GRAPES_MESO模式预报降水相态诊断及应用研究
投稿时间:2019-04-16  修订日期:2019-05-11  点此下载全文
引用本文:
摘要点击次数: 49
全文下载次数: 35
作者单位E-mail
佟华 中国气象局 国家气象中心, 北京 100081 tongh@cma.gov.cn 
张玉涛 中国气象局 国家气象中心, 北京 100081  
基金项目:科技冬奥专项冬奥会气象条件预测保障关键技术(2018YFF0300103);中国气象局气象关键技术集成与应用重大发展项目(CMAGJ2015Z10)
中文摘要:当地面气温在0℃附近时,降水相态通常比较复杂,降水相态预报一直是天气预报的难点之一。本文根据降水粒子在下落过程中的热力结构,利用BTC算法、修订的BTC算法、Ramer算法、Bourgouin算法4种计算方案,通过GRAPES_MESO中尺度模式提供所需参数,对落到地面的降水相态进行诊断,诊断的降水相态包括雨、雪、冻雨和冰粒4种,然后利用集成方法获得大概率且破坏程度大的降水相态预报产品。基于2个冬季降水个例,研究发现4种计算方案都能较合理地得到雨雪分界线,以及降雨、降雪落区范围。其中,修订后的BTC算法可以修正原始BTC算法对降雪的诊断偏差,并且预测的冰粒比原始冰粒更少。Ramer算法会比其他几种算法获得更多的冻雨事件,而Bourgouin算法最接近合成算法的结果。对于不确定性较大的中间态的冻雨和冰粒预报,鉴于其高影响特性,方案倾向于过高估计冻雨和冰粒的预报落区。集成的最终降水相态产品反映其发生的概率大或影响程度大,能为防灾减灾提供有效的警示信息。
中文关键词:GRAPES_MESO  模式预报  降水相态  冻雨  数值模式产品
 
Diagnosis of precipitation types and its application in the GRAPES-MESO forecasts
Abstract:Precipitation type is often complicated when the ground temperature alternates between 0℃,and the its prediction has been one of the major difficulties in weather forecasting for many years.In this paper,four diagnostic methods are used to study the precipitation types,by examining the vertical temperature structure to be encountered by a falling hydrometeor,so as to diagnose a single type when the ground is reached.The four methods are the BTC algorithm,revised BTC algorithm,Ramer algorithm and Bourgouin algorithm.According to the ability of the mini-ensemble,the results with the majority of the algorithm/s outputs and more dangerous winter weather are integrated as the final precipitation type,including rain,snow,freezing rain and hail.The GRAPES_MESO mesoscale model is used to provide the parameters that the four diagnostic algorithms require.Next,the precipitation type's products are verified by two winter precipitation cases.Compared with the observation,it is shown that the revised BTC algorithm can correct the diagnostic bias of the original BTC algorithm for snowfall,and less hail is forecasted than when using the original algorithm.The Ramer algorithm obtains a greater number of freeze-rain events than several of other algorithms.The Bourgouin algorithm is closest to the results of the synthesis algorithm.Therefore,it is shown that the four schemes can reasonably predict the boundary line of rain and snow,as well as the areas of rainfall and snowfall.However,the scheme intentionally over-predicts hail and freezing rain,viewing these two types as more dangerous varieties of winter weather.
The results are directly related to the accuracy of the forecasted temperature profiles and precipitation areas in the model.These diagnostic schemes can only determine one type of precipitation,but cannot diagnose the sleet,thus at the junction of rain and snow there are a greater number of intermediate phases such as hail and freezing rain.The final integrated precipitation type of the four schemes shows the high probability and high influence of the various precipitation types,thereby providing effective warning information for disaster prevention and mitigation.
keywords:GRAPES-MESO model  forecast  precipitation types  freezing rain  numerical model products
查看全文  查看/发表评论  下载PDF阅读器