Page 183 - The Petroleum System From Source to Trap

9. Overburden Rock, Temperature, and Heat Flow 175

0 geophysical well logs. In many instances, strong correla­
tions have been found between thermal conductivity and
TER TIARY one or more log parameters such as resistivity, seismic
. .
. .
velocity, and density (Houbolt and Wells, 1980; Reiter et
. .
I a!., 1980; Vacquier et a!., 1988; Blackwell and Steele,
I . . . .
I � · GA NNET
I >. : . 1989). In other cases, mineralogy has been estimated
I . .
•I, . from well logs, and the thermal conductivity of the bulk
. ' .
I ., .. PREUSS rock estimated from laboratory-derived values for
I
,- , ., . ;�· .:
2 different mineralogies (Brigaud and Vasseur, 1989;
I , • ...,. • • TWIN CREEK
I ,- .. , . • Brigaud et al., 1990; Demongodin et al., 1991). The limita­
E '-- · �- - - - � . . . tion of all of these methods is the lack of an accurate
.:It NUGGE T r -:-..;-• -- ,.,J I . . . . .
ANKA REH .. . .. mineralogy log. Matrix thermal conductivity is deter­
L - - - '• 1 •• • • . . .
J: 3 . . mined by mineralogic composition; correlations and
I
I- THAYNES I • . .
a.. L • . . . inferences found to be valid in specific instances cannot
·
LI.J WOOD. - DJNW ,.. - 7 · ,, • be generalized. Thus, at the present time, there is no
"¥
0 I -..· . .
PHOSPHORIA L. - - ' - • .. . . simple algorithm for estimating thermal conductivity
• _ .:.. 1 · ... .. ,
4 from well logs that is demonstrably accurate. However,
I
I well logs may prove useful in interpolating between
I .
WEBER I I . . measurement sites when log parameters can be cali­
I .
I brated by laboratory measurements.
I •
5 I . .
r- <t :
1 . .
MADISON . I I .
.. ,·. I CONTROLS ON TEMPERATURE IN
I
SEDIMENTARY BASINS
I 2 3 4 5 6 7
T H E R M A L C O N D U C T I V I T Y ( W / m K ) Heat Flow and Thermal Conductivity
Because the primary mode of heat transport in the
Figure 9.7. Thermal conductivity data, Anschutz Ranch well crust is conduction, both heat flow (determined from
34-02, Utah-Wyoming thrust belt. Dots are matrix conduc­ equation 2) and thermal conductivity (measured
tivities measured at 20°C in the laboratory; dashed lines
are estimated in situ thermal conductivity. (After Deming directly) are of first-order and equal importance in deter­
and Chapman, 9 88.) mining temperature in sedimentary basin fill.
1
Heat flow is inversely correlated to tectonic age
(Vitorello and Pollack, 1 9 80; Morgan, 1 9 84) and is
data measured on samples from the Anschutz Ranch 34- depressed by sedimentation (see later discussion). Heat
02 well in the Utah-Wyoming thrust belt (Deming and flow in young ( <25 Ma) rift basins can be as high as
-
Chapman, 1988). The discrete points represent matrix 90 1 2 0 mW m 2 or higher, but it decreases with
1
conductivities measured in the laboratory on drill chips; increasing age. Foreland basins are typically associated
the dashed lines show estimated in situ thermal conduc­ with post-Precambrian orogenic belts and therefore tend
tivities. The in situ estimates are lower than the labora­ to have heat flows in the range of 50-70 m WI m2.
tory measurements due to the effects of porosity and Intracratonic basins generally have modest heat flows in
temperature, and range from about 2 to 4 WI m K. The the range of 30-50 mW 1m2, reflecting their location on
wide scatter of measurements for any formation is old, stable cratons. Other types of basins, such as pull­
partially due to errors in measurement, but most of the apart or backarc basins (e.g., the Salton trough; Lachen­
scatter can be attributed to changes in lithology and bruch et al., 1985) may have young tectonic ages and can
mineralogy. have high heat flows. Heat flow in basins subject to sedi­
The number of measurements needed to determine mentation rates higher than 100 mlm.y. (e.g., passive
the average thermal conductivity of a geologic unit to an margins) can be extremely depressed.
acceptable level of precision depends on its lithologic Thermal conductivity varies by as much as a factor of
heterogeneity. For some marine Paleozoic units that are three or four among common lithologies. However, any
lithologically uniform over hundreds of kilometers, it relatively thick stratigraphic section tends to be
may be possible to make only 10-20 measurements for composed of a variety of different lithologies. Some of
an entire basin. However, large spatial variations in these may have thermal conductivities that are relatively
thermal conductivity are more typical because most sedi­ high and some relatively low. A useful rule of thumb is
mentary rocks tend to have facies changes that occur that the average thermal conductivity of a section
both vertically and laterally. It is therefore difficult in containing diverse lithologies is about 2.5 W lm K. It is
most cases to collect enough data to estimate how the unusual to find a lithologically diverse section of sedi­
thermal conductivity of a geologic unit changes mentary rocks with an average thermal conductivity
l
throughout a basin. lower than 1.5 W m l K or higher than 3.0 W m K.
To overcome this difficulty, concerted efforts have However, a variation of 100% is of first-order impor­
been made to estimate thermal conductivity from tance.