DI43A-1752
Behavior of Os at High P,T: a Metal Analog of Fe in the Earth's Core
The transition metal osmium (Os) can be considered as a material analog for the high pressure behavior of the primary core metal iron since it is in the same group but has higher electron density. We examined the phase stability, density, and thermoelasticity of Os metal at pressures up to 50 GPa, and temperatures up to 2900 K in the laser-heated diamond anvil cell in conjunction with X-ray diffraction at beamline 12.2.2 at the Advanced Light Source. NaCl was used as a thermal insulator and internal pressure marker in these studies. Our data show that Os is stable in the hexagonal-close-pack phase throughout the study. Preliminary analysis of the thermoelastic parameters of Os suggests that the room pressure linear thermal expansion of the c-axis is ~ 2.9×10-6 (±1.2×10-6) K-1 and the linear thermal expansion of the a-axis is ~ 2.4×10-6 (± 0.9×10-6) K-1. The results suggest that both the linear thermal expansion parameters decrease by 2-3×10-8 K-1 GPa-1. Potential sources of error in the thermoelasticity measurements will be discussed in terms of thermal pressure in the laser-heated diamond anvil cell, pressure calibration with the NaCl standard and temperature measurement. The implications in terms of structure and thermoelasticity of Earth's core will be discussed.
DI43A-1753
In Situ Determination of BCC-, FCC- and HPC-Iron Textures at Simultaneous High- Pressure and –Temperature by Means of the Resistive Heated Radial Diffraction Diamond Anvil Cell (RH-RD-DAC): Implications for the iron core.
Radial diffraction in the diamond anvil cell (DAC) has long been used to determine the stress state of materials under non-hydrostatic compression. This technique is also a major tool to investigate textures and infer deformation mechanisms in the earth mantle and core. However, most of these experiments have been conducted at ambient temperatures and therefore the results of these measurements may be difficult to extrapolate to the deep Earth. Here, we present texture data collected at HPCAT sector 16 BMD of the Advanced Photon Source during the plastic deformation of BCC-, FCC- and HPC-iron at simultaneous high-pressure and temperature in the new Resistive Heated Radial Diffraction Diamond Anvil Cell (RH-RD-DAC). Initial results from Rietveld refinements in MAUD indicate that BCC- iron develops a mixed {100} and {111} texture that remains active during heating. Latter is compatible with previous observations on BCC-iron and interpreted as slip along {110}<111>. Texture obtained after formation of FCC-iron at simultaneous high- pressure and temperatures show a pronounced maximum at {110} with minima at {100} and {111}. This texture is typical for FCC metals in compression with slip on {111}<110>. Processing of the HCP-iron textures at high-pressure and –temperature are under way. We will discuss the implications that the experimental results have for the deformation mechanisms of iron at pressure temperature conditions of the inner core.
DI43A-1754
Elastic Anisotropy of the Earth's Inner Core: Mineral-Physics Evidences on the Elasticity of hcp Metals at High Pressure and Temperature
Certainly, elastic anisotropy is one of the most important evidence of the Earth's inner core. Alignment of iron crystals is generally invoked to explain such a feature, however, the physical mechanisms responsible for the texture development are still unclear and highly debated. Last decades have witnessed several efforts, and even the most recent and accurate seismic analysis and sophisticated models are inconclusive. Ultimately, all the anisotropy models can be seen as series of orientation and rotational average of iron crystal to provide an estimate of the aggregate elasticity that reproduce the seismic observations, and thus, heavily rely on the iron elasticity. Complications arise from the fact that the stable phase of iron and its elastic tensor at core conditions are still unknown. Classically, the hexagonal-closed-packed (hcp) phase is considered in most of the models, and three of the more frequently used sources for the iron elastic moduli are the calculations of Stixrude and Cohen [1995] and Steinle-Neumann et al. [2001], and the early experimental work of Mao et al., [1998]. In this paper, we compare results of inelastic x-ray scattering and x-ray radial diffraction measurements on hcp iron and hcp cobalt with most recent mineral- physics literature, experiments and calculations, to show why all the three above-mentioned sets of values for the elastic moduli of hcp iron should not be used any more. Finally, assuming negligible high temperature effects at constant density, we argue that the elastic anisotropy in the meridian plane of hcp iron at core conditions is best described by a sigmoidal shape, with the longitudinal velocity along the c-axis about 10% higher than in the basal plane. [1] Stixrude, L., and R. E. Cohen (1995), High-pressure elasticity of iron and anisotropy of Earth's inner core, Science, 267, 1972 -- 1975. [2] Steinle-Neumann, G. et al., Elasticity of iron at the temperature of the Earth's inner core, Nature 413 (2001) 57 -- 60. [3] Mao, H. K. et al. (1998), Elasticity and rheology of iron above 220 GPa and the nature of the Earth's inner core, Nature, 396, 741 -- 743. (Correction, Nature, 399, 280, 1999.)
DI43A-1755
Influence of Elastic Deformations Within the Inner Core on the Free Nutations of the Earth
Two of Earth's free modes of nutation, the prograde free core nutation (PFCN) and the inner core wobble (ICW), owe their existence to the presence of the inner core. Their dynamics are controlled by coupling at the inner-core boundary (ICB) and by the torque exerted by the rest of the Earth on a tilted inner core. Previous theoretical estimates of the periods of these two modes were based on models that did not properly take into account elastic deformations taking place within the inner core when the latter is in a tilted position. In this work, I present theoretical estimates of the periods of the ICW and PFCN when such elastic deformations are included. Based on an oceanless, dissipationless, elastic Earth model, elastic deformations within the inner core contribute to a lengthening of the period of the ICW from 2410 to 2713 Solar days, in agreement with recent results obtained by Rochester and Crossley. The period of the PFCN is found to increase by a similar factor, from 476 to 536 days, when given with respect to a space-fixed reference frame. Although this new estimate for the PFCN is still about a factor 2 shorter than its period of ~ 1000 days that is inferred through the resonance of this mode with the forced nutations, the amplitude of the coupling from surface tractions that is commonly invoked to account for the difference is now reduced. If this coupling is interpreted solely in terms of electromagnetic forces, the RMS amplitude of the magnetic field at the ICB is reduced from 7.1~mT to approximately 5~mT.
DI43A-1756
Texture Study of the Uppermost Inner Core from Seismic Coda Waves
Recent studies have confirmed the existence of scattering by a fabric of small-scale heterogeneities in the uppermost inner core. Seismic waves interacting with the uppermost 300 km of the inner core reveal this region to have strong lateral variations in elastic structure, anisotropy, attenuation, and scattering. The detailed spatial distribution (or texture) of the uppermost inner core is important for understanding how the inner core is solidifying from the liquid outer core, since lateral variations in this texture may record variations in the solidification process of the inner core and fluid flow of the outer core at the inner core boundary. Fundamentally different sensitivities of forward- versus back-scattered body waves from regions of heterogeneity enable constraints to be placed on the anisotropy of heterogeneity scale lengths. In the case of the inner core, maps of the lateral variation in the anisotropy of heterogeneity scale lengths can separate regions of growth by active new crystallization perpendicular to the inner core boundary from regions of viscous flow and recrystallization parallel to the inner core boundary. Using Monte-Carlo simulations based on radiative transfer theory, we use the high frequency coda of the P wave reflected from the inner core boundary (PKiKP) and the pulse broadening of P waves transmitted through the inner core (PKIKP) to infer the spatial distribution of 1 to 100 km scale lengths of heterogeneity in the inner core. Lateral variations in the inferred distributions of heterogeneity can be used to identify lateral variations in the solidification processes of the inner core.
