Production rates of 3He in olivine derived from 3 lava flows 40Ar/39Ar dated
at 152, 281, and 1350 k years. From Dunai and Wijbrans, (2000).
Cosmogenic Exposure Dating and the Age of the Earth
Cosmogenic nuclides are nuclides formed by the interaction of 'target' atoms with cosmic radiation. Such nuclides are formed in space, in the atmosphere (e.g. 14C and 10Be), and in situ within minerals at or near the earth's surface (e.g. 10Be, 26Al, and 21Ne). The accumulation of cosmogenic nuclides in minerals at or near the earth's surface provides a basis for exposure 'dating' of landforms, the quantification of erosion rates, and other geologic applications (Bierman, 1994; Cerling and Craig, 1994; Gosse and Phillips, 2001). Independent evidence discussed below strongly suggests that production rates of these nuclides have remained constant or nearly so, validating their use in geochronometry. This essay focuses on cosmogenic exposure dating, a method of dating rock surfaces which has been compared to using the redness of someone's skin in order to estimate the duration of exposure to sunlight (an analogy attributed to Edward Evenson; Gosse and Phillips, 2001). Such analyses show that many landform surfaces have been exposed to cosmic radiation for at least 105 -106 years, far too long to be reconciled with the <10k yr geologic "timescale" promoted by Young Earth Creationists.
Cosmogenic Nuclide Production
The earth is constantly being bombarded by so-called galactic cosmic radiation. This radiation consists mostly of high-energy (~1 - 1010 GeV) protons and alpha particles emanating from within our galaxy. This radiation interacts with nuclei in the atmosphere to produce garden variety or 'meteoric' cosmogenic nuclides (e.g. 14C and 10Be). These interactions produce a cascade of secondary particles, primarily neutrons and muons, which interact with target nuclei within minerals such as quartz and olivine at the earth's surface, producing terrestrial cosmogenic nuclides (TCN). The primary nuclear processes by which cosmogenic nuclides are produced are spallation, muon capture, and neutron activation (Bierman, p. 13,887).
The cosmogenic nuclides most widely utilized for geologic applications are the radionuclides 10Be, 26Al, and 36Cl, and the stable nuclides 3He and 21Ne. 10Be is a radionuclide with a half-life of 1.5 Myr, is primarily produced by spallation from O, Mg, Si, and Fe, and is most commonly measured in quartz, olivine and magnetite. 26Al is a radionuclide with a half-life of 0.7 Ma, is primarily produced by spallation from Si, Al, and Fe, and is most commonly measured in quartz and olivine. 36Cl is a radionuclide with a half-life of 0.3Ma, is mostly formed by spallation from Ca and K and by neutron capture from 35Cl, and is commonly measured in whole rock samples. The stable nuclide 3He is produced primarily by spallation from O, Mg, Si, Ca, Fe, Al, and is most often measured in olivine and pyroxene. 21Ne is measured in quartz, olivine and pyroxene, where it is mostly produced by spallation from Mg, Na, Al, Fe, and Si. Each nuclide is useful for dating some target minerals, but not for others, for various reasons. For example, 3He is produced in both quartz and pyroxene, but it is not retained well in quartz.
| Note: The actual cosmic radiation flux, and hence TCN production rate, varies with geomagnetic latitude and altitude. For instance, the cosmic ray flux/TCN production rate at 20° latitude and 1500m above sea level is ~2 times the flux/production rate for the same TCN at high latitude (>58°) and sea-level. This latitude/altitude variation is described by scaling factors which are deduced from modern neutron-flux measurements (Lal, 1991; Dunai, 1999). For consistency, all production rates are reported here as atoms per gram per year at sea-level and high latitude, regardless of the location at which they have been measured. |
Obviously, the accumulation of cosmogenic nuclides can not be used to date anything unless we know something about the rates at which they are produced, both now and in the past. The rates at which cosmogenic nuclides are produced in rocks and minerals at the earth's surface have been estimated by three methods. The first method is by prediction from a numerical simulation based on probabilities of the nuclear interactions involved in the production of TCNs, such as the model of Masarik and Reedy (1995). The second method is by direct experimental measurements and/or extrapolation from direct experimental measurements (e.g. Nishiizumi et al., 1989). The third method is by geologic calibration -- using surfaces of independently well-constrained ages, such as landslide exposures, lava flows, and glacially- striated bedrock (e.g. Dunai and Wijbrans, 2000; Kubik et al., 1998).
