Federica Restelli
The flow in the mantle has a crucial impact on the evolution of our planet: it shapes the surface of the Earth, drives plate tectonics controlling the distribution of earthquakes and volcanoes, regulates the internal temperature and interacts with the fluid outer core. However, direct data of the Earth become more and more sparse with increasing depth, so the study of the deep Earth requires indirect methods. Seismic tomography is essential for imaging the Earth’s interior and to better understand the dynamic processes at work. Tomography models represent the perturbations in seismic wave velocity due to variations in physical properties within the Earth, such as density, temperature and elasticity. Tomographic images of the Earth's mantle show the presence of three-dimensional seismic structures, i.e. portions of the mantle where seismic waves travel faster or slower than a reference model. However, the link between velocity and physical properties is not straightforward and tomography models are often not consistent, so the chemical and mechanical features of these structures are still discussed. My project aims to build a new tomography model using a newly developed method (called the "SOLA method") and to interpret the results in terms of mantle flow and physical properties with the help of geodynamic models and mineral physics.
2. Seismic tomography is a powerful way for imaging the interior of the Earth: it exploits the waves produced by the earthquakes to create models - called tomography models - which represent the perturbations in seismic wave velocity. These perturbations are due to variations in physical properties within the Earth, such as density, temperature and elasticity, so seismic wave velocity can provide insights into the internal structure of the Earth. However, tomography models are not consistent and show discrepancies among them. For this reason, there are still a lot of open questions. The study of mantle flow can provide invaluable constraints useful to determine more robustly the Earth’s physical properties.
3. Seismic anisotropy expresses the dependence of seismic wave velocity upon direction. In the deep mantle it is caused by alignment of intrinsically anisotropic minerals due to mantle convection, so seismic anisotropy is the key to infer the trajectories of mantle flow.
4. A robust physical interpretation of tomographic images requires the model to have unbiased amplitudes and to be accompanied by uncertainties. Commonly-used techniques, such as damped least-square inversions, cause amplitudes to be biased and uncertainties are usually not computed. The aim of my project is to build a new tomography model using a newly developed method – called the SOLA method (Zaroli, 2016; Zaroli et al., 2017) to overcome these issues. I will focus on seismic anisotropy to interpret the results in terms of mantle flow.
5. The database of our model will be made up of standing waves. After very large earthquakes the Earth starts oscillating like a ringing bell for weeks, and these oscillations are called standing waves. Each one possesses a particular resonance frequency. The heterogeneities inside the Earth cause standing waves’ frequencies to deviate from the reference one, and this effect is called splitting. Splitting can be measured and used as input to build a tomography model.
6. Our model will be built using the SOLA (Subtractive Optimally Localizes Averages) method. With this method, amplitudes are constrained to be unbiased and uncertainties are always computed. Moreover, it is computationally efficient and little influenced by uneven data coverage.
7. Once the tomography model is built, I will interpret the results in terms of mantle flow and eventually physical properties with the help of geodynamic modelling and mineral physics.