Hacker News new | past | comments | ask | show | jobs | submit login

The half life of Carbon 14 is 5,730 years. So all it would be able to do is tell you the person is between 0 and 5,730 anyway right?

That wouldn’t help solve these cases.

No, the radioactive decay will follow a (nearly) continuous exponential curve. Half-life is just an arbitrarily chosen point on that curve: the point at which half of the atoms have decayed. With carbon dating you are basically solving this equation: percent of carbon 14 left = 0.5 ^ (years since death/5,730). If it has been 500 years since death, you would expect 94% of the original carbon 14 to remain.

That said, you generally have error boundaries spanning multiple decades and it measures since time of death (when new carbon stops being integrated into the body) not time of birth, so it would not be useful for this.

Note that you can use tooth enamel, which freezes its carbon at time of development (instead of death). https://news.ycombinator.com/item?id=20627335

That said, the error bars are simply too large to do anything reasonable for dates prior to 1955. The difference between 80 years and 110 years is 98.7% vs. 99.0% of the original C14 concentration... and while you may be able to get a very precise measurement of the remaining C14 you also need to very precisely know the baseline from that era to determine the percentage. Thus typical radiocarbon error bars are at least ±60 years.

Thanks for the details.

Applications are open for YC Winter 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact