KOUN Case Study Radar Beam Attenuation Author: Kevin Scharfenberg, CIMMS/OU Last Updated: 16 January 2003

Introduction

On 15-16 June 2002, a severe mesoscale convective system (MCS) moved from north to south across much of Oklahoma. The thunderstorms along this squall line produced wind damage, torrential rain, and small hail. In a few cases during the event, the NSSL Polarimetric WSR-88D (KOUN) radar beam was attenuated as its energy traversed through a long path of heavy rain and hail.

Two scans from KOUN during this event illustrate how such attenuation may be identified with a polarimetric radar, and how attenuation can affect polarimetric products. In the first case, attenuation was caused by a single large, intense cell. In the second case, the beam was gradually attenuated as the radar scanned down the length of a squall line.

Indentifing beam attenuation

Figure 1 shows a case of radar beam attenuation caused by a single, large, intense storm cell. In the reflectivity image (Figure 1A), a large severe storm cell with reflectivity values approaching 65 dBZ is clearly evident near the eastern edge of the image. It is not clear that beam attenuation is occurring in the reflectivity image alone.

The differential reflectivity image (Figure 1B) offers the first indication of beam attenuation. While most of the rainy region behind the squall line exhibits uniform differential reflectivity values between 0.5 and 1 dB, a sector of values as low as -1.5 dB is present behind the severe cell. This sector of low values appear as a shadow cast behind the echo. Negative differential reflectivity values suggest more vertical energy is being backscatterered than horizontal energy, which is not consistent with the horizontally-oriented shape of falling raindrops. The significant scattering of the horizontal energy by numerous rain drops in the large storm core has reduced the amount of horizontal energy making it to the more distant rain.

The correlation coefficient image (Figure 1C) confirms the large cell is causing beam attenuation. A shadow of low values extends beyond the cell, indicating a region where horizontal and vertical backscattering is out of proportion.

Figure 1: Radar beam attenuation by a single storm.
A: Reflectivity	B: Differential reflectivity	C: Correlation coefficient




Later in the MCS event (Figure 2), the radar beam is aligned with the major axis of the squall line. At the distant end of the line, a strong cell with reflectivity values approaching 60 dBZ is apparent (Figure 2A). Again, attenuation is not readily apparent in the reflectivity image.

The differential reflectivity image (Figure 2B) clearly shows a gradual reduction in differential reflectivity values down the line. By the time the distant cell is reached, the horizontal energy backscattered is much smaller than the vertical backscatter, given the large negative values of differential reflectivity.

The correlation coefficient image (Figure 2C) confirms our suspicion, with a gradual decrease in values with range when looking down the squall line. This shows the horizontal and vertical backscatter fields are indeed becoming out of proportion.

Figure 2: Radar beam attenuation down a squall line.

A: Reflectivity

B: Differential reflectivity

C: Correlation coefficient

Side effects of beam attenuation

Beam attenuation impacts fields derived from the polarimetric products. In the case of the single storm cell (Figure 3), the specific differential phase image (Figure 3A) shows a shadow of missing values in the beam attenuation region. This is because echoes associated with low correlation coefficient are usually not associated with precipitation, so the algorithm which produces the specific differential phase image does not consider such poorly correlated echoes.

The hydrometeor classification algorithm (Figure 3B) also contains some areas of missing values beyond the severe storm's core. This is because the algorithm relies on all of the polarimetric variables (as well as reflectivity) to determine the particle type detected. If several of those fields produce output which are not consistent with the other fields, the algorithm will not know which particle type to assign and will default to "none".

Figure 3: Radar beam attenuation by a single storm.
A: Specific differential phase	B: Hydrometeor classification algorithm

In the case looking down the squall line (Figure 4), the specific differential phase (Figure 4A) and the hydrometeor classification algorithm (Figure 4B) gradually show an increasing number of missing values down the line. Note both images have no output for the large cell at the end of the line.

Figure 4: Radar beam attenuation down a squall line.

A: Specific differential phase

B: Hydrometeor classification algorithm

Conclusions

KOUN reflectivity products alone did not reveal the presence of radar beam attenuation in this case. On the other hand, the polarimetric products did aide in the identification of the attenuated regions. Since the attenuation is caused by rain, the horizontal signal degrades more quickly than the vertical. The result is differential reflectivity values that become more negative as the beam is attenuated. The increasingly different backscattering characterics between the vertical and horizontal pulses also results in a decreasing correlation coefficient in the attenuating region.

Products derived from the polarimetric fields were adversely affected by the attenuation. Specific differential phase values may not be computed where the correlation coefficient values fall too low to be classified a usable signal. The hydrometeor classification algorithm's fuzzy logic scheme may fail to interpret the inconsistent signals and produce no output.

It is also important to note that attenuation negatively effects all precipitation estimation algorithms, both the traditional reflectivity-based algorithm and the new, polarimetric-based algorithms. However, the improved methods for detecting attenuation with polarimetric products will now allow the development of techniques to adjust the precipitation accumulation algorithms in cases of beam attenuation.

Related Links

• National Severe Storms Laboratory (NSSL)
• JPOLE: Joint POLarimetric Experiment
• NSSL Polarimetric Radar Page

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!