Morrison方案模擬的回波(ZHH)被發現高估了觀測。分析發現，這起因於Morrison方案產生了過大粒徑(質量權重粒徑Dm>0.7mm)的雪。因此即使在低估了雪的混和比(q)情況下，模擬仍能產生更強的回波。在接下來的部分，本文討論不同冷雨過程對雪的混和比以及質量權重粒徑增量的貢獻。發現受雲滴霜化的雪(cloud-riming snow)轉換(auto-convert)成冰霰(graupel)這個過程與其他冷雨過程相比在粒徑增量上占很重要的腳色。接著針對雪的數目濃度(number concentration)和雪對雲滴(cloud)的收集效率係數(eci)設計與實行兩組敏感度測試。而結果顯示，這些測試對雪的粒徑分布模擬改善非常輕微，這暗示了微物理過程很有可能不是最主要的過程。 ;Compared to warm-rain processes which is well understood in decades of advancement, cold-rain microphysics of precipitation is still challenging task in numerical model simulation. The deficient knowledge in cold–rain processes may result in incorrect ice-phased drop size distribution (DSD) of various hydrometers simulated in microphysics scheme. Past studies have proven the inseparable relationship between polarimetric variables and storm microphysics. In the research, Morrison two moment scheme which is a double-moment (DM) scheme is selected to simulate a MCS located at southwest Taiwan on 14 June 2008 (SoWMEX-IOP8). The simulation is validated quantitatively with the NCAR s-band polarimetric measurements and DSD retrievals of raindrops and snow particles. Simulation results from Morrison scheme are found overestimating the reflectivity (ZHH) comparing to observation. The analysis reveals that stronger ZHH is due to the exaggerated mean snow particle sizes (mass-weighted diameter, Dm > 0.7 mm), even though model underestimates the snow mixing ratio (q). The increments of mixing ratio and Dm of snow particle which contributed from different cold-rain microphysical processes are analyzed. The autoconversion of graupel from cloud-riming snow is one of the dominating processes. Two sensitivity experiments including snow concentration and coefficient of collection efficiency of snow for cloud (eci) were performed. The results indicate only slightly improvements of the simulated snow DSD.
