Towards Defining the Elevated Convection Problem

Operational meteorologists have provided feedback that most forecasts busts in the warm season are associated with elevated convection. These busts are from convection-allowing models and do not appear as prevalent in convection-parametrizing models, so why is this the case? To move towards a definition of the model deficiencies associated with elevated convection we first need to understand what is meant by “there is a problem with elevated convection”. Thus, a critical first step towards defining the “elevated convection problem” is to understand the perceptions of operational meteorologists. Using convective cases (identified, scored, and classified by operational meteorologists) that are elevated at some point within their lifecycle from 2017-2020 over the UK, and focusing on the events with weather warnings associated with them, will enable the perceptions to be understood. We show that there is a preference for UK elevated convection to form from continental plumes and in moderate large-scale forcing situations. However, we note that although continental plumes are the most common the frequency of the bust cases, when considering combined scores, is indistinguishable from plumes from the Atlantic. Splitting the scores into components shows that location, upscaling, timing and intensity (in that order) are the main reasons for classifying events as forecast busts and that location and timing errors tend to happen more frequently for Continental plumes compared to Atlantic plumes. Given the association with continental plumes the timing and positioning errors indicate that the land-sea-land interaction could be crucial in understanding the representation of elevated convection in the UK. However, we also need to be able to objectively identify elevated convection in the convection-allowing models. Thus we also present a comparison of four diagnostics and show that combining diagnostics related to CAPE ratio and the effective inflow layer properties could provide a beneficial forecasting tool that is spatially and temporally smooth whilst indicating environments that will likely produce convection instead. Understanding the problem from the perspective of an operational meteorologist, and having objective diagnostics, will enable targeted experiments to occur and will enable the “elevated convection problem” to be defined and will eventually lead to improvements to the operational model that could be seen on the bench.