Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
7 online now:
Newest Member: popoi
Post Volume: Total: 915,809 Year: 3,066/9,624 Month: 911/1,588 Week: 94/223 Day: 5/17 Hour: 1/1


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   How, exactly, is dating done?
Will_Drotar
Inactive Member


Message 1 of 58 (67092)
11-17-2003 2:29 PM


I don't need to know how half-lifes work and all that jazz.
I'm just curious- do you measure the amount of alpha/beta radiation, do you use a proportion of the father/daughter isotope present in a sample (if so, how do we know how much of the father/daughter isotope was present in it to begin with), what machine do you use, etc. Thanks. TTYL Jesus loves you!

Replies to this message:
 Message 2 by Rei, posted 11-17-2003 2:53 PM Will_Drotar has replied
 Message 8 by JIM, posted 11-17-2003 5:32 PM Will_Drotar has not replied

Rei
Member (Idle past 7012 days)
Posts: 1546
From: Iowa City, IA
Joined: 09-03-2003


Message 2 of 58 (67095)
11-17-2003 2:53 PM
Reply to: Message 1 by Will_Drotar
11-17-2003 2:29 PM


The methods vary depending on the isotopes, but in general you compare the father and daughter isotopic ratio. However, it's more than that. For example, in isochron dating methods, you also compare the ratio of a different isotope of either the parent or daughter that normally exists in nature at a fixed ratio to the radioisotope; this helps you confirm whether what you've determined as the amount of the parent that has decayed to daughter is correct. Also, on controversial dates, multiple methods are generally used whenever possible.
Certain types of dates have a number of constraints as to when and where they can be used to obtain valid data (something that creationists often abuse). For example, carbon dating is not to be used in the following situations:
- When dating marine fossils, or species that consume a lot of marine organisms. The oceans recycle a lot of deep, old carbon.
- When dating fossils in highly volcanic areas. Volcanos recycle a lot of old carbon.
- Not for use with for fossils that have been taking in carbon since the 1950s (nuclear testing has thrown off ratios) without a high margin of error.
- Like all dating methods, not with fossils that are beyond the acceptable error range for a given method (carbon dating is for relatively "young" fossils only, while your uranium dating methods are only for relatively "old" fossils, as an example).
Due to the unusual nature of carbon dating, it is one of the few methods that has to be calibrated (it is typically calibrated with tree ring chronology). However, calibration factors, to the best of my knowledge, are all under 20% for dates before the 1950s. Other methods of dating typically use no calibration factor.
Again, what equipment you use depends on what dating method you use; here's how carbon dating is done - as step by step guide, with pictures.
------------------
"Illuminant light,
illuminate me."
[This message has been edited by Rei, 11-17-2003]

This message is a reply to:
 Message 1 by Will_Drotar, posted 11-17-2003 2:29 PM Will_Drotar has replied

Replies to this message:
 Message 3 by Will_Drotar, posted 11-17-2003 3:17 PM Rei has replied

Will_Drotar
Inactive Member


Message 3 of 58 (67102)
11-17-2003 3:17 PM
Reply to: Message 2 by Rei
11-17-2003 2:53 PM


I always thought multiple methods was the isochron. Is there a difference?
Does the HCl acid mess with the parent-daughter isotope ratio? Might that be kind of risky?
How do we know how much of the parent isotope/daughter isotope was present to start with? YEC's are telling me that we estimate. Please don't tell me they're right.

