Search Results

You are looking at 1 - 10 of 13 items for

  • Author or Editor: Michael D. Eilts x
  • Refine by Access: All Content x
Clear All Modify Search
Michael D. Eilts

Abstract

The feasibility of using the next generation weather radar (NEXRAD) system to detect low-altitude wind shear near airports is investigated. We compare surface-measured horizontal shear with that observed aloft with Doppler radar to determine how the radar-estimated shear above the surface relates to the surface-measured shear. For five Oklahoma gust fronts, the Doppler radar estimate of shear (at heights between 50–600 m) averaged 1.6 times the shear measured at the surface. For none of 43 comparisons was the surface radial velocity difference across the gust front stronger than the radial velocity difference measured by Doppler radar aloft When the five gust fronts passed an instrumented tower a vertical profile through the lowest 440 m of the gust front could be determined. In all cases the wind speed and wind shear increased in the lowest 90 m of the atmosphere. In one case, the 90 m height had the peak wind shear, in all other case the peak wind shear was at a much higher altitude. The Federal Aviation Administration requires that NEXRAD radar coverage have a lowest scan of 60 m above the surface in the airport area (within 20 km of the airport), the strongest shears in the five gust fronts investigated in this study were at the 90 m or higher levels of the tower. Due to surface friction. it is expected that wind speeds and shears in downbursts will also be stronger aloft than at the surface; however, further study is necessary. It is suggested that a combination of Doppler radar data and information gleaned from a Low-Level Wind Shear Alert System (LLWSAS) would allow more accurate wind shear estimates in the terminal area of airports than would be possible with either system by itself.

Full access
Kurt D. Hondl
and
Michael D. Eilts

Abstract

The capability of Doppler weather radars to short-term forecast the initiation of thunderstorms and the onset of cloud-to-ground (CG) lightning is examined. Doppler weather radar data from 28 thunderstorms were analyzed from August 1990 in the central Florida environment. These radar echoes were associated with CG lightning strike locations from the National Lightning Detection Network and two lightning detection systems operated by the U.S. Air Force in the vicinity of Kennedy Space Center. From a time history of these radar echoes it was found that a 10-dBZ echo, first detected near the freezing level, may be the first definitive echo of a future thunderstorm. This thunderstorm initiation signature is often accompanied by low-altitude convergence and divergence at the top of the radar echo. The observed lead times between this thunderstorm initiation signature and the first detected CG lightning strike ranged from 5 to 45 min with a median lead time of 15 min. All lightning-producing radar echoes were detected using the thunderstorm initiation signature; however, some echoes exceeded the 10-dBZ threshold and did not produce any CG lightning. The characteristics of the WSR-88D and Terminal Doppler Weather Radar systems are evaluated for their capability to detect the thunderstorm initiation signature in central Florida with sufficient temporal and spatial resolution.

Full access
Michael D. Eilts
and
Steven D. Smith

Abstract

A Doppler velocity dealiasing algorithm is described that processes one radial at a time by comparing that radial with a previous radial. This technique has worked reliably on numerous Doppler radar datasets for clear air, thunderstorm, and severe thunderstorm situations. It was also tested on four volume mans from severe weather environments with difficult aliasing problems to determine statistically how well the algorithm performs in a worst case environment. Of some 1.2 million velocities in these severe storms, 0.2% were improperly dealiased, and 93% of those were above 13 km height in the storm-top divergent region where shears were extreme. Every tornado, mesocyclone, gust front, microburst, and storm-top divergent signature was preserved, and could be readily discerned by human analyst. No adverse impact was observed on the signature and automated signature detection algorithms would therefore be freed from contamination by velocity aliasing. The velocity dealiasing algorithms described is adaptive and therefore efficient because simple cheeks are made initially, and progressively more sophisticated and time-consuming checks are used only if they are needed.

