European Commission logo
français français
CORDIS - Résultats de la recherche de l’UE
CORDIS

Monitoring Earth Evolution Through Time

Periodic Reporting for period 2 - MEET (Monitoring Earth Evolution Through Time)

Période du rapport: 2022-05-01 au 2023-10-31

MEET stands for Monitoring Earth Evolution through Time.

The main goal of the MEET project is to investigate the Earth’s evolution since its creation over 4.5 billion years ago. The three principal investigators are Alexander Sobolev (Université Grenoble Alpes, France), Stephen Sobolev (GFZ Potsdam, Germany), and John Valley (University of Wisconsin-Madison, USA).

This project investigates two main questions: How has Earth’s chemical composition evolved over time? And what physical processes are responsible for these changes?

Previous attempts to understand the early Earth have been stymied because rocks that are archives from this time are either destroyed or altered so that the original chemical information is gone. However, there is a unique possibility to retrieve the chemical tracers most sensitive to changes of Earth’s mantle and crust. This information is preserved in the composition of melt inclusions in crystals of minerals olivine and zircon (Figure 1). These are tiny drops of melt that were trapped when the mineral crystallized. They typically measure less than 20 microns and weigh just a few nanograms.

Unique microanalytical equipment for in-situ chemical analysis of such inclusions has been installed in the Institut des Sciences de la Terre (ISTerre), Université Grenoble Alpes (UGA) (Figure 2) and, in conjunction with other in situ techniques such as Secondary Ion Mass Spectrometry (SIMS) at UW-Madison, deliver new information on the recycling of chemical elements in the Earth and on the formation of its crust from about 4.4 billion years ago to present day. The evolution of Earth has profound implications for questions in other disciplines, such as the origin of life and the conditions on exoplanets.
The establishment of a new microanalytical platform in ISTerre (Figure 2) we consider as one of the major achievements in the first two years of project. Using this new equipment, we developed a number of new in-situ analytical methods to analyze elemental and isotopic compositions of melt inclusions and host minerals. This allowed us to discover the most depleted in radiogenic 87Sr melt from Earth in a subset of melt inclusions in olivine from 3.3 billion years old mantle-derived melts (komatiites from Barberton Mountain land, South Africa, Figure 1). These melt inclusions correspond to a middle Hadean age (4.2 billion years) mantle source with compositional evidence of significant continental crust production. 

We performed two highly successful expeditions to remote areas of South Africa and Vietnam to collect samples of komatiites of different ages.  
A new model has been developed for Ni-Mg-partitioning between olivine and rich in alkalis melt based on new one-atmosphere experiments. 

We have experimented with annealing of devitrified melt inclusions in zircon. The temperature, pressure and duration of heating varied in 11 different experiments (900 - 1200 C, 0.1 MPa - 0.5 GPa, 2-24 hr). Zircons were selected from previously studied suites that vary in age from 4.4 Ga to 20 ka, including Hadean samples from the Jack Hills, W. Australia and Barberton, S. Africa. Selected glass inclusions were analyzed by laser-Raman to determine the water content of glass and radiation damage in zircon; electron microprobe to determine the chemical composition of glass; and SIMS to determine oxygen isotope ratios of glass and host zircon, the water content of the glass, and age and trace elements in host zircons. 

We developed new protocols for the analysis of water and of key trace elements such as Nb, Sc and Ta by SIMS.

We reported U–Pb/Hf/O data for zircons from Archean Saglek Block, North Atlantic craton, suggesting that a depleted mantle domain existed in the early Archean and crustal reworking in the Archean.

We develop numerical techniques and run global models of Earth’s evolution spanning its entire age. Our current key finding is that the likely dynamic regime in the Hadean and Eo-Archean times was oscillatory, with interchanging of long-lasted subduction-dominated regimes and shorter non-subduction regimes, rather than previously suggested continuous non-subduction stagnant-lid or squishy-lid regimes.
The rate of formation of the continental crust and the time of the onset of subduction and recycling of the lithosphere is an unsolved problem in the history of the Earth of paramount importance.

Our collaborations with Profs. Nadja Drabon and Annie Bauer have contributed to understanding an important geochemical change 3.7 billion years ago (3.7 Ga, Figures 3). The step-changes seen at ca 3.7 Ga in εHfT, U/Nb, and Sc/Yb correlate to εHfT patterns in nine other Archean terranes and indicate an influx of juvenile magmas into Hadean protocrust. These events were previously proposed to mark the unidirectional transition from "stagnant-lid" tectonics to subduction. However, Hf model ages in Eo-Archean zircons, coupled with extremely low initial amounts of radiogenic 87Sr in olivine-hosted melt inclusions from Barberton komatiites with fractionated Ce/Pb, Nb/U and Nb/Th suggest active subduction and massive production of continental crust in Hadean Eon. New geodynamic modeling reconсiles these apparently contradicting geochemical observations.

The synergy of the new geochemical data and new modeling suggests that the intensity of subduction in Hadean and early Archean was oscillatory rather than unidirectional. After an initial period of active subduction and crust formation at ca 4.4-4.1 Ga, there was a period of subdued subduction, the reworking of crust, and little magma from the mantle from ca 4.1-3.8 Ga. The step changes seen at 3.7 Ga mark the resurgence, rather than onset, of subduction with mantle-derived magma and continental crust. This result could be a real breakthrough in understanding early Earth's evolution if substantiated by further studies. We will continue to test this model as part of the ongoing MEET Project.
Archean komatiites of South Africa, and homogenized melt inclusions in olivine and zircon
https://www.isterre.fr/english/platforms-services-1235/isterre-microanalytical-platform-imap/
Synthesis of obtained data and modelling for early Earth evolution.