Transcript Has Northern Hemisphere Heat Flow Been Underestimated?

The Post-Glacial Warming
Signal in Heat Flow
IUGG Perugia, It,
6/7/2007
Will Gosnold and Jacek Majorowicz
University of North Dakota
Grand Forks, ND USA
Outline
• Introduction to the problem
• The signal is subtle and
difficult to detect
• Test of warming signals
• Impact on heat flow and
crustal temperature analyses
• Summary
Introduction
• Much of our understanding of the thermal
state of Earth’s crust derives from a mere
scratch of the surface.
• Most direct measurements of the
temperature distribution in the continents
come from boreholes less than 500 m deep.
• Fewer than 4 percent of heat flow
measurements in North America and only
slightly more than 14 percent of all
continental heat flow measurements were
made at depths greater than 2 km.
20,201 heat flow sites recognized by the International Heat Flow Commission
• How then, one might ask, is it that we can
use this “scratch of the surface” to calculate
temperature profiles for the crust and to
confidently apply these calculations in a
variety of geological, geochemical and
geophysical investigations?
• We assume the thermal state of the crust is
conductive, in thermal equilibrium, and that
we know with reasonable certainty the
thermal properties of the rocks that comprise
the crust.
• In general, these assumptions are valid for
the deeper part of the crust, and heat flow
researchers take precautions to avoid nonconductive regimes and transient
disturbances in the near-surface
measurements.
• What if the heat flow data
acquired from shallow boreholes
contain a systematic error?
• Would the error lead to inaccurate
projections of temperatures
deeper in the crust?
• Recent heat flow studies in Europe and
Russia lead us to hypothesize that postglacial warming in northern hemisphere
continents may have been of the order of 10°
to15° C rather than 3° to 5° C as is generally
accepted in terrestrial heat flow research.
• If this hypothesis is correct, some northern
hemisphere heat flow values may require
revision by as much as 30 to 60 percent.
• A significant challenge is that the signal is
difficult to detect and to quantify.
Conductive heat
flow at the surface is
described by
Fourier’s Law of
heat conduction
and common
practice is to use a
portion of the T-z
profile in which the
gradient is linear for
heat flow
calculations.
q  
λ= 1.3 W/m/K
q = 70 mW m-2
The effect of postglacial warming on the thermal
gradient is subtle.
40
35
30
Deg C
25
20
Steady-state T-z
3 Deg T-z
5 Deg T-z
10 Deg T-z
15 Deg T-z
15
10
5
0
0
400
800
Depth (m)
1200
1600
How detectable is the disturbance to the temperature gradient?
16
ΔT-z for a 15º temperature change at 10 ka
14
100 – 200 m
12
y = -0.0082x + 14.957
R2 = 0.9997
400 – 500 m
10
Deg C
200 – 300 m
y = -0.0052x + 13.974
R2 = 0.9986
y = -0.0076x + 14.831
R2 = 0.9997
8
500 – 600 m
300 – 400 m
y = to
-0.0037x
+ 13.245
Suppose we calculate a linear least squares fit
the
2
R = 0.997
y = 100
-0.0065x
+ 14.503
temperature gradient at
meter
intervals.
2
6
R = 0.9995
Will the postglacial warming signal be detected?
4
Answer:
Not at all!
600 – 700 m
y = -0.002x + 12.232
R2 = 0.9838
2
0
0
10 0
200
300
400
Depth (m )
500
600
700
800
25
y = 0.0102x + 13.899
R2 = 0.9998
20
y = 0.0073x + 14.976
R2 = 0.9998
y = 0.0084x + 14.695
R2 = 0.9996
y = 0.0112x + 13.268
R2 = 0.9997
y = 0.0092x + 14.355
R2 = 0.9997
15
Deg C
y = 0.0078x + 14.876
2
R = 0.9997
10
y = 0.022x
R2 = 1
5
0
0
100
200
300
400
Depth (m)
500
600
700
800
• Revision of some data may be needed even if
warming at the end of the last glaciation was
only 3 to 5 °C.
• This range of warming may be appropriate
globally, but we now know that the temperature
trends associated with climate change increase
toward the poles.
