Mother Nature is unpredictable, with actions almost as random as holiday shoppers scanning online sales, but both can be tracked with real-time online indicators. Social media listening is key to sourcing these events, particularly flood maps, and to getting help to those affected by natural disasters. And it helps first responders and businesses alike takes precautions – online and off – ahead of the storm.
When at least 75% of the famous lagoon city was underwater last month due to record-breaking floods, locals and tourists alike were sent scurrying.
It’s not like they aren’t used to flooding in Venice, as it’s actually a common and expected occurrence happening around four times each year. Residents and businesses are always ready to barricade against the floods to protect homes and shops, and it gets a bit wet, but is business-as-usual otherwise.
Last month was beyond the pale though, with water levels reaching more than five feet above sea level. It did little to deter some patrons (pictured below), beyond forcing them to don rain boots in place of more fashionable footwear:
Not to take the event lightly in any way, as it was far from ‘just another wet day’ for many business owners and tourists alike. People were stranded when transportation officials closed the water-bus system due to violent storms causing small tornadoes and hurricane-force winds – and there were uprooted trees, power loss and at least 11 people died.
Learning from Loss
In the wake of the event, the ability to have access to recent reports and understanding changing flood conditions is certainly something to consider. And social media offers a largely untapped resource for exactly that.
Sourcing disasters via social media has been done to an extent, but is far from mainstream as of this writing, which is unfortunate when considering the exceptional potential it offers.
Rather than relying on traditional sensors and models alone, data reported by local people experiencing real-world phenomena, such as flood events, would offer real-time insight around the current situation, including who is being affected and how. It would be key to directing first responders to those in need of help.
So why isn’t it being used more extensively already? It’s not a perfect system, as gathering reliable, actionable data requires a multitude of posts and tweets from many people to be considered a valid resource. And finding a good way to aggregate the data in one place has been a struggle.
Take the map designed by the MIT Urban Risk Lab that was used when Broward County, Florida, USA experienced extreme flooding during Hurricane Irma, for example: “Residents affected by flooding [could] add information to the publicly available map via popular social media channels. Using Twitter, Facebook, and Telegram, users submit reports by sending a direct message to the Risk Map chatbot. The chatbot replies to users with a one-time link through which they can upload information including location, flood depth, a photo, and description.” It even reads cumbersome. Submit a report to a specific chatbot and then watch for a link to use to upload additional info back to the chatbot, which then aggregates the data? Yikes.
There are too many steps in that process, but a process of some sort is necessary, as clearly laid out during the 12th International Conference for Hydroinformatics: “Tweets about actual floods and containing a reference to a location, can be considered as flood observations. However, the observations are not made by validated instruments or reliable observers. Therefore, a single observation has to be considered as being unreliable. Multiple unique observations reporting the same flood severity however increase the probability of the observations being correct.”
The volume of activity is already happening though, as evidenced by those tweeting about the Venice flooding during and immediately following the event.
Those affected where talking and interacting with each other, seeking solutions and sharing concerns. The data already shared on social media is available right now and exactly what’s needed to inform whatever checks and balances response services require.
Taking steps to secure solid social listening skills of existing data and understanding real-time dashboarding opportunities is really where these hydroinformatic experts need to focus. Managing future events in a way that reduces (or eliminates) the preventable loss of life during natural disasters is a sentiment analysis away!
We can help you understand important analytics affecting challenges of every variety. Reach out for a demo.