| 摘要: | 西南氣流 (Southwest Monsoon, SWM) ，每年為菲律賓帶來近過半的降雨，過往有許多相關研究，然而降雨中的微物理卻未被探索。本研究使用2018年六至九月期間，PARSIVEL 雨滴譜儀的觀測，以及位於Tagaytay的 C波段雙偏極化氣象雷達的資料，分析菲律賓南方呂宋島 (South Luzon) 降雨的微物理特徵。將降水分為四種型態：大範圍強對流 (SWC) 、獨立強對流 (SIC) 、淺對流 (WC) 與層狀 (WS) 降水。利用主成份分析法 (PCA) 分析雨滴粒徑分布 (Drop Size Distribution, DSD) 和降雨積分參數 (integral rain parameters) ，客觀地將DSD分群 (PCA Groups, PGs) 以分析和不同微物理過程的關聯。
透過比對PGs和雷達參數，如：回波 (ZH)、差異反射率 (ZDR) 、比差異相位差 (KDP) 與相關係數 (ρ_HV) 時間序列剖面圖，顯示雷達參數垂直分佈與時序變化和微物理過程特徵一致。SIC由高濃度的中至大雨滴組成，伴隨高ZH、ZDR與KDP數值，且發展達高度10公里，觀測到以冰相為基礎的PG對流分群，並伴隨降雨率 (R)、液態水含量 (LWC) 和ZH的極值，也解釋地面觀測到大雨滴的現象。SWC有較廣的雨滴粒徑分布區間，且許多分群會隨系統轉移，其log10Nw (雨滴濃度 參數)、Dm (質量權重平均粒滴直徑) 與LWC可與SIC相當，卻伴隨較弱降雨率。WC通常可見高濃度小雨滴的淺對流，強回波 (ZH > 35 dBZ) 不超過6公里高。WS主要由小至中型雨滴組成，低降雨率，有明顯亮帶訊號，其DSDs緊密分布於對流/層狀分界線以下。
本研究近一步使用CFAD (Contoured frequency by altitude diagrams) 找出雷達參數垂直結構與PGs的關係。常態對流 (PG1) 與以冰相為主的對流 (PG6) 在各雷達參數均有較高數值，PG6的回波值可達40dBZ，且出現顯著柱狀ZH、ZDR與KDP。至於淺對流 (PG3) 的強回波在5公里以下。在CFADs對層狀的分群中，弱層狀 (PG2) 與中度層狀 (PG4) 高頻率地出現低ZH、ZDR與KDP數值，中度層狀因為更強的融解作用，稍微有著更高的數值，較大的雨滴會反映在ZDR CFAD中。上述發現對於菲律賓過往降雨事件研究中未提及的部分，提供了更詳盡的觀點。
;The Southwest Monsoon (SWM), which brings nearly half of the Philippines′ annual rainfall, has been extensively studied; however, its rainfall microphysics remain largely unexplored. This study investigates the microphysical characteristics of rainfall in South Luzon, Philippines, during the 2018 SWM season (June to September), using data from a PARSIVEL disdrometer and the Tagaytay C-band dual-polarization weather radar.
Four precipitation types are presented: strong widespread convection (SWC), strong isolated convection (SIC), weak shallow convection (WC), and weak stratiform (WS). Applying principal component analysis (PCA) to the drop size distribution (DSD) and integral rain parameters, PCA groups (PGs) are identified and linked to dominant microphysical processes. PGs and the averaged time-height plots of reflectivity (ZH), differential reflectivity (ZDR), specific differential phase (KDP), and co-polar cross-correlation coefficient (ρHV) were time-matched to reveal these processes. SIC exhibited high concentrations of mid-to-large drops and with high ZH, ZDR, and KDP values reaching heights up to 10 km ¬¬¬—typical for deep convective systems. Most ice-based convection DSDs (PG6) were observed after the peak rainfall rate (R), liquid water content (LWC), and ZH, explaining the large drops observed at the surface. SWC showed wider range of drop diameters with multiple PGs observed showing the transitions within the system. The values of normalized intercept parameter (log10Nw), mass-weighted mean diameter (Dm) and LWC are comparable to SIC, but with lower R. High concentrations of small drops were typically observed in WC, but convective reflectivities (>35 dBZ) were confined at the lower levels, hence the shallow classification. Meanwhile, in WS, small to mid-sized drops dominate, with low R, with an evident bright band signature and DSDs are tightly clustered below the convective-stratiform (CS) separation line.
Contoured frequency by altitude diagrams (CFAD) were used to relate radar vertical profiles to specific PGs. Both normal convection DSD group (PG1) and PG6 have high values of the same radar parameters. However, reflectivity values for PG6 are centered around 40 dBZ and usually show pronounced columns of ZH, ZDR and KDP. For shallow convection (PG3), high values of ZH (≥35 dBZ) are confined below 5 km suggesting the less intense and shallow nature of this convection compared to other convective groups. CFADs for stratiform groups—weak (PG2) and moderate (PG4) — revealed high frequencies of low ZH, ZDR and KDP values. PG4 has slightly higher values, likely due to more intense melting, which produces larger drops as reflected in the ZDR CFAD. These findings provide a more detailed perspective on Philippine precipitation systems which were not presented in previous literature. |