Remarkably, production rates derived from the three methods are very similar, ranging from a high of ~119 atoms/gm/yr for 3He in olivine (Licciardi et al., 1999), to a low of about 5.1 ±0.3 atoms/gm/yr for 10Be in quartz (Stone, 1999). This shows that changes in TCN production rates have not been profound over the calibrated time range. For example, the production rate of 10Be in quartz predicted by the model of Masarik and Reedy (1995) is 5.97 atoms/gm/yr. The values derived from geologic calibration have ranged from about 4.74 - 6.4, with the current best-estimate being 5.1 ±0.3 (Stone, 1999). These values agree remarkably well with the modern production-rate value of 5.21±0.278 determined through the use of water targets (Nishiizumi et al., 1996). The production rate of 21Ne in quartz calculated by Masarik and Reedy (1995) is 18.4 atoms/gm/yr. The production rate derived from geologic calibration using a 13ka surface is 19±3.7 (Niedermann, 2000). Again, these values agree with the modern production rates of 16.3 derived from only a few years of data (Graf et al., 1996). The production rate of 3He in olivine is calculated by Masarik and Reedy (1995) as 105, while long-term, geologically-calibrated rates of ~116-120 are derived from radiometrically-dated lava flows ranging from a few thousand (Licciardi et al., 1999) to over a million (Dunai and Wijbrans, 2000) years.
Production rates of 3He in olivine derived from 3 lava
flows 40Ar/39Ar
dated
at 152, 281, and 1350 k years. From Dunai and Wijbrans, (2000).
| Addendum: For more recent times (Holocene
and latest Pleistocene), the concentration of atmospheric
10Be and 14C in tree rings, lake
and oceanic deposits, and ice cores can be used as proxy
measurements for in situ production rates. This is
because atmospheric cosmogenic nuclides (which are also
produced by interaction with cosmic radiation) are washed
out of the atmosphere and incorporated into such
deposits, leaving a semi-permanent record of variations
in atmospheric production (e.g. Beer et al., 1988; Frank
et al., 1997). For instance, tree rings provide a
continuous record of atmospheric 14C
production extending back to 9800 BCE (Becker, 1993).
Production rate variations during this time are quite
small, less than about 100 per mil. The record of
atmospheric 14C production has recently been
extended back all the way to 45,000 yrs BP (Kitagawa and van der Plicht, 1998) using the
varve chronology from Lake Suigetsu, Japan (Kitagawa and van der Plicht, 1998). Again, the
record shows that changes in production rates have
occurred, but they are far too small to
reconcile the cosmogenic nuclide evidence with a Young
Earth time scale.
|
In order to use TCNs to calculate an exposure age for a given sample, it is also necessary to determine the amount of the nuclide in the sample(s) which is due to in situ cosmogenic production. For instance, 10Be and some other TCNs are also produced in the atmosphere, and must be carefully removed by sample preparation techniques. Also, some nuclides produced cosmogenically can also be produced in situ radiogenically, for instance 3He. The non-cosmogenic inventory of such nuclides can be be assessed by different analytic procedures depending on the mineral/nuclide system, for instance by sampling the parent rock at completely shielded depths, where none of the nuclides can be attributed to cosmogenic production.
