Modeling Geoelectric Fields and Geomagnetically Induced Currents Around New Zealand to Explore GIC in the South Island's Electrical Transmission Network

Abstract

Transformers in New Zealand's South Island electrical transmission network have been impacted by geomagnetically induced currents (GIC) during geomagnetic storms. We explore the impact of GIC on this network by developing a thin‐sheet conductance (TSC) model for the region, a geoelectric field model, and a GIC network model. (The TSC is composed of a thin‐sheet conductance map with underlying layered resistivity structure.) Using modeling approaches that have been successfully used in the United Kingdom and Ireland, we applied a thin‐sheet model to calculate the electric field as a function of magnetic field and ground conductance. We developed a TSC model based on magnetotelluric surveys, geology, and bathymetry, modified to account for offshore sediments. Using this representation, the thin sheet model gave good agreement with measured impedance vectors. Driven by a spatially uniform magnetic field variation, the thin‐sheet model results in electric fields dominated by the ocean‐land boundary with effects due to the deep ocean and steep terrain. There is a strong tendency for the electric field to align northwest‐southeast, irrespective of the direction of the magnetic field. Applying this electric field to a GIC network model, we show that modeled GIC are dominated by northwest‐southeast transmission lines rather than east‐west lines usually assumed to dominate.

Assessment of GIC Based On Transfer Function Analysis

“…For field variations of 2.1 min period in this orientation, in response to a field variation of 100 nT GIC of nearly 40 A can be expected at SDN, while up to 10 A and 5 A can be expected at INV and MAN, respectively. These magnitudes differ, however, from those found for a period of 10 min by Divett et al () whose results suggested smaller GIC at SDN than at either INV or MAN. Although these differences may well reflect the acknowledged uncertainty in the conductance model in the lower part of the South Island, they may well be partly explained by the fact that the largest GIC are clearly associated with much more rapid variations of the magnetic field than 600 s period.…”

“…At SDN the real parts of both A ( A r ) and B ( B r ) are negative and become increasingly so at shorter period, whereas at both INV and MAN these parts of the transfer functions are both much smaller and, in general, tend to be positive. In terms of the direction of GIC at all three locations, this is in agreement with the results shown by Divett et al () based on thin sheet modeling and the interaction of induced electric fields and the actual transmission network. That study suggested that for field variations of 10 min period, the direction of GIC at SDN, on the east coast of the South Island, was opposite to that observed at INV and MAN on the south and west coasts, respectively.…”

“…The responsiveness of GIC to the orientation of the variations in the magnetic field is illustrated in Figure , which shows, for different periods of variation of the magnetic field, polar plots of the total magnitude of GIC produced by an assumed 100 nT variation in field oriented in different directions. Whereas Torta et al () suggested that the orientation of transmission lines relative to the induced electric fields is important in determining the magnitude of GIC, Figure is in agreement with the thin sheet modeling results of Divett et al () in showing that the largest GIC in the South Island of New Zealand are produced when the orientation of the geomagnetic field variations broadly aligns with the overall SW‐NE geological trend of New Zealand associated with its location on the boundary of the Australian and Pacific tectonic plates. For field variations of 2.1 min period in this orientation, in response to a field variation of 100 nT GIC of nearly 40 A can be expected at SDN, while up to 10 A and 5 A can be expected at INV and MAN, respectively.…”

“…These are then in turn used as input into a model of the electric transmission network to calculate the resulting GIC. This method has been used to study GIC in France (Kelly et al, ), Ireland (Blake et al, ), and the UK (McKay, ), as well as being recently applied in New Zealand (Divett et al, ). Although such modeling has the potential to allow investigation of mitigation strategies by testing the effect of modifications to the transmission network on GIC, it also has significant drawbacks.…”

Transformer‐Level Modeling of Geomagnetically Induced Currents in New Zealand's South Island

“…This approach had also been used successfully to calculate geoelectric fields for GIC in the United Kingdom, and in Austria by Beggan et al (), Mckay (), and Bailey et al (), respectively. Divett et al () compared the output of the thin‐sheet model to field studies by comparing against induction vectors measured by Chamalaun and McKnight (), finding good agreement in magnitude and direction. However, Divett et al () did note some errors in the direction of calculated fields at the coast compared to measured induction vectors, which was inferred to be due to the relative coarseness of the modeled coastline in the thin‐sheet conductance model.…”