Attendees: Matt Parker (NCSU), Justin Lane (GSP), Hunter Coleman (CAE), Frank Alsheimer (CHS), Jonathan Blaes (RAH), Brian Miretzky (ERH), Steve Nelson (FFC), Keith Sherburn (NCSU), Steve Zubrick (LWX)
I. Incoming NC State grad student Keith Sherburn introduced himself to the group. Keith is a recent graduate of OU and will be “officially” an NC State student in August. Jason Deese is the other student who will be participating in the project.
II. GSP and CAE unveiled the first iteration of the “Master” case study list. At this point, the list is simply a compilation of the event dates identified by each of the participating offices. After grouping together dates that were obvious “overlaps” (i.e., dates identified by multiple offices), 89 potential HSLC events have been identified so far. This number may change slightly as a few offices are still wrapping up the case identification process. GSP and CAE will continue to “fine tune” the list over the next month or so. Participants should forward any late additions to the case list to Justin.
Everyone agreed that this is a great accomplishment and that the collaboration among and effort put in by the participating WFOs has laid a solid foundation for what should be a great project. Matt asked if any epiphanies, interesting anecdotes, or “Eureka” moments had arisen from the case ID process. The short answer (among this call’s participants anyway) was, “not really.”
III. The majority of the call was devoted to “what now?” There are a multitude of paths and detours that the project can take from here, and it’s time to start determining these and what the role of the WFOs will be going forward.
Matt has identified four potential avenues of future research:
a. Case Studies- Regardless of the direction(s) we choose, comprehensive and thorough case studies will be a necessary first step in the process. What constitutes “comprehensive and thorough” is still up for some discussion, but detailed radar analyses as well as mesoanalysis are considered “musts.” Another possibility to consider is dual Doppler analyses of events that occur within TDWR domains.
The Case Study process is the area in which WFOs can play the biggest role going forward. If each WFO is able to tackle 2-3 events, that will result in approximately 20 case studies for the project. (It is still to be determined how many case studies would be “ideal.”) The first step in this process will be for each WFO to identify their 2-3 “highest priority” cases.
b. Discriminating “Hits” from “Nulls” and/or “Higher End” Events from “Marginal” Events. It was universally agreed among the call participants that this needs to be a focus of the project. There are a number of databases that can be accessed for this purpose, including SPC mesoanalyses (for analysis of atmospheric ingredients), SPC’s convective structures database, and WDSS statistics.
c. NWP/Predictability- recognition/analysis of events from a forecast modeling and/or ensemble perspective. There was general agreement that WFO recognition of HSLC environments in the 18+ hour “outlook” period has improved significantly in recent years, and that the main challenge lies in the ability to discriminate those events that will be marginal from those that will be relatively “high impact.” Therefore, this was generally considered the least critical of the four major categories.
d. Dynamics of HSLC Cases- storm scale modeling of case studies or “idealized” storms. This topic provoked the most debate. There was some thought expressed that the results of numerical simulation studies are often too theoretical and abstract and offer little in the way of operational significance. Others argued that such studies can be enlightening in terms of understanding the underlying storm-scale dynamics, which can have useful application to radar analysis. Matt feels a numerical simulation is worth investigating. However, he also noted that recent attempts to model tropical-cyclone “mini” supercells took approximately two years before a credible storm was successfully modeled (basically about the time that the grad student working on the project was ready to graduate). So, there is a cost:benefit aspect to consider to a storm-scale modeling study.
IV. Steve Nelson informed the group that QLCS warning services will be an upcoming focus topic for Southern Region SOOs. He will keep the CSTAR group informed of any significant activity in this arena.
V. Ongoing and near-future activities- The next conference call will be 28 July from 1000-1200 EDT. GSP and CAE will continue to “polish” the case list, and will also begin constructing a list of nulls. The goal is to complete these tasks by the time of the August conference call. Therefore, participating WFOs are asked to wrap up the case identification process by the time of the 28 July call. WFOs should also identify their 2-3 “high priority” cases by this time.
Bonus Material: Matt Parker’s SPC conference call notes.
I. It should be possible for us to use both the SPC mesoanalyses and the database of convective modes. The SPC mesoanalyses are archived 2D grids of surface fields and bulk parameters (CAPE, bulk shear, etc.). The full RUC sounding data are not archived by SPC, but we can get them from NOMADS if/when we need them. The benefit of the SPC RUC analyses, of course, is the improvements they add to the surface fields using real data. The convective modes database was created by Rich Thompson and Brian Smith, who documented convective mode and other details for every tornado report and every *sig* severe report. The reports have also already been assigned to the mesoanalysis environments, which means that we won’t need to process the grids for those reports; thus, it should make for a nice first-look dataset. It’s not clear what fraction of our HSLC cases would be covered by the “tornado or sig svr” criterion, but it’s at least a start. The thing I like about using these databases is that they will provide some uniformity of method across our entire region.
II. They were interested to know some of the nuts and bolts of “what is a null” and the like. Their input was largely in line with the conclusions we’ve made in our monthly HSLC teleconference calls.
III. I will be in China for a summer school through late July, so there may not be much action on data acquisition until August. But, the goal is to have things ready in mid-August so that Keith and Jason can come in and get their feet wet. In terms of ordering data from them, it would be helpful to be able to send them our completed database of events and nulls so that they can retrieve the subset of data that we need. I hope that late summer is a reasonable goal for including the nulls and getting the database into a polished form that we wouldn’t mind sharing with them.
IV. I asked a few questions about how *they* think the SPC mesoanalyses perform in HSLC scenarios. One thing they mentioned was interesting… they are somewhat skeptical of MLCAPE in these scenarios, and have a bit more faith in SBCAPE (although they suggest checking them both). Apparently, the MLCAPE can be strongly influenced by the RUC’s treatment of the boundary layer, and nobody really knows how good that is in these cool season low-topped events. Another related note is that they always assume a 100 mb deep mixed layer for their ML calcs, which may be too deep for these events. Their recommendation: if SBCAPE and MLCAPE are much different from one another, it’s important to look at the local RUC soundings and try to figure out what’s going on and whether the RUC ABL looks reasonable. Same would go for CIN and other related parameters.
V. Topics that we did not get to in the time allotted: How much will SPC be able to contribute to the analysis? What is the status of the recent request to add more contours to the SPC CAPE mesoanalyses? We can raise these issues when I next talk to them in late summer, once I’m back in Raleigh full time again.