This message is a reply to:
 Message 2 by Rei, posted 11-17-2003 2:53 PM Rei has replied

Replies to this message:
 Message 4 by NosyNed, posted 11-17-2003 3:38 PM Will_Drotar has not replied
 Message 5 by JonF, posted 11-17-2003 4:11 PM Will_Drotar has not replied
 Message 6 by Rei, posted 11-17-2003 4:16 PM Will_Drotar has not replied
 Message 7 by JonF, posted 11-17-2003 4:27 PM Will_Drotar has not replied

NosyNed
Member
Posts: 8996
From: Canada
Joined: 04-04-2003


Message 4 of 58 (67109)
11-17-2003 3:38 PM
Reply to: Message 3 by Will_Drotar
11-17-2003 3:17 PM


Please don't tell me they're right.
LOL, given the history of the YEC sources what are the odds they will get it right?
I would expect there is a YEC or two here who might disagree with the above. If so, please jump in and point to an online source that does "get it right". This seems to be an area that they are particulary bad at.

This message is a reply to:
 Message 3 by Will_Drotar, posted 11-17-2003 3:17 PM Will_Drotar has not replied

JonF
Member (Idle past 167 days)
Posts: 6174
Joined: 06-23-2003


Message 5 of 58 (67134)
11-17-2003 4:11 PM
Reply to: Message 3 by Will_Drotar
11-17-2003 3:17 PM


I always thought multiple methods was the isochron. Is there a difference?
Yes. Isochron methods require multiple samples (typically 6 to 8 is a good number) from the same source. When these samples are measured and plotted on an appropriate graph, two things can happen:
1. The points do not lie on a straight line and the test results in a failure to find an age.
2. The points do lie on a straight line.
There are three sub-cases of case 2:
2A. The points also lie on a straight line on a different graph called a "mixing graph", and the test results in a failure to find an age because the isochron line is a result of mixing of two sources of different isotopic composition.
2B. The points do not lie on a straight line on a mixing graph, and the slope of the isochron line is an indicator of the age of the samples and the Y-intercept of the isochron line is the ratio of parent-to-daugher at solidfication for all the samples. Note that the initial amount of daughter isotope is measured by the isochron technique, and therefore isochron methods are not affected by daughter product present at solidification.
2C. The points do not lie on a straight line on a mixing graph, and we think we have a good age determination, but pure random chance made the seven or so points line up or the samples were formed by a complex mixing scheme of three or more sources with exceptionally unlikely isotopic makeups or some other very unlikely occurence.
There may be a few case 2C's in the literature, but it's pretty obvious that very unlikely events don't happen over and over and over again, so most are 2B (and some 2A's are pblished as known mixing isochrons).
Now, in reality, for various reasons the only isochron technique that is widely used today for determination of extremely precise ages (the question of approximate ages, within a few percent, having been settled long ago) is Argon-Argon, in which it is rare to actually draw an isochron diagram, but the methods that are used are equivalent to an isochron diagram (but more difficult to explain).
There's a short basic discussion of all radioisotope techniques at Radiometric Dating: A Christian Perspective, and there's a fairly technical but readable explantion of isochron techniques at Isochron Dating
Does the HCl acid mess with the parent-daughter isotope ratio? Might that be kind of risky?
Nope, and nope.
How do we know how much of the parent isotope/daughter isotope was present to start with? YEC's are telling me that we estimate. Please don't tell me they're right.
Don't worry, they're wrong.
In isochron dating, the initial parent/daughter ratio is one of the results of the analysis.
We've never found a rock that's 4.55 billion years old. We have rocks that are around 4 billion years old, and we've found zircon grains that are 4.4 billion years old. The 4.55 billion year age of the Earth is derived from Pb-Pb dating, an isochron method (and therefore not affected by initial daughter problems) that is a little too complex to explain here.
Probably the most popular technique for age determination today is U-Pb concordia-discordia, which is not an isochron technique but is instead a simultanous application of two independent dating methods. This method is very widely applicable, the half-life of uranium is known to far greater precision than any other half-life (guess why), and it can even give a valid age if there has been "open system behavior" (e.g. loss of Pb to diffusiion or leaching and what-not).
U-Pb dating requires that there be no significant daughter isotope at solidfiication, so it is used solely on minerals which strongly reject lead at solidification, usually zircons. It's physically impossible to solidify a zircon with enough lead to screw up U-Pb dating, unless there's also so little uranium that the dating wouldn't work anyway. The sexiest U-Pb dating is done essentially in situ in a Sensitive High-Resolution Ion MicroProbe, or SHRIMP (see Centre for Excellence in Mass Spectrometry). SHRIMP is so sensitive, and needs so small a sample size, that it can date sedimentary rocks by looking at the zenotime that forms between grains when the rock lithifies (see U-Pb SHRIMP Dating of Diagenetic Xenotime. Essentially all other radioisotope techniques are restricted to igneous rocks.
Then there's K-Ar dating, beloved of creationists and almost the only one they ever discuss. K-Ar does require an "assumption" of no radiogenic argon at solidification, but since argon is a gas this is almost always a good "assumption". But some errors have occurred because of "excess argon". However, any sample that can be dated by K-Ar can also be dated by Ar-Ar, an isochron method, and this is often done.
K-Ar dating is quite low cost and simple, and is still done, but is essentially never published today without the cross-check of another dating method. For some examples of rocks dated by multiple methods, see Consistent Radiometric dates and Radiometric Dating.
For a fascinating confirmation of the accuracy of Ar-Ar dating, see Precise dating of the destruction of Pompeii proves argon-argon method can reliably date rocks as young as 2,000 years and 40Ar/39Ar Dating into the Historical Realm: Calibration Against Pliny the Younger (free registration required for full text).

