適當氣候平均統計模式應用於台灣地區溫度預報

Size: px

Start display at page:

Download "適當氣候平均統計模式應用於台灣地區溫度預報"

继裘
4 years ago
Views:

1 h t! 87 óq ì u p } Ô x ¹ r Õ v Œ ½ Õ v Œ n½ î õ v š(optimal Climate Normals Â½ OCN) Š (K v)³v žå ³Â fˆš Š ë î Ê ëp 1994 vo OCN f w w Ð Œˆo Ä OCN ˆšo Œv Ÿ³ žî ½³ ç ö n Á OCN o ž¾ÿv³ f á o ä ž¾ìå OCN 11 Œ³ f á i K=7 õ ³ö OCN ³ f á ä ÐˆŸvŒ ³Æz î Ì šœõœ 5-7 v³ ë 10 Œ ç v Ÿv 11 Œt h 5-7 v³ ë ò Ðˆ OCN f á 11 Œ³tï h Œ ½ öð ž w³ Ë i w ž 11 Œ Ÿ³ áx Î ˆ OCN w páºæ w Œ žð Ÿv³Î ýæ õ v f Åå

2 88 æ Í h šäå h fv³ ç Ä õ õ h³ Œh f Ú Æ}é ³h (External Forcing)x h w fš Ë Ìm ë (ENSO) á ÌxË Madden Julian Oscillation(MJO)(Madden and Julian, 1971;1972) Quasi-Biennial Osciallation (QBO) (Ebdon, 1963) á ôv æh ½ º h Ñœ n õœ v ³v é hv Œ Pacific Decadal Oscillation(PDO)(Mantua et al. 1997) Ì hä Œ Atlantic Multi-decadal Oscillation(AMO)(Kerr 2000) õ hpv³é yo e³ Œ x?î Ú õ eh Œ f? ënoð o ³ Œh ž³näˆšœåå Å å h (Troccoli et al. 2008) Ååš ½ Å (empirical)š(van den Dool, 2007), õ tå³ ³ ë žeº žˆš ˆ w ät ³ žˆš(hoskins et al. 2008) ÌÅå ˆ w žt Ååˆš ˆ w³ ž ³ ž w(mason 2008) w åà ³õ äžìˆ³ëo äž á ³³õ Åå žt f á³p³oµ z f É³ ft (van den Dool 2007) v š (climatology) Èš (persistence) eœ³åå ˆš v š 30 väžìˆ³v ž i ˆáŠxx ³v Ÿ ùìå n á o ËÌåõ ìå ˆ ž Œ u ³ Ë ëä Ç(WMO)Ä 100 v (van den Dool, 2007) n 30 vv Ÿv (climatological standard normals) v ˆ (WMO 1979) (2009)v Ÿv v³v v WMO šœ è³ ç n 30 v Ÿ ç vˆîhîœ áz Œ Œäžìˆ ³ž¾ˆ Œ ³hv 30 v³ö h Ð á³ˆš âƒ xw Ð Èš ½ Š ³äž x Š v¼³ žš h ÄÅhoŒ Èz ÈšŒŠ ³nÁ ö Œ ³ fÿ¹v v x ë õ ³ WMO ³ Ÿv š³ f ³eŒö ˆ Š ë h v Œ Ÿ ˆ v v ³vˆõ ˆ ˆ v i v ž¾ Œv Ÿ 2010 v Œ Ÿ³ f Îõ i vv ŸŸ¹ Ê ëp ž y(climate Prediction Center, Â ½ CPC)Ð 1994v j ž ô 10v³v ná f³å hv ³ ná Ÿv³ få u á ˆ õ³åå ˆš Huang et al. (1996) 10v Ëù Ê 344 ž¾³ná ã ¼v x ˆ Ì n³ n š oá e~æë} ³ z ö õ v (K)³ v «x ³ žx ƒ å K p õ v š(optimal Climate