DI43A-1757
The frequency dependent anisotropy in attenuation and the structure of the inner core
The anisotropy in attenuation may give important constraints on the texture of the inner core, thus on its growing mechanism. This anisotropy is investigated from the amplitude ratio PKP(DF)/PKP(BC), for paths with various orientations having their turning point beneath Africa, and sampling depths from 200 to 270 km beneath the inner core boundary (ICB). First the anisotropy in attenuation is investigated at period 3s, and compared with the level of heterogeneity, determined from a large number of paths parallel to the equatorial plane. The results confirm the anisotropy in attenuation, with amplitude ratios almost 10 times smaller for polar paths than for equatorial paths at 3s. The influence of frequency on this anisotropy is then investigated in the frequency range 0.2-2 Hz, together with the variation of the DF/BC ratio as a function of frequency. For equatorial paths, the decay of Ln(DF/BC) with frequency is mild and nearly linear, and agrees with the viscoelastic attenuation of the mean Earth model PREM, with Q independent of frequency. For paths with smaller angles with respect to the Earth rotation axis, the decay is much stronger, resulting in a very strong anisotropy at high frequency (DF/BC close to 0.02 at 2 Hz). Moreover, Ln(DF/BC) departs from linearity. For nearly polar paths, an abrupt decay occurs at frequency 0.8-1 Hz, suggesting the presence of scatterers as the dominant mechanism along these paths. The attenuation Q-1 is determined as a function of frequency and compared with models obtained for different textures. The results give constraints on the possible relative contributions of anelasticity and scattering as attenuation mechanisms in the inner core.
DI43A-1758
Timescales of inner core - mantle gravitational locking
Seismic observations of earthquake doublets indicate that the inner core is rotating faster than the mantle. However, hemispheric differences in seismic anisotropy within the inner core have also been observe and likely require a persistent orientation of the inner core with respect to the mantle over geologically long time scales. Gravitational coupling of the non-spherical density distributions provides a mechanism of locking the inner core to the mantle. We consider the timescales over which one would expect gravitational coupling to lock the inner core to the mantle. The locking depends on the strength of the gravitational coupling, the viscosity of the inner core and the geometry of the torque responsible for the differential rotation. We find two separate timescales associated with locking of the inner core and mantle. The first, with a period of approximately 6 years, corresponds to a normal mode of oscillation of the inner core with respect to the (nearly) stationary mantle. The second period, approximately 97 years, corresponds to an "abnormal" mode in which the mantle would oscillate with respect to a stationary inner core. For any forcing with a period longer than 97 years we find that the mantle and inner core will be locked, provided the inner core viscosity is at least on the order of 1017 Pa s. This result may support the idea that the observed inner core super-rotation is a short-term fluctuation in relative angular velocity and that the long-term mean of the velocity is expected to be close to zero.
DI43A-1759
Hemispherical structures in inner core anisotropy
Anisotropy in the inner core has been studied using both free oscillations of the Earth and compressional body waves which sample the core. Some body wave studies have suggested that anisotropy is not uniform, but confined by two lines of longitude to one 'hemisphere' of the inner core. Existing free oscillation models of inner core anisotropic structure, which are formally inconsistent with each other, are symmetrical about the Earth's rotation axis. This constraint occurs because current normal mode models of inner core anisotropy are created using the self-coupling approximation which assumes that the coupling of different inner core modes through Earth structure is negligible. The self-coupling approximation creates selection rules which prevent normal modes responding to the presence of hemispherical structure. We have shown (Irving et al 2008) that this approximation is inaccurate when inner core anisotropy is considered, and that full coupling between normal modes must be considered when studying inner core anisotropy. When full coupling of normal modes is allowed, inner core sensitive modes are permitted to react to the presence of hemispherical structure in the inner core. Here, we investigate the effect of confining inner core anisotropy to just one hemisphere of the inner core on the properties of free oscillations of the Earth. Hemispherical anisotropy is currently only seen in compressional wave velocity. The investigation of PKJKP modes, those normal modes which are sensitive to inner core shear wave velocity and not compressional wave velocity, allows us, for the first time, to investigate the potential effects of hemispherical variation in inner core shear wave anisotropy. By comparing body wave and normal mode data with inner core anisotropy models we hope to find a picture of inner core anisotropy that is consistent with both types of seismic observations. A more detailed understanding of inner core anisotropy is a prerequisite for the integration of results from seismology with mineral physics, heat flow and dynamo modelling. This will help to establish the details and causes of inner core anisotropy and better understand the large scale processes at work in the core.
DI43A-1760
Constraints on grain size and stable iron phases in the uppermost inner core from multiple scattering modeling of seismic velocity and attenuation
We propose a simple model of texture for the uppermost inner core, consistent with seismological and mineralogical data. We model the superficial part of the solid core as an untextured aggregate of iron "patches". Each patch may contain a large number of dendrites with strongly correlated orientation of crystallographic axes and is characterized by the anisotropic properties of individual iron crystal. The orientation of crystallographic axes varies randomly from one patch to another. This simple model accounts for the observed velocity isotropy of short period body waves, and offers a reasonable physical interpretation for the scatterers detected at the top of the inner core. Using a rigorous multiple scattering approach based on the Dyson equation, we show that the Voigt velocity of a single crystal can be up to 15% higher than the seismic velocity of elastic waves in the aggregate. Based on our improved theory, we perform a systematic search for iron models compatible with measured seismic velocities and attenuations. An iron model is characterized by its symmetry (cubic or hexagonal), elastic constants, and patch size. Independent of the crystal symmetry, we infer a most likely size of patch of the order of 400 m. Recent bcc iron models from the literature are in very good agreement with the most probable elastic constants of cubic crystals found in our inversion. Our study (1) suggests that the presence of melt may not be required to explain the low shear wavespeeds in the inner core and (2) supports the recent experimental results on the stability of cubic iron in the inner core, at least in its upper part.
DI43A-1761
Formation of a Cubic Iron-Sulfur Alloy at Megabar Pressures and its Equation of State
The details of binary iron-light element systems at pressures and temperatures relevant to the core can be used to constrain core composition and temperature. The addition of light elements to iron is known to affect the stability field of iron polymorphs. In this study, an iron plus 10 wt. percent sulfur sample was compressed and laser heated at 145 GPa in a diamond anvil cell at the GSECARS beamline of the APS. Phases present in the sample were monitored with x-ray diffraction. At this pressure, hcp-Fe was found to coexist with Fe3S. However, at 158 GPa, upon laser heating a new cubic phase formed at the expense of hcp-Fe until all hcp-Fe was consumed and a single cubic phase was left, apparently indicating solid solution behavior. The strongest x-ray diffraction lines closely resemble the x-ray diffraction pattern of bcc-Fe, however several additional weak lines rule out a structure as simple as bcc. The sample was slowly decompressed in order to measure the pressure-volume relationship. The unit cell volume of the metastable cubic phase began to expand rapidly below 100 GPa during decompression, and was completely amorphous below 80 GPa. Solid solution behavior would suggest that sulfur could be an important component of Earth's inner core; the implications of this, and the possible structure of the cubic phase in relation to the known iron polymorphs, will be discussed.
DI43A-1762
Bridging ab initio theory together with seismological models
Almost all the ab inito studies performed so far for various (possible) Earth's inner core materials were conducted by comparing together the computed longitudinal (Vp) and shear (Vs) wave velocities to the first- order velocity term of seismological models. We here demonstrate that further constraints on the Earth's inner core composition can be provided if first-principles data are also compared with the second-order velocity term. The proposed procedure represent a new way to enforce the agreement between existing seismological models and ab initio calculations and provides a more severe way to guarantee the uniqueness of the achieved results.
DI43A-1763
Fe-Ni-C system at high pressure
Apart from being technologically important, FeNi alloys introduce particularly interest to the material sciences as well as to the geosciences. It is generally accepted that the Earth's core is predominantly composed by FeNi alloy with 10-15 wt% Ni. The certain amount of the light element(s) is also known to be presented in the Core. A number of candidates for the light component have been proposed, including sulphur, oxygen, hydrogen and carbon. In favor of the last one the following arguments can be listed (i) high cosmic abundance, (ii) chemical affinity to iron even at low pressures and (iii) capability of lowering the density of molten iron. Although there is quite a bit of experimental and theoretical results on high pressure high temperature behavior of the system Fe-C, there is still lack of information about the phase relations in Fe-Ni- C system at elevated pressures and temperatures. Therefore we provided a series of compression experiments on the system Fe-Ni-C at pressures up to 53 GPa in temperature range 300 - 2600 K (combining diamond anvil cell and large volume press techniques) in order to investigate phase diagram of Fe-Ni system and the influence of carbon on the phase relations in the system at elevated pressures and temperatures. We observed that dissolution of even 1 wt% carbon in FeNi alloys with 10, 15 and 22 wt% Ni leads to dramatic changes in the system: presence of carbon stabilizes fcc-structured FeNi through the redistribution of nickel. Combining Mössbauer spectroscopy, XRD, TEM and chemical analyses by microprobe and SEM techniques we detected the formation of Ni-poor and Ni-enriched phases, with different elastic and structural properties.
