Using Social Sensing to Validate Flood Risk Modelling in England

Floods are amongst the most severe natural disasters. Accurate flood risk maps are vital for emergency response operations and long-term flood defence planning. Currently the validation of such maps is often neglected and suffers from a lack of high-quality data. The proliferation of social media usage worldwide in recent years has supplied access to large amounts of data linked to flooding, and the detection of real-world events using such data is termed ‘social sensing’. In this paper we investigate the use of social sensing for the validation of flood risk maps. We apply this methodology to 7 years’ worth of flood related Tweets in order to perform a comparison to long term planning flood risk maps in England. The results show that there is a low level of correlation between the collection of socially sensed floods and high-risk flood areas as well as highlighting areas with high levels of socially sensed flooding that have low levels of flood risk, showcasing the potential importance of social media data for use in flood risk validation and planning policy.