Bivariate frequency analysis of rainfall intensity and duration for urban stormwater infrastructure design

Changhyun Jun, Xiaosheng Qin*, Thian Yew Gan, Yeou-Koung Tung, Carlo De Michele

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

13 Scopus citations


This study presents a storm-event based bivariate frequency analysis approach to determine design rainfalls in which, the number, intensity and duration of actual rainstorm events were considered. To derive more realistic design storms, the occurrence probability of an individual rainstorm event was determined from the joint distribution of storm intensity and duration through a copula model. Hourly rainfall data were used at three climate stations respectively located in Singapore, South Korea and Canada. It was found that the proposed approach could give a more realistic description of rainfall characteristics of rainstorm events and design rainfalls. As results, the design rainfall quantities from actual rainstorm events at the three studied sites are consistently lower than those obtained from the conventional rainfall depth-duration-frequency (DDF) method, especially for short-duration storms (such as 1-h). It results from occurrence probabilities of each rainstorm event and a different angle for rainfall frequency analysis, and could offer an alternative way of describing extreme rainfall properties and potentially help improve the hydrologic design of stormwater management facilities in urban areas.

Original languageEnglish
Pages (from-to)374-383
Number of pages10
JournalJournal of Hydrology
StatePublished - 1 Oct 2017


  • Bivariate frequency analysis
  • Copula theory
  • Design rainfall
  • Rainfall DDF
  • Storm-event analysis

Fingerprint Dive into the research topics of 'Bivariate frequency analysis of rainfall intensity and duration for urban stormwater infrastructure design'. Together they form a unique fingerprint.

Cite this