Using citizen science reports to define the equatorial extent of auroral visibility

The article “Using citizen science reports to define the equatorial extent of auroral visibility” was recently accepted to the American Geophysical Union (AGU) journal Space Weather. The article, for which Nathan was the lead author, describes how citizen science reports from the Aurorasaurus project can be used to determine the extent of auroral visibility and predict how far it might be seen in the future.


An aurora may often be viewed hundreds of kilometers equatorward of the auroral oval owing to its altitude. As such, the NOAA Space Weather Prediction Center (SWPC) Aurora Forecast product provides a “view-line” to demonstrate the equatorial extent of auroral visibility, assuming that it is sufficiently bright and high in altitude. The view-line in the SWPC product is based upon the latitude of the brightest aurora, for each hemisphere, as specified by the real-time Oval Variation, Assessment, Tracking, Intensity, and Online Nowcasting (OVATION) Prime (2010) aurora precipitation model. In this study, we utilize nearly 500 citizen science auroral reports to compare with the view-line provided by an updated SWPC aurora forecast product using auroral precipitation data from OVATION Prime (2013). The citizen science observations were recorded during March and April 2015 using the Aurorasaurus platform and cover one large geomagnetic storm and several smaller events. We find that this updated SWPC view-line is conservative in its estimate and that the aurora is often viewable further equatorward than is indicated by the forecast. By using the citizen reports to modify the scaling parameters used to link the OVATION Prime (2013) model to the view-line, we produce a new view-line estimate that more accurately represents the equatorial extent of visible aurora. An OVATION Prime (2013) energy-flux-based equatorial boundary view-line is also developed and is found to provide the best overall agreement with the citizen science reports, with an accuracy of 91%.

You can download the article from Lancaster University’s repository.