Contents lists available at ScienceDirect Urban Climate journal homepage: www.elsevier.com/locate/uclim Frequency analysis of annual maximum hourly precipitation and determination of best fit probability distribution for regions in Japan Jihui Yuan ⁎ , Kazuo Emura, Craig Farnham, Md Ashraful Alam Dept. of Housing and Environmental Design, Graduate School of Human Life Science, Osaka City University, Osaka 5588585, Japan ARTICLE INFO Keywords: Japan EA weather data Annual maximum hourly rainfall Return period Probability distribution function Chi-square test ABSTRACT In the design of irrigation and other hydraulic structures, evaluating the extreme rainfall for a specific probability of occurrence is important. The capacity of such structures is usually designed to cater to the probability of occurrence of extreme rainfall during its lifetime. In this study, a frequency analysis of annual maximum hourly rainfall for 15 locations in Japan was carried out using the Expanded Automated Meteorological Data Acquisition System (EA) weather data of 20 years from 1981 to 2000. Eight formulas were used to expect the return period in years of annual maximum hourly rainfall. Five different probability distribution functions (PDFs) were adopted to predict the probability distribution of occurrence of annual maximum hourly rainfall. The goodness of fit was evaluated using Chi-square test. It indicated that the Log-Pearson type 3 (LP3) distribution is the overall best fit PDF for annual maximum hourly rainfall at most locations of Japan. 1. Introduction Extreme rainfall events and the resulting floods can take thousands of lives and cause billions of dollars in damage. Flood plain management and designers for flood control works, reservoirs, bridges, and other investigations need to reflect the likelihood or probability of such events. Hydrologic studies also need to address the impact of unusually low stream flows and pollutant loadings because of their effects on water quality and water supplies (Stedinger, 1983). Frequency analysis is used to predict how often certain values of a variable phenomenon may occur and to assess the reliability of prediction. It is a tool for determining design rainfalls and design discharges for drainage works and drainage structures, especially in relation to their required hydraulic capacity. Designers of drainage works and drainage structures commonly use one of two methods to determine the design discharge. One is to select a design discharge from a time series of measured or calculated discharges that show a large variation. Another is to select a design rainfall from a time series of variable rainfalls and calculate the corresponding discharge via a rainfall-runoff transformation (Oosterbaan, 1988). Frequency analysis is also an information problem: if one had a http://dx.doi.org/10.1016/j.uclim.2017.07.008 Received 27 August 2016; Received in revised form 19 July 2017; Accepted 22 July 2017 ⁎ Corresponding author at: Dept. of Housing and Environmental Design, Osaka City University, Graduate School of Human Life Science, 3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan. E-mail address: yuanjihui@hotmail.co.jp (J. Yuan). Abbreviations: P (X ≥ x), probability of occurrence; T, return period or recurrence interval; m, rank of a value in a list ordered by descending magnitude; n, total number of values to be plotted; b, a parameter which is different in different formulas; X T , maximum value of event corresponding to return period (T); μ, mean of annual maximum hourly rainfall of observed years; δ, standard deviation of annual maximum hourly rainfall of observed years; w, a value of an intermediate variable; z, a value corresponding to an exceedance probability of P; χ 2 , Chi-square test; F(x), cumulative density function (CDF); E i , expected annual maximum hourly rainfall; Q i , observed annual maximum hourly rainfall; i, the number of observations (1, 2, ………, k) Urban Climate xxx (xxxx) xxx–xxx 2212-0955/ © 2017 Elsevier B.V. All rights reserved. Please cite this article as: Yuan, J., Urban Climate (2017), http://dx.doi.org/10.1016/j.uclim.2017.07.008