博士 / 國防大學中正理工學院 / 國防科學研究所 / 90 / In Southeast Asia, during the summer monsoon season, heavy rainfalls induced by the Mei-Yu front and Mesoscale Convective Systems (MCSs) caused severe damage to regional economies, including those of Taiwan, Luzon in the Philippines, and Southern China. In this study, during the Mei-Yu season, the Tropical Rainfall Measuring Mission (TRMM) microwave Imager (TMI) data were used to retrieve oceanic rain rates. The oceanic rainfall characteristics were also analyzed. The retrieval results in combination with the infrared (IR) rainfall observations made by both the TRMM and the Geostationary Meteorological Satellite 5 (GMS-5) were used to estimate the quantitative rainfall over the waters near Taiwan. This approach was dedicated to providing quantitative rainfall information at least every hour. The goal was to reduce the loss in human lives and economics caused by heavy rainfalls.
Using the TMI data, a physical algorithm and an empirical algorithm were used to retrieve the oceanic rain rates. A screening methodology (combination technique) utilizing the Scattering Index, the Rain flag, and the Threshold Check techniques was used to distinguish rain regions from the TMI filed of views. Using a standard linear regression technique, the island rain gauge data and its corresponding TMI nine-channel brightness temperatures obtained during the Mei-Yu season of 1998 were used to develop a rain retrieval methodology. Using the island rain gauge data obtained during the Mei-Yu periods of 1999~2001, a quantitative validation of the rain retrievals and the rain identification performance using the TMI were then carried out. The overall successful rainfall identification rates (including rainy and no-rainfall events) were 99.2%, 97.3%, and 95.2% for 1999, 2000, and 2001 respectively. The coefficient of determination was 0.81 for the quantitative oceanic rain retrieval validation. The vertical hydrometeor profiles and surface rain rates retrieved by the physical algorithm (Goddard profiling algorithm, GPROF) using the TMI also presented advantages for the Mei-Yu rainfall study.
Rain retrievals using the TMI were assumed to be ground truth for the rain retrievals using satellite IR data. In this study, an IR algorithm: the GOES Precipitation Index (GPI) was used to estimate the oceanic rain rates. Using the Visible and Infrared Scanner (VIRS) 11-m IR data from the TRMM, the best IR threshold needed in the GPI computation was obtained using the linear regression technique. At times when the TRMM observations were not available, the following rain retrievals were made using the GMS-5 IR data within every 1o×1o grid boxes. This technique combining microwave and IR data from different satellites presented the ability to track MCS rainfalls during the Mei-Yu season. Although the spatial resolution might be not enough for analyzing oceanic MCSs and initializing numerical models, it still provided direct and reasonable threat information for heavy rainfall warnings and flash flood watches. With the addition of the results from this study, heavy rainfall forecasting during the Mei-Yu season could be provided with increased accuracy.
| Identifer | oai:union.ndltd.org:TW/090CCIT0584004 |
| Date | January 2002 |
| Creators | 李慶忠 |
| Contributors | 陳萬金 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 195 |
Page generated in 0.009 seconds