Radiocarbon dating occupies a narrow but unusually detailed part of geological time. It does not measure the age of most rocks, nor can it reach back to the dinosaurs or the formation of continents. Its strength lies in younger deposits containing carbon that once circulated through a living organism or another exchangeable carbon reservoir. Within that range, it can connect wood, charcoal, peat, shells, and other carbon-bearing remains to the chronology of sediments and environmental change.
The method’s upper boundary is often described as about 50,000 years, although some laboratories can detect radiocarbon in carefully prepared samples approaching 60,000 years old. That is not a sharp wall between datable and undatable material. Reliability fades progressively as the remaining signal becomes extremely small and increasingly difficult to distinguish from contamination and measurement background.
A Clock Begins When Carbon Exchange Ends
Cosmic-ray reactions continually produce carbon-14 in the upper atmosphere. It becomes carbon dioxide and joins the active carbon cycle alongside stable carbon-12 and carbon-13. Plants acquire it through photosynthesis; animals obtain it through food; and dissolved carbon enters aquatic systems. While an organism lives, exchange broadly maintains a relationship with the reservoir supplying its carbon.
After death, uptake stops while carbon-14 continues to decay into nitrogen-14. Its half-life is about 5,730 years. Measuring how much remains relative to stable carbon allows the elapsed time since carbon exchange ended to be estimated. Accelerator mass spectrometry counts carbon isotopes directly and can work with much smaller samples than older decay-counting techniques.
The event dated is therefore not automatically the burial of a layer. A long-lived tree’s inner ring formed before the tree died, and timber may have been reused. Reworked charcoal can enter sediment centuries after the plant grew. A valid measurement can still answer the wrong geological question if the sampled carbon is older or younger than the depositional event being reconstructed.
Why the Useful Signal Eventually Disappears
After one half-life, half of the original carbon-14 remains; after ten, only about one-thousandth survives. Near the method’s practical limit, traces of younger carbon can have an outsized effect. Rootlets, conservation substances, soil humic compounds, adhesives, and laboratory residues may all make an old sample appear younger. Pretreatment removes identified contaminants, but harsh cleaning can also leave too little original material to measure.
Background carbon inside instruments and reagents sets another constraint. Laboratories analyze blanks that should contain no detectable carbon-14 and standards with known compositions, then use those results to quantify correction and uncertainty. As a sample’s activity approaches the blank level, its calculated age becomes more sensitive to small assumptions. Reporting a finite number does not guarantee that the number resolves a real difference from background.
At the young end, the problem is different. Fossil-fuel emissions diluted atmospheric carbon-14 with ancient carbon lacking the isotope, while nuclear weapons testing sharply increased atmospheric levels during the mid-20th century. These changes complicate recent dates but also create a recognizable bomb-pulse marker that can resolve events from the last several decades when the correct atmospheric history is applied.
Radiocarbon Years Require Calibration
A conventional radiocarbon age is not identical to a calendar age. Atmospheric carbon-14 production has varied through time with solar activity, Earth’s magnetic field, and carbon-cycle changes. By convention, radiocarbon ages are calculated with the historically adopted Libby half-life of 5,568 years and reported in years before present, where “present” means 1950. Calibration converts that laboratory timescale into calendar years.
Calibration curves are assembled from independently dated records, especially tree rings for the younger interval and combinations of corals, cave deposits, lake or marine sediments, and other archives farther back. The current IntCal family extends calibration to 55,000 calendar years before 1950, with separate curves for Northern Hemisphere atmosphere, Southern Hemisphere atmosphere, and surface-ocean carbon.
Because the calibration curve contains bends, plateaus, and reversals, one measured radiocarbon age may correspond to more than one possible calendar interval. A calibrated result is commonly expressed as one or several probability ranges rather than a single exact year. Greater analytical precision cannot remove ambiguity created by a flat section of the curve.
Reservoirs and Context Define the Answer
Marine organisms can appear older than contemporaneous terrestrial material because ocean carbon may spend long periods away from the atmosphere. Regional upwelling and circulation alter this marine reservoir effect. Freshwater systems can also contain old dissolved carbon derived from carbonate rock or deep groundwater. Applying an atmospheric curve to such a sample without correction can shift the interpretation by centuries or more.
Material selection must therefore come before calculation. Researchers ask what carbon reservoir supplied the sample, whether it remained closed after formation, whether it was reworked, and how securely it belongs to the layer or event of interest. Multiple dates from a stratigraphic sequence can be combined with the known order of layers, but statistical modeling cannot rescue poorly understood samples.
Radiocarbon dating is most informative when its limits are treated as part of the evidence. It offers exceptional resolution across the late Quaternary, yet every result carries the history of a particular carbon source. The method measures isotope decay; geological reasoning establishes what that measurement actually dates.