Carbon dating

[Brad2009]

Can anyone give a reason why cosmic ray flux normal to the earth's surface for all time must be a constant?

 

[Assyrian]

It isn't, which is why carbon dating is calibrated with tree rings and varves.

Check out http://en.wikipedia.org/wiki/Radiocarbon_dating#Calibration for an overview.

 

[laconicstudent]

Is it even worth mentioning that radiocarbon isn't the only radiometric dating method?

Potassium-Argon anyone?

 

[Papias]

Since the OP of the thread has been answered (cosmic rays aren't constant, and this is corrected by calibration curves), perhaps a more direct question would be:

"why do the various dating methods (including C14, K-Ar, varves, dendrochronology, ice cores, obsidian, protein racecimization, speleotherms, superposition, geologic event dating, geomagnetic polarity, Pb/U, association, Rb/St, and others), agree with each other when more than one can be used on the same sample?"


If methods are wrong, they'll give wrong answers. It seems odd to suggest that they'll happen to all give the same "wrong" answer.

Papias.

 

[Willtor]

Since the OP of the thread has been answered (cosmic rays aren't constant, and this is corrected by calibration curves), perhaps a more direct question would be:

"why do the various dating methods (including C14, K-Ar, varves, dendrochronology, ice cores, obsidian, protein racecimization, speleotherms, superposition, geologic event dating, geomagnetic polarity, Pb/U, association, Rb/St, and others), agree with each other when more than one can be used on the same sample?"


If methods are wrong, they'll give wrong answers. It seems odd to suggest that they'll happen to all give the same "wrong" answer.

Papias.

This is a really good point, Papias. Multiple unrelated sources for the same information are difficult to deny.

A popular debugging technique in software engineering is to test the correctness of an implementation of an algorithm by writing a second implementation of a different algorithm that solves the same problem and compare their outputs for the same input. It is highly unlikely that two different algorithms will produce the same wrong answer.

Likewise, if multiple different dating techniques all come out with approximately the same results, it's difficult to argue they are wrong.

 

[juvenissun]

Since the OP of the thread has been answered (cosmic rays aren't constant, and this is corrected by calibration curves), perhaps a more direct question would be:

"why do the various dating methods (including C14, K-Ar, varves, dendrochronology, ice cores, obsidian, protein racecimization, speleotherms, superposition, geologic event dating, geomagnetic polarity, Pb/U, association, Rb/St, and others), agree with each other when more than one can be used on the same sample?"


If methods are wrong, they'll give wrong answers. It seems odd to suggest that they'll happen to all give the same "wrong" answer.

Papias.

You made it too simple and too dramatic. Normally, no more than three methods on you list are used in the same study. In most cases, two, or just one. However, your question is still a legitimate one.

A simple answer to your question is: a consistent result may not be a correct result. In order to evaluate the nature of geologic time, it is a better idea to consider it together with cosmic time. Because the time measured by astronomy and by physics is the same time as in geology, yet is evaluated on very different principles. So far, very few people are putting them together.

 

[philadiddle]

A simple answer to your question is: a consistent result may not be a correct result. In order to evaluate the nature of geologic time, it is a better idea to consider it together with cosmic time. Because the time measured by astronomy and by physics is the same time as in geology, yet is evaluated on very different principles. So far, very few people are putting them together.

And you think that this will affect the results of all of these independent dating methods? Not only that, but it will work proportionately on each one in such a way that they will all agree with each other on dates that paint a completely different picture of our past? Please explain how this alternate view of time will accomplish this. It sounds fascinating.

 

[gluadys]

A simple answer to your question is: a consistent result may not be a correct result. In order to evaluate the nature of geologic time, it is a better idea to consider it together with cosmic time. Because the time measured by astronomy and by physics is the same time as in geology, yet is evaluated on very different principles. So far, very few people are putting them together.

Radiometric dating is based on physics and it does agree with geology. Time measured by astronomy also uses physics. Sounds to me like they are all put together.

