Validation of Cloud-Resolving Model Background Data for Cloud Data Assimilation

Abstract

Results from a cloud-resolving model are systematically compared with a variety of observations, both ground based and satellite, in order to better understand the mean background errors and their correlations. This is a step in the direction of developing a background error covariance matrix for use in cloud data assimilation. Observation sources include the Geostationary Operational Environmental Satellite (GOES), the Atmospheric Emitted Radiance Interferometer (AERI), a microwave radiometer (MWR), radiosonde, and cloud radar. When exploring model biases in temperature, precipitable water vapor, and liquid water path, a warm and moist bias at night and a cool and dry bias during the day are observed. Values for the background decorrelation length of water variables are determined. In addition, a dynamic cloud mask is presented to give more control in the assimilation of cloudy satellite radiances, allowing different cloud types to be excluded from the assimilation as well as establishing values for the maximum residuals to be considered.