Reprint 531 利用廣東省自動氣象站網絡實時資料 增強香港暴雨臨近預報的能力 李炳華, 周志堅, 楊漢賢 第十八屆粵港澳氣象科技研討會, 中國, 香港,2004 年 2 月 16-18 日
Using the Real-time data of the Guangdong Automatic Weather Station Network to Improve Rainstorm Nowcasting in Hong Kong PW Li, CK Chow and HY Yeung Hong Kong Observatory Abstract Using real-time surface rainfall data to calibrate the radar rainfall estimation is one of the key steps in rainstorm nowcasting. The Hong Kong Observatory s nowcasting system SWIRLS has been employing local rain-gauge data to calibrate the rainfall intensities of all echoes within the radar range. This approach is quite effective for the radar echoes over the territory but could in some circumferences provide inaccurate rainfall estimation for echoes falling beyond Hong Kong. We discuss in this study the ways to employ the real-time rainfall data provided by the Guangdong Automatic Weather Station Network into SWIRLS, so as to see if the radar rainfall estimation and thereby the rainfall nowcast in Hong Kong could be improved.
1. 1999 2003 500 2. 2.1 Z-R Marshell Palmer Z=200R 1.6 NexRAD Z=300R 1.5 Z-R 2
Z R : (1) Probability Matching [1]; (2) Least Square Matching [2] 0 dbz n dbz 0 m n dbz m Z=aR b ab G R R-G ab min N ( dbr i dbg i ) i= 1 dbg=10log(g) dbr = ( 10log( a) dbz )/ b N ( ) ( ) ab ab [3][4] Z R ab [5] 50 50100 100150 150200 200250 ( ) 50 3 2
50100 100 150 150200 250 2.2 (TREC) TREC 2.1 Z-R [3][4] 3. 256 1 CAPPI 130 (1) 2002910 (2) 200392 (3) 200355 ( ) 4. 4.1 2002 9 10 18:00 21:00HKT 20039220:0023:00HKT 200355 01:0008:00HKT 200355 (e) ( 511 (f) ( 979 ) 0.890.96 4
5 0.660.57 0.73 0.740.75 0.78 200355 01:00 02:00 (a)(c) Z R(b)(d) ( 4.3 ) 4.2 Z R ab 55 ab ab 55 a01:00hkt200 04:00HKT235 b 1.5 1.4 01:0002:00HKT ab ( ) Z R a01:00hkt 267 02:00HKT270 04:00HKT203 b 1.3 1.5 ab ( :01:00HKT) b ( ) Z R (b)(d) ( a b b ) ab 55 ab a (6%) b (13%) b b 55 100 Z R
150 4.3 200355 (a)(b) 02:0003:00 Z R (c)(d) (e)(f) ( ) ( ) a b (4.2) 5. Z R [1] Rosenfeld, D., D.B. Wolff, D. Atlas, 1993. General probability-matched 6
relations between radar reflectivity and rain rate. J. Appl. Meteor., 32, 50-72. [2] Zawadzki, I., C. Desrochers, E. Torlaschi, 1986. A Radar-raingauge Comparison. Proceedings of the 23 rd Conf. on Radar Meteorology, Snowmass, Colorado, Amer. Meteor. Soc., 121-124. [3] Li, P. W., W. K. Wong, K. Y. Chan and Edwin S. T. Lai, 2000: SWIRLS An Evolving Nowcasting System, Technical Note, No.100, Hong Kong Observatory. [4] Li, P.W. & E.S.T. Lai (2004) Short-range Quantitative Precipitation Forecasting in Hong Kong. J. Hydrol., to appear. [5] Ciach, G.J., W.F. Krajawski, E.N. Anagnostou, M.L. Baeck, J.A. Smith, J.R. McCollum and A. Kruger, 1997. Radar Rainfall Estimation for Ground Validation Studies of the Tropical Rainfall Measurement Mission. J. Appl. Meteor. 36, 735-747. 7
Z-R (HKT) 10 9 2002 18:00 21:00 232 58 ( ) 2 9 2003 20:00 23:00 294 376 ( ) 5 5 2003 ( ) 01:00 08:00 511 468 a b 18:00 19:00 20:00 21:00 a b a b a b a b 200 1.6 312 1.4 295 1.3 303 1.3 + 200 1.6 315 1.4 298 1.3 306 1.3 a b 20:00 21:00 22:00 23:00 a b a b a b a b 200 1.6 250 1.4 226 1.4 222 1.5 + 200 1.6 235 1.4 247 1.4 201 1.5 a b 2003 5 5 01:00 02:00 03:00 04:00 a b a b a b a b 201 1.5 201 1.5 205 1.5 235 1.4 + 267 1.3 270 1.3 270 1.3 203 1.5 a b 2003 5 5 01:00 HKT ( ) a b < 50 202 1.5 50 100 213 1.4 100 150 214 1.4 150 200 214 1.3 200 250 214 1.3 8
Z-R () 9
(a) (c) (b) (d) (e) (f) ( ) () () (a)(b) 2002 9 10 18:00 21:00HKT (c)(d) 2003 9 2 20:00 23:00HKT (e)(f) 2003 5 5 01:00 08:00HKT 10
(a) (b) (c) (d) 2003 5 5 ( ) () Z-R () (a)(b) 2003 5 5 01:00 02:00HKT (c) (d) 02:00 03:00HKT 11
2:00 HKT 3:00 HKT (a) (c) (b) (d) (e) (f) 2003 5 5 (a)(b) 02:00 03:00 (c)(d) (e)(f) 12
18:00 HKT 19:00 HKT (a) (b) (c) (d) (e) (f) 2002 9 10 (a)(b) 18:00 19:00 (c)(d) (e)(f) 13
20:00 HKT 21:00 HKT (a) (b) (c) (d) (e) (f) 2003 9 2 (a)(b) 20:00 21:00 (c)(d) (e)(f) 14