ECM model upgrade 49r1 - 12th November 2024

[WYorksWeather]

Just seen on the ECMWF website that a new upgraded version of the model is due out on 12th November. There are a number of key changes to be aware of - I have quoted some of the most important below.

The first section quoted here is a discussion of the meteorological impacts of the changes:

Quote

Meteorological impact

Medium range

• 2m temperature and 10m wind speed forecasts are improved in the short and medium ranges (due to land surface model upgrades, and to data assimilation changes related to 2m temperature, soil moisture and snow) .
• For 2m temperature forecasts the largest impacts are for the winter months in the northern hemisphere, where the Continuous Ranked Probability Score (CRPS) is improved by 11% at day 1 and 2% at day 10.
• 10 m wind speed forecasts are improved throughout the year. The largest impacts are for the winter months in the northern hemisphere, where the CRPS is improved by 12% at day 1 and 6% at day 10
• Tropical Cyclones have (for multiple reasons) increased spread in depth and slightly increased spread in track in the ensemble in the first 3-4 days, whilst errors in extreme wind metrics are slightly reduced. Benefits are most noteworthy for strong systems. 
• There is more rapid (=improved) melting of small amounts of snow on the ground (due to changes in the snow cover and snow density representation).
• 2m temperatures in large built up areas have improved (due to inclusion of urban tiles in the land surface scheme).
• Soil moisture handling exhibits some notable changes (due to changes in land surface property specifications, in the handling of land surface processes and in soil moisture data assimilation):
• Less soil moisture across the world as a whole
• Large differences in certain regions compared to cycle 48r1
• Higher spatial variability in soil moisture levels
• Over oceans peak 10m mean wind and gust values around windstorms are slightly increased (related to changes in coupling between waves and winds)

Sub-seasonal range

• The most robust impacts on weekly mean forecast anomalies are small but statistically robust changes in the ensemble spread, which are driven by the switch from SPPT to SPP
• Positive impact on Madden-Julian Oscillation (MJO) at week 3-4

The second is around changes to the distribution / availability of the model:

Quote

 

HRES and ENS Control run will be computationally identical.

Both HRES and ENS will run:

• to day 15 (step 360) for the 00 and 12 runs, and
• to day 6 (step 144) for the 06 and 18 runs (as part of BC Optional Programme)
• There will be no change in vertical or horizontal resolution of the Atmospheric model 
• There will be no ENS configuration changes

 

In summary it looks like we should get better temperature forecasts, and a longer range for the main ECM output, out to day 15. Will be interesting to see how the new version performs in practice.

The full description of the upgrade is available here:

Implementation of IFS Cycle 49r1 - Forecast User - ECMWF Confluence Wiki

CONFLUENCE.ECMWF.INT

Other thoughts welcome - I thought this was worth a separate thread out of the main model thread to discuss.

[The PIT]

As you say how it actually performs in practise is the real question.

[Derecho]

I think this will help bust the myth that the GFS model is better at picking up long term trends. People tend to look at the OP runs so now the longer range output will be seen by more people.

[WYorksWeather]

Implementation time is below. Another detail I found is that the AIFS is not changing, so no changes to the AI model.

 

Quote

Cycle 49r1 will be implemented on 12 November 2024 run 06 UTC

I'd continue to advise that if you want the full details, keep an eye on this page.

Implementation of IFS Cycle 49r1 - Forecast User - ECMWF Confluence Wiki

CONFLUENCE.ECMWF.INT

For example, some new parameters being introduced - not sure whether these will be of interest. A lot relate to the 'thermal comfort' measures - we will have things like feels-like temperatures, heat index, and so on as part of the new upgrade, each of which have slightly different calculation methodologies.

New parameters: heat and cold indices, mean radiant temperature and globe temperature - Forecast User - ECMWF Confluence Wiki

CONFLUENCE.ECMWF.INT