This message is a reply to:
 Message 3 by Will_Drotar, posted 11-17-2003 3:17 PM Will_Drotar has not replied

Rei
Member (Idle past 7012 days)
Posts: 1546
From: Iowa City, IA
Joined: 09-03-2003


Message 6 of 58 (67137)
11-17-2003 4:16 PM
Reply to: Message 3 by Will_Drotar
11-17-2003 3:17 PM


quote:
I always thought multiple methods was the isochron. Is there a difference?
Not exactly. (Edit: JonF described this in a lot of detail, so I'll save your time).
quote:
Does the HCl acid mess with the parent-daughter isotope ratio? Might that be kind of risky?
No. Hydrochloric acid contains two elements: hydrogen, and chlorine. It does not contain carbon (at least not a statistically significant amount at the purities they use in carbon dating). Acid cannot change isotopic ratios, which are determined at the atomic level, not the molecular level. If chemical reactions could readily cause nuclear decay, the atomic bomb would have been a lot easier to build. Chemical reactions don't mess with the nucleus.
quote:
How do we know how much of the parent isotope/daughter isotope was present to start with? YEC's are telling me that we estimate. Please don't tell me they're right.
As I mentioned concerning carbon dating (which is different from most methods), there is the assumption of the same C12/C14 ratio in the atmosphere as there was before the atomic bomb tests, with a calibration factor. The calibration factor is always less than 20% for dates before the 1950s, and is usually just a few percent - consequently, this cannot be used as an excuse for making a young earth look old. These calibration factors are largely determined by dendrochronology - i.e., checking carbon ratios in tree rings. While individual trees don't live that long, tree rings are effected by environmental factors - drought, flood, fire, etc. The areas of damage can be lined up in multiple trees in a given region. Bristlecone chronologies, for example, go back 9,000 years (and they're working on extending it). They don't just use a single line of trees; they do a statistical analysis on all of the trees that they can examine, to ensure that no matter how you line up the rings, significant events always match up. Before then, we have to rely on other methods (such as ice cores, which have annual lines) to attempt to determine the calibration factor - but again, it's always rather small factor. Ice cores correspond to dendrochronology, over the period of time that we have tree ring data.
How do we know what the original ratios were? Well, it depends on the method; different methods have additional methods to confirm what original ratios were. However, all of them have the same thing: They correspond to each other amazingly well. Even on cases where it is expected to have a strong degree of difficulty (such as dating from rubble piles), it's rare to get more than a 10% error between different methods. In most cases, a 1% error is about all you'll get. Isochron dating adds another correlation. This cross-correlation between entirely different methods ensures confidence in them. Furthermore, there's another issue that all share in common: geographical correlation. You can date a trilobite fossil from anywhere in the world, anywhere you like. It doesn't matter what sort of sediments it was preserved in - you'll never find a single trilobite fossil that dates more recently than the upper permian, within a small margin of error, as an example. Not just the fossil, but you can date the *rocks* that surround the trilobite (in case one thinks that the fossil itself is storing minerals strangely), anywhere in the same layer, and get the same date.
To get a young earth, you not only need to show that the daughter product was mostly in the rock to begin with, and that the parent wasn't, but you need to show why the ratios of *several different* minerals, all with different ratios of parent to daughter, come up with being just the precise amount decayed to present the same age.
One theory that creationists have proposed is rapid decay - that for some God-given reason, radioactive decay went faster in the past, and then reached a steady slow state today which doesn't change. This is, of course, nonsense. Radioactive decay is what heats the earth. Just the amount of decay that would have had to occur in surface granites to reach our current matching ratios would have ensured that the surface of the planet was a molten slag to this day.
It's good to see someone being inquisitive - I'm glad to have you here.
------------------
"Illuminant light,
illuminate me."
[This message has been edited by Rei, 11-17-2003]

