Termite mound 

Discovering
Early
Earth
Project

Lake stromatolites 

Advisor: John Valley 

Research Area: Jack Hills Metasedimentary Belt, Western Australia 

Introduction 

Geologists are essentially historians that study the evolution of the Earth through through time by examining the preserved rock record. In some ways we aren't that different from anthropologists, who study the evolution of humans through examining preserved skeletal remains (also in the rock record). Where we differ somewhat is in the concept of time. Anthropologists deal with the time period of Earth's history during which humans have existed on the planet. This includes their earliest appearance, all stages of development and evolution up to today. At great risk of generalizing, this time span encompasses at most a few million years back in time. This 'anthropological time' can be further subdivided into 'historic' or 'prehistoric' time, depending on whether or not records were kept.

Geologists are concerned with the appearance and evolutionary development of the Earth, from it's inception in a swirling cloud of cosmic dust up until today, and as such we approach time somewhat differently. We (geologists) frequently refer to the concept of 'deep time', when describing older portions of Earth's history. Most geologists agree that the best estimate for the age of the Earth is about 4.56 billion years old. Yes billion- this means 4,560,000,000 years old. Numbers this large aren't frequently used in everyday conversation, unless you are either talking about Bill Gates' wealth (several billions, perhaps 4.56?), or the number of stars in the universe (remember Carl Sagan's famous quote: "billions and billions"?, or even the dollar amount our government spends on defending our national boders. Numbers in 'the billions' can and are represented in a variety of ways, for example Earth's age is 4.56 billion years old (b.y.o.), or 4.56 x 10^(9), b.y.o., or 4.56 Ga (Giga anna, or billion years in Latin). It is when considering time periods in the millions and billions of years that geologists refer to 'deep time'.

"So are there 4.56 billion year old rocks that I can go look at ?" Great question, but the answer, unfortunately, is "No" (or shall I say, none have yet been found). Due to the fact that our planet is essentially a continually moving 'Rubic's Cube', whereby plate tectonics recycle the floors of our oceans and sometimes even our land masses deep into the interior of the Earth, the oldest part of the Earth identified so far are not rocks, but are actually ancient grains of sand. 'Sand' acutally refers to the size of these grains, and does not mean that they are grains of quartz (SiO2) like you would find on a beach. The ancient 'sand-sized' grains are a different mineral, called 'zircon' (ZrSiO4) and have been found in a type of rock called a 'metaconglomerate' from the Jack Hills, in the barren Outback of Western Australia. Conglomerate is a type of sedimentary rock composed of all sizes of detritus, ranging in size from sand grains to boulders. When it becomes metamorphosed it is termed a 'metaconglomerate'. 

Why all the fuss about zircons? 

Zircon is a mineral that is particularly useful to geologists for dating rocks for several reasons: 
(1) Zircon is a common accessory mineral in igneous and metamorphic rocks. As these rocks erode, the zircons they contain end up in sedimentary rocks. So they are found in nearly all rock types on Earth. (2) Zircon is very hard (~7.0-7.5 on a scale of 1-10), meaning that it can survive intact as rock around it is eroded and transported away. For this reason zircons can survive in sedimentary rocks for a very long time. 

(3) When it forms, zircon incarcerates into it's crystal structure several radioactive 'parent' nuclides 

(such as ²³⁸U, ²³⁵U, and ²³²Th)

that have long-lived half-lives before eventually decaying into radiogenic 'daughter' isotopes of Pb, and

(4) Zircon incarcerates negligible unradiogenic (common) lead. This is important because a very precise measurement of U/Pb isotopic ratios are required to accurately date zircons (the instrument used to measure the U/Pb ratios is called SHRIMP- (Sensitive High mass Resolution Ion Microprobe). In simple terms, if we can accurately measure the concentrations of U, Th, and Pb, we can determine how long it has taken to produce all of the accumulated radiogenic Pb in the crystal. 

The bottom line is that zircons are essentially the 'Gold Standard' for assigning dates to many rock types and geologic events. 

Getting back to Australia... 

In several locations in Western Australia, sedimentary rocks containing zircons older than 4 billion years have been found. For geologists, this is exciting because until a recent discovery in Canada, there were no known rocks as old as 4 billion years (the Canadian development is exciting, as the rocks appear to be 4.03 billion years old). The single oldest zircon grain found to date is 4.404 +/- 0.008 Ga (billion years old). The sample it came from was collected in the Jack Hills in 1984, and many old zircons we dated from it (the oldest found was 4.276 billion years old). The oldest grain, however, was not dated until 2000, when previously undated zircons separated from the same sample were re-investigated. The exciting find of a 4.404 billion year old grain was subsequently described in the journal Nature (see January issue, Wilde et al., 2001 if interested). I am currently studying this 4.404 Ga grain, and other >4.0 billion year old zircon grains to determine what else we can learn about this period of Earth history from these early ambassadors. The very presence of these old zircons is a record of the existance of rocks far older than any currently exposed on the surface of the Earth. For the time being, this single 4.404 Ga zircon grain is the oldest piece of the Earth known to science.

The ~3.0 Ga sedimentary rocks exposed in the Jack Hills are significantly younger than the ancient >4.0 Ga detrital grains they contain . They are analagous to one of the many nice mountain stream deposits in the Rocky mountains of Colorado. Imagine if you will, that the modern stream bed that is currently collecting boulders, cobbles, pebbles, and sand-sized grains from all of the diverse rocks types in the Rocky mountain drainage basin will eventually be burried and become a rock some day, much younger than the cobbles that it contains. For example, in Rocky Mountain National Park (RMNP) there are stream bed deposits that are currently getting buried and turned into rock. The age of the rock would be 0.0 Ga (i.e. essentially today), but it would contain cobbles of rocks such as granite that occur in RMNP that are almost 2 billion years old! This is the kind of rock that we are studying in the Jack Hills, only is was getting fed by material over 4 billions years old!

I am currently doing various types of geochemical analyses on these detrital zircon grains, as well as other detrital minerals including quartz (SiO₂) and chromite (FeCr₂O₄). The analytical techniques include oxygen isotope analysis of silicates and oxides, EPMA (electron probe micro analysis) of silicates and oxides, XRF (x-ray fluorescence) analysis of whole-rock samples for major, minor and trace elements, and SHRIMP (see above) U/Pb dating of single zircons.

The goal of this work is to learn as much as possible about the source rocks of the ancient detrital grains, given that no equivalent 'rocks' exists. We can then use the inferences about the source rocks to understand geological processes active during the earliest, least understood period of Earth's history of which no rock record exists.

Stromatolites in Shark Bay 

A little larger termite mound... 

The work I am starting for a Ph.D. is a collaborative project with advisor Dr. John Valley at the Universtiy of Wisconsin and Dr. Simon Wilde at Curtin University.