heat flow in Earth

The Earth's heat flow is defined as the amount of heat escaping per unit time from the interior across each unit area of the Earth's solid surface. The movement of heat within the Earth and its eventual loss through the surface are central elements in the modern theory of plate tectonics. The plate movements over the surface of the Earth are seen as one manifestation of a heat engine at work in the interior, and the heat flow at the surface as the exhaust from the engine.  See also: Earth interior; Plate tectonics

That the interior of the Earth is a great reservoir of heat is a fact known from antiquity. The eruption of molten rock from volcanoes and the egress of thermal waters and steam from subterranean regions in the form of hot springs and geysers are manifestations of the Earth's internal heat. These phenomena have always been apparent to the human inhabitants on every continent. With the development of underground mining, it became uncomfortably apparent to workers in the deeper levels that the temperature within the Earth generally increased with depth.

Because heat flows from the warmer to the cooler parts of a body, the observation that the Earth's temperature increases with depth clearly implies that heat is escaping from the interior by conduction through the rocky crust. William Thomson (Lord Kelvin) used the heat flow in estimating the age of the Earth. He argued that as the Earth cooled, following its formation and solidification from molten rock, the rate at which the temperature increased with depth (the geothermal gradient) lessened over time, and thus by measuring the geothermal gradient the length of time that the Earth had been cooling could be estimated. Such an estimate is based on the assumption that all the heat being lost is drawn from the endowment of heat that the Earth possessed at the time of its initially molten condition. This assumption was later shown to be untenable because a significant fraction of the Earth's heat flow is derived not from the initial heat but from the decay of radioactive elements present in trace amounts within the Earth. Thus the Earth's heat should be thought of not as a finite quantity acquired at the time of formation and gradually dissipated over time, but as a quantity that in large part has been continually produced, albeit in diminishing amounts throughout the history of the Earth.  See also: Earth, age of; Earth, convection in

 

Measurement

 

The quantitative measure of heat flow, q, is obtained as the product of the geothermal gradient, grad T, multiplied by the thermal conductivity, K, of the material through which the heat is passing: q = K grad T. The determination of the gradient and conductivity differs in continental and oceanic settings.  See also: Conduction (heat); Heat radiation; Heat transfer

 

Continents

 

On continents the geothermal gradient, that is, the rate of increase of temperature downward, is ordinarily obtained by measuring the temperature at several points (or continuously) down a borehole. Most measurements are obtained in mineral exploration boreholes in the depth range 100–500 m (330–1650 ft); at shallower depths there are disturbances to the borehole temperature profile, principally due to circulating ground water, but also arising from the daily and annual temperature variations, irregular distributions of vegetation, topographic relief, erosion and deposition, climatic variations, and other causes. Holes of greater depth, while generally freer of disturbances in the deeper parts, are only infrequently available. Occasionally temperature measurements can be obtained in horizontal borings at different levels of an underground mine, and a geothermal gradient thereby determined. The thermal conductivity is measured on samples of rock obtained from the borehole as it was drilled. The conductivity is usually determined by a laboratory instrument in which a known quantity of heat is conducted through the sample and the temperature gradient established across the sample is measured. Other approaches to the determination of conductivity involve heating the surface of the rock sample with a heating wire or laser for a short time, and monitoring the rise and subsequent fall of the temperature of the rock.  See also: Drilling, geotechnical

 

Oceans

 

In oceanic measurements, a hollow probe several meters long equipped with temperature sensors at intervals along the length is allowed to plunge into the soft sediment of the ocean bottom. The temperature of the sediment is thus obtained at several depths, and the geothermal gradient established. As the probe plunges into the soft ocean sediment, a core is collected in the interior cavity and is brought to the surface when the probe is retrieved. The conductivity of the soft sediment is obtained by inserting a heating needle into it and monitoring the rise of temperature with time as heat is supplied to the core. Some oceanic heat flow probes contain the heating needle internally and allow the direct measurement of conductivity in place while the probe is immersed in the ocean bottom.