3 h t! 89 Normals Â½ OCN)y n³ ý ˆ OCN žš y šá ³ ½ Œh že Œˆš OCN ³ ž á ã f ˆš(i ô š)³eÿ ƒå i ˆš³ ž Ÿp OCN ä šœ á ù x nòh ½ pvòo Ÿv õ v š«xœ á³ žå xã œ opv h ÄÅ h õ v š³ Ë Œ Œˆ³ ³ n Ÿ³õ v v K ž¾ìˆ Ð ÒØ ž¾³ÿvìˆ Ë} Ÿvõ v v ná ½ y ¾o Äì ˆÑåÀˆš ¾o ÄŸv Œ ž õ v š³ žå Ñæù ¾ Œˆ ³Åç i¾ Œˆ ³ìˆŒ h ëp Ð ÒØž¾ võœv Ÿv ìˆ ˆ Ÿ ùœ ò ³opvé õ ž¾ ³ áž¾ OCN ³ ä y õ ìˆ o õ³vˆ K v K v¹n x K v³œn v ³ f K ³ nˆši åœ Š v T i = 1, 2, 3, N N áìˆ³ç vˆ õ Š v y i xx k Š ³v T i, k wáº T k 1 = T ik, i j k j= 1 k = 1, 2,, 30 ; i = 31, 32,, N v³ìˆ ç N = 66 fˆš y f áx Ì OCN ³ f õ ˆv³v åàìx ž¾ ³ ìˆ K = 1 Ð 30 pˆ Ä ìˆ Œ ½ v w¾ w³ ìˆä oõ Œ f ˆ w³åæœ (training period) v fœ(target period) fœ³ v(1971)³ K = 1 Ð 30 f v³ìˆåà ³ k n 1 Ð 30 v ˆ k îõ 30 x šœ ³ f á(huang, J. et al; Wilks, D. S. 1996) K = 30 ˆ v fš OCN ³ f á ËeŸ K ³ö åà k f T i, k ÑäžŠ v ïv ˆÑ ˆ çt x ³ ˆ o³ ˆ çt k Ä õ ³ v v n³ K ïv ˆˆ¼wi COR dep ( K ) ˆ f, = Ti Ti k CWMO ˆ ob ob Ti = Ti CWMO n f ob ˆ ˆ Ti Ti i = 31 = n n f 2 ob Ti ˆ ˆ [ ( ) Ti i= 31 i= / 2 ) ] õâ C WMO áº WMO në³ v 10 v ˆ ³ 30 vv p 10 v ï³ 30 vÿ v 2009 v C WMO vv o 1991 vì å C WMO v³v T ˆ f i i

4 90 v fïv T ˆ ob i väžìˆåà³ï i v ˆ Ñ ˆ çt ö õ³ K v x æ K vx få õ f á³ä Œˆƒ åà (3:4:3) f³ Heidke Skill Score (HSS) áx æ Ÿ(Wilks, D. S, 1995) HSS f æùeÿ Ó HSS 0 á f áñ p á(no Skill) Ó HSS 1 nê f(perfect Skill) h w¾ Ä ˆìˆõá OCN f æ o OCN f áx ³ æ Ë få º K ˆ 1-10 ³ HSS hä 0.2 t Ì v Ä 0.1 OCN f áx y hˆ 0.2À Œ á ³ f Õ» e¼ 1 ºÐ ž¾ K v 1-30 v Ñ Œ (1-12 Œ) v OCN fñäž Š v ïv ˆ (a)ñ ˆ çt (b) x á ˆ õõ 95% ³Åå n y ˆ x 1-2 Œ 6-8 Œ Œ Œ ³ ˆ õ Œ 1 Ð ž¾ v Œ ˆ(a) ˆçt(b) ˆ Ò á ˆ õõ 95% ³Åå n x ò áœ y ò á k

5 h ³ ˆ çt òo y ˆ çt 1-2 Œ 6-8Œ³ K ô Œ³ K õ e höœ 7-8 ô Ìå ôè Œ ³ K ˆ š¹n ³ K ŒˆåÀ K ³ 12 Œ ³ fx (HSS) Å º 2 ˆ 1 ³ ˆ ˆ ç t! t ³Å 2 º HSS òî½ OCN ³Œ Á º 11 Œ³ f áx ˆ f³x (HSS hˆ 0.2) x ò ³ K ø 1 ˆ ˆ çt ž ž¾ 11 Œ Ÿ ³ K 7 k = 7 k = 5 ³ HSS k = 5 ³ ˆšŒ r ž ¾( ) k = 7 žå ³ ˆ HSS ò 91 2 Ð ž¾ v Œ áx (HSS) Ò á hˆ 0 X ò áœ Y ò á k

