EMC multimodel ensemble TC track forecast Jiayi Peng
EMC multi-model ensemble TC track forecast Jiayi Peng*, Yuejian Zhu and Richard Wobus* *IMSG at Environmental Modeling Center /NCEP/NOAA, Camp Springs, MD 20746 Acknowledgements: EMC Ensemble Team EMC HWRF Team Hurricane Forecast Improvement Project Internal HWRF Meeting, May 26, 2011
Models and Data Resolution Members Daily Frequency Forecast Length 00, 06, 12, 18 UTC 16 days (384 hrs) NCEP ensemble GFS T 190 L 28 70 km - 20+1 CMC ensemble GEM(0. 9)L 58 -100 km 20+1 00, 12 UTC 16 days (384 hrs) ECMWF ensemble IFS T 639/319 L 62 -30/60 km 50+1 00, 12 UTC 15 days (360 hrs) NCEP deterministic GFS T 574 L 64 27 km 00, 16, 12, 18 UTC 192 /384 hrs CMC deterministic GEM(0. 45 x 0. 3) L 58 -33 km 1 00, 12 UTC 180/240 hrs (12/00 z) ECMWF deterministic IFS T 1279 L 91 -16 km 1 00, 12 UTC 240 hrs - 1 Data: 01/01~12/31/2010 NCEP, CMC and ECMWF TC tracks Our goal: Improve tropical-cyclone track prediction by using NCEP, CMC and ECMWF global ensemble forecast products !
2010 multi-model ensemble product (3 EMN). 21 hurricanes in Atlantic 1 Nautical Mile=1. 852 km Verification: 3 EMN and 6 EMX are much better than OFCL. Forecast hours GEMN 3 EMN AVNO 6 EMX OFCL NCEP (GEFS) 20 -member mean NCEP+CMC+ECMWF 90 -member mean NCEP (GFS) deterministic run (T 574 L 64) three ensemble+ three deterministic runs weighted-mean OFCL official forecast
2010 multi-model ensemble product (3 EMN). 12 hurricanes in East Pacific Forecast hours GEMN 3 EMN AVNO 6 EMX OFCL NCEP (GEFS) 20 -member mean NCEP+CMC+ECMWF 90 -member mean NCEP (GFS) deterministic run (T 574 L 64) three ensemble+ three deterministic runs weighted-mean OFCL official forecast
2010 multi-model ensemble product (3 EMN). 19 Typhoons in West Pacific Verification: 3 EMN and 6 EMX are much better than JTWC. Forecast hours GEMN 3 EMN AVNO 6 EMX JTWC NCEP (GEFS) 20 -member mean NCEP+CMC+ECMWF 90 -member mean NCEP (GFS) deterministic run (T 574 L 64) three ensemble+ three deterministic runs weighted-mean JTWC official forecast
2010 Atlantic: 3 EMN/6 EMN (3 EMN/6 EMN 0 -hour delay) #CASES 195 174 154 136 118 Forecast hours 85 69 3 EMN---NCEP, CMC, ECMWF 90 -member mean 6 EMN---three ensemble means+three deterministic runs, 6 -member mean 55
2010 Atlantic: 3 EMI/6 EMI (6 -hour TC track interpolation, 3 EMN/6 EMN 6 -hour delay) #CASES 179 165 145 130 115 Forecast hours 84 67 3 EMN---NCEP, CMC, ECMWF 90 -member mean 6 EMN---three ensemble means+three deterministic runs, 6 -member mean 51
2010 Atlantic: 3 EM 2/6 EM 2 (12 -hour TC track interpolation, 3 EMN/6 EMN 12 -hour delay) #CASES 173 155 137 122 104 Forecast hours 79 62 3 EMN---NCEP, CMC, ECMWF 90 -member mean 6 EMN---three ensemble means+three deterministic runs, 6 -member mean 48
2010 West Pacific: 3 EMN/6 EMN (3 EMN/6 EMN 0 -hour delay) #CASES 154 147 132 112 96 66 41 24 Forecast hours 3 EMN---NCEP, CMC, ECMWF 90 -member mean 6 EMN---three ensemble means+three deterministic runs, 6 -member mean
2010 West Pacific: 3 EMI/6 EMI (6 -hour TC track interpolation, 3 EMN/6 EMN 6 -hour delay) #CASES 138 134 122 105 91 63 38 22 Forecast hours 3 EMN---NCEP, CMC, ECMWF 90 -member mean 6 EMN---three ensemble means+three deterministic runs, 6 -member mean
2010 West Pacific: 3 EM 2/6 EM 2 (12 -hour TC track interpolation, 3 EMN/6 EMN 12 -hour delay) #CASES 133 126 114 99 84 57 32 21 Forecast hours 3 EMN---NCEP, CMC, ECMWF 90 -member mean 6 EMN---three ensemble means+three deterministic runs, 6 -member mean
Strike probability AL 06: Danielle (2010) Ensemble mean track GEMN----NCEP 20 -member mean TEMN----CMC 20 -member mean UEMN---- ECMWF 50 -member mean 3 EMN----NCEP+CMC+ECMWF 90 -mean
Strike probability WP 15: Megi (2010) Ensemble mean track GEMN----NCEP 20 -member mean TEMN----CMC 20 -member mean UEMN---- ECMWF 50 -member mean 3 EMN----NCEP+CMC+ECMWF 90 -mean
GEFS-T 254 next implementation in 2011 GFS T 574: GFSv 9. 01, next implementation GEFS T 190: GFSv 8. 0 , current operation GEFS T 254: GFSv 9. 01, next implementation CASES 119 114 100 85 75 56 40 27 Tropical Storm Tracks (Aug. – Sep. 2010, for AL, EP and WP)
Conclusions • The multi-model ensemble forecast products (NCEP, CMC and ECMWF ensemble forecast) could significantly reduce the 2010 TC track forecast error. • If the multi-model ensemble forecast products delay 6 -hour (12 -hour), they are still valuable to the 2010 TC track forecast. Please visit http: //www. emc. ncep. noaa. gov/gmb/jpeng/TC_ens. html 2011 plans: • Multi-model ensemble TC-track forecast in parallel. • GEFS(T 254) TC-track bias-correction based on 30 -year reforecast. • Add Navy Ensemble (FNMOC), Met. Office ensemble into this multiple model ensemble forecast system. • Add meso-scale models (HWRF, GFDL, SREF, etc. ) into this multiple model ensemble forecast system. • Develop new consensus /ensemble method. • Produce TC intensity multi-model ensemble forecast.
2010 Atlantic: Multi-model (6 EMX) VS. individual model forecast #CASES 196 174 154 136 118 85 69 55 Forecast hours GEMN/AVNO----NCEP TEMN/CMH----CMC UEMN/EMH----ECMWF
2010 East Pacific: Multi-model (6 EMX) VS. individual model forecast #CASES 81 66 53 44 36 28 16 9 Forecast hours GEMN/AVNO----NCEP TEMN/CMH----CMC UEMN/EMH----ECMWF
2010 West Pacific: Multi-model (6 EMX) VS. individual model forecast #CASES 154 147 132 112 96 66 41 24 Forecast hours GEMN/AVNO----NCEP TEMN/CMH----CMC UEMN/EMH----ECMWF
- Slides: 18