July 17, 2004

Kay, John, three students, and I spent the morning in Locality G discussing the geology of the area. Our reason for visiting was to see if we could figure out how the volcanic tuffs relate to the excavation at the Sandwich Site. This site is named for the fact that it's sandwiched between two tuffs. I realize I mentioned tuffs earlier and their importance in dating a site, but didn't get to describe them.

The Sandwich Site.

Tuffs are layers of old volcanic ash. We see them as grey layers in the strata. The volcanic ashes came from eruptions. They were then carried over the landscape by wind and the force of eruption, and in some instances they were carried by water to where we see the layers of ash in the sediments. Sometimes these layers have another kind of material that was ejected out of volcanoes, called pumice, which can also be dated. You can see the two tuffs at the Sandwich site in the photograph: they are the grey sediments above and below the tan layer in the middle.

How does the dating work? Some of the chemical elements found in the ash and pumice are radioactive. That is, they're unstable and these elements change (decay) from one chemical form (or isotope) to another. They do this in a regular manner that can be measured - the rate of decay from the radioactive isotope to its stable (daughter) product. Over a long period of time, the amount of radioactive decay from one chemical isotope to its daughter product provides a measure of time.

Let me tell you a little more. In the molten rocks (magma) beneath the surface of the rift valley, there's a lot of radioactive potassium-40 (written 40K), which decays into a gas called argon-40 (40Ar). Although 40Ar doesn't decay any further (it's stable), the gas trapped in the molten rock is all driven off during the incredible heat of a volcanic eruption. The radioactive potassium within the volcanic crystals continues to decay. As a result, new argon gas is trapped in each crystal as the volcanic particles cool down and fall to the earth. Two other things you should know: Number 1: The amount of argon gas can be measured really precisely. Number 2: The element argon also has another isotope, called argon-39 (39Ar), and its amount in a rock is directly related to the amount of radioactive potassium.

Hmm, so that means that you can calculate very precisely both the stable (or daughter) element and the original radioactive element by zapping the volcanic crystal with a laser and measuring the ratio of argon-40 to argon-39. Because the rate of decay is known, that ratio tells you how much time has passed since the volcanic eruption.

Are you ready for the name of this dating technique? It's called single-crystal, laser-fusion 40Ar/39Ar dating. Sorry for the complicated name, but the technique does give us nice, precise age estimates of the many volcanic layers at Olorgesailie.

With any scientific measurement and calculation, it's important to know about the margin of error - and the amount of error can also be calculated. For 40Ar/39Ar dating of the Olorgesailie tuffs, the error margin is usually from 1,000 to several thousand years. That may sound like a lot, but not really when you're dealing with hundreds of thousands of years. For example, the 40Ar/39Ar date for the pumice of Member 10 of the Olorgesailie Formation (where we have recovered fossil animal bones and some stone artifacts) is 662,000 plus-or-minus 4,000 years. This means that the age of the pumice (and the layer where we find it) is between 658,000 and 666,000 years old (subtracting and adding the error margin to the average date).

The presence of a tuff at an archeological site is a great benefit. That's because we can measure about how old the site is. With one tuff above and another below, the Sandwich Site is a special site. When my colleague Dr. Alan Deino joins us starting next week, I'll ask him to sample the pumice and ash from these two tuffs, and he'll do the hard work to find out the age of the site.