Media type: E-Article Title: Improvement of monthly and seasonal synthetic extreme values of the Neyman–Scott rainfall model Contributor: Mondonedo, Carlo A.; Tachikawa, Yasuto; Takara, Kaoru Published: Wiley, 2010 Published in: Hydrological Processes, 24 (2010) 5, Seite 654-663 Language: English DOI: 10.1002/hyp.7559 ISSN: 0885-6087; 1099-1085 Keywords: Water Science and Technology Origination: Footnote: Description: AbstractThe Neyman–Scott clustered Poisson rectangular pulse rainfall model was investigated on the basis of its performance regarding the generation of consistent synthetic extreme values relative to historical counterparts. Rainfall extremes are evaluated in 1‐, 6‐, 12‐, 24‐, and 48‐h durations. These maximum rainfall values were pooled per calendar month during the Japanese rainy season from June to October. Rainfall data were obtained from the Japan Meteorological Agency Automated Meteorological Data Acquisition System (AMeDAS) from the mountainous region of Kamishiiba in Kyushu, Japan, as well as from the maritime area of Naha in Okinawa. Based on the traditional parameter searches of the Neyman–Scott rainfall model, we demonstrate that the model could be further improved, mostly via applications of monthly data. Therefore, we reformulated the model rain cell intensity from a single‐parameter exponential distribution to a two‐parameter gamma distribution or a two‐parameter Weibull distribution. In addition, we investigated the synthetic maximum rainfall values when the third central moment equation of rainfall intensity is included in the parameter search. The results of modifying a rain cell distribution and parameter searches, including the third moment, show a significant improvement in the consistency of extreme rainfall values. Copyright © 2010 John Wiley & Sons, Ltd.