Nature vs Nurture: Understanding the Role of the Driving Ensemble on Under-Dispersive Convective-Scale Precipitation Forecasts

Abstract: Convective-scale ensembles are routinely used in operational centres around the world to produce probabilistic precipitation forecasts, but the added value provided by these models is limited by a lack of spread between members. Currently, it is unclear how much the behaviour of the driving ensemble determines spread towards the convective scale since very few studies have been conducted which compare the evolution of spread between ensembles of differing resolutions. This work focuses on understanding the correlations in spread between a nested convective-scale ensemble and its driving ensemble over the UK and will help researchers understand the extent to which spread characteristics are determined by the specific regional model configuration vs the driving ensemble. 

We have found that correlations are strongest in the first 24 hours of integration, with the nested ensemble typically displaying larger overall spread than the driving ensemble during this time. Further work is being conducted to study the cause of these correlations - for example, is the spread similarity caused by a strong resemblance between the corresponding nested and driving members (i.e., nature over nurture) or has each member evolved the initial state in its own way and the similarity is due to the environment (nurture over nature). It will also be necessary to understand which conditions favour stronger correlations over others, and whether there is leadtime dependence to these correlations. 

Biography: Adam is a third year PhD student whose work focuses on exploiting the operational benefits provided by high-resolution ensembles. While these models offer superior representation of convective events compared to coarser models, their usefulness in probing the uncertainty of these events is limited by a frequent lack of spread between members. Adam’s work is supported by the Met Office Research to Operations team which has identified this spread problem as one of the major issues affecting the guidance produced by these models.