REU Home | 2010 Home | Students & Mentors | Projects | Schedule | Travel Tips | Photos



NWC REU 2010

May 25 - July 30



2010 NWC REU Papers

Assessing Climate Change Impacts on the Blue River Basin of Oklahoma [PDF]

Christopher Bednarczyk
Mentors: Yang Hong (OU CEES)

A 16 GCM ensemble was used to assess the future climate of Oklahoma and its Blue River Basin under three IPCC emissions scenarios. Output from the models was then applied to a monthly water balance model to predict changes in the hydrologic cycle. By the end of the century ensemble median warming is predicted to be 2.2 to 4.6 °C for the state depending on the scenario. Precipitation trends depended on the emissions scenario, with the state experiencing almost no annual change. The Blue River Basin is expected to receive slightly more precipitation under the lower emissions scenario and less under the higher scenario. Change in temperature along with little change in precipitation led to predicted increase in both potential evapotranspiration and actual evapotranspiration. Soil moisture and runoff are both expected to decrease significantly. Runoff changes by 2100 ranged from ensemble mean of -9.6% for the lower emission scenario to -29.8% for the higher.

Evaluating High-Resolution NWP Forecasts of the Nocturnal Low Level Jet for Improving Wind Power Forecasts [PDF]

Jeffrey Deppa
Mentors: Richard Carpenter and Brent Shaw (WDT, Inc.)

The Nocturnal Low Level Jet (NLLJ) is a significant contributor to overnight wind power production in the Southern Great Plans. This region of the United States is expecting wind farm growth over the coming decades and therefore it is important to better understand how to forecast wind energy, and hence forecast the strength and location of the NLLJ. The Weather Research and Forecasting Model (WRF) is one tool that can be used for forecasting winds. This study investigates performance of a real-time, high-resolution (3-km grid spacing) configuration of the WRF for several NLLJ cases in southwest Oklahoma. Forecast location and intensity of the NLLJ and its interaction with moderate terrain features around the Blue Canyon Wind Farm, particularly the Wichita Mountains and Slick Hills, were evaluated. These model forecasts also provide insight into the relationship between NLLJ behavior as a function of wind magnitude and atmospheric stability. The study finds that errors in model forecasted boundary layer stability coupled with NLLJ terrain interactions could be the reason for wind forecast errors at Blue Canyon.

Evaluation of NWS Storm-Based Warnings Using Gridded Products [PDF]

Todd Ferebee
Mentors: Kiel Ortega (OU CIMMS) and Kevin Scharfenberg (NOAA NWS OCWWS)

In 2008, the National Weather Service began issuing storm-based polygon warnings instead of county warnings. Only one severe hail, wind, or tornado report is needed to verify an entire warning polygon. Few severe weather reports in the warning, and in turn for the storm which prompted the warning, makes difficult to determine the spatial extent of severe weather for a particular storm. Since 2006, the Severe Hazards Analysis and Verification Experiment (SHAVE) has been collecting severe weather reports at temporal and spatial resolutions much higher than those available in Storm Data. The National Severe Storms Laboratory (NSSL) produces several severe weather products, such as reflectivities at different isotherms and estimated hail size, on a grid for the entire contiguous United States. These grids could provide for synthetic verification of severe weather especially for the spatial extent of severe weather. This study will investigate how well the grids perform in determining where severe hail fell by using high resolution SHAVE reports. Discussed for applications of such grids for warning verification and improvement will also be included.

Updraft Helicity as a Forecast Parameter [PDF]

Stacey Hitchcock
Mentors: Patrick Marsh (OU SOM), Harold Brooks (NOAA NSSL), and Charles A. Doswell III (OU CIMMS)

Improved science and technology has created the opportunity to explore the impacts of different model diagnostic fields as indicators of convection developed in highresolution numerical models. Indication of the success of different diagnostic fields has been discussed (Kain et al. 2008, Sobash et al. 2008). Updraft helicity (UH) has shown a particular ability to identify supercell-like structure in convection allowing model observed locations. UH will be examined to determine the best integration layer over which to calculate UH.

Output of updraft helicity over different layers from the convection allowing 4-km National Severe Storms Laboratory- Weather Research and Forecasting Radar (NSSLWRF) Advanced Research WRF (ARW) from the Spring Experiment 2008 was compared to Storm Prediction Center (SPC) storm reports using contingency tables. Verification measures (Probability of Detection, False Alarm Ratio, Critical Success Index, bias) were calculated from the contingency tables and used to create several visual comparisons. These include Relative Operating Characteristic curves (ROC) (Mason 1982), and Performance Diagrams (Roebber 2008), as a comparison of different depth’s success as a forecast parameter.

