Three types of charts are shown below:

Cumulative bias and root-mean-square error for heights, temperatures, vector wind, and relative humidity versus rawinsondes for the 12-h, 24-h, and 48-h forecasts of operational NGM (black solid lines) and the parallel NGM (dashed red lines) at mandatory pressure levels. These statistics have been computed for all NGM model grid points inside NCEP grid #236 over the CONUS and inside NCEP grid #216 over Alaska.

Time series of 12-h, 24-h, and 48-h NGM forecast surface temperature bias and root-mean square error over the CONUS

Equitable threat and bias score of 24-h NGM forecast precipitation (valid at 12Z) using the NWS River Forecast Center rain gauge network for verification

TABLE 1: Statistics for the parallel test from 12Z 16 December 1999 - 12Z 22 February 2000

TABLE 2: Mean Bias/RMS surface temperature errors (deg C) for the parallel test from 12Z 16 December 1999 - 12Z 22 February 2000

TABLE 3: Statistics for warm season retrospective from 00Z 2 July 1999 - 12Z 2 August 1999

TABLE 4: Mean Bias/RMS surface temperature errors (deg C) for warm-season retrospective run from 00Z 2 July 1999 - 12Z 2 August 1999

TABLE 5: Statistics for 24-h accumulated precipitation