 

[juvenissun]

And you think that this will affect the results of all of these independent dating methods? Not only that, but it will work proportionately on each one in such a way that they will all agree with each other on dates that paint a completely different picture of our past? Please explain how this alternate view of time will accomplish this. It sounds fascinating.

The calculation of radiometric age involved a few worldwide recognized constants. And, these constants are continuously changed for one reason or another. So old dates need to be corrected by the updated constants.

So, all radiometric ages are simply model ages. They are easy to be changed based on new discoveries.

 

[Papias]

Juve wrote:

You made it too simple and too dramatic. Normally, no more than three methods on you list are used in the same study. In most cases, two, or just one. However, your question is still a legitimate one.

Right - thanks for clarifying that not all of these methods is usually applicable to a given sample. However, two, three, four or five is not uncommon. For instance, an arrow will often be testable by C14, dendrochronology, amino racimization, obsidian hydration, and association. Or a piece of wood in a lake could easily be tested by varves, C14, dendrochronology, superposition, and association. And other sample can easily have more - there are over three dozen methods, after all.

More to the point, however: Isn't the agreement of even three methods pretty powerful evidence? I mean, if a smashed clock at a car wreck was stuck at 5:32, the nearby security camera taped the crash at 5:28, and a a receipt with a time of 4:56 was found in the car, wouldn't it be hard to suggest a time other than around 5:30 for the crash? That's only 2+ pieces of evidence (the receipt only indicates a time after 4:56), and though the measurements don't agree exactly (5:28 isn't exactly the same as 5:32), they all are consistent with a crash near 5:30. And that's not to mention samples with methods, or the fact that sample after sample, in place after place, hundreds of times, is shown to be consistent.

A simple answer to your question is: a consistent result may not be a correct result.

It's really hard to imagine several different methods just happening to all give the same "wrong" answer, isn't it? Are you saying that a bunch of different methods, that all give the same answer, are likely just "happening" to all agree? If so, I don't understand how.

In order to evaluate the nature of geologic time, it is a better idea to consider it together with cosmic time. Because the time measured by astronomy and by physics is the same time as in geology, yet is evaluated on very different principles. So far, very few people are putting them together.

Gluadys pointed out that they are put together. There is no problem here.

The calculation of radiometric age involved a few worldwide recognized constants. And, these constants are continuously changed for one reason or another. So old dates need to be corrected by the updated constants.

Yes, that's true. However, to be honest we need to recognize that these "changes" are small adjustments. It's not like the changes are taking a 20,000 year old date and changing it to a 10,000 year old date.


More important than the fact that this calibration is small, is the fact that it is consistent - as mentioned earlier in this thread, many samples give the same calibration, so the calibrated date is verified. You can see the calibration, based on many different dated samples (which all give about the same value for the calibration), on the graph at the bottom of this page:

Dr. Gary Hurd (Gary Hurd) on creationism, evolution, religion in society, religion and science: The Age of Rocks, Not the Rock of Ages

So, all radiometric ages are simply model ages. They are easy to be changed based on new discoveries.

See, now it sounds like you are saying that the many dating methods (I only mentioned a dozen or so, there are actually dozens of methods), which all agree with each other in samples that can be tested by more than one, all "happen" to all agree? If you are saying that dates confirmed by several other methods are often found to be incorrect, could you please provide some examples?

Thanks, and Merry Christmas-

Papias

 

[laconicstudent]

One does wonder if the OP was actually interested in the discussion.

 

[juvenissun]

Juve wrote:


Right - thanks for clarifying that not all of these methods is usually applicable to a given sample. However, two, three, four or five is not uncommon. For instance, an arrow will often be testable by C14, dendrochronology, amino racimization, obsidian hydration, and association. Or a piece of wood in a lake could easily be tested by varves, C14, dendrochronology, superposition, and association. And other sample can easily have more - there are over three dozen methods, after all.

More to the point, however: Isn't the agreement of even three methods pretty powerful evidence? I mean, if a smashed clock at a car wreck was stuck at 5:32, the nearby security camera taped the crash at 5:28, and a a receipt with a time of 4:56 was found in the car, wouldn't it be hard to suggest a time other than around 5:30 for the crash? That's only 2+ pieces of evidence (the receipt only indicates a time after 4:56), and though the measurements don't agree exactly (5:28 isn't exactly the same as 5:32), they all are consistent with a crash near 5:30. And that's not to mention samples with methods, or the fact that sample after sample, in place after place, hundreds of times, is shown to be consistent.