This message is a reply to:
 Message 3 by Will_Drotar, posted 11-17-2003 3:17 PM Will_Drotar has not replied

Replies to this message:
 Message 9 by NosyNed, posted 11-17-2003 5:51 PM Rei has replied

JonF
Member (Idle past 167 days)
Posts: 6174
Joined: 06-23-2003


Message 7 of 58 (67142)
11-17-2003 4:27 PM
Reply to: Message 3 by Will_Drotar
11-17-2003 3:17 PM


FWIW, here's some real isochron diagrams: Some Real Isochron Graphs. Comments:
Each data point is a sample, and all samples on a graph were analyzed by the same technique.
Some of these are a little old, and the straight-line fit to the data might not be good enough to pass muster today.
Note that the Y-intercepts of the graphs for each parent-daughter pair are approximately the same.

This message is a reply to:
 Message 3 by Will_Drotar, posted 11-17-2003 3:17 PM Will_Drotar has not replied

JIM
Inactive Member


Message 8 of 58 (67180)
11-17-2003 5:32 PM
Reply to: Message 1 by Will_Drotar
11-17-2003 2:29 PM


I'm just curious- do you measure the amount of alpha/beta radiation, do you use a proportion of the father/daughter isotope present in a sample (if so, how do we know how much of the father/daughter isotope was present in it to begin with), what machine do you use, etc. Thanks.
There are a few things to keep in mind when you're dating objects using isotopic methods such as the one you describe.
What you do with isotopic dating is to count the number of atoms of the "parent" nuclide and the number of atoms of the "progeny" (also called "daughter") nuclide. This is what can tell you the age of the rock. Say, for example, you are dating a rock using the uranium-lead dating method. To find the age, you will simply count the number of uranium atoms and the number of lead atoms. There are two primary isotopes of uranium, U-235 (with a half-life of about 700 million years) and U-238 (with a half-life of 4.38 billion years). U-235 decays to lead (Pb-207, to be precise). If you count exactly the same number of atoms of both U-235 and Pb-207, then you know that the uranium has been in the rock for exactly one half-life because, after one half-life, one half of the U-235 atoms will have decayed to lead, and the other half will still be uranium-235. So, if you count equal numbers of parent and progeny atoms, you know that the rock is exactly one half-life old. In this case, it would be 704 million years in age.
As the rock gets older, you have fewer and fewer parent atoms and more progeny. So, a rock with a 3:1 ratio of progeny to parent atoms would be 2 half-lives old (because there would be 1/4 of the parent atoms left, so 3/4 would be progeny, giving a 3:1 ratio).
For more information about isotopic dating, you can look in just about any high-school or college-level geology text book. One that is particularly good is "Earth", by Press and Siever. Another book that is outstanding at discussing isotopic dating is "The Age of the Earth", by Brent Dalrymple. And, on a more technical level, there are two college textbooks that are also outstanding. One is called "Isotope Geology" by Gunter Faure, and the other is called "Radiogenic Isotope Geology", by Alan Dickin.
[This message has been edited by JIM, 11-17-2003]