6 92 MSE ò k = 7 ò ö h 10 Œ K = 25 Œ ˆ HSS ºšŒ f á ä f á x ³yäz ž¾³å Ð ž¾ o íô ž¾ Œ K ³ v f áx º 3 ä f áx 11 Œ K Œ Œò ³ áx õ Œ ³ ˆ õõ 95% æ ³Åå n Œ o³ ˆ çt ÒØž¾ f áx 11 Œò Œõ 0.2 õ OCN fˆš å OCN o á ¾ 11 Œ Œ ³ f áx Œ ˆ³ 10 Œ šœ 11 Œ³ò ³ f áx oùoð ž¾ç nž¾³{š Ð 3 2 ž¾ v Œ áx (HSS)

h t! 93 1971-2006vÐ ž¾³ 10Œ 11ŒŸv ÄoÆ º 4a Ñ 4b ÃoÆ K = 7 ³ OCN f Æ K = 25 ³ f K = 7 K = 25 ³äžÑ få ³ Íá ºˆ ˆ K = 7 11 Œ³ K K = 25 10 Œ ³ K 10 Œ³ K = 7 K = 25 f Ñäž ³ ˆ x K = 25 0.189 K = 7 0.

7 h t! vÐ ž¾³ 10Œ 11ŒŸv ÄoÆ º 4a Ñ 4b ÃoÆ K = 7 ³ OCN f Æ K = 25 ³ f K = 7 K = 25 ³äžÑ få ³ Íá ºˆ ˆ K = 7 11 Œ³ K K = Œ ³ K 10 Œ³ K = 7 K = 25 f Ñäž ³ ˆ x K = K = Íá º K = 25 f äž ³o~ ò K = 7 ç «f³ 15 Ÿ o ³ ³ Œ 6 HSS ¹o OCN f 4a ³ 10 ŒŸ vš v á ³ é i v vÿv Ì vÿ v õ½ ³ é OCN f ³ ä OCN ƒ K v v á f o é ³ ~õˆö ú i vÿvy vv³ ïv óé ìïv K = 7 ³ f Æ³ŸvöÐ v o ³ Ÿ 1994 v x t j K = 25 ³ f Æé uvx o n Î ~o ³é 11 Œ³ fá 5 10 Œh äž Ñ K = 7 K = 25 f ³ ˆ õ 0.48 K = 7 Íá º OCN f³ 15 Ÿ o ³ ³Œ 9 n ³ HSS õ 0.36 º OCN ˆ ž ò 4a Ñ 4b 10 Œ 11Œ h³ ˆ Ë 5-7v ÌxË³ÆzgŸî Â Æz ô x 11 Œ³Ÿv õ«v C 10 Œõ«0.17 C/ v ³ 4 11 Œ³Ÿv î õõ 95% yv³ Åå n ºõ î Åå ³ OCN f³ Ÿv ˆ 1980 Ì Ÿv x 1995 x Š î Ð f áx ò ³ Óo Š î ùˆ f 6 ºÐ ž¾ K vñ Œ HSS f áx î x î 2 ò 11 Œ³t p õ 11 Œ K = 7 ³ fx ö ˆ ž 4 Ð ž¾ 10 ŒäžŸv K 7 Ñ K 25 Š v ÄoÆ äž ÃoÆ K 7 Ã Æ K 25 t åˆn äž ô î äž Ñ K=7 ˆÑäž Ñ K=25 ˆ äž Ñ K=7 äž Ñ K=25 3 á á -: o o: u +: h HS á 3 f áx