Comparison of cosmogenic and non-cosmogenic ages
In a number of instances it has been possible to compare cosmogenic exposure ages with independent ages derived from different methods, such as thermoluminescence (TL) and radiometric dating. For instance, Laughlin et al. (1994) derive 3He dates for 3 basalt flows of the Zuni-Bandera volcanic field of New Mexico, and show that those ages are concordant with 14C and U-Th dates for the same flows (but younger than K-Ar dates, likely as a result of excess Argon). Fenton et al. (2001) use 3He to date 7 basalt flows in the Uinkarent volcanic field, Grand Canyon, and show that those ages agree within uncertainties with TL and 40Ar/39Ar ages for the same flows. The oldest of the flows, the Upper Prospect, is estimated to be 402±36ka. The weighted mean of 21 36Cl exposure ages for a volcano on the northern Tibetan Plateau was 66±1ka, while 40Ar/39Ar analyses of the same volcano yielded an age of 62±1.3ka (Gosse and Phillips, p. 1550). Phillips et al. (1996) derive 36Cl ages for 33 seperate basalt flows, glacial erratics, and other geomorphic surfaces. The resulting ages are compared with independent ages (2.1-55ka) for the same features derived from 14C (28 dates), TL (4 dates), and K-Ar (1 date). In most of these cases the agreement is quite good, with descrepancies of 15% or less. Markewich et al. (1998) use 10Be analyses to date several intervals within thick, paleosol-bearing loess deposits within the Lower Mississippi Valley, USA. The resulting ages --between 25 and 200ka-- are shown to be in good agreement with TL and 14C dates of the same intervals. The agreement in cosmogenic exposure ages and those derived from non-cosmogenic methods provides strong support for the validity of the method.
Comparison of 33 36Cl exposure ages with independent
TL
and 14C dates for the same surfaces.
From Phillips et al., 1996.
Meteor Crater: Using cosmogenic nuclides to date a hole in the ground.
Meteor Crater in Arizona is a 1.2km wide crater punched 200m or so into the through the Permian Kaibab and Toroweap Formations. Since the crater penetrates Permian strata, it is Permian or younger. Since the crater contains some Pleistocene lake deposits, it is Pleistocene or older. This is an ideal situation for the application of cosmogenic dating techniques, since the the cratering event would quickly exposure previously shielded rock to the cosmic ray flux, and since the morphology of the crater itself indicates only a small amount of erosion. Nishiizumi et al. (1991) report a minimum age of 49.2±1.7ka, based pm 10Be and 26Al analyses of samples from the crater walls and ejecta blocks at the crater rim. Phillips et al. (1991) report an indistinguishable 36Cl exposure age of 49±0.7ka for dolomite ejecta on the crater rim. Both sets of TCN dates are in turn indistinguishable from quartx thermoluminescence dates of 49±3ka reported by Sutton (1985) for shocked sandstone breccia st the crater floor.
Paleoseismology: Using cosmogenic nuclides to date ancient earthquakes.
A fault scarp is a sheer rock face produced when rock on one side of a fault is moved vertically relative to the other side. Movement along these fault planes usually occurs during earthquakes. By determing the distribution of exposure ages along a fault scarp as a function of depth, the number and magnitude and recurrence interval of the faulting events which produced the scarp can be determined. The exposure ages should increase from bottom to top, and the should cluster into discrete groups or steps representing the episodic nature of faulting events. As an example of this method, Zreda and Noller (1998) calculate 36Cl exposure ages along a limestone (Meagher Formation, Middle Cambrian) fault scarp on the Hebgen Lake fault in Montana, using whole rock samples. The total vertical offset along the fault is about 12-13m. The results of Zreda and Noller (1998) indicates an age progression from ~0ka at the base of the scarp, to about 35ka at the top. The cluster of ages vertically on the scarp confirm field evidence of ~6 paleo-earthquake/slip events, dating them at 0.4ka, 1.7ka, 2.6ka, 7.0ka, 20.3ka, and 23.8 ka, respectively.
