An agent-based modelling framework to explore the role of social media and stubborn people on evacuation rates during flooding events
Abstract
Flood warnings from various information sources are important for individuals to make evacuation decisions during a flood event. In this study, we develop a general opinion dynamics model to simulate how individuals update their flood hazard awareness when exposed to multiple information sources, including global broadcast, social media, and observations of neighbors' actions. The opinion dynamics model is coupled with a traffic model to simulate the evacuation processes of a residential community with a given transportation network. Through various scenarios, we investigate how social media affect the opinion dynamics and evacuation processes. We find that stronger social media can make evacuation processes more sensitive to the change of global broadcast and neighbor observations, and thus, impose larger uncertainty on evacuation rates (i.e., a large range of evacuation rates corresponding to sources of information). For instance, evacuation rates are lower when social media become more influential and individuals have less trust in global broadcast. Stubborn individuals can significantly affect the opinion dynamics and reduce evacuation rates. In addition, evacuation rates respond to the percentage of stubborn agents in a non-linear manner, i.e., above a threshold, the impact of stubborn agents will be intensified by stronger social media. These results highlight the role of social media in flood evacuation processes and the need to monitor social media so that misinformation can be corrected in a timely manner. The joint impacts of social media, quality of flood warnings and transportation capacity on evacuation rates are also discussed.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.H33L..05D
- Keywords:
-
- 1813 Eco-hydrology;
- HYDROLOGY;
- 1834 Human impacts;
- HYDROLOGY;
- 1855 Remote sensing;
- HYDROLOGY;
- 1880 Water management;
- HYDROLOGY