During the tsunami warning that hit Japan in November 2016, the top issue facing disaster response teams was (surprisingly) having to manage the impact of social media trolls. These trolls were masquerading footage from the 2011 disaster as if it was real-time, creating confusion within the online community.
Misinformation is not just a problem for Japan. New Zealand has had similar struggles surrounding tsunami warnings issued after a significant earthquake hit the country earlier in the same month and Australian emergency management teams also have combating trolls as a top priority.
One NZ Signal customer told us:
“Monitoring social media during emergencies such as a tsunami threat is important in order to obtain critical information, identify potential impacts and limit the spread of misinformation. This helps us respond to emergencies more effectively.”
Another from Australia said:
"Signal has been used to investigate fires when we weren’t quite sure what the exact location was. It has allowed us to debunk new footage from the old and clarify misinformation surrounding the location of fires."
How do you spot Trolls in a sea of information?
Running a search on the tsunami warning in Japan to showed how open source intelligence can be used to stop the spread of misinformation and trolling when natural disasters strike.
Over 500,000 mentions found:
- 283,246 tsunami-related mentions in 24 hours. (Source: Signal)
- 303,512 posts tagged with #tsunami hashtag (Source: Signal)
How do you cope?
As you can see, it would be impossible for a even a large disaster response team to sift through over 500,000 mentions during a 24 hour period, and accurately identify and act upon trolls and misinformation.
However, with the right software you can automate a number of tasks to make this task manageable and efficient, allowing your teams the time to concentrate on saving lives.
What does this trolling look like?
Here are some of the examples of misinformation during the recent Tsunami warning: