4:00pm  4:20pmKeynote: Geomagnetic Data Assimilation And Modeling Of Core Field Changes
Nicolas Gillet
ISTerre, France
The past decade has seen the advent of geomagnetic data assimilation techniques. These aim at considering together information from both magnetic records (historical, groundbased, satellite...) and from a dynamical model advecting the state of the Earth's outer core. I will review the several avenues considered by our community. Important advances have been recently performed concerning the forward integration of threedimensional geodynamo simulations, that timestep primitive equations (induction, momentum, heat). Run at today's extreme parameters they show Earthlike features (e.g. nonaxisymmetric equatorial westward drift, torsional waves), but nevertheless struggle to produce a magnetic energy as important as it is the case in the core, enhancing dissipation and thus filtering MHD waves possibly important for the interpretation of magnetic observations. Alternative startegies have thus been followed, with promissing (although not yet operational) reduced models involving e.g. the quasigeostrophic assumption or largeeddy simulations. I will also show that whatever the employed model, it is mandatory to consider the unmodelled physics (through e.g. stochastic representation of the unresolved quantities) in order to obtain an unbiased estimate of the core dynamics.
4:20pm  4:35pmAn Accelerating Highlatitude Jet in Earth's Core
Phil Livermore^{1}, Chris Finlay^{2}, Rainer Hollerbach^{3}
^{1}School of Earth and Environment, University of Leeds, United Kingdom; ^{2}DTU Space, Technical University of Denmark, 2800 Kgs. Lyngby, Copenhagen, Denmark; ^{3}School of Mathematics, University of Leeds, United Kingdom
Observations of the change in Earth's magnetic field, the secular variation, provide information on the motion of liquid metal within the core that is responsible for its generation.
The very latest highresolution observations from ESA's Swarm satellite mission show intense field change at highlatitude localised in a distinctive circular daisychain configuration centred on the north geographic pole.
Here we explain this feature with a localised, nonaxisymmetric, westwards jet of 420 km width on the tangent cylinder, the cylinder of fluid within the core that is aligned with the rotation axis and tangent to the solid inner core. We find that the jet has increased in magnitude by a factor of three over the period 20002016 to about 40 km/yr, and is now much stronger than typical largescale flows inferred for the core.
The current accelerating phase may be a part of a longer term fluctuation of the jet causing both eastwards and westwards movement of magnetic features over historical periods, and may contribute to recent changes in torsional wave activity and the rotation direction of the inner core.
4:35pm  4:50pmModelling of Geomagnetic Secular Variation with Swarm: Past, Present and Future
William Brown
British Geological Survey, United Kingdom
The magnetic field generated by the motion in Earth’s fluid outer core is by far the largest contribution to the geomagnetic field. The shape and intensity of this field changes through time (known as secular variation), occasionally in unpredictable ways. We observe this field evolution with missions such as the Swarm constellation. From such measurements, models of the geomagnetic field can be built to study the temporal and spatial variations, from the core’s surface to satellite altitudes.
We present results derived from the latest iteration of the BGS Model of the Earth’s Magnetic Environment (MEME), updated with the latest Swarm and ground observatory data from 2017 as well as data from previous satellite missions CHAMP and Ørsted. Given that recent secular variation has been significant in some regions, with rapid variations known as geomagnetic jerks observed in 2014 and 2015, we assess how well these changes are captured by this model, particularly when in close proximity to the end of the data span. We also look ahead to the state of the geomagnetic field in the near future as predicted by extrapolation of MEME and provide an outlook with respect to the possible future orbit evolutions of the Swarm satellites.
4:50pm  5:05pmUltra Low Viscosity Geodynamo Models With Scale Separation
Andrew Jackson^{1}, Andrey Sheyko^{1}, Christopher Finlay^{2}
^{1}ETH Zurich, Switzerland; ^{2}Danish Technical University, Denmark
The mechanism by which the Earth’s magnetic field is generated is thought to be thermal convection in the metallic liquid iron core. Energy is converted into magnetic fields by motional induction, which creates electric currents from the convection and thus creates magnetic fields. Computational considera tions previously restricted most numerical simulations to a regime where the diffusivities of momentum and electric current are roughly equal, leading to similar spectra for both velocities and magnetic fields. Here we present results of spherical shell computations where, in some cases, there is a twentyfold difference in the aforementioned diffusivities, leading to significant scale separation between magnetic and velocity fields, the latter being dominated by small scales. When the magnetic diffusivity is larger than the momentum diffusivity by a large factor (a regime rarely simulated in a spherical dynamo), this leads to a likelihood that selfexciting dynamos will die; such dynamos are, however, possible when the Ekman number is similarly reduced to values lower than previously used, of O(10^{}^{7}). Our dynamos dissipate energy primarily through Ohmic dissipation and we show how this scales with magnetic energy. This permits a new estimate of the Ohmic dissipation in the core of 25TW.
5:05pm  5:20pmApplication of Swarm Measurements to Data Assimilation Studies of Core Dynamics
Christopher Finlay^{1}, Olivier Barrois^{2}, Magnus Hammer^{1}, Nicolas Gillet^{2}
^{1}DTU Space, Denmark; ^{2}ISTerre, Université Grenoble 1, CNRS 1381, France
Time variations of the coregenerated magnetic field can be monitored, on timescales of months and longer, using robust mean estimates of the vector magnetic field on a global grid of reference locations. Here, we present results from this approach, known in the literature as "Virtual Observatories" (VO) (Mandea and Olsen, 2006; Olsen and Mandea, 2007), as applied to data from Swarm and CHAMP, and taking advantage of alongtrack and acrosstrack field differences. Comparisons with ground observatories and the CHAOS6 field model will be used to illustrate the quality of the secular variation point estimates.
The next generation of models of core dynamics will be based on data assimilation techniques, that is the combination of magnetic observations with physicsbased models of core MHD. A serious obstacle to this goal is presently the lack of suitable observationbased data covariance information  this is essential in order to optimally adjust the model to fit the observations. Preliminary attempts at data assimilation have been based primarily on sphericalharmonic field models but these typically have no (or very limited) covariance information due to the difficulty of estimating covariance properties for the very large number of instantaneous satellite data. Point estimates of secular variation from Swarm data provide a way round this problem: deriving observationbased covariances for a global grid of say 250 locations, and considering mean values over a month or longer rather instantaneous measurements, is feasible. As an example, we shall briefly discuss our efforts to assimilate Swarm data into a model of core dynamics based on geodynamo simulation statistics.
5:20pm  5:35pmOn Zonal Flows and Axial Dipole Field Changes
Mathieu Dumberry^{1}, Nathanael Schaeffer^{2}
^{1}University of Alberta, Canada; ^{2}ISTERRE, Université Grenoble Alpes, France
First noted a few decades ago, there is a good temporal correlation between the changes of the axial dipole magnetic field and changes in the length of day (LOD) over the past 100 years. LOD changes are carried by zonal (axially symmetric) azimuthal flows which, by themselves, should not produce changes in the axially symmetric part of the magnetic field, including the axial dipole part. As shown by core flow models, changes in the axial dipole can be accounted for by the globally integrated effect of the nonlinear interactions between local flow eddies and magnetic field. The correlated changes in LOD and dipole field may then simply reflect that they are both the product of a common underlying dynamical system. An alternative view is that there is a more direct connection between the zonal flows and the dipole. Such a view has been suggested recently, in which zonal flows are a manifestation of MAC waves in a stratified layer at the top of the core and are connected to a northsouth axially symmetric flow. The latter is then responsible for the observed dipole field changes. In this scenario, the zonal flows deduced from the secular variation only reflect those at the top of the core. Yet the good fit between observed and predicted LOD changes is based on rigid flows, extending deep inside the core and having little variations in the direction of rotation. Here, we explore the possibility that rigid zonal flows  generated by the convective dynamics  can generate a similar northsouth flow at the top of the core through a boundary layer effect.