This message is a reply to:
 Message 1 by Will_Drotar, posted 11-17-2003 2:29 PM Will_Drotar has not replied

NosyNed
Member
Posts: 8996
From: Canada
Joined: 04-04-2003


Message 9 of 58 (67189)
11-17-2003 5:51 PM
Reply to: Message 6 by Rei
11-17-2003 4:16 PM


that for some God-given reason, radioactive decay went faster in the past
I won't bother with finding the sites unless someone really wants to but:
There is a creationist site that actually sites specific scientific papers about difference in the rate of radioactive decay. They talk about "bound state beta decay" and another condition which causes very significant changes in the rate of change of nucluli. This is, of course, used to support the suppostition that dating of rocks could be wrong because decay rates can TOO change.
Unfortunately for the credibility of this source it leaves out a couple of facts. I would say that leaving them out makes it dishonest but perhaps that is in the mind of the beholder.
Bound state beta decay (only one of the ways for something to decay) requires complete or nearly complete ionization of the atom in question. This is not mentioned (odd that) nor are the conditions required to produce this mentioned. Not mention is made of the chances of the rocks of the earth being in a plasma state anytime during the past 4 Gyrs made.
The other change does mention the conditions --- I don't remember the exact numbers -- something like 200 million degress if I remember right. That fact that this might apply to a supernova and not even the core of the sun seems to have gotten left out. Something of an oversight perhaps?

This message is a reply to:
 Message 6 by Rei, posted 11-17-2003 4:16 PM Rei has replied

Replies to this message:
 Message 10 by Rei, posted 11-17-2003 6:03 PM NosyNed has not replied
 Message 11 by JonF, posted 11-17-2003 6:12 PM NosyNed has not replied

Rei
Member (Idle past 7012 days)
Posts: 1546
From: Iowa City, IA
Joined: 09-03-2003


Message 10 of 58 (67195)
11-17-2003 6:03 PM
Reply to: Message 9 by NosyNed
11-17-2003 5:51 PM


quote:
The other change does mention the conditions --- I don't remember the exact numbers -- something like 200 million degress if I remember right. That fact that this might apply to a supernova and not even the core of the sun seems to have gotten left out. Something of an oversight perhaps?
You know, I bet if you took the atoms, ripped off all their electrons but kept them tightly contained in a penning trap, cooled them to near becoming a bose-einstein condensate (so that they won't move around to much), and then fired an intense barrage of high energy radiation at them, you could probably increase their half lives that way, too. Or perhaps if you took them and put them in a particle accelerator, ramped them up to a few TeV's, and collided them with a wall.... See? There's lots of ways that you could cause the atoms to break down faster!
... so that they can release their energy faster, and give the Earth's crust a nice meltdown....
------------------
"Illuminant light,
illuminate me."
[This message has been edited by Rei, 11-17-2003]

This message is a reply to:
 Message 9 by NosyNed, posted 11-17-2003 5:51 PM NosyNed has not replied

JonF
Member (Idle past 167 days)
Posts: 6174
Joined: 06-23-2003


Message 11 of 58 (67200)
11-17-2003 6:12 PM
Reply to: Message 9 by NosyNed
11-17-2003 5:51 PM


something like 200 million degress if I remember right
David Ewan Kanaha, in the first link below, says 109 degrees K.
Some links that are relevant to your post:
Modifications of Nuclear Beta Decay Rates
Re: Some more on radioactive decay, and my references
Changing Decay Rates?
[This message has been edited by JonF, 11-17-2003]