The deep ocean floor is a thermally stable environment, generally free from diurnal and seasonal temperature variations that disturb equally shallow temperature measurements on continents. However, it is not free of temperature disturbances arising from the circulation of water through the fractures in the volcanic rock of the oceanic crust. This circulation, actually observed as ocean-bottom hot springs, is most active along the mid-ocean ridges, comprising the youngest areas of the oceanic crust. The hydrothermal circulation on the ridges gives rise in places to the spectacular “black smoker” venting of water at temperatures up to around 350°C (660°F), with important implications for life in the deep oceans. At the ridge crest, unusual biological communities exist at the hot water vents. Because of their relative youthfulness, the ridges have received less of a sediment cover than the older areas of ocean floor, where the ocean water is denied entry into the crustal fractures by a thick sedimentary blanket.  See also: Mid-Oceanic Ridge

 

 

Global data set

 

In the first review of heat flow measurements in 1954, the total count of measurements both on continents and in oceans numbered only in a few tens. Subsequently, the data set has grown substantially, principally because of the recognition of the role of the Earth's heat in the internal dynamics and surficial tectonics of the planet. Measurements have been made at more than 20,000 sites around the globe in both continental and oceanic settings (Fig. 1). On a 5 × 5° grid, the observations cover 62% of the Earth's surface. The most intensive continental coverage has been in the United States, Europe, and Japan, with moderate coverage in Canada, Australia, India, and southern Africa. Gaps include Antarctica and much of Asia, Africa, and South America. The oceanic measurements have been concentrated in the latitude range 60°N to 30°S; the Arctic and southern oceans are only weakly represented with measurements. Despite the uneven distribution and incomplete coverage, a number of useful results bearing on the thermal and tectonic history of the continents and oceans have been obtained from studies of the heat flow data.

 

 

Fig. 1  Geographic distribution of heat flow measurement sites. (After H. N. Pollack, S. J. Hurter, and J. R. Johnson, Heat flow from the Earth's interior: Analysis of the global data set, Rev. Geophys., 31(3):267–280, 1993)

 

 

 

 

 

 

 

Heat flow and age relationships

 

The most significant empirical relationship to emerge from the data studies is that heat flow generally decreases with increasing crustal age at the site of the measurement (Fig. 2). In the oceans the significant age is the time of magmatic emplacement of the basaltic crust at a mid-ocean ridge. On the continents it is usually the age of the last major tectonic, magmatic, or thermal metamorphic event to have affected the measurement site. The pattern of the decrease in heat flow with age differs from ocean to continent. While both settings show a heat flow in the older segments of about 40–45 milliwatts per square meter, the oceanic heat flow shows a greater range over a shorter time interval than does the continental heat flow.

 

 

Fig. 2  Heat flow–versus–age relationships for continents and oceans. (After D. Sprague and H. N. Pollack, Heat flow in the Mesozoic and Cenozoic, Nature, 285:393–395, 1980)

 

 

 

 

 

 

The oceanic heat flow–versus–age curve is somewhat obscured in the younger oceanic crust because of seawater circulating through the basalt; however, measurements at sites where sediments have sealed the crust and prevented seawater penetration verify that the inverse square-root dependence on crustal age is valid in the younger crust. The decay of continental heat flow with age extends over a much longer period of time and is more complex than that of the oceans. The continental heat flow principally comprises (1) a radiogenic component derived from the enrichment of the continental crust in heat-producing radioactive isotopes, (2) a component derived from the cooling of tectonothermally mobilized lithosphere, and (3) a background heat flow of probable deep origin.  See also: Radioisotope (geochemistry)

 

Radiogenic heat in Earth's crust

 

The heat production from radioactive decay contributes significantly to the heat flow from the Earth's interior. In fact, the entire surface heat flow on continents could arise within the crust if the surface concentrations of the important heat-producing isotopes such as are found in granites and gneisses persisted throughout the full thickness of the crust. However, probable lower crustal rocks such as granulites and migmatites have lesser concentrations of these isotopes, and so radiogenic heat production apparently decreases with depth in some manner. The existence of radiogenic heat and its generation throughout the history of the Earth proved to be the weak link in the Kelvin theory of the age of the Earth.

 

Radioisotopes

 

The principal heat-producing isotopes are thorium-232 (232Th), uranium-238 (238U), potassium-40 (40K), and uranium-235 (235U) with respective half-lives of 14.0, 4.47, 1.25, and 0.70 × 109 years. Other radioactive isotopes do not presently contribute significant heat because their decay chains are not sufficiently energetic, or their abundances are insignificant, or their half-lives are too short. The concentrations of uranium and thorium in rocks are generally in trace amounts measured in parts per million, while potassium is much more abundant, in the range of a few percent, of which a small but well-known fraction is 40K (see table). A range of values for the upper mantle is characteristic of various possible mantle rocks as inferred from xenoliths brought to the surface by igneous and tectonic processes. If the entire mantle had the lesser concentrations, then crustal and mantle radiogenic heat production would be less than half the present-day heat loss; whereas if the greater concentrations were representative of the mantle as a whole, radiogenic heat production would be about equal to the present-day heat loss.

The abundances of these isotopes are commonly obtained from a multichannel gamma-ray spectrometer, which separates and counts the gamma-ray emissions on the basis of energy. Certain steps in the decay scheme of each isotope involve gamma radiation with characteristic energies, and thus the number of counts at certain levels of the energy spectrum is indicative of the isotopic abundance in the sample. Atomic absorption spectrometry can also be used to determine the amounts of the more abundant potassium, and mass spectrometry to determine the much smaller abundances of uranium and thorium.  See also: Atomic spectrometry; Mass spectrometry

The explanation for the relatively higher concentrations of uranium, thorium, and potassium in the continental crust resides in the relatively large ionic radii of these isotopes compared to the size of silicon, aluminum, magnesium, calcium, and iron, which combined with oxygen make up the bulk of the Earth's mantle as oxide minerals. The larger ions fit less easily into the compact crystal lattices of the mantle oxides, and in magmatic and metamorphic events they are more easily mobilized and tend to follow the magmatic products and metasomatic volatiles upward into the crust, where they are more easily incorporated in the more open crystal structures characteristic of the lower-pressure environment of the crust.

 

Heat flow province

 

The crustal radiogenic component of continental heat flow has been studied extensively. The principal observation is that, within a heat flow province, the regional variation in heat flow is well correlated with the heat production of the surface rocks. The relationship is expressed as q0 = qr + bA0, where q0 is the surface heat flow, A0 is the heat production of the surface rocks, qr is the reduced heat flow (the heat flow intercept for zero heat production), and b (the slope of the line) is a quantity with dimension of length that historically has been interpreted as a vertical scale length which characterizes the distribution of the heat sources with depth. The reduced heat flow is then interpreted as that flux that arises below the depth range defined by the empirical scale length. A heat flow province comprises a geographic area in which the heat flow and heat production are linearly related (Fig. 3); each heat flow province displays a characteristic reduced heat flow and scale length parameter. However, the traditional one-dimensional (vertical) interpretations have been questioned by investigators who argue that two- and three-dimensional heat transfer can also give rise to an apparent linear heat flow–heat production relationship, depending on both the horizontal and vertical scale lengths of crustal heterogeneities. In the multidimensional interpretations the linear relationship reveals more about the characteristic dimensions of the crustal heterogeneity than it does about the distribution of heat sources with depth.

 

 

Fig. 3  Geographic distribution of 17 heat flow provinces where the linear relationship between measured heat flow and radiogenic heat production from near-surface rocks has been empirically determined. (After I. Vitorello and H. N. Pollack, On the variation of continental heat flow with age and the thermal evolution of continents, J. Geophys. Res., 85(B2):983–995, 1980)

 

 

 

 

 

 

 

 

Global heat loss

 

The heat flow–versus–age relationships (Fig. 2) that provide the critical constraints for thermal history models of the crust and upper mantle of both oceans and continents can be put to another useful purpose: estimating the heat flow in areas where no measurements have been made but for which the age of the crust is known. In a general way, the distribution of ages in the continental crust has been known for some time, and with the advent of the geomagnetic reversal time scale, large areas of the oceanic crust have been dated on the basis of their magnetic signatures. Thus it has become possible to estimate the heat flow in unsurveyed areas and map the heat flow over the entire surface of the Earth (Fig. 4).

 

 

Fig. 4  Present-day global heat flow (degree 12 spherical harmonic representation). Contours are in milliwatts per square meter. (After H. N. Pollack, S. J. Hurter, and J. R. Johnson, Heat flow from the Earth's interior: Analysis of the global data set, Rev. Geophys., 31(3):267–280, 1993)

 

 

 

 

 

 

The oceanic ridge system, where new oceanic lithosphere is produced, is important in delivering internal heat to the surface; fully half the Earth's current heat loss comes from oceanic lithosphere produced in the last 66 million years, representing only 31% of the Earth's surface. Other regions of above-average heat flow include the cordillera of western North America, alpine Europe, and the marginal basins of the western Pacific. Areas of low heat flow include all major continental shields and platforms and older ocean basins. Regions where lithospheric plates slide past each other, such as the San Andreas Fault of western North America, show no special heat flow signature, suggesting that friction between the plates is insufficient to generate anomalous heat. The mean heat flow through continents (including marine continental shelves) and oceans is 65 and 101 mW · m−2, respectively. The speculation that the heat flow through oceanic and continental crust is approximately equal, advanced before the relationships between heat flow and tectonic age were well established, is no longer tenable.  See also: Lithosphere