Full access
Michael D. Eilts
and
Richard J. Doviak

Abstract

Doppler radar data collected each spring in 1979–1984 with the two Doppler radars operated by the National Severe Storms Laboratory (NSSL) are used to investigate the asymmetry of low-altitude divergent outflows of convective storm downbursts in central Oklahoma. Outflows in Oklahoma storms can be highly asymmetric with horizontal shear along the axis of maximum divergence as much as 5.5 times the shear along the axis of minimum divergence. The downbursts observed in central Oklahoma, all large-scale (4–10 km) events, were superposed with the maximum reflectivity core of the storms. However, scanning strategies may have precluded detection of smaller scale (>4 km) microbursts. Typical downbursts observed during the Joint Airport Weather Studies (JAWS) Project were of small scale (>4 km) and were often associated with little or no rain at the surface. The mechanism for the initiation of the majority of JAWS microbursts was evaporative cooling, which occurred when precipitation fell into a dry, deep and nearly adiabatic boundary layer it appears that other mechanisms are responsible for the initiation of the observed Oklahoma downbursts because of a lower cloud base and a moister and slightly more stable boundary layer.

Full access
Harold E. Brooks
,
Arthur Witt
, and
Michael D. Eilts

The question of who is the “best” forecaster in a particular media market is one that the public frequently asks. The authors have collected approximately one year's forecasts from the National Weather Service and major media presentations for Oklahoma City. Diagnostic verification procedures indicate that the question of best does not have a clear answer. All of the forecast sources have strengths and weaknesses, and it is possible that a user could take information from a variety of sources to come up with a forecast that has more value than any one individual source provides. The analysis provides numerous examples of the utility of a distributions-oriented approach to verification while also providing insight into the problems the public faces in evaluating the array of forecasts presented to them.

Full access
Laurie G. Hermes
,
Arthur Witt
,
Steven D. Smith
,
Diana Klingle-Wilson
,
Dale Morris
,
Gregory J. Stumpf
, and
Michael D. Eilts

Abstract

The Federal Aviation Administration's Terminal Doppler Weather Radar (TDWR) system was primarily designed to address the operational needs of pilots in the avoidance of low-altitude wind shears upon takeoff and landing at airports. One of the primary methods of wind-shear detection for the TDWR system is the gust-front detection algorithm. The algorithm is designed to detect gust fronts that produce a wind-shear hazard and/or sustained wind shifts. It serves the hazard warning function by providing an estimate of the wind-speed gain for aircraft penetrating the gust front. The gust-front detection and wind-shift algorithms together serve a planning function by providing forecasted gust-front locations and estimates of the horizontal wind vector behind the front, respectively. This information is used by air traffic managers to determine arrival and departure runway configurations and aircraft movements to minimize the impact of wind shifts on airport capacity.

This paper describes the gust-front detection and wind-shift algorithms to be fielded in the initial TDWR systems. Results of a quantitative performance evaluation using Doppler radar data collected during TDWR operational demonstrations at the Denver, Kansas City, and Orlando airports are presented. The algorithms were found to be operationally useful by the FAA airport controllers and supervisors.

Full access
J. T. Johnson
,
Pamela L. MacKeen
,
Arthur Witt
,
E. De Wayne Mitchell
,
Gregory J. Stumpf
,
Michael D. Eilts
, and
Kevin W. Thomas

Abstract

Accurate storm identification and tracking are basic and essential parts of radar and severe weather warning operations in today’s operational meteorological community. Improvements over the original WSR-88D storm series algorithm have been made with the Storm Cell Identification and Tracking algorithm (SCIT). This paper discusses the SCIT algorithm, a centroid tracking algorithm with improved methods of identifying storms (both isolated and clustered or line storms). In an analysis of 6561 storm cells, the SCIT algorithm correctly identified 68% of all cells with maximum reflectivities over 40 dBZ and 96% of all cells with maximum reflectivities of 50 dBZ or greater. The WSR-88D storm series algorithm performed at 24% and 41%, respectively, for the same dataset. With better identification performance, the potential exists for better and more accurate tracking information. The SCIT algorithm tracked greater than 90% of all storm cells correctly.

The algorithm techniques and results of a detailed performance evaluation are presented. This algorithm was included in the WSR-88D Build 9.0 of the Radar Products Generator software during late 1996 and early 1997. It is hoped that this paper will give new users of the algorithm sufficient background information to use the algorithm with confidence.

