Tanvi Misra is a staff writer for CityLab covering immigrant communities, housing, economic inequality, and culture. She also authors Navigator, a weekly newsletter for urban explorers (subscribe here). Her work also appears in The Atlantic, NPR, and BBC.
They may not provide comprehensive information, but they can help save lives.
While Twitter may not offer a comprehensive picture of a phenomenon or an event, it can provide fast, real-time information about a large-scale disaster—like a flood—as it unfolds.
A new study by Dutch flood monitoring organizations Deltares and Floodtags finds that the tweets in Jakarta, Indonesia, rose to 900 per minute after a recent flood hit, and contained important information about location and depth of the flood waters. After an analysis of these tweets, Dutch researchers designed a method to produce a real-time map using information from Twitter, land elevation stats, and water motion data, Reuters reports:
"This method is really fast," Deltares flood expert Dirk Eilander told the Thomson Reuters Foundation. "It can produce a map within around a minute of messages being posted."
Not everyone can tweet during a disaster. But where there is Twitter activity, a tweet-based map can help save lives in an event where every minute counts. Plus, once the researchers double-checked and cross-referenced information from the tweets, the final flood map was pretty close to a photograph of the flood, they say.
PetaJakarta.org, an Indonesia-based open-source platform, used crowd-sourced information from Twitter to make a real-time map during the floods this year. According to the Jakarta Post, the organization received around 6,000 flood-related tweets in January and February.
"People are really getting into it; they go right into the flood and report the depth [of floodwater] and even take a selfie,” Etienne Turpin, the organization's co-principal investigator told the Jakarta Post in February.
Here's an example of a Tweet-based real-time flood map, posted on Twitter's blog in December 2014: