NAERINGSPH-Nærings-phd

Companies searching for oil and gas collect offshore electromagnetic data by deploying receivers on the seabed and a powerful source in the water. When the source is off or far away from the receiver, the natural background electromagnetic signal that contain MT data is recorded. Exploration companies do not usually use marine MT data because the method is too low-resolution to detect oil and gas reservoir and rather sense deeper in the crust, down to the Earth mantle. In this project, we extract and analyze electromagnetic data from the industry and unleash the scientific potential of MT data on the NCS. The results are so far promising as high-quality MT data can be extracted for ~330 receivers laid on the seafloor of the Barents Sea. A detailed data analysis has shown that we can image the subsurface electrical down to 100 km, exceeding our initial expectations and allowing for a full reconstruction of the Earth brittle crust and the lithosphere across the Barents Sea region. After in-depth data analysis, we explore the model space and find a 3D electrical model of the subsurface that can numerically replicate the observations. This is done with a 3D inversion code using an iterative process that converges to an optimal solution. We use seismic and potential field data to constrain the chosen 3D conductivity model with acoustic and density constraints. The project is a scientific journey to the depth of the Earth, leading to a realistic conductivity structure of the crust beneath the NCS emanating from three independent and complementary geophysical methods. 3D deterministic inversion of MT data is performed along two profiles across the Bjørnøya Basin, a hyper-extended rift basin that failed to reach break-up in Early Cretaceous times. We demonstrate the robustness of two highly conductivity features in our best-fit resistivity models. The first conductor is located in the Early Cretaceous marine sediments deposited in the basin and the second one lies in the uppermost mantle, where the continental crust is the most severely thinned. To optimize the exploration of the solution space, we perform 1D Bayesian inversion for a selection of 21 MT receivers across the Bjørnøya Basin. To handle of the 1D modelling assumption, we trained a machine learning model to recognize and compensate for 2D and 3D patterns in MT data. The analysis of the posterior model distributions reveals a very high probability of finding anomalously high conductivity in the syn-extension, early Aptian - Albian marine shales of the Bjørnøya Basin. Using two-phase rock-fluid models, the predicted salinity of the pore fluid in the marine shales and upper mantle rocks has to exceed seawater salinity to account for the conductivity anomalies in MT models. Finally, we find that residual fluids from 25% serpentinization of the upper mantle could reconcile resistivity, density models, and seismic observations.

This research demonstrates that marine MT is valuable geophysical method to study rifted systems. In the hyper-extented Bjørnøya Basin, our results point towards a partial serpentinization of the upper mantle and migration of saline fluid along active faults during rifting. The ambiguity in the nature of basement rocks from seismic velocity and density is resolved with resistivity models. The new approach and knowledge are applicable to the understanding of numerous hyper-extended rift systems in the North Atlantic. In the context of the transition to low-carbon energies, the conclusions of this thesis have direct implications on the understanding of natural hydrogen formation from serpentinization-driven systems.

The structure and composition of the Earth’s crust is key to understanding the geological evolution of tectonic plates. Magnetotellurics (MT) is a large-scale geophysical method that image the electrical conductivity of the Earth’s crust using the natural variation of the electromagnetic field. For half century, a wealth of electromagnetic data have been acquired onshore for natural resource exploration but marine MT data remain scarce. In this project, we plan to process offshore electromagnetic data acquired by the exploration company in the North Atlantic margin and Barents Sea and extract MT data. Dimensionality analysis of the magnetotelluric impedance tensor should lead to a better understanding of the nature and distribution of telluric currents flowing in the crust. Through inversion and modelling studies, we will seek robust electrical conductivity models of the crust along selected area and transects that can reproduce the MT measurements. We will refine our conductivity model using seismic and potential field data. Using a multi-disciplinary modelling and interpretation approach, we aim at finding a crustal model that honors three independent geophysical measurement. The project will lead to a better understanding of magmatic and hydrothermal processes occurring in the Earth’s crust in a variety of geological settings of the NCS. The study will impact petroleum and mineral exploration by developing new concepts on deep crustal mechanisms and basin evolution.