So, we have to assume these are annual rings, I am assuming for the most part they are, but over a long period (say thousands of years), a few extra rings here and there could give an older date for the tree.
But, you see, this method of calibration shows that, if there are any extra rings -- that are not detected (and it is possible to see them) or any missing rings (which can occur) doing a comparison of radiocarbon dates to tree ring counts shows that, at worst, the erroneous ring count can't be too far out -- that is less than the 10 % that the variations in C14 content can produce. If there were many extra rings (say) scattered in the tree we'd get a ring count that is too old and there would not be any kind of good match between the C14 date and ring date.
E.g., Let's say there are twice as many rings as years.
That means that the "ring year" line would be twice as steep as the "real year" line on a graph.
(see
Message 1 for an example of years from varve counting (like tree rings in behavior)
So if that were the case and we plotted C14 age against this where would the C14 line run?
There are lots of cases:
1) It could be utterly random. If the C14 in the atmosphere was changing drammatically year to year this might be the result.
2) It could match (more or less) the "real years" which is the hypothosis we started with. In this case it would be half as steep as the "ring years" line and very far from it pretty quickly. But if C14 decays as both the mathematics and tests say it does then it would still fall on a line and not be random.
3) It could match the false "ring years". This would happen if C14 decayed just about twice as fast as we have measured it to do. How likely is it that it would be measured wrong by just enough to match the mismeasure in the ring count?
What we see (as in the linked example) is that the C14 date more or less (around 10% variance) the counted date.
Not only that, it does it for many different trees, of different kinds, in different climates and in different places. There seems to be no possible explanation for these matches other than that both the C14 measurements and the tree ring counts are both tracking pretty close to the actual correct years. Can you think of another one?
Again, an assumption has to be made that all the “daughter” mass actually is daughter mass (came from the parent isotope), and was not already present in the sample.
This has nothing to do with C14 dating. In C14 dating we are measuring what we might call the "parent". Other forms of dating measure ratios of parents to daughters. It is not an issue here.
How this estimation can be done with any accuracy is beyond me.
It may be beyond you but in fact this can be checked in a number of ways. It is not a problem for the majority of dating tests.
™ Version 4.2