Investigation of the predictability and physical mechanisms of an extreme-rainfall-producing mesoscale convective system along the Meiyu front in East China: An ensemble approach

Forecast uncertainties and physical mechanisms of a quasi-linear extreme-rain-producing mesoscale convective system (MCS) along the Meiyu front in East China, during the midnight-to-morning hours on 8 July 2007, are studied using ensembles of 24 h convection-permitting simulations with a nested grid spacing of 1.11 km. The simulations reveal a strong sensitivity to uncertainties in the initial state despite the synoptic environment being favorable for heavy rainfall production. Linear changes of a less skillful member's initial state toward that of a skillful member lead to a monotonic improvement in the precipitation simulation, with the most significant contribution arising from changes in the moisture field. Sensitivity to physics parameterizations representing subgrid-scale processes fail to account for the larger simulation errors (missing the MCS) with the physics variation examined but could result in a large spread in the location and amount of accumulative rainfall. A robust feature of the best-performing members that reasonably simulate the MCS-associated heavy rainfall is the presence of a cold dome ahead of the Meiyu front generated by previous convection. The cold dome promotes nocturnal convective initiation by lifting high equivalent potential temperature air in the southwesterly flow to its level of free convection. The skillful members reproduce the convective backbuilding and echo-band training processes that are observed during this event and many other heavy rainfall events over China. In contrast, the less skillful members that miss the development of the MCS either do not simulate the previous convection or produce a cold dome that is too shallow to initiate the MCS.