Spatial Analysis of Tornado Vulnerability Trends in Oklahoma and Northern Texas [PDF]

Eric Hout
Mentors: May Yuan, John MacIntosh, and Chris Weaver (OU CSA)

Determination of effective ways to reduce vulnerability from tornadoes is one of the fundamental drivers for tornado research. This study analyzes spatial vulnerability in the context of past tornado events with aims to enhance the understanding of tornado casualties in Oklahoma and Northern Texas. Many previous studies on tornado vulnerability have provided insight on how individual factors influence overall social and spatial vulnerability. However, few studies have been conducted to evaluate the aggregated effect on vulnerability when these factors coincide. Additionally, a definition of vulnerability has been absent from the meteorological literature. Thus, to provide a more comprehensive view of vulnerability, this study proposes a mathematical definition for spatial vulnerability, and then uses tornado casualty data from 1950 through 2009 to calculate vulnerability on a county level for seven time periods. Overall vulnerability trends are then calculated and visualized by averaging changes and by k-means clustering. This study shows the existence of spatial patterns in vulnerability between counties both when analyzing each individual F-scale and when all F-scales are combined. These spatial patterns are likely caused by the existence of multiple variables working together.

Radar-Disdrometer Comparison to Reveal Attenuation Effects on CASA Radar Data [PDF]

Christopher Kerr
Mentors: Guifu Zhang (OU SOM)

X-band radar provides a high resolution image at the cost of significant attenuation. This is due to the X-band's short wavelength. In this study, a two dimensional video disdrometer (2DVD) was deployed to the center of a triangle formed by three CASA dual-polarization X-band radars. The CASA radars provide measurements of reflectivity, differential reflectivity, specific differential phase, and copolar cross-correlation coefficient of precipitation. Using drop size distributions obtained from the 2DVD, the radar variables are calculated and treated as the ground truth. The radar and disdrometer measurements are compared to reveal discrepancies. Biases and errors are calculated, and possible causes are investigated. These results can be used to further minimize the attenuation obstacle in X-band radar.

An Analysis of Southern U.S. Ice Storm Frequency From 2000-2009 [PDF]

Carly Kovacik
Mentors: Mark Shafer and James Hocker (OCS)

The Southern Climate Impacts Planning Program (SCIPP) is a climate research program that focuses on helping the public improve planning for weather and climate-related disasters. SCIPP focuses on the high frequency of hazardous weather events, including extremes in precipitation. Over the past several years, SCIPP has speculated that there has been an increase in the number of ice storms within the region each winter. This paper analyzes trends in ice storm frequency and intensity for the years 2000-2009 using data from the National Climatic Data Center’s Storm Events and Storm Data datasets. For this period of study, it was found that an ice storm maximum stretches from southwestern Texas through Oklahoma, northwestern Arkansas, southeastern Kansas and central Missouri. It was also found that there is no consistent trend associated with the number of ice storms, the ice thickness values of recorded ice storms, or the number of ice storm catastrophes over the last ten years. Ice storm frequency was also briefly compared to atmospheric signals and El Nino-Southern Oscillation (ENSO) events. This project also identified discrepancies in ice storm reporting across National Weather Service office boundaries as evidenced through Geographic Information Systems mapping. This project provides preliminary results that can be incorporated into more extensive studies to create national criteria for documenting ice storms.

Verifying Model Forecasts of Arctic Fronts in Advance of Winter Storms in the Southern Plains [PDF]

Willliam E. Leatham, IV
Mentors: Patrick Burke and Andrew Taylor (NOAA NWS Norman)

Arctic fronts and associated freezing line positions are of concern in winter storm forecasting. In the southern Great Plains of the United States the arrival of shallow arctic air plays a major role in the development of severe ice storms. At other times, the cold air becomes deep enough to support snowstorms and even blizzard conditions. Forecasters' providing at least twelve to twenty-four hour advanced warning allows the public and other groups time to prepare for these potentially dangerous events. Therefore, determining how operational forecast models perform in these situations is crucial to improving forecast accuracy and increasing our understanding of shallow cold air. This paper compares the observed surface freezing line and cold front location with model forecasts of both these features during the twenty-four hour period leading up to the onset of four winter storms. The model forecasts tend to move arctic fronts southward much too slowly. This has strong implications for the southward extent of winter storm warnings based on model forecasts, and their associated lead time.

