Enhanced Automated Quality Control Applied to High-Density Satellite-Derived Winds

Abstract

The coverage and quality of atmospheric motion vectors (AMVs) derived from geostationary satellite imagery have improved considerably over the past few years. This is due not only to the deployment of the new generation of satellites, but is also a result of improved data processing and automated quality control (AQC) schemes. The postprocessing of the Geostationary Operational Environmental Satellite (GOES) derived displacement vectors at the National Oceanic and Atmospheric Administration/National Environmental Satellite, Data, and Information Service (NOAA/NESDIS) has been fully automated since early 1996. At the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) AQC was used as support to the manual quality control (MQC) for Meteosat vector fields until September 1998 when the MQC was discontinued and fully replaced with the automated procedure. The AQC schemes at the two organizations are quite diverse. Based on a method developed at the University of Wisconsin?Cooperative Institute for Meteorological Satellite Studies, the NOAA/NESDIS AQC involves an objective three-dimensional recursive filter analysis of the derived wind fields. The fit of each vector to that analysis yields a recursive filter flag (RFF). The AQC scheme employed at EUMETSAT derives a quality indicator (QI) for each individual vector based on the properties of the vector itself and its consistency with other AMVs in close proximity. Mainly relying on satellite data, this QI-based scheme has been proven to provide a good estimate of the reliability of the derived displacements, but it fails to identify fleets of winds that are consistently assigned to a wrong height. The RFF-based scheme is capable of readjusting the heights attributed to the wind vectors, which yields a better fit to the analysis and ancillary data. These quality estimates can be employed by the user community to select the part of the vector field that best suits their application, as well as in data assimilation schemes for optimizing the data selection procedures. Even though both schemes have already been successfully implemented into operational environments, the possibility exists to exploit the advantages of both approaches to create a superior combined methodology. In order to substantiate the differences of the two methodologies, the two schemes were applied to the high-density AMV fields derived during the 1998 North Pacific Experiment from GOES and the Geostationary Meteorological Satellite (controlled by the Japan Meteorological Agency) multispectral imagery data. The performance of the two schemes was evaluated by examining departures from the analysis fields of the European Centre for Medium-Range Weather Forecasts (ECMWF) assimilation scheme, and a new combined methodology was further evaluated by looking at the impact on medium-range forecasts verified against the operational forecasts. It will be shown that the new combined AQC approach yields superior ECMWF forecast results.