Transcript AGU - Global Heat Flow Database

A Multidisciplinary Approach to Assessing the Causal Components of Climate Change
William D. Gosnold,
ABSTRACT. Separation of climate forcing by
anthropogenic greenhouse gases from natural radiative
climate forcing is difficult because the composite
temperature signal in the meteorological and multiproxy temperature records cannot be resolved directly
into radiative forcing components. To address this
problem, we have initiated a large-scale,
multidisciplinary project to test coherence between
ground surface temperatures (GST) reconstructed from
borehole T-z profiles, surface air temperatures (SAT),
soil temperatures, and solar radiation. Our hypothesis
is that radiative heating and heat exchange between
the ground and the air directly control the ground
surface temperature. Consequently, borehole T-z
measurements at multi-year intervals spanning time
periods when solar radiation, soil temperatures, and
SAT have been recorded should enable comparison of
the thermal energy stored in the ground to these
quantities. If coherence between energy storage, solar
radiation, GST, SAT and multi-proxy temperature data
can be discerned for a one or two decade scale,
synthesis of GST and multi-proxy data over the past
several centuries may enable us to separately
determine the anthropogenic and natural forcing of
climate change. The data we are acquiring include: (1)
New T-z measurements in boreholes previously used in
paleoclimate and heat flow research in Canada and the
United States from the 1970's to the present. (2)
Meteorological data from the US Historical Climatology
Network and the Automated Weather Data Network of
the High Plains Regional Climate Center, and
Environment Canada. (3) Direct and remotely sensed
data on land use, environment, and soil properties at
selected borehole and meteorological sites for the
periods between borehole observations. The project
addresses three related questions: What is the
coherence between the GST, SAT, soil temperatures
and solar radiation? Have microclimate changes at
borehole sites and climate stations affected
temperature trends? If good coherence is obtained,
can the coherence between thermal energy stored in
the ground and radiative forcing during the time
between T-z measurements be extended several
centuries into the past?
1
Jr. ,
1
Todhunter ,
1
Dong ,
1
Rundquist ,
1
Majorowicz ,
Paul
Xiquan
Bradley
Jacek
and David D.
1 University of North Dakota, Northern Plains Climate Research Center
2 Southern Methodist University
Fig. 7. Theory
The geothermal gradient (Γ) is determined by the
Fig. 2. Climate forcing components, Intergovernmental Panel on Climate
ratio of heat flow (Q) to thermal conductivity (κ) and
Change, Climate Change 2001: Working Group I: The Scientific Basis. may be expressed by rearranging Fourier’s law of
heat conduction as: Γ = Q/κ. On a planetary surface,
the controlling quantities in Fourier’s law are thermal
1.5
conductivity and heat flow from the interior. Thermal
1
conductivity is a physical property of rocks and
0.5
generally falls within the range of 1.0 W m-1 K-1 to 3.5
W m-1 K-1 depending on rock composition and fabric.
0
Heat flow from the interior is partly heat acquired
-0.5
during formation of the planet and partly heat
-1
generated by long-lived radioactive isotopes of
-1.5
potassium, thorium and uranium that are present in
1860 1880 1900 1920 1940 1960 1980 2000 2020
trace amounts. Due to the nature of these sources,
surface heat flow is constant on time scales of
Fig. 3. Global Temperature anomaly 1880 -2002. The
hundreds of millions of years and the background
anomaly was calculated by removing the mean value of all the
geothermal gradient is linear. Therefore, in the
data from each monthly data point.
absence of heat transfer by ground water flow,
curvature in the near surface T-z profile indicates a
recent change in the bounding surface temperature.
Cooling trends cause the surface intercept to swing
towards cooler temperatures and warming trends
cause the surface intercept to swing towards warmer
temperatures. Due to the exponential diffusion of the
signal with depth and the low thermal diffusivity
values for rocks, surface temperature changes
occurring 100 years ago affect the T-z profile to about
100 m depths and changes occurring 500 years ago
affect the T-z profile to about 500 m depths. Fig. 7,
shows the effect on Γ of a 2 degree change
persisting for different time periods (red = one year;
green = ten years ; blue =100 years.)
Fig. 4. The multi-proxy temperature anomaly is shown in
dark blue and the 1σ range is shown in light blue. The
observed temperature anomaly is shown in red. Compare
scales with Fig. 3. (Modified after Mann et al., 2000.)
Fig. 9. The ground-surface temperature history (GSTH)
can be calculated from the T-Z profile using a non-linear
Bayesian formulation based on the method of least squares
called Functional Space Inversion (FSI) (Shen, Pollack and
Huang, 1996). In a regional study designed to compare the
GSTH with predictions of GCM models of global warming,
Gosnold, et a., (1997) used the FSI inversion scheme on a
group of specially drilled heat flow boreholes in a 1000 km
wide transect in the mid-continent of North America. Their
results, shown Table 1 and graphically in Fig. 9 above,
indicate a century-long warming trend that increases
systematically with latitude as predicted by GCM’s.
HCN
LATITUDE Deg. C (no.)
41 - 42.9
0.59
(18)
43 - 44.9
0.72
(13)
45 - 46.9
1.14
(13)
47 - 49.6
1.34
(7)
σ
0.55
0.62
0.54
0.40
GSTH
Deg. C (no.) σ
0.49 (10) 0.31
1.03 (10) 0.66
2.00
(1)
---2.44
(8) 0.52
Table 1. Observed warming in the past century from the
Historical Climatology Network (HCN) and from borehole
data (GSTH).
The goals of this research are to determine the heat flux
that caused the warming increases observed in
boreholes, e.g. at Minot, Glenburn, and Landa, ND
shown above, extract the solar component and thus
quantify greenhouse gas forcing of climate change. An
example of a heat flux calculation is shown below. A 20
m W m-2 heat flux into the ground accurately matched
observed T-Z profiles from 1995 and 2002.
1369
1368
1367
W/m^2
2
Blackwell
1366
1365
1364
1363
1362
1978
1983
1988
1993
1998
9
Fig. 5. Long wave solar irradiance observed by
satellite. Note the 11-year cycle in the data.
Minot heat flux model
8.5
15 meters
20 meters
25 meters
Deg C
1
Fig. 8. T-z profiles from sites in North Dakota,
Manitoba, and Saskatchewan showing warming in
the upper 100 meters. The profiles are spaced apart
on the temperature scale so all may be seen clearly.
0.5
0
Deg C
1.5
2002
1995
20 mW/m^2
8
7.5
-0.5
-1
1975
Fig. 1. Climate stations shown in green and boreholes shown in red.
1980
1985
1990
1995
2000
Date
Fig. 6. Subsurface temperature anomaly
resulting from the solar signal in shown in Fig. 5.
7
0
20
40
Depth
60
80
100