표제지
연구보고서 : 진보된 태풍분석 및 예측시스템 개발(II)
목차 3
CONTENTS 5
요약문 13
SUMMARY 16
제1장 서론 19
제2장 태풍-해양 상호작용 22
제1절 태풍강도변화와 해양상호작용 22
1. 서론 22
2. 태풍관련 해양열량지수의 정의 23
3. 자료 및 사례선정 25
4. 결과 26
5. 토의 및 결론 51
제3장 마이크로파 위성자료를 이용한 태풍강도 연구 53
제1절 TRMM TMI 관측과 태풍강도와의 관련성 53
1. 서론 53
2. 자료 및 분석 방법 56
3. 태풍강도와 공간상관관계 분석 57
4. 태풍강도와 시간상관관계 분석 65
5. 결론 71
제2절 상륙태풍 구조변화 특성연구를 위한 레이더자료 사전분석 73
1. 서론 73
2. 레이더자료 활용 태풍구조 분석법과 사례분석 73
3. 결론 82
제4장 태풍모델의 예측성 개선 83
제1절 차세대 태풍모델(TWRF) 개발 및 사례적용 83
1. 서론 83
2. 태풍영역 및 보거싱 방법 84
3. 분석 결과 87
4. 요약 및 토의 95
제2절 이동격자태풍모델 개선 96
1. 서론 96
2. 실험 설계 및 방법 98
3. 결과 100
4. 요약 108
제5장 태풍의 온대저기압화 특성 연구 110
제1절 저기압 위상공간도를 이용한 온대성 저기압 분석 110
1. 서론 110
2. 자료 및 분석방법 111
3. 분석 결과 116
4. 2006 사례에 대한 검증 결과 127
5. 요약 및 토의 128
제6장 요약 및 결론 130
참고문헌 133
APPENDIX(제목없음) 137
학술용역 과제 137
1. 태풍정보 콘텐츠 개발 - 태풍바람의 통계적 특성 연구 - 137
제출문 138
보고서 요약서 139
요약문 140
SUMMARY 143
CONTENTS 147
목차 149
제1장 연구개발과제의 개요 155
1.1. 연구의 필요성과 목적 155
1.2. 연구의 내용 및 범위 155
제2장 국내외 기술개발 현황 157
제3장 연구개발 수행 내용 및 결과 158
3.1. 태풍 최대풍속 예측 모형 개발 158
3.1.1. 분석자료 158
3.1.2. 분석자료에 대한 탐색적 분석 162
3.1.3. Best-track data에서의 최대풍속 모형 169
3.1.4. Best-track data 최대풍속 수정 모형 171
3.1.5. 요약 및 제언 175
3.2. 태풍 바람의 반경 예측 모형 개발 176
3.2.1. 태풍바람 반경의 정의 및 자료 176
3.2.2. 태풍바람 반경 관련 변수의 특성분석 176
3.2.3. 태풍바람 반경예측 모형 181
3.2.4. 요약 및 제언 186
3.3. 태풍 바람 예측 모형 개발 187
3.3.1. 태풍바람 187
3.3.2. 태풍바람 예측 모형 190
3.3.3. 태풍바람 이진예보 194
3.3.4. 요약 및 제언 202
제4장 목표달성도 및 관련분야에의 기여도 203
제5장 연구개발결과의 활용계획 204
참고문헌 205
부록 I : 지상관측지점 일람표 206
부록 II : AWS 지점 일람표 208
부록 III : 분석 및 예측모형 개발을 위한 SAS 프로그램 218
부록 IV : 태풍바람 예측치 생성 FORTRAN 프로그램 226
2. Numerical Study of Landfalling Typhoons in the Western Pacific Ocean with WRF Model 230
SUMMARY 231
CONTENTS 237
1. Introduction 240
2. Overview of Typhoon "Haitang" 243
2.1. Track 243
2.2. Radar imagery 243
2.3. Infrared cloud-top bright temperature (TBB) imagery of Goes-9 satellite 244
3. Numerical simulations 244
3.1. The model and experimental design 244
3.2. Simulation results 246
3.2.1. The contrastive analyses between simulations and observations 246
3.2.2. Asymmetric structure during landfall 247
4. Diagnostic analysis of some physical parameters during landfall 248
4.1. Helicity 248
4.2. Moist Potential Vorticity (MPV) 249
4.3. Convective Vorticity Vector (CVV) and Moist Vorticity Vector (MVV) 249
5. Conclusions and discussion 250
References: 266
Table 3.1.1. Characteristics of DMSP SSM/I and TRMM TMI. 55
Table 3.1.2. Regression coefficient derived from satellite observation (TB, PCT RR, PW) and typhoon intensity. 63
Table 3.1.3. Lag Correlation Coefficient (LCC) between maximum wind speed with observation of 0.5˚ circle mean for typhoon Rusa and MAEMI(0314). 71
Table 3.2.1. The Typhoons affected on the korea peninsula. 77
Table 4.1.1. Experiment design. 85
Table 4.2.1. Description of MTM and five experiments (EXP1, EXP, EXP3, EXP4, EXP5). 99
Table 4.2.2. Mean error distance of typhoon track prediction of ORG, EXP3, JGSM, JTYM, GDAPS, DBAR. 106
Table 4.2.3. Training time of ORG and five experiments(EXP1, EXP2, EXP3, EXP4, EXP5). 108
Table 5.1.1. RSMC best track of CHANCHU(0601). 112
태풍정보 콘텐츠 개발 : 태풍바람의 통계적 특성 연구 153
Table 3.1.2.1. Observation locations with the maximum velocity of the wind 163
Table 3.1.2.2. Basic Statistics for high press category(p_cls=0) 165
Table 3.1.2.3. Basic Statistics for low press category(p_cls=1) 165
Table 3.1.2.4. Basic Statistics for all data 166
Table 3.1.2.5. Basic Statistics for right sided data 166
