A Method for Indentifying Convective Initiation in Forecasts and Observations

A threshold-based, object-oriented CI detection and tracking algorithm was developed and the results of its application on CIWS and HRRR products were presented. A rule set for defining and tracking of CI was developed using the CIWS analysis product with high temporal (5 min) and spatial (1 km) resolution, providing a refined approach to the understanding of the complex CI phenomenon. The definition and detection of CI is achieved using VIL thresholds (instead of radar reflectivity, as was used in previous studies) within an aviation framework, where CI is defined by and tracked with respect to proximity to existing hazards (i.e., thunderstorms).

The results show that the algorithm is applicable to both CIWS analysis and HRRR model output even with the presence of complex weather systems. It has been demonstrated that the developed CI detection and tracking technique can be applied to a forecast product and produce sensible results for comparison to observations. As was seen with the need for adjustment of the tracking cutoff threshold for the HRRR, the effectiveness of the technique depends upon the appropriate choice of proximity and tracking cutoff thresholds, which are, in turn, dependent on the temporal and spatial characteristics of the product to which they are applied. Therefore, some analysis will be necessary to determine the appropriate thresholds for application to specific convective forecast products.