Operational Models: Readiness Now Martin Köhler, ECMWF with Hua-Lu Pan, NCEP and Shouping Wang, NRL

Transcription

1 Operational Models: Readiness Now Martin Köhler, ECMWF with Hua-Lu Pan, NCEP and Shouping Wang, NRL observational data: EPIC DYCOMS-II GLAS cloud top GCSS Pacific Cross-Section ocean buoys AOSN field campaign (COAMPS) cloud vector winds QuikSCAT winds radiosonde winds SST for coupled GCMs model evaluation: (dt=12h to 10yr) the good q mix ± 0.5 g/kg θ l ± 0.5 K LWP ± 50 g/m 2 CC ± 10 % ΔT inv ± 2 K the bad SST ± 2 K u sfc + 1 m/s inv. height to 500m the unknown w ± 50% u 850 ± 2 m/s drizzle? aerosols? PBL transports and clouds Martin Köhler 1

2 EPIC: Peruvian stratocumulus model comparison LWC [g/m 3 ] q v [g/kg] θ [K] Bretherton et al, BAMS 2004 EPIC obs NCAR CAM2.0 GFDL AM2.10 ECMWF 23r4 NCEP PBL transports and clouds Martin Köhler 2

3 Stratocumulus: EPIC column from 3D forecasts PBL transports and clouds Martin Köhler 3

4 Stevens et al 2007: DYCOMS vs ECMWF vs NCEP PBL transports and clouds Martin Köhler 4

5 Shouping Wang, NRL: COAMPS Forecast Statistics (East Pacific) COAMPS E_PAC Grid 2 (27km) Temperature June 2006

6 Shouping Wang, NRL: COAMPS Forecast Statistics (East Pacific) COAMPS E_PAC Grid 2 (27km) Wind Speed June 2006 COAMPS E_PAC Grid 2 (27km) Relative Humidity June 2006 PBL transports and clouds Martin Köhler 6

7 ECMWF vs GLAS observations: cloud top height ECMWF strcu fraction ECMWF cloud top height GLAS strcu fraction GLAS cloud top height cloud top < 2km, cld > 80% Maike Ahlgrimm SC top too low! PBL transports and clouds Martin Köhler 7

8 Stevens et al 2007: DYCOMS vs ECMWF vs NCEP Taylor diagram Divergence θ PBL v PBL u PBL q PBL θ 850 ECMWF NCEP PBL transports and clouds Martin Köhler 8

9 IMET buoy (Anton Beljaars & Bob Weller) T 2m [ C] q 2m [ C] wind [m/s] EPIC PBL transports and clouds Martin Köhler 9

10 Shouping Wang, NRL: Field Temp. Speed Dir. Bias 3-km km COAMPS vs 2 buoys off Monterey RMS 3-km 9-km AOSN field campaign in the vicinity of Monterey Bay, California

11 GCSS Pacific Cross-Section Intercomparison pressure [hpa] GFLD AM2 NCEP C. Hannay HadGAM ECMWF ARPEGE ERA Equator Dateline California Equator Dateline California Equator Dateline California Joao Teixeira Equator Dateline liquid water [g/kg] pressure [hpa] NCAR CAM3 California JJA1998 PBL transports and clouds Martin Köhler 11

12 Cloud cover against ISCCP D2 CY32R3 - ISCCP ISCCP D2 PBL transports and clouds Martin Köhler 12

13 ECMWF buoy verification model mean: 6.08 m/s buoy mean: 6.12 m/s bias: m/s RMSE: 1.12 m/s correlation: PBL transports and clouds Martin Köhler 13

14 QuikSCAT winds: Bias and RMSE Bias [m/s] 0 bias RMSE [m/s] 1.4 m/s PBL transports and clouds Martin Köhler 14

15 QuikSCAT winds bias=0.06m/s QuikSCAT=7.17m/s PBL transports and clouds Martin Köhler 15

16 6 4 RMSE [m/s] GCSS Pacific Cross-section Project BIAS [Pa/s] W Bias at 500hPa U850hPa W500hPa RMS error BIAS ECMWF 1d/5d forecast NCAR 1d/5d forecast UKMO 1d/5d forecast ECMWF 1d/5d forecast NCAR 1d/5d forecast UKMO 1d/5d forecast V-Bias: METEOSAT-8, es6w: WVM V1: V2: N = V-Bias: METEOSAT-8, es6w: sfc vector wind: QuikSCAT vs ECMWF FG points: CA to EQ WVM V1: V2: N = points: EQ to CA ITCZ V-Bias: METEOSAT-8 vs ECMWF FG 200 WVcloud1: N = p [hpa] 600 Equator p [hpa] p [hpa] p [hpa] p [hpa] Lat [deg] Hans Herbach 200 IR3: N = Lat [deg] Lat [deg] IR3: N = VIS2: N = Claire Delsol. PBL transports and clouds Martin Köhler µm a] p a] [hpa] p a] [hpa]

17 SE Pacific U-profiles Analysis 48h Forecast CY31R1 <DIFF pressure [hpa] pressure [hpa] U wind [m/s] U wind [m/s] PBL transports and clouds Martin Köhler 17

18 Cloud Vector Winds versus Radiosondes Ascencion & St. Helen French Pacific Islands PBL transports and clouds Martin Köhler 18

19 Surface Temperature [ DEMETER CGCM Surface Bias [K] o C] Bias: EXP(UKMO) regarding ERA-40 reanalysis Forecast start dates: 05/ , FC period: months 4-6 (ASO), ens: 0-8 ECMWF (ERA40 cycle) Surface Temperature [ o C] Bias: EXP(MPI) regarding ERA-40 reanalysis Forecast start dates: 05/ , FC period: months 4-6 (ASO), ens: 0-8 UKMO MPI Surface Temperature [ o C] Bias: EXP(CNRM) regarding ERA-40 reanalysis Forecast start dates: 05/ , FC period: months 4-6 (ASO), ens: month forecasts Aug/Sep/Oct ensemble members comparison to ERA Surface Temperature [ o C] Bias: EXP(LODYC) regarding ERA-40 reanalysis Forecast start dates: 05/ , FC period: months 4-6 (ASO), ens: Meteo France LODYC PBL transports and clouds Martin Köhler 19

20 coupled ECMWF-ERA40 Surface temperature years 3-4 of integration Antje Weisheimer ENSEMBLES CY31R1: Sept 2006 CY32R1: June 2007 CY32R3: Fall 2007 model cycle: impl. date PBL transports and clouds Martin Köhler 20

21 EC EARTH (ECMWF uncoupled) future 2XCO2 scenario Pier Siebesma, KNMI Δ CC Standard model (cy31r1) Δ CC Enhanced topentrainment model PBL transports and clouds Martin Köhler 21

22 Operational Models: Readiness Now Martin Köhler, ECMWF with Hua-Lu Pan, NCEP and Shouping Wang, NRL observational data: EPIC DYCOMS-II GLAS cloud top GCSS Pacific Cross-section ocean buoy AOSN field campaign (COAMPS) cloud vector winds QuikSCAT winds radiosonde winds SST for coupled GCMs model evaluation: (dt=12h to 10yr) the good q mix ± 0.5 g/kg θ l ± 0.5 K LWP ± 50 g/m 2 CC ± 10 % ΔT inv ± 2 K the bad SST ± 2 K u sfc + 1 m/s inv. height to 500m the unknown w ± 50% u 850 ± 2 m/s drizzle? aerosols? PBL transports and clouds Martin Köhler 22