改進對流系統在高解析全球模式中的表現

Abstract

熱帶洋面的衛星雲圖經常呈現不同水平尺度對流系統共存的現象，展現了大氣濕對流過程的多樣性。想要在天氣預報和氣候推估中掌握極端降雨事件，需要讓全球數值模式能正確的模擬不同水平尺度對流系統相對應的降水特徵。隨著衛星降水遙測技術和高速平行運算資源的進展，現今衛星反演的降水資料和全球數值模式具有相近的空間解析度(約10公里)。在此解析度下，可以使用衛星反演資料評估全球數值模式所解析的對流系統內部結構。本研究以衛星反演降水資料為基礎發展全新的評估方法，量化判斷數值模式是否正確地掌握不同水平尺度的對流系統中應有的降水特徵。評估結果顯示，高解析中央氣象局全球模式(CWBGFS)所模擬的大尺度對流系統降水強度偏弱，且極端降水事件的強度隨對流系統的水平尺度未能顯著增加。為了改進此模擬偏差，本研究整合了CWBGFS的濕對流效應參數化及模式解析的垂直運動場，藉由同時呈現不同尺度的濕對流系統來掌握對流發展過程。此方法在診斷模式網格內的對流上升比例後，調整濕對流過程中直接解析垂直運動的比重。整合後的CWBGFS所模擬的對流系統發展具有系統內部不均勻的水氣水平分布，產生了網格尺度的局地環流，增強了系統中降水強度的變異度。統計結果顯示，整合後的CWBGFS在極端降水強度與對流系統水平尺度的關係、海洋大陸區域降水強度的日夜變化等評估指標更優於其他全球雲解析模式。本研究突顯了未來在發展高解析全球數值模式時，應更著重了解參數化的濕對流效應及模式解析的垂直運動過程之間的關係。

Tropical convective systems play an essential role in the global climate system due to their efficient redistribution of energy, moisture, and momentum over the globe. Accurately representing the convective systems is invaluable for the simulation of extreme precipitation in weather forecasts and climate projections. Based on an object-based precipitation evaluation metric derived from satellite rainfall estimates, excessive large systems with weak precipitation have been found in the Central Weather Bureau Global Forecast System (CWBGFS) at the horizontal resolution of 15 km. The bias has been attributed to the ambiguity between the sub-grid scale and grid-scale convective processes. To alleviate this bias, we have developed and implemented the unified parameterization (UP) in the CWBGFS based on the relaxed Arakawa-Schubert scheme (RAS) to introduce a continuous transition from parameterization to an explicit simulation of the convective processes. The results show that the moisture hotspots within convective systems contribute to the enhanced local circulation and the more significant variability of precipitation. The relationship between the extreme precipitation and the horizontal scale of convective systems is improved. The UP also leads to a more realistic precipitation diurnal cycle over the land regions of the Maritime Continent because the grid-scale processes continue to produce precipitation even when the convective instability decreases during the evening. Moreover, the CWBGFS can better capture the observed object-based precipitation statistics over the Maritime Continent than the other global cloud-resolving models in the DYAMOND intercomparison. Our results highlight that investigating the physical constraints on the sub-grid cumulus effects is a key step toward the development of global convection-permitting models.