The average heat flow over the entire Earth is 87 mW · m−2 and is comparatively a trickle; it is sufficient to bring a thimbleful of water to a boil in about 2 years, or if collected over the area of a football field, would be adequate to light four 100-watt incandescent bulbs. The radiant energy from the Sun intercepted by the Earth is some 4000 times greater than the geothermal flux. The absorption and reradiation of the incident solar energy are the principal processes that determine the surface temperature of the Earth. The geothermal energy is of little significance to the surface temperature but is of paramount importance in considering the Earth's internal thermal condition.  See also: Geologic thermometry; Geothermal power

 

Heat flow in the past

 

The present-day heat flow from the Earth has commonly been considered a useful if not precise upper bound on the quantity of heat being produced radiogenically in the interior. However, if the heat flow has a temporal variability on a scale that is short compared to the half-lives of the principal heat-producing isotopes, the present-day heat flow must be seen as a much coarser constraint on the Earth's internal heat generation.

Just as the present-day heat loss can be determined from the heat flow–versus–age relationships and the present-day distribution of crustal ages, so can the heat loss at a past time be calculated, provided the age distribution of the crust at that time can be estimated. For the past 180 million years, the age distribution of the oceanic crust at a given time can be determined by subtracting from the present-day distribution all oceanic crust produced from that time to the present and resurrecting and reinstating crust which has been subducted in the interval from that time to the present. The age distribution of the restored lithosphere can be estimated from the present-day age distribution, and the probability that crust produced at an earlier time will have survived to the present day. Continents can be treated similarly, but with different survival probabilities which reflect the longevity of the continents. This method has yielded estimates of the mean heat flow in oceans and continents and the global flux for the past 180 million years. The global mean heat flux, presently 87 mW · m−2, has fluctuated in a range of −5% to +15% around the present-day value. The mean heat flow exceeded the present-day flux in the interval 60–100 million years before present, when the rate of production of oceanic crust was high. The mean age of the ocean floor, presently 62 million years, has ranged between 45 and 65 million years, accounting for most of the variation in the global heat loss.

For times earlier than 180 million years ago, an analysis based on the present-day age distribution and survival probabilities becomes unfeasible because in the oceans, where 70% of the global heat loss takes place, there is no oceanic crust surviving with which to estimate past rates of production of the sea floor. Moreover, in studying the Earth's heat loss over intervals of several hundred million to billions of years before the present, it is necessary to consider the half-lives and relative abundances of the principal heat-producing isotopes and the loss of primordial heat acquired during planetary accretion and core segregation. The isotopic endowment of the bulk Earth is unknown but is sometimes assumed to be similar to the chondritic meteorites (see table), which are thought to be representative of the primitive material from which the Earth differentiated. However, the rocks of the Earth's crust have a lower potassium-uranium ratio than do the chondrites. Because of the relatively short half-life of 40K, the heat production 3 billion years ago from a chrondritic composition would have been more than four times the present-day heat production, while a terrestrial composition, based on observed potassium-uranium ratios of crustal rocks, would yield heat at only somewhat more than twice the present rate.  See also: Meteorite

Generalized model calculations of the thermal history of the Earth using the potassium-uranium ratio of terrestrial crustal rocks show that the present-day heat production replaces much but not all of the present-day heat loss. The inequality between heat production and heat loss arises because a finite transit time is required between the time when heat is produced within the body of the Earth and when it escapes across the outer surface, during which interval the heat sources have decayed. The present-day heat loss thus can be seen as older radiogenic heat, delayed in its escape from the Earth's interior by the heat transfer process. Because the replacement heat production is less than the heat loss, the Earth is necessarily cooling. In spite of the various uncertainties about the blend of the nuclear fuel, the thermal history calculations indicate that the heat loss from the Earth 3 billion years ago was at least twice as great as the present-day heat loss. However, the augmented heat flow may not have been distributed uniformly over the Earth. Rather, the increased heat loss may have been best accommodated by a more rapid rate of production of oceanic crust, with only a modest increase of heat flow through the continents. This latter speculation derives from crustal temperatures inferred from metamorphic rocks in ancient continental terrains, which suggest that the heat flow entering the base of the continental lithosphere was approximately the same as today.

Henry N. Pollack

 

Bibliography

 

 

  • H. N. Pollack, S. J. Hurter, and J. R. Johnson, Heat flow from the Earth's interior: Analysis of the global data set, Rev. Geophys., 31(3):267–280, 1993
  • Alifazeli=egeology.blogfa.com