Modeling the Coupling between Large-Scale Dynamics and Convection in the 2010 Pakistan Flood

The coupling between large-scale dynamics and convection in the 2010 Pakistan flood is examined with a cloud-resolving model (CRM) in the Column Quasi-Geostrophic (CQG) framework. In this framework, the CRM is coupled with large-scale dynamics through the Quasi-Geostrophic (QG) Omega equation and is driven by corresponding large-scale variations, thus allowing us to separate the total precipitation into components due to large-scale triggering and due to convective heating feedback. With the 2010 events being successfully reproduced, numerical experiments show that the first event is triggered by the upper-level potential vorticity (PV) advection and orographic lifting and the second event is mainly triggered by orographic lifting. In both events, the triggered convective heating has a greater contribution to omega than the large-scale forcing, and this feedback is essential in the extreme events. Besides, the environmental humidity has strong effects on the precipitation amount of these events. Comparing to the CRM results, representing convection by a convective parameterization or effective static stability leads to substantial discrepancies with observation. The convective parameterization underestimates the triggering effects due to the upper-level PV advection and is subject to a much weaker sensitivity on environmental humidity. One may tune the effective static stability to reproduce the triggering effects due to the upper-level PV advection. However, using effective static stability can not appropriately capture the triggering effects due to orographic lifting and the dependences of precipitation on the environmental humidity.