The currents flowing in the magnetosphere-ionosphere (M-I) form a complicated multi-scale geosystem that contains temporal and spatial scales from seconds to days over meters to global lengths. Ground-based magnetometers have long been the main tool to observe the M-I current system and a number of indices have been introduced to characterize the variations of specific current components. Due to the nature of this current system, the magnetic effects recorded on the magnetograms are multi-scaled, impulsive, and asychroneous with non-stationary frequency spectra. This leads to cross-talk among the traditional geomagnetic indices. In this statistics-geosciences collaborative research, the investigators develop new statistical techniques based upon wavelet analysis, which is especially suitable for analyzing signals that are impulsive and multi-scaled and have time-dependent spectra. Their new methodology is based on the analysis of a global distribution of magnetograms in the INTERMAGNET database. This investigation develops a technique for filtering multivariate time series with an objective of removing cross-talk signatures that contaminate the present-day geomagnetic indices. New approaches are developed for modeling time series with complex scaling and temporal structures, which may offer a simpler alternative to or complement various multi-fractal models. In addition to providing new insights into the structure of the magnetometer records, their methodology will be applicable to similar problems in which multivariate time series with intricate time-varying scaling and distributional properties need to be understood.
Solar storm impacts in space and on Earth are usually adverse to humans due to enhanced radiation levels and also technology via electromagnetic charging-discharging and induced currents. These adverse effects are categorized as space weather. Civilian and defense weather centers specify and forecast the magnitude of their impacts to a broad range of commercial, military, and scientific users. Geomagnetic indices are the main measure of the strength of these impacts. For the first time in over 50 years, modern statistical and computational techniques are applied to resolving spatial and temporal structures in these indices on a global scale. This investigation develops improved magnetic indices of a significantly higher resolution. These new indices will be made available to the scientific and space weather forecast communities.
