What was done
The authors used a coupled atmosphere-ocean model simulation of the climate of the past millennium as a surrogate climate to test the skill of the empirical reconstruction methods used by Mann et al. (1998, 1999) in deriving their thousand-year "hockeystick" temperature history of the Northern Hemisphere (NH).  This they did by (1) generating a number of pseudo-proxy temperature records by sampling a subset of the model's simulated grid-box temperatures representative of the spatial distribution of the real-world proxy temperature records used by Mann et al. in creating their hockeystick history, (2) degrading these pseudo-proxy records with statistical noise, (3) regressing the results against the measured temperatures of the historical record, and (4) using the relationships thus derived to construct a record they could compare against their original model-derived surrogate temperature history.

What was learned
Von Storch et al. report that the centennial variability of the NH temperature was underestimated by the regression-based methods they applied, suggesting, in their words, that past variations in real-world temperature "may have been at least a factor of two larger than indicated by empirical reconstructions."  The unfortunate consequences of this result are readily evident in the reduced degree of Little Ice Age cooling and Medieval warming that result from the fault-prone techniques employed by Mann et al.

What it means
In an accompanying commentary, Osborn and Briffa (2004) state that "if the true natural variability of NH temperature is indeed greater than is currently accepted," which they appear to suggest is likely the case, "the extent to which recent warming can be viewed as 'unusual' would need to be reassessed."  That this reassessment is sorely needed is also suggested by the fact that what von Storch et al. refer to as "empirical methods that explicitly aim to preserve low-frequency variability (Esper et al., 2002)" show much more extreme Medieval warming and Little Ice Age cooling than do the reconstructions of Mann et al., which suffer from the problems elucidated in this important new study of von Storch et al.

In light of these observations, it is becoming ever more evident that the temperature record of Esper et al. is likely to be much more representative of reality than is the IPCC-endorsed record of Mann et al., and that the lion's share of the warming experienced since the end of the Little Ice Age occurred well before mankind's CO2 emissions significantly perturbed the atmosphere, which indicates that the majority of post-Little Ice Age warming was due to something other than rising atmospheric CO2 concentrations, which in turn suggests that the lesser warming of the latter part of the 20th century may well have been due to something else as well.

References
Esper, J., Cook, E.R. and Schweingruber, F.H.  2002.  Low-frequency signals in long tree-ring chronologies for reconstructing past temperature variability.  Science 295: 2250-2253.

Mann, M.E., Bradley, R.S. and Hughes, M.K.  1998.  Global-scale temperature patterns and climate forcing over the past six centuries.  Nature 392: 779-787.

Mann, M.E., Bradley, R.S. and Hughes, M.K.  1999.  Northern Hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations.  Geophysical Research Letters 26: 759-762.

Osborn, T.J. and Briffa, K.R.  2004.  The real color of climate change?  www.scienceexpress.org/ 30 September 2004.