This message is a reply to:
 Message 9 by NosyNed, posted 11-17-2003 5:51 PM NosyNed has not replied

Replies to this message:
 Message 12 by AdminNosy, posted 11-17-2003 7:04 PM JonF has not replied

AdminNosy
Administrator
Posts: 4754
From: Vancouver, BC, Canada
Joined: 11-11-2003


Message 12 of 58 (67211)
11-17-2003 7:04 PM
Reply to: Message 11 by JonF
11-17-2003 6:12 PM


Thank you, you did my work for me! I should be embarrassed but I didn't think it mattered all that much.
Oops posted under the wrong name.
[This message has been edited by AdminNosy, 11-17-2003]

This message is a reply to:
 Message 11 by JonF, posted 11-17-2003 6:12 PM JonF has not replied

Will_Drotar
Inactive Member


Message 13 of 58 (67757)
11-19-2003 3:21 PM


I read Henry Morris's rebuttal to isochrons in one of his books. It was like one paragraph long and it said that he didn't want to get into all the technicalities. Which means he didn't have anything better to say.
But he did mention, very briefly, pseudo-isochrons. Can someone explain to the ignorant-minded, namely me, what those are and how those are drawn? Thanks a bunch- TTYL Jesus loves you!

Replies to this message:
 Message 14 by Rei, posted 11-19-2003 3:43 PM Will_Drotar has not replied
 Message 15 by JonF, posted 11-19-2003 5:04 PM Will_Drotar has not replied

Rei
Member (Idle past 7012 days)
Posts: 1546
From: Iowa City, IA
Joined: 09-03-2003


Message 14 of 58 (67766)
11-19-2003 3:43 PM
Reply to: Message 13 by Will_Drotar
11-19-2003 3:21 PM


Here's an excellent page on radiometric dating. It covers pseudo-isochrons.
The Skeptic Files - SkepticFiles Setting
------------------
"Illuminant light,
illuminate me."

This message is a reply to:
 Message 13 by Will_Drotar, posted 11-19-2003 3:21 PM Will_Drotar has not replied

JonF
Member (Idle past 167 days)
Posts: 6174
Joined: 06-23-2003


Message 15 of 58 (67779)
11-19-2003 5:04 PM
Reply to: Message 13 by Will_Drotar
11-19-2003 3:21 PM


I read Henry Morris's rebuttal to isochrons in one of his books. It was like one paragraph long and it said that he didn't want to get into all the technicalities. Which means he didn't have anything better to say
Woodmorappe/Peczkis has written a large-format book of about 90 pages, "The Mythology of Modern Dating Methods", in which he gives some fairly detailed criticisms and a lot of quote-mining and misrepresentation, as is his wont. Kevin Henke has posted several rebuttals to parts of this book, many at No Answers in Genesis (there's a search box near the bottom of the page).
But he did mention, very briefly, pseudo-isochrons. Can someone explain to the ignorant-minded, namely me, what those are and how those are drawn?
The most common cause of a false isochron is a mixing isochron, where two sources with different isotopic makeup mix. There's a good discussion and simple example at Isochron Dating: Mixing of Two Sources. You might have to read the beginning part to understand the notation.
Some creationist once published a scenario in which he demonstrated that mixing of three sources of peculiar composition (IIRC, one source had to contain only radiogenic daughter and no other isotopes of the daughter, which never happens) could form an isochron and the mixing plot described in the reference above wouldn't detect it. I can't find a reference now. But the idea that such things happen regularly is pretty ludicrous.
IMHO, the major indicator of the reliability of radioisotope dating is the excellent correlations between different radioisotope methods and the excellent correlations between radioisotope methods and non-radioisotiope methods.

This message is a reply to:
 Message 13 by Will_Drotar, posted 11-19-2003 3:21 PM Will_Drotar has not replied

Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024