Rocks, Zircon Zapping and a Big Shrimp

John Goodge Weathering of Ferrar dolerite at Mount Sirius.

What are our scientific findings? Great question, but at this point we don’t have definite conclusions. That’s because our fieldwork this season is mainly geared toward collecting rock samples out of the glacial moraines for lab analysis later on. Other than noting the general variety of rock types transported by glacial process, it’s difficult to interpret their specific origins. One granite looks pretty much like another, even if they are of completely different ages and origins. In this post I try to give a sense of the lengthy lab process yet to come, but for now this is mostly a collecting trip. Some of our earlier results from a pilot study a few years ago, mentioned here and in my first post, offer a taste of the kinds of things we expect to find.
If it’s possible to personify the different moraines, probably our most obvious finding is that they are quite fickle. Some contain only local rock types, and in several places this consists of Beacon Supergroup sedimentary rocks and the Jurassic Ferrar sills that intrude them. These geologic units overlie the craton we want to study and so don’t tell us very much. The resident experts on these rocks, David Elliot and John Isbell, think that the Beacon strata do not extend too far toward the plateau from the Transantarctic Mountains, so it is reasonable to predict that the glaciers might do a good job eroding the crystalline basement beneath. Disappointingly, this is not always the case. Near faster ice streams, like Byrd Glacier, we found a rich assortment of igneous and metamorphic rocks, probably eroded from the upstream craton, even though the local nunatak geology exposes only Beacon and Ferrar. But in other cases, we’ve come up essentially empty-handed. This is likely to be a combination of the thick sediment cover as well as the fact that the ice coming into the mountains in some places does not have as large a catchment area behind it, in which case the ice is not moving as quickly and cannot erode as effectively. So, not surprisingly, sometimes we have been successful in finding basement rock types, and other times not at all.

John Goodge Locally derived moraine scoured from outcrop of Beacon strata with Ferrar sills, Dominion Range.

What’s next? Now that we have collected our rock samples, our research has just begun. We have learned some important things from our fieldwork, but the real answers to our questions won’t come until after lab analysis on the samples. In addition to field support provided by the United States Antarctic Program, our continuing lab work will be supported by a grant to Jeff and me from the National Science Foundation. From McMurdo, our samples will go by ship to California, and will then be hauled as truck freight to my home campus in Duluth, Minn. From there, it will be at least a year before we know the first results that will tell us something new. The process begins with unpacking our samples, laying them out to sort, photograph and catalog. Next we cut them all with a rock saw — like a mason’s saw, a rock saw has a circular steel blade with diamonds embedded in the cutting surface to slice through rock samples. We then make slides of the samples for microscopic study. When sliced very thin, to approximately 30 microns (a micron is one-thousandth of a millimeter), most minerals and rocks are translucent and can be observed with a polarizing microscope. We will study these “thin sections” to identify minerals and textures in the rocks that tell us something of their origin, and we will also use light microscopy and a scanning electron microscope to search for the mineral zircon.
Zircons hold the key to this project. Zircon is a zirconium silicate mineral that is very hard, physically durable and chemically resistant to modification. It also incorporates atoms of uranium, lutetium, titanium and other rare earth elements in its crystal structure during growth. This is quite valuable, because isotopes of uranium (so-called parent) naturally decay by radioactive process to isotopes of lead (the daughter), and the balance of uranium to lead isotopes is directly related to the age of the zircon by a factor known as a decay constant. In a similar way, lutetium decays to the element hafnium, and the relative abundances of these elements in zircon provide us information about the earlier origin of the rock host (Jeff Vervoort will write more about this soon). Once we find zircons in the samples, we will take the thin sections to Jeff’s lab at Washington State University in Pullman, where we will analyze the isotopes in the zircons with a mass spectrometer equipped with a laser that can ablate the zircons and ionize the elements of interest for U-Pb geochronology. Informally, this is known as “zapping” the zircons. This method provides a quick way to screen the samples by their age, so we can focus on the oldest ones. There are, for example, large masses of Cambrian granite (about 500 million years old) in the Transantarctic Mountains, and we are mainly interested in learning about the history of the East Antarctic continent that precedes formation of the granites. So if we see preliminary isotopic evidence that a sample is older than this age threshold, we will move it on to the next step.

The Shrimp-II in the Research School of Earth Sciences at Australian National University. This is one of four models of ion microprobe built at A.N.U.

