The rain, it is a changing: non-stationarity of precipitation and the effect on water infrastructure design

Abstract

With climate change concerns, the question arises as to whether precipitation is changing with time. This has implications for the design of water infrastructure and hydraulic structures. The question of whether it is necessary to enlarge water infrastructure in the future has safety and cost implications, with larger budgets required in the future. To investigate the non-stationarity of precipitation, this paper analyzes all the precipitation records for the entire period of record for approximately 100 of the most populated cities, both in the USA and throughout the world. In particular, the 10-year and 100-year return period precipitation magnitudes were calculated marching through time to see if they are changing. The 10-year return period is commonly used as a design flow for small structures, while the 100-year precipitation is used for larger structures and flooding concerns. It was found that for some cities, both the 10- and 100-year precipitation magnitudes were both increasing, while for other cities, both were decreasing. Similarly, there was a lesser number of cities for which one or the other was increasing and the other decreasing. Results are presented on a newly introduced structure enlargement plot (SEP) that shows each category of city precipitation trends in a separate quadrant for easy visualization of both trend and magnitude. It was found that Dhaka, Bangladesh had the highest increase of both the 10- and 100-yar events, therefore indicating the need to enlarge both small and big water infrastructure. Houston had the largest in the USA. Conversely, Ho Chi Min City, Vietnam (Portland for the USA) had the most negative increase of both measures, thereby indicating no need for either small or large water infrastructure.