[A Conventional Radiocarbon Age or CRA, does not take into account specific differences between the activity of different carbon reservoirs. A CRA is derived using an age calculation based upon the decay corrected activity of the absolute radiocarbon standard (1890 AD wood) which is in equilibrium with atmospheric radiocarbon levels (as mentioned previously, 1890 wood is no longer used as the primary radiocarbon standard, instead Oxalic Acid standards I and II were correlated with the activity of the original standard). In order to ascertain the ages of samples which were formed in equilibrium with different reservoirs to these materials, it is necessary to provide an age correction. Implicit in the Conventional Radiocarbon Age BP is the fact that it is not adjusted for this correction. In this page, we consider natural reservoir variations and variations brought about by human interaction].

Natural Corrections

Reservoir effects

Radiocarbon samples which obtain their carbon from a different source (or reservoir) than atmospheric carbon may yield what is termed apparent ages. A shellfish alive today in a lake within a limestone catchment, for instance, will yield a radiocarbon date which is excessively old. The reason for this anomaly is that the limestone, which is weathered and dissolved into bicarbonate, has no radioactive carbon. Thus, it dilutes the activity of the lake meaning that the radioactivity is depleted in comparison to 14C activity elsewhere. The lake, in this case, has a different radiocarbon reservoir than that of the majority of the radiocarbon in the biosphere and therefore an accurate radiocarbon age requires that a correction be made to account for it.

One of the most commonly referenced reservoir effects concerns the ocean. The average difference between a radiocarbon date of a terrestrial sample such as a tree, and a shell from the marine environment is about 400 radiocarbon years (see Stuiver and Braziunas, 1993). This apparent age of oceanic water is caused both by the delay in exchange rates between atmospheric CO2 and ocean bicarbonate, and the dilution effect caused by the mixing of surface waters with upwelled deep waters which are very old (Mangerud 1972). A reservoir correction must therefore be made to any conventional shell dates to account for this difference. Reservoir corrections for the world oceans can be found at the Marine Reservoir Correction Database, a searchable database online at Queen's University, Belfast and the University of Washington. Human bone may be a problematic medium for dating in some instances due to human consumption of fish, whose C14 label will reflect the ocean reservoir. In such a case, it is very difficult to ascertain the precise reservoir difference and hence apply a correction to the measured radiocarbon age.

Spurious radiocarbon dates caused by volcanic emanations of radiocarbon-depleted CO2 probably also come under the category of reservoir corrections. Plants which grow in the vicinity of active volcanic fumeroles will yield a radiocarbon age which is too old. Bruns et al. (1980) measured the radioactivity of modern plants growing near hot springs heated by volcanic rocks in western Germany and demonstrated a deficiency in radiocarbon of up to 1500 years through comparison with modern atmospheric radiocarbon levels. Similarly, this effect has been noted for plants in the bay of Palaea Kameni near the prehistoric site of Akrotiri, which was buried by the eruption of the Thera volcano over 3500 years ago (see Weninger, 1989). The effect has been suggested as providing dates in error for the eruption of Thera which has been linked to the demise of the Minoan civilisation in the Aegean. One modern plant growing near the emanations had an apparent age of 1390 yr. The volcanic effect has a limited distance however. Bruns et al. (1980) found that at 200 m away from the source, plants yielded an age in agreement with that expected. They suggested that the influence of depleted CO2 declined rapidly with increasing distance from the source. Radiocarbon discrepancies due to volcanic CO2 emissions are a popular source of ammunition for fundamentalist viewpoints keen to present evidence to show that the radiocarbon method is somehow fundamentally flawed.