Table 3.1.2.6. Basic Statistics for left sided data 166
Table 3.1.2.7. Basic Statistics for the 1st quadrant data 167
Table 3.1.2.8. Basic Statistics for the 2nd quadrant data 167
Table 3.1.2.9. Basic Statistics for the 3rd quadrant data 167
Table 3.1.2.10. Basic Statistics for the 4th quadrant data 167
Table 3.1.2.11. Frequency table for the B and category of central press 168
Table 3.1.2.12. Frequency table for the Q and category of central press 168
Table 3.1.3.1. Result of regression analysis of best-track data 170
Table 3.1.4.1. Result of regression analysis between maximum velocity of the wind(SMWS) in observatories with variables in best-track data 172
Table 3.1.4.2. AIC according to number of hidden node 173
Table 3.1.4.3. Parameter estimates of NN 3-3-1 174
Table 3.1.4.4. Result of regression analysis on maximum velocity of the wind(SMWS) in observatories with estimates variables obtained by neural networks 174
Table 3.2.2.1. Basic Statistics for all data 176
Table 3.2.2.2. Basic Statistics for right sided data 177
Table 3.2.2.3. Basic Statistics for left sided data 177
Table 3.2.2.4. Basic Statistics for high press category(p_cls=0) 177
Table 3.2.2.5. Basic Statistics for low press category(p_cls=1) 178
Table 3.2.2.6. Merged data after calculation of maximum distance (in part) 178
Table 3.2.2.7. Correlation analysis of variables 180
Table 3.2.3.1. Results of regression analysis of best-track data and typhoon wind radius 181
Table 3.2.3.2. AIC according to number of hidden node 182
Table 3.2.3.3. Results of fitted statistics 183
Table 3.2.3.4. Parameter estimates of NN 3-4-1 183
Table 3.2.3.5. Results of regression analysis of best-track data and typhoon wind radius 184
Table 3.3.1.1. Basic Statistics for all data 187
Table 3.3.1.2. Correlation analysis among variables with p-values for H : not correlated. 188
Table 3.3.2.1. Results of regression analysis of best-track data and wind speed 190
Table 3.3.2.2. Results of regression analysis of observation and predicted value 191
Table 3.3.2.3. AIC according to number of hidden node 192
Table 3.3.2.4. Results of fitted statistics 192
Table 3.3.2.5. Parameter estimates of NN 5-6-1 193
Table 3.3.2.6. Results of regression analysis of best-track data and typhoon wind 193
Table 3.3.3.1. Results of regression analysis for binary forecast 195
Table 3.3.3.2. Results of regression analysis for binary forecast 195
Table 3.3.3.3. 2x2 Tables(threshold=0.5) 196
Table 3.3.3.4. AIC according to number of hidden node 196
Table 3.3.3.5. Parameter estimates of NN 4-5-1 197
Table 3.3.3.6. Results of regression analysis for binary forecast by the NN 4-5-1 199
Table 3.3.3.7. 2 X 2 Tables(threshold=0.5) 200
Table 3.3.3.8. Threshold skill score and threshold 200
Table 3.3.3.9. 2 X 2 Tables(threshold=0.025) 201
Table 3.3.3.10. 2 X 2 Tables(threshold=0.1) 202
Fig. 2.1.1. Climatological distribution on the depth of the 26℃ isotherm on July - September. 24
Fig. 2.1.2. An example of the ocean water temperature profile to define the typhoon-related oceanic heat content index. Shaded area indicates the index of the heat content. 25
Fig. 2.1.3. SST and TOHCI distribution for typhoon SONCA(0503):... 27
Fig. 2.1.4. SST and TOHCI distribution for typhoon NESAT(0504) :... 28
Fig. 2.1.5. SST and TOHCI distribution for typhoon NALGAE(0506) :... 29
Fig. 2.1.6. SST and TOHCI distribution for typhoon MAWAR(0511) :... 30
Fig. 2.1.7. SST and TOHCI distribution for typhoon GUCHOL(0512) :... 31
