15^th Annual Mini-Technical Conference

SC DHEC Central Office, Columbia, SC

Friday, April 3^rd, 2009

Click Title to View Presentation

Abstracts for Each Presentation Below Agenda

Time		Speaker, Affiliation
9:30am	Welcome & Opening Comments	David Werth, 2008 PAMS President
	SESSION 1 – (Moderator: David Werth, 2008 PAMS President)
9:35	The Application of an Evolutionary Algorithm to the Optimization of a Mesoscale Meteorological Model	David Werth, SRNL
10:00	Assimilation of Doppler Radar Data into High Resolution Numerical Model Forecasts	Steven Chiswell, SRNL
10:15	An Analytical Screening Technique to Estimate the Effect of Cooling Ponds on Meteorological Measurements: A Case Study	Carl Mazzola, Shaw Environmental
10:30	Trends in Estimated Mixing Depth Daily Maximums	Robert Buckley, SRNL
10:55	BREAK (15 min.)
	SESSION 2 (Moderator: Erik Kabela, 2008 PAMS Vice President)
11:10	A Tornadic Mini-Supercell with No Cloud-To-Ground Lightning	Patrick D. Moore, GSP NWS
11:35	A Case Study of the 15 March 2008 South Carolina Supercell Outbreak: Pre-Storm Environment	Michael Cammarata, CAE NWS
11:50	A Case Study of the 15 March 2008 South Carolina Supercell Outbreak: Radar Analysis	Steven Naglic, CAE NWS
12:05	A Case Study of the 11 June 2008 Severe Pulse Thunderstorm Over Newberry County, South Carolina	Michael Cammarata, CAE NWS
12:20	LUNCH (Subway, in house, 40 min.)
1:00	PAMS Business	David Werth, 2008 PAMS President Erik Kabela, 2008 PAMS Vice Pres.
	SESSION 3 (Moderator: David Werth, 2008 PAMS President)
1:10	An Examiniation of a Non-Convective High Wind Event on 24 October 2008 Over South Carolina	Daniel Miller, CAE NWS
1:25	Ozone Concentrations Over Columbia, SC and Charlotte, NC	Casey Zuzak, USC
1:40	Extratropical Transitions of 19^th Century Tropical Cyclones	Matthew Rodgers, USC
1:55	Operational Utility of Atmospheric Favorability for Severe Storms (AFS) Gridded Data	Jeffrey P. Taylor, GSP NWS
2:10	Synoptic-Scale Conditions Leading to Flooding in South Carolina: A Case Study of October 22-23, 1990	Erik Kabela, SRNL/USC
2:25	ADJOURN

SESSION ONE

The Application of an Evolutionary Algorithm to the Optimization of a Mesoscale Meteorological Model

David Werth

Savannah River National Laboratory

Aiken, SC

Improvement in mesoscale atmospheric model simulations is often thought to be primarily a matter of finer spatial resolution. While this is generally true, there is a limit to the improvement one can obtain by simply decreasing the grid size of a numerical model. Further improvements in forecasts may be achieved with better model parameterizations, but this leaves the mesoscale modeler with the task of determining which parameterizations to use for a specific problem and what values to use for individual model parameters. The accuracy of a given numerical simulation is often a matter of a judicious choice of these values.

In this presentation, we show how a simple evolutionary programming (EP) algorithm can optimize a given set of parameters in a mesoscale atmospheric model with respect to agreement between simulation and observations. This is illustrated using the Regional Atmospheric Modeling System (RAMS). As an initial test case, data from a RAMS simulation with a default set of parameters, rather than actual data, is used to calculate an objective function. Ideally, the model parameters will evolve toward those of the default simulation as the objective function is minimized. This type of experiment also tests the ability of EP to find the global minimum of the objective function since the optimum is known.

Our primary goal was to demonstrate that an EP algorithm can provide a systematic and objective method for optimizing a mesoscale atmospheric model. We are now working towards exploring the path the model follows as it is pushed towards its target, and how such a scheme could be applied operationally.