8 94 æ 5 4 Ð ž¾ 11 Œ Š v Ñ 3 á 6 2 Ð ž¾ Œî x Œ HSS

9 h é 10 Œ 11 Œ Ñ Ÿy 11 Œ j ÆÄÅx òhœ Chen et al. (2008)³ ¼n ŸŒ ³ v é ë ³õŒ 11 v uhv v œÿ Œ ³õŒz é œÿ³é õhv Ñ ³ u gé õì xð ô³ p f é Š Ÿ Š Ÿ Ÿy 11 Œ j ÆÄÅx ò õœ Î é 10 Œ 11 Œ ³ õ å Œ ³Ÿv z 7 ò Ð ž¾ Œ ³Ÿv v(iqr) Æz î IQR ³åÀˆw vìˆƒv ³ v 75 Ñ 25 ³ t 7 n 2-5 Œ Œ IQR h áºõ ná³ÿv v Œòh³é hn Œ IQR t! ò hn uäå oì ÆvÄÅ o n u Æv³x Œ IQR ò ~ v ³ u ÆvÄÅo x ¼v 10 Œ 11 Œ³ IQR tï n 10 Œv Ìåö yw 11 Œ jx x Ÿv y Œ³t Œ Ÿ v³š ¼vt 11 Œ³Š w 9 ŒŠ w áºÿv³ ÍŠ h x hpv ÄÅ³x ùä 11 Œ w³š ë OCN f áx ³ ä 10 Œ³ OCN f át Š î òwì ëòw³æ Š î oÿvïv³x 11 Œ 7 Œ Œ ³î x žw Š î oÿv³x 95 7 Ð ž¾ Œ IQR(o )ÑŸvî ( Æ) Š v IQR áœìˆ v³ ³ 75 Ñ ³ 25 ³t X ò Œ Y ò IQR Ñô î

10 96 q Œˆ Ð ÒØ ž¾ ³Ÿvìˆƒå Ÿ³õ v v (K) ³ ÒØž¾h ä ž¾ õ³ K ž 12 Œ x³ÿ v ž á ˆ š š(k = 30) È š(k = 1) p 11 ŒŒ ³ f áx 11 Œ ³ f á Œ Œò ³ î Æz î x³ fx Œ ½³ õœ³v é y hh OCN ³ f áx õ ë 10 Œ ³Ÿvp ŸvÜˆõŒ ³óéŒ OCN n fî ÒØž¾³ 11 Œ Ÿ î Ä 0.4 C/decade nž¾ 0.7 C/decade o OCN ž HSS x ˆ ž¾ õ Î ÒØž¾³ 11 Œ Ÿv é òh p v Ñ ž¾³t h x OCN ³ f áò ÒØž¾³ 11 Œ Ÿ v é hˆä ³ ž¾ Œ Î n x Ð ¹ xxè ½å OCN Åå ž w Ë} Œ é zåå³ žˆš w š ã žˆš áä ³ Ì ž¾ ná õ ³ K Ó f áäîî š Èš foëy ˆ šœ ³ 5-7 v ˆ Œ 10 v ³é î Œä ³ 11 Œ Ÿvé õ ³ wéœ 11 ŒŸv ž f w ž á³ æ i Û 1998 ~ ô ³p Œ f h Ñ f å çˆxæ(87), Barnston, A. G., 1994: Linear statistical short-term climate predictive skill in the northern hemisphere. J. Climate. 5, Chen, J.-M., F.-C. Lu and C.-F. Shih, 2008: Decadal oscillation of fall temperature in Taiwan. Terr. Atmos. Ocean Sci. 19, Ebdon, R. A., 1963: The tropical stratospheric wind fluctuation. Weather, 18, 2-7. Hoskins, B., P. Schopf, and A. Navarra 2008: Modelling the Atmospheric, Oceanic and Coupled System. In Troccoli, A., M. S. J. Harrison, D. L. T. Anderson, and S. J. Mason (Eds), Seasonal Climate Variability: Forecasting and Managing Risk, Springer Academic Publishers, Dordrecht, Huang J., Huug M. van den Dool and Anthony G. Barnston. 1996: Long-Lead Seasonal Temperature Prediction Using Optimal Climate Normals. J. Climate. 9, Kerr, R. A., 2000: A North Atlantic climate pacemaker for the centuries, Science, 288 (5473), Lee, Yung-An, 2002: A T-EOF based prediction Method. J. Climate. 15,

