AboutDr Thomas Bell Expertise I can answer questions regarding surface earth processes and the chemical transformations that sediments and rocks undergo with burial. I can also answer questions regarding deep time, the evolution of the elements, and the last 4.5 billion years of earth history. I specialize in metallic ore forming processes, the major geologic time periods when they were produced and what they tell us about the evolution of our planet. Learn more about my professional interests at Stratamodel.com.
Experience I am a professional consulting geologist with a background in the petroleum, mining, environmental, and geotechnical industries with over 25 years of experience.
Education/Credentials Ph.D., Geology, University of California at Berkeley, 1984
M.A., Geology, University of California at Berkeley, 1980
B.S., Geology, San Jose State University, 1978
I'm an undergraduate student who is working towards a degree in Ecology and Evolutionary Biology. Through my pleasure reading, I've become interested in the idea of using "molecular clocks" to infer ages since divergence from common ancestry. From what I understand, this method seems to hinge on using dated strata as a guide to yield results. I understand how radioactive decay works and that we use calibrations for 14 C readings, for example, since the rate of atmospheric nitrogen has fluctuated over long periods of time. I have many questions about dating methods and how we calibrate but I will ask my broad question: How do we know that the date readings we're obtaining using radiometric methods are accurate?
I'm training to be a biologist but I'm fascinated with geology as well. Any knowledge you can give me, or related sources you can recommend for further study, will be much appreciated. Thank you.
Answer Tom,
The theory upon which radiometric dating is well established and it sounds as if you have a good grasp of this. Your question seems to be more closely related to the errors associated with the measurements themselves. All measurements of natural systems have two sources of error. The total error is the sum of sampling error and analytical error. Sampling error is generated in the field through the procedures used to collect material to analyze in the lab. Analytical error is generated in the lab as the field sample is processed.
Lets deal with analytical error first. Any time we measure something, we get a slightly different result. The range of results may be of any magnitude and this usually depends upon the care we take in measurement and the precision of the instruments we use. The error associated with this phenomenon is random so given enough measurements, the average value approaches the 'true' value. If there is a systematic error like one caused by an improperly calibrated instrument, it can be harder to recognize and may only become apparent when a sample is divided into subsamples and those are analysed on two different instruments. In general, a careful analyst guards against systematic errors by running blank samples, known standard samples, duplicate samples, and replicate samples. If this is done correctly, the remaining error will be reduced to the random error associated with all measurements.
Sampling error is generated at the time material for dating is collected in the field. Consider for example a dig into a cave floor with many layers some of which contain charcoal from ancient fires. Cross contamination from one layer to another can take place before sampling by natural processes or during the dig itself. Quantifying these errors is very difficult. Or, consider a thin layer of volcanic ash containing small crystals of feldspar that is interbedded with other ashfalls. Isolating the crystals from a single cm thick layer without getting a few strays from above or below will also introduce sampling error. Collecting duplicate samples at individual sample sites and collecting material from several sites will help quantify this source of error.
Back at the lab, when natural material arrives, separating the specific component to be dated from a geologic sample can also introduce sampling error. There is the potential for contamination from previous samples run through the equipment, atmospheric contamination, and many other sources.
And, the fundamental assumption that all radiometric dating is predicated on is that the sample has behaved as a closed geochemical system since the radioisotope was deposited. Geochemical fractionation after deposition usually results in younger age determinations than the real age of the site or sample.
If you follow the literature on dating, you will quickly see that several things have changed with time. First, many important places have been dated repeatedly and often by different methods. This leads of course to discussions in the literature and hopefully resolution and ever more accurate age determinations. Second, the parameters associated with known standards change gradually as more measurements are made with ever more sophisticated instruments.
At the end of the day, dating requires precise instrumentation and sound field practices. Often, subjective determinations in the field play a big role in the measurement process itself.
So, I haven't really answered your question because that is not really possible without writing a book. If you are deeply interested in this subject, I suggest you take a class or perhaps even align yourself with a professor who is actively dating geologic materials, work in their lab, go into the field, and get a real feel for how this is all done.