DI43A-1764
Phase Diagram of Wustite at High Pressures and Temperatures
The Earth's core is comprised primarily of iron and nickel, with a small amount of one or more lighter elements present. Oxygen is a strong candidate for inclusion in the core, so it is important to fully understand the behavior of iron oxides, including Fe1-xO (wüstite), at high pressures and temperatures. However, portions of the phase diagram of wüstite remain somewhat poorly constrained. Previously published melting curves disagree significantly (Ringwood and Hibberson, 1990; Knittle and Jeanloz, 1991; Boehler, 1992; Shen et al., 1993; Seagle et al., 2008), and there is apparent disagreement, possibly related to variations in stoichiometry, on the presence of a B1 to iB8 phase transition (Murakami et al., 2004; Seagle et al., 2008). To resolve some of these issues, we investigated the phase diagram of wüstite using single- sided laser heating in the diamond anvil cell to achieve high P-T conditions in an Ar pressure medium. A new technique to obtain a two-dimensional temperature map of the laser-heated spot (Campbell, 2008) allowed the determination of phase boundaries from analysis of temperature and emissivity data, as well as from visual observation of the sample during heating. The measured melting curve is in very close agreement with the results of Boehler (1992) and Ringwood and Hibberson (1990). The subsolidus (B1 to iB8) phase boundary in wüstite, Fe1-xO, has a negative slope of -51 K/GPa, and intersects the melting curve at 45 GPa, 2750 K.
DI43A-1765
Acoustic velocities and thermoelastic properties of FeSi at high P and T
Velocity and density profiles of the Earth's deep interior show a marked discrepancy from pure iron under similar conditions of high pressure and temperature, indicating that the core of the planet must contain some amount of some light element(s). In order to assess which elements these may be, and in what proportions, the physical properties of some iron-light element alloys must be experimentally ascertained under extreme conditions. Among those light elements, Si has been strongly suggested as a possible constituent of the Earth's core, which is primarily supported by current data on bulk sound speed and density derived from static compression; the shear properties on these iron-light element alloys have been rarely investigated at relevant pressure and temperature conditions. Acoustic velocity measurements using ultrasonic interferometry, in conjunction with synchrotron X-radiation techniques at beamline X17B2 of NSLS, Brookhaven National Lab, allow for simultaneous measurements of the bulk and shear elastic properties of these materials at high pressures and temperatures. Using this technique, we have determined the acoustic velocities and elastic properties of ε-FeSi at high pressures and temperatures, yielding the first complete and thermodynamically consistent data set of thermoelastic properties (adiabatic bulk and shear moduli, and their pressure and temperature derivatives) for this material. These new data will be presented and used to draw comparisons to seismic profiles of the Earth's core.
DI43A-1766
High Temperature Theory of Iron and Iron-Nickel from Surface to Core
The structure of iron and its alloys at high pressures and temperatures is still a fundamental question[1,2] and may help explain the seismologically observed inner core structure[3]. Fe has been long thought to be hcp under core conditions[4]. We have performed first-principles DFT linear-response computations for iron and iron-nickel over a wide range of pressure (0-400GPa) and temperature (0-6000K), within density-functional theory in the generalized-gradient approximation using the projector augmented wave (PAW) method with the ABINIT code. We computed the free energies of hcp, bcc and fcc phases for iron and nickel including the thermal excitation of electrons and quasiharmonic phonons. Fe-Ni ordered and quasirandom alloys were studied to obtain the Fe-Ni phase diagram. Under core conditions the free energy difference between fcc and hcp is smaller than the thermal energy, and we find that the inner core phase is likely a mixed phase with random fcc and hcp layers. We find that the stacking fault energy is smaller than one millihartree per atom. At the high temperatures of the inner core, we predict that Fe has a disordered layer structure, and is neither hcp nor fcc. [1] A. S. Mikhaylushkin et al., Physical Review Letters, vol. 90, 165505(2007). [2] L. Dubrovinsky et al., Science, vol. 316, 1880(2007). [3] J. Aubert et al., Nature vol. 454, 758(2008). [4] L. Stixrude and R. E. Cohen, Science, vol. 267, 1972(1995).