Public Response to the 10 May 2010 Norman, Oklahoma Tornado [PDF]

Sarah Stalker
Mentors: Heather Lazrus (OU SSWIM), Kristin Kulhman (OU CIMMS), Randy Peppler (OU CIMMS), and Kim Klockow (OCS)

The purpose of this project is to gather initial actions and reactions from the public in response to the 10 May 2010 Norman, Oklahoma tornado. This is done in support of the National Severe Storms Laboratory's Warn-on-Forecast project for severe thunderstorm, tornado, and flash flood events. The tools and products that will be developed as part of the project are needed to improve warning for both the public and community stakeholders i.e. emergency managers, hospitals, and schools. This research study consisted of formally interviewing 6 individuals impacted by the May 10 storm and analyzing their responses. The majority of the interviewees did not feel any direct threat from the tornado during the early stages of storm development and advisories. Interestingly, with a longer lead-time promised by Warn-on-Forecast, most said they would still probably wait to obtain more information before taking any form of shelter or enact a safety plan. Most of the participants said it would be beneficial to see the expected track information Warn-on-Forecast will provide to help make their decisions on whether they felt the need to take safety measures. The results in this study will help to aid the National Severe Storms Laboratory in further development of the Warn-on-Forecast system with respect to public perspectives on longer lead times and other information needs.

Toward an Analysis of the Influence of the Urban Heat Island Effect on Single-Cell Convective Cloud Trajectories [PDF]

Joshua Turner
Mentors: Brian Vant-Hull (CUNY)
*Exchange with CUNY's REU

This is an initial analysis of how the Urban Heat Island (UHI) Effect influences convective cloud trajectories. Case studies have been done in the past that demonstrate the observed movement of convective storms around various urban areas like Atlanta and New York City. This analysis uses the GOES-East satellite to track single-cell convective cloud towers (SCCTs) and the algorithm for following the SCCT tracks them using infrared data from satellite observations taken at 15-minute intervals. Matlab was then used to filter and extract suitable trajectories by using thresholds on both speed and direction. Following the filters to determine viable trajectories, another process was developed to determine where along the trajectory there existed a significant deviation in the propagation of the particular SCCT. After these specific perturbations in trajectory angle were found, they were then cross-referenced with locations of urban areas in the domain using a density plot to determine where the highest concentration of significant perturbation points occurred.

Comparisons of Flood Affected Area Derived From MODIS and Landsat Imagery [PDF]

Kevin Van Leer
Mentor: John Galantowicz (AER, Inc.)

During the first few weeks of June 2008, the Midwest experienced a weather system that dropped large amounts of rain across the region. In southern Indiana the Wabash and White Rivers went several feet above their flood stage and many people were displaced from their homes and businesses. This study uses the event as a test case for comparisons of resolutions and data from the MODIS and Landsat 5 TM sensors. A k-means classification scheme is created to cluster the data to identify the flood region in the imagery. Calculations are then made to estimate a flood area for each resolution. A statistical study is then performed to analyze false positive and and false negative rates using Landsat imagery as "ground truth." The results of the area estimate and statistical study support a claim that coarse resolutions, 1 and 2 kilometers, provide the most
accurate measuring of area in large scale flood events, but the overall location of the fine details of the flood are lost. The finer resolutions (500 and 250 meters), while more accurate about locations of fine details, have a higher false positive and false negative rates that raise questions about their ability to effectively use this classification scheme to measure overall area. The conclusions of this research promote further questions as to what resolutions could be effectively used gain anaccurate map and measurement of the innundated area's extent.

Measured Severe Convective Wind Gust Climatology of Thunderstorms for the Contiguous United States, 2003-2009 [PDF]

Andrew Winters
Mentors: Bryan Smith and Corey Mead (NOAA NWS SPC)

A severe convective wind gust climatology spanning 2003-2009 for the contiguous United States is developed using measured Automated Surface Observing System (ASOS), Automated Weather Observing System (AWOS), and Oklahoma Mesonet wind observations. National Lightning Detection Network and Radar Mosaic/Level II data are used amongst other quality control checks to identify and remove erroneous observational data. The filtered observations are then time matched with a number of diagnostic mesoanalysis fields from the Storm Prediction Center (SPC) for assessment of the severe convective wind gust environments. These data are then binned based on season and geographic region in order to identify atmospheric regimes characteristic to different parts of the country. The filtered observations are compared to storm reports archived by the SPC. Finally, a relatively denser surface observing network in Oklahoma is utilized to determine how consistently severe convective wind gusts are recorded by differing networks (i.e. ASOS/AWOS and Oklahoma Mesonet).

This study characterizes and contextualizes observations associated with southeast weak shear environments and contiguous U.S. strong deep layer shear, higher CAPE atmospheric regimes. Additionally, results exemplify the usefulness and necessity of a dense observing system network and demonstrate that the highest frequency of measured wind gusts occur throughout the southern and central High Plains and in a corridor from South Dakota across the southern Great Lakes region.



Copyright © 2010 - Board of Regents of the University of Oklahoma