Community experiences of the Mid North Coast Floods: A social media analytics approach

This research examines how communities experienced and responded to the 2025 Mid North Coast floods by analysing social media activity surrounding the event. Drawing on publicly available posts and comments from Facebook, Instagram and X, the project explores how residents communicated, interpreted warnings and coordinated practical decision-making. The analysis provides insight into both the benefits and limitations of social media as a disaster communication ecosystem. The project focuses on several interconnected aspects of community response.

It investigates how flood warnings were perceived, interpreted and acted upon, paying particular attention to the relationship between official emergency messaging and community-led information sharing. The research found that third-party intermediaries emerged as highly influential actors in shaping how residents understood and responded to evolving risks. Their prominence is amplified by platform algorithms that reward highly engaging content and, in some cases, incentivise commercially motivated actors to compete with official agencies for attention during disasters.

The study also examines how people navigated evacuation decisions, sought route information and coped with delays and inconsistencies across emergency systems, revealing how social media became a tool for coordination, reassurance and critique. Findings show that social media communication during the floods was characterised by a strong reliance on peer-to-peer exchanges and intermediary pages, alongside selective engagement with formal emergency authorities. While official posts were valued for authoritative information, users frequently turned to conversational threads and locally focused pages to clarify whether warnings applied to them.

At the same time, the study highlights challenges created by information overload, fragmented information systems, inactive or under-maintained official accounts, and platform algorithms that disadvantage posts directing users off-platform to agency-controlled disaster apps and maps, ultimately weakening the visibility of official messages and strengthening engagement-optimised intermediaries.

Overall, the results suggest that emergency management agencies and authorities must adapt their strategies to a social media environment that is increasingly shaped by opaque algorithms. To improve effectiveness, agencies should strengthen locally targeted messaging, maintain active and trustworthy platform presences, collaborate carefully with influential intermediaries, and reduce reliance on off-platform calls to action where possible. Doing so will help ensure that agencies remain authoritative voices in social media discussions that can provide accurate, timely information that is visible to communities navigating the complex circumstances of natural disasters in Australia.