Fig. 2.1.8. SST and TOHCI distribution for typhoon TALIM(0513) :... 32
Fig. 2.1.9. SST and TOHCI distribution for typhoon NABI(0514) :... 33
Fig. 2.1.10. SST and TOHCI distribution for typhoon KHANUN(0515) :... 34
Fig. 2.1.11. SST and TOHCI distribution for typhoon KIROGI(0520) :... 35
Fig. 2.1.12. SST and TOHCI distribution for typhoon BILIS(0604) :... 36
Fig. 2.1.13. SST and TOHCI distribution for typhoon SAOMAI(0608) :... 37
Fig. 2.1.14. SST and TOHCI distribution for typhoon BOPHA(0609) :... 38
Fig. 2.1.15. SST and TOHCI distribution for typhoon I0KE(0612) :... 39
Fig. 2.1.16. SST and TOHCI distribution for typhoon SHANSHAN(0613) :... 40
Fig. 2.1.17. SST and TOHCI distribution for typhoon CIMARON(0619) :... 41
Fig. 2.1.18. SST and TOHCI distribution for typhoon MAN-YI(0704) :... 42
Fig. 2.1.19. SST and TOHCI distribution for typhoon USAGI(0705) :... 43
Fig. 2.1.20. SST and TOHCI distribution for typhoon FITOW(0709) :... 44
Fig. 2.1.21. SST and TOHCI distribution for typhoon NARI(0711) :... 45
Fig. 2.1.22. Scatter diagram between the TOHCI and central pressure difference(/6hrs) during the intensifying period. 47
Fig. 2.1.23. Prediction example of intensity for typhoon MAN-YI(0704) 48
Fig. 2.1.24. Prediction example of intensity for typhoon USAGI(0705) 49
Fig. 2.1.25. Prediction example of intensity for typhoon NARI(0711) 49
Fig. 2.1.26. Prediction example of intensity for typhoon WIPHA(0712) 50
Fig. 2.1.27. Prediction example of intensity for typhoon KROSA(0715) 50
Fig. 3.1.1. The structure of typhoon Rusa (30, Aug 2002) in (a) GMS visible (b) infrared and (c) TRMM microwave 85 GHz channel imagery. 54
Fig. 3.1.2. Correlation coefficient between satellite observation and typhoon intensity on (a) 2000, (b) 2001, (c) 2002, (d) 2003, and (e) 2004. Upper left(right) panel indicates 85 GHz TB (PCT) and Lower left(right) panel indicates RR(PW), respectively. 56
Fig. 3.1.3. The structure of tropical cyclone. 58
Fig. 3.1.4. Correlation coefficient between satellite observations of (a) TB, (b) PCT, (c) RR, and (d) PW and typhoon intensity from 2000 to 2004 with respect to spatial distance. 59
Fig. 3.1.5. Categorized correlation coefficient between satellite observations of (a) TB, (b) PCT, (c) RR, and (d) PW and typhoon intensity from 2000 to 2004 with respect to spatial distance. 60
Fig. 3.1.6. Same as figure 3.1.5 for each year. 61
Fig. 3.1.7. Scatter diagrams and regression equations of (a) TB, (b) PCT, (c) RR, and (d) PW and observed maximum wind speed. 62
Fig. 3.1.8. Comparison of root mean square error (RMSE) of regressed typhoon intensity (maximum surface wind) for typhoon (a) Rusa (0215) and (b) Maemi (0314), adopted from METRI (2006) annual report. (c) and (d) are same value as (a) and (b) but calculated by new regression coefficient... 64
Fig. 3.1.9. Correlation coefficient between satellite observations of TB and typhoon intensity from 2000 to 2004 with respect to spatial distance and time lag. 66
Fig. 3.1.10. Same as Fig. 3.1.9 but for the typhoons of TY degree. 67
Fig. 3.1.11. Correlation coefficient between satellite observations of RR and typhoon intensity from 2000 to 2004 with respect to spatial distance and time lag. 68
Fig. 3.1.12. Same as Fig. 3.1.11 but for the typhoons of TY degree. 69
Fig. 3.1.13. Time series of maximum wind speed and observations of (a) TB and (b) RR for typhoon Rusa (c) and (d) are for typhoon MAEMI(0314). 71
Fig. 3.2.1. The radar spiral by Least square method. 74
