A moist turbulence model based on the shallow Boussinesq equations with a simple condensation scheme is introduced. Its key advantage is its ability to express the major dynamical effects of moisture on turbulence while maintaining computational efficiency. Because of its simple condensation scheme and periodic boundary conditions, the majority of the computational and memory expense can go towards increased resolution.
Sensitivity experiments were performed on a test case of moist bubble simulations. We find that timestep choices that satisfy the CFL condition give adequate temporal resolution. Also addressed are other numerical issues regarding spurious oscillations in fields, in particular in the moisture variables. A ‘hole-filled’ experiment, which removes the negative liquid water and partially smooths the vapour and liquid water fields, indicates that these issues are not important for such a high-resolution model and field-smoothing schemes are not worth the increase in computation expense or lowered accuracy.
The moist bubble test cases also span a large range of resolutions, from 903 to 3843. The higher resolutions show shallow liquid water spectra, implying that resolution is key to correct modelling of moist atmospheric dynamics.
Acknowledgements
The authors would like to thank M. Waite, K. Ngan, L. Bourouiba and D. Straub for helpful discussions. Comments from an anonymous reviewer were also greatly appreciated. Support from the Natural Sciences and Engineering Research Council of Canada, through a Post-Graduate Scholarship (K.S.), and from the Canadian Foundation for Climate and Atmospheric Sciences, through the Quantitative Precipitation Forecasting network (P.B. and M.K.Y.) is gratefully acknowledged.