Fig. 1: Comparison of the globally averaged latitude-weighted ACC (first and second rows) and RMSE (third and fourth rows) of the HRES (dark green lines), GraphCast (organge lines), and FuXi (light blue lines) for 4 surface variables, such as MSL, T2M, U10, and V10, and 4 upper-air variables at the pressure level of 500 hPa, including Z500, T500, U500, and V500, using testing data from 2018. | npj Climate and Atmospheric Science

Fig. 1: Comparison of the globally averaged latitude-weighted ACC (first and second rows) and RMSE (third and fourth rows) of the HRES (dark green lines), GraphCast (organge lines), and FuXi (light blue lines) for 4 surface variables, such as MSL, T2M, U10, and V10, and 4 upper-air variables at the pressure level of 500 hPa, including Z500, T500, U500, and V500, using testing data from 2018.

From: FuXi: a cascade machine learning forecasting system for 15-day global weather forecast

Fig. 1

FuXi and GraphCast are evaluated against the ERA5 reanalysis dataset, and ECMWF HRES is evaluated against HRES-fc0.

Back to article page