Linking Borehole Temperatures to Air Temperatures

1
Linking Borehole Temperatures to Air
Temperatures
Invited presentation at EGU - Vienna, April, 2006

• Will Gosnold, Xiquan Dong, Shannon Heinle,
• Jacek Majorowicz, Julie Popham, Rachel Robinson,
• Brad Rundquist, Paul Todhunter

This research is supported by National Science
Foundation Award ATM - 0318384
2
Overview

• Introduction
• Research Plan
• Background and Theory
• Observations
• Results
• Conclusions

3
The key to understanding climate forcing lies in
linking the paleoclimate record with the modern
meteorological record. However, the various
methods do not appear to agree with the precision
needed to quantify and distinguish among various
forcing signals.
von Storch et al., 2004
Gosnold, Todhunter, and Schmidt, 1997
4

• This project focuses on correlating borehole
  temperature profiles with the meteorological
  record.
• The ultimate objective is to correlate GST and
  SAT with the proxy records, solar irradiance and
  models of climate forcing.

5
Working Hypothesis

• Radiative heating and heat exchange between
  the ground and the air directly control the
  ground surface temperature
• A time-series of borehole T-z measurements
  spanning time periods when solar radiation, soil
  and air temperatures have been recorded will
  enable comparison of the thermal energy stored in
  the ground to these quantities.

6
Research Plan

• Locate and relog boreholes.
• Obtain daily meteorological data from AWDN sites.
• Test short-period correlation of SAT and borehole
  T-z
• Correlate the long-term temperature record from
  deep boreholes with the meteorological record.

7
Borehole sites included in present study red
circles, AWDN sites green boxes
8
A key reason for focusing on the mid-continent of
North America is that climate varies with
latitude as does change predicted by GCMs using
greenhouse gas forcing.
Source NOAA NCDC
9
Inversion of borehole T-z data to recover the GST
indicates warming beginning about a century ago
between 46N and 50N.
10
Binning GST by latitude shows a pattern
consistent with GCM predictions based on
greenhouse gas forcing of climate.
11
One of the difficulties in comparing GSTH and SAT
is that the initial T-z profile is not known and
is assumed to be undisturbed.
When we first tried this analysis, SAT and GST
data did not agree when compared as change per
century. Adjustment using a POM that minimized
the misfit made the two data sets agree. (Harris
and Gosnold, 1999)
12

• This study uses observed T-z profiles as the
  initial condition and the SAT as forcing signals.
  The profiles below are from North Dakota.

13
Multiple T-z logs from a site in Saskatchewan
14
Daily temperature record covering time span of
borehole logs
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
Cochrane, Northern Ontario
20
Borehole sites included in present study red
circles, AWDN sites green boxes
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
Results of short period test

• The short-period comparison of multiple
  observations of borehole data to climate data
  indicates that the boreholes accurately record
  trends in air temperatures.
• Do the T-z profiles modeled using 110 years of
  SAT data for agree with observed T-z profiles?
• To answer this question we used both monthly and
  annual temperature data from the US HCN and
  compared the results to boreholes logged at the
  appropriate times.

27
US Historical Climatology Network Stations with
Daily and Monthly Data
28
Monthly Data from US HCN
29
Monthly data filtered by thermal diffusion
30
(No Transcript)
31
(No Transcript)
32
(No Transcript)
33
(No Transcript)
34
Synthetic T-z profiles based on 110 y of SAT
forcing show nearly the same warming as the
observed T-z profiles.
35
Some observations on the global borehole data set

• Borehole Percentages-Possibly Problematic
  Boreholes
• Total boreholes examined in United States 130
• Boreholes with terrain effects 33/130 25.38
• Breakdown
• City/Urban effects 6/130 4.615
• Proximity to large bodies of water 5/130
  3.546
• Hill Effects-Ravines, Valleys, Buttes 16/130
  12.308
• Mining Operation 3/130 2.308
• Mixed Effects 3/130 2.308

36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
Even with some non-climatic effects in the
borehole data, the overall record appears to give
a good record of air temperature changes during
the critical period of significant anthropogenic
climate forcing.
40
Conclusions

• We tested our basic hypothesis by synthesis of
  daily meteorological data from an array of 89
  automated meteorological stations and repeat
  measurements of T-z profiles in boreholes
  initially logged ten to twenty-six years ago.
• Repeat temperature vs. depth measurements at a
  number of borehole sites over a 26 year period
  indicate that changes in borehole temperatures
  closely agree with changes in surface air
  temperatures and soil temperatures on
  multi-decade timescales.
• The tests show that subsurface temperatures
  contain an accurate record of the surface air
  temperature if precautions are taken to avoid
  terrain and non-climate related transient
  signals.

41
Linking Borehole Temperatures to Air Temperatures
Will Gosnold, Xiquan Dong, Shannon Heinle, Jacek
Majorowicz, Julie Popham, Rachel Robinson, Brad
Rundquist, Paul Todhunter
This research is supported by National Science
Foundation Award ATM - 0318384