The Age of the Earth

Strengths and weaknesses of radiometric and other dating methods

See also the discussions on Dating a Young Earth and Biblical Earth Dating

NASA's Blue Marble of Earth

NASA's BlueMarble NASA's Earth Observatory

Most people accept the current old-earth (OE) age estimate of around 4.6 billion years. This age is obtained from radiometric dating and is assumed by evolutionists to provide a sufficiently long time-frame for Darwinian evolution. And OE Christians (theistic evolutionists) see no problem with this dating whilst still accepting biblical creation, see Radiometric Dating - A Christian Perspective. This is the crucial point: it is claimed by some that an old earth supports evolutionary theory and by implication removes the need for biblical creation. Some claim Genesis in particular, and the Bible in general looks mythical from this standpoint.

A full discussion of the topic must therefore include the current scientific challenge to the OE concept. This challenge is mainly headed by Creationism which teaches a young-earth (YE) theory. A young earth is considered to be typically just 6,000 years old since this fits the creation account and some dating deductions from Genesis. The crucial point here is: if YE theory can be established scientifically, then macroevolutionary theory falls!

Accepted Dating Methods

Here we outline some dating methods, both absolute and relative, that are widely accepted and used by the scientific community. Absolute dating supplies a numerical date whilst relative dating places events in time-sequence; both are scientifically useful.

Radiometric Dating

This is based upon the spontaneous breakdown or decay of atomic nuclei. Radioactive parent (P) atoms decay to stable daughter (D) atoms e.g. the carbon isotope C-14 decays to nitrogen-14 and the uranium isotope U-235 decays to the lead isotope Pb-207. The time required for half the original number of parent atoms to decay is called the half life. Some half-lives are listed below:

  • U-235 to lead-207: 0.704 billion years
  • U-238 to lead-206: 4.47 billion years
  • Potassium-40 to Argon-40: 1.25 billion years
  • Rubidium-87 to Strontium-87: 48.8 billion years
  • Carbon-14 to Nitrogen-14: 5,730 years

It follows that uranium-lead, potassium-argon (K-Ar), and Rubidium-Strontium (Rb-Sr) decay can be used for very long time periods, whilst radiocarbon dating can only be used up to about 70,000 years. The K-Ar method is often used for rock dating. This uses a simple exponential decay formula linking the original number, Po, of parent atoms in rocks and minerals to the P atoms now present, thereby enabling an estimate of geological age. Mathematically, P = Po exp(-lambda T), where lambda = the decay constant and T = the period of decay. Assuming Po = P + D i.e. a zero initial number of D atoms, where P(or D) = the current number of P(or D) atoms, it follows that the age of a rock or mineral is computed as T=(1/lambda)ln(1 + D/P). Using radiometric techniques, the oldest dated minerals (4.0 - 4.2 billion years) are zircon crystals found in sedimentary rocks in western Australia.

One problem with earth dating is that the original earth surface is assumed to have eroded long ago. But assuming the earth was formed at the time of the rest of our solar system, then recovered moon rock and meteorites can also be used to estimate the age of the earth. These estimates give 4.4-4.5 billion years for moon rock, and 4.54 billion years for iron metreorites.

Non-radiometric Dating

These techniques utilize the physical parameters of the earth, such as ice cores, annual lake sediments, and astronomical cycles.

Ice cores from Greenland and Antarctica show annual layers (varves) and can be traced up to about 40,000 years before the layers become too thin due to compaction. Similarly, annual lake sediments can be used to estimate relative age and conventional interpretation for the Green River varves suggests they have been formed over some 20 million years. This implies the earth is at least 20 million years old.

Astronomical cycles can also be used to measure relative age. The earth precesses (wobbles like a spinning top) around the sun in a series of cycles. These cycles affect sunlight and hence long-term can form layers in rock. In some cases these astronomical cycles in rock appear to have been laid down over some 25 million years (and radiometric dating puts the absolute age of the rock at some 200 million years).

Dating Anomalies

Here we outline a few dating methods or 'clocks' that present a dating anomaly when referenced to the widely accepted OE age of 4.6 billion years. They appear to be inconsistent with an old earth.

Radio Carbon (C-14) Dating

At the outset we note C-14 cannot be used to directly date the earth for the simple reason that the unstable C-14 isotope has a half-life of just 5,730 years. In other words, half of the radioactive isotope in a sample would have decayed to Nitrogen-14 (N-14) in just 5,730 years. C-14 dating of carbon-bearing materials is therefore limited to roughly 50,000 years.

But YE scientists point out some anomalies in relation to C-14 and a very old earth. For instance, measurable amounts of C-14 have been found in fossil material, such as coal (traditionally Carboniferous period c300 mya). In fact, organic samples from every portion of the Phanerozoic record (spanning the last 500 million years on OE dating) show detectable amounts of C-14. The implication is that this organic material was either contaminated by new C-14, or it was buried much more recently and OE dating methods are suspect.

Ocean Measurements

One early approach was based upon ocean salinity [John Joly, 1800's]. This assumed the ocean was initially pure water and that it's salinity was derived from continental erosion. The technique gave 90 million years, but took no account of the non-constant erosion rate, or the loss and recycling of salt, or the fact that salt is obtained from other sources as well as continents. More recently, work has been done on ocean sediments [S. Nevins, Institute for Creation Research]. This suggests that, given the current annual rates of erosion (some 27.5 billion tons), all earth's continents would be delivered into the oceans in just 14 million years. Clearly, this seems incompatible with an ocean billions of years old. However, this may be a simplistic computation since there is Sediment Recycling as sediments accumulate and cause continental plates to collide, resulting in land uplift and subsequent errosion.

Earth's Magnetic Field

The Earth's magnetic field is thought to arise from circulating electic currents in the Earth's molten metalic core, and scientists agree that the field is weakening. Some claim that this decrease began two or three thousand years ago, and since monitoring began in the 1830's, scientists have observed a 10% decline in the magnetic dipole. At the current rate of decline it could take just 1,500 years to disappear, with increasing effects upon the electronic systems of satellites and spacecraft. Magnetic field decrease applies to other planets. For instance, recent satellite measurements show that Mercury's magnetic field is rapidly decaying and YE Creationists claim this indicates a young field. OE scientists believe that a weakened magnetic field could herald a new magnetic pole reversal.

Magnetic pole reversals are rebutted by YE creationists. Instead they claim that the field decrease can be used as a clock to date the earth since it has been decaying since the origin of the earth. Taking the half-life of the decaying magnetic moment at 1400 years, the field is now only about one third as strong as it was at the time of Christ. Working further back in time, the value of the earth's magnetic field approaches that of a magnetic star at 10,000 years ago. Since this would need a huge nuclear power source, it seems magnetic field decay places an upper age limit on the earth of the order of 10,000 years.

Whilst the rigour of this YE theory is disputed (see talkorigins), recent spacecraft observations have confirmed its planetary magnetic predicions made in 1984!

Atmospheric Helium

The decay of uranium and thorium isotopes results in a net build-up of Helium-4 atoms in the atmosphere. It is claimed to be increasing at an annual rate of 3.5 x 10^11 grams, and that there are some 3.5 x 10^15 grams of Helium-4 currently in the atmosphere [Nuclear Geology, H. Faul][Nature 179,213, M. Cook]. From these figures and known decay rates, it can be shown that the current amount of atmospheric Helium can be generated in just 11,000 years (not billions of years). The use of this approach to measure absolute geological age has been controversial for over 100 years and the YE it suggests has been attributed to Helium loss (see U-Th/He thermochronometry). On the other hand, some claim recent research supports a Helium Diffusion Age of 6,000 2,000 years.

Short-Period Comets

Short-period comets orbit the sun in less than 200 years (the Halley comet orbits about every 76 years). Each time they come close to the sun they lose material (the comet tail) and disintegrate. If no new comets are being generated, it would appear that no short-period comets can survive more than about 10,000 years - implying a young earth. This claim is countered by the fact that the origin of short-period comets is still uncertain and that there may be a source of short-period comets e.g. the Kuiper Belt Objects (KBOs) in the Uranus-Neptune zone, or the Oort Cloud.

Population Growth

This is perhaps one of the more challenging anomalies for OE science. It is claimed that Homo sapiens appeared some 600,000 to 200,000 years ago. But doesn't it seem strange that after more than 200,000 years earth's population is still only 7 billion? After all, the population increased from 1 billion in 1804 to 7 billion in 2011 - a span of just 207 years! Of course, population growth is exponential, but even then the numbers don't add up. Some claim a world-wide catastrophe may have occurred around 70,000 years ago, reducing the human population to maybe just 1,000 breeding pairs.

