数值天气预报模式的行星边界层方案热力预报变量选择OA北大核心CSTPCD

In numerical weather prediction models,atmospheric boundary layer turbulence is parameterized via planetary boundary layer(PBL)schemes.Traditional PBL schemes often adopt potential temperature θ as the prognostic thermal variable,compute sensible heat flux and the corresponding θ-tendency.As such,most PBL schemes are designed by considering turbulent mixing of θ in a dry boundary layer.In fact,boundary layer convection is driven by buoyancy rather than sensible heat.The former accounts for moisture effects,and is represented by the virtual potential temperature θv.Based on turbulence-resolving large-eddy simulations,this study evaluates the applicability of sensible heat flux-based turbulence parameterizations in the moist boundary layer,focusing on counter-gradient correction related model coefficients.The performance of buoyancy flux-based parameterization is also investigated.It is found that buoyancy fluxes are consistent for the dry and moist boundary layers,with identical model coefficients for both cases.This study therefore recommends the use of θv instead of θ as the prognostic thermal variable for PBL schemes.