高级检索
吴志鹏, 李跃清, 李晓岚, 等. 2021. WRF模式边界层参数化方案对川渝盆地西南涡降水模拟的影响[J]. 大气科学, 45(1): 58−72. doi: 10.3878/j.issn.1006-9895.2005.19171
引用本文: 吴志鹏, 李跃清, 李晓岚, 等. 2021. WRF模式边界层参数化方案对川渝盆地西南涡降水模拟的影响[J]. 大气科学, 45(1): 58−72. doi: 10.3878/j.issn.1006-9895.2005.19171
WU Zhipeng, LI Yueqing, LI Xiaolan, et al. 2021. Influence of Different Planetary Boundary Layer Parameterization Schemes on the Simulation of Precipitation Caused by Southwest China Vortex in Sichuan Basin Based on the WRF Model [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 45(1): 58−72. doi: 10.3878/j.issn.1006-9895.2005.19171
Citation: WU Zhipeng, LI Yueqing, LI Xiaolan, et al. 2021. Influence of Different Planetary Boundary Layer Parameterization Schemes on the Simulation of Precipitation Caused by Southwest China Vortex in Sichuan Basin Based on the WRF Model [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 45(1): 58−72. doi: 10.3878/j.issn.1006-9895.2005.19171

WRF模式边界层参数化方案对川渝盆地西南涡降水模拟的影响

Influence of Different Planetary Boundary Layer Parameterization Schemes on the Simulation of Precipitation Caused by Southwest China Vortex in Sichuan Basin Based on the WRF Model

  • 摘要: 应用WRF v4.0模式五种边界层参数化方案(YSU、MYJ、MYNN2、ACM2和SH),对2016 年汛期(5~9月)在川渝盆地东部造成暴雨的所有西南涡过程进行了数值模拟,检验评估了它们对各量级降水的预报能力,并基于加密的L波段秒级探空资料对比分析了模拟与实况边界层结构的差异,结合各方案对湍流运动的算法特点探讨了其差异的原因,最后对ACM2方案进行了湍流强度调整,由此改善其对于川渝盆地边界层与西南涡降水的模拟能力。结果表明:ACM2和YSU方案TS评分表现较好,相对其它方案ACM2空报较少,这种可以根据周围环境的稳定性切换局地或非局地算法的方案更适合于盆地西南涡降水模拟,但边界层方案对西南涡降水的空报都较普遍,尤以大量级降水更明显;精细的探空资料进一步表明,所有方案模拟的白天边界层高度都偏高,湍流混合强度都偏强。通过参数调整而降低混合强度的ACM2方案,模拟的边界层温湿结构则更符合实际观测,其边界层下部温度更低、湿度更高,减少了大量级降水的空报,使盆地西南涡降水模拟有一定改善;边界层参数化方案对西南涡模拟的差别主要体现为不同的西南涡位置与降水强度,但归根到底都源于方案的局地或非局地特性、不同的混合强度这两方面原因。因此,根据不同特定区域下垫面环境与气候状况合理选择方案的特性和混合强度是准确模拟边界层结构及其降水过程的关键。

     

    Abstract: Five planetary boundary layer (PBL) parameterization schemes Yonsei University (YSU), Mellor–Yamada–Janjic (MYJ), Mellor–Yamada–Nakanishi–Niino Level 2.5 (MYNN2), Shin-Hong (SH), and the asymmetric convective model, version 2 (ACM2) in the Weather Research and Forecast model (WRF v4.0), were used to simulate all well-developed Southwest China vortex (SWCV) processes in the eastern Sichuan basin in 2016. Each level of precipitation prediction was verified, and the L-band radiosonde data with temporal resolution of 1 s were used to reveal the fine structure of the PBL during a midday. The differences between the observation and simulation are assessed, and the reasons are discussed based on the characteristics of the turbulence algorithm used in each scheme. Finally, the parameter of turbulence intensity was adjusted for the ACM2 scheme to improve the structure of the PBL that influences the simulation of the precipitation in the eastern Sichuan basin. The results show that the ACM2 and YSU schemes show a relatively better TS performance. Compared with other schemes, ACM2 has fewer false alarms. The attribute of ACM2 that can modify local or nonlocal algorithms according to the stability of the surrounding environment seems to be more suitable for the Sichuan basin precipitation simulation than the other schemes. However, all PBL schemes show a high false-alarm rate in the prediction of the SWCV precipitation, especially when the precipitation is heavy. The sounding data with 1 s temporal and 3 m spatial resolution further show that all the PBL schemes predict a higher PBL height compared with that of the observations, which means that the simulation has a stronger mixing intensity compared with that of the real atmosphere. By parameter adjustment, using the ACM2 scheme with reduced mixing intensity, the potential temperature and humidity structure in PBL are more aligned with the observations. Further, the potential temperature of the low PBL is low, humidity is high, and false alarm reports of heavy precipitation are reduced, which leads to an improvement regarding the precipitation simulation in the Sichuan basin. The different characters of the PBL schemes that are used in the simulation of the SWCV mainly lead to different positions of the vortex and precipitation intensity. Essentially, these characters are derived from a local or nonlocal attribute and the intensity of vertical mixing. A selection based on regional features of a research object is the key to the accurate simulation of a PBL structure and precipitation process.

     

/

返回文章
返回