DI43A-1767
Modeling the Inner Core Boundary with Antipodal Seismic Waves
Antipodal seismic waves interacting with the inner core boundary (ICB) sample almost the entire surface of the ICB. Hence, models of antipodal observations of PKIKP, PKIIKP, and PKP-C diffracted waveforms can provide an estimate of the global average velocity and density discontinuities and their vertical gradients at the ICB. PKIIKP and PKP-C diffracted phases arriving at the antipode are strongly focused due to constructive interference of their wavefields that arrive from all azimuths. These interference patterns can be used to draw constraints on lateral heterogeneities and sphericity of the ICB. To demonstrate the effectiveness of this approach we synthesize core phases in the distance range 170o-180o and compare them with those observed from an event that occurred in 1997 November 03 of Mw = 6.2 at lat 30.74S and lon 71.22W with a depth of 45km. Seismograms are synthesized in a 2-step process. First we derive the source time function (STF) for the event of interest using observed seismograms in the 30o-90o distance range. Then we convolve this STF with seismograms synthesized using a high-frequency (up to 2 Hz) full wave theory, accurate for frequency dependent wavefields at grazing incidence to the ICB. Modeling experiments include tests of different models of heterogeneity in the uppermost 100 km of the inner core and a test for evidence of lateral heterogeneity extending into the liquid outer core near the ICB.
DI43A-1768
Could K and Rb be in Earth's Core?
K and Rb undergo high-pressure electronic transitions that suggest the possibility for them to be found in Earth's core. Currently, without radioactivity in the core, the age of the Earth's core is 1-2 billion years younger than currently believed based on the understanding of Earth's interior dynamics. The presence of a radionuclide in the core, such as 40K, provides a heat source to balance the planetary heat budget. Rb regularly substitutes for K in minerals due to its similar ionic charge and radius. 87Rb's decay product, 87Sr, is a tracer for geochemical reservoirs. Geochemical modeling shows that a two-reservoir model where the upper mantle interacts chemically with the lithosphere is more likely than whole mantle mixing. The decay of 87Rb to 87Sr allows us to model Rb sequestration in the core to constrain these two scenarios: Was Rb sequestered into the core upon formation or is it continually entering the core via mantle interaction? Iron and either Rb- or K- feldspar were loaded into Diamond Anvil Cells (DACs), and laser heated to the iron melting temperature at pressures of 12-75 GPa in order to test the reactivity of iron with K and Rb at a variety of possible magma ocean conditions. Synchrotron X-ray diffraction was performed on the quenched samples to determine the lattice expansion of iron due to the incorporation of feldspar components into its structure. Complementary TEM and SEM measurements constrain the concentrations of K, Rb, O, Al, and Si found in the recovered iron. By exploring the behavior of the Fe with the K and Rb found in these samples, a better understanding of Earth's inner dynamics is achieved.
DI43A-1769
Seismic Wave Velocity in Earth's Shallow Core
Studies of the outer core indicate that it is composed of liquid Fe and Ni alloyed with a ~10% fraction of light elements such as O, S or Si. Recently, unusual features, such as sediment accumulation, immiscible fluid layers or stagnant convection, have been predicted in the shallow core region. Secular cooling and compositional buoyancy drive vigorous convection that sustains the geodynamo, although critical details of light-element composition and thermal regime remain uncertain. Seismic velocity models can provide important constraints on the light element composition, however global reference models, such as Preliminary Reference Earth Model (PREM), IASP91 and AK135 vary significantly in the 200 km below the core-mantle boundary. Past studies of the outermost core velocity structure have been hampered by traveltime uncertainties due to lowermost mantle heterogeneities. The recently published Empirical Transfer Function (ETF) method has been shown to reduce the uncertainty using a waveform stacking approach to improve global observations of SmKS teleseismic waves. Here, we apply the ETF method to achieve a precise top-of-core velocity measurement of 8.05 ± 0.03 km/s. This new model accords well with PREM. Since PREM is based on the adiabatic form of the Adams-Williamson equation, it assumes a well mixed (i.e. homogeneous) composition. This result suggests a lack of heterogeneity in the outermost core due to layering or stagnant convection.
