Mesoscale Convective Systems: PRobabilistic forecasting and upscale IMpacts in the grey zonE (MCS:PRIME)

Lead Research Organisation: University of Oxford
Department Name: Oxford Physics


A Mesoscale Convective System (MCS) is an organisation of many convective thunderstorms, each a few km in scale, into a coherent entity on scales of hundreds of km. We use the term to encompass a range of organised convective phenomena, including squall lines, supercells, and mesoscale convective complexes.

MCS sit at the intersection between weather and climate. On weather timescales, these long-lived systems produce extreme precipitation and flash flooding. Through their coupling to the large-scale circulation, they play a key role in climate phenomena including the Madden Julian Oscillation (MJO), the Intertropical Convergence Zone (ITCZ), and the Monsoons. The dynamical coupling is two-way: large-scale environmental conditions dictate the likelihood of convective organisation occurring, while in turn the MCS strongly feedback on the dynamics and thermodynamics of the environment.

Global numerical weather prediction (NWP) models, with grids of 15-20 km, and climate models, with grids of 50-100 km, cannot represent MCS. Our models operate in the "grey zone" where the phenomenon occurs at scales similar to the grid scale. This means that MCS are not fully resolved, but cannot be parametrised using conventional approaches, which assume that the unresolved process occurs on scales much smaller than the grid scale. Biases in the representation of the MJO, Asian Monsoons and ITCZ, as well as too few strong precipitation events, have been linked to deficiencies in the representation of MCS in models. Furthermore, "forecast busts" over the UK, for which the five- to six-day lead time forecast skill drops to around zero across the world's leading NWP centres, have been linked to a poor representation of MCS upstream over North America. We must improve the representation of MCS in weather and climate models.

This project addresses the representation of MCS in the grey zone in a comprehensive and coordinated manner. We will first combine a new global satellite-derived database of MCS with analysis products to assess the predictability of MCS formation and evolution conditioned on the large scales, taking a novel, probabilistic approach. Secondly, several theoretical frameworks have recently been developed which describe the dynamical impact of MCS back onto the large scales. We will critically assess these frameworks, making innovative use of analysis increments from within the data assimilation cycle, to measure the upscale impacts of MCS that are missing from current models.

We will use the fundamental understanding gained to develop a new parametrisation of the dynamical coupling between MCS and the larger scales. We will couple our approach to the new CoMorph convection scheme, which is undergoing trials for operational implementation in the UK Met Office's model. While CoMorph shows substantial improvements in initiating organisation, coupling of MCS to the large scales remains a problem. The representation we develop will be stochastic: we will represent the probability of different MCS tendencies conditioned on the resolved scale flow. Stochastic schemes are well suited to the grey zone, where parametrised motions are poorly constrained by the grid-scale variables, and so are very uncertain.

Evaluating the new parametrisation will critically test the knowledge gained throughout the project. Having validated our knowledge, we will use the scheme to measure the importance of the dynamical impacts of MCS on climate phenomena including the ITCZ and the MJO.

This project will produce a new understanding of the dynamics of MCS formation and upscale impacts. Through close collaboration with the Met Office, we intend to translate this into improved probabilistic forecasts for the UK and wider world. Only with reliable probabilistic forecasts can industry, policy makers, and the humanitarian sector quantify the risks of natural hazards, and act appropriately to protect against those hazards.


10 25 50