Recent Human derived disturbances

Suess or Industrial effect

Since about 1890, the use of industrial and fossil fuels has resulted in large amounts of CO2 being emitted into the atmosphere. Because the source of the industrial fuels has been predominantly material of infinite geological age ( e.g coal, petroleum), whose radiocarbon content is nil, the radiocarbon activity of the atmosphere has been lowered in the early part of the 20th century up until the 1950's. The atmospheric radiocarbon signal has, in effect, been diluted by about 2%. Hans Suess (1955) discovered the industrial effect (also called after him) in the 1950's. A number of researchers found that the activity they expected from material growing since 1890 AD was lower. The logical conclusion from this was that in order to obtain a modern radiocarbon reference standard, representing the radiocarbon activity of the 'present day', one could not very well use wood which grew in the 1900's since it was affected by this industrial effect. Thus it was that 1890 wood was used as the modern radiocarbon standard, extrapolated for decay to 1950 AD.

Atom bomb effect

Since about 1955, thermonuclear tests have added considerably to the C14 atmospheric reservoir. This C14 is 'artificial' or 'bomb' C14, produced because nuclear bombs produce a huge thermal neutron flux. The effect of this has been to almost double the amount of C14 activity in terrestrial carbon bearing materials (Taylor, 1987).

De Vries (1958) was the first person to identify this 'Atom Bomb' effect. In the northern hemisphere the amount of artificial carbon in the atmosphere reached a peak in 1963 (in the southern hemisphere around 1965) at about 100% above normal levels. Since that time the amount has declined owing to exchange and dispersal of C14 into the Earth's carbon cycle system. The presence of bomb carbon in the earth's biosphere has enabled it to be used as a tracer to investigate the mechanics of carbon mixing and exchange processes. Ellen Druffel has called this the silver lining in thermonuclear bomb testing. The GEOSECS (Geochemical Ocean Section Study) oceanographic programme, for example, involved the collection and measurement of samples of ocean water along a number of Pacific and Atlantic transects to map the presence of bomb carbon and enable modellers to analyse the pathway of radiocarbon and its exchange and residence times. Currently, at the Woods Hole AMS Laboratory, the World Ocean Circulation Experiment (WOCE) is underway, this link shows the transects across the East Pacific ocean where C14 measurements of dissolved inorganic carbon have been obtained. You can see the dispersal of bomb carbon into the upper layers of the Pacific. Reidar Nydal and Knut Lovseth have compiled 14C data sets of changes in atmospheric carbon dioxide between 1962 and 1993.

More relevant information:

Radiocarbon Calibration

Pretreatment and Contamination

Isotopic fractionation

Summary: The accuracy of radiocarbon dates (modified from Polach, H.A. 1976).

Sources of Error Effect upon Age Determination Measures to minimise the error incurred
1. Precision of age determination Statistical:Typically 1%Modern or less Big samples, longer count times, repeat sample assays
2. Inherent
a. C14 half-life

Libby half life 3% too low

Multiply CRA's by 1.03 if necessary
b. C13/C12 fractionation Variable, up to 450 yr for shell. Stable isotope analyses using Mass Spec.
c. C14 Modern standard Variable > 80 yr International crosscheck of secondary standards.
d. Variation in past C14 production rates 0-800 yr, beyond ca12 ka not determined Tree ring calibration; otherwise interpret results in radiometric timescale.
e. Distribution of C14 in nature Surface ocean latitudinal dependence -400 to -750 yr. Deep ocean -1800 yr. Interpretation of results.
f. Changes of C14 concentration in the atmosphere. Industrial effect ca -2.5% and atom bomb effect +160% in atmosphere Interpretation of results
3. Contamination. Nil to 300 yr up to 15 ka; >20 ka possible beyond 25 ka. Interpretation of results, analysis and dating of extracted pretreated fractions.
4. Biological age of material <10 yr to>1000 yr Identification of species of material in the case of wood and charcoal to short lived samples only.
5. Association of sample and event Intermediate Interpretation of results
6. Human Intermediate Care in field and laboratory
7. Interpretation of results Intermediate Care in interpretation, interdisciplinary approach and collaboration

 INDEX | Introduction | Measurement | Applications | WWW Links | k12 | Publication | Corrections |
Age calculation | Calibration | Pretreatment | References | Awards | Credits | What's new? | Email |