ECMWF IFS Cycle 49r1: Major Upgrades – Technical Content


This article would be more suited for a meteorological technician or expert. It is also ideal for a weather enthusiast with a good knowledge about meteorology.

Summary

On 12 November 2024, there was an upgrade of the IFS (Integrated Forecasting System) of the ECMWF (European Centre for Medium-Range Weather Forecasts). This upgrade was the Cycle 49r1.

The major gains from this upgrade are in the near-surface wind predictions. Temperature predictions also improved. These gains are especially notable for the winter months in the northern hemisphere.

For the first time it has included the assimilation of the 2-metre temperature observation data, which, through a change in the Ensemble of Data Assimilation, helped to set and give estimates of the initial conditions and its uncertainty.

This upgrade also helped to determine the uncertainty of the ensemble forecast, through a new scheme.

Other improvements have been achieved in the use of observations. These observations are for the initial conditions of forecast. Additionally, there have been changes in the surface wind wave model.

Near-surface wind and temperature predictions

After assimilating the 2-metre temperature data in the 4D-Var system, an improvement was observed. This enhancement was seen in the physical representation of processes which, thus, positively affected the temperature forecast.

The impact of IFS Cycle 49r1 on two-metre temperature forecasts verified against SYNOP observations (percentage change in root-mean-square error) for December 2021 to February 2022, 20°–90°N. The grey rectangles show 95% confidence intervals.
The impact of IFS Cycle 49r1 on two-metre temperature forecasts verified against SYNOP observations (percentage change in root-mean-square error) for December 2021 to February 2022, 20°–90°N. The grey rectangles show 95% confidence intervals.
The chart shows the root-mean-square error (RMSE) of the ensemble mean for ten-metre wind speeds from Cycle 48r1 (blue) and Cycle 49r1 (red), verified against observations and averaged over the northern hemisphere. All scores are calculated using 50 perturbed members from 75 forecasts initialised daily between 1 December 2022 and 13 February 2023. The vertical bars represent 95% confidence intervals.
The chart shows the root-mean-square error (RMSE) of the ensemble mean for ten-metre wind speeds from Cycle 48r1 (blue) and Cycle 49r1 (red), verified against observations and averaged over the northern hemisphere. All scores are calculated using 50 perturbed members from 75 forecasts initialised daily between 1 December 2022 and 13 February 2023. The vertical bars represent 95% confidence intervals.

A new scheme for model uncertainty, called SPP ( Stochastically Perturbed Parametrization), replaced an old one, closing more the gap towards the sources of errors.

Relative differences in fair CRPS (continuous ranked probability score) of an ensemble using SPP and an ensemble using SPPT in the northern extratropics, for geopotential at 500 hPa verified against analyses. Positive values show higher forecast skill for the ensemble using SPP. Shown are combined scores for northern winter 2021/2022 and summer 2022 (282 start dates). The experiments use a resolution of 9 km (TCo1279) and eight perturbed members initialised from operational initial conditions. The vertical bars show 95% confidence intervals for the score differences.
Relative differences in fair CRPS (continuous ranked probability score) of an ensemble using SPP and an ensemble using SPPT in the northern extratropics, for geopotential at 500 hPa verified against analyses. Positive values show higher forecast skill for the ensemble using SPP. Shown are combined scores for northern winter 2021/2022 and summer 2022 (282 start dates). The experiments use a resolution of 9 km (TCo1279) and eight perturbed members initialised from operational initial conditions. The vertical bars show 95% confidence intervals for the score differences.

The use of new observations was also important. One example is the use of microwave imaging radiances over sea-ice surfaces for the 4D-Var.

Data coming from the AMSR2 (Advanced Microwave Scanning Radiometer 2) and GMI (Global Precipitation Measurement Microwave Imager), for sea ice and surroundings , improved the forecast near Antarctica (averaging 0,5% up to day 4) .

The ocean wind wave model has been improved, both scientifically and technically, in the horizontal grid which now measures 9 and 36 km in the medium-range and sub-seasonal forecasts, respectively.

That also helped the forecast of the wave height and, then, the atmospheric temperature forecast performance.

Normalised change in temperature forecast root-mean-square (RMS) error, measured against own analysis, showing the impact of all wave-model-related changes, for combined winter and summer seasons. Cross-hatching indicates statistical significance at a confidence level of 95%. Blue areas show a reduction in RMS error and hence a beneficial impact.
Normalised change in temperature forecast root-mean-square (RMS) error, measured against own analysis, showing the impact of all wave-model-related changes, for combined winter and summer seasons. Cross-hatching indicates statistical significance at a confidence level of 95%. Blue areas show a reduction in RMS error and hence a beneficial impact.

Final remarks

Wave parameters are passed to other Earth system components in sea-ice conditions. This points at the possible importance of wave–sea-ice interactions. Work has started on adding these interactions, with potential for future progress.

Future work should investigate whether introducing some processes, e.g.:  the role of sea spray and active wave breaking in the calculation of ocean surface fluxes is beneficial. These occur under extreme wind situations. So, when modeling such extremes becomes possible, it can improve calculations.

One important aim of ECMWF is to run forecasts at kilometre scale !

Images are by ECMWF and are licensed under CC BY 4.0

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.