|
HailTrax
Technical Paper
Co-developers of the Hail Detection Algorithm for
the NEXRAD Weather Radar Program have come together at Weather Decision
Technologies to develop HailTrax the first fully automated
report generator capable of determining the areal extent and size
of hail that fell based upon a four-dimensional data set available
from multiple radars, observed hail sightings, and unique and patented
algorithms developed by leading meteorologists at WDT.
HailTrax has the capability to accurately map where hail
has fallen utilizing sophisticated computer algorithms written by
world leading meteorologists at WDT. These algorithms utilize Doppler
Weather Radar data and other proprietary data sets to determine,
with great accuracy, where hail has fallen and the size of the largest
hailstones in a given area. When ordering a HailTrax report you
define the area of concern and the time of interest. Once the report
is produced, a meteorologist will personally check the results to
ensure the accurate mapping of hailswaths before the report is delivered
to you. The report will contain a map showing hailswaths and hail
sizes near your location of interest, and a note detailing the size
of hail that is estimated to have occurred at that location.
An example of the HailTrax product image is shown below. The
area where hail fell is mapped according to the estimated hail size.
HailTrax maps areas that hail fell that had diameter greater than
¾ inch. It is also able to delineate between hail that is
¾ inch to 1 and ¾ inches and hail greater than or
equal to 2 inches in diameter.

Figure 1: Example of HailTrax image from large
hail event near Lubbock, Texas on May 30, 2001.
Background
The United States National Weather Service, Federal Aviation Administration
and Department of Defense deployed a network of 140 state-of-the-art
Doppler Weather Radars (called NEXRAD) throughout the Continental
United States (Figure 2). The NEXRADs collect data continuously
and provide full 3-D coverage of their local region every 6 minutes.
In addition, the NEXRAD has a number of algorithms, many developed
by scientists at the National Severe Storms Laboratory (NSSL), that
automatically detect severe weather phenomena. A number of these
scientists are now employed by WDT. One of those algorithms is called
the Hail Detection Algorithm (HDA). The HDA determines which storms
are likely producing hail at the present time. In addition, the
HDA provides an indication of the size of hail presently being produced
by a storm. This information is used by the National Weather Service
(NWS) to aid in the process of warning the public of storms that
have the potential to produce hail that may damage property or result
in injury.

Figure 2: Nationwide coverage provided by 140 NEXRAD Doppler
Weather Radars.
Studies have shown that the HDA is skillful in determining whether
a given storm will produce hail. One study performed in Colorado
showed the HDA correctly determined which storm was producing hail
92% of the time with an extremely low false alarm rate of 4 %; (Witt
et. al. 1998). Also, based upon a geographically diverse radar dataset,
the HDA has demonstrated very accurate detections of severe hail
including the probability of detecting hail greater than 1 inch
(87% of the time) and 2 inches (96% of the time). These accuracy
numbers are based on using only one radar to detect hail. WDT's
proprietary algorithm utilizes overlapping data from multiple NEXRADs
that cover the area of interest. This allows us to achieve probability
of detection of hail nearing 100% over the lifetime of hail generating
storms. Of course, performance is a function of atmospheric conditions,
proper operation and calibration of the radar and quality reports
of hail.
The HailTrax Algorithm
When we receive your order, our experts determine which NEXRAD
radars provide coverage over your area of interest. We then extract
from our extensive archive of NEXRAD data the appropriate information
for input into the HailTrax algorithm. A second algorithm
examines the HailTrax output, takes into account all hail
observations made during the event of interest and plots hailswaths
showing the areal extent of hail and the estimated size of the largest
hailstones for each hailswath. Local landmarks such as state and
count boundaries, interstates, state highways and county roads are
added to the images for reference. Custom features or user assets
may also be added to the maps, for a small fee, such as city streets,
significant industrial centers/structures, customer locations, and
even proprietary geographically based information. The HailTrax
product is a valuable tool for accurately determining which locations,
assets and areas were impacted by hail, especially large damaging
hail.
Because of WDT's unique knowledge base and our archive and computing
infrastructure, we are able to provide the HailTrax product within
2 business days of your order. The unique HailTrax product is
the only product available anywhere in the world that accurately
maps the areal extent and size of hailstones.
References:
Conway, John W.*, Dusan S. Zrnic, 1993: A
Study of Embryo Production and Hail Growth Using Dual-Doppler and
Multiparameter Radars. Monthly Weather Review: Vol. 121,
No. 9, pp. 25112528.
Witt, Arthur, Michael D. Eilts*, Gregory J. Stumpf,
J. T. Johnson*, E. DeWayne Mitchell*,
Kevin W. Thomas, 1998: An Enhanced Hail Detection Algorithm for
the WSR-88D. Weather and Forecasting: Vol. 13, No. 2,
pp. 286303.
* WDT Employees
|