to the National Center for Environmental Prediction (NCEP) Spectral Statistical Interpolation (SSI) analysis system. Forecasts skills using fvSSI are state-of-the-art, i.e. comparable to those from the operational NCEP system. The goals of this part are 1) to improve the accuracy numerical weather forecasts and 2) to better represent fields in the assimilated data that are important for climate-related research such as stratospheric temperature and upper-tropospheric humidity.  Preliminary results using the standard SSI configuration with modifications to the current operational AIRS data stream and radiance errors show positive impact from AIRS data both hemispheres. We plan additional experiments to optimize the assimilation configuration and to evaluate the impact of AIRS radiances that have been 1) cloud screened with a new scheme developed at NASA GSFC and 2) cloud-cleared using an interactive variational algorithm also developed at NASA GSFC. All modifications to the fvSSI system will be adaptable to the next-generation assimilation system being developed at the GMAO.  Secondly, we propose to use radiances from AIRS and retrievals from the Tropospheric Emission Sounder (TES) to evaluate tropospheric processes within the Global Modeling Initiative (GMI) chemical-transport model (CTM). We plan to develop a set of metrics for this evaluation based on the differences between AIRS observations (level 1B radiance data) and model calculations using a radiative transfer model. Specifically, we will use radiances sensitive to carbon monoxide (CO) at 4.67 microns to assess the credibility of pollution plume transport (e.g. from boreal fires) within the CTM. We will then extend the analysis by producing 1D AIRS CO retrievals using model fields as the first guess.  We will derive additional metrics based on CO profile increments (retrieved minus model) from AIRS and TES to assess the CTM. Radiance and retrieval-based metrics can be used to compare CTM runs that incorporate different input meteorological fields or physical/chemical parameterizations.  They are also useful steps towards full assimilation of this data.  Biases in the CTM, the meteorological input fields, the observations, and forward models may be identified and potentially corrected.

+ Back to Participant Listing