Abstract
Strong electric currents that flow in the near-Earth space and close through the upper atmosphere can be generated during geospace magnetic storms. The magnetic field of these currents induces currents in the Earth's surface called Geomagnetically Induced Currents (GICs), the intensity of which depends on the distribution of the electrical properties in specific areas of the Earth's solid crust. GIC occurrence can potentially result in causing serious disrupts or damages to the electricity transmission and distribution network. GICs are the ground end of the space weather chain: Sun– solar wind – magnetosphere – ionosphere – Earth's surface. Hence, nowadays they constitute an integral part of the space weather research. Traditionally, it was thought that only electricity networks located in high latitudes (Northern America, Scandinavia) are affected by GICs. But this cannot explain the recently reported existence of electrical power issues in areas of low latitudes (e.g., Spain, South ...
Strong electric currents that flow in the near-Earth space and close through the upper atmosphere can be generated during geospace magnetic storms. The magnetic field of these currents induces currents in the Earth's surface called Geomagnetically Induced Currents (GICs), the intensity of which depends on the distribution of the electrical properties in specific areas of the Earth's solid crust. GIC occurrence can potentially result in causing serious disrupts or damages to the electricity transmission and distribution network. GICs are the ground end of the space weather chain: Sun– solar wind – magnetosphere – ionosphere – Earth's surface. Hence, nowadays they constitute an integral part of the space weather research. Traditionally, it was thought that only electricity networks located in high latitudes (Northern America, Scandinavia) are affected by GICs. But this cannot explain the recently reported existence of electrical power issues in areas of low latitudes (e.g., Spain, South Africa, Italy, Japan, China),that is latitudes similar to those of Greece. Here, we primarily investigate the possibility of GIC development in Greece. We analyze magnetic field timeseries from ground-based magnetometers located mainly in Europe, focusing on the Mediterranean region, spaceborne magnetometers from the European Space Agency (ESA) Swarm satellites, as well as geomagnetic activity indices during magnetic storms and substorms. For the analysis, we use advanced signal processing methods based on wavelet transforms, Hurst exponent calculations, and entropy measures to capture the potential existence of characteristic signatures prior to the occurrence of magnetic storms. Since GICs are related to space weather events, indications of impending storms contribute to early warnings for the development of GICs in areas hosting critical technological infrastructure. Next, we calculate the GIC index for Greece and South Europe, in general, focusing on magnetic storms that occurred during the previous (24th) solar cycle. Furthermore, we calculate the intensity of the geoelectric field E, taking into account the electrical conductivity of the ground and, consequently, conducting an analytical assessment of the vulnerability of the electrical power system in our country as well as in southern Europe during the occurrence of extreme space weather events, such as magnetic storms. The question we are called upon to answer is whether Greece's electrical power grid, located in a middle geomagnetic latitude region, is at risk and to what extent from GICs that are expected to develop in the case of a strong magnetic storm. Calculating a GIC index can significantly contribute to the implementation of protective measures by electricity providers and distribution network operators in Greece, as well as in Europe, on a future basis. This also contributes to research towards space weather prediction. This is the first attempt to assess GIC development in the Greek territory, a study that aligns perfectly with similar research conducted worldwide. Therefore, this dissertation largely fills the existing gap in the literature by providing an estimation and model construction of GIC distribution patterns in Greece. In detail, after an introductory chapter dedicated to the necessary physical framework regarding the study of space weather phenomena and specifically geospace magnetic storms and GICs, we present the geomagnetic field timeseries (and geomagnetic activity indices) analysis methods used for this dissertation. In particular, we discuss the wavelet spectral analysis, the Hurst exponent, and the entropy measures of Shannon entropy, nonextensive Tsallis entropy, and Fisher information. In Chapter 3, we investigate the dynamical complexity of geomagnetic activity indices, using Information Theory. The methods described in Chapter 2 are applied to the SYM-H and AE indices of geomagnetic activity, as well as to the Swarm-derived SYM-H and Swarm-derived AE indices, two geomagnetic activity indices emanating from spaceborne data from the ESA Swarm constellation of satellites. Chapter 4 is dedicated to the GIC index calculation. The data and methodology are presented and the GIC index is calculated for the strongest magnetic storms (i.e., minimum Dst index -150 nT) of solar cycle 24 in magnetic stations / observatories of Greece and the wider Mediterranean area. Results are displayed in plots where a five-level color scale is used to match the five risk levels (very low to