Today, when extreme weather affects an urban area, huge numbers of digital data are spontaneously produced by the population on the Internet. These “digital trails” can provide insight into the interactions existing between climate-related risks and the social perception of these risks. According to this research “big data” exploration techniques can be exploited to monitor these interactions and their effect on urban resilience. The experiments presented in this paper show that digital research can amplify key issues covered by digital media and identify the stakeholders that can influence the debate, and therefore the community's attitudes towards an issue. Three corpora of Web communication data have been extracted: press articles covering the June 2016 Seine River flood, press articles covering the October 2015 Alpes-Maritimes flood, and tweets on the 2016 Seine River flood. The analysis of these datasets involved an iteration between manual and automated extraction of hundreds of key terms, aggregated analysis of publication incidence and key term incidence, graph representations based on measures of semantic proximity (conditional distance) between key terms, automated visualisation of clusters through Louvain modularity, visual observation of the graph, and quantitative analysis of its nodes and edges. Through this analysis we detected topics and actors that characterise each press dataset, as well as frequent co-occurrences and clusters of topics and actors. Profiling of social media users gave us insights into who could influence opinions on Twitter. Through a comparison of the three datasets, it was also possible to observe how some patterns change over time, in different urban areas and in different digital media contexts.