A paper in April’s edition of the Bulletin of the American Meteorological Society published today and co-authored by Peter Thorne shows that citizen scientist classifiers can help unravel the mysteries of tropical cyclones and climate change.
Taken from Figure 3 in paper illustrating how uers are asked to classify the cloud pattern and perceived organisation of the Tropical Cyclone.
The historical tropical cyclone record consists of a patchwork of ocean basin analyses by different forecasting centres using similar but non-identical rules and processes. Worse still, these processes have changed through time and analysts have come and gone. So, it is unclear which aspects of variability and trends in the records we have to hand are real and which are simply an artefact. To better understand this the Citizen Science Alliance and a group of scientists primarily from NOAA’s National Centers for Environmental Information instigated cyclonecenter in 2012. The basic premise is to employ a simplified set of the Dvorak rules that forecasters use and get hundreds of pairs of human eyes to reclassify the whole record.
The www.cycloneceter.org project has amassed over 440,000 classifications from volunteers since its inception. The paper by Hennon and colleagues provides some initial analyses and shows that the volunteers provide skilful estimates of cyclone intensity. Because each image is classified by ten unique users the volunteers also provide a useful tool to begin to assess how confident we are in our intensity estimates. There are still many images left to classify so it is not too late to get involved. Go classify!
Paper is available at http://journals.ametsoc.org/doi/pdf/10.1175/BAMS-D-13-00152.1