Wednesday, 25 January 2012
Ensemble Perspective of the 27 April Tornado Outbreak From the Hazardous Weather Testbed Experimental Forecast Program
Hall E (New Orleans Convention Center )
The 2011 Hazardous Weather Testbed Experimental Forecast Program tested unique displays and guidance from the Center for Analysis and Prediction of Storms (CAPS) 50 member convection-allowing Storm Scale Ensemble Forecast (SSEF) system. Standard displays of CAPE, 0-6km bulk wind difference (SHR), CIN, and the product of CAPE and SHR (CASH) revealed that this outbreak was extreme when compared to both prior significant outbreaks (e.g., 3 May 1999) and the overall climatology of significant severe storm episodes. An object based algorithm was applied to hourly maximum updraft helicity (UH) to detect significant simulated storms with rotating updrafts, and the maximum (CAPE, SHR, CASH) and minimum (CIN) environmental parameters were extracted from these object tracks during the prior hour. In addition, observed tornado report locations were used to extract grid point values for the same variables in the ensemble. From 12 UTC 27 April to 12 UTC 28 April, the SSEF produced similar numbers of significant UH tracks as tornado reports (216 tornado reports while the membership averaged 217 UH tracks ranging from 155-312). CASH and SHR were exceptionally high in both samples with the upper 25 percent exceeding 90000 m2s-2 and 45 ms-1. The SSEF outbreak resembled the observed, but qualitatively lacked multiple simultaneous corridors of supercell storms. However, the new output field "convective activity" (any grid point where 35 dBz reflectivity is exceeded at -10C, essentially a mixed phase updraft) showed that the members were initiating new short-lived storms in the warm sector. These storms were often in close proximity to observed storms (with 1 to 2 hour lag) but suggest that the SSEF was capable of predicting the likelihood of convective initiation in the warm sector and that multiple corridors of supercells were possible. Several unique diagnostic displays (i.e. non-map based perspectives) are shown where the entire ensemble membership can be compared simultaneously. These can allow forecasters to detect outliers, representative members, and diagnose timing and longevity differences amongst the membership. In addition, an ensemble sounding tool was developed to visualize the profiles from members utilizing various planetary boundary layer and microphysics parameterization schemes. Visual and quantitative impacts of these SSEF physics variations will be presented.
Supplementary URL: