台灣近幾年來飽受水災與乾旱所苦。水的問題影響民生生活甚俱,因此如何解決此問題,最根本的方法是從水何時落下及水量有多少等處著眼。而大氣中的水氣含量雖然只佔地表所有水分的0.001%,但卻控制著許多天氣的現象,如颱風與雷雨等劇烈天氣的發生。因此知道水氣的分佈及變化對於瞭解氣候及改善氣象預報有極大的幫助。 從水循環過程中發現,地球上的水藉由陽光的熱能所產生的蒸發作用,轉換成水氣傳送到大氣層中,大氣層中的水氣再經由凝結(雲、霧)與降水(雨、雪)的形式回到地表,由此可知水氣與降雨必定存在某種程度的關係。因此本篇論文主要的目的為探討可降水量與降雨的關係程度。首先分析1993~2002十年的台北氣象站探空氣球資料,得到可降水量與降雨量在統計上有相當好的關係,其相關係數高達0.83。而在缺水嚴重的2002年,夏季可降水明顯較過去少,及納莉颱風侵台的2001年,可降水量則是歷年來最高。而一年之中以夏季的可降水量與降雨量最高,春秋次之,冬天最低。而冬季可降水量在50公釐以上,秋季可降水在65公釐以上和夏季可降水在70公釐以上時,降雨機率皆高達100%,同時發現降雨機率隨著可降水量增加而增加,當可降水量愈大時,不僅容易降雨且發生強降雨的機率也較高。由於探空氣球一天只有兩筆資料,但水氣隨著時間及空間變化大,因此只有探空氣球的資料是不足以代表水氣變化情況。然而知道水氣隨時間變化的情況,有助於預測當地的天氣及了解全球的天氣系統。因此接下來利用有較好空間及時間解析度的GPS資料,即內政部和氣象局2002年至2003年GPS網形觀測衛星及氣象資料做分析,發現一天之中溫度大約於清晨6:00時開始上升,隨後溫度開始上升,於中午時達到最大值,因此水氣含量在中午和下午最多。夏季時東部沿海及山區測站由於受到局部環流的影響,水氣的日變化較西部及南部測站大。最後和逐時雨量資料比較發現,一天之中水氣較多的時刻也較容易產生降雨,且發生強降雨時,可降水量會在降雨之前持續增加,當降雨發生後,可降水量則會迅速降低。 It is a common knowledge that there exists some correlation between the amount of water vapor in the air and rainfall. The purpose of this study is to investigate the relationship between precipitable water (PW) and rainfall. Two data sets are utilized, namely radiosonde data acquired at the Banchiao weather station of Central Weather Bureau (CWB) from year 1993 to year 2002, and data of ground based GPS network operated by CWB and Ministry of Interior 2002-2003. Many interesting features are observed from this investigation. Some of them are briefed here as example. First, there is a trend that rainfall is likely to occur when the amount of PW is increased. This does not infer that rainfall occur when PW is high. Second, the probability of rainfall occurrence is 100% if diurnal average PW exceeds 70 mm in summer and 55 mm in winter. Third, PW is generally higher during daytime especially in the afternoon. Amplitude of diurnal variation in PW is higher in the eastern Taiwan and mountain areas possibly due to more active local circulation associated with complex topography and land-sea breeze. While some trends are observed, we must admit that clear relationship between PW and rainfall is quite site and time dependent. Further investigation is very much needed.