One of the biggest forecasting challenges for Great Plains forecasters is correct prediction of convective initiation, especially when it is elevated at night. This study examines a subset of nocturnal elevated convective initiation events that occur without direct influence from surface boundaries or pre-existing convection. 287 of these events were examined over a four-month period during the summer of 2015 (May, June, July, and August). Events were first classified into one of four types based on apparent formation mechanisms and location relative to any low-level jet. A climatology of each type was performed, noting general spatial tendencies over a large Great Plains domain and initiation timing trends. Additionally, analysis of initiation elevation was performed. Five convection allowing models available during the Plains Elevated Convection At Night (PECAN) field campaign were analyzed for location and timing errors. Additionally, four versions of a 4km WRF model, each with a different planetary boundary layer (PBL) scheme were run for a few representative cases to explore predictability of these types of convective initiation.

The climatology revealed a dual-peak pattern for when initiation occurred, with one peak near 0400 UTC and another near 0700 UTC. The times that the two peaks occurred along with the prominence of the individual peaks shifted depending on the region of the domain being analyzed. Subtle patterns in location were identified, as well as elevation tendencies for the initiation. Model deficiencies tended to be more with location rather than timing for the 5 PECAN models, with the four 4km WRF models showing similar location issues plus issues with initiating convection lower than what actually happened.