Convective triggering is key to understanding the diurnal cycle of convection and the associated rainfall. Traditional triggering schemes in mesoscale models do not account for roll circulations because they are poorly resolved, and the impact rolls have on convection are mostly unknown. In this study, we use several cases of convective roll circulations observed in January 2016 from the Himawari-8/9 satellite over the Gulf of Carpentaria in Northern Australia to test the convective triggering process. These cases were simulated using the Weather Research and Forecasting (WRF) Model (version 3.7.1) starting from a resolution of 9 km with convective parameterization and downscaled up to a resolution of 333 m using three additional nested grids with explicitly resolved convection. The simulations were evaluated with satellite-derived outgoing longwave radiation used as a proxy for deep convection. We then use the finest simulations closest to observations as “truth” to determine the minimum cloud depth and the critical perturbation temperature in the presence of roll circulations; which are both required by the convective trigger function in the cumulus scheme. Several experiments using the vertical velocity-based trigger are tested in an attempt to improve convection over the observed cases. Results from the improved trigger function and its sensitivity to moisture in the boundary layer influencing convection will also be discussed.