Full access
Arthur Witt
,
Michael D. Eilts
,
Gregory J. Stumpf
,
J. T. Johnson
,
E. De Wayne Mitchell
, and
Kevin W. Thomas

Abstract

An enhanced hail detection algorithm (HDA) has been developed for the WSR-88D to replace the original hail algorithm. While the original hail algorithm simply indicated whether or not a detected storm cell was producing hail, the new HDA estimates the probability of hail (any size), probability of severe-size hail (diameter ≥19 mm), and maximum expected hail size for each detected storm cell. A new parameter, called the severe hail index (SHI), was developed as the primary predictor variable for severe-size hail. The SHI is a thermally weighted vertical integration of a storm cell’s reflectivity profile. Initial testing on 10 storm days showed that the new HDA performed considerably better at predicting severe hail than the original hail algorithm. Additional testing of the new HDA on 31 storm days showed substantial regional variations in performance, with best results across the southern plains and weaker performance for regions farther east.

Full access
Gregory J. Stumpf
,
Arthur Witt
,
E. DeWayne Mitchell
,
Phillip L. Spencer
,
J. T. Johnson
,
Michael D. Eilts
,
Kevin W. Thomas
, and
Donald W. Burgess

Abstract

The National Severe Storms Laboratory (NSSL) has developed a mesocyclone detection algorithm (NSSL MDA) for the Weather Surveillance Radar-1988 Doppler (WSR-88D) system designed to automatically detect and diagnose the Doppler radar radial velocity patterns associated with storm-scale (1–10-km diameter) vortices in thunderstorms. The NSSL MDA is an enhancement to the current WSR-88D Build 9.0 Mesocyclone Algorithm (88D B9MA).

The recent abundance of WSR-88D observations indicates that a variety of storm-scale vortices are associated with severe weather and tornadoes, and not just those vortices meeting previously established criteria for mesocyclones observed during early Doppler radar studies in the 1970s and 1980s in the Great Plains region of the United States. The NSSL MDA’s automated vortex detection techniques differ from the 88D B9MA, such that instead of immediately thresholding one-dimensional shear segments for strengths comparable to predefined mesocyclone parameters, the initial strength thresholds are set much lower, and classification and diagnosis are performed on the properties of the four-dimensional detections. The NSSL MDA also includes multiple range-dependent strength thresholds, a more robust two-dimensional feature identifier, an improved three-dimensional vertical association technique, and the addition of time association and trends of vortex attributes. The goal is to detect a much broader spectrum of storm-scale vortices (so that few vortices are missed), and then diagnose them to determine their significance. The NSSL MDA is shown to perform better than the 88D B9MA at detecting storm-scale vortices and diagnosing significant vortices.

Operational implications of the NSSL MDA are also presented. In light of the new WSR-88D observations of storm-scale vortices and their association with severe weather and tornadoes, it is clear that the operational paradigms of automated vortex detection require changes.

Full access
E. De Wayne Mitchell
,
Steven V. Vasiloff
,
Gregory J. Stumpf
,
Arthur Witt
,
Michael D. Eilts
,
J. T. Johnson
, and
Kevin W. Thomas

Abstract

The National Severe Storms Laboratory (NSSL) has developed and tested a tornado detection algorithm (NSSL TDA) that has been designed to identify the locally intense vortices associated with tornadoes using the WSR-88D base velocity data. The NSSL TDA is an improvement over the current Weather Surveillance Radar-1988 Doppler (WSR-88D) Tornadic Vortex Signature Algorithm (88D TVS). The NSSL TDA has been designed to address the relatively low probability of detection (POD) of the 88D TVS algorithm without a high false alarm rate (FAR). Using an independent dataset consisting of 31 tornadoes, the NSSL TDA has a POD of 43%, FAR of 48%, critical success index (CSI) = 31%, and a Heidke skill score (HSS) of 46% compared to the 88D TVS, which has a POD of 3%, FAR of 0%, CSI of 3%, and HSS of 0%. In contrast to the 88D TVS, the NSSL TDA identifies tornadic vortices by 1) searching for strong shear between velocity gates that are azimuthally adjacent and constant in range, and 2) not requiring the presence of an algorithm-identified mesocyclone. This manuscript discusses the differences between the NSSL TDA and the 88D TVS and presents a performance comparison between the two algorithms. Strengths and weaknesses of the NSSL TDA and NSSL’s future work related to tornado identification using Doppler radar are also discussed.

Full access