Fig. 3.2.2. Schematic of the precipitation features. 76
Fig. 3.2.3. The radar precipitation at 16. Sep., 2007. 78
Fig. 3.2.4. The radar of PPI VR(radial velocity) at 16. Sep., 2007. 79
Fig. 3.2.5. The radar precipitation at10. JUL., 2006. 81
Fig. 3.2.6. The radar of PPI VR(radial velocity) at 10. JUL., 2006. 81
Fig. 3.2.7. The precipitation at 10. JUL., 2006. 82
Fig. 4.1.1. TWRF Domain 84
Fig. 4.1.2. Flow chart of Typhoon initializing in TWRF 87
Fig. 4.1.3. The forecast track of MAN-YI(0704) on 13 JUL., 2007. 88
Fig. 4.1.4. Comparison of between Result of Track prediction (60hr) included bogucing and not bogucing. 90
Fig. 4.1.5. Comparison of intensity prediction. 92
Fig. 4.1.6. Comparison of between Result of sea level pressure prediction included bogucing and not. 93
Fig. 4.1.7. Comparison of between Result of precipitation prediction included bogucing and not. 94
Fig. 4.2.1. Flow chart of MTM. 97
Fig. 4.2.2. Track of NARI (0711) 98
Fig. 4.2.3. Sea Surface Temperature at 12 UTC 13 September 2007.... 100
Fig. 4.2.4. Track prediction (left), error distance (right upper), pressure prediction (right lower) of NARI (0711).... 101
Fig. 4.2.5. Mean error distance of typhoon track prediction of ORG, EXPI, EXP2, EXP3, EXP4, EXP5. 105
Fig. 4.2.6. Mean error distance of typhoon track prediction of ORG, EXP3, JGSM, JTYM, GDAPS, DBAR. 106
Fig. 4.2.7. Comparison of intensity Prediction of ORG, EXPS, JGSM, JTYM, GDAPS.... 107
Fig. 5.1.1. The ideal result of CPS method when was Extratropical Transition occurred 115
Fig. 5.1.2. Comparison between the result of Evans' method(2003) and this study. 116
Fig. 5.1.3. The Satellite images and weather chart of EWINIAR(0603) when was Extratropical Transition. 119
Fig. 5.1.4. The result of CPS analysis of EWINIAR(0603) with - VUT vs - VLT and B vs. - VLT.(이미지참조) 120
Fig. 5.1.5. The Satellite images and weather chart of MARIA(0607) when was Extratropical Transition. 122
Fig. 5.1.6. The result of CPS analysis of MARIA(0607) with - VUT vs - VLT and B vs - VLT.(이미지참조) 123
Fig. 5.1.7. The Satellite images and weather chart of SHANSHAN(0613) when was Extratropical Transition. 125
Fig. 5.1.8. The result of CPS analysis of SHANSHAN(0613) with - VUT vs - VLT and B vs - VLT.(이미지참조) 126
Fig. 5.1.9. The result of CPS method for Typhoons during 2006. 128
태풍정보 콘텐츠 개발 : 태풍바람의 통계적 특성 연구 151
Figure 3.1.1.1. Locations of central pressure of the Typhoon Wind in the best-track data(Left : Data of land and sea, Right : Data of the land) 159
Figure 3.1.1.2. Locations of the observation data in Korea 161
Figure 3.1.2.1. Plot of locations with the maximum velocity of the wind 162
Figure 3.1.2.2. Exploration data analysis for the central pressure 164
Figure 3.1.2.3. Plot of relative locations of maximum velocity of the wind from the center of typhoon for each category of central press 169
Figure 3.1.3.1. Scatter plot for the maximum velocity of the wind and Estimate of regression model in best-track data(left : scatter plot, right : residual plot) 170
Figure 3.1.3.2. Scatter plot of between maximum velocity of the wind(SMWS) in observatories with maximum velocity of the wind(MWS) in Best-track data 171
Figure 3.1.4.1. Scatter plot between the maximum velocity of the wind and estimate of regression model in best-track data(left : scatter plot, right : residual plot) 172
Figure 3.1.4.2. Structure of neural networks : NN 3-3-1 173
