Climate Dynamics

, Volume 44, Issue 5, pp 1661-1684

First online:

Open Access This content is freely available online to anyone, anywhere at any time.

Calibration of a convective parameterization scheme in the WRF model and its impact on the simulation of East Asian summer monsoon precipitation

  • Ben YangAffiliated withSchool of Atmospheric Sciences, Nanjing University
  • , Yaocun ZhangAffiliated withSchool of Atmospheric Sciences, Nanjing University Email author 
  • , Yun QianAffiliated withPacific Northwest National Laboratory
  • , Anning HuangAffiliated withSchool of Atmospheric Sciences, Nanjing University
  • , Huiping YanAffiliated withPacific Northwest National LaboratorySchool of Atmospheric Sciences, Lanzhou University


Reasonably modeling the magnitude, south–north gradient and seasonal propagation of precipitation associated with the East Asian summer monsoon (EASM) is a challenging task in the climate community. In this study we calibrate five key parameters in the Kain–Fritsch convection scheme in the WRF model using an efficient importance-sampling algorithm to improve the EASM simulation. We also examine the impacts of the improved EASM precipitation on other physical process. Our results suggest similar model sensitivity and values of optimized parameters across years with different EASM intensities. By applying the optimal parameters, the simulated precipitation and surface energy features are generally improved. The parameters related to downdraft, entrainment coefficients and CAPE consumption time (CCT) can most sensitively affect the precipitation and atmospheric features. Larger downdraft coefficient or CCT decrease the heavy rainfall frequency, while larger entrainment coefficient delays the convection development but build up more potential for heavy rainfall events, causing a possible northward shift of rainfall distribution. The CCT is the most sensitive parameter over wet region and the downdraft parameter plays more important roles over drier northern region. Long-term simulations confirm that by using the optimized parameters the precipitation distributions are better simulated in both weak and strong EASM years. Due to more reasonable simulated precipitation condensational heating, the monsoon circulations are also improved. By using the optimized parameters the biases in the retreating (beginning) of Mei-yu (northern China rainfall) simulated by the standard WRF model are evidently reduced and the seasonal and sub-seasonal variations of the monsoon precipitation are remarkably improved.


Parameter calibration The East Asian summer monsoon precipitation Convection scheme WRF model simulation