摘要
如何將我們對聖嬰的瞭解應用到區域氣候預報上是近來受到廣大關切的課題。本報告嘗試利用比較完整的分析資料重新探討聖嬰與台灣異常氣候的關係。分析步驟有三：（1）按照Nino4指標特徵判定「聖嬰」、「反聖嬰」並分類，（2）按照「聖嬰」、「反聖嬰」類別對台灣18個測站的溫度和降水百分序作合成分析，（3）挑選在聖嬰現象條件下台灣氣候明顯異常的個案分析其所對應的環流與海溫合成距平特徵。結果顯示，台灣僅有一些特定月份的異常氣候和聖嬰現象有明顯的關係。在溫度方面，「聖嬰11月」以前的5、6月偏冷，以後的12-4月偏暖。「反聖嬰11月」以前的5、6月則偏暖，以後的12-4月偏冷，並且偏冶的現象以4月最明顯。在降水方面，「聖嬰11月」以前的6月偏濕，9、10月偏乾，以後的6月偏乾，9、10月偏濕。「反聖嬰11月」以前的6月偏乾，10月偏濕，以後的10月也是偏濕。但是特別要注意的是，這些台灣溫度和雨量的特徵在各個合成個案中也有程度上的明顯差異。 在環流和溫度的關係方面，台灣偏冷時低層有北風距平，偏暖時則有南風距平；而高層的南北風距平對於台灣的溫度異常似乎並不重要。造成低層南、北風距平的系統在4、6月是以南海、東海、菲律賓海與西太平洋等海上環流系統為主；在12月則是海上與亞洲大陸上的環流系統同等重要。至於雨量方面，我們發現台灣的偏濕或偏乾除了要看低層環流是否配合之外，高層擾動的特性也很重要。台灣偏濕時，上層大氣若不是在一串流函數波列型式距平的負區內，便是在波長較長的長波槽區。特別要注意的是，若在6月南海上有低層反氣旋環流，台灣會偏乾。然而，南海上的低層反氣旋環流若是在10月出現，台灣不但不會偏乾，在高層短波活躍的情況下還會出現偏濕現象。異常海溫所導致的低層異常環流系統在不同的季節可能會造成區域氣候極端不同的反應，在探討異常海溫與區域氣候關係時不能不留心。
並列摘要
Taiwan ENSO signals are documented in this paper. The investigation follows three steps. First, the warm and cold phases of ENSO are determined and categorized by the percentile ranks of Nino4 index. Second, the temperature and precipitation percentile ranks of the 18 stations in Taiwan are composited according to the ENSO categories. Third, the corresponding circulation and sea surface temperature anomalyus patterns are analyzed for the selected months with clear ENSO signals in Taiwan. Our results indicate that the ENSO signals in Taiwan are particularly clear in certain months. It is abnormally cold in the May and June before the ”E1 Nino November”, but abnormally warm from December to April after the ”E1 Nino November”. It is abnormally warm in the May and June before the ”La Nina November”, while abnormally cold from December to April after the ”La Nina November”. The abnormally cold temperature is particularly clear in April. In the precipitation aspect, it is abnormally wet in the June before the ”E1 Nino November”, while abnormally dry in the September and October before the ”E1 Nino November”. The June after the ”E1 Nino November” is abnormally dry, while the September and October after the ”E1 Nino November” are abnormally wet. The June(October)before the ”La Nina November” is abnormally dry(wet),and the October after the ”La Nina November” is also abnormally wet. Although variations from case to case exist, there is no evidence showing the opposite signals in these abnormal cases. Few specific months are selected for depicting the circulation and anomalous sea surface temperature patterns related to the abnormal climate of Taiwan. The selected months are the June before the ”ENSO November” and the December, April and October after the ”ENSO November”. We found that the abnormal temperature in Taiwan is strongly related to the low-level wind direction. When Taiwan is abnormally cold(warm), the corresponding low-level wind anomaly is northerly(southerly). The upper-level wind anomalies seem not much affecting Taiwan temperature anomalies. In April and June, the abnormal climate in Taiwan is associated with the low-level circulation anomalies over the South China Sea, the East China Sea, the Philippine Sea and the western margin of the western Pacific. In December, however, in addition to the low-level circulation anomalies over the marginal seas, the anomalies over the north and central part of mainland China also play a critical role in shaping the low-level wind anomalies over Taiwan. In the aspect of precipitation, we found that in addition to the low-level wind anomalies, the pattern of upper-level perturbations may also be a critical factor. In the abnormally wet condition, Taiwan is under a clear upper-level deep trough that may be a part of a wave train of short waves or a sector of the trough of a long wave. Particular attention should be paid that an anomalous low-level anticyclone over the South China Sea in June may result in abnormal dry situation in Taiwan, however the similar anomalous circulation in October can result in abnormal wet situation. Therefore, the abnormal circulation patterns caused by abnormal sea surface temperature can cause opposite climate situations in Taiwan. Our results suggest that the abnormal Taiwan local climate is sensitive to season, in other words, to the characteristics of the East Asian monsoon system.