北京一次冬季极端降水过程中相态转换预报的误差分析

Errors in the forecast of precipitation type transition in an extreme winter precipitation event in Beijing

中文关键词:  极端降水  降水相态  0℃层高度  模式预报误差

英文关键词:Extreme precipitation  Precipitation type  Height of 0℃ level  Model forecast errors

中文摘要:

      应用多种常规和非常规观测气象资料以及再分析资料对2020年2月13日夜间至14日白天北京地区一次极端雨雪过程的成因进行了分析，并重点探讨了模式降水相态预报的误差及其原因。结果表明：（1）本次降水过程中，低涡系统深厚，强度异常强，移速慢，影响时间长，导致北京地区部分站点降水持续12 h左右。异常偏强的东南风急流向北京西部山前输送水汽，配合与急流相伴的较强低空风切变形成的对称不稳定，产生高降水率的斜升对流降水。较长的降水时间以及冬季夜间罕见的高降水率共同造成了此次极端日降水。（2）北京凌晨0℃层高度和地面气温下降缓慢，北京西部处于两股冷空气间的暖舌中，冷空气从东路入侵造成北京东部降温时间较西部早，且降温辐度较大，导致0℃层高度呈西高东低形势，故转雪时间东部早于西部。（3）模式预报的东路冷空气较观测偏强偏早，降水的对流性也显著弱于观测，导致其预报的凌晨地面气温较观测低，0℃层高度下降过快，从而过早预报转雪时间，高估了降雪量和积雪深度，利用非常规温度观测对模式温度廓线预报误差进行检验，可为订正模式相态转化时间预报偏差提供依据。

英文摘要:

      Based on both conventional and unconventional observations as well as reanalysis data, a case study of the mixed rain-snow storm that occurred in February 2020 over Beijing was conducted. This study aims to explore the mechanism for the formation of the extreme precipitation event and analyze errors in the forecast of precipitation type and their possible reasons. The result indicates that: (1) the slow moving of the extremely strong and deep low vortex resulted in the persistent precipitation event that lasted for up to 12 h. The anomalous southeasterly jet transported water vapor to the east of the mountainous area in western Beijing, which, combined with strong vertical shear, formed symmetric instability that was conductive to slantwise convective precipitation with high precipitating efficiency rarely occurring in winter night. The above two factors jointly contributed to this extreme rain-snow event; (2) in the early morning, the 0℃ level height and surface temperature decreased slowly. The cold air mass from Northeast Hebei exerted earlier and stronger influence on eastern Beijing, resulting in lower 0℃ level height and thus earlier rain-snow phase transition there; (3) the influence of cold air mass from Northeast Hebei predicted by the model was earlier and stronger than observations, causing a stable stratiform precipitation in the prediction while convection was significantly weaker than observations. The model predicted surface temperature was lower than observations, and the 0℃ level height decreased more rapidly than observations. As a result, the predicted rain-snow transition occurred earlier with higher snowfall and snow depth. Verification of the model predicted temperature profile against unconventional observations provided helpful information for correcting model forecasting errors in precipitation type transition.