Blake Allen
Mentor: Ted Mansell

Micophyiscs Complexity Effects on Storm Evolution and Electrification

Microphysics parameterization schemes used in numerical storm models vary greatly in complexity. One of the ways in which these schemes vary is in the number of microphysical moments that they predict for each of the hydrometeor categories included in the scheme. This study analyzed the effects of enabling prediction of a second microphysical moment, number concentration, for each hydrometeor category in a mixed phase, bulk microphysics scheme. The addition of number concentration prediction for cloud droplets was found to have a large influence on the early development of the simulated storm, while the addition of number concentration prediction for rain was found to have the largest impact on the storm’s reflectivity structure. The electrification of the storm was also found to be quite sensitive to changes in the microphysics complexity, due at least in part to variations in cloud ice and graupel production in the different model runs.

While the forecasting ability of the Storm Prediction Center (SPC) has increased through the years, and the watches that are issued have become increasingly more accurate and precise, it is what people do with the information that is given to them that ultimately makes the difference. The best product that the SPC issues can become ineffective if the people who are supposed to use it are not aware of it or do not respond to it in an appropriate manner. For this study, students from both Northern Illinois University and the University of Oklahoma were surveyed in order to discover more about their knowledge of and response to the severe weather watches issued by the SPC. They were asked questions concerning what they knew about severe weather watches and how they learned of and responded to them. In addition, other questions regarding their demographics and some personal information, such as their age, gender, residence, and personal experience with severe weather were asked. The data that was collected through the survey is in no way complete or meant to give a complete picture of how people make decisions regarding severe weather watches. However, it will hopefully provide some insight that can be built upon in future studies.

Timothy Bonin
Mentor: Deke Arndt

Are Freezing Rain Patterns in the South Central United States Changing?

Daily surface climatological reports in conjunction with radiosonde data was analyzed from a period of 1950-2007 for five stations in the South Central United States. From this data, the annual number of possible snow, rain, and icy precipitation events was determined by analyzing characteristics of the troposphere for two mandatory radiosonde launches for each wintertime precipitation day. The number of “potentially significant freezing rain events” was also determined for each location. It was determined that there is an “ice belt” in which icy precipitation, which includes sleet, graupel, and freezing rain, is more likely to fall. This ice belt has moved northwest with time. The occurrence of potentially significant freezing rain events is not associated with this ice belt. Instead, the frequency of these events has its own trend and has been generally increasing, principally from the 1980s onward. However, there was a drastic decline in the number of these events in the early 2000s. This result is counterintuitive due to the fact that there have been several very damaging ice storms in the early and mid 2000s. The correlation between icy precipitation events/freezing rain and various teleconnections is also evaluated in this study.

The fall and winter months of 2007-2008 were particularly active in terms of tornado events and the number of tornado-related fatalities in the
United States. The media portrayed the severe weather during this period in time as unusual, if not record-breaking. Journalists and reporters asked questions about the causes of such an active cool season and how this season compared with other years.

Defining “cool season” as the 6 month period from October to March, this study analyzes the 2007-2008 cool season tornado outbreaks in terms of tornado frequency, number of fatalities, and tornado strength based on Enhanced Fujita scale ratings. Data from the Storm Prediction Center’s Storm Event Database was utilized to determine the uniqueness of the 2007-2008 cool season tornado outbreaks as compared to the last 50 cool seasons. Our results will more accurately define the significance of this season with respect to the past 50 seasons in order to aid the media and other interested parties in quantifying this period of time. Trends in cool season tornado outbreaks are also investigated.

Many of the techniques that have been developed for lake-effect snow forecasting have been designed for regions where lake-effect snow is common, such as western Michigan and upstate New York. In this paper forecasting parameters developed by the NWS forecast offices in Buffalo and Detroit are applied to lake-effect snow cases on the west side of Lake Michigan to see if the parameters accurately depict conditions that are favorable for lake-effect snow development. North American Mesoscale (NAM) and Rapid Update Cycle (RUC) model data at two points near Chicago and Milwaukee are used in the evaluation. Northeast and north- northeast 850 mb and 925 mb winds are found to be common to lake-effect snow events in this region. In addition, the minimum -13°C temperature difference between 850 mb and the lake surface is present during most of the lake-effect snow cases. Low directional wind shear between the surface and 850 mb is also present, but is not an absolute requirement for lake- effect snow to occur in this region.

