Session CD8 - Measuring and modelling geoelectric fields for GIC studies

Juliane Huebert (British Geological Survey, UK), Joana Alves Ribeiro (University of Coimbra, Portugal), Ciaran Beggan (British Geological Survey, UK), Ellen Clarke, onsite (British Geological Survey, UK)

Large geoelectric fields are induced by variations of the geomagnetic field over a range of periods from seconds to minutes particularly during severe space weather events. Accurate representation of the geoelectric field is vital for correctly estimating Geomagnetically Induced Currents (GICs) in grounded technologies. The geoelectric field can be measured directly using magnetotelluric methods for example or inferred using large-scale conductivity properties of the subsurface. At present, there is a general lack of high-quality data and models specifically for space weather applications. In this session, methods for measuring and modelling the geoelectric field are examined, including new 1D/3D models of conductivity, data collection methodology or resources. We also solicit talks on improvements in real-time modelling of the geoelectric field over wide areas and new techniques for using space weather forecasts to estimate extreme geoelectric values.

Poster Viewing
Thursday October 27, 08:30 - 13:30, Poster Area

Talks
Wednesday October 26, 14:15 - 15:15, Earth Hall

Click here to toggle abstract display in the schedule

Talks : Time schedule

Wednesday October 26, 14:15 - 15:15, Earth Hall

14:15	3-D modelling of the geoelectric field and geomagnetically induced currents in Fennoscandia with laterally nonuniform inducing sources	Marshalko, E et al.	Oral
	Elena Marshalko[1], Ari Viljanen[1], Mikhail Kruglyakov[2], Alexey Kuvshinov[3]
	[1]Finnish Meteorological Institute, Helsinki, Finland; [2]University of Otago, Dunedin, New Zealand; [3]Institute of Geophysics, ETH Zurich, Zurich, Switzerland
	In this study, we carry out three-dimensional (3-D) geoelectric field (GEF) modelling in Fennoscandia for three days of the Halloween geomagnetic storm (29-31 October 2003) with the use of a recently developed real-time 3-D GEF modelling technique. This method relies on the factorization of the inducing source by spatial modes (SM) and time series of respective expansion coefficients and exploits precomputed GEF kernels generated by corresponding SM. We invoke a high-resolution 3-D conductivity model of Fennoscandia and consider a realistic source constructed using the Spherical Elementary Current Systems (SECS) method as applied to magnetic field data from the International Monitor for Auroral Geomagnetic Effect (IMAGE) magnetometer network. The factorization of the SECS-recovered source is then performed using the principal component analysis. We then calculate geomagnetically induced currents (GIC) based on the obtained GEF and compare modelling results with GIC observations at the Mäntsälä pipeline recording point. We further introduce an alternative method of the inducing source reconstruction, which is based on the approximation of magnetic fields observed by IMAGE magnetometers via a linear combination of magnetic fields generated by SM in a given 3-D conductivity model. This approach allows us to obtain more accurate SM expansion coefficients. After this, we calculate the GEF using the real-time modelling technique mentioned above and compute GIC at the Mäntsälä recording point based on the modelled GEF. The correlation between observed GIC and GIC obtained using 3-D GEF modelling is high for both methods. However, a higher correlation with measurements is observed in the case of the second modelling technique. Our results also show that the largest uncertainty of GIC estimates arises from the uncertainty of the conductivity model, which has a strong effect on the GEF.
14:30	European-wide geo-electric field and geomagnetically induced current modelling developments	Richardson, G et al.	Oral
	Gemma Richardson, Ciarán Beggan, Guanren Wang, Ewelina Florczak and Ellen Clarke
	British Geological Survey
	The EUHFORIA2.0 Horizon 2020 project aims at developing an advanced space weather forecasting tool, combining state of the art models from CME evolution all the way through to ground effects. As part of this project we have been developing the capability to use forecasts of magnetic field perturbations at ground level to compute electric fields, and geomagnetically induced currents (GIC) in the interconnected European power networks. Using a modified version of the EURHOM model (Adam et al, 2012), a conductivity model has been constructed, providing a series of 1D layered blocks of conductivity. This can then be combined with forecasts of ground-level magnetic field variations across Europe to compute the geo-electric fields across the continent. We have also developed a new model of the European power grid network, which when combined with the geo-electric field model, provides estimates of GIC right across Europe. This full process has been completed for a series of hindcasts of historical geomagnetic storms to allow us to compare the results to the limited geo-electric and GIC measurements that exist, and to identify the areas in Europe more vulnerable to GIC effects during geomagnetic storms. This work is funded by the European Commission Horizon 2020 (H2020) Grant Agreement No. 870405.
