Observation of Hailstorm with Higher Spatial Resolution Using a Compact Dual Polarimetric X-Band Weather Radar

 

1. Introduction

 

Recently heavy snow and hailstorm in a localized region cause a big damage on our life and assets as well as traffic transport systems such as highways, railways and airports. Observation of hailstorm with higher spatial resolution using a dual polarimetric Doppler weather radar can give us valuable information about classification of hydrometeors. It can give us information such that how much precipitation is expected and how much mixing ratio of rain, snow, hail and so on, it is composed of.

 

2. One Case Study

 

One interesting case study of 14th April 2015 is presented when a cold vortex has passed through the Kagoshima Prefecture. According to the local news reports, a hailstorm was observed near Kagoshima City about 5:00 p.m. on 14th April. The hailstorm was observed using a dual polarimetric weather radar which has been deployed in the west part of Kagoshima. The observed data are processed by multi-parameter analysis. Fig 1 (a) shows the observation area (circle) and the location of Kagoshima in Japan. The blue square of 20km indicates the area focused in this study. The conditions of the observation are indicated in Fig 1 (b).

Fig 1: (a) The observation area and the location of Kagoshima and (b) The conditions for the observation.

 

3. Results

 

Fig 2 indicates horizontal distributions of Zhh, Zdr, RhoHV and Kdp at the elevation of 6 deg inside of square region shown in Fig 1 (a). The following results focus on two square regions, A and B (1km x 1km) indicated in Fig 2. The region A is placed on the area where Zhh is more than 50 dBZ, and the region B is placed on the area where Zhh is ranging between 40 and 46 dBZ.

Fig 2: The horizontal distributions of (a) Zhh, (b) Zdr, (c) RhoHV and (d) Kdp around region A and B. A star-mark indicates a radar site.

 

(1) Analysis using scatter diagrams

The points sampled inside of region A and B are plotted as scatter diagrams in Fig 3 (a)-(c). The points of region A are drawn in red circles and those of region B are drawn in blue circles. As can be seen from Fig 3 (a), regarding Zhh-Zdr the differences between A and B are too small to distinguish each other. On the other hand, Fig 3 (b) indicates that Zhh of region A is higher than those of region B, and that RhoHV of region A is relatively lower than those of region B and distributed between 0.92 and 0.98. Therefore it is possible to roughly estimate that region A might be covered with hail, based on the analysis using the scatter diagram of RhoHV.

Fig 3: The scatter diagrams of (a) Zhh-Zdr, (b) Zhh-RhoHV and (c) Zhh-Kdp inside of region A and B.

 

(2) Analysis using histograms

In general ice particles like hail are expected to have various and non-uniform shapes and have a broader distribution of multi-parameter than water particles like rain. So it is very useful to analyze the spread of multi-parameter. In addition a radar observation with higher spatial resolution enables us to acquire more sample points than that of conventional radars. So it helps us create a histogram with more sample points. Therefore points sampled inside of region A and B are plotted as histograms in Fig 4 (a)-(c). The bar of region A is drawn in red and that of region B are drawn in blue. Fig 4 shows following results.

(a) Regarding Zdr, the points of region B are distributed around 0.5 dB. On the other hand those of region A are distributed between -0.5 dB and 2 dB, and have a broader distribution than that of region B.

(b) Regarding RhoHV, the points of region B are distributed in more than 0.98. On the other hand those of region A are distributed between 0.92 and 0.98, which are lower than those of region B and have a broader distribution than that of region B.

(c) Regarding Kdp, the points of region B are distributed around 0.0 deg/km. On the other hand those of region A are distributed between 0.0 deg/km and 1.6 deg/km, and have a broader distribution than that of region B.

In addition to the analysis using scatter diagrams, Fig 4 (a)-(c) indicates that distributions of region A are relatively broader than those of region B. As a result it is possible to recognize that the region A is mainly covered with hail and the region B is covered with rain, based on the analysis using both scatter diagrams and histograms.

Fig 4: The histograms of (a) Zdr, (b) RhoHV and (c) Kdp inside of region A and B.

 

4. Conclusions

 

One interesting case study of 14th April 2015 is presented when a hailstorm due to a cold vortex is observed in Kagoshima. The hailstorm was observed using a dual polarimetric weather radar.

The two regions, A and B are defined based on the threshold of Zhh, 50 dBZ. A radar observation with higher spatial resolution helps us create a histogram with more sample points. The results of multi-parameter analysis using both scatter diagrams and histograms of Zdr, RhoHV and Kdp indicate that it is possible to recognize that the region A is mainly covered with hail and the region B is covered with rain.

The reasons are as follows. Analysis using scatter diagrams indicates that Zhh of region A is higher than those of region B, and RhoHV of region A is relatively lower than those of region B. Analysis using histograms indicates that regarding Zdr, RhoHV and Kdp the distributions of region A are relatively broader than those of region B.

 

5. Acknowledgement

 

This research is supported by Adaptable and Seamless Technology Transfer Program through target-driven R&D, JST.