If you want to judge when the next major hurricane is about to make landfall, Flickr might be just as useful as atmospheric pressure.
A just-published study found that uploads to the photo-sharing website during the days prior to and after Superstorm Sandy's landfall last year strongly corresponded with changes to the atmospheric pressure during the same time period.
Graphs from the study show similar trajectories for both the number of Flickr photos mentioning "hurricane," "Sandy" or "Hurricane Sandy" and shifts in the atmospheric pressure as the storm made landfall.
The researchers from the University of Warwick, Boston University and University College London said they found a "striking correlation" between the two measures.
"Notably, the time of landfall of Hurricane Sandy not only marks the time of lowest air pressure, but also the time at which the largest number of Flickr photos labeled with terms related to Hurricane Sandy were taken," the researchers say. (Air pressure decreases during a hurricane.)
The findings "suggest that Flickr can be considered as a system of large scale real-time sensors documenting collective human attention," the researchers say, and such information could be useful to emergency-management officials.
The study tracked Sandy-related Flickr photos and atmospheric pressure in New Jersey from Oct. 20, 2012, to Nov. 20, 2012, showing the time leading up to and following Sandy's landfall near Atlantic City.
The researchers said Flickr users may have taken more photos as the storm became stronger, suggesting the number of photos could be used to gauge the severity of a problem.
Alternatively, users may have been expecting the storm's landfall due to media coverage and were paying more attention to the disaster at that time.
"This would equally open the possibility that increases in Flickr photo counts with particular labels may reveal notable increases in attention to an issue, such that issues which have received less extensive media coverage but which may merit further investigation may be identified," the researchers say.