Twentieth Century Reanalysis (V2c): Opportunities for Improvement

  1. 07/25/2018: Incorrect values were discovered at 3Z and 6Z for Aug. 31st, 1950 cduvb and duvb. These files have been replaced for 8xDaily, daily, and monthly time scales.
  2. 11/16/2015: Incorrect values were discovered at 12z and 15z for Sep 14th 1990 single level Gaussian gridded files. These files have been replaced for the 8xDaily, daily, and monthly time scales.
    Incorrect values were discovered for single level Gaussian spread files for 3z Aug18 1855, 6z May 8 1910, 3z May 22 1950, and 9z Jun 2 1990. These files have been replaced for the 8xDaily, daily, and monthly time scales.
  3. 10/22/2015: Incorrect values were discovered at 6z and 9z for Aug 19 1910 single level Gaussian gridded files. These files have been replaced for the 8xDaily, daily, and monthly time scales.
  4. A low pressure bias in marine pressures from the US Maury Collection (Woodruff et al. 2005 , Wallbrink et al. 2009, ) appears to have affected the resultant 20CRv2c mass-related fields (e.g., pressure, geopotential height) from 1851 to about 1865. The figure illustrates the effect on the globally-averaged annual mean sea level pressure.

    slp difference 20cr vs 20crv2c SLP

    Globally and annually averaged sea level pressure from 20CR version 2 (labelled 20CR, red dashed) and the new 20CR version 2c (blue). The figure was generated using the Web-based Reanalysis Intercomparison Tool (WRIT, https://reanalyses.org/atmosphere/writ, Smith et al. 2015)

    Future work will address how to correctly calibrate these measurements and make the period 1851-1865 consistent with later fields.
  5. Increase the spread of the analysis ensemble members.