Extended Data Table 2 The forecast time gain of Pangu-Weather

From: Accurate medium-range global weather forecasting with 3D neural networks

  1. It lists the 10 weather variables shown in Fig. 1, namely, Z500/T850/T2M/U10/V10 on which both operational IFS3 and FourCastNet2 reported, and Z850/T500/Q500/U500/V500 on which only operational IFS reported. Here, Z500/T500/Q500/U500/V500 indicates the geopotential, temperature, specific humidity, and u-component and v-component of wind speed at 500 hPa. Z850/T850 indicates the geopotential and temperature at 850 hPa. T2M indicates the 2 m temperature, and U10/V10 indicates the u-component and v-component of 10 m wind speed. To calculate the values for a given variable, we first fetched the RMSE value reported by Pangu-Weather in a 168-hour forecast, and then estimated the forecast time gain, denoted by ∆t, so that the compared algorithm (e.g. operational IFS) report the same RMSE value when the lead time is 168 h minus ∆t.
Back to article page