Severe Weather Session 1 Abstracts (Friday)

1:45pm - 2:00 pm // Climatology of intense convection in Texas during past decades from satellite observations // Chuntao Liu // Texas A&M Corpus Christi
    About mid 1990s, a group of researchers and graduate students at Texas A&M started a project of counting storms on SSM/I passive microwave brightness temperature images. Computer programs were designed to automatically group adjacent cold brightness temperature pixels into precipitation features. For the first time, global precipitation was quantitatively described from the perspective of precipitation systems and their properties. Since then, this methodology has been improved and adapted to the TRMM, SSMIS, CloudSat, and GPM satellites by several generations of students and researchers. This presentation will review the development of precipitation feature databases during past two decades. The applications of these databases over Texas are shown with the climatology of the intense convection indicated with low passive microwave brightness temperatures, strong radar echo at high altitudes, and high lightning flash rates.

2:00pm - 2:15 pm // Building a Weather Ready Rio Grande Valley from the Ground Floor: “The Spanish Language Early Warning Severe Weather Notification System” // Maria Torres // NWS Brownsville
    The Rio Grande Valley has been growing rapidly in the last two decades. Hidalgo County, for example, is the eighth most populous county in Texas, and the Rio Grande Valley region (Starr, Hidalgo, Cameron, and Willacy) is the 5th most populous metropolitan region, behind Houston/Galveston, Dallas/Fort Worth, Austin, and San Antonio. More than 90 percent of the residents in the region are Hispanics, and Spanish is their native language.
    Prior to 2014, NOAA Weather Radio (NWR) served only English listeners with three transmitters providing hazardous weather notifications in Rio Grande City, Pharr, and Brownsville. Fewer than half of the residents in the most vulnerable communities in the Rio Grande Valley were unable to understand the messages since they either spoke or wrote English poorly or not at all. Several Spanish commercial radio stations described the arduous process required to manually translate warnings, which reduced valuable time for families to find safety.
    In 2014, the NWS in Brownsville launched the Spanish Language Early Alert/Warning System, whose centerpiece was two new Spanish-Language NWR transmitters located in Harlingen and Pharr, with signals strong enough to cover nearly 100 percent of the Spanish-language listening community. This presentation will describe the collaborative project that was implemented with funds from a Hazard Mitigation Grant and local government and private sector contributions, and supported by local, state, and federal governments and elected officials. Software upgrades aimed at improving voice quality, and planned marketing strategies to put NWR receivers in the hands of the most vulnerable Spanish-language listeners, and assistance to other interested NWS offices to bring more Spanish-language NWR to more communities will also be discussed.

2:15pm - 2:30 pm // Lessons Learned from Space Flight Accidents and What It Means for Integrated Warning Teams // Tim Oram // NWS Southern Region
    The post-accident reviews of the Space Shuttle Columbia (February 1st, 2003) and Atlas/Centaur 67 (March 26, 1987) accidents provided valuable lessons learned on communications and decision-making in a complex environment. At the core, these accidents were caused by failures in communication and teamwork. The lessons learned from these accidents drew upon knowledge of other non-space related disciplines and apply to any team involved in collaborated communications and decision-making such as our Integrated Warning Teams (IWTs). The primary three components of the IWT are the National Weather Service Weather Forecast Office (WFO), Emergency Management officials, and the broadcast media. Each of these components is itself a team. The IWT challenge is, therefore, complex with a distributed network of teams focused on the four elements of an effective Integrated Warning System for weather and its impacts: forecast, detection, dissemination, and response. 
    One concept to improve teamwork taught to NASA Flight Controllers after the Columbia accident that can be used to improve the effectiveness of IWTs is “Crew Resource Management” (CRM). CRM consists of five team skills: Communications, Situational Awareness, Leadership/Followership, Decision-making, and Debriefing. CRM provides IWTs a common framework to discuss their strengths and areas for improvement while the IWT provides the means through which the team members can work together to develop these skills. Some IWTs are already conducting debriefings after significant events to discuss what went well and what can be improved with regards to communications and situational awareness. In addition, a critical component of developing the CRM skills is the use of exercises and simulations. A few IWTs are also conducting exercises to provide additional training opportunities to improve teamwork.  The consistent use of these CRM practices by IWTs is recommended as a means to improve our teamwork and, ultimately, the effectiveness of IWTs in protecting lives and property.

2:30pm - 2:45 pm // Avoiding Information Overload: Effective Interpretation of Model Ensemble Output for Impact Weather Events // Mark Conder // NWS Lubbock
    Ensemble Forecast Systems represent a collection of atmospheric model forecasts obtained by varying the initial conditions or model parameters. They are used to account for imperfections in the models and the uncertainty inherent in our observations of the state of the atmosphere. As computing power, storage and display capabilities have increased, ensemble systems have become more widespread and complex. For example, the European Centre for Medium Range Weather Forecasts (ECMWF) ensemble consists of 51 separate model forecasts, or members. THE ECMWF ensemble is just one of an increasing number of ensembles available to the forecaster. Thus, these ensembles present a challenge to the forecaster as he or she simply does not have the time to examine each member thoroughly. To mitigate this problem, forecasters have typically relied on applying two statistics, the mean and the standard deviation, to forecast variables such as temperature and wind speed. The mean represents the average of all the model members, while the standard deviation measures the spread, or degree of uncertainty, in the model forecast. Recently, additional techniques for analyzing ensemble data have emerged; these include the use of plume diagrams, probabilities, and percentiles. 
    This presentation will demonstrate some of these new techniques using the Texas Tech University Ensemble Modeling System, and how these can be used to clearly convey the potential impacts of hazardous weather events.

2:45pm - 3:00 pm //Analysis of Sub-Domain Inconsistencies in Simulated Outflow Trajectories // Andrew Vande Guchte // Texas Tech University
    Analysis of modeled parcel trajectories has become commonplace in the study of mesoscale convective dynamics. While advances in computing has precipitated the development of cloud models capable of realistically simulating convection at high spatial and temporal resolution—and are thus capable of simulating parcel trajectories with a high degree of realism—the treatment of parcels beneath the lowest model level (referred to as the ‘sub-domain’) continues to be an unsolved problem for a number of reasons. First, the horizontal wind profile beneath the lowest model level cannot be determined by the prognostic equations of the model, and therefore must be parameterized analytically. Second, the profile of non-velocity scalar quantities (such as vorticity and potential temperature) are similarly unknown beneath the lowest model level, such that a convention to record these scalar values along sub-domain parcel trajectories must be specified. The problem with the treatment of parcels that traverse the model sub-domain is well documented, however the errors that occur along these trajectories have not been quantified, or even qualified. This study determines that no matter the treatment of the wind profile in the sub-domain, parcels that pass through it are immediately and irreversibly subject to inconsistencies with the rest of the dynamically-driven model. Furthermore, other scalar quantities that are calculated on the model grid are incorrectly assigned to parcels in the sub-domain. It is determined that the best way to avoid these problems is to reject analysis of any parcels that spend any amount of time in the sub-domain. This is accomplished by either increasing the model vertical resolution or simply removing parcels that traverse the sub-domain from the analysis completely.