Datasets
The reliability diagram shows the reliability of the ECMWF seasonal forecast system (SEAS5) with ...
Interval/period: N/A
This chart shows the Relative Operating Characteristics (ROC) diagram for the three-month ...
Interval/period: N/A
This chart shows the spatial variation in the Relative Operating Characteristics (ROC) skill ...
Interval/period: N/A
The ECMWF seasonal forecasts (SEAS5) are produced every month with a 51-member ensemble at a ...
Interval/period: N/A
The 850 hPa level is usually just above the boundary layer and at this level the day-night variation in temperature is generally negligible...
Interval/period: N/A
This chart shows 7-day mean anomalies of 500hPa geopotential height from the ECMWF Sub-seasonal ...
Interval/period: N/A
500-1000 hPa thickness is a measure of the mean temperature of a column of the atmosphere between these pressure levels and can be used to distinguish between warm and cold air masses and...
Interval/period: N/A
This chart shows the spatial variation in the Anomaly Correlation Coefficient (ACC) for the ...
Interval/period: N/A
Aurora: a deep learning-based system developed by Microsoft. It is initialised with ECMWF analysis. Aurora operates at 0.1° resolution.
Interval/period: N/A
Aurora: a deep learning-based system developed by Microsoft. It is initialised with ECMWF analysis. Aurora operates at 0.1° resolution.
Interval/period: N/A
Aurora: a deep learning-based system developed by Microsoft. It is initialised with ECMWF analysis. Aurora operates at 0.1° resolution.
Interval/period: N/A
FourCastNet v2-small:a deep learning-based system developed by NVIDIA in collaboration with researchers at several US universities.It is initialised with ECMWF analysis. FourCastNet operates at 0.25° resolution.
Interval/period: N/A
FourCastNet v2-small:a deep learning-based system developed by NVIDIA in collaboration with researchers at several US universities.It is initialised with ECMWF analysis. FourCastNet operates at 0.25° resolution.
Interval/period: N/A
FourCastNet v2-small:a deep learning-based system developed by NVIDIA in collaboration with researchers at several US universities.It is initialised with ECMWF analysis. FourCastNet operates at 0.25° resolution.
Interval/period: N/A
GraphCast (Google DeepMind): a deep learning-based system developed by Google DeepMind.It is initialised with ECMWF analysis. GraphCast operates at 0.25° resolution.
Interval/period: N/A
Interval/period: Fri, 01/01/1982 - Tue, 12/31/2024
These diagrams compare Continuous Ranked Probability Skill Scores (CRPSS) of ECMWF with ...
Interval/period: N/A
These charts show clustering of ENS members based on the 500 hPa height forecasts. ...
Interval/period: N/A
**Note:** In **June 2023** ECMWF implemented a **major upgrade ...**
Interval/period: N/A
These plots compare recent IFS and experimental AIFS daily verification scores for 500 hPa ...
Interval/period: N/A
The 850 hPa level is usually just above the boundary layer and at this level the day-night variation in temperature is generally negligible...
Interval/period: N/A
In the northern hemisphere the air rotates anti-clockwise around low contour centres and clockwise around the high contour centres. In the southern hemisphere the air rotates clockwise around low contour centres and ...
Interval/period: N/A
The air flow is not the same at all levels and can be very different in strength and direction when associated with vigorous weather systems...
Interval/period: N/A
These charts show the difference between the latest HRES Long Window Data Analysis (LWDA) for the main synoptic hours of 00, 06, 12 and 18UTC, and its corresponding background field taken from the previous HRES LWDA run with the same validity time...
Interval/period: N/A
The 850 hPa level is usually just above the boundary layer and at this level the day-night variation in temperature is generally negligible...
Interval/period: N/A