Home page  
Home   Your Room   Login   Contact   Feedback   Site Map   Search:  
Discover this product  
About Us
Overview
Getting here
Committees
Products
Forecasts
Order Data
Order Software
Services
Computing
Archive
PrepIFS
Research
Modelling
Reanalysis
Seasonal
Publications
Newsletters
Manuals
Library
News&Events
Calendar
Employment
Open Tenders
   
User Guide to ECMWF Forecast Products > Derived products from the EPS > The Extreme Forecast Index (EFI) > 
The Extreme Forecast Index (EFI) The cumulative distribution function  
   

The EFI reference climate

Browse
The EFI reference climate
The cumulative distribution function
Calculating the EFI
The interpretation of the EFI
EFI maps
Plans
 
 

In the EFI scheme, the probabilities are compared to the model climate (M-climate) distribution for the chosen location, time of year and lead time. The underlying assumption is that, if a forecast is anomalous or extreme, relative to the M-climate, the real weather is also likely to be anomalous or extreme compared to the real climate.

The M-climate is based on five consecutive weekly 32-day re-runs of the current EPS model, with four perturbed and one unperturbed initial conditions and started from ERA-Interim re-analyses from each of the last 18 years. The resolution decreases with forecast range exactly as in the operational EPS. This procedure allows seasonal variations and model changes to be taken into account, as well as model drift. An additional advantage of the EFI is that it compensates for areas where the ECMWF model climate is less accurate, such as in the tropics.

The model climate for the EFI calculations on Sunday 31 October 2010 at 12 UTC is, for example, prepared from five forecast runs from five Thursdays, centred on the 28 October, i.e.14, 21 and 28 October and 4 and 11 November, for all the 18 years, totalling 450 forecasts (see Figure 50).




Top of page 08-02-2012
 
   Compare Pages Page Details         © ECMWF
The Extreme Forecast Index (EFI) The cumulative distribution function  
shim shim shim