On the Short- to Medium-Range Predictability of Thunderstorm Formation

Thunderstorm formation, hereafter referred to as convection initiation, is a classic scale interaction problem requiring the favorable superposition of the micro-, meso-, and synoptic-scales. Each scale provides a set of necessary but insufficient conditions for convection initiation. To first order, the synoptic- and meso-alpha-scales establish the large-scale thermodynamic and kinematic environment in which convection initiation may occur; the meso-beta-scale modulates horizontal variability within the large-scale environment; and the meso-gamma- and microscales control local boundary layer lifting, moistening, and environmental variability crucial to determining the occurrence, timing, and location of convection initiation. Convection initiation is highly sensitive to the atmospheric state, particularly within the boundary layer, in which it occurs. Indeed, previous studies have shown that modifications to boundary layer temperature and moisture content of magnitudes below those of observational uncertainty and microscale variability can influence convection initiation's occurrence. Thus, given controls upon convection initiation that range from the inherently more-predictable synoptic-scale down to the inherently less-predictable microscale, we pose the question: to what extent is convection initiation predictable?

In this presentation, a review of contemporary studies conducted into the short- to medium-range predictability of convection initiation, emphasizing initiation within warm-season, continental mid-latitude environments, will be presented. Current understanding regarding the extent to which convection initiation is believed to be predictable, and factors believed to limits its predictability, will be presented. Improvements in forecast skill and practical predictability fostered by convection-permitting numerical forecasts for convection initiation occurrence, timing, and location will be demonstrated. The control exerted upon the predictability of convection initiation by synoptic-scale forcing, topographical features, and planetary boundary layer parameterization will be discussed. The influence of assimilating targeted synoptic- to meso-alpha-scale observations in the upstream, pre-convective environment upon the predictability of convection initiation will be illustrated. Characteristics of particularly skillful forecasts of convection initiation occurrence, timing, and location will be presented and contrasted against forecasts lacking meaningful skill.

Despite improved understanding of the predictability of convection initiation, there remains much opportunity for further advances to be made. The presentation will close with a brief discussion of several such opportunities.