Analysis Crisis Twitter — Snurb, 16 March 2011

Given the recent focus on this blog on the use of social media during major disasters, a closer look at the use of Twitter during the devastating Christchurch earthquake on 22 Feb. 2011 is long overdue (and sadly, more on the horrible Japanese earthquake and tsunami will follow soon, too). I’m following our previous methodology here – much as we’ve employed it for looking at the Queensland floods and other major events before. However, rather than relying on data from Twapperkeeper (which may not remain available for much longer, given the recent changes in how Twitter governs its data), we’re now using our own install of yourTwapperkeeper, which (with a few tweaks – more on that at another time) I can thoroughly recommend so far.

First, an overall look at the volume of tweets. As soon as news of the earthquake broke, I set our yTK to archive #eqnz, the major hashtag associated with the disaster, so we have data from exactly 10:28:16 (AEST) onwards; the earthquake itself took place at 9:51 AEST, or 12:51 local time. (Working across timezones will be a little confusing in what follows, so I’ll stick with Australian Eastern Standard Time (AEST) here – for New Zealand time, add three hours.) In other words, we’re missing only the first 37 minutes here:


Very obviously, then, Twitter coverage of the earthquake spikes within the first hours of the event, at nearly 7500 tweets/hour – this is the phase when locals and more distant onlookers alike are likely to be tweeting and retweeting the first reports in order to get the news out; a very clear illustration of what Alfred Hermida has called Twitter’s ‘ambient journalism‘ role: it may lie dormant for most of the time, but it’s there when you need it.

So who were the major participants in the Twitter conversation, then? As always, an analysis of @replies (including old-style retweets) provides a very good indication of the central sources of information (click on the image for a full-size graph):


Some interesting patterns emerge here, especially also in comparison with the data from the Queensland floods emergency: again, we see government and emergency authorities as well as mainstream media as central sources of information on Twitter. However, compared to Queensland, where the Queensland Police Service (@QPSmedia) led the field, here it is a newspaper, the NZ Herald, with more than twice as many @replies as the next account. In second place is another account of a kind we haven’t seen in this form in Queensland: CEQgovtnz, the New Zealand government’s official Canterbury Earthquake Twitter account which (as I understand it) was set up to provide information about the previous major earthquake in September 2010. Having already been in place from this previous disaster, it’s no surprise that it would swing into action again, and become a major source of information, during this second earthquake – however (and this is by no means a criticism, given what the people running the Canterbury Earthquake site no doubt had to deal with in the immediate aftermath), it is also notable that its first post-quake tweet we see in the data was only at about 8:10 a.m. AEST on 24 Feb. – two days after the quake hit. @CEQgovtnz is more important as a source for the long-term relief and recovery effort, therefore, than for the immediate response to the disaster.

Many other highly notable accounts follow established patterns from the Queensland floods – we see accounts for the Christchurch City Council (@ChristchurchCC), the Fairfax-owned news Website @NZStuff, @TVNZNews, @NZcivildefence, and the @NZRedCross, for example, and (unsurprisingly, given the close relations and the significant attention paid to the disaster by Australians) also the Australian public broadcaster’s account @abcnews, for example. Also of interest are @TelecomNZ and @VodafoneNZ, though they are prominent here mainly because their advisories, e.g.

RT @TelecomNZ: Please keep ALL calls nationwide to minimum to save capacity for emergency services. Txt instead if you can #eqnz

were widely retweeted by other users. Similarly, various individuals show up here for relatively idiosyncratic reasons:

RT @georgedarroch: Incredible image of Christchurch, from the hills, moments after the quake. #eqnz

was frequently retweeted, as were

RT @anthonybaxter: Google has people finder up for #eqnz #christchurch please RT widely

and Stephen Fry’s message of support:

RT @stephenfry: Oh dear, poor Christchurch. Another horrific earthquake. #chch #eqnz

A further worthy attempt was the @safeinchch account, set up to gather and retweet information about people thought missing in Christchurch – e.g.

