Relying on social media for data in disaster response

Working for Médecins Sans Frontières (MSF), I’m required on the occasional weekend to check MSF’s Twitter account for questions and activity to respond to. The account is currently followed by 100,000 followers.

On a recent weekend, I came across a tweet at @MSF which got me thinking:

My response was as follows:

In the last few weeks, the world has been hit by two Category five hurricanes through the Caribbean and the United States (Wikipedia, 2017)– Hurricanes Irma and Maria – and two powerful earthquakes in Mexico. (Médecins Sans Frontières, 2017) My thought was, how much information do we get in disaster response such as the events in the Caribbean and in Mexico based on tweets such as these? How often do we see people pointing responders – using social media – in the direction of people who need help? People flagging what the scale of disaster is? And then when the response does come, how much is social media a tool for where can people go to help, or even criticise the repose?

In Hurricane Irma, first the scene was set:

Roisin Read and her colleagues outlined in a paper on humanitarian response information systems that the hurricane that struck Haiti in 2010 first outlined the potential of social media in humanitarian response:

At first the goal was simply to map the unfolding crisis and identify where people had moved, and it was not connected to any official humanitarian response efforts. However, as the digital map grew, emergency responders began to see how it might assist them. The processes of the digital humanitarians began to change to take a more active (though geographically remote) role in the response… the Haitian crisis highlighted the fact that real time data could now feature in humanitarian responses. (Roisin Read, 2015)

Social media – in this case Twitter, using data on phones, perhaps one of the few ways to get information out in the aftermath of a disaster like this – allowed Barbuda to tell the world just how bad things were:

The situation throughout much of the Caribbean after Irma was desperate; but social media tools allowed responders to consider what to do next – “good contextual knowledge is essential in designing humanitarian responses.” (Roisin Read, 2015)

Next, comes the help – targeted to those areas that need it most:

Büscher et al note that crowd-sourced information in the hands of digital volunteer networks ‘can support faster and more detailed awareness of the needs of affected communities and the nature and extent of damage. (Roisin Read, 2015)

…or the fundraising campaigns…

And few disasters pass by without some criticism being levelled at one or more responding institutions. In this case, the British government received stinging rebukes on their slow response, in contrast to the French government – and in contrast to one of the main points of using social media tools in disaster response “is that data can be gathered and conveyed at greater speed, with an impact on the timeliness of humanitarian responses.” (Roisin Read, 2015)

So is social media tools like Twitter used in response to disasters? Yes. Are they useful for responders? Yes, absolutely. As Read and her colleagues conclude, “The promise of greater accuracy and speed of information gathering, together with the novelty aspect that technology can bring, may constitute material power and demand-resource reallocation within international organisations and INGOs.” (Roisin Read, 2015)

The power of social media behind disasters is that they can tell the whole story. From initial disaster – often even capturing the disaster itself – to the consequences, to the response, to the response assessment.


Médecins Sans Frontières. (2017, September 21). Mexico: MSF assists people following Mexico City earthquake. Retrieved from Médecins Sans Frontières:
Roisin Read, B. T. (2015, December 22). Data hubris? Humanitarian information systems and the mirage of technology. Third World Quarterly.
Wikipedia. (2017, September). 2017 Atlantic Hurricane Season. Retrieved from Wikipedia: