The Convective Storms Initiation Project (CSIP), whose field phase took place in summer 2005 was an attempt to gather observations relevant to the initiation of convection in the southern UK. Here we describe a study of Intensive Observation Period (IOP) 1 (15th June 2005) which was a case of showers developing in a SW flow over southern England behind a front. Only one of the showers developed into a significant thunderstorm. This case was chosen because the with the formation of only one storm, at first sight at least, the case looked relatively simple and tractable to analysis. The initiation and development of the storm was well forecast by a rerun of the 1.5km gridlength version of the Met Office Unified Model (MetUM).
An earlier observational study, Morcrette et al (MWR 2007) looked at the deepening phase of the storm. This study identified the key features of this case to be a tongue of low level theta-w air associated with the front with overrunning lower theta-w air above, a convergence line and a "lid" of high static stability air at about 700hPa which the shower was initially constrained below but later broke through.
In this paper we look at the initiation of the shower in question which can be traced back to a region of high ground (Dartmoor) at around 07 UTC. We investigate the development in more detail using model sensitivity studies with the MetUM. We establish that the convergence line was initially (07 UTC) caused by roughness effects but had a significant thermal component later. The convergence line at 07 UTC was essential for the initiation of the shower which later developed into the storm.
The presence of Dartmoor also had a key role in the development of the thunderstorm. However the mechanism by which orography encouraged convective initiation was not one of the more usual mechanisms of lifting or elevated heating. A period of asymmetric flow over the high ground due to an area of high static stability air associated with the front lead to a hole in the cloud downstream. The surface heating through this hole, in combination with the tongue of low level high theta-w air behind the front, caused the shower to initiate with sufficient lifting to enable it to later break through the lid.
The unexpected role of Dartmoor in the formation of this storm, along with all the other factors, shows that even in this relatively simple case the initiation and development of the storm was surprisingly complicated yet it was predictable in the high resolution model.