Main Page

Weather Map Discussion

Satellite Imagery

Radar Imagery

Forecast/Warning Services


Newspaper Accounts

Media Accounts

Public Accounts

Storm Chaser Accounts

A Message from Dick Elder, NWS Wichita

Suzanne M. Fortin - Perspectives from a New NWS Forecaster

Dr. Keith Ballard - Rebuilding the Oologah School


Radar Imagery

The April 26th, 1991, tornado outbreak occurred in the midst of a period of transition for the National Weather Service as an organization. The National Weather Service Modernization led to reorganization of offices and brought up-to-date technologies into everyday operations. One important part of the modernization was the development of a nationwide network of Doppler radars, known as the WSR-88Ds or NEXRADs. These radars replaced the older WSR-57 and WSR-74 units, neither of which had Doppler capabilities, that had been in use for many years. In April of 1991, the Norman, Oklahoma, WSR-88D was the lone WSR-88D, although it had not yet been cleared for use in everyday operations. WSR-88D provided forecasters at NWS Norman a unique perspective certainly for storms in north central Oklahoma, but even storms as far away as southern Kansas and northeast Oklahoma. However, this also meant that the older technology remained in use at both NWS Wichita (WSR-57) and Tulsa (WSR-74). The Wichita WSR-57 was decommissioned in November 1995 and the Tulsa WSR-74 in April 1995.

In post-event analysis, comparison of the WSR-88D data to that of the Wichita WSR-57 and the Tulsa WSR-74 proved how beneficial the new radar technology would be to warning operations. All Tornado Warnings issued by the Wichita Weather Service Office during the afternoon of April 26th were issued as a result of spotter reports, with none issued solely because of the storm’s appearance on radar. The technology simply did not allow for that in most circumstances. As a result, the first tornado produced by a certain supercell had no warning lead time. Armed with the knowledge that a storm had already produced at least one tornado and continued receipt of good spotter reports, lead times were much higher for succeeding tornado touchdowns. For example, Andover officially had 10 minutes lead time, providing enough time for local officials to compensate for a failed outdoor warning siren. This undoubtedly saved lives in that community. In addition, limitations of the WSR-57 sometimes provided obscured looks at certain storms and storm features. This was the case for the storm that produced the Andover tornado.

Wichita WSR-57 Image - 633 pm CDT April 26, 1991

Wichita WSR-57 Image - 633 pm CDT

Note: The Andover storm is obscured by ground clutter.

The WSR-88D in Norman proved that increased warning lead times – especially for initial tornado touchdowns – would likely result from a new national radar network. Important storm-scale features, including mesocyclones and various reflectivity features, could be clearly seen in storms even great distances from the radar. Despite being 110 nautical miles from the radar, a mesocyclone was clearly visible in the Cowley County storm at the same time it was producing a tornado. However, data from the Wichita radar show no significant features indicative of a tornadic storm, despite being much closer to the storm itself. Using this new technology, NWS Norman was able to issue most Tornado Warnings before any tornadoes touched down as a result of a storm’s appearance on the WSR-88D. Twenty-four minutes of lead time was provided in one case.

Half Degree Reflectivity - Norman WSR-88D - 429 pm CDT April 26, 1991

4-panel Reflectivity - Norman WSR-88D - 633 pm CDT April 26, 1991

4-panel Storm Relative Velocity - Norman WSR-88D - 633 pm CDT April 26, 1991

4-panel Reflectivity - Norman WSR-88D - 844 pm CDT April 26, 1991

0.5 degree Reflectivity - 429 pm CDT

4-panel Reflectivity - 633 pm CDT

4-panel Storm Relative Velocity - 633 pm CDT

4-panel Reflectivity - 844 pm CDT

4-panel Storm Relative Velocity - Norman WSR-88D - 844 pm CDT April 26, 1991

4-panel Reflectivity and Storm Relative Velocity - Norman WSR-88D - 856 pm CDT April 26, 1991

4-panel Reflectivity - Norman WSR-88D - 937 pm CDT April 26, 1991

4-panel Storm Relative Velocity - Norman WSR-88D - 937 pm CDT April 26, 1991

4-panel Storm Relative Velocity - 844 pm CDT

4-panel Reflectivity and Storm Relative Velocity - 856 pm CDT

4-panel Reflectivity - 937 pm CDT

4-panel Storm Relativity Velocity - 937 pm CDT

The performance of the WSR-88D during the April 26th, 1991, tornado outbreak led to the following recommendation in the National Weather Service Assessment of the event:
“The NWS should continue to implement the Next Generation Radar (NEXRAD) network across the Nation. This event illustrates the usefulness of the WSR-88D velocity fields and better azimuthal resolution reflectivity data.”
Today, there are 159 WSR-88Ds in operation across the country and its territories. The national average Tornado Warning lead time is currently 13 minutes, undoubtedly due to the Doppler radar technology and other scientific and technological advances. Upgrades to the WSR-88D network have occurred since its deployment. The upgrades have allowed forecasters to resolve even smaller scale and faster evolving details in storms. Current and future enhancements to the network, including the inclusion of dual polarization technology, are expected to increase detection of certain storm-related phenomena and reduce false alarms. is the U.S. government's official web portal to all federal, state and local government web resources and services.