11 h t! 97 Madden, R. A., and P. R. Julian, 1971: Description of a day oscillation in the zonal wind in the tropical Pacific. J. Atmos. Sci., 28, Madden, R. A., and P. R. Julian, 1972: Description of global-scale circulation cells in the tropics with a day period. J. Atmos. Sci., 29, Mantua, N. J. and S. R. Hare, Y. Zhang, J. M. Wallace, and R. C. Francis 1997: A Pacific interdecadal climate oscillation with impacts on salmon production. Bull. Amer. Meteor. Soc., 78, Mason, S. J., 2008: From dynamical predictions to seasonal forecasts. In Troccoli, A., M. S. J. Harrison, D. L. T. Anderson, and S. J. Mason (Eds), Seasonal Climate Variability: Forecasting and Managing Risk, Springer Academic Publishers, Dordrecht, Michaelson, J., 1987: Cross-Validation in statistical climate forecast models. J. Appl. Meteor., 26, Torrence, C. and C. P. Compo, 1998: A Practical Guide to Wavelet Analysis. Bul.l Amer. Meteor. Soc., 79, Troccoli, A., M. Harrison, D. L. T. Anderson, and S. J. Mason, 2998: Seasonal Climate Variability: Forecasting and Managing Risk, Springer Academic Publishers, Dordrecht, 467pp. Van den Dool, H., 2007: Empirical Methods in Short-Term Climate Prediction. Oxford Press. 215pp. Wilks. D. S. 1995: Statistical Methods in Atmospheric Sciences. Academic Press, 467pp. Wilks. D. S. 1996: Statistical Significance of Long-Range Optimal Climate Normal Temperature and Precipitation Forecasts. J. Climate. 9, WMO 1979: WMO Technical Regulations. Volume I General. Basic Documents No. 2. WMO-No. 49.

12 98 æ Using Optimal Climate Normals to Forecast Taiwan Temperature Ching-Teng Lee Mong-Ming Lu Central Weather Bureau, Taiwan (manuscript received 5 August 2009 in final form 20 October 2009) ABSTRACT The Optimal Climate Normals (OCN) is a simple statistical forecast model that requires data only from the station where the forecast is needed. The OCN has been used as a formal method for US seasonal forecasts at the Climate Prediction Center/NOAA since late Until the present, this method is still used as a bench mark for forecast skill measurement mainly due to strong warming trend after late 1970s. In this study, the forecast skill of OCN used for predicting Taiwan temperatures is analyzed. Four stations: Taipei, Taichung, Kaoshiung and Hualien are used to represent temperature characteristics in north, central, south and east Taiwan, respectively. Our results suggest that the forecast skill of the OCN method is generally poor over east Taiwan. However, over other regions, OCN shows robust high forecast skill in November, when the number of years (K) used in making the climate normal is 7. The November high skill can be explained by strong warming trend in this particular month. The trend is less clear in other months. An interesting contrast appears in October, of which the temperature shows clear 5-7 years variations that results in lower forecast skill. It is suggested that the OCN for November temperature over western Taiwan can be used as a baseline for evaluating the monthly or seasonal climate forecast models. Key Word: Optimal Climate Normals (OCN), Climate Forecast, Climate Statistics, Taiwan Climate

福建省中部沿海地區前汛期持續性暴雨低頻特徵

福建省中部沿海地區前汛期持續性暴雨低頻特徵 h ù! g 171 Ôä ¾ } Ìƒ ³ ÈŠ ú 1,2 i 1 ÕÇ 3 ü 2» 2 1 Î Ê 2 Õå vs 3 Õå ô h Š «Š Ø î žãþš ç 10 Ô NCEP/NCAR ä ä ãþ Š ç õ 5h6 ë³ Ôõ Ý ò ë ä õ Ý ò ³ ë ô ãðâ (1)ãþ Š ç õ â ë Ô (BWO)ÔØ (ISO) ³ ¹ 24% 70% (2) ³ Ž