Reconstruction of Antarctic near-surface temperatures (1958-2012)


This project has been financially supported by the National Science Foundation through grants PLR-1049089 and PLR-1341695, and by NASA through grant NN12AI29G.

This webpage provides the reconstructed temperature data used in the following paper:
Nicolas, J. P., and D. H. Bromwich, 2014: New reconstruction of Antarctic near-surface temperatures: Multidecadal trends and reliability of global reanalyses. J. Climate, 27, 8070-8093, doi:10.1175/JCLI-D-13-00733.1.
The pdf of the paper (along with its supplementary material) can be downloaded from our publication webpage.

Please read this before downloading the data:
  • If you publish work that uses the reconstruction data, please acknowledge the J. Climate paper mentioned above.
  • Whenever possible, please include the URL of this webpage as the source of the data.
  • If you also use (some of) the reanalysis data provided on this webpage, please cite the relevant papers [Saha et al. (2010), Rienecker et al. (2011), Dee et al. (2011)].
The reconstruction files are provided both in NETCDF and ASCII formats and include the following data sets:
  • The data from the four reconstructions described in the paper (each one based on a different reanalysis data set). These data consist of monthly mean temperature anomalies (in K) from Jan. 1958 through Dec. 2012 distributed on a 60x60 km Cartesian grid covering the Antarctic continent. The temperature anomalies are estimated over land only. Ocean grid points are assigned missing values (-999) in the NETCDF files and are not included in the ASCII files (to reduce the file size).
  • The respective 2-meter temperature estimates from the four reanalysis data sets (CFSR, MERRA, ERA-Interim analysis, ERA-Interim forecast) used to produce the four reconstructions. The data consist of (absolute) temperatures (in K) from Jan. 1979-Dec. 2009 regridded onto the 60x60 km grid (ocean grid points are masked out).
  • The lat-lon coordinates of the grid points of the 60x60 km Cartesian grid.
The layout of the NETCDF files containing reconstruction data is described here with the output from the ncdump -h command. Similar information for the NETCDF files containing reanalysis data can be found here.

In the ASCII files, the 4163 rows represent the spatial dimension (i.e., number of land grid points) and the columns represent the time dimension (months). The reconstruction files contain 660 columns (i.e., 660 months between Jan 1958 and Dec. 2012) while the reanalysis files contain 372 columns (372 months between Jan. 1979 and Dec. 2009).

I hope to have provided all the information needed to understand the content of the files. If you have any questions or problems downloading the files, feel free to contact me at nicolas.7[at]osu.edu. And thanks again for your interest in the reconstruction! J.P.N.

Reconstructions files in NETCDF format:
  • recon_cfsr_195801-201212.nc.gz
  • recon_merra_195801-201212.nc.gz
  • recon_eraia_195801-201212.nc.gz
  • recon_eraif_195801-201212.nc.gz
Reconstructions files in ASCII format:
  • recon_cfsr_195801-201212.dat.gz
  • recon_merra_195801-201212.dat.gz
  • recon_eraia_195801-201212.dat.gz
  • recon_eraif_195801-201212.dat.gz
Reanalysis files in NETCDF format:
  • cfsr_t2m_197901-200912.nc.gz
  • merra_t2m_197901-200912.nc.gz
  • eraia_t2m_197901-200912.nc.gz
  • eraif_t2m_197901-200912.nc.gz
Reanalysis files in ASCII format:
  • cfsr_t2m_197901-200912.dat.gz
  • merra_t2m_197901-200912.dat.gz
  • eraia_t2m_197901-200912.dat.gz
  • eraif_t2m_197901-200912.dat.gz
Coordinate file in NETCDF format:
  • lat_lon.nc.gz
Coordinate files in ASCII format:
  • lat.dat.gz
  • lon.dat.gz

Other data sets used in the paper include:
  • the reconstruction from Monaghan et al. (2008) revised in 2010, courtesy of Andrew Monaghan (NCAR).
  • the main reconstruction from Steig et al. (2009), available at http://www.usap-data.org/entry/NSF-ANT04-40414/2009-09-12_11-10-10/.
  • the RLS reconstruction from O'Donnell et al. (2011), available at http://www.climateaudit.info/data/odonnell/. If for some reason this link doesn't work, a copy of the files can be found here.

Last updated: 12 November 2014