Kein Folientitel

1
Welcome !

• SRNWP-PEPS workshop
• 6th April 2005, ARPA-SIM, Italy
• products
• validation
• further developement
• rights of use

Santuario della Madonna di S. Lucia
2
Workshop
products
Sebastian Trepte
3
Workshop
validation
Question Can we observe a clustering of the
forecasts according to the membership of a
consortium ?
Values of 0 lt c lt 1 indicate more similar
forecasts within the consortium than in the total
ensemble
Parameters Maximum temperature Tmax, total
precipitation RR (12 h accumulation) Forecast
period 00 UTC 06...18h daily Domain
maximum overlapping area (see above) Analysis
period Average over February 2005
4
Workshop
validation
Areas having maximum number of ensemble members
for each consortium
ALADIN
COSMO
3 models
8 models
UKMO
HIRLAM
2 models
2 ? 6 models
5
Workshop
validation

• All COSMO members use boundary conditions from
  different
• models (GME, ECMWF, EuroLM)
• The other Consortia have a single global model

6
Workshop
validation

• Comparison with COSMO-LEPS
• Scoring probabilistic forecasts
• - error measures
• - FBI, POD, FAR, ETS, HSS, Odds Ratio
• - BS, BSS, RPS, ROC
• Scale-/Object oriented techniques
• - contiguous rain area method (Ebert McBride)
• Severe weather Problem
• - linear error in probability space (LEPS)
• Online verification

Coordination of activities ? -gt WG Verifcation
7
Workshop
further developement
Ensemble Calibration
Calibrated Intervals or events that we declare
to have probability P happen a proportion P of
the time Sharp Prediction intervals are
narrower on average than those obtained from
climatology the narrower the better
Methodologies Bayesian Model Averaging
(BMA),...... ?
8
Workshop
further developement
Using Bayesian Model Averaging (BMA) to
calibrate forecast ensembles
Adrian E. Raftery, Fadoua Balabdaoui, Tilmann
Gneiting and Michael Polakowski Department of
Statistics, University of Washington, Seattle,
Washington
The model is estimated from a training set of
recent data by maximum likelihood using the EM
algorithm. Good results with a 25-day training
period.
9
Workshop
further developement
In experiments with temperature and pressure in
the Pacific Northwest, BMA was calibrated, sharp,
and gave good deterministic forecasts work on
precipitation is in progress Software R package
EnsembleBMA is available Source www. stat.
washington. edu/ raftery www. stat. washington.
edu/ MURI

10
Workshop
further developement
Concerning GLOBAL PEPS According to most skill
measures, these hybrid configurations outperform
the ECMWF-EPS at short range for most variables,
regions and thresholds from Test of a Poor
Mans Ensemble Prediction System for short range
probability forecasting Arribas, A., Robertson,
K.B., Mylne, K.R.

• Hybrid LAM-Ensemble ?
• concatenate SRNWP-PEPS with other ensemble
  systems
• INM Ensemble
• Meteo France PEACE Ensemble
• UK-Met Office LAM
• met-norway LAMEPS
• COSMO-LEPS
• ?

11
Workshop
further developement
Hydrological Ensemble Forecasts for the MULDE
catchment
Hybrid Ensemble COSMO-LEPS (120h) SRNWP-PEPS
(48h)
consistent forecast scenarios of precipitation
for the Mulde catchment up to 120h
12
Workshop
further developement

• The SRNWP-PEPS consits of different model grids
  with different horizontal and vertical
  resolutions.
• Question
• How can we account for these differences in an
  appropriate way ?
• Estimate a statistical relation between low and
  high resolution analysis (or initial
  states) ?
• Neighbourhood Ensemble ?

13
Workshop
further developement
Neighbourhood Ensemble ? consider all gridpoints
within a given distance of a point
Neighbourhood members from different grids should
not have equal weights Systematic errors (e.g.
due to orography) should be corrected
14
Ensemble
Generation PEPS grid with a grid spacing of
0.0625 (7 km) covering Europe
Methodology
The ensemble size depends on location and every
PEPS grid point has its own probability
distribution
15
Workshop
further developement
Additional Contributions ?
Comments ?
16
Workshop
rights of use
Request to DWD Distribute request to
contributing NWS NWS give their permission
Jean Quiby
17
Thank you to all contributing Weather Services !
any further questions or remarks ?
Thank you to ARPA-SIM to host the meeting !!!!
18
2.8 km grid spacing forecasts every 3 hours
(18h)
Workshop
further developement
LMK
Milestones

• Summer 2005Prototype version of the LMK-System
  with LHN is running in a quasi-operational
  mode.
• Early 2006Start of a pre-operational
  test-phase.
• End 2006Start of the operational application