Let's do the maths on these 'catastrophe' figures. Starting with a population of No, after t years the population will have grown to N = No exp(rt), where r is the natural (non-immigration) percentage rate of population increase (r = birth rate - death rate). Of course, there are many factors that affect r, such as climate, disease, war, standard of living and so on. Typically, population growth rates are between 0.1% and 3% annually. For a conservative estimate we will take r = 0.1% and No = 1,000. Over a 70,000 year period we find N = 2.5 × 10 power 33. In words, earth's population should be some million, billion, billion, billion. This an impossibly large number when compared to the earth's current 7 billion people. Either the population growth calculation is hopelessly wrong, or the theory of human evolution is suspect!

On the other hand, taking the biblical data of just No = 8 people after the Flood (around 2,300 BC), we need a mean growth rate of r = 0.5% to achieve the present world population of 7 billion. This computation appears much more realistic.

Scientific Debate

Earth dating via ocean sediments, magnetic field decay, atmospheric helium, short-period comets (and other techniques) point to a young earth. However, the scientifc accuracy of YE claims are frequently challenged e.g. Evidences for a Young Earth? and Talk Origins. In order to balance the discussion we should also challenge the currently accepted radiometric dating methods. If these are suspect then the disputed methods take on more meaning.

Assumptions and Weaknesses of Radiometric Dating

Most rock dating methods rely on the following basic assumptions:

  • Initial conditions are known
  • Initial ratio of daughter/parent isotopes is known (zero date problem)
  • A constant decay rate
  • There is no leaching or addition of parent or daughter isotopes

There are several causes for concern here. The K-Ar method dates rocks by measuring the accumulated Ar-40. It is claimed the advantage of this method is that it circumvents the zero date problem i.e. there is little concern for the initial presence of the daughter isotope, Ar-40, since it is assumed that the inert gas escaped as the rock solidified. In other words, all Ar-40 in a rock is assumed to have been produced by in-situ radioactive decay of K-40 within the rock since it formed and there was zero Ar-40 in the rock when it solidified. However, this primary assumption has been challenged e.g. in Radiometric Dating Methods, Pitman, 2004 and in Radio Isotopes and the Age of the Earth, Vardiman et al. This 'zero Ar-40' problem has also been identified by Snelling who comments for one research project:

"Available evidence indicates the excessively old 'ages' are due to excess Ar-40 in the basalt which was not derived from in- situ decay of parent K-40 but inherited by the lava from its source."

Furthermore, since the Ar gas is not chemically bound to other atoms, it may leak in or out of samples and violate the assumption of no leaching/addition (see Radio Isotopes and the Age of the Earth). Certainly it is known to diffuse easily from deeper rocks under pressure so surface rocks tend to have a higher Ar-40 concentration than would be expected. This, coupled with the fact that potassium is easily washed out of minerals, suggests this technique can give an artificially high age for the earth and leads some to conclude that:

"... all K-Ar and Ar-Ar 'dates' of crustal rocks are questionable ..." [D. Pitman, Radiometric Dating Methods, 2004]

If we question these techniques, there is an alternative method called isochron dating. The isochron dating method attempts to combat the zero date problem by using ratios of isotopes and samples of different minerals from the same rock. For example, a plot of Sr-87/Sr-86 against Rb-87/Sr-86 for different minerals yields a straight line, and the slope of the line is simply (lambda)T where T is the age estimate. However, it still relies on certain basic assumptions, and in particular on the assumption that the specimen was entirely homogenous when it formed i.e. not layered or incompletely mixed. This is questionable for many isochron-derived dates, see Radiometric Dating Methods and Radio Isotopes and the Age of the Earth. The method also assumes that all mineral samples will have the same initial Sr-87 to Sr-86 ratio, but this is not always the case. So whilst isochron dating can give a straight line, the slope may have no significance [Vardiman et al].

Constant Decay Rate and Uniformitarianism

What about the radiometric assumption of constant decay rate? Such an assumption rests on the old evolutionary concept of uniformitarianism. In broad terms this means the observed geological features are the result of slow geological forces of the same kind and intensity as those found today. And for radiometric dating it means that the decay constant of the parent has not changed over earth's history. Scientific justification for this assumption is found for example in Radiogenic Isotope Geology, A. P. Dickin. The overall theme is that of a very old earth.

