Periodic Reporting for period 2 - MIMATOM (Paleomagnetism and rock-magnetism by Micro-Magnetic Tomography)
Okres sprawozdawczy: 2021-08-01 do 2023-01-31
The Earth’s magnetic field protects our planet against harmful electromagnetically charged particles from the Sun. Without the protection of the Earth’s magnetic field our atmosphere would be stripped away slowly and live on our would become impossible. These electromagnetically charged particles also pose a threat to our technological advances. The interaction between them and the Earth’s magnetic field may lead to induction in long telecommunication cables and may interfere with wireless communication. Also, many satellites are vulnerable to excess amounts of electromagnetically charged particles that interfere with their electronic circuits on board.
The Earth’s magnetic field is generated in the liquid outer core of our planet. Because it is generated in a liquid, the Earth’s magnetic field can vary quickly -even on human timescales- in both strength and direction. To understand and possibly even predict the behavior of the Earth’s magnetic field, it is paramount to reconstruct its past behavior. The Earth’s magnetic field is recorded in volcanic rocks when they cool and they can retain that information over millions of years. Active volcanic regions with well-dated lava flows are therefore an invaluable archive of the past behavior of the Earth’s magnetic field. Volcanic rocks, however, consist of many different minerals, some have magnetic properties and can act as a recorder of the Earth’s magnetic field, while others cannot. And even within the group of magnetic minerals, factors like size, shape and chemical composition influence the way the information on the past state of the Earth’s magnetic field is recorded and stored. When measuring samples from volcanic rocks, all these minerals are measured together. The presence of minerals with different recording properties hamper obtaining reliable magnetic information from such rocks. In experiments that aim to determine the past strength of the Earth’s magnetic field usually 80% of the data is unreliable.
With MIMATOM we develop and establish a new way of obtaining paleomagnetic data from volcanic rocks. By combining a magnetic surface scanning technique with a micrometer resolution with microCT data, we are able to calculate the magnetic moments of individual magnetic grains in the sample. This enables us to only consider and interpret magnetic signals of reliable grains in a sample, while disregarding the contributions of adverse-behaved magnetic grains.
The Earth’s magnetic field is generated in the liquid outer core of our planet. Because it is generated in a liquid, the Earth’s magnetic field can vary quickly -even on human timescales- in both strength and direction. To understand and possibly even predict the behavior of the Earth’s magnetic field, it is paramount to reconstruct its past behavior. The Earth’s magnetic field is recorded in volcanic rocks when they cool and they can retain that information over millions of years. Active volcanic regions with well-dated lava flows are therefore an invaluable archive of the past behavior of the Earth’s magnetic field. Volcanic rocks, however, consist of many different minerals, some have magnetic properties and can act as a recorder of the Earth’s magnetic field, while others cannot. And even within the group of magnetic minerals, factors like size, shape and chemical composition influence the way the information on the past state of the Earth’s magnetic field is recorded and stored. When measuring samples from volcanic rocks, all these minerals are measured together. The presence of minerals with different recording properties hamper obtaining reliable magnetic information from such rocks. In experiments that aim to determine the past strength of the Earth’s magnetic field usually 80% of the data is unreliable.
With MIMATOM we develop and establish a new way of obtaining paleomagnetic data from volcanic rocks. By combining a magnetic surface scanning technique with a micrometer resolution with microCT data, we are able to calculate the magnetic moments of individual magnetic grains in the sample. This enables us to only consider and interpret magnetic signals of reliable grains in a sample, while disregarding the contributions of adverse-behaved magnetic grains.
To establish our new method to obtain paleomagnetic information from rocks (from now on referred to as Micromagnetic Tomography or MMT), we had to overcome five main challenges: (1) install a Quantum Diamond Microscope; (2) increase the resolution of the MicroCT scans; (3) ease the mapping between the magnetic and MicroCT datasets; (4) assess and improve the accuracy of the calculated magnetic moments; and (5) optimize the computational efficiency to allow processing of natural samples. Then, the method would be usable to answer rock-magnetic and paleomagnetic research questions.
We made significant progress with all five challenges that we identified. The Quantum Diamond Microscope was installed in our laboratory, but with some delay due to the onset of the COVID-19 pandemic a couple of weeks after the start of this project. This microscope has now been fully operational and is producing data for our project – this completes challenge (1). Shortly after the start of this project the EU-funded Excite-network was launched: a pan-European network providing access to (amongst others) high-resolution MicroCT scanners. Last summer we made MicroCT scans with a resolution that was sufficient to identify the magnetic grains of interest in our samples – this accomplishes challenge (2). The mapping between the magnetic and MicroCT datasets is done mainly by hand. We are experimenting with image recognition routines and applying markers to the samples – addressing challenge (3) is work-in-progress. We provided a statistical framework to scrutinize MMT results that allows us to determine which grains produce accurate results and which do not. Challenge (4) is therefore addressed, although experience in the second half of the project will lead to an optimized and refined statistical framework. Upscaling the computational routines necessary for MMT experiments has been achieved: at the start of the project, we could solve the magnetic moment of eight grains in two days computational time, now we can solve for 10,000 grains in a matter of minutes. This completes challenge (5).