RT @publicaddress: Just confirmed that [name redacted] is okay. Yay! #eqnz

Tweets directed to @safeinchch mostly retweeted information about its existence, though, and the account itself made 235 tweets between 22 Feb. and 8 March; while not taken up very widely, then, it was no doubt a valuable service for anyone it managed to help…

That’s it for part 1 – I’m exploring Alfred Hermida’s idea of Twitter as ‘ambient journalism’ further in part 2.

About the Author

Dr Axel Bruns leads the QUT Social Media Research Group. He is an ARC Future Fellow and Professor in the Creative Industries Faculty at Queensland University of Technology in Brisbane, Australia. Bruns is the author of Blogs, Wikipedia, Second Life and Beyond: From Production to Produsage (2008) and Gatewatching: Collaborative Online News Production (2005), and a co-editor of Twitter and Society, A Companion to New Media Dynamics and Uses of Blogs (2006). He is a Chief Investigator in the ARC Centre of Excellence for Creative Industries and Innovation. His research Website is at, and he tweets as @snurb_dot_info.

Related Articles


(8) Readers' Comments

  1. Pingback: Mapping Online Publics » Blog Archive » Twitter in the Christchurch Earthquake, Pt. 2

  2. Pingback: Mapping Online Publics » Blog Archive » Twitter in the Christchurch Earthquake, Pt. 3

  3. Good article but I feel some clarification is in order. @safeinchch’s sole purpose was to get messages out from people who were reporting as being safe following the quake, and from people searching for people who were in the quake zone. As a result, it was a retweeting service rather than an account for original tweets. The tweets sent to @safeinchch about missing people were entered into the Google People Finder app. If messages came in, from that app or from Twitter, to say someone was known to be safe, the message was passed on to the person who tweeted the original message. Information, such as hotlines and embassy contact details for missing and safe people was passed on. Some people from countries outside of New Zealand used Twitter DM’s to @safeinchc for information to be forwarded to the Red Cross missing person’s register, where the costs of international telephone calls to the hotline were of concern.

    The account wasn’t used like a standard Twitter account but did prove very useful in connecting people and – perhaps just as importantly – enabling people to do something and feel something was being done. New Zealanders in New Zealand, Australia and Poland each contributed time. Along with a concerned Twitterer from India, this allowed the account to be retweeting and updating the Google app in real time, around the clock, in the immediate aftermath of the quake.

    At no time was @safeinchc tweeting about its existence (although, some Twitter clients made editing out the @safeinchc part of the original tweet problematic with the retweets). The fact of its existence spread virally, from the tweet I made when I said I had set it up. No other promotion was used.

    • Hi Elpie,

      thanks for that additional information – very good to hear about @safeinchch directly from you, and many thanks for running the service.

      Just as a clarification: in the post I didn’t say that tweets _by_ @safeinchch were mostly about the service itself, but that a large number of tweets _at_ @safeinchch were promoting the service – the viral spread if information that you mention in your comment. Tweets like

      RT @Mr_Madness: Please follow @SafeInChch A twitter a/c set up for those with friends & family in Christchurch to check in #ChCh #eqnz # …

      were widely retweeted, for example.

      So, no slight on @safeinchch itself intended – merely an observation about how the spread of information _about_ a service like this often seems to surpass the actual use _of_ the service itself. Perhaps that’s a necessary pattern, in fact – without a great many people hearing about the service, those who could actually make use of it would never find out about it…


  4. Pingback: Mapping Online Publics » Blog Archive » Social Media and Emergencies, in German(y)

  5. Pingback: Mapping Online Publics » Blog Archive » Broader Twitter Patterns during Acute Events

  6. Pingback: Mapping Online Publics » Blog Archive » Emergency Media and Public Affairs Conference

  7. Pingback: Mapping Online Publics » Blog Archive » Our Social Media and Crisis Communication Research in the Media