One-Dimensional Variational Retrievals from SSMIS-Simulated Observations

Abstract

Retrievals using synthetic background fields and observations for the Special Sensor Microwave Imager Sounder (SSMIS) instrument are performed using a one-dimensional variational data assimilation (1DVAR) scheme for clear and cloudy nonprecipitating skies over open oceans. Two retrieval techniques are implemented in the 1DVAR and are extensively tested. Profiles of temperature, marine surface wind speed, and skin temperature are retrieved with both techniques. In addition, with technique A, profiles of the natural logarithm of specific humidity and liquid water path are also retrieved. With technique B, the natural logarithm of total water content (sum of specific humidity and liquid cloud water content) is retrieved instead of the natural logarithm of humidity and liquid water path. A function specifies how total water content is split among its two components. In essence, excess water vapor oversaturation leads to cloud formation. Retrievals in clear and cloudy conditions for a variety of experiments thoroughly demonstrate how technique A works. The choice of humidity control variable, the presence of biases in the moisture retrievals, and the impact of applying a supersaturation constraint are also discussed. Furthermore, in the presence of clouds, it is shown that little temperature information can be extracted with this technique if the a priori cloud vertical distribution is not known well. With technique B, however, temperature information can be extracted from the observations even in the presence of clouds. Because of its more physically based parameterization, it has some skill at positioning the cloud in the vertical direction.