The Marshall Automated Wind Algorithm: Error Analysis, Quality Control and Climate Applications

The Marshall Automated Wind (MAW) algorithm for wind determination from geostationary satellite data is described and contrasted with other methods. The major differences between the MAW technique and others are found in the target selection, template size, search area constraints, and in the quality control procedures. Using appropriate data dependent tracking parameters and quality controls, random wind errors are held to around 3.0 ms as estimated with structure function analysis. It is shown that structure function analysis is also a useful tool to monitor the reduction in wind errors as a result of tracking procedure improvements. A tracking error parameter is developed to provide guidance for the trade-offs between spatial and temporal resolution when rapid scan satellite imagery is used. Major sources of error implicit in wind tracking are discussed.