In contrast, Humphreys has proposed an accelerated decay (higher decay rate) early in earth's history, leading to a younger earth. This idea has been rebutted by those who claim there is no known scientific mechanism to produce such a change, see for example Claims of Accelerated Radioactive Decay and Tim-Thompson: decay rate. Others disagree and say that studies in theoretical physics suggest accelerated nuclear decay can occur e.g.:

  • Variation in compactified dimensions could affect coupling constants
  • Consequent variation in coupling constants could cause accelerated decay
  • Changes in potential well depth change the alpha-particle wave function
  • Changes in the alpha-particle wave function change decay half-lives

Uniformitarianism is also challenged if we invoke the concept of a world-wide flood (for which there is much evidence). Vardiman et al claim that this would result in unreliable radioisotopic dating. Vardiman et al conclude from their research that:

"Conventional radioisotopic dating methods are unreliable. The chief reason is that uniformitarianism is not a legitimate model of earth history. Observational evidence supports the recent occurrence of a global catastrophic flood."

From an Old Earth to a Young Earth?

The current scientific argument for an old earth is popular (especially in the media and education) whilst the concept of a young earth (as held by Creationism) is given low profile and so seems relatively weak. For example, non-radiometric dating techniques using ice cores do indeed appear to date the earth well in excess of 100,000 years. But there are several factors in favour of a young earth. These are largely ignored by mainstream science but could be the key to the massive discrepancy when it comes to dating the earth.

Biblical Earth Changes

The Bible records two dramatic, worldwide physical changes to the earth:

  • Physical changes to the earth at the Fall of Man
  • Physical changes to the earth during the Flood

At the Fall of man the whole of creation, including the earth, was suddenly subjected to corruption or decay (Rom 8.20-22). Man suddenly had a limited lifespan (Gen 2.17, 3.19) and the ground itself was 'cursed' resulting in cultivation problems (Gen 3.17,18). Also, at the Flood there were catastrophic geological changes, see for example geological evidence for the flood and scientific evidence. Some see these physical events as being related to changes in physical laws e.g. the introduction of the Second law of Thermodynamics, which essentially states that open or closed systems tend to deteriorate with time i.e. entropy increases [Creation Scientists Answer their Critics, D.T. Gish]. In short, the earth's order is deteriorating with time, and "the earth is wearing out like a garment" (Isa 51.6). This concept seems to be supported by theoretical physics, which suggests that a decrease in the speed of light, c, (see Is the Velocity of Light Constant in Time?) together with a change in other physical constants with time is a possible cosmological model of the universe, see Physical Constants and the Evolution of the Universe, Troitskii, 1987. Physical changes are also suggested from the biblical accounts of man living to over 900 years prior to the Flood (Gen 5), followed by an exponential decrease in age after the Flood. Some suggest this could be from a significant increase in radioisotopes in the atmosphere after the Flood.

Could these biblical events and the associated physical changes have caused accelerated radiometric decay, and by implication destroy uniformitarianism, the bedrock of radiometric dating? If so, standard radiometric dating must be corrected for an early accelerated decay rate, reducing millions of years to thousands! These biblically-implied abrupt physical changes in the earth are largely ignored in radiometric dating, which may be the source of the OE and YE discrepancy. These physical changes also affect the assumptions in radiocarbon dating and ice core dating. For more detail see A Young Earth Model.


For many Christians the jury is still out. The OE theory (and associated evolutionary theory) is well supported by high profile scientific bodies such as The Royal Society, and by the media. But there are serious dissenting scientific voices on evolutionary theory, and conventional earth dating techniques, and a growing Creation Science community make a good case for a Young Earth. Various dating clocks, such as the earth's decaying magnetic field and population growth suggest a young earth, and the classical radiometric dating assumption of Uniformitarianism has to be questioned given possible change in physical constants. Also, theologically it seems difficult to accept OE creationism (theistic evolution) and dismiss YE creationism when the Bible is read literally and when Jesus Himself implied a young earth (see biblical earth dating). The basic question seems to be "where is one's starting point?". For all Christians this should be:

'In the beginning God created the heavens and the earth' (Gen 1.1)

Bible quotations are from the New American Standard Bible