14:45	The assessment of GICs based on time-domain transfer functions	Kruglyakov, M et al.	Oral
	Mikhail Kruglyakov[1], Craig J. Rodger[1], Daniel H. Mac Manus[1], Michael Dalzell[2], Tanja Petersen[3]
	[1]Department of Physics, University of Otago, Dunedin, New Zealand, [2]Transpower New Zealand Limited, Wellington, New Zealand, [3]GNS Science, Lower Hutt, New Zealand.
	Following Maxwell equations one can express mathematical relationships between GICs and the changing geomagnetic field in the form of convolution-type integrals of the magnetic fields and some kernels. We call these kernels “time-domain transfer functions” (TDTFs) as they are independent of the source. When both the GICs and magnetic field measurements are available, these TDTFs can be obtained by solving the corresponding systems of linear equations. One of the advantages of this approach, when contrasted with the conventional frequency domain transfer functions, is that one can use time series with gaps and also properly deal with non-uniform time-steps in the raw measurements as is --- that is, without any interpolation or averaging requirements. When TDTFs have been obtained by data for some time intervals, they can be used for now-casting or even forecasting GICs by the use of measured or forecasted magnetic field. Unfortunately, their usefulness for this is rather limited due to the fact that it appears that (relatively) often changes in the power network configuration also affect the TDTF parameters. However, the analysis of TDTFs allow us to study different aspects of the GICs behaviour, considering a route for proper simulation without concern for the influences inaccuracies in the conductivity and magnetic field source models. Examples are: How long is the “memory” of GICs in the studied region? What time resolution do we really need for the magnetic field? Should we use only horizontal components of magnetic field, as in mainstream impedance-based methods, or should we include the vertical component? And last but not least, how many (or which) magnetic observatories provide us enough data to compute GICs at a particular site with the requested accuracy? We have applied the approach described above to GICs and magnetic field data measured at multiple locations in the South Island of New Zealand. We obtain remarkable agreements between the modelled and observed data, with correlation near 0.9 and $R^2$ near 0.8, if all three components of magnetic field with 1s resolution are used.
15:00	Nowcasting Geoelectric Fields in Ireland using Magnetotelluric Transfer Functions	Malone-leigh, J et al.	Oral
	John Malone-Leigh [1,2], Joan Campanya[3,4], Peter T. Gallagher[1], Maik Neukirch[5],Colin Hogg[1], Jim Hodgson[4]
	[1] Dublin Institute for Advanced Studies, Ireland [2] Trinity College Dublin, Ireland [3] South-East Technological University, Ireland, [4] Geological Survey Ireland, Ireland [5] University of Oslo, Norway
	Geomagnetically induced currents (GIC) driven by geoelectric fields pose a hazard to ground-based infrastructure, such as power grids and pipelines. Here, a new method and implementation is presented for modelling geoelectric fields in near real time, with the aim of providing valuable information to help mitigate the impact of GIC. The method uses magnetic field measurements from the Magnetometer Network of Ireland (MagIE), interpolates the geomagnetic field variations between magnetometers using spherical elementary current systems (SECS), and estimates the local electric field using magnetotelluric transfer functions (MT-TF). The method was optimised to work in real time by comparing to a standard non real-time model and adding a correction to the output electric fields to account for an underestimation close to real time, due to a lack of long period information. This approach proved successful, providing accurate electric fields up to a 1-minute delay, with high coherence (> 0.7) and signal to noise ratio (SNR > 3) relative to measured electric field validation time series. This was comparable to a standard non real-time geoelectric field model (coherence > 0.8 and SNR > 4). The usefulness of a galvanic distortion correction for space weather was also briefly evaluated, with a galvanic correction demonstrated to give a more homogeneous representation of the direction of the electric field at a regional scale, which could help improve GIC modelling.