Using cosmogenic nuclides to date glacial moraines
Moraines are bodies of glacially-eroded and transported rock left behind when a body of ice advances and then retreats. Assuming that the boulders were shielded prior to the advance (i.e. no inheritance), and that the boulders have been exposed continuously since the retreat, the glacial retreats can be dated by cosmogenic dating of boulders on moraine surfaces. Unlike 14C dates on organic material buried in moraines, which is useful only to ~45ka BP, cosmogenic methods can be used to date moraines over a million years old (Briner et al., 2001; Shanahan and Zreda, 2001). Shanahan and Zreda (2001) for instance use 36Cl exposure dating to date moraines on Mount Kenya and Kilimanjaro in equatorial east Africa. The oldest moraines on both Kenya and Kilimanjaro date to >350ka. On Mount Kilimanjaro, the younger moraines are dated at 20±1ka, 17.3±2.9ka, 15.8±2.5ka, and 13.8±2.3ka. On Mt Kenya, the younger moraines are dated at 255-285ka, 64±40ka, 55±23ka, 28±3ka, 14.6±1.2ka, 14.1±0.6ka, 10.2±0.5ka, and 8.6±0.2ka, respectively. The youngest moraine on Mount Kenya, the Lewis Moraine, was deposited by a readvance of the Lewis Glacier during the 'Little Ice Age' (roughly 1500-1800). Excluding one outlier date of 1.2ka, the moraine dates to only 0.21±0.21ka, consistent with historical observations (Mahaney, 1990). Dating of boulders at the current margin of the Lewis Glacier are exposure dated at 60±80 yrs, with one outlier date of 630 years, indicating insignificant prior exposure.
Using cosmogenic nuclides to date the earth's oldest landscapes
The oldest exposure ages come from extremely arid regions, where a lack of water makes chemical weathering processes extremely slow. The oldest of all known exposure ages come from the arid polar environments of Antarctica. 21Ne dating of a summit in Daniel's Range, Victoria Land, has yielded exposure ages of up to 11.2Ma, making this the oldest and most stable known surface on earth (Van der Wateren, 1999). 21Ne exposure ages from Mt Fleming in the Dry Valleys region are up to 10.08 ±0.24Ma (Schafer et al., 1999). Bruno et al. (1997) obtain minimum 21Ne and 3He exposure ages of 6.5 and 6Ma respectively for the Mt Fleming and Table Mountain tillites. Beacon Valley is unique in that it preserves an ancient body of glacial ice at least 80m deep, buried beneath a thin (>1m) layer of till. The ice is thought to be a remnant of the Taylor glacier, which once filled the valley. Analyses of cosmogenic 3He and 21Ne in dolerite boulders in the surface till yield minimum exposure ages of 2.3Ma, making this the oldest known body of ice on earth (Schafer et al., 2000).
Very old landscapes are also documented in Africa and Australia. The hyper-arid Namib desert in Africa stretches over 2000km along the Atlantic coast, receiving from about 10mm of rain/yr near the coast to 100-200mm at the foot of the Great Escarpment. A quartz vein near Hope Mine has yielded a minimum 21Ne of 5.18Ma, with erosion rates ~0.1m/My (Van de Wateren, 2001). Bierman and Caffee (2002) use 10Be and 26Al to determine exposure ages and erosion rates for many granitic landforms along a 20 degree N-S transect of the Australian continent. Their analyses indicate minimum exposure ages ranging from 105±16 to 1310±190Ka, and maximum erosion rates of 0.3±0.1 to 5.7±1m/my. As expected, erosion rateslinearly increase with increasing annual precipitation. Dating of granitic domes (inselbergs) such as Mount Wudinnna rising aboved the peneplained Eyre Peninsula in the south indicate that there "have been and are eroding so slowing that they may well be direct descendents of Cenozoic and perhaps even Mesozoic forms" (p. 787), as previously claimed by Twidale.
References
Becker, B., 1993. A 11,000-year German Oak and Pine dendrochronology for radiocarbon calibration: Radiocarbon 35, pp. 201-213.
Beer, J., Siegenthaler, U., Bonani, G., Finkel, R.C., Oeschger, H., Suter, M. and Wölfli, W., 1988. Information on past solar activity and geomagnetism from 10Be in the Camp Century ice core. Nature 331, pp. 675-679.
Bierman, P.R., 1994. Using in situ produced cosmogenic isotopes to estimate rates of landscape: evolutionA review from the geomorphic perspective. Journal of Geophysical Research 99, pp. 13885-13896.
Cerling, T.E., and Craig, H., 1994, Geomorphology and in-situ cosmogenic isotopes: Annual Reviews of Earth and Planetary Sciences 22, pp. 273-317.
Frank, M., Schwarz, B., Baumann, S., Kubik, P.W., Suter, M. and Mangini, A., 1997. A 200 kyr record of cosmogenic radionuclide production rate and geomagnetic field intensity from 10Be in globally stacked deep-sea sediments. Earth and Planetary Science Letters 149, pp. 121-129.
Graf, T., Marti, K. and Wiens, R.C., 1996. The 21Ne production rate in a Si target at mountain altitudes. Radiocarbon 38(1), p. 155.
Kitagawa, H., and van der Plicht, J., 1998. Atmospheric Radiocarbon Calibration to 45,000 yr B.P.: Late Glacial Fluctuations and Cosmogenic Isotope Production, Science 279, pp. 1187- 1190.
Lal, D., 1991, Cosmic ray labeling of erosion surfaces: In-situ nuclide production rates and erosion models. Earth and Planetary Science Letters 104, pp. 424-439.
Mahaney, W.C., 1990. Ice on the Equator: Quaternary geology of Mount Kenya. W. Caxton Ltd., 386 p.
Nishiizumi, K., Kohl, C.P., Shoemaker J.R., Arnold, J.R., Klein, J., Fink, D. and Middleton, R., 1991. In situ 10Be and 26Al exposure ages at Meteor Crater, Arizona. Geochimica et Cosmochimica Acta 55, pp. 2699-2703.
Nishiizumi, K., Finkel, R.C., Klein, J. and Kohl, C.P., 1996. Cosmogenic production of 7Be and 10Be in water targets. Journal of Geophysical Research 101, pp. 22225-22232.
Phillips, F.M., Zreda, M.G., Elmore, D. and Sharma, P., 1996. A reevaluation of cosmogenic 36Cl production rates in terrestrial rocks. Geophysical Research Letters 23, pp. 949-952.
Phillips, F.M., Zreda, M.G., Smith, S.S., Elmore, D., Kubik, P.W., Dorn, R.I. and Roddy, D.J., 1991. Age and geomorphic history of Meteor Crater, Arizona, from cosmogenic Cl-36 and C-14 in rock varnish. Geochimica et Cosmochimica Acta 55, pp. 2695-2698.
Stone, J.O., 1999. A consistent Be-10 production rate in quartz - muons and altitude scaling. AMS-8 Proceedings Abstract Volume, Vienna, Austria.
Sutton, S.R., 1985. Thermoluminescence measurements on shock-metamorphosed sandstone and dolomite from Meteor Crater, Arizona. Journal of Geophysical Research 90(B5), pp. 3690-3700.
Links
Lecture notes on in situ cosmogenic
nuclides [PDF]
Lecture notes: introduction to
cosmogenic nuclides
Cosmic ray exposure dating with in situ
produced cosmogenic 3He: Results from young Hawaiian lava flows
Burial dating with 26Al and 10Be
Catchment-wide denudation rates from
cosmogenic nuclides in river sediment [PDF]
Erosion of the Rio Puerco Basin, New
Mexico from cosmogenic nuclide analysis
A 30,000 year erosion rate record from
cosmogenic nuclides in river terrace sediments, Massif Central,
France [PDF]
Cosmogenic isotope dating of a Sioux
Quartzite erosion surface, Southwestern Minnesota
Measuring Noble Gases with the MAP-215
Mass Spectrometer
The
paleoseismic history of the Nahef East Fault, Northern Israel,
using cosmogenic 36Cl
Cosmogenic Dating of Fault Faces - a New Tool for
Paleoseismology
Marek Zreda publication list
Terrestrial in-situ cosmogenic nuclide
page
