利用WRF模式不同参数化方案组合模拟玉树地区降水研究

Study on Simulating Precipitation in the Yushu Region by Combining Different Parameterization Schemes with the WRF Model

青藏高原是全球气候变化的敏感地区,该地区的降水变化对全球及区域气候系统影响深远。运用WRF模型对高原的降水状况进行模拟研究,可以揭示降水形成机制和变化规律。但目前缺少利用WRF模式的不同组合改进青藏高原降水高估。为了改善降水高估的问题,本研究利用区域气候站逐时降水实况资料和NCEP 0.25度的3小时全球再分析格点资料,采用WRF4.0,对2019年7月28日发生在玉树地区的降水事件进行了对比分析,涉及多种积云对流方案与分辨率的组合。通过比较各种方案对降水量及降水分布的模拟效果,从而得到了相关的研究结论:不同积云对流参数化方案对降水落区和强度模拟的影响显著,GD方案在降水区域模拟上效果最好,与实际降水空间分布相似性最大,而BMJ方案效果最差。总体来看,各方案均普遍高估降水量,提高模式水平分辨率可以提高降水模拟精度,但对雨带走向和范围影响不大。模式能很好地模拟降水演变过程及其后的温度、露点、CAPE和风场等物理量,除NKF-5 km方案外,其余方案的CAPE模拟结果基本与实际一致。NKF方案倾向于产生更高的CAPE,这可能是其降水模拟效果较差的原因。The Tibetan Plateau is a sensitive region for global climate change, and variations in precipitation in this area have far-reaching impacts on both global and regional climate systems. Conducting simulation studies on precipitation conditions in the plateau using the WRF model can reveal mechanisms and patterns of precipitation formation. However, there is currently a lack of research utilizing different combinations of the WRF model to improve the overestimation of precipitation on the Tibetan Plateau. To address this overestimation issue, this study utilized hourly precipitation data from regional climate stations and 3-hourly global reanalysis grid data from NCEP with a resolution of 0.25 degrees. The WRF model was employed to conduct comparative analyses of a precipitation event that occurred in the Yushu region on July 28, 2019, involving various combinations of cumulus convection schemes and resolutions. By comparing the simulation effects of different schemes on precipitation amount and distribution, the study drew relevant conclusions: different cumulus convection schemes significantly affect the simulation of precipitation regions and intensity. The GD scheme performed the best in simulating precipitation areas, exhibiting the highest similarity to the actual spatial distribution of precipitation, while the BMJ scheme performed the worst. Overall, all schemes tended to overestimate precipitation amounts. Increasing the model’s horizontal resolution can enhance precipitation simulation accuracy but has a minimal impact on the direction and extent of rain belts. The model effectively simulates the evolution of precipitation and subsequent physical quantities such as temperature, dew point, CAPE, and wind fields. With the exception of the NKF-5 km scheme, the CAPE simulation results of the other schemes were generally consistent with actual observations. The NKF scheme tends to produce higher CAPE, which may be a contributing factor to its poorer precipitation simulation performance.