DI43A-1770
Detection of Multiply Reflected Core Phases at Broadband and Short Period Arrays in Australia
Australia is not favorably positioned relative to most global seismicity for the recording of PKP waves at long epicentral distances. However, the surrounding seismicity and a large number of recorders present vast opportunities to record and study seldom observed core-sensitive phases, such as the whispering gallery of P waves that reflect from the lower side of the core mantle boundary (PnKP waves). Prior reports of these seismic phases recorded at analog stations during the sixties and the seventies suggest that digital recordings of these phases in Australia could be abundant, especially with the advent of newer signal processing techniques. In addition, the lack of anthropogenic noise for most stations, and the presence of low attenuation in parts of the upper mantle, enable good signal-to-noise ratio. In the last 10-15 years, short- period and broadband arrays have been deployed simultaneously throughout different parts of the continent, which increases the probability of observing complex core phases and provides a good basis for comparative analysis. We design a procedure for detecting core-sensitive phases using seismograms recorded at different deployments. We search for PcP and PnKP in a systematic nature with a variety of techniques, such as sliding filters in the time domain and spectrograms in the frequency domain. In addition, utilizing the adaptive stacking method increases the visibility of the phases and aids in detection. Preliminary investigations of a partial dataset detect a significant number of arrivals. In particular, findings show several P4KP arrival times on the order of 5-10 sec earlier than predicted arrival times from ak135. A number of observed P4KP waveforms have smaller amplitude precursors several seconds prior to the main arrival. These precursors have the same slowness and similar frequency content as compared to the main arrival, all of which reinforces the precursory observation. We aim for further analysis of the PnKP and precursory arrivals toward interpreting the structure of the outer core.
DI43A-1771
Geodynamo Models with Core Evolution
Numerical dynamos with time-variable parameters that simulate the secular evolution of the core are used to interpret long-term trends in geomagnetic field behavior. We find dynamos with slow variations in convective forcing and decreasing rotation rate show trends in average dipole field intensity and fluid velocity, sometimes with systematic variations in polarity chrons. Similar dynamos driven by constant forcing and constant rotation have statistically stationary dipole field intensity and fluid velocity, with random polarity chrons. A reversing dynamo with steady rotational deceleration constrained by tidal friction, and a decreasing (regular) inner core growth rate constrained by core thermal history, evolves over 10 ~Myr with minor trends in average dipole intensity, fluid velocity, and polarity chron length. In contrast, a dynamo started from a non-reversing initial state and subject to an increasing (anomalous) inner core growth rate and constant rotation evolves in 20 ~Myr to a reversing state with a secular increase in variability, decrease in dipole intensity, and decrease in average polarity chron length. The dispersion of polarity chron lengths in the dynamo model with anomalous evolution is qualitatively similar to the observed dispersion of geomagnetic polarity chrons since the end of the Cretaceous Normal Superchron, and the model dipole field spectra are qualitatively similar to recent estimates of the geomagnetic spectrum at very low frequencies. Regular monotonic core evolution tends to produce slower (possibly neglgible) dynamo evolution, which may account for the observed stationarity of the dipole moment intensity over the last several hundred million years. The regular evolution dynamo also shows nearly constant reversal statistics, implying that the regime boundary separating reversing and non-reversing dynamo behavior is also regular, so that an episode of anomalous evolution may be required to initiate or end a magnetic superchron.
