PolyU Micro-LARGE Observing Water Vapor (WV) Variation During Heavy Precipitation Events in Hong Kong by GPS Tomography Biyan Chen 1 , Zhizhao Liu 1 , Wai-kin Wong 2 , and Wang-chun Woo 2 1 Department of Land Surveying and Geo-Informatics The Hong Kong Polytechnic University 181 Chatham Road South, Hung Hom, Kowloon, Hong Kong E-mail: by.chen@connect.polyu.hk/lszzliu@polyu.edu.hk 2 Hong Kong Observatory, 134A Nathan Road, Kowloon, Hong Kong WMO WWRP 4th International Symposium on Nowcasting and Very-short-range Forecast 2016 (WSN16), 25-29 July 2016, Hong Kong
Outline PolyU Micro-LARGE Research background Principle of WV tomography Observing WV variation during heavy rain Case study 1: 22 July 2010 Case study 2: 21 May 2013 Case study 3: 30 March 2014 Conclusion 2
Significance of WV in meteorology PolyU Micro-LARGE Weather phenomena cloud, drizzle, rain, snow, sleet, and hail Latent heat release source of atmospheric energy transportation fuel for storms Water vapor distribution and variation Water vapor variation evolution of atmospheric storms during heavy rain the vertical stability of atmosphere 3
Climate in Hong Kong PolyU Micro-LARGE Highly populated and extremely humid coastal city Hong Kong Average humidity: 78% Annual rainfall: 2400 mm More extreme weather events occur in recent years Landslip and urban waterlogging To enhance the monitoring of atmospheric WV 3D WV fields from the tomography 4
WV observation techniques PolyU Micro-LARGE large spatial coverage Aerosol Robotic Microwave Water relative low accuracy NETwork Vapor Radiometer (AERONET) (WVR) Satellite remote sunphotometer sensing high temporal resolution good accuracy unavailable rain only works with direct sunlight or heavy fog Atmospheric Global Navigation Water Vapor Satellite System Observation (GNSS) low operational cost high accuracy Numerical weather all-weather operability Radiosonde prediction (NWP) high accuracy high spatial resolution model low temporal resolution Relative low accuracy 5
Outline PolyU Micro-LARGE Research background Principle of WV tomography Observing WV variation during heavy rain Case study 1: 22 July 2010 Case study 2: 21 May 2013 Case study 3: 30 March 2014 Conclusion 6
Principle of WV tomography PolyU Micro-LARGE GNSS satellites Tomography technique enables us to precisely probe the atmosphere: 3D WV distribution Voxel Under all weather conditions With high temporal and spatial resolution Inversion process: Slant wet delay (SWD) Wet refractivity troposphere GNSS receivers 7
GNSS signal delayed by troposphere PolyU Micro-LARGE Zenith tropospheric delay (ZTD): ~90% Zenith hydrostatic delay (ZHD) + Zenith ~10% Zenith wet delay (ZWD) Tropospheric Delay Saastamoinen ZHD model: Pressure Latitude Height 8
Estimation of WV from GPS obs PolyU Micro-LARGE GPS data processing GPS data processed by Bernese 5.0 software Zenith tropospheric delay (ZTD), wet delay gradients and the residuals Parameters are estimated once an hour The slant wet delay (SWD) can be derived from: ∂ f = − ⋅ + ⋅ ⋅ φ + ⋅ φ + SWD ( ZTD ZHD ) f ( z ) ( G cos( ) G sin( )) R ∂ N . W E . W z zenith post-fit hydrostatic residuals Wet delay delay satellite zenith gradient in the azimuth distance northern and angle eastern directions 9
GPS stations in Hong Kong PolyU Micro-LARGE 22.6 Shenzhen HKFN 22.5 Geographical HKKT HKSS HKLT distribution of HKWS HKST 22.4 HKSL the 12 GPS Hong Kong Latitude ( ° N) stations used HKSC Radiosonde in this study 22.3 HKPC HKNP HKOH HKMW 22.2 22.1 113.8 113.9 114.0 114.1 114.2 114.3 114.4 Longitude ( ° E) 10
Principle of WV tomography PolyU Micro-LARGE satellite Length of ray in voxel 2 Wet refractivity in voxel 2 = A ⋅ y x G G receiver 11
Tomography equation PolyU Micro-LARGE y A GPS G G y A NWP N N y A Sun photometer A A = ⋅ y A x Microwave radiometer W W y A Synoptic observations Solved by using s s Least-Squares y A Vertical constraint R R method 0 H Horizontal constraint More information about the tomographic modeling, please refer to: Chen, B. and Liu, Z.: Assessing the Performance of Troposphere Tomographic Modeling Using Multi-Source Water Vapor Data During Hong Kong’s Rainy Season from May to October 2013, Atmos. Meas. Tech. Discuss., doi:10.5194/amt-2016-158, in review, 2016. Chen, B. and Liu, Z.: Voxel-optimized regional water vapor tomography and comparison with radiosonde and numerical weather model, J. Geod., 88(7), 691–703, doi:10.1007/s00190-014- 12 0715-y, 2014.
