This directory contains selected results from the paper ("mann et al 1998a"):

Mann, M.E., Bradley, R.S., and Hughes, M.K., Global-Scale Temperature Patterns
 and Climate Forcing Over the Past Six Centuries, Nature, 392, 779-787, 
 1998.

and from the book chapter ("mann et al 1998b"):

Mann, M.E., Bradley, R.S., and Hughes, M.K., Long-term variability in the El Nino
 Southern Oscillation and associated teleconnections, Diaz, H.F. & Markgraf, V.
 (eds) El Nino and the Southern Oscillation: Multiscale Variability and its Impacts on
 Natural Ecosystems and Society, Cambridge University Press, Cambridge, UK, in
 press, 1998. 


Included are:

I. from mann et al 1998a: 

nhmean.dat: reconstructed annual mean northern hemisphere areally-weighted
            mean temperature 1400-1980, actual instrumental nh annual
            areally-weighted mean temperature 1902-1995, and +/- 1 and
            2 sigma uncertainty intervals for NH mean reconstruction.

series used in  Figure 7:

fig7-nh.dat: this version of the nh series is slightly older than that
        shown in our figure 6 and given above, because figure 7 was not 
        updated in the final version of the paper (in which results
        from an expanded proxy data network were used to obtain slightly
        better calibrated/cross-validated climate pattern reconstructions).
        A correction pointing out this small oversight (and showing that
        none of the important conclusions are in any away affected) should
        be appearing in Nature.

fig7-solar.dat: raw solar irradiance series 
fig7-co2.dat: raw co2 data
fig7-dvi.dat: raw dvi series
fig7-corrs.dat: moving correlations with forcings, and confidence 
          levels for significance as determined from resampling (see paper).
          
          There are 7 columns: Column 1 is year,
          column 2 is co2 partial correlation, column 3 is solar partial
          correlation, column 4 is volcanism partial correlation, and
          columns 5,6,7 are one-sided positive 90%,95%,99% confidence 
          intervals for co2 (the values for solar are virtually
          identical).
 
          Volcanic forcing has lower conf thresholds as noted in
          paper. The one-sided (negative) confidence intervals are: 
          90% = -0.13, 95% = -0.16, 99% = -0.21

II. from mann et al 1998b

nino3.dat: reconstructed annual mean NINO3 index 1650-1980,
            actual instrumental nino3 index 1902,1993
            and +/- 1 and 2 sigma uncertainty intervals for 
            reconstructed NINO3 index.

III. Other results (from the reconstructions of Mann et al 1998a, but not 
shown therein):

latbands.dat: reconstructed annual mean temperature series 1600-1980 for 
           the 3 latitudinal bands: (1) 0-30 deg N, (2)30-70 deg N, and
           (3) southern hemisphere

latbands-raw.dat: actual instrumental data (1902-1993) for (1) and (2)
           above.

regional.dat: reconstructed annual mean temperature series 1650-1980 for
           the North American and European areally-averaged regions.
           Note that appreciable variance is lost in these regional
           reconstructions before 1750, where EOF #3 (an important
           contributor) has dropped out. Nonetheless, the regionally-resolved
           variance back to 1650 is useful.

regional-raw.dat: actual instrumental data (1902-1993) for the North
           American and European regions described above.

mannetal-latbands.ps: postscript file showing comparison between
           reconstructed entire NH mean and extratropical 30-70 N lat 
           band. Noteable is a greater mean warming for the 30-70N
           band than the NH as a whole since pre-industrial times,
           and a better-defined cooling during the early 19th century.
           This is attributed the larger trends and 19th century
           cooling over the large continental regions of the
           Northern hemisphere relative to other regions.