Geologic Column

By absolute dating methods, mostly radiometric.{Added in Edit} Typically we cannot date the sedimentary rocks in which fossils are found (although in some cases we can). So we date related igneous layers, usually but always by radiometric methods, and take advantage of Steno's Laws. It's probably easiest to understand from a few examples:We find fossils in layer A. Igneous layer B lies above layer A and is dated to 16,000,000 years. Igneous layer C lies below layer A and is dated to 18,000,000 years. There is no evidence that the three layers have been overturned or disturbed since formation. The fossils are between 16,000,000 and 18,000,000 years old.We find fossils in layer D. Igneous layer E lies below layer D and is dated to 48,000,000 years. Igneous "layer" F lies at about a 45 degree angle and cuts through both layer D and layer E, obvioulsy was formed after both of tehem, and is dated to 47,500,000 years. Other than "layer" F there is no evidence of overturning or disturbance to the layers. The fossils are between 47,500,000 and 48,000,000 years old.We find fossils in layer G. Igneous layer H lies below layer G and is dated to 32,000,000 years. We find the exact same fossils in layer I, somewhere far away from layers G and H. Igneous layer J lies above layer I and is dated to 31,000,000 years. These fossils are known to occur only in very thin strata and therefore only existed for a short period of time. The fossils are between 31,000,000 and 32,000,000 years old.Note that the dating is always done by absolute methods that have nothing to do with fossils, and the index fossils are only used to correlate between layers in different places, as in the last example. No. Nobody uses methods that are wrong any significant percentage of the time. The vast majority of the methods used are "age-diagnostic"; the method produces an age and an indication of how reliable that age is. If the samples are contaminated so as to prevent getting a good date, that's automatically detected.Wherever possible (and that's quite often) dates are obtained by multiple methods, both radiometric and non-radiometric, and almost always these different methods agree. (In the rare cases when they don't that's when things get really interesting, because that's where the real opportunities for new research are.) It's just not possible for there to be so much agreement unless the dates are correct. This agreement between methods is never addressed by creationists, no matter how often they are asked; they just try to deny that it exists. But it exists, and is a fact that must be explained by any viable theory. It is very rare for radiometric dating to give erroneous results for things for which we know the dates by other means. A few, a very few, erroneous dates obtained by creationists in "studies" that are usually if not always intentionally fraudulent is not "happens so much". When real scientists study the potential issues they do things like G.B. Dalrymple, “40Ar/36Ar Analyses of Historic Lava Flows,” Earth and Planetary Science Letters, 6 (1969): pp. 47-55. Many of the results are available at Ar-Ar Dating Assumes There Is No Excess Argon?, and I can send a PDF to anyone who is interested. Dalrymple studied 26 lava flows that occurred in historic times. He found that 19 of them had no excess argon, and would be dated correctly by K-Ar dating as soon as they get to be old enough. Of the remaining 8, 7 had so little excess argon that K-Ar dating would not be significantly affected once the rocks got old enough to be dated. One out of 26 lava flows had significant excess argon. So in the vast majority of cases K-Ar dates are correct ... and if only one out of the hundreds of thousands of dates we have is correct, the Earth is old.And he did all this using Ar-Ar, which is just one of the many widely used methods that detects contamination and open systems.Edited by JonF, : Add significantly to answer to first question