The search for records of Earth's earliest crust to test terrestrial planet early global differentiation models.

点击翻译按钮获取中文摘要

The present-day Earth is known to be composed of a dense metallic core surrounded by a partially molten mantle upon which a buoyant outer crust lies. In order for this separation to have occurred, at some point in time our planet must have physically differentiated. We also know that differentiation occurred on other terrestrial planets. Critical to enhancing our understanding of terrestrial planetary evolution is the timing of differentiation. Evidence from meteorites strongly suggests that other terrestrial planets differentiated into a core, mantle and crust within 30 million years of the formation of the Solar System, and there is some evidence to suggest that that was equally true on Earth. However, whereas very old crustal rocks appear to be present on the Earth's Moon and possibly Mars, the oldest crustal rocks exposed on Earth are much younger. The Solar System is around 4570 million years old, whereas the oldest crustal rocks found on Earth are around 4030 million years old. This raises the question as to whether there is any evidence locked within Earth's crustal rocks to suggest that they formed as a result of reworking of much older pre-existing crust, or whether Earth's crust actually formed in a more piecemeal fashion over an elongated timescale. Answering this question is the main focus of this proposed research so that Earth's evolution can be more confidently compared with that of other terrestrial planets. To do this, we can use isotopic techniques to monitor the extraction of crust from the mantle in several ways. The isotopic composition of the element Neodymium (Nd) can determine whether very old crust was formed from a mantle that had already undergone depletion due to differentiation and Calcium (Ca) isotopes can determine whether early crust was available to contaminate later formed crust. In addition, isotopes of Oxygen (O) can further determine the nature of these pre-existing crustal rocks. Also, dating by Uranium to Lead (U-Pb) radioactive decay methods is required to interpret the isotope data. Part of the problem is identifying a very old sample that has not been subjected to later isotopic resetting. Single minerals within rocks may act as a repository for important isotopic information if they are robust enough to withstand later reworking. The mineral zircon has been used previously and through this, evidence of the oldest crustal rocks on Earth has been extended back to 4400 million years ago. However, zircon does not contain abundant Nd and Ca. Therefore, we propose to use the mineral titanite, which can be dated by U-Pb methods and contains abundant Nd, Ca and O. Titanite is not quite as robust as zircon, but one way of circumventing this problem is analysing small areas within single crystals. Since individual grains of titanite can contain distinct cores that retain important isotopic information even during later reworking of the crust, we can therefore use a novel approach to extract this information. We propose to work on samples from West Greenland where we may find evidence within titanite of the very oldest remnants of Earth's crust. Isotopic evidence from these rocks will confirm whether or not the Earth did differentiate very early. We then propose to work on an array of rocks from around the world that are around 3000 to 2500 million years old. These rocks should contain a distinctive Ca isotopic signature if an early crust on Earth did form and survived and the sample coverage will allow us to determine the extent of this early crust. In combination with Nd isotopes within these rocks, it will also be possible to determine whether new crust was generated throughout this important time period. In this way, it should be possible for the first time to build up an accurate picture of when Earth's crust was formed, whether the early crust survived in large volume and whether later formed crust significantly contributed to the overall growth of Earth's crust.