MMT is therefore ready to start addressing paleomagnetic and rock-magnetic research questions. This summer, we started working on three datasets that do not produce reliable results with conventional paleomagnetic methods: 3.7 billion year old volcanic dykes from Greenland; 400 million year old pillow lavas from Germany; and 600 million year old dykes from Norway.
We made significant progress with all five challenges that we identified. The Quantum Diamond Microscope was installed in our laboratory, but with some delay due to the onset of the COVID-19 pandemic a couple of weeks after the start of this project. This microscope has now been fully operational and is producing data for our project – this completes challenge (1). Shortly after the start of this project the EU-funded Excite-network was launched: a pan-European network providing access to (amongst others) high-resolution MicroCT scanners. Last summer we made MicroCT scans with a resolution that was sufficient to identify the magnetic grains of interest in our samples – this accomplishes challenge (2). The mapping between the magnetic and MicroCT datasets is done mainly by hand. We are experimenting with image recognition routines and applying markers to the samples – addressing challenge (3) is work-in-progress. We provided a statistical framework to scrutinize MMT results that allows us to determine which grains produce accurate results and which do not. Challenge (4) is therefore addressed, although experience in the second half of the project will lead to an optimized and refined statistical framework. Upscaling the computational routines necessary for MMT experiments has been achieved: at the start of the project, we could solve the magnetic moment of eight grains in two days computational time, now we can solve for 10,000 grains in a matter of minutes. This completes challenge (5).
MMT is therefore ready to start addressing paleomagnetic and rock-magnetic research questions. This summer, we started working on three datasets that do not produce reliable results with conventional paleomagnetic methods: 3.7 billion year old volcanic dykes from Greenland; 400 million year old pillow lavas from Germany; and 600 million year old dykes from Norway.
Our efforts until the end of the project will be in three directions: (1) improving both the hardware and software necessary for MMT experiments; (2) further developing the statistical framework that is necessary to interpret MMT results reliably; and (3) applying MMT to paleomagnetic and rock-magnetic research questions to show the potential of this new technique.
To address direction (1), there are several improvements possible. For example, during the installation of our Quantum Diamond Microscope it became evident that the resolution can be increased while reducing the signal-to-noise ratio with a new generation camera. These optimizations are now being installed and we expect to be able to use our improved set-up by the summer of 2023. The software that does the inversion to determine the magnetic moments of the individual grains in a sample can be further optimized so it utilizes the graphics card in our computational system best. This will be done in the spring of 2023. We expect that the entire workflow to obtain MMT results from volcanic samples will be optimized and finalized by the end the summer 2023.
The statistical tools and theories to interpret MMT results (direction 2) are completely new to paleomagnetism and rock-magnetism. We therefore started a collaboration with the Center for Mathematics and Informatics in Amsterdam, that is home to world-leading experts on statistics for inversion problems. The first manuscript providing a first-order statistical approach to MMT results was published in 2022, but we foresee to refine this framework in the second half of the MIMATOM project.
To illustrate the potential of MMT (direction 3), we selected three paleomagnetic and rock-magnetic research questions. The common ground between them is that the samples potentially hold magnetic information to solve standing questions in paleomagnetism, but traditional methods yield uninterpretable data. We are using these samples for MMT studies to assess whether an interpretation at mineral level may unlock crucial information that is currently inaccessible.
To address direction (1), there are several improvements possible. For example, during the installation of our Quantum Diamond Microscope it became evident that the resolution can be increased while reducing the signal-to-noise ratio with a new generation camera. These optimizations are now being installed and we expect to be able to use our improved set-up by the summer of 2023. The software that does the inversion to determine the magnetic moments of the individual grains in a sample can be further optimized so it utilizes the graphics card in our computational system best. This will be done in the spring of 2023. We expect that the entire workflow to obtain MMT results from volcanic samples will be optimized and finalized by the end the summer 2023.
The statistical tools and theories to interpret MMT results (direction 2) are completely new to paleomagnetism and rock-magnetism. We therefore started a collaboration with the Center for Mathematics and Informatics in Amsterdam, that is home to world-leading experts on statistics for inversion problems. The first manuscript providing a first-order statistical approach to MMT results was published in 2022, but we foresee to refine this framework in the second half of the MIMATOM project.
To illustrate the potential of MMT (direction 3), we selected three paleomagnetic and rock-magnetic research questions. The common ground between them is that the samples potentially hold magnetic information to solve standing questions in paleomagnetism, but traditional methods yield uninterpretable data. We are using these samples for MMT studies to assess whether an interpretation at mineral level may unlock crucial information that is currently inaccessible.