Posters

1	Geomagnetically induced currents in the German power grid	Pick, L et al.	Poster
	Leonie Pick, Aline Guimaraes Carvalho, Aoife E. McCloskey, Jens Berdermann
	German Aerospace Center (DLR), Institute for Solar-Terrestrial Physics, Department for Space Weather Impact
	Space weather driven geomagnetically induced currents (GICs) occur under conditions of intense geomagnetic disturbance (GMD), which is particularly strong in the vicinity of the high-latitude ionospheric electrojets during magnetospheric substorms. By now, it is well established that even mid-latitude regions are at risk of GICs disrupting ground conducting systems in response to the storm-time expansion of the auroral electrojets. Nevertheless, GIC research in Germany is still in its infancy. Here, we present our current progress toward a comprehensive assessment of the GIC risk in the German high-voltage power grid. Our analyses comprise both the geophysical as well as the engineering side of the problem. The first part is concerned with the quantification of the GIC hazard imposed by critically large induced electric fields. Given the scarcity of direct measurements, we calculate the induced electric field by applying the plane wave approach with a 1-D conductivity model (EURISGIC) to maps of GMD generated from spatially interpolated INTERMAGNET observatory measurements. The engineering part of the problem deals with the exposure and vulnerability of the power grid to the hazard imposed by the induced electric fields. This requires knowledge of the technical specifications of the German high-voltage transmission network, which we collect from publicly available sources. Future efforts will include the validation of the outlined calculation scheme by direct measurements of GICs at German transformer substations.
2	Expected Geomagnetically Induced Currents in the Spanish islands power transmission grids	Torta, J et al.	Poster
	J.M. Torta[1], S. Marsal[1], P. Piña-Varas[2], R. Hafizi[2], A. Martí[2], J. Campanyà[3], V. Canillas-Pérez[1], J.J. Curto[1], J. Ledo[2], P. Queralt[2], A. Marcuello[2]
	[1]Observatori de l’Ebre (OE), Univ. Ramon Llull - CSIC, Roquetes, Spain;[2]Institut Geomodels, Dept. Dinàmica de la Terra i de l’Oceà, Universitat de Barcelona, Barcelona, Spain; [3] Department of Build Environment, South East Technological University, Ireland
	After having recently modelled in detail and indirectly measured geomagnetically induced currents (GICs) in the power transmission system of mainland Spain, including the 400 and 220 kV voltage levels, and to strictly complete the assessment at the national level, the aim of this study is to evaluate the GIC hazard in the power transmission networks of the Canary Islands and Balearic archipelagos. We constructed models for the transmission networks in each of the individual systems and used electrical resistivity models of the lithosphere for each group of islands, from which we calculated the impedances at the surface. These models were obtained from MT data inversion, as in the case of Tenerife, or were constructed based on geological and geophysical information. The respective models of electrical admittances of the power grids have been combined with the geoelectric field derived from the convolution of the recorded (or expected in an extreme scenario) geomagnetic storms and the impedances calculated from the geoelectrical models to derive the expected GIC in the power lines, substations, and transformers of those islands. The low geomagnetic latitude of the Canary Islands combined with the small size of their power transmission networks, makes the archipelago one of the least likely electrified locations in the world to record significant GICs. The amplitudes of 3-4 A expected on the most susceptible power line of the Tenerife power grid in the 100-year return period are certainly insignificant. Even the nearly 20 A that could be reached for the upper limit of the 95% confidence interval at the 500-year return period does not seem likely to have a significant impact on the network. On the contrary, in addition to the notably higher latitude (which makes them directly more prone to higher geomagnetic disturbances), the Balearic Islands network has long AC (mostly submarine) power transmission lines connecting the islands, which make the system length of approximately 300 km on its main axis. This means that GIC signals of moderate amplitude can be reached there.
3	Extreme values and return levels of modelled geoelectric fields at the UK observatories	Beggan, C et al.	Poster
	C. Beggan, J. Huebert, G.S. Richardson
	British Geological Survey, Edinburgh, UK
	During geomagnetic storms, rapid variations in the Earth’s magnetic field induce large ground electric fields that pose a hazard to grounded technological infrastructure. These geoelectric fields can vary significantly depending on the conductivity structure of the local geology. Understanding how the ground electric field varies during geomagnetic storms allows for improved hazard estimation from rare and extreme space weather events. Using simultaneous geoelectric and magnetic field measurements from six months of data, magnetotelluric (MT) transfer functions for the three magnetic observatories (Lerwick, Eskdalemuir and Hartland) operated by the British Geological Survey in the UK were computed. These relate changes of the magnetic field to the induced geoelectric field at each site and so can be used to determine the geoelectric field during historic storms. Digital magnetic records of the magnetic extend from January 1983 to present covering almost 40 years. We compute the modelled geoelectric field from the magnetic data for 40 years to examine the distribution of the extremes and to estimate the 1 in 50- and 100-year return values. We find the largest geoelectric field values are: 3.8 V/km, 1.7 V/km and 0.2 V/km at Lerwick, Eskdalemuir and Hartland respectively, all experienced during the March 1989 storm.
4	The UK’s long-period magnetotelluric field campaign for improved ground electric field modelling	Eaton, E et al.	Poster
	E. Eaton [1], J. Huebert [1], C. Beggan [1], A. Montiel-Alvarez [2], A. Thomson [1], C. Hogg [3] and D. Kiyan [3]
	[1] British Geological Survey, Edinburgh, UK [2] University of Edinburgh, Edinburgh, UK [3] Dublin Institute for Advanced Studies, Dublin, Ireland
	During geomagnetic storms, rapid variations in the Earth’s magnetic field can induce large ground electric fields that pose a hazard to grounded technological infrastructure, via Geomagnetically Induced Currents (GICs) in the UK power transmission networks. These ground electric fields can vary significantly depending on the electrical conductivity structure of the continental lithosphere. Accurately modelling the ground electric field spatially during geomagnetic storms allows for improved hazard estimation to ground technological infrastructure, yet currently in the UK, continuous observations are only made at the three magnetic observatories operated by the British Geological Survey. Part of the UK-funded SWIMMR-SAGE (N4) project involves improving the conductivity model of the subsurface based on magnetotelluric (MT) data for modelling the ground electric field. Using a rolling deployment of five Lemi-417 MT systems, we are collecting long-period MT data at more than 45 sites with approximate 60 km spacing across England, Southern Scotland, and Wales between 2021 - 2023. Here, results from a selection of the 30 stations collected thus far are shown. The magnetotelluric impedance tensors, characterising the conductivity structure at each station, are used to model induced ground electric fields during the 03 November 2021 geomagnetic storm. Measured ground electric field data during larger space weather events between May 2021 and August 2022 are also presented. This new data set will be used as part of a nowcast and forecast service on potential space weather impacts on ground-level infrastructures.