Assimilation of Doppler Radar Data into High Resolution Numerical Model Forecasts

Steven R. Chiswell

Savannah River National Laboratory

Aiken, SC

During 2008, the United States National Weather Service (NWS) radars implemented a significant upgrade which increased the real-time level II data resolution to 8 times their previous “legacy” resolution, from 1 km range gate and 1.0 degree azimuthal resolution to “super resolution” 250 m range gate and 0.5 degree azimuthal resolution. These radar observations provide reflectivity, velocity and returned power spectra measurements at a range of up to 300 km (460 km for reflectivity) at a frequency of 4-5 minutes and yield up to 13.5 million point observations per level in super-resolution mode. The migration of National Weather Service (NWS) WSR-88D radars to super resolution is expected to improve warning lead times by detecting small scale features sooner with increased reliability; however, current operational mesoscale model domains utilize grid spacing several times larger than the legacy data resolution, and therefore the added resolution of radar data is not fully exploited.

Assimilation of radar velocity and precipitation fields into high-resolution model simulations can improve precipitation forecasts with decreased "spin-up" time and improve short-term simulation of boundary layer winds which is critical to improving plume transport forecasts. Accurate description of wind and turbulence fields is essential to useful atmospheric transport and dispersion results, and any improvement in the accuracy of these fields will make consequence assessment more valuable during both routine operation as well as potential emergency situations. The assimilation of super resolution reflectivity and velocity data into high resolution numerical weather model forecasts where grid spacing is comparable to the radar data resolution is investigated here to determine the impact of the improved data resolution on model predictions. As the increase in computational power and availability has made higher resolution real-time model simulations possible, the need to obtain observations to both initialize numerical models and verify their output has become increasingly important.

The assimilation of super resolution radar observations therefore provides a vital component in the development and utility of these models.

An Analytical Screening Technique to Estimate the Effect of Cooling Ponds on Meteorological Measurements: A Case Study

Stephen A. Vigeant, CCM

Carl A. Mazzola, CCM

Shaw Environmental and Intrastructure

An onsite meteorological measurement program supporting the licensing of a nuclear power generation facility outside of the United States consists of two separate meteorological towers. One tower monitors data at a height of 58 meters while the other at a height of 10 meters. In proximity to the smaller 10-meter tower are two discharge water ponds with elevated water temperatures, each sized at 12 meters by 12 meters and open to the atmosphere. The nuclear regulatory agency expressed a concern that the ponds were in close enough proximity to the meteorological towers to exert a micrometeorological influence on the relative humidity and ambient temperature measurements, and possibly on other parameters as well. These possible biases could have an effect on some environmental analyses required to license the facility. Thus, it became important to ascertain the extent that the monitored parameters would be biased and whether or not some or all of the parameters would no longer meet spatial representativeness requirements.

In order to establish whether this effect of the ponds was meaningful, a quantitative determination needed to be made to compare the effect of the discharge water ponds on the measurements to parameter accuracy requirements in ANSI/ANS-3.11(2005), a voluntary consensus standard on meteorological data measurements. Should the effect be significantly below these requirements, the meteorological data monitored at the 10-meter tower can be judged as spatially representative, essentially unaffected by the ponds, and thus of sufficient fidelity to perform environmental analyses.

The objective of this evaluation was to develop a defensible analytical technique that could be used to quantify the potential effect of the discharge ponds on the measurements taken at the 10-meter onsite meteorological tower located approximately 62 meters from the cooling ponds. The methodology would first have to establish “source terms” of fluxes of moisture and sensible heat at the surface of the ponds. Once the source terms have been established, a steady-state Gaussian transport and dispersion model could be applied to determine concentrations of moisture and heat at the 10-meter meteorological tower to be compared to ambient conditions. Since the distance from the ponds to the meteorological tower was less than 100 meters, a specially-developed Gaussian dispersion model, ARCON96, was applied, since this is the only code that accounts for the effects of plume meander and aerodynamic building wake on the horizontal and vertical dispersion magnitudes at downwind distances within 100 meters. Once the moisture and sensible heat effects are calculated, impacts on the ambient relative humidity and temperature measurements, respectively, can be developed for each hour of a 1-year onsite data base and subsequently defended in a regulatory setting.

