The April 2015 high-shear, low-CAPE (HSLC) CSTAR conference call featured a guest presentation from Amos Dodson, forecaster from the NWS Northern Indiana (IWX) office. Amos recently presented work at the Central Iowa NWA’s Severe Storms and Doppler Radar Conference investigating HSLC events across the IWX CWA, including an evaluation of SHERBS3 performance therein, and he was gracious enough to share his work with the CSTAR group this past Tuesday. We are excited to share some highlights from his presentation as it shows the SHERBS3 performing well and it provides a great example of the CSTAR HSLC research being evaluated and utilized away from our Mid-Atlantic cluster.
Figure 1. List of recent HSLC tornado outbreaks within IWX and pie chart showing the percentage of IWX tornadoes associated with a given storm type over the 30-year period 1980-2010.
Within the IWX CWA, HSLC environments have accounted for half of the total tornadoes since 2010, including three recent notable outbreaks (11/17/13, 4/19/11, and 10/26/10; see Figure 1). To address this considerable forecasting concern, Amos examined the performance of the SHERBS3 parameter developed through recent CSTAR initiatives at discriminating between hits (i.e., days in which more than one severe report occurred in IWX) and nulls (days when zero or one severe reports occurred in IWX or SVR and TOR warning FAR in IWX was 100%) from 2005-2014. In addition, Amos created composite charts of various fields to determine differences between the synoptic setups associated with events and nulls.
Figure 2. Composite charts showing the mean 300 hPa wind speeds and vectors (m/s) for IWX HSLC hits and nulls.
Figure 3. Composite charts showing the mean 700 hPa wind speeds and vectors (m/s) for IWX HSLC hits and nulls.
Synoptically, hits were associated with a coupled jet feature aloft (Figure 2), a deeper 500-hPa trough, a stronger 700-hPa and 850-hPa jet (Figure 3), and slightly more negative LIs. This corroborates recent work within the CSTAR group suggesting that significantly severe HSLC events are associated with stronger forcing than those limited to non-severe convection. Parameter-wise, the SHERBS3 performed well in discriminating between hits and nulls within both datasets, as shown in Figures 4 and 5, though there were one or two puzzling misses and false alarms. Despite these exceptions and the limited dataset, the SHERBS3 showed promise as a guidance tool across IWX during HSLC environments, particularly when used in tandem with other techniques.
Figure 4. SHERBS3 distribution for hits (red) and nulls in the first null dataset (blue). The y-axis shows the number of events in a given SHERBS3 bin.
We would like to thank Amos again for sharing his work with the CSTAR group, and we look forward to additional dialogue in the future focused on similar projects across our collaborating CWAs!
The entire set of powerpoint slides can be viewed here: Dodson High Shear Low CAPE