Development and evaluation of mei-yu season quantitative precipitation forecasts in Taiwan river basins based on a conceptual climatology model

Chung Chieh Wang, Chu Ying Kung, Cheng Shang Lee, George Tai Jen Chen

Research output: Contribution to journalArticle

5 Citations (Scopus)


Using rain gauge data during 134 mei-yu frontal cases in May-June 1991-2006, a rainfall climatology in relation to the positions of fronts every 0.5° in Taiwan is obtained, showing widespread precipitation with maxima over windward mountain slopes associated with frontal passages. For sixmajor river basins, rainfall characteristics and synoptic factors are further analyzed to build a conceptual climatology model for shortterm quantitative precipitation forecasts (QPFs). The four basins in central-southern Taiwan exhibit increased heavy-rainfall frequencies (.25%) roughly from 2° south to 1° north of the front, while the two basins in northern Taiwan have lower frequencies with different characteristics mainly due to the differences in their topography. The synoptic factors in the checklist employed by the Central Weather Bureau and important to heavy rainfall are identified for each of the six basins through statistical tests and their threat score (TS). These factors include those related to mei-yu fronts, low-level jets and moisture, upperlevel divergence-diffluence, and short-wave troughs. A conceptual climatology model that uses both synoptic and probability forecasting guidance is developed, and in practice the average rainfall climatology is replaced by one obtained for heavy-rainfall periods if either of the two guidance schemes indicates heavyrainfall possibility. This model for 0-6- and 0-12-h QPFs is also evaluated for its usefulness using cases during the 2007-08 seasons. With typical TSs of 0.2-0.3 (for heavy rainfall), this approach outperforms simple climatology in all six basins especially toward higher thresholds (about 20-50 mm) and for 12-h events, where it also shows advantages over model QPFs in southern Taiwan. Thus, the model can provide useful information for operational use.

Original languageEnglish
Pages (from-to)586-607
Number of pages22
JournalWeather and Forecasting
Issue number3
Publication statusPublished - 2012 Jun 1


ASJC Scopus subject areas

  • Atmospheric Science

Cite this