KOUN Case Study 18 September 2002: Large hail Author: Kevin Scharfenberg, CIMMS/OU Last Updated: 13 January 2003

Introduction

On 18 September 2002, numerous thunderstorms moved across Oklahoma. Several of these thunderstorms became severe, and prompted numerous reports of hail. The NSSL polarimetric WSR-88D radar (KOUN) operated throughout the event, and detected hail signatures in a number of storms. Three thunderstorms sampled by the KOUN radar on 18 September are compared, illustrating the power of polarimetric radar in the detection of hail.

Giant Hail-producing Supercell Thunderstorm

Figure 1 shows KOUN data taken as a thunderstorm produced hailstones the size of softballs, 4.5 inches in diameter. The reflectivity image, Figure 1A, shows a region of reflectivity between 60 and 65 dBZ at the core of the storm. Figure 1B, the differential reflectivity (Z_DR) image, shows a classic "Z_DR hole", or local minimum in the differential reflectivity field, that coincides with the location of the reflectivity maximum in 1A. This "hole" marks where spherical hydrometeors (hail) is surrounded by horizontally oriented hydrometeors (rain).

Figure 1C, specific differential phase (K_DP), shows little differential phase shift between the horizontal and vertical radar pulses near the storm's core. This indicates there is little or no rain present in the hail region. Farther north and northeast, away from the hail core, K_DP values near 2 deg/km mark a region of moderate to heavy rain.

In figure 1D, a large region of correlation coefficient values between 0.95 and 1 indicates the region of uniform rain hydrometeors. Closer to the storm's core, correlation coefficient values decrease sharly to near 0.6. Values this low, in area suspected of containing hail, likely indicate resonance (Mie) scattering is taking place.

Figure 1E, the hydrometeor classification algorithm, shows the hail region in the expected location, with rain detected in the surrounding region.

Exhaustive studies relating polarimetric radar data to hail size have not yet been attempted. However, the well-defined nature of the "Z_DR hole", as well as the possibility of resonance scattering, coincident with very high reflectivity values, can lead the forecaster to a high confidence forecast of large hail in this supercell thunderstorm.

A: Reflectivity (Z)	B: Differential Reflectivity (ZDR)	E: Hydrometeor Classification Algorithm (HCA)
C: Specific Differential Phase (KDP)	D: Correlation Coefficient (pHV)

Figure 1: Supercell thunderstorm producing softball-size hail.

Marginally Severe Hail-producing Thunderstorm

A second severe thunderstorm sampled by KOUN (Figure 2) was producing much smaller hail, 0.88 inches in diameter. Figure 2A, the reflectivity image, shows a storm core with maximum reflectivity approaching 60 dBZ. The differential reflectivity image, Figure 2B, shows another local minimum in the Z_DR field is associated with this maximum in reflectivity. Note that this storm is farther from the radar than the first storm discussed, so we are slicing through this hail core at a higher altitude.

In Figure 2C specific differential phase (K_DP) values in the storm's core are large, approaching 3 deg/km. While the Z_DR image suggests spherical hydrometeors (hail), the K_DP image suggests a considerable amount of hydrometeor content in the horizontal (rain). Taken together, this suggests a mixture of rain and hail is present at this altitude in the storm's core.

The correlation coefficient, Figure 2D, is around 0.95 in the storm core, and is generally greater than 0.95 farther east. The presence of a mixture of hail with the rain in the core lowers the correlation coefficient from the higher values in the pure rain regime. Figure 2E, the hydrometeor classification algorithm, confirms the presence of hail in the storm's core with a surrounding region of rain.

A: Reflectivity (Z)

B: Differential Reflectivity (ZDR)

pHV)

Figure 3: Non-severe thunderstorm.

Conclusions

Using KOUN data, forecasters were able to determine in real time the relative hail threats from these thunderstorms. Strongly-worded warnings were issued for the supercell, a warning was in effect for the marginal storm, and no warning was issued for the hail-free storm. At a future date, these principles may be used to significantly improve the WSR-88D Hail Detection Algorithm. In addition, automatic classification of hail versus rain can significantly improve Quantitative Precipitation Estimation (QPE) algorithms, by removing hail contamination.

Related Links

• National Severe Storms Laboratory (NSSL)
• JPOLE: Joint POLarimetric Experiment
• NSSL Polarimetric Radar Page
• Hondl, K.D., 2002: Current and planned activities for the Warning Decision Support System - Integrated Information (WDSS-II). Preprints, 21st conference on severe local storms, San Antonio, TX, American Meteorological Society, 146-148.

Acknowledgements

This case study would not have been possible without the dedicated work of the scientists who maintain, operate, and collect data from KOUN radar. Thank you!