| 摘要: | 在東北季風盛行期間，從東亞長程傳輸的空氣污染物會影響多個下風處地點，包括臺灣。許多研究嘗試量化長程傳輸污染的比例及濃度，但受限於地面測站之分布及氣象預報模式的預測極限，對於較長時間尺度或較大的空間尺度的分析上具有一定的複雜度及困難性。相較之下，衛星具有長時間且大範圍觀測的優勢，故本研究提出結合衛星遙測的方法提供更全面的見解。首先應用一個根據測站資料建立的分類方法將2017年至2023年東北風期間的污染分成三種來源模式，進一步利用衛星的氣膠光學厚度、氣膠種類和PM2.5分析長程傳輸事件對臺灣的影響及分布，發現能有效地辨識出長程傳輸污染物的種類，同時結合軌跡模型提出境外污染的預警區域 (28-31°N, 121-125°E) 及閾值 (氣膠光學厚度大於1)。另外一方面，當長程境外傳輸事件影響臺灣地區時，臺灣西半部細懸浮微粒 (PM2.5) 相較於其他事件類型增加了25%，而沙塵氣膠的PM2.5濃度增加了約60%，是在境外污染影響下增加最為明顯的氣膠種類，可作為東北季風帶來的境外污染的指標。而人為污染氣膠的PM2.5在臺灣西北部的地區沒有增加的趨勢，反而集中在臺灣西南部，也就是尾流弱風區。該處的風速較弱，擴散條件會受到抑制。本研究提出的方法實現了對長程傳輸污染的長期分析，對氣膠的境外污染傳輸有了更明確的理解，進一步提高了長程境外污染事件預測的準確性，對於跨境合作的空氣污染控制政策具有重要參考價值。;Long-range transport (LRT) air pollution from East Asia, driven by the prevailing northeast monsoon can impact multiple downwind regions, including Taiwan. Numerous studies have attempted to quantify the contribution of LRT. However, analyzing longer time scales or larger spatial scales remains complex and challenging due to limitations in the spatial distribution of ground-based measurements and the prediction of simulation models. In contrast, satellite have the advantage of long-term and wide-area observations. Therefore, this study presents an approach that integrates satellite data to provide more comprehensive insights. Initially, an efficient method based on ground-based measurement was employed to classify particulate matter (PM2.5) source patterns into three types during northeast monsoon. Subsequently, satellite-based aerosol optical depth (AOD), aerosol types and PM2.5 data were used to analyze the impact and distribution of LRT on Taiwan. The results demonstrated that this approach effectively identifies the aerosol types of LRT. By integrating the trajectory model, it can propose a warning region (28-31°N, 121-125°E) and threshold (AOD greater than 1) for LRT. Other key findings include a 25% increase in PM2.5 concentration in western Taiwan during LRT events compared to other patterns. The PM2.5 concentration of dust type, in particular, increased most significantly by approximately 60%, serving as an indicator of LRT during the northeast monsoon. However, no increasing trend was observed in PM2.5 levels of anthropogenic pollutants (AP) in northwest Taiwan. Instead, the pollution is concentrated in southwestern Taiwan, a downwind region of northeast wind, which experiences weaker wind and poorer dispersion conditions. This research achieves long-term analysis of LRT and enhances understanding of aerosol type transport. Moreover, it improves the accuracy of forecasting LRT events and proves valuable for cross-border cooperation in air pollution control policies. |