It's really hard to imagine several different methods just happening to all give the same "wrong" answer, isn't it? Are you saying that a bunch of different methods, that all give the same answer, are likely just "happening" to all agree? If so, I don't understand how.




Gluadys pointed out that they are put together. There is no problem here.



Yes, that's true. However, to be honest we need to recognize that these "changes" are small adjustments. It's not like the changes are taking a 20,000 year old date and changing it to a 10,000 year old date.


More important than the fact that this calibration is small, is the fact that it is consistent - as mentioned earlier in this thread, many samples give the same calibration, so the calibrated date is verified. You can see the calibration, based on many different dated samples (which all give about the same value for the calibration), on the graph at the bottom of this page:

Dr. Gary Hurd (Gary Hurd) on creationism, evolution, religion in society, religion and science: The Age of Rocks, Not the Rock of Ages



See, now it sounds like you are saying that the many dating methods (I only mentioned a dozen or so, there are actually dozens of methods), which all agree with each other in samples that can be tested by more than one, all "happen" to all agree? If you are saying that dates confirmed by several other methods are often found to be incorrect, could you please provide some examples?

Thanks, and Merry Christmas-

Papias

You mixed too many issues in the discussion. I only addressed the nature of radiometric dating.

And you did not get what I said. If two dating methods obtained the same age on one sample, it only confirmed the model age is precise. But it may not give the true age. (in here, I use the term "true age" more strictly than what it means to a geochronologist)

 

[shernren]

I think carbon dating is a pretty reliable thing.

I can't imagine myself making out with any kind of silicon-based lifeform, for example.

 

[Assyrian]

Wasn't that the one alien life form Captain Kirk didn't try to snog?

 

[sfs]

The calculation of radiometric age involved a few worldwide recognized constants. And, these constants are continuously changed for one reason or another. So old dates need to be corrected by the updated constants.

Please list the constants involved in radiometric dating whose estimated values have changed substantially(*) in the last thirty years. (Note that this does not include changes to C14 calibration, which is not a matter of a few constants.)

(*) I'll define "substantially" to mean, by enough to change radiometric dates by more than 5%

So, all radiometric ages are simply model ages. They are easy to be changed based on new discoveries.

No, they're actually extremely difficult to change based on new discoveries. That would require fundamental changes to our understanding of nuclear physics, and that hasn't happened in a very long time. No one has a working model of how they could be wrong and still be so consistent with each other, so your statement amounts to little more than, "Well, maybe they're all wrong somehow." Not exactly compelling science.

 

[juvenissun]

Please list the constants involved in radiometric dating whose estimated values have changed substantially(*) in the last thirty years. (Note that this does not include changes to C14 calibration, which is not a matter of a few constants.)

(*) I'll define "substantially" to mean, by enough to change radiometric dates by more than 5%


No, they're actually extremely difficult to change based on new discoveries. That would require fundamental changes to our understanding of nuclear physics, and that hasn't happened in a very long time. No one has a working model of how they could be wrong and still be so consistent with each other, so your statement amounts to little more than, "Well, maybe they're all wrong somehow." Not exactly compelling science.

For example: the continuous update on the value of decay constant in nearly all dating systems; and also the significant variation on isotope ratios. All these changes are supposed to show the improvement of precision.

The variation may be small, even it is incremental. But they are changing for one reason or another. Even this "age correction" does not affect the relative age relationship, so it may not cause any practical problem. But it does show that the accuracy of radiometric dating is limited to known physics. And, this is enough for me to say: even we do not know how could the value of dating parameters be dramatically changed, but it is possible. A similar argument is: even we do not know how could the time travel be practical, but we do not deny somehow it is possible. We are not talking about engineering, we are talking about science. Science can sense the existence of God by a one-in-a-billion probability, engineering can not.

