SO&P: Climate reconstructions

This section contains links to reconstructions of large-scale climate patterns or averages, from sets of multiple proxies. For reconstructions of local climate, based on single proxy records, see SO&P proxy data.

Data that are password-protected are indicated by

The World Data Center for Paleoclimatology, run by the National Climatic Data Centre (NCDC, formerly run by the National Geophysical Data Centre, NGDC), make many climate reconstructions available through their web site. It is not the purpose of this SO&P web site to reproduce such a comprehensive data base. Reconstructions that are not available at WDC-Paleoclimatology will be placed here, plus a limited range of other key data sets.

Please cite the references given in the data descriptions below when using any of these data!

• Comparison of many Northern Hemisphere temperature reconstructions from Briffa and Osborn (2002).
• Regional and hemispheric temperature reconstructions without loss of low-frequencies from tree-ring densities, from Briffa et al. (2001).
• Maps of April-September temperature anomalies, 1400-1960, from Briffa et al. (2002).
• Reconstruction from 1100-present of April-September temperatures averaged over western Europe [Guiot et al., 2005]
• Reconstruction from 1500-present of seasonal temperatures for a grid covering Europe [Luterbacher et al., 2004; Xoplaki et al., 2005]
• Reconstruction from 1500-present of seasonal precipitation for a grid covering Europe [Pauling et al., 2005].

• Briffa KR and Osborn TJ (2002) Blowing hot and cold. Science 295, 2227-2228.

• Briffa and Osborn (2002) showed a comparison of most of the published reconstructions of Northern Hemisphere temperature, to put the (then) new reconstruction of Esper et al. (2002) into the context of the others.

• To make the comparison between series that had been calibrated in different ways, Briffa and Osborn (2002) re-calibrated all series using the same target data.

• The choice of what target data to calibrate against is difficult to make. Briffa and Osborn (2002) chose to present results calibrated against annual temperature averaged over all land north of 20°N. Their annual results are available for use here.

• For many of the reconstructions, it could be argued that a better target for calibration would have been summer or growing-season temperature, because of the sensitivity of many of the proxies that were used. We have re-done the analysis of Briffa and Osborn (2002), but this time re-calibrating all records against the April-September temperature averaged over all land north of 20°N. These April-September results are available for use here.

• Briffa KR, Osborn TJ, Schweingruber FH, Harris IC, Jones PD, Shiyatov SG, Vaganov EA (2001) Low-frequency temperature variations from a norther tree ring density network. Journal of Geophysical Research 106, 2929-2941.

• Briffa et al. (2001) produced reconstructions of regional (sub-continental) scale April-September temperatures from 1400-present, using the tree-ring density measurements from the "Schweingruber" tree-ring network. They used a novel method of removing the age-related component of tree growth that did not remove the low-frequency (century timescale) climate signal.

• The calibrated regional data are available from Keith Briffa's website.

• Briffa KR, Osborn TJ, Schweingruber FH, Jones PD, Shiyatov SG and Vaganov EA (2002) Tree-ring width and density data around the Northern Hemisphere: part 2, spatio-temporal variability and associated climate patterns. The Holocene 12, 759-789.

• Briffa et al. (2002) produced reconstructions of gridded spatial temperature fields for every year from 1400 to 1960, using tree-ring density measurements from the "Schweingruber" tree-ring network. They reconstructed April-September temperature anomalies, and showed maps of every year from 1600-1887.

• The full set of maps, from 1400 to 1960, are made available here, in postscript, PDF and PNG, and individually or in decadal groups.

• The data themselves will not be made available until the method used to calibrate the reconstruction has been published (Briffa et al., 2002, used the gridded data, which is why the maps were published there, but did not describe their generation).

• Note that, as explained in Briffa et al. (2002), the tree-ring density data used to make these reconstructions had been standardised using a detrending-type method that removes multi-century climate variations. These maps should, therefore, be used mainly for inter-annual and inter-decadal time scale studies. The "age-band decomposition" method of Briffa et al. (2001, see above) has not been applied to these data.

• The reference for this data set is (see SO&P publications for reprints):
  • Guiot J, Nicault A, Rathgeber C, Edouard JL, Guibal F, Pichard G and Till C (2005) Last-millennium summer-temperature variations in western Europe based on proxy data. The Holocene 15, 489-500.
• The data available here are:
  • guiot05_eurot.xls (Microsoft Excel format).
  • guiot05_eurot.txt (ASCII plain text format).
    • A time series of April-September temperature anomalies averaged over western Europe (10W-40E, 35N-55N). Temperatures are given as anomalies in K relative to the 1961-1990 mean. The 90% uncertainty range is also given (subtract "lower bar" from reconstructed value to obtain the 5%ile and add "upper bar" to reconstructed value to obtain the 95%ile), as are the observed instrumental temperatures. Missing values are indicated by -999.99.

