Datasets

Scores of forecasts of upper-air parameters by AIFS - experimental machine learning model

These plots compare recent IFS and experimental AIFS verification scores for 500 hPa ...

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Scores of forecasts of upper-air parameters by experimental machine learning models

These plots compare recent IFS and experimental AIFS verification scores for 500 hPa ...

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: Aurora ML model: 500 hPa geopotential height and 850 hPa temperature

Aurora: a deep learning-based system developed by Microsoft. It is initialised with ECMWF analysis. Aurora operates at 0.1° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: Aurora ML model: Temperature and geopotential at various pressure levels

Aurora: a deep learning-based system developed by Microsoft. It is initialised with ECMWF analysis. Aurora operates at 0.1° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: Aurora ML model: Wind and geopotential heights at various pressure levels

Aurora: a deep learning-based system developed by Microsoft. It is initialised with ECMWF analysis. Aurora operates at 0.1° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: FourCastNet ML model: 500 hPa geopotential height and 850 hPa temperature

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.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: FourCastNet ML model: Temperature and geopotential at various pressure levels

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.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: FourCastNet ML model: Wind and geopotential heights at various pressure levels

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.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: GraphCast ML model: 500 hPa geopotential height and 850 hPa temperature

GraphCast (Google DeepMind): a deep learning-based system developed by Google DeepMind.It is initialised with ECMWF analysis. GraphCast operates at 0.25° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: GraphCast ML model: Temperature and geopotential at various pressure levels

GraphCast (Google DeepMind): a deep learning-based system developed by Google DeepMind.It is initialised with ECMWF analysis. GraphCast operates at 0.25° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: GraphCast ML model: Wind and geopotential heights at various pressure levels

GraphCast (Google DeepMind): a deep learning-based system developed by Google DeepMind.It is initialised with ECMWF analysis. GraphCast operates at 0.25° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: Pangu-Weather ML model: 500 hPa geopotential height and 850 hPa temperature

Pangu-Weather: a deep learning-based system developed by Huawei. It is initialised with ECMWF analysis. Pangu-Weather operates at 0.25° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: Pangu-Weather ML model: Temperature and geopotential at various pressure levels

Pangu-Weather: a deep learning-based system developed by Huawei. It is initialised with ECMWF analysis. Pangu-Weather operates at 0.25° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Experimental: Pangu-Weather ML model: Wind and geopotential heights at various pressure levels

Pangu-Weather: a deep learning-based system developed by Huawei. It is initialised with ECMWF analysis. Pangu-Weather operates at 0.25° resolution.

calendar_today Interval/period: N/A

open_in_newview in Open Charts

Daily scores of forecasts of upper-air parameters by experimental machine learning models

These plots compare recent IFS and experimental AIFS daily verification scores for 500 hPa ...

calendar_today Interval/period: N/A

open_in_newview in Open Charts