B5 · Publication Volume 10

Common Geochronological Systems

U–Pb, Ar–Ar, Rb–Sr, Sm–Nd and Re–Os applicability

Learning goals

After this lesson, you should be able to select a dating system by matching parent-daughter chemistry, half-life, mineral host, event and disturbance risk. You should be able to explain the different questions addressed by U-Pb, K-Ar or Ar-Ar, Rb-Sr, Sm-Nd, Lu-Hf and Re-Os approaches without treating any one system as a universal clock.

The goal is system literacy, not instrument operation. A robust project may use multiple systems because they date different minerals or boundaries. Agreement can strengthen an event model when the expected meanings are compatible; disagreement may be the primary evidence for inheritance, cooling, fluid reaction or later resetting.

Selecting a system from the geological question

Decision framework linking geological event, mineral host, parent-daughter system, closure behaviour and internal tests
Decision framework linking geological event, mineral host, parent-daughter system, closure behaviour and internal tests

Start with the event: magma crystallisation, metamorphic growth, cooling, sediment deposition, hydrothermal mineralisation, weathering or source differentiation. Identify minerals or materials that formed or changed during that event. Then ask whether they contain enough parent, exclude or constrain initial daughter, preserve useful internal domains and have suitable decay timescale.

Selection criteria include:

  • parent and daughter abundance and their crystal-chemical hosts;
  • half-life relative to expected age and analytical resolution;
  • common or initial daughter and disequilibrium corrections;
  • grain size, zoning, inclusions, alteration and diffusion behaviour;
  • availability of concordance, isochron or mineral-domain tests;
  • reference materials, calibrated constants and traceable data reduction; and
  • whether the dated material can be tied to field and petrographic sequence.

A high-precision system applied to the wrong object answers the wrong question precisely. An age should be planned backwards from the event claim and its falsification tests.

U-Pb systems and concordance

Uranium has two long-lived parent isotopes that decay to different lead isotopes. Minerals that incorporate uranium while excluding much initial lead can provide two related clocks. Zircon is widely used because it can accept uranium, commonly contains little initial lead, resists physical and chemical breakdown, and preserves growth domains. Those same strengths allow inheritance: an old core can survive younger magmatism.

Other uranium-bearing minerals can target different events and closure behaviour. Each material has its own common-lead, disequilibrium, diffusion and alteration problems. Imaging is essential when domains differ. A bulk grain dissolution can average core and rim; a small spot can isolate a domain but may intersect cracks, inclusions or mixed depth.

Concordance tests whether the related isotope systems agree with a closed-system history under the adopted constants and corrections. Discordance is not merely poor quality. Lead loss, mixing, inheritance or analytical effects can produce structured discordance. Chemical abrasion or domain selection may reduce damaged material in some workflows, but every treatment changes sampled support and must be documented.

K-Ar and Ar-Ar systems

Potassium-bearing minerals and rocks can accumulate radiogenic argon. Argon is a noble gas and can be lost by diffusion or fluid-assisted processes; excess argon can also produce old apparent ages. A K-Ar calculation requires potassium and radiogenic argon inventories on compatible material. An Ar-Ar approach irradiates material to produce a proxy related to potassium and can analyse step-heating spectra or spatial domains, using a calibrated irradiation parameter.

A plateau-like sequence of heating steps is not automatically a cooling age. Steps may mix diffusion domains, alteration, recoil, inclusions and excess components. Isochron relations and released-gas patterns provide tests, but geological interpretation requires mineral texture and thermal history. Fine grains, deformation and alteration can lower retention relative to an ideal intact crystal.

Decay constants and standards connect the numerical scale. Ages from different calibration conventions must be recalculated or compared with systematic differences included. Quoting small internal precision while omitting standard and constant uncertainty creates false comparability.

Rb-Sr, Sm-Nd, Lu-Hf and Re-Os systems

Rubidium and strontium partition differently among common minerals and can support mineral or whole-rock isochrons. Initial strontium isotope composition is also a source tracer. Open-system alteration and mixing can create scatter or pseudo-isochrons, so co-genesis and initial-ratio assumptions require field and petrographic support.

Samarium-neodymium and lutetium-hafnium systems use rare-earth or high-field-strength element behaviour and often retain source information through geological processes. Mineral isochrons can date equilibration or growth; whole-rock arrays may address differentiation but can also mix reservoirs. Model ages depend on assumed reservoir evolution and are not direct event ages.

Rhenium-osmium is useful in materials where these elements are concentrated, including some sulfides and organic-rich rocks. Initial osmium and common-component structure, mineral intergrowths, later fluid exchange and heterogeneity matter. A precise sulfide date is not automatically the age of all mineralisation in a deposit; it dates the analysed domain under its system model.

No list is exhaustive. Young deposits and surface processes may require shorter-lived or disequilibrium systems. The same selection logic applies: event, material, clock boundary, correction and independent test.

Sampling and analytical design

Build a paragenetic framework before separating minerals. Preserve photographs, thin-section or grain images and spatial coordinates. Separate inherited cores, metamorphic rims, alteration patches and fractures where the question requires them. Do not pool grains solely because they share a mineral name.

Use enough analyses to reveal populations and internal inconsistency, not merely enough to calculate a mean. Measure complementary chemistry that identifies domains and common components. Retain blanks, tracer or monitor data, reference-material results, interference corrections, covariance and rejected analyses. If a method destroys material, archive splits and images that preserve geological context.

Design cross-system tests around expected event order. A crystallisation date should not be younger than an unequivocally cross-cutting phase unless resetting, inheritance assignment or field interpretation is wrong. A cooling system is expected to record a later boundary than a high-retentivity crystallisation system, but deformation or fluid reaction can reverse simple assumptions. The prediction must be explicit before results are compared.

Worked selection exercise and review

Consider an invented igneous-metamorphic body with inherited zircon cores, igneous zircon rims, a later metamorphic mica fabric, cross-cutting sulfide veins and still later alteration along fractures.

  • To test magma crystallisation, target imaged zircon rims with U-Pb methods and retain inherited cores as a source constraint rather than averaging them.
  • To test metamorphic mineral growth or cooling, date the mica only after showing whether it grew with the fabric and whether later alteration affected argon retention.
  • To test source evolution, whole-rock Sm-Nd or zircon Hf information may be relevant, but model ages must remain separate from crystallisation ages.
  • To test sulfide-vein timing, a Re-Os approach may be suitable if the dated sulfide generation, common component and later exchange are characterised.
  • The fracture alteration may require another material and method; it must not be assigned the youngest existing date by assumption.

Review questions:

  1. Which result would be an analytical date, crystallisation age, cooling age, model age or mineralisation age?
  2. What texture must connect each analysed domain to the event?
  3. Which pairs should agree, which should differ, and why?
  4. How would inherited, excess or lost daughter components change apparent ages?
  5. What result would falsify the proposed sequence?

Interpretation workflow and uncertainty

  1. Write the event hypothesis and required temporal resolution.
  2. Map minerals and domains into a relative sequence.
  3. Select systems from parent-daughter chemistry, half-life and preservation.
  4. Predict expected relations among systems before measurement.
  5. Apply system-specific corrections and internal tests.
  6. Separate populations using geology and statistics together.
  7. Include calibration and constant uncertainty when comparing systems.
  8. Interpret discordance and disagreement rather than hiding them.
  9. Assign each age only to the boundary recorded by its material.

Failure modes

Frequent failures are choosing a system from availability rather than question, pooling domains, assuming zircon always dates crystallisation, calling every mica date cooling, treating a source model age as an event, and comparing ages on incompatible calibration scales. Another is using agreement as proof when both systems could have been reset by the same later process.

Systematic uncertainties can shift a whole dataset while internal errors remain small. Geological heterogeneity can create real dispersion. Report analytical, calibration and interpretation uncertainty separately and retain the full domain evidence.

Sources and further reading