Improving Ionospheric Source Models for Imaging Upper Mantle/Transition Zone Resistivity
Abstract
Information about deep Earth (> 200 km) resistivity comes almost exclusively from magneto-variational data, a superposition of magnetic fields generated by complex processes in the ionosphere and magnetosphere, and by induction in the spatially heterogeneous conducting Earth. Frequencies of 0.5-10 cycles per day, required to image the upper mantle from the aesthenosphere to transition zone, mostly have their origin in the ionospheric dynamo region at 100-150 km height, controlled by the spatial and temporal varying thermospheric neutral wind and the ionospheric conductivity distribution. To interpret the relatively subtle induced internal signals, these spatially complex ionospheric signals must be properly accounted for. Here we present progress on a collaboration between specialists in ionospheric physics and in EM induction, to develop methods to constrain better external source spatial structure, so as to reveal solid Earth heterogeneity. There are two novel aspects to our approach. First, we use robust frequency domain Principal Components Analysis (PCA) to reduce a large and heterogeneous set of historical and modern geomagnetic observatory data to the dominant spatial modes present in the daily variation band. The PCA scheme, massively reduces the size of the dataset, and allows data from different eras to be merged into more densely sampled spatial modes. Second, we use a mature physics based numerical model, the Thermosphere-Ionosphere-Electrodynamics General Circulation Model (TIE-GCM) to constrain source spatial structure. TIE-GCM external magnetic field outputs are analyzed in the frequency domain using PCA. Results are used to define a reduced-rank covariance, which is used with optimal interpolation to define global source structure for each data mode. Initially, we allow for internal fields with a global thin-sheet model of ocean/continent conductivity variations. The estimated sources can then be used to invert the data for Earth conductivity, with refined Earth models used to iterative improve source estimates. Initial efforts to model sources will also feed-back to refine our computations with TIE-GCM - e.g., variability in the observations may suggest need for enhanced forcing of one sort or another.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2016
- Bibcode:
- 2016AGUFMGP24E..02E
- Keywords:
-
- 1515 Geomagnetic induction;
- GEOMAGNETISM AND PALEOMAGNETISMDE: 2799 General or miscellaneous;
- MAGNETOSPHERIC PHYSICSDE: 4313 Extreme events;
- NATURAL HAZARDSDE: 7904 Geomagnetically induced currents;
- SPACE WEATHER