In the wake of devastating Typhoon Haiyan in the Philippines, a newly released scientific paper investigates how to trace the progress of a natural disaster through crowdsourced and tagged photographs.
Researchers tracked the number of photos posted to Flickr with the tags Hurricane, Sandy, or Hurricane Sandy, in the days before and after the storm made landfall on the East Coast in 2012. They found the number of photos spiked exactly when the storm made landfall. In fact, tracking the number of photos posted to Flickr, as an indicator of the progress of the storm, was just as accurate as tracking the barometric pressure in New Jersey.
The researchers speculate that crowdsourced storm data might one day reveal more dimensions of the impact of an incident to governments, insurance companies, and emergency responders. “We suggest that Flickr can be considered as a system of large-scale real-time sensors documenting collective human attention. The analysis of other examples of catastrophic events, beyond this case study of Hurricane Sandy, is however needed to evaluate whether an appropriate leverage of such a system could be of interest to policy makers and others charged with emergency crisis management,” they write. Of course, the accuracy of such a system of “sensors” depends on the ratio of Flickr users and the amount of Internet access in a given area, something that may be very different in rural Philippines, for example, than in the northeastern United States.