A Study of Flood Disaster Risk Communication Model and Adaptive Behavior for River-Watershed residents in Taiwan.

Abstract

  Along with the global climate change, the rainfall patterns become more centralized and cause serious damages more frequently and heavily. From the experiences of Typhoon attacks in Taiwan, the importance of risk communication with residents, especially in the vulnerable river watershed area, become the main issues of disaster prevention policy, and also lack of related academic research. This study aims to analyze key factors in risk communication mechanism and their relationships with influencing decision-making of adaptive behaviors.

  Firstly, this study build a conceptual framework of risk communication process to realize how adaptive behaviors was triggered and guided with risk communication. And then, the questionnaires were sent to the households in Kaoping River Watershed area to verify the framework by utilizing SEM model. Based on the framework, an empirical analysis was performed to analyze the key factors influencing decision-making of adaptive behaviors using multinomial logistic regression,.

   The result finds that adaptive behaviors are affected by internal cognition of disaster and meanwhile it was affected by risk communication mechanism and external environment. The crucial communication channels from Family, friends, neighbors and local governments are with higher effectiveness. The key factors influencing decision-making of adaptive behaviors are cognition of disaster and adaptive behavior. People with higher cognition of disaster and adaptive behavior would more likely to take positive adaptive behaviors. Besides, the interaction of environments, risk communication patterns and socioeconomic attributes, people from different communities would take different adaptive behaviors. Based on the empirical results, some risk communication measures are proposed to improve disaster-prevention strategies.

Keywords:

Risk communication; adaptive behaviour; structural equation modelling; multinomial logistic regression.