Annual air temperatures from the US HCN for the north central United States show a high degree of interannual variability and overall trends of warming from 1895 until the mid 1930

1
A Test of Borehole Paleoclimatology William D.
Gosnold, Jr., Xiquan Dong, Julie Popham
PP52A-0659Climate Change in the Recent Past
Integrating Meteorological, Proxy, Borehole, and
Modeled Climate Reconstructions
This research is supported by National Science
Foundation Award ATM - 0318384
It has been argued that differences between
multi-proxy reconstructions of temperature change
and inversions of borehole temperatures are
largely due to the fact that snow cover decouples
air and ground temperatures resulting in biased
trends in the GST recorded in boreholes Mann et
al., 2003 Mann and Schmidt, 2003 Schmidt and
Mann, 2004. Chapman et al., 2004 counter that
this interpretation of the differences are based
in selective and inappropriate presentation of
model results by Mann and Schmidt, and provide
examples of good correlations between SAT and
GST. We address the argument by assessing the
accuracy of direct coupling between ground
surface temperatures (GST) and surface air
temperatures (SAT) in regions affected by
seasonal snow cover in two critical tests (1)
Comparison of borehole T-z profiles to synthetic
T-z profiles generated from century-long daily,
monthly, and annual SAT data from the north
central US and Canada. (2) Comparison of
changes in borehole T-z profiles over two or more
decades to changes in SAT recorded at automated
weather stations near the borehole sites. The
upper section of this poster focuses on the first
test and the lower section of the poster focuses
on the second test. The results from Alberta and
Saskatchewan were presented by Jacek Majorowicz
this morning in the oral session of PP19.
Annual air temperatures from the US HCN for the
north central United States show a high degree of
interannual variability and overall trends of
warming from 1895 until the mid 1930s followed
by cooling from the late -1930s to the early
1970sa and warming from 1970 to present. The
climate divisions shown are those that contain
borehole sites.
The plots above show synthetic T-z profiles at
10-year intervals for the period of record
generated by the SAT data in the top panels.
Two-dimensional finite-difference conductive
models used monthly temperatures from the period
of record to generate the synthetic profiles.
Linear least-squares fits to SAT time series were
used to translate the SAT data so that the
initial temperatures, i.e., the pre-observational
mean, were 0 C.
Fig. 1 (below). Borehole sites included in this
study are shown as red circles and climate
stations are shown as green squares. Boreholes
in Kansas, Nebraska, South Dakota, North Dakota,
Manitoba, and Ontario were drilled specifically
for heat flow measurements. Sites in
Saskatchewan and Alberta are holes of opportunity
that were drilled for mineral or oil exploration.
Climate stations are part of the US Historical
Climatology Network and climate data were
obtained from National Climatic Data Center, High
Plains Regional Climate Center, and Environment
Canada. A critical aspect of the heat flow
boreholes in the US is that they were sited
specifically to be free from microclimate
effects, land use change, and terrain effects
that could contaminate the temperature-depth
profiles.
The plots above compare synthetic T-z profiles
generated by climate data from 1895 to the date
of borehole logs with borehole T-z profiles.
Background heat flow was removed from the
borehole profiles so the curves reflect only the
climate signal and variation in thermal
conductivity. Sites in North Dakota and South
Dakota penetrated relatively homogeneous bedrock
(Pierre Shale) and have lesser noise due to
variability in thermal conductivity.
Interestingly, the best agreement between GST and
SAT is found in North Dakota and South Dakota,
i.e., regions with significant seasonal snow
cover.
A critical comparison in this study is that of
the multi-proxy record with the borehole record.
Here we compare the synthetic T-z profiles for
four proxy records with a synthetic borehole
profile based on the GST record of Huang and
Pollack (2000). We also compare the 1895-2000
climate data for the North American mid-continent
with borehole records from the north central US.
A critical point is that both the borehole data
and the climate data show that warming during the
past century has increased in a pattern predicted
by GCM models based on increases in greenhouse
gases. That is warming increases with latitude
in the mid continent region.

• Conclusions
• Observed changes in multiple T-z profiles at
  sites which experience seasonal snow cover in the
  Northern US and Canada match closely with SAT
  data.
• Synthetic T-z profiles derived from SAT data
  from the US and Canada match closely with
  observed T-z profiles.
• Contamination of the climate signal in borehole
  data due to terrain effects and changes in
  microclimate and land use at borehole sites can
  be recognized and avoided.
• Both the borehole data and the climate data show
  that warming during the past century has
  increased in a pattern predicted by GCM models
  based on increases in greenhouse gases.

Multiple T-z profiles recorded between 1990 and
2002 compared to synthetic T-z profiles using
ensembles of climate data from the automated
arrays depicted in Figure 1 show that the
borehole profiles changed in close agreement with
the SAT. The two examples shown above are two of
a total of 51 that have been examined to date.
Of this number only three show disagreement with
predicted changes. In each case, the reason for
disagreement was found to be change in the
microclimate of either the borehole or the
climate station. In addition to the sites in
South Dakota, multiple T-z logs from a site in
New Hampshire show excellent correlation in
overall between observed T-z profiles and
synthetic profiles based on climate data. New
Hampshire site US_NH-72 shows a signal opposite
to predictions and to observations at USNHa. The
disagreement is caused by terrain effects at that
site. See poster by Heinle and Gosnold.
References Chapman, D. S., M. G. Bartlett, and
R. N. Harris (2004), Comment on Ground vs.
surface air temperature trends Implications for
borehole surface temperature reconstructions by
M. E. Mann and G. Schmidt, Geophys. Res. Lett.,
31,L07205, doi10.1029/2003GL019054. Huang, S.,
H. N. Pollack, and P. Y. Shen (2000), Temperature
trends over the past five centuries reconstructed
from borehole temperatures, Nature, 403,
756758. Mann, M. E., and G. Schmidt (2003),
Ground vs. surface air temperature trends
Implications for borehole surface temperature
reconstructions, Geophys. Res. Lett., 30(12),
1607, doi10.1029/2003GL017170. Mann, M. E., R.
S. Bradley, and M. K. Hughes (1998), Global-scale
temperature patterns and climate forcing over the
past six centuries, Nature, 392, 779 787.