The impact of elevation scan spacing on observations of heat bursts sampled by the National Weather Radar Testbed Phased Array Radar

As part of the 2009 Phased Array Radar (PAR) Innovative Sensing Experiment (PARISE), a customized scanning strategy was developed using an unusually large number of elevation scans. This “dense sampling” strategy is intended to provide a large amount of vertical detail for analyzing hailstorms and other events. One such event is a heat burst, which is defined as a region of air that experiences significant warming as it descends rapidly. Heat bursts can create localized temperature spikes and dew point depressions of 10°C in as little as 15 min. People sensitive to heat may be threatened by this sudden change in temperature. Wind gusts of 25 m s^-1 or higher may also occur, leading to potential property or agricultural damage. Because rapidly descending air precedes heat bursts, it is likely that analyses of detailed radar observations in elevation will lead to better understanding of how and when they may occur.

In the early morning hours of 13 May 2009, a series of heat bursts was observed by approximately 70 Oklahoma Mesonet stations. Rawinsonde data obtained from Norman, Oklahoma at 0000 UTC 13 May 2009 indicate a nearly dry-adiabatic lapse rate and relatively dry air between 750 and 400 mb. As thunderstorms moved into the region and dissipated, precipitation evaporated aloft and generated descent. The lapse rate allowed the descending air to increase in velocity and warm rapidly, resulting in heat bursts that spread over a large area. At 0320 UTC, a 15-min temperature increase of 8.0°C, dew point depression of 6.7°C and maximum wind gust of 23 m s^-1 were observed in southwestern Oklahoma. Seven hours later, another 15-min temperature spike of 6.0°C, dew-point depression of 6.0°C and maximum wind gust of 15 m s^-1 were observed in north-central Oklahoma. The other Mesonet stations reported weaker heat burst activity over this time period.

During this event, the NWRT PAR performed scans for a 2.5-hr period using the dense sampling strategy. To evaluate the usefulness of the closely spaced elevation scans, it is necessary to compare these results with those from a scan with fewer elevations. For this study, individual elevations will be selectively removed from the PAR dense sampling data to form a second “sparse” scan. Vertical cross-sections will be constructed at selected Mesonet sites, in order to demonstrate the detail obtained from both scans. Vertical reflectivity profiles will also be produced from both scans in order to provide a quantitative comparison of heat burst observations. The analysis will provide insight into which radar features are significant in analyzing and predicting heat burst activity. In addition, the results will provide guidance on how to implement PAR scanning when a large amount of vertical detail is necessary.