Preparing for natural disasters and adapting to climate change can save lives. However, little research has examined how organizations can effectively communicate climate change adaptation and disaster risk mitigation behaviors. This dissertation employs two studies to examine how to effectively communicate disaster mitigation and preparedness to help at-risk publics better prepare for natural disasters. Fragmented studies so far have not provided an integrated model to identify the most effective factors for explaining and predicting disaster preparedness behaviors and policy support. Moreover, studies have not yet developed and tested communication messages that can motivate publics’ disaster risk mitigation through experiments. Thus, in Study 1, this dissertation attempts to build an integrated model and identify the key factors that motivate disaster preparedness behaviors and policy support through three large-scale online surveys (N = 3,468). Two of the most common federally declared disasters in large disaster-prone states are studied: wildfires and hurricanes with floods. Study 1 finds that social norms and self-efficacy strongly motivate disaster preparedness behaviors, while response efficacy strongly motivates policy support behaviors. Then, based on Study 1 and consultation with eight communication experts, Study 2 develops messages using social norms and efficacy. Study 2 tests the social norms and coping appraisal messages through four between-subject online experiments (2 X 2 X 2 X 2) with an additional vicarious experience condition in flood- and hurricane-prone states (N = 5,027). Injunctive norms and disapproval rationale strongly encourage at-risk publics to take mitigation behaviors, and vicarious experience seems promising for message design. Additionally, this dissertation reveals at-risk publics’ awareness, behavioral engagement, preferred communication channels, and information sources for preparing for hurricanes and wildfire risks. Weather forecasters and the National Oceanic and Atmospheric Administration (NOAA), including the National Weather Service (NWS), were the preferred information sources for preparing for hurricane risks. Conversely, local and state fire departments were the preferred information sources for preparing for wildfire risks. By developing and testing messages on the strongest factors using preferred information sources, the dissertation provides guidance for risk communication researchers and professionals.