CycloneCenter: Harnessing the Power of Citizen Science to Analyze Hurricane Imagery

Tropical cyclones are among the most destructive weather phenomena. Whenever possible, the intensities of these storms have been determined from in situ data or aircraft reconnaissance. More often, however, they are estimated subjectively from satellite data using a technique developed in the late-1970s and early-1980s called the Dvorak technique. Using this method, intensities may be accurately estimated by observing the properties of clouds in a single satellite image and applying a set of rules.  However, heterogeneities are introduced into the historical record with the evolution of operational procedures, personnel, and observing platforms. In some cases, multiple agencies even arrive at different estimates for the same storm. These uncertainties impede our ability to identify the relationship between tropical cyclone intensities and climate change.

The National Oceanic and Atmospheric Administration’s (NOAA) National Climatic Data Center (NCDC) has produced a 32-year (1978–2009) homogeneous dataset (HURSAT) of tropical cyclone imagery from geostationary satellites. This dataset has the potential to address some of the uncertainties in the recent tropical cyclone record. However, it would take over 12 years for a dedicated expert to apply the Dvorak technique to the entire set of nearly 300,000 images to obtain a homogeneous record of intensities.  To that end, Cyclone Center harnesses the power of thousands of volunteers (called Citizen Scientists) utilizing a web-based interface with the help of the Citizen Science Alliance.  Using this approach, the same task can be completed dozens of times over in a matter of months. This presentation will explain how the Dvorak technique was adapted for Citizen Scientists, and how their skill will be evaluated relative to the operational analyses by trained experts.