DI43A-1772
Effects of external imposed magnetic fields on 3-dimensional self-sustained numerical dynamos
As a general rule, planetary dynamos in the solar systems are not affected by external currents or electromagnetic forces. These sources are too far from the dynamo region and currents too weak to have a significant effect in the magnetic field generation. Mercury, with its large core and small magnetosphere, may be an exception to the rule. Interaction between the interplanetary solar wind with Mercury's internally generated magnetic field, gives rise to Chapman-Ferraro currents and an associated secondary magnetic field that can significantly modify the operation of Mercury's dynamo. To study the influence of Chapman-Ferraro curents we modeled the effect of an external constant field on solutions of self-sustained numerical dynamos. We explored three regimes where the self-sustained solutions yield: stable dipolar; stable dipolar with a significant multipolar component; and reversing dipolar dominated dynamos. The external field was chosen to oppose the direction of the original axial dipole, and its strength was varied in order to find the minimum necessary to affect the original dynamo magnetic field. We find the most dramatic effect on the overall magnetic energy in cases where the external field acts on the most stable dipolar regime. In those cases, the external magnetic field can cause a polarity reversal if its magnitude is at least 3% of the original dynamo field magnitude at the core-mantle boundary (CMB). A stable dipolar solution with a significant multipolar component requires less than 1% of the time-averaged (CMB) surface field in order to be affected by the external field. For the reversing regime studied, the external field does not affect the total magnetic energy significantly. However, an external field with a magnitude of about 1% of the original field magnitude reverses the polarity rapidly and changes the reversal rate of the solution, causing the field to be preferentially aligned with the external field. In order to have a better understanding of Mercury's magnetic field, it is important to understand the feedback mechanism between internal and external fields. Thanks to spacecraft measurements from MESSENGER and Mariner 10 we will have better measurements of averaged internal and external magnetic fields as well as accurate estimates of the magnetospheric currents, which are critical in order to determine the dynamo regime of Mercury and thus unravel its magnetic field generation dynamics.
DI43A-1773
Torsional Oscillations in a Numerical Geodynamo Operating in a Regime of Low Ekman and Magnetic Prandtl Numbers
Numerical simulations of geodynamo are carried out to understand the dynamic state of the Earth's core. Our model core consists of an electrically conducting, rotating Boussinesq fluid spherical shell and a solid inner core which has the same conductivity as the outer core and is free to rotate about the rotation axis by magnetic and viscous torques. Thermal convection occurs by simultaneous effects of secular cooling and heating at the inner core boundary. The Ekman number is 5 × 10-7, the magnetic Ekman number is 2.5 × 10-6, and the magnetic Prandtl number is 0.2, by which we aim to analyze a dynamic state in which the kinetic/magnetic energy ratio and the contribution of the viscous force are both small. A main target of this study is to elucidate torsional oscillations in the core. Theoretical study predicts that the torsional oscillations have a decadal time-scale in the Earth's core and might be responsible for observed variations of length of day through exchange of angular momentum between the core and the mantle. We expect that viscous damping of waves can be reduced by lowering the magnetic Prandtl number and advection plays a secondary role in the wave equation because of low magnetic Ekman number. As the Rayleigh number is increased from 640 to 6400, the kinetic and the magnetic energies increase almost linearly with the Rayleigh number, by their ratio being about 0.1 throughout. At the highest Rayleigh number, the volume-averaged Elsasser number in the core reaches 0.7. The zonal velocity averaged on a cylindrical surface coaxial with the rotation axis is analyzed as a function of radius of the cylinder s and time t. Outside the tangent cylinder (s>0.35), a clear wave-like signature is found in st-space. There are both ingoing and outgoing waves whose phase velocity agrees with theoretical estimates. The Lorentz force acting on the cylindrical surface dominates the viscous force, as anticipated. However, the advection term is still of the same order as the Lorentz force in the wave equation, indicating a magnetostrophic balance is incomplete. The dynamical regime may be compared with that of Dumberry and Bloxham (Phys. Earth Planet. Inter. 140, 29, 2003). We will discuss the results of wave analysis and will present results of numerical calculations at lower Ekman numbers.
DI43A-1774
Instabilities and Small-Scale Flow in Earth's Core
Aiming to better understand the behaviour of small-scale turbulence in Earth's core, we examine a very small cube of fluid in the core with a strong uniform magnetic field and periodic boundary conditions. Solving the hydrodynamic stability problem gives the qualitative picture of the dominant structure of small-scale turbulence. In this work, we solve the secondary stability problem for this system, which gives an idea of how the primary structures are ultimately broken down by non-linear effects. To solve this problem we find the spectral coefficients for the eigenfunctions, and we demonstrate how this can be done using an efficient numerical method. The eigenvalues and eigenfunctions contain information about the transfer of energy between scales during convection, which will be used for modelling the sub-grid scale in dynamo simulations.