An Investigation into the Short-Range Predictability of Convection Initiation: Model Verification and Case Study Analyses

Improvements in numerical forecasts of deep, moist convection have been notable in recent years and are in large part due to increased computational power and improved physical process parameterizations. Accurately forecasting the timing and location of convection initiation (CI), however, remains a substantial forecast challenge given the highly non-linear nature of the atmospheric processes that influence CI and the limited amount of data available on the scales that exert the greatest local control on its occurrence.

In this work, numerical simulations conducted using the Advanced Research Weather Research and Forecasting (WRF-ARW) numerical model, version 3.3.1, are used to investigate the short-range (0-24 h) predictability of CI. Such simulations utilize a horizontal grid spacing of 429 meters, enabling for the representation of meso-gamma-scale to microscale phenomena thought to influence CI. A set of thirty CI events occurring between May-August 2010 across the central High Plains, including several 'null' cases in which CI was forecast to occur but did not do so, was randomly obtained from the set of all days between May-August 2010. This region was chosen for study given the wide array of convection triggering mechanisms, both orographic and atmospheric in nature, observed there during what was an anomalously active convective period.

Numerical forecasts of CI are verified utilizing a spatiotemporal object-based verification method. This verification is used to assess modes of forecast success and forecast failure as stratified by geographic location, CI triggering phenomenon, time of day, and time of season. Selected cases with relatively low and relatively high forecast skill will be presented with emphases on assessing the evolution of thermodynamic and kinematic structures associated with the simulated CI events and comprehensively evaluating the numerical modeling system's ability to accurately forecast CI over a diverse range of events.