碩士 / 國防大學理工學院 / 大氣科學碩士班 / 105 / Fog has harmful impact on sea and air traffic due to its low visibility near low levels. Therefore, the forecast of its formation and dissipation is a very important issue for Taiwan and its surrounding areas, especially in late winter and early spring.
There are two phases in this study. The first one is to statistically analyze the characteristics of fog weather during years 2004 to 2016, using CWB observations. The second one is to examine the performance of the numerical forecast of fog using the WRF model and two fog prediction equations for the whole year of 2016. Model sensitivity tests have been carried out for two fog cases in that year.
In conclusion, first, using the CWB surface meteorological observations from years 2004 to 2016, fogs occurred more often in Kinmen and Matsu islands and Chiayi and Tainan in western Taiwan. For all stations, most fogs formed between December to April in the 13 years. The most common type of fog was advection fog in off-shore islands and radiation fog in western Taiwan. Under the fog situation, relative humidities were over 80% and the temperature-dew point differences were less than 6 degrees of Celsius.
Secondly, the numerical fog forecasts in 2016 were verified with observations of 9 CWB ground stations marked on the google map with a 50 km-diameter circle. The successful rate in low-visibility forecast was over 80% in average.
The final conclusion of this study, for the model sensitivity tests, the initial time had great impact on fog formation timing, but little on the area and strength. On the other hand, the micro-physics schemes may have influence on the strength forecast of fog.
Identifer | oai:union.ndltd.org:TW/105CCIT0022001 |
Date | January 2017 |
Creators | LIANG, CHIA-LING, 梁佳齡 |
Contributors | LIAO, CHI-CHANG, WANG, JIAN-LIANG, 廖杞昌, 汪建良 |
Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
Language | zh-TW |
Detected Language | English |
Type | 學位論文 ; thesis |
Format | 116 |
Page generated in 0.0114 seconds