 

[The Barbarian]

In that sense, it could be possible that orange leprechauns are hiding in your attic.

But in the absence of any evidence for it...

 

[Papias]

Juve wrote:

You mixed too many issues in the discussion. I only addressed the nature of radiometric dating.

Sure, that's what I'm talking about - whether or not radiometric dating can be trusted. The non radioactive methods are relevant because they confirm the radioactive methods. For instance, if I mentioned the use of historical records to confirm a C14 date of 1223 CE +/- 45 years, then my use of historical records is relevant to the topic, even though it is a non-radioactive dating method.

If two dating methods obtained the same age on one sample, it only confirmed the model age is precise. But it may not give the true age. (in here, I use the term "true age" more strictly than what it means to a geochronologist)

What do you mean by "model age"? If three dating methods agree on the age of a sample, why would any rational person not say that this is the real age? You are aware that geologists overwhelmingly agree that the Earth is 4.6 billion years old (real age, not "model age"), right?

You are aware that we often convict and sentence people to death on the agreement of two pieces of evidence, right? Such as, say, DNA evidence of the victim's blood in their car, consistent with, say, the suspect's DNA in hair found on the victim's clothes? With the dating situation we have three, four, or more sets of evidence in agreement, hundreds of times over - why would you, me or any sane person call the result a "model" age instead of a real age? I'm sorry if I misunderstood what you mean by "model" age - are you using that to just mean "real age"? Thanks-

Papias

 

[sfs]

For example: the continuous update on the value of decay constant in nearly all dating systems; and also the significant variation on isotope ratios. All these changes are supposed to show the improvement of precision.

The variation may be small, even it is incremental. But they are changing for one reason or another. Even this "age correction" does not affect the relative age relationship, so it may not cause any practical problem.

Right -- it does not cause any practical problem. There are small uncertainties on some of the physical constants involved. As new measurements are made, the values of those constants will be changed slightly and the uncertainties reduced. All of the changes will be tiny on the scale we're talking about, and negligible compared to other, routine experimental errors.

So why are you mentioning them, if they make no real difference to age estimates?

But it does show that the accuracy of radiometric dating is limited to known physics.

So is the accuracy of speedometers, and meat thermometers, and yardsticks. What's your point?

And, this is enough for me to say: even we do not know how could the value of dating parameters be dramatically changed, but it is possible.
A similar argument is: even we do not know how could the time travel be practical, but we do not deny somehow it is possible. We are not talking about engineering, we are talking about science.

It's also possible that the earth is a giant rutabaga, that a giant conspiracy of kittens controls the world's economy, and that Joe Biden is Elvis in disguise (and a two-headed space alien, of course). Lots of things are possible. Why bother mentioning this particular possibility, when you have no evidence that it's true?

Science can sense the existence of God by a one-in-a-billion probability, engineering can not.

What does God have to do with any of this? I thought we were talking about scientific estimates of the age of old objects.

 

[juvenissun]

Juve wrote:



Sure, that's what I'm talking about - whether or not radiometric dating can be trusted. The non radioactive methods are relevant because they confirm the radioactive methods. For instance, if I mentioned the use of historical records to confirm a C14 date of 1223 CE +/- 45 years, then my use of historical records is relevant to the topic, even though it is a non-radioactive dating method.




What do you mean by "model age"? If three dating methods agree on the age of a sample, why would any rational person not say that this is the real age? You are aware that geologists overwhelmingly agree that the Earth is 4.6 billion years old (real age, not "model age"), right?

You are aware that we often convict and sentence people to death on the agreement of two pieces of evidence, right? Such as, say, DNA evidence of the victim's blood in their car, consistent with, say, the suspect's DNA in hair found on the victim's clothes? With the dating situation we have three, four, or more sets of evidence in agreement, hundreds of times over - why would you, me or any sane person call the result a "model" age instead of a real age? I'm sorry if I misunderstood what you mean by "model" age - are you using that to just mean "real age"? Thanks-

Papias

As I said, consistent dates may not be the true age. All clocks pointed toward 3 PM may not mean it is 3 PM. There are other reasons to give consistent dates.