Quality control of weather radar data by using dual-polarization

Transcription

1 WMO/ASEAN Training Workshop on Weather Radar Data Quality and Standardization Quality control of weather radar data by using dual-polarization 7 February 2018 Hiroshi Yamauchi Observation Department Japan Meteorological Agency Japan Meteorological Agency Bangkok, Thailand, 5-13 February 2018

2 WMO/ASEAN Training Workshop on Weather Radar Data Quality and Standardization Contents Ground clutter removal Sea clutter removal Clear air echo removal JMA s practical case Japan Meteorological Agency Bangkok, Thailand, 5-13 February 2018

3 Removal of ground clutter (GC) MTI filter cannot fully remove GC MTI filter may degrade data. Normal MTIed Mountains Power lines Buildings 3

4 Clutter removal (MTI filter) MTI filter : moving target indication (MTI) is one of the method to mitigate the clutter and identify moving targets It removes and interpolates the power component around 0 m/s in velocity-power space (periodogram) Normal Power Ground Clutter MTIed Power Remained GC 0m/s Doppler velocity 0m/s Doppler velocity Met. echo Power Met. echo Power 0m/s Doppler velocity 0m/s Doppler velocity Suppressed met. echo 4

5 Removal of ground clutter (GC) Φ DP is spatially fluctuating in ground clutter regions. Φ DP is spatially smooth in precipitation echo. Z Fdp Mountains Power lines Buildings 5

6 Removal of ground clutter (GC) GC can be efficiently identified and removed using standard deviation of Φ DP. Mountains Power lines Buildings 6

7 Removal of ground clutter (GC) GC can be efficiently identified and removed using standard deviation of Φ DP. Mountains Power lines Buildings 7

8 Removal of ground clutter (GC) GC can be efficiently identified and removed using standard deviation of Φ DP. Mountains Power lines Buildings 8

9 Removal of sea clutter (SC) SC can be efficiently identified and removed using r hv and standard deviation of Φ DP. Mountains Sea clutter Precipitation 9

10 Removal of sea clutter (SC) SC can be efficiently identified and removed using r hv and standard deviation of Φ DP. Mountains Sea clutter Precipitation 10

11 Removal of sea clutter (SC) SC can be efficiently identified and removed using r hv and standard deviation of Φ DP. Mountains Sea clutter Precipitation 11

12 Removal of sea clutter (SC) SC can be efficiently identified and removed using r hv and standard deviation of Φ DP. Mountains Sea clutter Precipitation 12

13 Removal of clear air echoes Biological echoes and chaffs are efficiently identified and removed using Zdr, r hv and standard deviation of Φ DP. Precipitation Biological Chaffs 13

14 Removal of clear air echoes Biological echoes and chaffs are efficiently identified and removed using Zdr, r hv and standard deviation of Φ DP. Precipitation Biological Chaffs 14

15 Removal of clear air echoes Biological echoes and chaffs are efficiently identified and removed using Zdr, r hv and standard deviation of Φ DP. Precipitation Biological Chaffs 15

16 JMA s practical case for QC with dual-pol parameters Clutter removal Standard deviation of Φ DP ( S(Φ DP )) is used Clear air echo removal Zdr and r hv are used Textures are not used but it seems to be useful 16

17 Clutter removal 17

18 Clutter removal Start No Other threshold for GC No yes Other threshold for GC No yes S(Fdp(NOR))>10 No Meteorological echo Normal data yes S(Fdp(MTI))>30 No Contaminated by GC Clutter filtered data yes Contaminated by strong GC Invalidated End Normal :S(Φ DP ) <10 Clutter filtered : 10< S(Φ DP ) <30 Probability density distribution of S(Φ DP ) Invalidated : S(Φ DP ) > Sugier and Tabary (2006)

19 Clear air echo removal START Attenuation correction Input data ρhv, Zhh, Zdr, Kdp, Φdp Zhh Zhhc Zdr Zdrc No Zhhc > < Zdrc < 7.0 ρhv > 0.7 Yes Kdp 0.6 Zhhc 38 No Yes Clear air echo removal Note: this threshold is for C-band R(Zhhc) R(Kdp)* NaN END * : R(K DP ) = 129(K DP / f)

20 Clear air echo removal (ex.) Courtesy of Mr. Umehara Zhh Courtesy of Mr. Koike It seems all parts of this echo are meteorological echo but 20

21 Clear air echo removal (ex.) No Zhhc > < Zdrc < 7.0 ρhv > 0.7 Yes RR (mm/h) NaN R(Zhhc) Courtesy of Mr. Umehara 21

22 Clear air echo removal (ex.) Most of clear air echo can be removed Clear air echo might be still remained No threshold RR (mm/h) B A RR (mm/h) B Courtesy of Mr. Umehara Zhhc > < Zdrc < 7.0 ρhv > 0.7 A 22

23 Summary Ground clutter removal S(Fdp) is useful Sea clutter removal S(Fdp) and rhv are useful Clear air echo removal S(Fdp), rhv, Zdr and S(Zdr) are useful JMA s practical case 23