• The references for this data set are (see SO&P publications for reprints):
  • Luterbacher J, Dietrich D, Xoplaki E, Grosjean M and Wanner H (2004) European seasonal and annual temperature variability, trends and extremes since 1500. Science 303, 1499-1503.
  • Xoplaki E, Luterbacher J, Paeth H, Dietrich D, Steiner N, Grosjean M and Wanner H (2005) European spring and autumn temperature variability and change of extremes over the last half millennium. Geophysical Research Letters 32, L15713.
• The data available here are:
  • temp_luterbacher_wi.txt.gz (Winter fine-resolution 0.5° grids, 5MB)
  • temp_xoplaki_sp.txt.gz (Spring fine-resolution 0.5° grids, 5MB)
  • temp_luterbacher_su.txt.gz (Summer fine-resolution 0.5° grids, 5MB)
  • temp_xoplaki_au.txt.gz (Autumn fine-resolution 0.5° grids, 5MB)
    • Gridded (0.5° resolution, lon x lat = 130 x 70) temperature reconstructions over all European land areas for winter, spring, summer and autumn.
    • Please note that the summer and winter reconstructed values are slightly different from Luterbacher et al. (2004) for earlier centuries because the reconstructions have been redone (i) fitting to Mitchell et al. observational data rather than New et al. (2000), (ii) using only instrumental temperature series and temperature-sensitive documentary and natural proxies, and (iii) including a few more instrumental series compared to Luterbacher et al. (2004).
    • These files have been compressed with gzip.
    • All these files cover the area 35.25N-69.75N / 24.75W-39.75E (all coordinates given here denote the centre of each box). Non-land grid boxes are indicated by values set to -99.999.
    • Data run from 1500 to 2002. Data from 1901-2002 are observational estimates from the Mitchell et al. (2004) dataset. Data prior to 1901 are reconstructed values (see data set references for details).
    • The files are ASCII plain text, with the following format:
      

      line 1: year season (first year=1500, season is 13=winter, 14=spring, 15=summer, 16=autumn)
      line 2: top row of grid (69.75N, 24.75W-39.75E)
      line 3: 2nd row of grid (69.25N, 24.75W-39.75E)
      etc.
      line 71: 70th row of grid (35.25N , 24.75W-39.75E)
      
      The following 70 lines describe the grid for the second year and so on.
      

    • Here is an example of the data set contents (reconstructed temperature for winter 1500):
      

• The reference for this data set is (see SO&P publications for reprints):
  • Pauling A, Luterbacher J, Casty C and Wanner H (2005) 500 years of gridded high-resolution precipitation reconstructions over Europe and the connection to large-scale circulation. Climate Dynamics (accepted).
• The data available here are:
  • prec_pauling_wi.txt.gz (Winter fine-resolution 0.5° grids, 8MB)
  • prec_pauling_sp.txt.gz (Spring fine-resolution 0.5° grids, 8MB)
  • prec_pauling_su.txt.gz (Summer fine-resolution 0.5° grids, 8MB)
  • prec_pauling_au.txt.gz (Autumn fine-resolution 0.5° grids, 8MB)
    • Gridded (0.5° resolution, lon x lat = 140 x 82) precipitation reconstructions over all European land areas for winter (prec_pauling_wi.txt), spring (prec_pauling_sp.txt), summer (prec_pauling_su.txt) and autumn (prec_pauling_au.txt).
    • These files have been compressed with gzip.
    • All these files cover the area 30.25N-70.75N / 29.75W-39.75E (all coordinates given here denote the centre of each box). Non-land grid boxes are indicated by values set to -99.999.
    • Data run from 1500 to 2000. Data from 1901-2000 are observational estimates from the Mitchell et al. (2004) dataset. Data prior to 1901 are reconstructed values (see data set reference for details: Pauling et al. (2005)).
    • The files are ASCII plain text, with the following format:
      

      line 1: year season (first year=1500, season is 13=winter, 14=spring, 15=summer, 16=autumn)
      line 2: top row of grid (70.75N, 29.75W-39.75E)
      line 3: 2nd row of grid (70.25N, 29.75W-39.75E)
      etc.
      line 83: 82nd row of grid (30.25N , 29.75W-39.75E)
      
      The following 83 lines describe the grid for the second year and so on.
      

    • Here is an example of the data set contents (reconstructed precipitation for winter 1500):
      
  • precmean_pauling_wi.txt (Winter European-mean time series)
  • precmean_pauling_sp.txt (Spring European-mean time series)
  • precmean_pauling_su.txt (Summer European-mean time series)
  • precmean_pauling_au.txt (Autumn European-mean time series)
    • Area average of the European precipitation reconstructions for winter (precmean_pauling_wi.txt), spring (precmean_pauling_sp.txt), summer (precmean_pauling_su.txt) and autumn (precmean_pauling_au.txt). The area average if calculated over all land grid boxes in the 30N-71N, 30W-40E domain.
    • For these spatial averages uncertainties are provided as labelled in the file (see data set reference for details: Pauling et al. (2005)).
    • Data run from 1500 to 1983. All data are reconstructed values (including values after 1901).