Once we have identified a smaller pool of samples not represented in the local outcrop geology, we will then ship them to Canberra, Australia. There, at the Australian National University, is a group of instruments generally referred to as ion probes. These machines focus a beam of ions at a target (in this case zircon) to liberate atoms into the flight path of a mass spectrometer. Ionized atoms from the sample travel through electrostatic filters and a large magnet, which separate the ions by mass and charge; after separation, the ions of different masses are counted by precisely positioned collectors. In this case, the A.N.U. has been building a special class of ion probes since the 1980s called Shrimp, which stands for sensitive high-resolution ion microprobe. The name Shrimp belies their actual size. These instruments cover a floor space of about 13 feet by 20 feet. It is by virtue of their size, the so-called turning radius of the mass spectrometer’s magnet, that they are so powerful in dating zircons of only a few tens of microns in size by the U-Pb method. In other words, analyzing something very small sometimes takes something very big. Mark Fanning is a leading expert with Shrimp and will lead up the U-Pb geochronology.
To work on zircons at the A.N.U. facility, we will crush the rocks, separate the zircon crystals, mount them in epoxy disks and polish them so that cross sections of their interiors are revealed. In many cases, zircons grow by an accretionary process akin to the formation of hailstones — the core of the hailstone grows first, followed by successive outer layers. Zircons can form in the same way, but each successive growth phase may be separated by millions to even billions of years. Of course, hailstones grow by condensation of water, but zircons most commonly grow by crystallization from a silicate melt at temperatures in excess of 700 degrees Celsius. The ion probe allows us to analyze each part of the zircon separately, giving us a full history of the rock in which it formed.

Created by Digital Micrograph, Gatan Inc. Images of zircon in cross section, as imaged by cathodoluminescence detector on a scanning electron microscope. These zircon crystals are about 100 microns long, and show internal structures related to their growth.

Once we determine the age history of the zircons, and thus their host rocks, we can then measure their ratio of different isotopes of oxygen, which can tell us whether the zircons originated in the earth’s crust or mantle. To bring us full circle, we will next return to Jeff’s lab to measure the hafnium isotope compositions in these zircons. Together, the U-Pb age, oxygen and hafnium isotope data will help us understand when and where within the earth a zircon formed, and what its complete history was preceding its latest stage of formation. All of these steps are quite involved, and in a best-case scenario, we will not have all of the analyses done before late in 2012. We have a great collection of material to work with, but instant gratification is not a characteristic of analytical geochemistry!

John Goodge Outcrop exposure of gneiss from Milan Ridge in the Miller Range, with lens-shaped felsic zones bordered by mafic selvedges, indicative of local melting. The question is, when did melting occur, and is there a record of this event in the glacial clasts?

By combining these data with the rock’s mineralogy and texture, we can write a history for each small piece of crust we have collected from the glacial moraines. With the hundreds of samples we have to work with, we hope to find patterns that we can use to reveal the architecture of the ice-covered craton that we cannot see directly. As part of a pilot study, we found clasts with ages of 1,100 million, 1,460 million, 1,580 million and 1,880 million years, along with a slew of Ross Orogen igneous and metamorphic rocks dating to about 500 million years ago. The first four ages are completely unknown in exposed rocks found over about 2,500 miles of the Transantarctic Mountains, so they represent a completely new chapter in the geologic history of East Antarctica. These data might help determine what kinds of rocks make up the enigmatic Gamburtsev Subglacial Mountains, a hidden mountain range of great interest to the glaciology community where the ice cap first started growing 34 million years ago. They can also provide us with a way to match the ages of hidden rocks in East Antarctica with those known from the now distant continents of Australia and North America.

Field party of Project G-503 at Mount Sirius. From left to right, Dylan Taylor, John Goodge, Jeff Vervoort, Tanya Dreyer and Mark Fanning.

For now, we have a few things to tidy up at CTAM and McMurdo. It’s been a great field season, but as with any lengthy trip away — now nearly seven weeks long — we are all eager to wrap things up and head home. As much as I enjoy being in the field and am happy to live in a tent camp in the snow, spending each night in a sleeping bag with a meager pad on hard snow gets old. Of course, in Antarctica, there is no such thing as a scheduled return. We won’t book seats on the flight home until we land in New Zealand, and just as on our way down here, we can get held up by weather and aircraft availability at any time. So we have no idea when we will actually return home. The more anxious you get, the more likely you are to be delayed. For now, we can enjoy the bright and beautiful day opening the new year, set against a backdrop of the snow-covered buttresses of the Transantarctic Mountains.