资源环境科技发展态势分析平台

Anthropogenic climate change influences the nature and probabilistic behavior of extreme climate phenomena over time. Current infrastructure design of water systems, however, is based on intensity-duration-frequency (IDF) curves that assume extreme precipitation will not significantly change. To sustain the reliability of infrastructure designs in a changing environment, time-varying nonstationary-based IDF curves must replace the static stationary-based IDF curves. This study outlines a fully time varying risk framework using Bayesian Markov chain Monte Carlo techniques to incorporate the impact of different complex nonstationary conditions on the occurrence of extreme precipitation in the Great Lakes area. The results demonstrate the underestimation of the extreme precipitation using stationary assumptions and the importance of updating infrastructure design strategies in a changing climate.