Studies of the seismic record indicate that seismic waves travel fastest along a north-south axis from the southern hemisphere to Siberia (and more recently suggesting an easterly shift to over Alaska) implying that over the the last century it appears that the earths solid core is rotating in an easterly direction. The conclusion made from this data is that the earths solid inner core can be said to be rotating west to east at approximately 1 degree longtitude a year faster than the mantle.(Song, Richards 2005) This implies then, that relative to Earths solid inner core, the mantle and the outer part of the liquid core can be said to be moving in a westerly direction. This also implies that there must be speed gradient in the liquid core across its radius (relative to the mantle), of approximately 0 degrees/year near the mantle to 0.1 degreeslongtitude/year near the solid core. Thus using Lenzs` law, one can predict that the larger volume of the outer part of the liquid core will induce a residual westerly electrical current in the inner part of the liquid core (ie the outer liquid core moves in a westerly direction relative to the inner part of the liquid core). Flemings Right Hand rule predicts this westerly direction of the electrical current will in turn induce a magnetic south pole near Earths geographic North Pole. Recent experiments confirm this, in that a freely rotating liquid metal will generate a self sustaining magnetic field without a neccesary seed magnetic field to kickstart the dynamo effect as previously thought.(Generation of a Magnetic Field by Dynamo Action in a Turbulent Flow of Liquid Sodium. R. Monchaux et al 2007). (It is also possible that the induced current would travel from core to crust or vice versa depending on how one interprets Lenz and Fleming vis a vis the circulating liquid core. However the cores reversal would still induce the neccesary N-S flip in this interpretation)As the seismic record shows that sound waves travel fastest across earths N-S axis, this could also be consistent with the solid inner core being slightly elongated in the earths N-S axis. (sound waves travel fastest in a denser medium) Following from this, it can be hypothesised that contrary to the speculation that the earths core is a rigid crystal, one could alternatively give the `solid` inner core non rigid, or elastic properties. Observations support this possibility (Andrew Jephcoat and Keith Refson (2001-09-06). "Earth science: Core beliefs". Nature 413: 27-30. doi:10.1038/35092650.) If so its shape could then be said to oscillate over large timescales between that of elongated in the N-S axis to stretched along the equatorial plane. Obviously this implies that the inner cores` rotational speeds would vary between slower and faster than the mantle, when the solid core goes from wider at the equator to stretched at the poles respectively. The historical record of Earths pole reversals and changing field strength can thus be well modeled by this variable core speed model. In that if currently it is accepted that the `solid core` is stretched is the north south axis and rotates faster than the mantle then it follows that when the cores volume changes shape to that of more stretched in the equatorial axis, It must then be forced to slow its rotational speed and therefore rotate slower than the mantle . Not faster as currently observed. (ie the core then will rotate in a westerly direction relative to the mantle)In turn this would reverse the relative motion between the inner and outer parts of the liquid core (as described above)and in turn reverse the induced equatorial direction of the induced electrical current which would then reverse the direction of earths induced magnetic field from N-S to S-N. THis is observed in the historical seismic record.The current weakening of Earths magnetic field must then be due to the solid cores` rotational speed slowing relative to the mantle. In other words if it is assumed that currently the core is observed to be travelling faster than the mantle in an easterly direction at 0.1 longitude per year than that rate must have been faster in the past and will be slower in the future. This can only be explained if it is assumed that the `solid` core itself is becoming less stretched in the N-S axis and tending towards becoming stretched in the equatorial axis.To explain the observed off axis magnetic North pole using this `variable core speed model` one can again refer to the seismic record which shows that the axis of anisotropy, corresponding to the fastest direction for seismic waves, tilts about 10 degrees from the Earth's N-S axis of rotation, towards Siberia. From this data one can interpret that the inner core not only could be `stretched` in the N-S axis but also be physically tilted 10 degrees off axis towards Siberia as it rotates within the earths core. In effect while it is rotating faster than the mantle, its axis of mass is also tilted off center by 10 degrees and stretched slightly in the N-S axis. This will in turn effect the equatorial rotation of the liquid inner core which acts as a boundary between the mantle and the solid inner core. Modeling the liquid core flow is always speculative for any model, but it is not unreasonable to assume that a 10 degree off axis solid core that rotates around its N-S center of mass could create a flow in the inner part of the liquid core that was also tilted in an equal but opposite 10 degrees `off axis` in the equatorial plane. Resulting in an equal and opposite tilt in the induced magnetic North pole of 10 degrees towards the Arctic, using Lenz and Fleming as described previously. This predicted tilt in the magnetic pole towards the arctic is consistent with observation. As is the prediction that the axis of the solid core is tilted towards Siberia.The final element is to explain the northerly drift of the magnetic North pole. There is insufficient seismic data to confirm or refute this prediction but the variable core speed model presented in this paper predicts that this would be due to an equal northerly drift of the solid inner cores rotational axis from more than 10 degrees off axis as currently observed to one that will be tilted by 0 degrees, or directly N-S. In other words if the solid core is said to be currently rotating tilted by 10 degrees off axis and this induces a off axis electrical current and off axis magnetic pole. Then a solid core that is aligned to earths geographic N-S axis as it rotates will generate no off axis electrical current and therefore no off axis magnetic field. In other words the cores volume is stretched in the N-S axis and tilted by 10 degrees but is changing shape as it rotates so that its N-S stretch is tending towards 0 degrees. In fact it would be more correct to say that its nvolume is changing from stretched 10 degrees off axis to not stretched in either the N-S or equatorial axis. This would be in line with a slowing down of the solid cores rotational speed as seen in the observed weakening of the overal magnetic field. (The variable speed core model predicts that when the core is perfectly circular in volume it rotational speed relative to the mantle will be the same as the mantle ,ie 0 degrees per year. And *no* magnetic field will be generated)The conclusion this `variable speed core` model makes is that all the combined observations of earths core and magnetic field indicate that the solid core is currently stretched in the N-s axis, tilted off center by 10 degrees and rotating faster than the mantle in an easterly direction by 0.1 degrees per year. But also observations indicate that it must be slowing down, relative to the mantle, tilting its axis more towards earths vertical N-S geographical axis, and most likely becoming less stretched in the N-S axis.
see this url for an accompanying graphic explanation of the variable speed core model..
http://www.youtube.com/watch?v=CiCBrXKIH_0