The application of this analytical technique demonstrated that the effect of the cooling ponds on ambient temperature and moisture for these circumstances was much smaller than the accuracy requirements of ANSI/ANS-3.11(2005).

Trends in Estimated Mixing Depth Daily Maximums

Robert L. Buckley

Amy Dupont

Robert J. Kurzeja

Matthew J. Parker

Savannah River National Laboratory

Aiken, SC

Mixing depth is an important quantity in the determination of air pollution concentrations. Fire-weather forecasts depend strongly on estimates of the mixing depth as a means of determining the altitude and dilution (ventilation rates) of smoke plumes. The United States Forest Service (USFS-Savannah River) routinely conducts prescribed fires at the Savannah River Site (SRS), a heavily wooded Department of Energy (DOE) facility located in southwest South Carolina. For many years, the Savannah River National Laboratory (SRNL) has provided forecasts of weather conditions in support of the burn program, including an estimated mixing depth using potential temperature change with height at a given location.

This paper examines trends in the average estimated mixing depth daily maximum at the SRS over an extended period of time (>3 years) derived from numerical atmospheric simulations using two versions of the Regional Atmospheric Modeling System (RAMS). This allows for differences to be seen between the model versions, as well as trends on a multi-year time frame. On a monthly basis, seasonal trends are as expected with the highest mixing depth predictions occurring during the summer months, although lower maximum summer values are predicted to occur in the earliest year (2003). The variation in average mixing depth during the summer months agrees with annual variability in local observations (i.e. average temperature, precipitation, and heat stress). In addition, comparisons of predicted mixing depth for individual days in which special balloon soundings were released are also discussed.

SESSION TWO

A Tornadic Mini-Supercell with No Cloud-To-Ground Lightning

Patrick D. Moore

NOAA/NWS Forecast Office

Greer, SC

An unusually small but well-organized thunderstorm moved over the southern Foothills and Piedmont of North Carolina during the evening of 19 April 2008. The thunderstorm developed ahead of a cold front in an environment that was initially thought to be not conducive to vigorous updrafts because of a relative lack of buoyancy with Convective Available Potential Energy less than 500 J kg^-1 on a nearby upper air sounding. Although the storm was shallow with a high radar reflectivity core (40 dBZ) extending only to 15,000 feet above ground level, it exhibited several characteristics commonly associated with supercells, including a persistent (but weak) mesocyclone, a weak echo region, and a low level reflectivity appendage and inflow notch. Two tornadoes were produced as the storm moved across the eastern part of Cleveland County and the western part of Lincoln County. An absence of cloud-to-ground lightning was noted in the thunderstorm for 20 minutes prior to tornadogenesis and while the tornadoes were on the ground. The first cloud-to-ground lightning flashes were detected after the second tornado lifted, when the high reflectivity core of the storm extended to a height greater than 10,000 feet above the freezing level. No significant increase in rotational velocity or horizontal shear, defined as an increase of 50 percent or greater between radar volume scans, was noted until the first tornado was already on the ground. A successful warning was issued for the second tornado only after several human factors were overcome and a careful interrogation of the data from the Terminal Doppler Weather Radar near the Charlotte – Douglas International Airport was performed.

A Case Study of the 15 March 2008 South Carolina Supercell Outbreak: Pre-Storm Environment

Hunter Coleman

A. W. Petrolito

Michael W. Cammarata

NOAA/NWS Forecast Office

Columbia, SC

David A. Glenn

NOAA/NWS Forecast Office

Gray/Portland, ME

On 15 March 2008 a supercell thunderstorm outbreak occurred across Georgia and South Carolina. Seven long-track supercells produced numerous minor tornadoes and several stronger (EF2-EF3) tornadoes causing an estimated 40 million dollars in damage. Storm initiation and development occurred along a west-east oriented surface moisture gradient extending across north-central Georgia into southern North Carolina ahead of an approaching cold front. The potential for supercell development leading up to the event was due to strong mid and low-level wind shear coupled with ample lift, surface convergence, and strong to extreme instability. This presentation will address the specific synoptic and mesoscale environmental conditions preceding tornadogenesis across the National Weather Service Columbia area of responsibility.

