Collective Digital Traces Could Predict Impact of Emergencies

Research undertaken by three institutions has uncovered a link between the number of images posted to photo sharing site Flickr and the intensity of the storm being captured in those Flickr images. The focus of the study was Hurricane Sandy, which hit 24 states in late October 2012. New Jersey was among those hardest hit.

Researchers from the United Kingdom-based Warwick Business School, University College London, and Boston University found that as the number of images posted to Flickr grew, the atmospheric pressure in New Jersey dropped by a comparable degree, portending the peak of the storm.

  New research suggests that digital traces, such as the number of photos uploaded to Flickr referencing Hurricane Sandy, could help researchers predict the consequences of natural disasters and other real-world events.
New research suggests that digital traces, such as the number of photos uploaded to Flickr referencing Hurricane Sandy, could help researchers predict the consequences of natural disasters and other real-world events.
 

"Quantifying the Digital Traces of Hurricane Sandy on Flickr," published this week in Scientific Reports, stated that 32 million photos were posted on Flickr tagged "Hurricane Sandy," "hurricane," or "sandy" between October 20 and November 20. The highest number of images appeared on the site in the same hour in which Hurricane Sandy made landfall in New Jersey.

The report suggested that using online indicators similar to this one could help governments measure the impact of disasters.

Warwick researchers Tobias Preis, associate professor of behavioral science and finance, and Suzy Moat, assistant professor of Behavioral Science, led the project. Both have previously been involved in other projects to find connections between what people seek out online and what they're experiencing in the real world. For example, the frequency with which online visitors search for financial information on Google and Wikipedia could be used as a signal of stock market shifts.

"This study would suggest that in cases where no external sensors are available, it may be possible to use the number of Flickr photos relating to a topic to gauge the current level of this category of problems. Flickr can be considered as a system of large-scale real-time sensors, documenting collective human attention. Increases in Flickr photo counts with particular labels may reveal notable increases in attention to a particular issue, which in some cases may merit further investigation for policy makers," said Moat. "Appropriate leverage of such online indicators of large disasters could be useful to policy makers and others charged with emergency crisis management: in particular if no secondary environmental measures are available."

About the Author

Dian Schaffhauser is a former senior contributing editor for 1105 Media's education publications THE Journal, Campus Technology and Spaces4Learning.

Featured