Christina Holt
Mentor: Kevin Kloesel

Radar Characteristics of Tornado Producing Mini-Supercell in Tropical Storm Erin (2007)

During the 2007 tropical season, Tropical Storm Erin re-strengthened over Oklahoma after making landfall along the coast of Texas. This tropical storm produced a total of seven tornadoes in Texas and Oklahoma. Its first tornado in Oklahoma was within the range of the Multifunction Phased Array Radar (MPAR) in Norman, OK. Using a scanning strategy that updates every 43 seconds with 14 elevation angles, the MPAR allowed us to measure the physical characteristics of a tornado producing mini-supercell in a tropical cyclone environment. The physical aspects of this mini-supercell include a shallow circulation only 4.5 km in diameter extending through a depth of 3 km. The reflectivity signatures were more subtle than typical Great Plains supercells and maximum reflectivity values of 50-54 dBZ extend over a small area. Based on these criteria, this cell is consistent with previous studies. The most noticeable difference form recorded events is the sampling of this storm at such a high temporal resolution. A rapid intensification to tornadic over a three-minute period exemplifies the need for up-to-the-minute radar data. Using better sampling strategies and higher resolution will lead to an increased understanding of the hurricane-spawned tornado and improved forecast and warning accuracy.

Kelly Keene
Mentors: Paul Schlatter, Jack Hales and Harold Brooks

Evaluation of NWS Watch and Warning Performance Related to Tornadic Events

Two organizations within the National Oceanic and Atmospheric Administration’s (NOAA) National Weather Service (NWS) are responsible for disseminating critical information to their customers prior to and during severe weather. The Storm Prediction Center (SPC) focuses on
conducting an overall assessment of the environmental conditions on the national scale, and issues the appropriate severe weather watches. At the local level, Weather Forecast Offices (WFO) are responsible for issuing warnings to cover imminent severe weather threats. The focus of this research is primarily on the evaluation of the NWS watch and warning performance in relation to tornadic events from 1997 to 2007. Data are obtained from the NOAA Performance Management Website and the SPC watch database. The watch and warning records are matched to each tornadic event, allowing an evaluation of tornado warning performance in relation to watch type and lead time. Statistical analysis of the data reveal interesting results on the nature of the relationship between tornado warnings and whether a watch was in effect, and if so, what type. Tornado warning performance and lead time increases with increasing F-scale intensity and also when there is a tornado watch in effect. The lowest tornado warning statistics occur when no watch is in place. Having a tornado watch in effect greatly increases the Probability of Detection (POD), while slightly decreasing the False Alarm Rate (FAR). For all tornado warnings from 1997-2007, there is an increasing of POD, while maintaining a steady FAR. Finally, the entire tornado warning data set is visualized using a new technique (Paul Roebber, 2008), and also added to a previous study (Brooks 2004).

Jennifer Newman
Mentors: Daphne LaDue, Pam Heinselman

Identifiying Critical Strengths and Limitations of Current Radar Systems

The Next-Generation Weather Radar (NEXRAD) network of Weather Surveillance Radar-1988 Doppler (WSR-88D) radars is nearing the end of its expected 20-year engineering design lifetime. One replacement system currently under consideration is multifunction phased array radar (MPAR). The purpose of this study is to illustrate the critical capabilities of current radar systems to decision-makers and to determine the suitability of MPAR to users’ needs. Interviews were conducted with National Weather Service (NWS) forecasters and broadcast meteorologists to collect stories which exemplify radar strengths and limitations.

The roles served by participants strongly affected their use of radar. NWS forecasters use radar to provide information to a variety of user groups across their forecasting area. During severe weather events, they use volumetric data to monitor storm evolution and make warning decisions, and also to evaluate rapid low-level scans to warn for small-scale signatures. Broadcast meteorologists use radar to anticipate and interpret warnings issued by NWS, make decisions about cutting into regular programming to provide severe weather coverage, and illustrate the current threat to viewers. Broadcast participant use their own radars to obtain rapidly updating low-level reflectivity data so that they can pinpoint the time and location of NWS warnings.

Limitations identified in the study include radar horizon issues, de-aliasing problems, beam spreading, and detection of precipitation type. PAR capabilities will help to mitigate the effects of many of these limitations.

