For the last few years, our research team has been data-mining social media to assess the impact of earthquakes on infrastructure and people. We use machine learning approaches to analyze millions of tweets and tease out the content that is of value to assess the condition of infrastructure following earthquakes. We find that there is valuable content in social media that is posted by users that can be used for post-earthquake reconnaissance and infrastructure condition assessment within minutes after the earthquake. We will be posting more about this in the future.

For now, we are glad to share this video from our ongoing work. The video below shows social media activity related to earthquakes in Twitter for the year 2018. In total this includes 4,380,121 tweets. The location of the tweets as well as the observed spikes in social media activity track extremely well with major earthquake events that occurred during the year. More damaging earthquakes are observed to have higher tweet activity compared to less damaging earthquakes. At the end of the video, one can see the location of all tweets in 2018. What do you see?

Answer: The tweet locations map out nicely the boundaries of the tectonic plates.