Figure 3.1.4.3. Scatter plot of maximum velocity of the wind and fitted value(left) and residual plot(right) 175
Figure 3.2.2.1. Distribution of typhoon wind radius 179
Figure 3.2.2.2. Distribution of typhoon wind radius for each category 179
Figure 3.2.2.3. Scatter gram matrix of variables 180
Figure 3.2.3.1. Scatter plots of typhoon wind radius and fitted value(left) and residual plot(right) 181
Figure 3.2.3.2. Structure of neural networks : NN 3-4-1 182
Figure 3.2.3.3. Scatter plot of typhoon wind radius and fitted value(left) and residual plot(right) 185
Figure 3.3.1.1. Scatter plot matrix among variables 189
Figure 3.3.2.1. Scatter plot of wind speed and fitted value(left) and residual plot(right) 191
Figure 3.3.2.2. Structure of neural networks : NN 5-6-1 192
Figure 3.3.2.3. Scatter plot of typhoon wind and fitted value(left) and residual plot(right) 194
Figure 3.3.3.1. Structure of neural networks : NN 4-5-1 197
Figure 3.3.3.2. Structure of neural networks : NN 4-5-1 199
Figure 3.3.3.3. Plot of TSS for each threshold 201
Numerical Study of Landfalling Typhoons in the Western Pacific Ocean with WRF Mode 238
Fig. 1. Tracks of typhoon "Haitang" from 00 UTC 19 to 12 UTC 20 July 2005.... 253
Fig. 2. The reflectivity from the "Changle" radar in Fujian province (dbz) at (a) 01 UTC 1'57" and (b) 09 UTC 8'25" 19 July 2005. 254
Fig. 3. The Goes-9 infrared cloud-top brightness temperature (oC) at (a) 00 UTC 19 July; (b) 06 UTC 19 July; (c) 12 UTC 19 July; (d) 18 UTC 19 July; (e) 00 UTC 20 July; (f) 06 UTC 20 July; (g) 12 UTC 20 July 2005. 256
Fig. 4. The evolution of (a) sea level pressure (hPa); (b) surface horizontal wind velocity (m s-1) from 00 UTC 19 to 12 UTC 20 July 2005. 257
Fig. 5. (a) horizontal wind velocity (vector arrow, m s-1) and geo-potential height (isoline, m) (b) sea level pressure (hPa); (c) vorticity (10-4 s-1), (d) divergence (10-4 s-1) at 800 hPa at 00 UTC 19 July 2005. 258
Fig. 6. (a) horizontal wind velocity (vector arrow, m s-1) and geopotential height (isoline, m) (b) sea level pressure (hPa); (c) vorticity (10-4 s-1), (d) divergence (10-4 s-1) at 800 hPa (e) 1-hour accumulated precipitation (mm) (f) the total water mixing ratio (g g-1) at 800 hPa at 09 UTC 19 July 2005. 259
Fig. 7. As Fig. 6, but at 12 UTC 20 July 2005. 260
Fig. 8. (a) The zonal-vertical cross sections of the equivalent potential temperature (solid line, K) and vertical velocity (dashed line, hPa s-1); (b) relative humidity (solid line, %) and vertical velocity (dashed line, hPa s-1) along 26oN at 09 UTC 19 July 2005. 261
Fig. 9. (a) The zonal-vertical cross section of the helicity (isoline, 10-4 hPa s-2) and vertical velocity (shaded, hPa s-1) along 26oN; (b) helicity (isoline, 10-4 hPa s-2) at 800 hPa at 09 UTC 19 July 2005. 262
Fig. 10. (a) The zonal-vertical cross section of the moist potential vorticity (isoline, PVU) and vertical velocity (shaded, hPa s-1) along 26oN; (b) moist potential vorticity (isoline, PVU) at 800 hPa at 09 UTC 19 July 2005. 263
Fig. 11. The zonal-vertical cross section of (a) the vertical component of convective vorticity vector (isoline, 10-8 K s-1) and vertical velocity (shaded, hPa s-1); (b) the vertical component of moist vorticity vector (isoline, 10-11 s-1) and vertical velocity (shaded, hPa s-1) along 26oN at 09 UTC 19 July 2005. 264
Fig. 12. The distributions of the vertical components of (a) convective vorticity vector (10-8 K s-1); (b) moist vorticity vector (10-11 s-1) at 800 hPa at 09 UTC 19 July 2005. 265