Climate changes have induced extreme weather conditions, storm surge, and sea level rise, such as enormously low temperature, strong wind, heavy snow, flash flood, fog, hurricane, tsunamis, etc. in the CSET consortium states (Alaska, Hawaii, Washington, and Idaho) and tremendously impacted traffic safety performance by resulting in more frequent and serious traffic crashes in Rural, Isolated, Tribal, or Indigenous (RITI) communities. This project aims to develop a new hybrid approach to integrate multinomial Logit model with Bayesian network to discover the underlying patterns behind rural crash data and investigate the impacts of significant contributing attributes on crash severities impacted by changing climates and weather extremes. The developed data visualization platform will provide user-friendly interfaces for further studies on rural crash safety implications of behavioral characteristics of RITI drivers. Cost-effective countermeasures will be recommended for reducing rural crash severities impacted by changing climates and weather extremes.