Appendix: Carbon-14 dating of the Newport Tower

Carbon-14 dating of lime mortar is a relatively new method to date buildings where no suitable organic material (construction timber, charcoal in the mortar, etc.) is present. […]

Mortar contains carbon from carbon dioxide absorbed from the atmosphere at the time of construction, which in principle makes it ideal for C-14 dating.

There are, however, well-known risks associated with this dating method. The mortar may contain old limestone, either from incompletely burnt limestone or from carbonaceous sand used as filler, leading to ages that are too old. […] The risk of contamination from unburnt limestone is reduced by mechanical separation. […]

[…] Subsequent to this mechanical separation, an additional chemical separation is performed

The C-14 dating method is based upon the measuring of the concentration of unstable C-14 atoms in the material that should be dated. The concentration may be measured by two methods, the "conventional" one, or the AMS (Accelerated Mass Spectometry) method. The latter, which was applicated in the case of the Newport Tower, demands only 1/1000 of the amount of carbon needed for the conventional method.[…]

The measurement yields a "C-14 age" in "C-14 years BP" (before present = 1950). To be converted into reliable calendar years, the result needs "calibration" (ajustment) by means of a curve established by measurements of the C-14 content of wood samples of known age, basically yearrings from a special group of very old pinetrees (Pinus aristata) in the White Mountains of California. As the C-14 production in the upper atmosphere varied through the ages, the calibration curve has many "wiggles." Therefore, the calibration of C-14 ages may sometimes lead to a dating that falls within a rather broad range of years. In other cases, the dating can be rather exact. Precision-dating also includes a minor correction because of small variations due to chemical processes that may change the original relation of the "normal" C-12 and C-14 in the atmosphere. The size of the variations can be determined by measurement of the content of another carbon isotope, the stable C-13. Datings are normally given within plus/minus one "standard deviation" (±17), corresponding statistically to a 68% confidence level.

Conclusion

Calibration of the mean age of the Newport Tower samples yields a calendar age of AD 1665 calibrated by intercept with the calibration curve as shown on Fig. 2. When the measuring uncertainty of 48 years is taken into account, the calibrated time interval corresponding to ls or 68% confidence level is AD 1651-1679 calibrated (ls) as later intercepts with the calibration curve may be disregarded because of the known minimum age of the tower. The time interval corresponding to 2s or 95% confidence level is AD 1635 - 1698 calibrated (2a).

In summary we estimate that, based purely on the measuring uncertainty in the C-14 determination itself, the real age of the dated mortar has less than about 5% probability of being outside the latter quoted time interval. […] A pre-Columbian date can almost certainly be excluded as the calibration curve rises steeply further back in time.