Assessing Satellite Estimates of Tropical Cyclone Intensity and Structure

A growing number of objective satellite-based estimates are available to operational forecasters for tropical cyclone (TC) intensity and structure analysis, including Synthetic Aperture Radar (SAR) in recent years. For an advisory package, geostationary satellite data is available nearly 100% of the time, and polar satellite and aircraft data are available less than 40% of the time. Given the variation in data availability from one advisory to another, there is a need to understand how to best integrate all of these estimates into the TC analysis process.

Geostationary satellite imagery enables intensity estimates based on the Dvorak technique, both subjectively from human analysts and objectively from automated programs. Polar-orbiting satellites use various techniques to estimate the surface wind field and provide estimates of the intensity and wind structure, but they have limited coverage in the tropics and are at best only available a couple of times each day. Additionally, an expanding number of machine-learning based and consensus estimates exist based on input from both geostationary and polar satellite data. This talk will analyze all of the satellite-based intensity and structure estimates currently available, or soon to be available, to National Hurricane Center (NHC) hurricane specialists. Performance of a variety of techniques relative to the NHC official best track and variability during time-coincident estimates will be assessed, as well as the differences of these results with and without in-situ aircraft data.