• For example global temperatures have warmed
about 1°C during the past century, but high
latitude temperatures (e.g.,Alaska) have warmed
by up to 5 °C.
• Could we determine an appropriate scale for
post glacial warming in high latitudes?
How can the hypothesis be tested?
• Empirical evidence from existing sites.
• Data from deep boreholes in which
continuous temperature logs can be
matched with continuous thermal
conductivity measurements.
Empirical evidence for large
magnitude postglacial warming
• T-z measurements in parts of Europe and
North America show a systematic increase
in heat flow with depth.
Heat flow increases with depth in Europe
160
140
120
Heat flow
100
80
60
40
20
0
0
1000
2000
3000
4000
Depth (m)
5000
6000
7000
Heat flow
increases with
depth in the North
American craton
Empirical evidence for large
magnitude postglacial warming
• Heat flow in southern hemisphere shields
averages approximately 61.4 mWm-2, but
heat flow in northern hemisphere shields
averages 37 mWm-2.
Differences by Craton
• Brazil 64.8 ± ? mW m-2 (86)
• Africa 52.3 ± ? mW m-2 (145)
• Australia 68.1 ± ? mW m-2 (157)
• N. America 33.1 ± ? mW m-2 (315)
• Fennoscandia and East European
Craton 35 - 40 mW m-2 (1,352)
Optimum sites would be in periglacial regions
in rocks having minimal variability in λ.
The Tertiary and Upper
Cretaceous sections of the
Williston basin consist of
continuously deposited
marine shales. These
units have a thermal
conductivity of 1.1 to 1.2
W/m/K.
Theoretical and empirical studies
suggest that as porosity is reduced
by compaction, thermal
conductivity increases.
If heat flow is constant, the
temperature gradient should
decrease with depth and the T-z
curve should be convex up.
•LSQ analyses of 200 m
segments of a temperature
log from the Williston basin
all appear linear.
•The geothermal gradient
increases systematically with
depth.
•The surface intercept on the
temperature scale decreases
systematically with depth.
•Does the change in surface
temperature show the
amount of warming that has
occurred at the surface?
•If so, the minimum warming
has been at least 12 K.
•Three T-z profiles from the Williston
basin exhibit an increase in heat flow
with depth.
•The synthetic T-z shows an expected
profile for constant heat flow with the
effects of compaction on thermal
conductivity.
•The 15 degree signal is a modeled
curve for warming since 10 ka.
•The Glacx curve results from
superposition of 3 glacial/interglacial
cycles (90 ka/10) ka warming signal on
the steady-state synthetic T-z. It
appears to match closely the observed
T-z profiles.
Detection of the signal from variable conductivity strata is more difficult.
Curve fits to six
sites in periglacial
regions in North
America
Knowledge of Earth’s internal temperature
is applied in a variety of geological,
geochemical and geophysical
investigations including but not limited to:
•
•
•
•
•
•
•
Geodynamics
Tectonics
Rheology
Seismology
Curie Temperature
Groundwater Flow
Radioactive Waste
•
•
•
•
•
Petrology
Mineralogy
Volcanology
Geothermal Energy
Maturation of Oil
and Gas Reservoirs
• Paleoclimatology
Percentage of q vs. depth as a result of warming
100%
90%
percent Q
80%
15 deg
10 deg
5 deg
3 deg
70%
60%
50%
40%
30%
0
400
800
1200
1600
2000
Depth (m)
2400
2800
3200
Summary
•
Evidence for large magnitude post-glacial warming in northern Europe
and Asia is growing.
•
Some North American sites show indications of large magnitude postglacial warming.
•
Many northern hemisphere heat flow values may require revision by
as much as 60 percent because they were determined from boreholes
too shallow for recognition of the gradient disturbance caused by a
large post-glacial warming signal.
•
The hypothesis requires further testing with data from deep boreholes
in periglacial regions.
•
If the hypothesis holds, can a general correction to heat flow
measurements be determined?
Low heat flow in North American craton
Mareschal and Jaupart (2004)
Low heat flow in
North America
(Blackwell and
Richards, 2004).
Note that the low
heat flow area in the
Canadian Shield is
coincident with the
center of the
Wisconsinan ice
sheet.