PolyU Micro-LARGE Evolution of water vapor field over Hong Kong solved by tomography with an interval of 30 min 13
Outline PolyU Micro-LARGE Research background Principle of WV tomography Observing WV variation during heavy rain Case study 1: 22 July 2010 Case study 2: 21 May 2013 Case study 3: 30 March 2014 Conclusion 14
Case study 1: 22 July 2010 PolyU Micro-LARGE 2900 houses destroyed in Guangdong province 23 July 2010 economic losses ~ US$ 350 million in Guangdong 22 July 2010 province 20 July 2010 Heavy rain in HK 21 July 2010 UT 07:00~10:30, 22 July 2010 ( LT=UT+8 ) 19 July 2010 rainfall of 169 mm 150 mm (UT 08:30 ~10:30) Track of the Typhoon Chanthu （颱風燦都） 15
Meteorological situation UT 06:00, 22 July 2010 (a) 22/07/2010 UTC 06:00 500 hPa (b) 22/07/2010 UTC 06:00 850 hPa 30 30 50 0 75 0 4 80 70 75 5890 0 8 PolyU Micro-LARGE 60 7 80 85 0 7 7 1460 1510 0 80 80 1480 7 1450 1470 1490 1 70 1530 80 5880 80 5870 5 5 80 8 6 0 5 60 0 50 0 0 0 0 6 1520 8 80 8 7 6 70 70 40 0 0 8 70 5 5 75 0 0 7 7 60 8 6 Rain: UT 07:00~10:30, CHINA CHINA 60 5 60 80 8 7 5 6 0 1510 5 7 90 0 25 25 5860 70 9 5 8 5 5 8 0 5850 8 0 95 80 85 90 22 July 2010 Latitude ( ° N) Latitude ( ° N) 8 60 5870 70 8 0 8 0 80 5 4 8 90 5 80 80 90 9 80 9 1450 95 0 85 0 3 8 5 0 1510 5820 95 Hong Kong Hong Kong 95 5 5810 8 70 8 Chanthu 85 5 90 8 85 0 Chanthu 90 60 0 0 0 9 1440 85 90 9 3 1460 9 0 80 4 85 1 1450 1470 0 8 20 20 9 1 0 1 80 5 5 8 4 2 0 8 5 90 1480 1490 0 90 0 80 0 0 0 85 0 9 5870 5 8 3 0 7 8 80 80 85 • Black contours 80 0 85 80 90 0 85 5840 1460 9 75 8 9 0 90 80 8 0 5 0 8 0 8 90 5850 SOUTH CHINA SEA SOUTH CHINA SEA 85 represent the 5 1470 90 8 80 80 90 85 7 90 8 5 8 5 8 80 70 0 90 0 5 0 85 85 0 9 65 8 8 15 15 104 108 112 116 120 124 104 108 112 116 120 124 geopotential heights. Longitude ( ° E) Longitude ( ° E) (c) 22/07/2010 UTC 12:00 500 hPa (d) 22/07/2010 UTC 12:00 850 hPa 30 30 80 70 80 6 0 70 70 75 75 65 70 70 70 85 1 1530 7 75 0 0 90 70 • Coloring contours are 5890 1 75 6 5 0 5 7 1 1500 7 70 70 5 8 9 0 4 80 3 7 5 4 65 75 50 70 70 70 1500 65 4 90 0 5 0 1490 0 0 70 4 9 1 0 6 70 85 0 5 80 80 5880 50 0 65 7 8 50 85 75 4 80 6 0 0 0 7 8 5 1480 8 9 1520 60 1470 5870 0 0 40 6 5880 6 0 85 40 1470 1520 6 80 0 1510 1510 4 4 0 1 8 80 8 0 0 7 6 5 4 5870 0 0 0 5880 7 1 5 relative humidity, 7 5 5 5 75 8 90 7 4 60 60 0 CHINA CHINA 1450 80 7 6 80 0 0 50 80 5860 50 80 7 0 70 5860 5 0 8 7 5 5 5 60 0 1450 5 7 1450 9 0 0 90 0 60 4 75 60 1 80 70 60 60 85 95 80 80 70 85 0 9 25 5850 25 80 0 8 90 95 0 5 1 80 which are marked in 5 85 5850 90 7 85 85 0 0 5 9 5840 0 80 70 90 85 85 85 Latitude ( ° N) Latitude ( ° N) 0 5 0 100 5840 9 100 95 90 60 0 1 90 4 9 70 4 90 95 6 70 70 85 1 1450 85 0 80 1 1510 0 1440 70 3 1430 4 1470 5 1440 8 0 9 0 95 6 8 5 2 5810 60 5 9 0 80 4 95 1440 red 8 9 0 90 Hong Kong 0 Hong Kong 0 5 9 5830 2 1470 9 Chanthu 0 Chanthu 8 8 60 0 90 90 1460 8 0 5 5 5 8 1450 7 90 0 95 85 85 70 80 8 0 8 0 90 80 3 90 85 8 80 8 9 0 0 80 80 0 8 0 0 0 0 8 8 90 5880 1430 8 5 85 5820 7 8 0 70 85 5 20 20 80 70 80 0 0 8 0 3 70 0 90 8 80 • Data are from ECMWF 5 5870 85 9 8 80 1440 0 1500 80 0 1 4 4 0 8 5860 1490 1490 8 0 5850 5860 1480 0 7 7 0 75 70 8 75 85 5870 85 9 1450 90 0 70 0 0 0 5840 0 75 80 9 5 5 0 8 5 90 5840 4 1500 8 8 1 4 7 60 9 60 60 80 85 1460 85 5 5 80 1 5 0 90 0 8 75 8 8 75 8 85 95 SOUTH CHINA SEA 0 8 70 7 SOUTH CHINA SEA 5 5 0 1 9 4 6 0 75 1470 1470 90 8 0 80 85 8 80 8 70 0 0 80 0 90 90 8 85 85 80 8 0 9 5 5 6 5 0 6 9 15 15 104 108 112 116 120 124 104 108 112 116 120 124 Longitude ( ° E) Longitude ( ° E) UT 12:00, 22 July 2010 16
Recommend
More recommend