A Case Study of the 15 March 2008 South Carolina Supercell Outbreak: Radar Analysis

Steven Naglic

NOAA/NWS Forecast Office

Columbia, SC

David A. Glenn

NOAA/NWS Forecast Office

Gray/Portland, ME

On 15 March 2008 a supercell thunderstorm outbreak occurred across Georgia and South Carolina. Seven long-track supercells produced numerous minor tornadoes and several stronger (EF2-EF3) tornadoes causing an estimated 40 million dollars in damage. Storm initiation and development occurred along a west-east oriented surface moisture gradient extending across north-central Georgia into southern North Carolina ahead of an approaching cold front. This presentation will examine the three dimensional radar signatures associated with some of these storms across the National Weather Service Columbia area of responsibility. Striking examples of hook echoes, bounded weak echo regions, and tornadic vortex signatures will be shown.

Using GR2Analyst together with AWIPS D2D displays, forecasters can gain a more complete insight to storm structure and morphology. This allows the radar operator to determine which D2D radar products may be better suited to identify parameters and characteristics used to issue warnings in a timely fashion. A comparison of D2D radar products with GR2Analyst products will demonstrate how they can be used in concert with each other to help improve the warning process.

A Case Study of the 11 June 2008 Severe Pulse Thunderstorm Over Newberry County, South Carolina

Michael W. Cammarata

NOAA/NWS Forecast Office

Columbia, SC

Jamie Christopher

Student Intern, NOAA/NWS Forecast Office

Columbia, SC

On the afternoon and evening of 9 July 2008, thunderstorms developed in a weakly sheared environment over central South Carolina. One of these storms became severe over Newberry County, SC producing 0.88 in hail. This study examines the synoptic, and mesoscale environment in which this storm formed and the three dimensional radar signatures that show the evolution of the updraft and hail core associated with this storm.

SESSION THREE

An Examination of a Non-Convective High Wind Event on 24 October 2008 Over South Carolina

Daniel C. Miller

Richard J. Linton

NOAA/NWS Forecast Office

Columbia, SC

On the late afternoon and early evening of 24 October 2008, an apparent gravity wave affected portions of Georgia and South Carolina, including the National Weather Service Columbia, SC area of responsibility. Strong non-convective wind gusts resulted in downed trees and other structural damage. Over one million dollars in property damage occurred. A case study of this event was performed utilizing archived data, including upper air and surface observations and analyses, GOES Satellite data, WSR-88D Doppler Radar products, and model forecast data. Meteorological features that contributed to the event are discussed. Based on past research and technical writings on the subject, the overall meteorological conditions for this event appeared consistent with those of previous gravity wave events.

Tropospheric Ozone Days and Atmospheric Pattern Relationships in Columbia, South Carolina vs. Charlotte, North Carolina

Casey Zuzak

University of South Carolina, Department of Geography

Columbia, SC

Atmospheric pollution has become a worldwide problem, as seen in the preparation for the 2008 Beijing Summer Olympic Games. China is not the only country that is currently effect by atmospheric pollution, but India, Taiwan, Spain, Greece, Canada, and the United States all contribute large amounts of air pollution. Tropospheric ozone has been known to cause adverse effects to humans and plants in the green areas of cities. For this study, the Daily maximum 8-hour mean ozone concentrations will be recorded along with the air mass present over Columbia, South Carolina and Charlotte, North Carolina from 1 January 2002 to 31 October 2007. This study will address two main questions, how much does population affect the surface ozone and what weather patterns have similar ozone days. This allows for forecaster to have a better understanding of what type of weather patterns are favorable for high surface ozone days, and if there is a link between population and surface ozone.

Extratropical Transitions of 19^th Century Tropical Cyclones

Matthew Rodgers

University of South Carolina, Department of Geography

Columbia, SC

New Englanders of the 19th century witnessed numerous cyclones of tropical origin.  The accounts of these storms were documented by residents and shipping interests caught in the turmoil.  Historical accounts through archival newspapers such as the Newport Mercury, Barnstable Patriot and the New York Herald describe the impact on local cities and villages.  Instrumental records from ship's logs, forts, and personal journals provide a wealth of knowledge into the state of the atmosphere at the surface.

