An Ultra-high Resolution Synthetic Precipitation Data for Ungauged Sites

• Published in: Asia-Pacific journal of atmospheric sciences 2018, 54(2), , pp.215-223
• Main Authors: Kim, Hong-Joong, Choi, Kyung-Min, Oh, Jai-Ho
• Format: Article
• Language: English
• Subjects: Atmospheric precipitations; Atmospheric Sciences; Climatology; Correlation coefficient; Detection; Earth and Environmental Science; Earth Sciences; Geophysics/Geodesy; Heavy rainfall; Hurricanes; Hydrologic data; Meteorological data; Precipitation; Precipitation data; Rain; Rainfall intensity; Resolution; Statistical analysis; Statistical analysis of data; Statistical methods; Typhoons; 대기과학
• ISSN: 1976-7633; 1976-7951
• DOI: 10.1007/s13143-018-0005-3

Despite the enormous damage caused by record heavy rainfall, the amount of precipitation in areas without observation points cannot be known precisely. One way to overcome these difficulties is to estimate meteorological data at ungauged sites. In this study, we have used observation data over Seoul city to calculate high-resolution (250-meter resolution) synthetic precipitation over a 10-year (2005-2014) period. Furthermore, three cases are analyzed by evaluating the rainfall intensity and performing statistical analysis over the 10-year period. In the case where the typhoon “Meari” passed to the west coast during 28-30 June 2011, the Pearson correlation coefficient was 0.93 for seven validation points, which implies that the temporal correlation between the observed precipitation and synthetic precipitation was very good. It can be confirmed that the time series of observation and synthetic precipitation in the period almost completely matches the observed rainfall. On June 28-29, 2011, the estimation of 10 to 30 mm h −1 of continuous strong precipitation was correct. In addition, it is shown that the synthetic precipitation closely follows the observed precipitation for all three cases. Statistical analysis of 10 years of data reveals a very high correlation coefficient between synthetic precipitation and observed rainfall (0.86). Thus, synthetic precipitation data show good agreement with the observations. Therefore, the 250-m resolution synthetic precipitation amount calculated in this study is useful as basic data in weather applications, such as urban flood detection.