Geomagnetically induced current model validation from New Zealand's South Island

Divett, T.; MacManus, D.H.; Richardson, G.S.; Beggan, C.D.; Rodger, C.J.; Ingham, M.; Clarke, E.; Thomson, A.W.P.; Dalzell, M.; Obana, Y.. 2020 Geomagnetically induced current model validation from New Zealand's South Island. Space weather, 18 (8), e2020SW002494.

Before downloading, please read NORA policies.
Text (Open Access Paper)
2020SW002494.pdf - Published Version
Available under License Creative Commons Attribution 4.0.

Download (7MB) | Preview


Geomagnetically induced currents (GICs) during a space weather event have previously caused transformer damage in New Zealand. During the 2015 St. Patrick's Day Storm, Transpower NZ Ltd has reliable GIC measurements at 23 different transformers across New Zealand's South Island. These observed GICs show large variability, spatially and within a substation. We compare these GICs with those calculated from a modeled geolectric field using a network model of the transmission network with industry‐provided line, earthing, and transformer resistances. We calculate the modeled geoelectric field from the spectra of magnetic field variations interpolated from measurements during this storm and ground conductance using a thin‐sheet model. Modeled and observed GIC spectra are similar, and coherence exceeds the 95% confidence threshold, for most valid frequencies at 18 of the 23 transformers. Sensitivity analysis shows that modeled GICs are most sensitive to variation in magnetic field input, followed by the variation in land conductivity. The assumption that transmission lines follow straight lines or getting the network resistances exactly right is less significant. Comparing modeled and measured GIC time series highlights that this modeling approach is useful for reconstructing the timing, duration, and relative magnitude of GIC peaks during sudden commencement and substorms. However, the model significantly underestimates the magnitude of these peaks, even for a transformer with good spectral match. This is because of the limited range of frequencies for which the thin‐sheet model is valid and severely limits the usefulness of this modeling approach for accurate prediction of peak GICs.

Item Type: Publication - Article
Digital Object Identifier (DOI):
Date made live: 06 Nov 2020 11:31 +0 (UTC)

Actions (login required)

View Item View Item

Document Downloads

Downloads for past 30 days

Downloads per month over past year

More statistics for this item...