Jonathan Poterjoy
Mentors: Ross Hoffman and Mark Leidner

Estimating Correlations from a Coastal Ocean Model for Localizing an Ensemble Transform Kalman Filter

Data assimilation is the process of using past and present data to estimate the current synoptic state of a dynamical system. Current data is merged with a previous model forecast or “background” field to produce the best estimate of a system’s state called an “analysis”. For cases where the probability distribution of observation and background errors are normally distributed, a Kalman filter can be shown to produce the best estimate of a variable and its uncertainty. A type of data assimilation system called the Ensemble Kalman Filter (EnKF) approximates the background covariance field using only a small ensemble of forecasts. Since a limited number of samples are used, many spurious correlations exist between an observation at one point and forecast errors at various locations within the model domain. To limit spurious relationships the Local Ensemble Transform Kalman filter (LETKF) limits the region considered in a process called “localization”. But a question arises regarding the optimal localization size for analyses within complex model domains. Using the Estuarine Coastal Ocean Model (ECOM) coupled with the LETKF, we examined correlations between simulated state variables on various locations and depths within a domain that spans the New York Harbor region. Distributions of correlation coefficients surrounding an analysis point were used to determine the optimal localization domain for each particular relationship. Since spurious correlations tend to diminish after 1 to 2 days of simulation, results observed during days 3 and 4 of this experiment were taken to be a good estimate of true relationships between variables. Given the large amount of dynamical and bathymetric variability within this model domain, correlation structures of mixed shapes and sizes were observed. In many instances, the parameterized localization domain was either too small or too large to capture the actual correlations. Results from this study provide incentive to pursue an automated solution to optimal localization within the LETKF/ECOM that tailors a unique localization volume for each analysis. If successful this solution can be applied to various other prediction systems that rely on ensemble data assimilation.

Christopher Wilson
Mentors: Kiel Ortega and Valliappa Lakshmanan ("Lak")

Evaluating Multi-Radar, Multi-Sensor Hail Diagnosis with High Resolution Hail Reports

Low resolution verification data, as available from the Storm Data database, has hindered the development and evaluation of high resolution hail algorithms as well as the assessment of hail forecasting techniques. Previous studies have highlighted the inadequacies and inaccuracies associated with this verification data. This study uses high resolution ground-truth hail verification data from the Severe Hazards Analysis and
Verification Experiment (SHAVE) to evaluate gridded synthetic hail verification and different radar derived parameters used in predicting severe hail.

MESH is found to have limited skill as a synthetic verification tool due to a high probability of false detection and a wide distribution of MESH values for each reported hail size range. In addition, radar-derived parameters are found to provide little skill in the prediction of severe hail as the probability of false detection associated with these parameters leads to low skill scores. The predictive skill of these parameters is also found to decrease with time, limiting the lead time in which sksurface hail fall is possible using radar derived parameters.

Jeffrey Zuczek
Mentor: Mark Shafer

Applying Mesonet Wind Climatology to Oklahoma Prescribed Burns

Since 2001, twelve prescribed burning associations have been enacted in the state of Oklahoma. These burn associations perform a variety of tasks, such as the control of invasive plant species. The Eastern Red Cedar tree is especially notorious in Oklahoma for breaking up pastures and wildlife habitats. Prescribed burning is also a major tool used by farmers for crop preparation, via the controlling of invasive weeds. The benefits achieved from prescribed burning in Oklahoma are vast.

A burn must be carried out in a safe and predictable manner in order to reap its environmental benefits. Understanding prevailing weather conditions is a must when it comes to burning. This study specifically focuses on winds that exist during the time of burn. Wind climatology, via Oklahoma Mesonet data that extends from January of 1994 through May of 2008, is used to predict the likelihood of having several consecutive favorable burn days. A three-pronged criterion was developed to determine the constraints of a “favorable burn day” that is based primarily on information gathered from officials of the individual burn associations. A resulting burn calendar shows both daily and monthly trends of favorable burn days for February, March, and April. These specific months are desired by the burn associations for a variety of reasons, such as burning before native birds begin to nest and when relative humidities are still low. The daily burn calendars present a weak downward trend in the data. This trend suggests that of the three months considered in this study, February is the most favorable month to conduct prescribed burns. Monthly burn calendars, however, show a more pronounced downward trend. They present clear evidence that the frequency of favorable burn days declines from February through April. These results suggest that from a purely climatological perspective, it is wise to conduct burns earlier, rather than later.