COSMO-SREPS COSMO Priority Project

C. Marsigli, A. Montani and T. Paccagnella

ARPA-SIM, Bologna, Italy

COSMO-SREPS

COSMO-LEPS (10 km) mainly designed for the 2-5 days range

The spread is coming only from EPS initial and boundary conditions (large scale uncertainties)

COSMO-SREPS (10 km) has been thought as a limited-area ensemble system for the short-range:

add more spread in the short-range => initial and boundary condition from a Multi-Model Multi-Analysis ensemble (INM SREPS)

take into account also small scale uncertainty => limited-area model (LM) errors

Link with data assimilation

1D-Var DA of satellite data to provide retrieval to be nudged within 7km LM (currently carried on by COSMO): investigate the possibility of using COSMO-SREPS to estimate a flow-dependent B matrix

SIR project: COSMO-SREPS will provide boundary conditions for the LM-K ensemble data assimilation runs

limited-area model error: schemes, parameters,

surface forcings

COSMO-SREPS

i.c. and b.c. perturbations -> INM multi-model multi-analysis

model perturbations -> Lokal Model perturbations

LAM ensemble

driving model errorsanalysis error at the global model scale

larger scale errors

smaller scale errors

ic and bc: multi-model approach

a MUlti-Model MUlti-Boundaries ensemble is operational at INM (Spain): 5 different limited-area models (HIRLAM, HRM, MM5, UM, LM) are nested on 4 different global models (IFS, GME, AVN, UM) => 20 members ensemble (25 km horizontal resolution)

IFS LM 25km

GME LM 25km

AVN LM 25 km

Courtesy of INM

UM LM 25 km

Good example of inter-consortia collaboration:

Formal agreement between COSMO and INM:

LM included into INM SREPS => more members

INM provides the LM-on-SREPS forecasts to COSMO

SREPS output made available by INM from May 2006: 3-hourly initial and boundary conditions for COSMO-SREPS are provided daily at ECMWF on ECFS (4 LM runs at 25 km on an Euro-Atlantic area, nested on 4 different global models interpolated with int2lm at 10 km over the COSMO-LEPS area)

Perturbation of model trajectories

• Schemes (physics and numerics): Some schemes have been selected to be perturbed: Tiedtke, Tiedtke with cape closure, Kain-Fritsch, 2-time-level, Runge-Kutta

• Parameters:

• a first selection of the parameters has been done

• new tuneable parameters will be tested (provided by the COSMO physics people + new namelist)

• Surface forcings: possible cooperation with Luca Bonaventura to test a technique for the perturbation of surface parameters

+ exploration of methodologies to perturb initial conditions at high resolution

Parameter listparameter name parameter description range default value usedPHYSICS

crsminminimal stomata resistance (part of 'terra'; sensible for much vegetation and not too dry soil [50,200] s/m 150 50, 200

c_soilsurface area index of the evaporating soil (part of 'terra'; sensible for less vegetation and not too dry soil ]0,c:lnd[ 0, clnd (=2)

plcovplant cover (part of 'terra' and 'turbtran'; controls transpiration) [0,1]

laileave area index (part of 'terra' and 'turbtran', controls transpiration) [0,10]

rlam_heat scaling factor of the laminar layer depth [0.1,50] 0.1, 50

c_difffactor for turbulent diffusion of TKE (may influence mixing at an inversion top) [0,2] 1 0, 2

pat_lenlength scale of thermal surface patterns (influences mixing in the stable stratified atmosphere) [0,10000] m 500 0, 10000

tur_lenmaximal turbulent length scale (influences mainly stratospheric mixing) [100,1000] m 1000

rat_canscaling parameter for the diagnosed canopy height (rather big influence on Td2m, but will be obsolete shortly) [0,10]

z0m_diaroughness length at a synop station (rather big influence on V10m, will very likely be obsolete in future) [0.05,0.5] m

NUMERICSl2tls 2-time-level scheme T or F false trueepsass Asselin filter coefficient [0.05,0.2] 0.15 0.05

COSMO-SREPS (Short-Range Ensemble

Prediction System) short range (up to 3 days)

16 LM runs at 10 km

LM at 25 km on IFS

IFS – ECMWF globalb

y I

NM

Sp

ain LM at 25 km on GME

GME – DWD global

LM at 25 km on UM

UM – UKMO global

LM at 25 km on NCEP

AVN – NCEP global

P1: conv. scheme (KF)

P2: turb. parameter 1

P3: time filter

P4: turb. parameter 2

13 15 UTC

47N

44N

6W 9W

13 18 UTC

47N

45N

7W

10W

14 00 UTC

46.5N

45N

10.5W 13W

14 06 UTC

13 May 12-18 UTC

COSMO-SREPS PROB. MAPS fc. range 12-18 18-24 24-30 h

COSMO-LEPS PROB. MAPS fc. range 0-6 6-12 12-18 h

> 10mm/6h

> 10mm/6h

10

30

60

90

100

COSMO-SREPS fc. range 24-30

14 May 0-6 UTC

ecmwf

gme

avn

ukmo

ensemble spread

12/05

12UTC

13/05

00UTC

13/05

12UTC

member behaviour (“father”)

intra-group distance

1-month test:autumn 2006

16 LM runs at 10 km

LM at 25 km on IFS

IFS – ECMWF globalb

y I

NM

Sp

ain LM at 25 km on GME

GME – DWD global

LM at 25 km on UM

UM – UKMO global

LM at 25 km on NCEP

AVN – NCEP global

P1: control (ope)

P2: conv. scheme (KF)

P3: turb. parameter 1

P4: turb. parameter 2

1-month test: model pert. with the same father

10 LM runs at 10 km

LM at 25 km on IFS

IFS – ECMWF global

by I

NM

Sp

ain

P1: control (ope)

P2: conv. scheme (KF)

P3: turb. parameter 1

P4: turb. parameter 2

P5: …

Future plans

• run of COSMO-SREPS with model perturbations for 1 month of Autumn 2006:

• statistical analysis of the system

• provide b.c. for EELMK (DWD) tests

• comparison with COSMO-LEPS and PEPS

• test a flow-dependent B-matrix for 1D-Var of satellite data

• MAP D-PHASE testing period (summer-autumn 2007) :

• perturbations selected on the basis of the 1-month test

• verification -> availability of observations

• comparison with the other available mesoscale ensemble systems