HURDAT, the modern tropical cyclone record, currently extends into the period of interest, though does not contain storms earlier than 1850.  The information gathered in this study, would aid in any HURDAT expansion by

creating a better understanding of the synoptic environment during each storm.  The historical information confirms synoptic features such as frontal boundaries and asymmetry of temperature and precipitation fields that we know potentially signify extratropical transition.  New England is the region where the highest likelihood of transition may occur close to shore.  The synoptic analysis leads to the production of daily weather maps that aid in understanding each storm.  Such features easily discernable in these maps include snowfall recorded during the storm of 1841, and the interaction with a frontal zone in 1849 and 1858.  Though their origins are tropical, these storms have transitioned into still powerful extratropical cyclones, and represent a danger to life, property, and the environment in the region.  For this reason, it is important that discussion occur on the same level as non-transitioned storms.

Operational Utility of Atmospheric Favorability for Severe Storms (AFS) Gridded Data

Jeffrey P. Taylor

Harry Gerapetritis

NOAA/NWS Forecast Office

Greer, SC

Meteorologists at the National Weather Service Forecast Office in Charleston, West Virginia have developed a technique that attempts to identify locations and times with the greatest likelihood of severe thunderstorm development. This technique, called “Atmospheric Favorability for Severe Storms” (AFS), has been implemented on an experimental basis since June of 2008 at the National Weather Service Weather Forecast Office in Greer, SC. The technique uses hourly computer model forecasts of the basic ingredients for severe thunderstorms, such as low level convergence of the environmental wind field and instability, to produce a threat level grid for each hour of the day. 

AFS data were archived for every hour of the day using the latest available model run at the time.  The forecast projection could be as short as one hour from model initialization time or as long as 10 hours, or anything in between. Threat levels are indicated as: “None,” “Low,” “Medium,” “High,” or “Extreme” and can be derived from a specific computer model or a composite of several models. The goal of the technique is to provide environmental information to warning forecasters that enhances their situational awareness and permits longer lead-time warnings based on cell proximity to higher risk areas. This study compares average AFS grid values over selected areas of interest to the occurrence of severe thunderstorm events and warnings. This is done in order to assess the utility of the technique and to determine which models perform best with regards to identifying threat areas using the AFS methodology. AFS values were positively correlated to the occurrence of severe thunderstorm events and warnings in time and space, but the correlation was weak across the full spectrum of AFS values. AFS values computed from the NAM and RUC generally outperformed the GFS, SREF, and local workstation WRF in accurately assessing threat levels and risk areas. Furthermore, higher-end AFS values exhibited good reliability in predicting the occurrence of severe events and warnings.

Synoptic-Scale Conditions Leading to Flooding in South Carolina: A Case Study of October 22-23, 1990

Erik D. Kabela

Savannah River National Laboratory

Aiken, SC

University of South Carolina, Department of Geography

Columbia, SC

Joanne Stevenson

University of South Carolina, Department of Geography

Columbia, SC

Wide-spread flooding can be a common occurrence throughout the southeastern United States, and South Carolina is no exception. Typically, South Carolina experiences its highest rainfall totals during the late spring and early summer months (June through August) when tropical activity is at its maximum. Minimum rainfall during the seasonal transition periods in April/May and October/November are typical of the Midlands and Low Country of South Carolina due to the modification of air masses associated with mid-latitude cyclones as they pass over the Appalachian Mountains.

October 1990 ranks as the wettest October for the modern record (1895-2008) in South Carolina and was characterized by three separate weather systems: from October 10-13 the remnants of Hurricane Klaus and Tropical Storm Marco moved northwards through the Southeast U.S. along a stationary front, and October 22-23 which was the result of an intense synoptic-scale system moving from the west. The effects of Hurricane Klaus and Tropical Storm Marco on South Carolina have been well documented; however the synoptic event later in the month of October has received little attention and will be the focus of this presentation.

This work illustrates a real-world example of a Maddox synoptic-type flash flood event coupled with a Konrad pattern 2 heavy rain event. The association of these event types in this case study illustrates how Southeast region forecasters can use pattern recognition of these idealized types in conjunction with an understanding of pre-existing conditions to forecast heavy rainfall and flooding.

NOTES: