Entraining CAPE for Better Assessment of Tornado Outbreak Potential in the Warm Sector of Extratropical Cyclones

Convective available potential energy (CAPE) is known to lack skill in distinguishing the environment of tornadic and nontornadic storms, or that for tornado outbreaks and nonoutbreaks. Recently, Sueki and Niino (2016, GRL) showed that entraining CAPE (E-CAPE; Molinari et al., 2012) which considers the effect of entrainment of environmental air has much better skill in distinguishing tornadic and nontornadic typhoons. Motivated by their study, we examine whether E-CAPE can also be useful for discriminating extratropical cyclones that spawn 15 or more tornadoes (outbreak cyclones: OCs) and that spawn 5 or fewer tornadoes (non-outbreak cyclones: NOCs).

The results of composite analyses in which the centers of OCs and NOCs at the time when the largest number of tornadoes occurred during 24 hours are simply superposed and averaged, show that E-CAPE is indeed useful in distinguishing the environments of OCs and NOCs: E-CAPE in the warm sector of OCs is larger than that for NOCs with statistical significance, where the entrainment rate of 20 percent per 1km is found to give the best result. Moreover, the regions with large E-CAPE for both OCs and NOCs explain the locations of tornadoes much better than those with large CAPE. The larger E-CAPE near the center in the warm sector of OCs is due to greater moisture at low- and mid-levels, which are caused by advection by stronger southerly and by cyclone-scale updrafts, respectively.

The composite analysis also shows that E-EHI, E-SCP and E-STP, in which traditional CAPE used in energy helicity index (EFI), supercell composite parameter (SCP), and significant tornado parameter (STP) is substituted by E-CAPE, are more useful for estimating tornado potential than original EHI, SCP, and STP, respectively.