severe). Correlations with the Storm Sudden Commencement (SSC) of each storm are also presented. In Chapter 5 contour maps of activity are presented. They focus on the Mediterranean region during the strongest magnetic storm of solar cycle 24, that occurred on 17 March 2015 (St. Patrick's Day storm). Contour maps present either the GIC index along with the conductivity of the ground (1-D model) or the calculated geoelectric field E along with the electricity network (https://www.entsoe.eu/data/map/).The conclusions of this dissertation can be summarized in the following:1. Wavelet spectral analysis, Hurst exponent analysis and entropic analysis of spaceborne and ground-based Earth's magnetic field time series, as well as geomagnetic activity indices. The spectral analysis in terms of wavelet transforms revealed that concurrently with each storm there is intense spectral content, characteristic of these events. In the case of the geomagnetic activity indices SYM-H and Swarm-derived SYM-H, the Hurst exponent and the entropy measures of Shannon entropy, nonextensive Tsallis entropy, and Fisher information indicated the existence of two different patterns: • A pattern associated with strong magnetic storms, characterized by high values of the Hurst exponent, implying higher "organization" in the magnetosphere. • A pattern related to the quiet periods of the magnetosphere, characterized by lower values of the Hurst exponent, indicating less "organization" in the magnetosphere. In the case of AE and Swarm-derived AE indices the wavelet spectral analysis revealed similar underlying features in the power spectra for the three storms, despite the fact that we are dealing with substorm indices: the big picture of the preconditioned magnetosphere is still present. However, the Hurst exponent and entropic analyses did not result in a clear depiction of two distinct patterns. This finding is attributed to the fact that these indices are related to substorms, which are more transient and dynamic, occur more frequently than storms and have different characteristic time scales and generation mechanisms compared to magnetic storms.2. GIC index analysis of ground-based Earth's magnetic field time series. Regarding the GIC index, our results showed a good correlation between the SSC and the increase in the GIC index. Furthermore, the maximum values of the GICy and GICx indices occur within the first four minutes from the abrupt onset of each respective storm at all the magnetic stations / observatories under study. At first glance, based on the GIC index values calculated for the time periods of the four magnetic storms, it appears that despite the elevated GIC index values, the expected detrimental effects due to GICs remain at low levels for the areas covered by the specific magnetic stations / observatories. However, the GIC index provides us with an initial estimate of the level of risk posed by the development of such currents to critical technological infrastructure without taking into account the geoelectric structure of the broader region, i.e., the distribution of electrical conductivity with depth, which could contribute to the variation in GIC values during a magnetic storm.3. Comparisons between GIC index values and electrical conductivity of the Earth's crust. As for the contour maps, it is observed that the contour lines are relatively sparse before (March 16, 2015) and after (March 18, 2015) the storm. However, they tend to become more horizontal and graded in intensity (increasing from south to north) during the storm on March 17, 2015, confirming the presence of higher "organization" in the Earth's magnetosphere. Taking into account the values of ground resistance (conductivity) and the GIC index allows us, to some extent, to visualize the impact of GICs on southern Europe. In the contour maps, it is noticeable that the contour lines of the GIC index appear smooth but traverse areas with significantly different ground conductivities. Therefore, this may have varying consequences concerning the development of GICs in critical infrastructure located in these regions. Hence, this specific parameter should be considered in assessing the risk associated with GICs.4. Comparisons between GIC index values and geoelectric field EComparing the estimated E fields and GIC indices, using measurements from the magnetic stations Chambon la Forêt (CLF), Castello Tesino (CTS), Dionysos (DIO), Ebro (EBR), San Fernando (SFS), and Velies (VLI) over a period of three days covering the magnetic storm of March 17, 2015 we found that their correlation coefficients range between 0.54 and 0.65. This suggests that while there is a discernible positive linear relationship between E fields and GIC indices, other factors may also influence the geoelectric field, resulting in moderate variability. The large disparity of ground conductivity values of 1D (layer) ground models for Europe means that local conductances can vary by a factor of over 100 even on the scale of countries, such as Spain or Greece. The calculated E fields vary locally by at least as much. This variability is added to local variations by a factor of around 1.5 to 2 in the calculated GICx and GICy indices for those two countries during the height of St. Patrick's Day storm. This shows the inadequacy of using a single observatory for the calculation of a nationwide index. Therefore, it is suggested that multiple magnetometers per big European country are needed to capture the complexity of induced E fields.
show more
