For the present case and for previous EDMF studies (e.
This new version of the EDMF approach based on TKE is tested for an idealized dry convective boundary layer case, similar to the one described above.
u] at simulation hours 3, 5 and 7 from the EDMF model.
Improvements are found throughout the profile, and a notable feature of the EDMF results is the good representation of the slightly stable upper half of the convective boundary layer, illustrating well that the mass-flux term is able to represent realistically the counter-gradient fluxes.
In order to have a better understanding of the performance of the new EDMF scheme, the buoyancy flux is presented in Fig.
The EDMF approach is able to simulate realistically both processes: counter-gradient fluxes and top-entrainment.
6 shows the evolution of the boundary layer height, diagnosed in the LES and in the 1D model with the BL89 and the EDMF parameterizations.
7, where it is shown that the EDMF approach is in better agreement with the LES than the BL89 approach, in both the maximum magnitude and the overall profile of the vertical velocity variance.
In order to study the sensitivity of the EDMF closure to the vertical resolution, the same case was simulated using the EDMF one-dimensional model with the vertical resolution of the ECMWF 40 and 60 level model, which has 12 levels below 1500 km height (see Teixeira 1999 for more details on the L40 and L60 ECMWF vertical resolution).
This approach differs from previous studies using EDMF methods in the sense that, by linking the eddy-diffusivity and mass-flux coefficients to the prognostic TKE, it leads to a more integrated and general approach, avoiding the use of empirical expressions, and giving a more consistent physical support to the different steps of the formulation.
This version of the EDMF scheme, implemented into the MesoNH model, produces realistic results for a dry convective boundary layer case study.