TROPICAL GEOGRAPHY ›› 2016, Vol. 36 ›› Issue (4): 673-681.doi: 10.13284/j.cnki.rddl.002846

Previous Articles     Next Articles

Comparative Analysis of Three Lightning Parameterization Schemes in WRF Model

SHI Dongdong1,YANG Zhongjiang1,XIANG Chunyan1,YANG Hu2,XU Yongbin2   

  1. (1.a.Key Laboratory for Aerosol-Cloud-Precipitation of China Meteorological Administration;b.School of Atmospheric Physics,Nanjing University of Information Science & Technology,Nanjing 210044,China;2.Tibet Lightning Protection Center,Lhasa 850000,China)
  • Received:2015-05-22 Online:2016-07-05 Published:2016-07-05

Abstract:

A squall line that occurred in Jiangsu Province on July 5, 2013, was successfully simulated with the WRF model by using three kinds of lightning parameterization schemes, which were based on empirical formulae.Comparative analysis of different schemes in simulating the distribution and frequency of lightning has beendone. Furthermore, we have discussed the reasons of simulation errors, which lay good foundation for lightningmonitoring and prediction. The results show that PR92w scheme exhibiting the location of lightning is in goodagreement with the observations, but the flash count has certain difference with the actual situation.Coincidentally, the results of PR92z scheme is opposite with that of PR92w scheme. Although forecast in highflash density using LPI is not so correct, it can indicate the distribution of lightning to some extent. Error analysishas revealed that vertical velocity and radar echo are the main factors of influencing the forecast in flash.According to the above results, increasing the grid precision and assimilating radar data is the main approach forimproving the simulation results of PR92 schemes.

Key words: WRF model, lightning parameterization, vertical wind, radar echo